Datasets:

keerthanapacific

/

sample6

like 0

539331

Major Structural Differences and Novel Potential Virulence Mechanisms from the Genomes of Multiple Campylobacter Species

Sequencing and comparative genome analysis of four strains of Campylobacter including C. lari RM2100, C. upsaliensis RM3195, and C. coli RM2228 has revealed major structural differences that are associated with the insertion of phage- and plasmid-like genomic islands, as well as major variations in the lipooligosaccharide complex. Poly G tracts are longer, are greater in number, and show greater variability in C. upsaliensis than in the other species. Many genes involved in host colonization, including racR/S, cadF, cdt, ciaB, and flagellin genes, are conserved across the species, but variations that appear to be species specific are evident for a lipooligosaccharide locus, a capsular (extracellular) polysaccharide locus, and a novel Campylobacter putative licABCD virulence locus. The strains also vary in their metabolic profiles, as well as their resistance profiles to a range of antibiotics. It is evident that the newly identified hypothetical and conserved hypothetical proteins, as well as uncharacterized two-component regulatory systems and membrane proteins, may hold additional significant information on the major differences in virulence among the species, as well as the specificity of the strains for particular hosts.

Introduction The Gram-negative, spiral-shaped bacterium Campylobacter jejuni is commensal in cattle, swine, and birds [ 1 ]. Campylobacter species, however, are the major cause of human bacterial gastroenteritis, and may be responsible for as many as 400–500 million cases worldwide each year [ 2 ]. Although the genus Campylobacter is composed of 16 described species [ 3 ], human illness is associated primarily with C. jejuni and C. coli and infrequently with C. upsaliensis, C. lari, and C. fetus . Filtration-based isolation techniques have revealed C. upsaliensis to be associated with human disease more than previously known [ 4 ]. The majority of C. jejuni infections result in uncomplicated gastroenteritis, but the development of the peripheral neuropathies, Guillain-Barré and Miller-Fisher syndromes is often associated with prior C. jejuni infection [ 5 , 6 ]. All clinically relevant Campylobacter spp. are considered to be thermotolerant in nature. C. jejuni, C. coli, C. lari, and C. upsaliensis also grow readily under microaerophilic conditions (5% oxygen) at 37 °C, and the majority of strains from these species will also grow at 42 °C. The thermotolerant Campylobacter spp. can also be distinguished by their host range. C. jejuni and C. coli are commensal in cattle, swine, and birds [ 1 ]; however, C. jejuni is often the predominant species in poultry, and C. coli in swine [ 4 , 7 ]. C. lari is prevalent in birds (seagulls in particular) [ 8 ], but has also been isolated from dogs and swine [ 9 , 10 ]. C. upsaliensis has frequently been isolated from domestic dogs and cats [ 11 , 12 , 13 , 14 , 15 ]. The main route of C. jejuni and C. coli human infection is through improperly handled or undercooked poultry, although illnesses caused by the consumption of livestock meat, unpasteurized milk, and contaminated water have also been reported [ 1 ]. C. lari has been isolated infrequently from poultry, ox and pork livers [ 16 , 17 , 18 ], and produce [ 19 ], in contrast to frequent isolation at moderate to high levels from fresh water, seawater, and shellfish [ 20 , 21 ]. C. upsaliensis has been isolated infrequently from poultry, ducks, and shellfish, and not from other food sources [ 4 , 22 , 23 ]. The main reservoir of C. upsaliensis appears to be dogs and cats, with reports of transmission of C. upsaliensis from animal to person [ 24 , 25 ] or person to person [ 26 , 27 ]. Human illness caused by C. lari and C. upsaliensis, unlike C. jejuni and C. coli, may be due to proximity to water and shellfish, and handling of pets, livestock, or livestock carcasses. The genome sequence of C. jejuni strain NCTC 11168 [ 28 ], a human clinical isolate, provided a starting point for studying the proteins involved in outer surface structures and glycosylation [ 29 ], and the expression of contingency gene products such as glycosyl transferases and restriction enzymes. However, in contrast to the current understanding of the pathophysiology of other enteric bacteria, that of Campylobacter species remains poorly understood. The genome of one C. jejuni strain is insufficient to provide a complete picture of the major aspects of Campylobacter biology, including the colonization of reservoir hosts [ 30 ], variation in lipooligosaccharide (LOS) and capsule, and potential adaptations of Campylobacter in poultry production and processing environments. In addition, information on the basis of Campylobacter virulence and potential targets for drug and vaccine design is still lacking. Therefore, we sequenced and finished the genome of C. jejuni strain RM1221 ( ATCC BAA-1062), and compared it with the genomes of C. coli strain RM2228 ( ATCC BAA-1061), C. lari strain RM2100 ( ATCC BAA-1060), and C. upsaliensis strain RM3195 ( ATCC BAA-1059) sequenced to at least 8-fold coverage. Strain RM1221 was sequenced because it was isolated from a chicken carcass and minimally passaged [ 31 ]. In addition, experimental work with this isolate has identified a number of unique features not present in the previously sequenced C. jejuni strain NCTC 11168, including the colonization of chicken skin and ceca, invasion of Caco-2 cells [ 31 ], unique LOS and capsule loci, and other unique open reading frames (ORFs) (unpublished data). C. coli RM2228 was sequenced because it is a multi-drug-resistant chicken isolate. Both C. lari RM2100 (CDC strain D67, “case 6” [ 32 ]) and C. upsaliensis RM3195 were selected for sequencing because they are clinical isolates. C. upsaliensis RM3195 was isolated from a patient with Guillain-Barré syndrome, using a filtration-based method of selection [ 33 ], and may have been responsible for this disease. Results/Discussion Comparative Genome Features The genome of C. jejuni RM1221 is a single circular chromosome, 1,777,831 bp in length, with an average G+C content of 30.31%. There are a total of 1,884 predicted coding regions in the genome with an average ORF length of 885 bp. Ninety-four percent of the genome represents coding sequence. Putative role assignments could be made for 1,124 of the ORFs (60%) ( Table 1 ; Figure S1 ). The bacterium was found to belong to multilocus sequence type (MLST) 354 and FlaA short variable region (SVR) 33, which belongs to clonal complex 354, whose members are associated with human disease or chickens/chicken meat ( Table 1 ) [ 34 ]. The genome features for the unfinished Campylobacter genomes were based on automated analysis and are presented in Table 1 . The average coverage of the unfinished genomes was found to be 8.5-fold for C. coli RM2228, 16.5-fold for C. lari RM2100, and 9.0-fold for C. upsaliensis RM3195 for those contigs used to construct the pseudomolecules. The ambiguity rate (number of consensus-altering ambiguities per basepair) was determined to be between 1:54,000 and 1:93,000 for these unedited, unfinished genomes at 8-fold depth of coverage. The genomic structure of C. jejuni RM1221 is syntenic with the genome of C. jejuni NCTC 11168, and is disrupted by inserted prophages/genomic islands in RM1221 (see below), and ORFs within the capsular (extracellular) polysaccharide (EP) loci in NCTC 11168 ( Figures 1A and S2 ). The C. coli RM2228 genomic structure also has a considerable amount of synteny with C. jejuni RM1221, sharing similar breakpoints, as observed in the C. jejuni comparisons, but displaying evidence of rearrangements about the oriC, as described for other bacterial genomes [ 35 ]. In contrast, C. lari and C. upsaliensis possess little if any synteny with C. jejuni RM1221. Figure 1 Whole-Genome Comparison of Five Campylobacter Strains Line figures depict the results of PROmer analysis. Colored lines denote percent identity of protein translations and are plotted according to the location in the reference ( C. jejuni RM1221, x-axis) and query genomes ( C. jejuni NCTC 11168 [upper y-axis] and C. coli RM2228 [lower y-axis]) (A). The Venn diagrams show the number of proteins shared (black) or unique (red) within a particular relationship for all five Campylobacter strains (B) and for members of the sequenced ɛ-Proteobacteria compared in this study (C). Protein sequences binned as “unique” are unique within the context of the genomes plotted and the cutoffs used to parse the BLASTP data. The pie charts plot the number of protein sequences by main functional role categories for C. jejuni RM1221 ORFs. A frequency distribution of protein percent identity (D) was computed: specifically, the number of protein sequences within class intervals of 5% amino acid identity from 35% to 100% that match C. jejuni RM1221 reference sequences were plotted. Table 1 Genome Features of Five Campylobacter Genomes a MLST and complex designations follow the PubMLST Web site ( http://pubmlst.org/ ) [ 101 ] b FlaA SVR ( http://phoenix.medawar.ox.ac.uk/flaA/ ) [ 34 , 102 ] c Estimate (number based on manual inspection of only a subset of genes) d From [ 28 ] e Disrupted ORF f Contingency gene present g Based on TIGR role category h See Table S10 for gene lists MOMP, major outer membrane protein Comparison of C. jejuni RM1221 protein sequences with those of other fully sequenced members of the ɛ-Proteobacteria revealed 540 shared protein sequences, many of which are proposed to have “house-keeping” functions ( Figure 1 C). Of the 1084 protein sequences shared by all the Campylobacter species in this study, 46 had no match to any other organism in the database ( p -value cutoff ≤ 10 −5 ) ( Figure 1 B). Eleven of these were assigned functions related to cell envelope biosynthesis, or fatty acid and phospholipids metabolism. Further analysis revealed 44 proteins considered C. jejuni -specific, of which 12 mostly hypothetical proteins were truly novel, having no match to other organisms in the database. Of the 300 C. jejuni RM1221-specific protein sequences, only 95 were not in phage or genomic island regions. To quantify relatedness among the sequenced ɛ-Proteobacteria, the average protein percent identity was computed for all proteins matching the reference strain C. jejuni RM1221 with a p -value less than or equal to 10 −5 , identity of 35% or more, and match lengths of at least 75% of the length of both query and subject sequence. Not surprisingly, C. jejuni NCTC 11168 had the highest average protein percent identity (1,468 proteins averaging 98.41% identity) with C. jejuni RM1221 proteins. C. coli RM2228 was second, with 1,399 proteins averaging 85.81% identity. Surprisingly, C. upsaliensis RM3195 had the third highest average protein percent identity with C. jejuni RM1221 (1,261 proteins; 74.72% average identity), followed by C. lari RM2100 with 1,251 proteins having 68.91% average identity. This was surprising since a 16S rRNA tree depicts C. upsaliensis to be more dissimilar to C. jejuni, C. coli, and C. lari [ 3 ]. Wollinela succinogenes DSMZ1740 was next, with 838 proteins averaging 53.77% identity, followed by Helicobacter hepaticus ATCC 51449 (770 proteins; 53.66% average identity), H. pylori 26695 (675 proteins; 52.39% average identity), and H. pylori J99 (682 proteins; 52.28% average identity). Phylogenetic Comparisons To resolve the apparent discrepancy regarding the relatedness of the ɛ-Proteobacteria between the results of average protein percent identities from this study and the previously published 16S rRNA tree based on percent sequence similarity [ 3 ], a consensus boot-strapped maximum-likelihood tree was generated based on trimmed alignments with gaps removed ( Figure 2 A). One of the advantages of generating whole-genome sequence is the magnitude of information available for resolving differences between closely related organisms. To better resolve the Campylobacter species, we took advantage of the wealth of sequence information to construct a maximum-likelihood concatenated protein tree using a set of 12 conserved protein sequences that have been previously shown to be reliable markers for phylogenetic analysis ( Figure 2 B) [ 36 , 37 ]. A frequency distribution of protein percent identity was plotted with 5% class intervals to visualize the similarities of these genomes at the protein level (see Figure 1 D). The 16S rRNA tree of sequenced members of the ɛ-Proteobacteria suggests that C. jejuni RM1221 is more closely related to C. coli RM2228 than to the other C. jejuni strain, NCTC 11168. However, the concatenated protein tree of these same organisms showed the two C. jejuni strains to be more closely related to each other than either is to C. coli RM2228, agreeing with the distributions of protein percent identities (see Figure 1 D). Both trees indicate that W. succinogenes is more closely related to Helicobacter than to Campylobacter . Most likely, the protein tree is more accurate and the rRNA tree is incorrect because the 16S rRNA does not have enough variation to resolve these close relationships [ 37 ]. Whole-genome sequencing of more members of the ɛ-Proteobacteria will enable a clearer resolution of the evolutionary relationships within this group of related organisms. Figure 2 Phylogenetic Analysis and Frequency Distribution of Protein Percent Identity Concensus maximum-likelihood trees are depicted using multiple alignments of 16S rRNA (A) or 12 concatenated protein datasets (B). The numbers along the branches denote percent occurrence of nodes among 100 bootstrap replicates. The scale bar represents the number of nucleotide (A) or amino acid (B) substitutions. Phages/Genomic Islands The major difference between the C. jejuni NCTC 11168 and C. jejuni RM1221 genomes is the presence within the strain RM1221 genome of four large integrated elements ( Figures 3 and S3 ). This characteristic has been observed in whole-genome intra-species comparisons of both Gram-positive and Gram-negative microorganisms [ 38 , 39 , 40 , 41 , 42 ]. The first element, Campylobacter Mu-like phage (CMLP1) (30.5% G+C content), located upstream of argC (CJE0275), encodes several proteins with similarity to bacteriophage Mu and other Mu-like prophage proteins [ 43 ], including putative MuA and MuB transposase homologs. Another feature consistent with the identification of CMLP1 as a novel Mu-like prophage is the presence of terminal 5′-TG-3′ dinucleotides flanked by a five-base direct repeat ( TATGC). Preliminary results suggest that this prophage is inducible with mitomycin C and that other C. jejuni strains harbor a related prophage (unpublished data). Genetic manipulation of this phage could yield useful molecular tools analogous to the Mu derivatives for the construction of random gene fusions or mini-Mu elements for in vivo cloning. Although this Mu-like prophage contains no characterized virulence determinants, it could potentially alter pathogenicity or other phenotypes via insertional inactivation. Figure 3 Linear Representations of Prophage Regions Regions are (from top to bottom): CMLP1, CJIE2, CJIE4, CLIE1, and CUIE1. Colors of ORFs are indicated in the key by putative phage function. Connecting lines represent those ORFs whose protein sequences match at a BLASTP of 30% identity or better. These lines do not indicate the coordinates of match, merely that there is a match. In contrast to CMLP1, C. jejuni RM1221 integrated elements 2 and 4 (CJIE2 and CJIE4) have integrated into the 3′ end of arginyl- and methionyl-tRNA genes, respectively. Several ORFs predicted to encode phage-related endonucleases, methylases, or repressors are present within these elements; however, unlike CMLP1, few ORFs encoding phage structural proteins were identified within CJIE4. CJIE4 is similar to a putative prophage contained within the C. lari RM2100 genome ( C. lari integrated element 1 [CLIE1]); 66% (35/53) of predicted proteins have BLASTP matches ( p -value ≤ 10 −5 ; identity ≥ 30%) ( Figure 3 ). CLIE1 is integrated into a leucinyl-tRNA. The inability to identify matches to major capsid, portal, and scaffold protease proteins within CJIE2 or C. upsaliensis RM3195 integrated element 1 (CUIE1) suggests that they represent either intact prophages with novel head morphogenesis proteins, satellite phages, or nonfunctional prophages or genomic islands. The absence of any phage-related ORFs within CJIE3 (located within an arginyl-tRNA), suggests that CJIE3 is not a prophage but rather a genomic island or integrated plasmid. Seventy-three percent (45/62) of the CJIE3 predicted proteins are similar to predicted proteins encoded on the C. coli RM2228 megaplasmid (pCC178) ( Figure S4 ; see below), suggesting that CJIE3 was plasmid-derived. However, the observed lack of synteny between CJIE3 and the C. coli RM2228 megaplasmid suggests that CJIE3 was not derived from pCC178 but possibly from a related Campylobacter megaplasmid. Although most of the ORFs contained within CJIE3 encode hypothetical proteins (23% 14/62), many are similar to proteins encoded within the 71-kb H. hepaticus ATCC 51449 genomic island (HHGI1), suggesting this genomic island could also be plasmid-derived [ 44 ]. Furthermore, 33% (23/70) of HHGI1 proteins match pCC178-encoded proteins. Bacteriophages are vehicles for the lateral or horizontal movement of genes that can increase bacterial fitness [ 45 , 46 ]. Additionally, it has been demonstrated that bacteriophage-carried genes can play a role in many aspects of bacterial virulence (adhesion, invasion, host evasion, and toxin production) [ 47 ]. Though only one of the Campylobacter prophages (CMLP1) has been shown to be inducible, we cannot predict whether the other putative prophages or plasmid-like element can be excised. Because the majority of ORFs that lie within prophage regions are hypothetical proteins, we are unable to deduce any putative functions from them; however, we cannot rule out possible functions that either directly impact virulence or increase the fitness of the host in a particular environment. Plasmids C. coli RM2228 and C. lari RM2100 each contain a single plasmid (pCC178; approximately 178 kb, and pCL46, approximately 46 kb, respectively), whereas C. upsaliensis RM3195 contains two plasmids (pCU3, approximately 3.1 kb, and pCU110, approximately 110 kb; Tables 1 and S1 ). In the current study, neither C. jejuni isolate harbors a plasmid; however, a C. jejuni virulence plasmid, pVir from C. jejuni strain 81–176, was previously sequenced and shown to play a role in pathogenesis [ 48 ]. The coding regions of pVir are entirely in one orientation except for a single coding region, which is uncharacteristic for a plasmid of this size. The coding regions of pCU110 and pCL46, like pVir, show a similar coding strand bias. In pCC178, the lack of coding region bias may be explained by the presence of antibiotic resistance genes ( Tables 2 and S2 ) flanked by putative mobile genetic elements. Only the 3.1-kb plasmid of C. upsaliensis RM3195 (pCU3) has a defined plasmid replication region. The single-stranded binding (Ssb) proteins are conserved among all of the plasmids, alluding to a common evolutionary origin; however, the nickase proteins on the plasmids are not conserved, suggesting that nickase may be specific to the plasmid or strain. Table 2 Relevant Drug Resistance Profiles a Resistance mechanism: CCOA0067/CCOA0068—aminoglycoside 3′-phosphotransferase from pCC178 b Sensitivity likely due to fragmentation of a class D β-lactamase c Resistance mechanism: 23S rRNA (A2122G), corresponding to position 2,143 of H. pylori sequence [ 84 ] d Resistance mechanism: T86V mutation in gyrA (CLA1521) e Resistance mechanism: CCOA0206—tetracycline resistance protein (tetO) from pCC178 I, intermediate resistance; R, resistant; S, susceptible One conserved feature of all of the large Campylobacter plasmids is the presence of a Type IV secretion system (T4SS), possibly involved in conjugative plasmid transfer or secretion of virulence factors [ 49 ] ( Figure S4 ). The plasmid-encoded T4SSs in the non– C. jejuni species are most similar to each other based on synteny and amino acid identity; however, they share only synteny with the T4SS encoded by pVir or the Agrobacterium tumefaciens Ti plasmid [ 50 ]. The non– C. jejuni plasmid T4SSs may be involved in conjugation rather than secretion of virulence factors because they are more similar to T4SSs known to mobilize DNA than to T4SSs that secrete effectors [ 50 ] ( Figure S4 ). Unlike pVir, the other Campylobacter plasmids encode proteins similar to VirB2 of the Ti plasmid, which is responsible for pilus formation [ 49 ] ( Figure S4 ) and has recently been shown to be essential for DNA transfer, further hinting at a role in DNA mobility [ 51 ]. Additionally, pCU110 appears to contain a number of other proteins that are similar to conjugal transfer proteins of other plasmids, which may function independently or in concert with the T4SS to transfer plasmid DNA to donor cells. Transposable Elements Both C. jejuni NCTC 11168 and C. jejuni RM1221 are notable for the apparent absence of intact insertion sequence (IS) elements. With the exception of one copy of a degenerate transposase resembling IS 605, located between the tonB gene and a 5S rRNA gene, their genomes are devoid of IS elements. In contrast, C. coli RM2228 contains five copies of an IS element (IS Cco1 of the IS 605 family) at three positions in the chromosome and at least two positions in the megaplasmid pCC178, hinting at recent acquisition and transposition competence. Both the C. upsaliensis RM3195 and C. lari RM2100 pseudomolecules lack the tonB –5S rRNA locus; however, since these are not closed genomes, we cannot accurately assess the status of the IS 605 family in these genomes. CRISPR Analysis The chromosomes of all five Campylobacter strains in this study were examined for the presence or absence of clustered regularly interspaced short palindromic repeats (CRISPRs) in intergenic regions. A strain was considered CRISPR-positive if it contained two or more direct repeats of a 21-bp or larger DNA segment separated by unique spacer sequences of a similar size. We identified CRISPR elements in only C. jejuni NCTC 11168 and C. jejuni RM1221. However, a previous study found that CRISPR elements are sometimes detectable in C. coli [ 52 ]. Also consistent with the previous study, the two strains of C. jejuni examined here can be differentiated by both the unique sequence of the spacer sequences ( Figure S5 ) and the number of CRISPR repeats in the element (five in C. jejuni NCTC 11168 and four in C. jejuni RM1221). It is noteworthy that the previous study did not include C. lari or C. upsaliensis, which appear not to contain CRISPR elements, unless they are in a different region of the genome from the C. jejuni CRISPRs and are in unsequenced areas. This further demonstrates the limited utility of CRISPRs in genotyping studies of Campylobacter species. Restriction–Modification Systems The Type I restriction–modification (RM) loci from 65 C. jejuni strains have been characterized previously [ 53 ]. In contrast to the C. jejuni, C. coli, and C. lari strains sequenced in this study, the C. upsaliensis RM3195 genome is predicted to contain at least three Type I RM loci ( Table S3 ). C. upsaliensis RM3195 also contains a putative fourth locus where the hsdR gene is absent. The sequenced genomes of the C. jejuni strains NCTC 11168 and RM1221, C. coli RM2228, and C. lari RM2100 encode few Type II or Type III RM systems. C. upsaliensis RM3195 encodes one putative Type II and two putative Type III restriction enzymes. In addition, C. upsaliensis RM3195 encodes 15 putative adenine- or cytosine-specific DNA methyltransferases. It is noteworthy that the sequenced genome of H. hepaticus ATCC 51449, like C. jejuni RM1221, C. coli RM2228, and C. lari RM2100, has a paucity of RM loci [ 44 ] and would therefore be considered “ Campylobacter -like” whereas C. upsaliensis RM3195 would be considered “ Helicobacter Pylori -like” with respect to RM systems. At least four of the C. upsaliensis RM3195 RM systems lie within regions of atypical nucleotide composition, suggesting recent horizontal transfer as selfish mobile elements [ 54 ]. Diversity within the Campylobacter RM systems has implications for Campylobacter biology, specifically DNA uptake and phage infection. Campylobacter spp. are naturally competent [ 55 ], and horizontal gene transfer via natural transformation is thought to play an important role in the evolution of C. jejuni [ 56 ]. Natural competence, as well as experimental introduction of DNA by electroporation, would be influenced presumably by host RM systems. Indeed, strain-specific differences in competence have been noted in Campylobacter [ 1 , 57 ]. RM system variation would also impact infection by both lytic and lysogenic bacteriophages. Future studies will be able to determine the functional status of the RM systems and their role in natural competence and phage restriction. Campylobacter Metabolism There have been relatively few studies of the metabolic capabilities of Campylobacter spp., but they are known to have a respiratory type of metabolism, with some strains growing under both aerobic and anaerobic conditions [ 58 , 59 ]. Carbohydrates in general are not utilized. Comparative analysis of the genomes of C. jejuni RM1221, C. coli RM2228, C. lari RM2100, and C. upsaliensis RM3195 revealed that these species have very similar metabolic profiles, with the main variation being the presence of a complete or partial tricarboxylic acid cycle ( Figure S6 ). In C. jejuni RM1221, the tricarboxylic acid cycle appears to be intact and most likely serves a dual role of generating biosynthetic compounds and providing intermediates that feed into electron transport. C. coli RM2228, C. upsaliensis RM3195, and C. lari RM2100 apparently lack a succinate dehydrogenase, and none of the strains appear to encode SucAB (oxoglutarate dehydrogenase). All four sequenced strains have pathways for the metabolism and biosynthesis of a number of amino acids ( Figure S6 ), and acetate, formate, and lactate appear to be the main end products of carbon metabolism. Preliminary Biolog data demonstrate differences in substrate utilization patterns across the Campylobacter strains in this study. C. jejuni RM1221, C. coli RM2228, and C. lari RM2100 all respire in the presence of arabinose, fucose, and formic and lactic acid. In addition, C. jejuni RM1221 respires in the presence of fructose, mannose, hydroxybutyric acid, asparagine, and aspartic acid, in contrast to the other species. These observed phenotypic differences from the preliminary Biolog data may be a reflection either of the conditions under which the substrates were tested or of C. jejuni having pathways that are lost in the other strains. Because of the lack of complete genomes from the other strains, we cannot say with confidence what the reason is for the observed differences, but variable patterns in substrate utilization by Campylobacter species have previously been described [ 60 ]. Some of these substrate utilization differences might stem from strain- and species-specific ORFs present in these isolates, or from simple gene mutations that cannot be detected at the genome level. In C. jejuni NCTC 11168, for example, the inability to grow on sugars that are added to the growth medium is felt to be a reflection of the missing phosphofructokinase that is necessary for glycolysis [ 28 ]. Interestingly, for all the ɛ-Proteobacteria included in this study, no phosphofructokinase could be identified except for W. succinogenes, enabling Wolinella to metabolize a wider range of carbohydrates than Campylobacter . Chromosomally Encoded Protein Secretion Systems The five Campylobacter strains analyzed in this study have the Sec-dependent and Sec-independent (twin-arginine translocation “TAT”) protein export pathways for the secretion of proteins across the inner/periplasmic membrane. In addition, Campylobacter has the signal recognition particle pathway. We have found no evidence for chromosomally encoded lol, Type III, or Type IV secretion systems other than the flagellar export apparatus [ 61 ]. In all five strains, there are putative proteins that comprise components of a transformation system with similarity to Type II secretion systems [ 62 ]. A putative pre-pilin peptidase and several putative pseudopilins have been identified based on BLASTP similarity or the presence of an N-terminal pre-pilin peptidase cleavage signal ( Table S4 ). The two-partner secretion/single accessory pathway [ 63 ] is used by Gram-negative bacteria to secrete adhesins and cytolysins [ 63 ]. There are undisrupted copies of putative pore-forming single accessory factors (generically termed TpsB homologs) in C. coli RM2228 (CCO0190), C. lari RM2100 (CLA0150), and C. jejuni NCTC 11168 (Cj0975); however, CCO1305 in C. coli and CJE0841–CJE0843 and CJE1056 in C. jejuni RM1221 are disrupted ( Figure S7 ). It is unclear whether these disruptions are real in the unfinished genomes or whether there would be any consequence for the disruption in C. jejuni RM1221. Virulence The pathogenic mechanisms responsible for acute intestinal infections by Campylobacter, although still poorly understood, are thought to involve adherence, cellular invasion, and toxin production, but not all clinical isolates of C. jejuni are able to invade cultured human cells or produce defined toxins [ 64 ]. However, a common feature of Campylobacter infectious enterocolitis is a localized acute inflammatory response that can lead to tissue damage and may be responsible for many of the clinical symptoms [ 64 ]. Motility is the major factor that has been implicated directly in intestinal colonization [ 65 ]. Of the 580 ORFs conserved between the Campylobacter and Helicobacter species included in this study (see Figure 1 C), 27 ORFs involved in flagellar biosynthesis and function were conserved between Campylobacter and Helicobacter . Another set of 18 ORFs involved in chemotaxis and motility was found to be conserved across the Campylobacter strains, but with no bidirectional match in Helicobacter (criteria: p -value ≤ 10 −5 , identity ≥ 35%, match lengths of at least 75% of the length of both query and subject sequence), emphasizing the importance of bacterial motility and adhesion for virulence [ 66 ]. Two-component regulatory (TCR) systems are used commonly by bacteria to respond to specific environmental signals. We identified five TCR systems (pairs of adjacent histidine kinase and response regulator genes) that appear to be conserved across the Campylobacter spp.: CJE0968–CJE0969, CJE1357–CJE1358, CJE1361–CJE1362, racR–racS (CJE1397– CJE1398), and CJE1664–CJE1665. In addition, another four putative response regulator genes (CJE0746, CJE0404, CJE1168, and CJE1780) and one putative histidine kinase gene (CJE0884) could be found in the finished C. jejuni genomes. Brás et al. [ 67 ] showed that the RacR–RacS system is involved in a temperature-dependent signaling pathway and is required for the organism to colonize the chicken intestinal tract. The high degree of conservation of these ORFs suggests an importance in the Campylobacter pathogenicity, not surprising given the likely exposure of the bacteria to temperature stress during the infectious process. Adherence of C. jejuni to epithelial cells is mediated by multiple adhesins, including CadF (CJE1651), PEB1 (CJE0997–CJE1000), JlpA (CJE1065), and a 43-kDa major outer membrane protein (CJE1395). Fibronectin (FN) has been implicated in C. jejuni adherence to epithelial cells via the protein CadF [ 68 ]. In addition to CadF, we found two putative FN-binding proteins (CJE1415 and CJE1538) that are conserved across the five Campylobacter strains. The FN host cell-surface receptor is the α5β1 integrin. In intact epithelia, α5β1 integrins are restricted to the basolateral membrane and thus are not available for interaction with luminally positioned microbial pathogens [ 69 ]. However, Monteville et al. showed that adherence and internalization of C. jejuni were greatly increased by exposure of cellular basolateral surfaces, and that FN was the receptor [ 70 ]. This suggests that C. jejuni invasion preferentially occurs via a paracellular route, rather than via an intracellular route. Additionally, inspection of loci adjacent to putative TpsB proteins revealed two intact filamentous hemagglutinin (FHA)–like adhesions: in C. lari RM2100, CLA0151, and in C. coli RM2228, CCO1312. The regions upstream of the remaining TpsB-like proteins have fragmented adhesion-like ORFs ( Table 1 ; Figure S7 ). Only C. lari RM2100 has both an undisrupted TpsB-like transporter (CLA0150) and an adjacent putative FHA-like adhesion (CLA0151), which, if functional, could enable C. lari RM2100 to attach to cell surfaces. Cytolethal distending toxins from enteropathogenic Escherichia coli have been shown to disrupt the barrier function of host intestinal epithelial tight junctions [ 71 ]. The three cytolethal distending toxins A, B, and C (CJE0075, CJE0074, and CJE0073) were conserved across the five Campylobacter strains. In addition, C. lari RM2100 encodes a single peptide (CLAA0034) in pCL46 that is similar to the Yersinia invasin proteins that enable Yersinia to penetrate host cells [ 72 ], suggesting that this C. lari strain might also have the ability to penetrate host cells. Identification of a Novel Campylobacter Putative Virulence Locus Examination of the C. upsaliensis RM3195 sequence revealed a putative licABCD (CUP0277–CUP0274) locus with varying, but significant, identity to genes present in Haemophilus influenzae [ 73 ], commensal Neisseria species [ 74 ], and Streptococcus pneumoniae [ 75 ]. licABCD genes in these microorganisms encode proteins involved in the acquisition of choline ( licB, CUP0276), synthesis of phosphorylcholine (PCho) ( licA, CUP0277; licC, CUP0275), and transfer of PCho ( licD, CUP0274) to LOS or teichoic/lipoteichoic acids to facilitate attachment to host cells [ 74 ]. Preliminary studies indicate that other strains of C. upsaliensis from South Africa also contain licA (unpublished data). It is noteworthy that licA expression in Haem. influenzae is regulated by variation in the number of intragenic tandem tetranucleotide repeats ( CAAT) at the 5′ end, resulting in translational on/off synthesis of PCho and expression on LOS [ 76 ]. A poly G tract within the licA gene (bp 132–146) of C. upsaliensis RM3195 probably regulates synthesis of PCho and decoration of LOS by a similar mechanism. Hypervariable Homopolymeric Tracks The presence of the homopolymeric repeat sequences in the genome of C. jejuni NCTC 11168 has been described [ 28 ]. However, in comparing these five Campylobacter strains, a number of other phenomena related to these repetitive regions were observed. First, when a homopolymeric repeat region was associated with a potential coding region, the base mostly included in the repeated region on the coding strand was G, resulting in poly-glycine, not poly-proline, in the peptide. Secondly, the C. upsaliensis RM3195 genome contains nearly three times as many variable homopolymeric repeats (22) as C. jejuni RM1221 (8), seven times as many as C. lari RM2100 (3), and 22 times as many as C. coli RM2228 (1) ( Table 1 ). These varied C. upsaliensis RM3195 poly G:C tracts come from a pool of almost five times as many total poly G:C tracts ( Table 1 ) as C. jejuni RM1221 and C. coli RM2228, and nearly ten times as many total poly G:C tracts as C. lari RM2100. Of these 22 varied poly G:C tracts, 11 (50%) are strain-specific (Tables S5 and S6 ). It appears that excess variable poly G:C tracts are due to the presence of unique ORFs; however, it is unclear as to why C. upsaliensis RM3195 contains so many more total homopolymeric repeated regions, since only 61 of the 209 regions are within unique ORFs. These variable regions encode a combination of hypothetical, cell envelope, and virulence-associated ORFs ( Table S6 ), which in other pathogenic bacteria has been shown to be the molecular basis of lipopolysaccharide phase variation [ 77 ], has been used to identify novel virulence genes in Haem. influenzae [ 78 ], and has been speculated to have a similar role in C. jejuni [ 28 ]. However, these observed differences could be the result of different culturing conditions prior to library construction. LOS and EP Biosynthesis LOSs and EPs are important surface structures in C. jejuni that function in the interactions of the organism with the environment. Interesting aspects of C. jejuni LOSs are their molecular mimicry of host gangliosides and their presumed roles in evasion of host immune responses and autoimmunity [ 79 ], decreased immunogenicity [ 80 ], and attachment and invasion [ 48 ]. The capsule of C. jejuni 81–176 has been reported to have a role in increasing serum resistance, invasion of cell lines, and surface hydrophilicity [ 81 ]. The LOS biosynthesis loci of all sequenced Campylobacter spp. are organized as previously observed in other C. jejuni strains [ 82 ]. At either end of the loci are the heptosyltransferase genes, waaC and waaF, that surround regions exhibiting significant variation in ORF content. Thus, these organisms likely synthesize novel LOS structures [ 82 ]. In particular, the LOS of C. jejuni RM1221 is distinct from the LOS of NCTC 11168, as seen on polyacrylamide gels, in that it possesses three LOS bands while NCTC 11168 possesses only one (unpublished data). Two LOS genes from C. jejuni RM1221 possess homopolymeric G:C tracts that may explain the additional bands. Comparison of the LOS genes from the sequenced Campylobacter spp. with those from C. jejuni strains that produce ganglioside mimics [ 29 ] demonstrates that these four strains do not possess the genes involved in the synthesis of N -acetylneuramic (sialic) acid or the associated sialic acid transferase, and are not likely to produce ganglioside mimics. Within the LOS loci of C. lari RM2100 and C. upsaliensis RM3195, there are ORF clusters that have homologs in NCTC 11168 that are unrelated to LOS biosynthesis. It is unclear what role this genomic reorganization plays in the biosynthesis of LOS. C. jejuni RM1221, C. coli RM2228, and C. lari RM2100 possess kps orthologs like the EP locus of C. jejuni NCTC 11168 that are involved in polysaccharide export; however, many putative EP biosynthesis genes from C. jejuni RM1221 and C. coli RM2228 are unique to these strains. The kps orthologs are present in C. upsaliensis RM3195, but they are not clustered with other polysaccharide biosynthetic genes as observed in the other strains. Specifically, there are three clusters of EP genes: CUP0615–CUP0619, CUP1248–CUP1270, and CUP1328–CUP1329. The second cluster contains many ORFs that are unique to C. upsaliensis ( Table S5 ), including two of the three copies of a putative GDP-fucose synthetase (CUP1255, CUP1257, and CUP1258). Only C. jejuni strains (Cj1428c and CJE1612) and C. upsaliensis RM3195 encode this enzyme. Of these GDP-fucose synthetases, only CUP1257 was shown to contain variable poly G tracts ( Table S6 ). Antibiotic Resistance The sequenced Campylobacter strains have adapted or acquired many mechanisms of antibiotic resistance ( Tables 2 and S2 ). All strains are resistant to cloxacillin, nafcillin, oxacillin, sulfamethoxazole/Tm, trimethoprim, and vancomycin, and this resistance is likely inherent to all Campylobacter spp. ( Table S2 ). Every strain but C. upsaliensis RM3195 is resistant to most β-lactam antibiotics. This general lack of resistance to β-lactam antibiotics for RM3195 is likely due to the disruption of a class D β-lactamase matching GenBank accession AAT01092 (CUP0345), which was found as an intact single copy in NCTC 11168 (Cj0299), RM1221 (CJE0344), and RM2100 (CLA0304). The corresponding sequence in C. coli RM2228 may reside in unsequenced regions. Only C. lari RM2100 was resistant to a broad range of quinolone/fluoroquinolone antibiotics ( Table 2 ). This broad quinolone/fluoroquinolone resistance is most likely the result of adaptation via a mutation of DNA gyrase (gyrA) that changed codon 86 from threonine to valine [ 83 ]. The macrolide antibiotics azithromycin, clindamycin, erythromycin, and tilmicosin were effective against all but C. coli RM2228. This is likely due to a mutation in all three copies of the 23S rRNA (A2122G), corresponding to position 2,143 of the H. pylori sequence [ 84 ]. C. coli RM2228 has acquired resistance to the aminoglycosides kanamycin and neomycin, tetracycline, oxytetracycline, minocycline, and presumably hygromycin B (but not gentamicin) from the megaplasmid pCC178 ( Table 2 ). It is possible that C. coli has acquired resistance to macrolides and tetracyclines as a result of the application of these drugs during poultry production. The resistance of C. upsaliensis RM3195 to oxytetracycline and its intermediate resistance to tetracycline may be due to the action of multi-drug efflux pumps or a novel mechanism, since there is no evidence for tetracycline resistance genes [ 85 ], and there are no known mutations in the 16S rRNA [ 86 ]. Similarly, no known mutations in gyrA or gyrB were found in C. upsaliensis RM3195 to explain the resistance to nalidixic acid [ 83 ] and novobiocin [ 87 ]. There were no obvious known mutations of dihydropteroate synthase (folP) [ 88 ] to explain the observed variable resistance to sulfonamide-class drugs ( Table 2 ). Rifampin resistance was observed in all strains but C. lari RM2100, but was not due to the classic mutations in the β subunit of RNA polymerase [ 89 ]. Conclusions The comparison of five sequenced Campylobacter genomes has provided the core genetic blueprint of the genus. Although the blueprint reveals obvious differences in genome structure and content, additional epidemiological data are needed to correlate these differences, and other, more elusive differences (e.g. differences in regulation and point mutations), with differences in virulence. Some obvious differences were the presence of drug resistance genes that may have been the result of adaptation in the animal production environment, where antibiotics are frequently used to eliminate bacterial infections. It is anticipated that the analysis of the Campylobacter genomes presented here will lay the foundation for the development of systems for fingerprinting strains for phylogenetics, epidemiology, and source tracking, as well as the development of alternative treatments for controlling Campylobacter in food production and in human infection. Materials and Methods Strain isolation and propagation C. jejuni strain RM1221 ( ATCC BAA-1062) was isolated from the skin of a retail chicken using methods modified from those described previously for isolation of Campylobacter from chicken products [ 31 ]. C. coli strain RM2228 ( ATCC BAA-1061) was isolated from a chicken carcass obtained from an inspected slaughter plant. A rinse sample was streaked on 5% sheep blood agar plates, and the plates were incubated at 37 °C for 48 h under an atmosphere of 5% O 2 , 10% CO 2 , and balance N 2 . An isolated single colony was picked and maintained on sheep blood agar plates. Three rounds of mixing and sonication of single colony picks were done as described [ 31 ]. C. lari strain RM2100 ( ATCC BAA-1060) is a human isolate obtained from the Centers for Disease Control and Prevention, Atlanta, Georgia, United States (CDC strain D67, “case 6” [ 32 ]). The strain was maintained on Brucella agar amended with 5% (v/v) laked horse blood (Hema Resource and Supply, Aurora, Oregon, United States). Three rounds of mixing and sonication of single colony picks were done as described [ 31 ]. C. upsaliensis strain RM3195 ( ATCC BAA-1059) was obtained from the feces of a 4-y-old boy confirmed clinically to have Guillain-Barré syndrome. The isolation procedure involved a filtration method with selection of Campylobacter cells in diluted feces by their migration through a 0.6-μm membrane filter and subsequent growth on nonselective medium [ 33 ]. Genome sequencing The four species of Campylobacter were sequenced by the random shotgun method [ 38 ]. The genome of C. jejuni RM1221 was sequenced to closure, whereas the genomes of strains C. lari RM2100, C. coli RM2228, and C. upsaliensis RM3195 were sequenced to 8-fold coverage of an estimated 1.8-Mbp genome. Briefly, one small insert plasmid library (1.5–2.5 kb) and one medium insert plasmid library (10–12 kb) were constructed for each strain (except RM1221, which had only a small insert library) by random nebulization and cloning of genomic DNA. In the random sequencing phase, 8-fold sequence coverage was achieved from the two libraries (sequenced to 5-fold and 3-fold coverage, respectively). The sequences from the respective strains were assembled separately using TIGR Assembler [ 90 ] or Celera Assembler [ 91 ]. All sequence and physical gaps for C. jejuni RM1221 were closed by editing the ends of sequence traces, primer walking or transposon-primed sequencing [ 92 ] on plasmid clones, and combinatorial PCR followed by sequencing of the PCR product. The correct nucleotide sequences for repetitive regions greater than the maximum insert size of 2.5 kb (i.e., rRNA operons) for C. jejuni RM1221 were confirmed by sequencing PCR products that spanned each repeat unit. Pseudomolecules for the draft sequences were constructed using NUCmer [ 93 ] and BAMBUS [ 38 , 94 ] as previously described [ 38 ]. Ambiguity rate The ambiguity rate for the unfinished genomes was determined using the following procedure. First, the consensus of the contigs was recalled using the consensus caller included in the AutoEditor package ( http://www.tigr.org/software/autoeditor/ ) [ 95 ] by executing “autoEditor—noedit” on the final contigs. This step was necessary because the contigs as produced by the Celera Assembler were made with a consensus caller which does not assign ambiguity codes, but instead assigns a base call arbitrarily in the event of a tie or near tie situation. The AutoEditor consensus caller recomputes the consensus at each position and assigns an ambiguity code if there is sufficient conflicting information. Using a custom script, a count was made of both the overall number of positions and the number of ambiguous positions with at least the specified depth of coverage. This was necessary because the depth of coverage in the assemblies is not uniform, but directly influences the ambiguity rate. For example, under the AutoEditor ambiguity model, there are no ambiguous positions at 1-fold coverage. The ambiguity rate is then reported as the ratio of the two counts, as a close approximation to the error rate of the true consensus sequence. Annotation An initial set of ORFs that likely encode proteins was identified using GLIMMER [ 96 ], and those shorter than 90 bp or those with overlaps were eliminated. ORFs were searched against a nonredundant protein database; frameshifts and point mutations were processed only for C. jejuni RM1221 [ 38 ]. Two sets of hidden Markov models were used to determine ORF membership in families and superfamilies [ 38 ]. Comparative genomics For the identification of species-specific ( Table S7 ) and strain-specific ( Table S5 ) ORFs, all predicted proteins (excluding pseudogenes) from the four TIGR-sequenced Campylobacter genomes and C. jejuni NCTC 11168 [ 28 ] were searched against an in-house database composed of 734,467 protein sequences encoded by 19 archaeal, 192 bacterial, 146 eukaryotic, three phage, and 17 virus chromosomes, as well as 145 plasmid, 29 mitochondrial, 17 plastid, and three nucleomorph genomes, using WU-BLASTP ( http://blast.wustl.edu ) [ 97 ]. To identify genus-specific ORFs, the protein sequences from the above five Campylobacter genomes plus three Helicobacter genomes ( H. pylori 26695 [ 98 ], H. pylori J99 [ 99 ], and H. hepaticus ATCC 51449 [ 44 ]) and the genome of W. succinogenes DSMZ1740 [ 100 ] were compared. Specifically, only bidirectional best matches that met the following prerequisites were scored: a p -value less than or equal to 10 −5 , identity of 35% or more, and match lengths of at least 75% of the length of both query and subject sequence. Match tables were created that were later used to generate the Venn diagrams (Tables S8 and S9 ). Novel ORFs encoded proteins that had no WU-BLASTP match. Regions of synteny were identified by first finding the maximum unique matches with a minimum length of five amino acids using PROmer, followed by visualization of the data using MUMmerplot ( http://www.tigr.org ) and Gnuplot version 4.0 ( http://www.gnuplot.info/ ). MLST and FlaA SVR typing The MLST of C. jejuni RM1221 was determined by searching the nucleotide sequences of aspartate ammonia-lyase ( aspA, CJE0082), glutamine synthetase type I ( glnA, CJE0798), citrate synthase ( gltA, CJE1851), serine hydroxymethyltransferase ( glyA, CJE0451), phosphoglucosamine mutase ( pgm/glmM, CJE0409), transketolase ( tkt, CJE1817), and ATP synthase F1 alpha subunit ( uncA/atpA, CJE0100) on the PubMLST Web site ( http://pubmlst.org/ ) [ 101 ]. The sequence of the C. jejuni RM1221 FlaA SVR was found by searching the flaA (CJE1528) nucleotide sequence using the sequence of primers FLA242FU and FLA625RU [ 34 ]. This nucleotide sequence was used to query the flaA allele database ( http://phoenix.medawar.ox.ac.uk/flaA/ ) to elucidate the FlaA SVR type [ 34 , 102 ]. Phylogenetic analysis The programs SEQBOOT, DNAML, PROML, and CONSENSE are part of the PHYLIP version 3.62 package ( http://evolution.genetics.washington.edu/phylip.html , http://fink.sourceforge.net/ ) [ 103 ]. Both the 16S rRNA and concatenated protein trees were rooted to the δ-Proteobacterium Desulfovibrio vulgaris subsp. vulgaris strain Hildenborough sequences [ 104 ]. One hundred bootstrapped datasets were generated using the SEQBOOT program, and consensus trees were determined using CONSENSE. The final trees with preserved branch lengths were computed with the user tree option of DNAML and PROML. 16S rRNA trees were generated by first creating a multiple alignment using the “PHYLIP Interface” option of the Ribosomal Database Project release 8.1 ( http://35.8.164.52/cgis/phylip.cgi , which aligns user-supplied 16S rRNA sequences against the Ribosomal Database Project alignment. The produced alignment was trimmed and gaps removed using an in-house PERL ( http://www.perl.org ) script. Maximum-likelihood trees were generated using DNAML (R = gamma-distributed rate of variation [coefficient of variation, 1.41; four hidden Markov model rate categories] and S = NO). Protein trees were generated from concatenated multiple alignments of 12 conserved proteins (initiation factor 2 [InfB]; elongation factors G [FusA] and Tu [Tuf]; ribosomal proteins L2 [RplB], S5 [RpsE], S8 [RpsH], and S11 [RpsK]; DNA topoisomerase I [TopA]; signal recognition particle protein [Ffh] [ 36 ]; DNA gyrase B subunit [GyrB]; GTP-binding protein LepA; and CTP synthase [PyrG] [ 37 ]). Each protein was aligned separately using CLUSTALW version 1.82 [ 105 ], using the slow, more accurate option. The alignments were trimmed to remove gaps using BELVU version 2.16 ( http://www.cgb.ki.se/cgb/groups/sonnhammer/Belvu.html ). Each organism's aligned sequences were concatenated using an in-house PERL script. Maximum-likelihood trees were generated using PROML (P = Jones-Taylor-Thornton model of change between amino acids, R = gamma-distributed rate of variation [coefficient of variation, 1.41; four hidden Markov model rate categories], and S = NO). Hypervariable homopolymeric G or C tracts Hypervariable homopolymeric G or C tracts were identified by analyzing the underlying sequences for each nucleotide within a tract of six or more G or C nucleotides. A hypervariable tract was considered of high quality if its underlying sequence comprised at least three sequencing reads with an average Phred score greater than 30 [ 106 ]. Supporting Information Figure S1 Circular Representation of the Closed C. jejuni RM1221 Genome Each concentric circle represents genomic data and is numbered from the outermost to the innermost circle. Refer to the key for details on color representations. The first and second circles represent predicted ORFs on the plus and minus strands, respectively. The third circle shows the GC-skew. The fourth circle depicts genetic loci with characteristics or functions of interest: CRISPRs, DNA competence, EP, LOS, prophage and genomic island regions, motility, repeats, and Type I restriction/modification regions. The fifth circle demarcates C. jejuni– specific and C. jejuni RM1221–specific ORFs. The sixth circle plots atypical regions (χ 2 value). The seventh circle denotes tRNA, rRNA, and sRNA (tmRNA and 4.5S RNA) loci. (2.6 MB EPS). Click here for additional data file. Figure S2 Linear Illustration of C. jejuni Genome Comparisons (274 KB PDF). Click here for additional data file. Figure S3 Comparison of Plasmid-Like Genomic Islands of C. jejuni RM1221 CJIE3 (top linear figure) and H. hepaticus ATCC 51449 HHGI1 (bottom line) against pCC178 megaplasmid of C. coli RM2228 (middle line). Colors of ORFs are indicated in the key by putative function. Connecting lines represent those ORFs whose protein sequences match at a BLASTP of 30% identity or better. These lines do not indicate the coordinates of match, merely that there is a match. (76 KB PDF). Click here for additional data file. Figure S4 T4SS Is Shared among the Large Campylobacter Species Plasmids but Is Not the Same as C. jejuni T4SS (A) shows a conceptual diagram indicating where each of the proteins thought to be involved in the T4SS interact. Each corresponding loci is color-coded in each of the plasmids. (B) The T4SS in each of the plasmids demonstrates that a number of the core proteins are conserved in all of the Campylobacter plasmids; however, the non– C. jejuni plasmids contain a structure that is more similar to the Agrobacterium tumefaciens T4SS. (In the Campylobacter plasmids, black ORFs are those not directly involved in the T4SS; however, many are similar to plasmid transfer proteins). (5.3 MB EPS). Click here for additional data file. Figure S5 DNA Sequences of the CRISPR Elements Found in the Two Strains of C. jejuni, RM1221 and NCTC 11168 The characters in italics indicate the 32-bp spacer sequences that are unique to the two strains; the spacer sequences for NCTC 11168 are 1 bp longer than presented by others [ 52 ]. The bold characters represent the CRISPR repeat region in RM1221 ( n = 4) and NCTC 11168 ( n = 5). The characters in roman typeface indicate regions flanking the repeat region that are identical in the two strains. (20 KB DOC). Click here for additional data file. Figure S6 Main Pathways for Metabolism Derived from an Analysis of Five Campylobacter Genomes The tricarboxylic (TCA) cycle has major variations based on comparative analysis across the strains (please refer to text). Differences in substrate respiration based on an analysis of Biolog data and species-specific pathways are also presented in the text. (51 KB PPT). Click here for additional data file. Figure S7 Putative Two-Partner/Single Accessory Secretion Loci FhaC, the single accessory protein that secretes the Bordetella pertussis FHA across the outer membrane, was used as the query for BLASTP searches against a database containing Campylobacter protein sequences. Fragments of single accessory proteins were found as matches in the Campylobacter match table (see Table S8 ). Putative single accessory protein/TpsB family proteins (teal) and putative FHAs/hemolysins (red) are noted, as well as putative proteins with weak matches to metacaspases or toxins (tan). The small red ORFs suggest fragmentation of a larger, full-length ORF. (1.6 MB EPS). Click here for additional data file. Table S1 Comparison of Campylobacter Species Plasmids (19 KB XLS). Click here for additional data file. Table S2 Antibiotic Susceptibility Profiles (22 KB XLS). Click here for additional data file. Table S3 C. jejuni, C. coli, C. lari, and C. upsaliensis Restriction-Modification (22 KB XLS). Click here for additional data file. Table S4 Putative DNA Competence Genes (16 KB XLS). Click here for additional data file. Table S5 Strain-Specific Genes with Annotations (238 KB XLS). Click here for additional data file. Table S6 Hypervariable Homopolymeric Sequences Found in Campylobacter Genomes (57 KB XLS). Click here for additional data file. Table S7 C. jejuni –Specific Genes with Annotations (31 KB XLS). Click here for additional data file. Table S8 Match Table Depicting Bidirectional Best Matches of Campylobacter Species (647 KB XLS). Click here for additional data file. Table S9 Match Table Depicting Bidirectional Best Matches of Sequenced ɛ-Proteobacteria (894 KB XLS). Click here for additional data file. Table S10 Arg-Gly-Asp, Lipoprotein, Outer Membrane Protein Signal, Secretion Signal, and Transmembrane Motif Results (155 KB XLS). Click here for additional data file. Accession Numbers The nucleotide sequence for the closed genome of C. jejuni RM1221 has been deposited at the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp/ , the European Molecular Biology Laboratory Nucleotide Sequence Database (EMBL; http://www.ebi.ac.uk/embl/ , and GenBank ( http://www.ncbi.nlm.nih.gov/Genbank/ ) under accession number CP000025. The whole-genome shotgun projects for the genomes of C. lari RM2100, C. coli RM2228, and C. upsaliensis RM3195 that were sequenced to at least 8-fold coverage were deposited at DDBJ, EMBL, and GenBank under accession numbers AAFK00000000, AAFL00000000 and AAFJ00000000, respectively. The versions described in this paper are the first versions, AAFK01000000, AAFL01000000 and AAFJ01000000, respectively. Additionally, all sequence traces and assemblies were deposited at the National Center for Biotechnology Information assembly archive ( http://www.ncbi.nlm.nih.gov/Traces/assembly ). The contig separator that was used to create the pseudomolecules for the unfinished genomes is NNNNN TTAATTAATTAANNNNN.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539331.xml

549603

Equitable Allocation of Antiretrovirals in Resource-Constrained Countries

Antiretroviral drugs change the lives of patients with HIV/AIDS—if they have access to them. Most patients in resource-poor countries cannot afford the drugs. Major initiatives are under way to expand access to antiretrovirals in developing countries, but the number of individuals in need of the drugs currently vastly exceeds the supply, and will continue to do so for the foreseeable future. These circumstances make for difficult decisions about treatment allocation. David Wilson and Sally Blower have shown how it is possible to design an equitable antiretroviral allocation strategy, that is, to come up with a plan that would give each individual with HIV an equal chance of receiving antiretrovirals. Their novel spatial model enables them to model the “spatial diffusion” of antiretrovirals in a resource-constrained country. Modeling allocation of antiretrovirals in KwaZulu–Natal Based on the premise that only a limited number of drugs will be available and only a limited number of health-care facilities can be used for drug distribution (each of them serving the population in a specific area), they determine an optimal equitable allocation strategy. They then apply this approach to a practical example—the equitable allocation of antiretrovirals to patients with HIV/AIDS in the South African province of KwaZulu–Natal. Using data from a detailed rollout plan for antiretrovirals designed by the South African government, they come up with an allocation strategy that differs substantially from the current governmental plan for the province. KwaZulu–Natal has a total of 54 health-care facilities, of which 17 are assigned to allocate antiretrovirals under the current plan. It is the largest province in South Africa, with a population of about 9.4 million, and it has more people with HIV than any other province (about 21% of all cases in South Africa). Wilson and Blower assume that the available amount of antiretrovirals can treat 10% of the individuals with HIV in KwaZulu–Natal. Modeling the 17 health-care facilities and the 51 communities of individuals with HIV, they determine the amount of drugs to allocate to each facility to achieve equitable access by patients throughout the province. They then extend the analysis assuming that additional health-care facilities could be made available to distribute drugs. They conclude that in order to achieve the greatest degree of treatment equality, all 54 health-care facilities should be used, and they should, on average, each serve the population within a radius of 50 km. Wilson and Blower discuss how their model can be adjusted and therefore used by policy makers in resource-constrained countries to determine a scientifically based allocation strategy for limited resources based on a number of specific objectives. They also recognize that there are other considerations that influence ethical treatment allocation besides equity, for example, the desire to maximize epidemic reduction, or the imperative to give priority to the least advantaged individuals They believe that their model can be adjusted and therefore “used by policy makers to determine an optimal scientifically based allocation strategy” for a number of specific objectives. Another possibility would be to apply the equity strategy to allocate drugs to particular health-care facilities (thereby achieving equality in accessibility), and then take additional ethical considerations into account at the community level.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549603.xml

545777

Spatial patterns of transcriptional activity in the chromosome of Escherichia coli

Analysis of the transcriptional activity in Escherichia coli K12 revealed an asymmetry in the distribution of transcriptional patterns along the bacterial chromosome and showed that spatial patterns of transcription could be modulated pharmacologically and genetically.

Background Chromosomes have evolved to effectively retrieve and transmit genetic information stored in DNA. Significant progress has been made recently in our understanding of how DNA is packaged into chromosomes, particularly at low compaction levels determined by supercoiling and/or protein-dependent condensation [ 1 ]. Available structural information about bacterial chromosomes indicates that the chromosome is supercoiled in vivo [ 2 ] and organized in topologically constrained domains [ 3 ]; diffusion of supercoils over the chromosome is impeded in actively replicating cells [ 4 ]; the chromosome is mildly condensed in vivo [ 5 , 6 ]; chromosomal loci inside the cell are specifically organized and arrayed in linear order according to the linear genetic map [ 7 , 8 ]; and at least two chromosomal loci are actively moved and positioned inside the cell [ 7 , 9 ]. Although the molecular bases of these structural features are not known, the bacterial chromosome can be viewed as a cellular organelle, whose dynamics may be coupled to the state of the cell. In turn, the state of the cell is reflected in the whole-genome transcriptional activity [ 10 ]. Therefore, genome-wide transcription can be used to probe chromosomal organization. Global transcriptional profiles have been successfully used to probe the organization of transcriptional units into operons [ 11 ] and regulons [ 12 , 13 ]. However, such analysis is limited by assumptions about the nature of transcriptional units. Covariation in transcriptional activity along the chromosome determine spatial transcriptional patterns [ 14 , 15 ]. Such co-variation might result from differing DNA accessibility along the chromosome [ 16 , 17 ]. The variation in accessibility, in turn, may be determined by chromosomal structural features. By analogy, chromosomal regions that do not reveal any spatial covariation could represent unstructured portions of the chromosome. Using signal processing and statistical techniques, we systematically examined transcriptional activity of genes as a function of their position in the bacterial chromosome. Here we report the discovery of stable, short- and long-range patterns in genome-wide transcription in Escherichia coli K12. Moreover, we demonstrate that such patterns are affected by genetic and environmental factors, thereby offering the first biologically relevant insights into the nature of the spatial organization of transcription in bacteria. Results Local structure in the spatial series of transcriptional activity We modeled transcriptional activity of the chromosome as a one-dimensional spatial series of transcript abundances. Transcript abundances were measured in cells grown in batch cultures to OD 600 = 0.5 in LB or M9 medium supplemented with 0.2% glucose (cultures reached stationary phase at OD 600 = 3.5 in LB medium and at 2.5 in M9 medium). Samples of total RNA extracted using hot-phenol method [ 18 ] were labeled with Cy-fluorophors and hybridized against genomic DNA as a reference. We carried out two types of hybridization: one using genomic DNA from the same cells from which we extracted the RNA and the other using genomic DNA from cultures with arrested initiation of DNA replication that completed ongoing rounds of replication. Two different types of genomic reference produced indistinguishable results in spectral analysis, and for the sake of simplicity we present here the analysis of results obtained in hybridizations against genomic DNA isolated from non-replicating cultures. The mRNA abundances have been recorded for almost every gene in the chromosome by two-color hybridization on whole-genome DNA microarrays. The data are publicly available at the Gene Expression Omnibus [ 19 ], accession numbers GSE1730 and GSE1735. To determine the degree of similarity in transcriptional activity of individual genes as a function of their position on the chromosome we calculated the autocorrelation function (ACF) as a function of the distance between genes, where the distance is measured as the number of intervening genes (Figure 1 ). Independent of the growth conditions, the ACF could be characterized as a decaying function whose largest statistically significant portion assumes positive values and corresponds to relatively short gene-to-gene distances (fewer than 100 genes; Figure 1a , data not shown). By definition [ 20 ], the portion of the ACF that constitutes significant correlations may reflect the existing stable structure in the series. Thus, it is likely that the transcription of any two genes, separated on the chromosome by a distance less than or equal to the length of the structured portion of the series, is similarly affected on average across the entire chromosome. Scrambling the linear order of genes leads to the loss of structure in the series (Figure 1b ), indicating that the relative positioning of genes determines the autocorrelation properties. If the autocorrelation reflects dominant patterns of co-transcription, then by determining the properties of the ACF we should be able to describe these patterns. Any one-dimensional pattern could be characterized by the stability (the distance between two genes at which transcriptional similarity drops to 50% of the average correlation between immediate neighbors) and by the range of stable correlations (maximal length of an average stable co-transcribed chromosomal region). Following the two-parameter exponential fitting of at least eight independently measured ACFs, we determined that the transcriptional correlations decay by half in the rich medium over 7.5 kilobases (kb) (Table 1 , 0 minute entry). The range of local transcriptional patterns was measured as the maximal length of a continuous, significantly correlated region of the chromosome that is not affected by experimental error. We determined that in rich medium up to 16 genes in a row could demonstrate apparently coherent transcriptional activity (Figure 1c ). The short-range correlations of the transcriptional series obtained in minimal medium had comparable characteristics (data not shown). While it is expected that genes organized in operons would show significant autocorrelation, the stability as well as the significant range of autocorrelations observed here extend far beyond those expected if gene expression was only coordinated within operons, which have an average size of three genes [ 21 ]. Analysis of long-range correlations In addition to continuous transcriptional correlations over short distances, we also observed individual, statistically significant spikes in correlations of transcriptional activities of genes located about 100 and 700 kilobases (kb) apart. To investigate such long-range transcriptional patterns in more detail we decomposed the original signal of transcript abundances into a series of harmonics. The frequency spectrum of transcript abundances is shown in Figure 2 . Four frequencies appear to be significant (at 95% confidence level) in the spatial series obtained from the cultures grown in LB medium: 690 -1 kb -1 ; 129 -1 kb -1 , 115 -1 kb -1 and 103 -1 kb -1 . The sequential signal recorded in the mid-exponential cells grown in M9 salts supplemented with glucose contained similar significant frequencies (Figure 2b ): 690 -1 kb -1 , a clump of frequencies around 115 -1 kb -1 and a free-standing frequency of 414 -1 kb -1 . The Fourier transform provides signal average characteristics and does not determine the frequencies at a particular spatial locality. To localize significant frequencies determined by Fourier transform and to find potentially significant local spikes of transcriptional activity, we subjected the spatial series of transcript abundances to a wavelet transform. The wavelet analysis revealed significant spectral components at the scales very similar to significant periods in the Fourier spectrum: approximately 125 kb and 600-700 kb (Figure 3a ). In addition, in the range of scales from 100 to 1,000 kb, the wavelet transform identified pronounced local patterns at frequencies corresponding to 235 -1 kb -1 , 300 -1 kb -1 , 365 -1 kb -1 and 555 -1 kb -1 . The wavelet spectrum also shows that patterns of transcriptional activity are not symmetrical with respect to the chromosome. The dominant patterns are localized largely in the left replichore (the half chromosome divided by the replication axis counter-clockwise from the oriC ) and appear to be bounded by the origin of replication. The most dominant pattern wave in transcriptional series, represented by a period of about 600 kb, spreads for 2.3 megabases (Mb), from the origin of replication to the terC site. The second most pronounced pattern (around 125 kb) consistent with the Fourier spectrum results from the transcriptional activity between the origin of replication and terG , about 1.3 Mb away. While significant components of the wavelet spectrum consistent with the Fourier spectrum were largely distributed on the left replichore, the scales unique to the wavelet spectrum were narrowly distributed along the scattered parts of the right replichore. Modulation of transcriptional spectra Inhibition of transcription initiation by rifampicin completely eliminates significant frequencies from the spectra after 30 min of treatment (data not shown). We rationalized that if transcription is not only inhibited, but modulated, globally, we might be able to track changes in the transcriptional spectra. We used a topoisomerase inhibitor, norfloxacin, to try to modulate transcription in the cell by inhibiting topoisomerase activities. As supercoiling and transcription in the cell are expected to be tightly coupled [ 22 ], we anticipated that inhibition of DNA topoisomerization would affect spatial transcriptional patterns. Norfloxacin was used at a concentration that ensured 50% and 90% killing of a bacterial population after 10 minutes and 30 minutes treatment, respectively. We examined the local correlations in transcription following norfloxacin treatment and found that by 30 minutes the range of significant autocorrelations has been reduced by about 50% from 16.5 to 7.7 kb and the stability of local patterns was reduced from 7.5 to 4.3 kb (Table 1 ). We have observed changes in the amplitude (the magnitude of squared amplitudes integrated over space - the power - is plotted along the vertical axis in Figure 3 ) as well as in spatial ranges of significant wavelet components (Figure 3b,c ). The most obvious changes in the wavelet spectrum can be described as a recession of the largest-scale wavelets from the terminus of replication in the left replichore. At a higher resolution, as seen in the local wavelet power spectrum in Figure 4 , the spatial range of the characteristic 100-125 kb wavelet narrowed by 25 to 90%. Similarly, the Fourier analysis revealed that the main periods in untreated cells, including 115 kb and 690 kb, were significantly ( p < 0.05) diminished by (in some case) 30 minutes with the drug (Figure 5 ). Global modulation of transcription by a single point mutation in DNA gyrase In addition to modulation of the transcriptional spectra by non-equilibrium perturbations, we were interested in steady-state alterations in the process of transcription that would not be associated with irreversible changes in bacterial physiology. Such alterations may result from a compensation to a partial reduction in gyrase activity. A transient increase in transcription of the gyrase genes has been observed following inhibition of gyrase activity [ 23 ]. We rationalized that a partial loss of function of the gyrase enzyme would be accompanied by a compensatory steady-state increase in transcription of the gyrase genes. We predicted these compensatory changes in transcription would be genome-wide rather than confined to the gyrase genes. Using resistance to low concentrations of norfloxacin as a screening method, we selected mutants with increased levels of gyrA and/or gyrB transcripts. We characterized one spontaneous mutant of gyrase, resistant to 0.8 μg/ml of norfloxacin, which carries the D82G mutation in the gyrA allele and causes elevated levels of gyrA and gyrB mRNA in vivo and lowers supercoiling activity of DNA gyrase in vitro (K.S.J, Hiroshi Hiasa and A.B.K, unpublished work). This mutant had a normal rate of growth and cell density in stationary phase (less than 5% different from the isogenic wild-type strain in LB). As predicted, the compensatory transcriptional mechanism was not limited to transcriptional regulation of gyrase expression but had a global effect. Using microarrays we estimated that steady-state transcriptional activity of as many as 847 genes had changed in the mutant relative to the isogenic wild-type strain. The microarray results were in part confirmed by reverse transcription-PCR (RT-PCR) for 10 out of 10 randomly chosen differentially expressed genes with an observed change in transcript abundance of 50% or greater. The distribution of differentially expressed genes along the chromosome is shown in a histogram in Figure 6 . Interestingly, transcription in the area of the chromosome spanning about 1.5 Mb from the vicinity of the terE site through the terG site appeared to be most affected in the mutant. If transcriptional activity of the chromosome is differentially perturbed, that is, some regions are being more affected than others, then patterns that existed in unperturbed chromosome may no longer spread across regions with such an 'out of sync' activity, resulting in a partial spatial confinement of originally wider patterns. Moreover, if transcriptional activity is inhibited in a spatial locality, it may cause reduction, or even complete elimination, of local patterns. The wavelet power spectrum in Figure 3d unambiguously confirmed such truncation of wider patterns and the disappearance of significant local patterns in the area of the chromosome where transcription was spatially differentially affected. No changes in transcription have been observed in the gyrase mutant carrying an S83L mutation, a naturally occurring mutation that is not accompanied by a compensatory increase in gyrase transcription (data not shown). Spatial patterns of DNA gyrase distribution on the chromosome While global and local patterns of transcription are likely to be due to multiple causes, our data suggest DNA gyrase as an important factor in pattern formation. It has been suggested that DNA gyrase is not randomly distributed along the chromosome of E. coli [ 24 ]. We examined the distribution of DNA gyrase along the chromosome in a chromatin immunoprecipitation (ChIP) chip assay with GyrA-specific antibodies. The averaged and de-noised spatial signal recorded in eight ChIP-chip microarray experiments were analyzed through a wavelet transform. The contours of significant scales calculated from the gyrase-binding signal and the transcriptional signal obtained under identical growth condition are overlaid in Figure 7a . Similarities between the two spectra were quantified as the dot product of power vectors at corresponding scales (Figure 7b ). We observed high positive correlation between the wavelet power spectra of gyrase distribution and transcriptional signal across multiple scales. We found no correlations between wavelet spectra of transcriptional signal and chromosomal distributions of several sequence-specific or nonspecific proteins, including Lrp, Topo IV, FtsK and LexA (data not shown; the results of these ChIP-chip experiments will be summarized in a separate paper). Discussion The development of new high-throughput technologies for parallel analysis of gene expression [ 25 ] and the completion of the full E. coli genome sequence [ 26 ] have enabled us to study the activity of the entire bacterial chromosome simultaneously. Owing to the physical limitations of some techniques and the invasiveness of others, the study of a system is limited to the analysis of signals coming from the system. The bacterial chromosome is a perfect example of a system that cannot be studied directly without interfering with its properties. Therefore, in order to obtain insights into the macroscopic properties of the bacterial chromosome, we chose to study the transcriptional signal that is generated by the chromosome and can be recorded on DNA microarrays. Transcript abundance along the chromosome can be represented as a one-dimensional signal in a spatial domain. The most pronounced feature of this transcriptional signal is a high degree of correlation between genes close together on the chromosome. While such a correlation is expected from genes organized in operons, the observed stability and range of correlations extends far beyond the expected size of the average operon. The stability of the short-range correlations can be significantly reduced by norfloxacin (Table 1 , Figure 8 ) suggesting that short-range correlations depend on negative supercoiling. Such dependence offers an intriguing hypothesis about the physical basis of the short-range transcriptional correlations: the transcription of the genes within a confined supercoiled domain is more similar to each other than to genes in other such domains, with the size of a domain being approximated by the linear stability of the short-range correlations. While this paper was in preparation, Postow et al . have shown that the size of a supercoiled domain in the E. coli chromosome is of the order of 10 kb [ 27 ]. The similarity between the dimensions of supercoiled domains and short-range transcriptional patterns makes our hypothesis even more plausible. Also consistent with our hypothesis is the observation that in a gyrase mutant with a transcriptionally compensated supercoiling function (plasmid DNA supercoiling is normal in the mutant, data not shown) the autocorrelation stability is statistically identical to that in the wild-type cells. By the same argument, however, other characteristics of transcriptional patterns do not appear to be strictly supercoiling-dependent. The observed changes in the long-range correlations could not be explained by changes in supercoiling because they are observed in the mutant as well as after drug treatment. It is more likely that changes in the medium- and long-range transcriptional patterns are associated with a change in the distribution of gyrase binding to the chromosome. The similarity between patterns of gyrase binding and transcription provides the basis for such a conjecture. It also seems plausible that coherent transcription, or changes in transcription, among clusters of functionally related genes [ 28 ] could be associated, in part, with their apparent regular spacing. Co-regulation of such clusters could contribute to the formation of observed local and global transcriptional patterns. We also note that transcription within operons is not sufficient to account for observed short- and long-range patterns. Randomization of the order of operons, as well as of individual genes, completely abrogates both types of correlations. It remains to be seen whether short-range patterns or nonrandom distribution of any sequence features, including extreme secondary structures [ 29 ], can contribute to the modulation of long-range correlations. Analysis of the structure in any signal can be complicated by instrument biases. Such biases in microarray measurements were originally pointed out by Speed and co-workers [ 30 ]. Although it has been argued that such systematic effects may preclude adequate spatial-temporal analysis of microarray data [ 31 , 32 ], we offer several reasons why our results are not a microarray artifact. First, the documented patterns could only be observed following subtraction of intensities in the reference DNA channel from the abundance transcript channel and not in the reference channel alone, suggesting that the patterns are not a property of hybridization efficiency. Second, the patterns could be significantly changed. Third, the patterns become more pronounced following the removal of the systematic biases and do not depend on the array design-specific periodicities whose removal by low-pass filtering has no effect on transcriptional patterns; and fourth, modeling of promoter activities carried out by Allen et al . [ 33 ] revealed the existence of at least one frequency component in the corresponding spatial series that was identical to the lowest-frequency component identified in this study through a direct modeling of transcript abundances. Conclusions This study demonstrates the existence of spatial patterns of transcription in the E. coli chromosome. These patterns can be classified on the basis of overall similarity in transcriptional activity of individual genes as well as on the basis of regional similarities. Three major spatial patterns have been identified: short-range correlations that are stable, on average, over 7 kb and could extend up to 15-16 kb; medium-range correlations over 100-125 kb; and long-range correlations over 600-800 kb. These patterns are experimentally stable and can be reproducibly detected in mid-exponential cells grown in batch culture. The growth rate and medium composition appear to have very minimal effects on pattern formation. However, these patterns could be modulated by perturbing DNA gyrase. The significant patterns of gyrase distribution on the chromosome match those of transcriptional activity. Among several proteins (see Results) whose distribution we mapped on the chromosome in mid-exponential culture, the pattern of gyrase binding was the only one coinciding with the patterns of transcription. Although it remains to be seen whether the observed patterns resulted from coherent transcription of functionally related genes regularly distributed along the chromosome and/or through chromosomal organization, the findings presented here are the first evidence of physiologically determined higher-order organization of transcription in any chromosome studied to date. Materials and methods Strains and culture conditions All experiments in this study were carried out in E. coli K12 strain MG1655 obtained from the ATCC. Mutant gyrA R alleles used in this study carried D82G (this study) and S83L [ 34 ] mutations. Wild-type and mutant gyrA R alleles were linked to a Tn 10 marker and P1 transduced into E. coli K12 MG1655 as described previously [ 35 ]. MG1655 dnaC2 [ 36 ] was used to obtain the DNA reference sample for transcript abundance measurements. Bacterial cultures were grown at 37°C in LB or M9 supplemented with 0.2% glucose. Microarray analysis Whole-genome DNA microarrays of E. coli were designed, printed and probed as described previously [ 13 , 18 ]. To ensure the success of PCR amplification and to minimize cross-hybridization we redesigned more than 700 primer pairs from the original set of primers supplied by Sigma-Genosys [ 37 ]. The relative transcript abundances were determined as described [ 38 ]. The RNA samples were extracted from the cultures grown to an OD 600 of 0.5-0.6 using the hot-phenol method [ 18 ]. The experimental error of the measurements of RNA abundances was assessed from at least three independent replicates, where one replicate corresponds to the RNA sample from a bacterial culture grown from a separate colony. Differentially expressed genes were identified using two-class comparisons of the adjusted relative expression values by SAM [ 39 ] at 1% false-discovery rate at the 90th percentile. RT-PCR was carried out on ABI Prism 7900 according to the Applied Biosystems protocol with SYBR Green dye as a fluorescent probe. Spectral analysis of spatial series Following the removal of the array-, pin- and dye-specific effects [ 40 , 41 ], the estimated relative abundance values were ordered according to the position of a corresponding gene on the chromosome and subjected to spectral analysis. The spatial domain is defined as a function of the position of the center of mass of the open reading frame (ORF) or operon. In the search for significant frequencies, 2,071 positive frequencies were examined in a signal consisting of 4,143 samples corresponding to individual genes. The autocorrelation function of transcriptional spatial series was calculated as in [ 20 ]: with j = 0,1,2,..., J , where y x is the series value at the index corresponding to a given ORF location, is the mean over all N observations and J is the number of genes in the genome minus 1. The standard error of autocorrelation estimates is determined as ( N ) -1/2 , where N is the number of samples in the series. Information about the process of transcription could be extracted by identifying a pattern in the observed variations in gene activities followed by a search for the cause or explanation of the pattern. For instance, the pattern may consist of a defined dependence of transcriptional activity as a function of a chromosomal position, or it may consist of the dependence of transcriptional activity as a function of time following a treatment or during the cell cycle. This functional dependence may express itself as a linear variation or harmonic oscillation partially hidden behind the noise. We considered a physical variable Y that corresponds to the relative abundance of mRNA. This variable could be measured as a function of the position, x , on the E. coli chromosome. Values of Y are discretely recorded following two-color hybridization on the whole-genome DNA microarrays. Thus there is a finite series of values of x , { x i }, i = 1, 2, 3,...., N ( N corresponds to the total number of genes) and corresponding values of Y , { Y i }, i = 1, 2, 3,...., N . We call { Y i } a spatial series or sequential series. For the sake of convenience we approximated positions of the centers of mass of individual genes, { x i }, as evenly spaced. A process is a rule or procedure that generates a sequential series, that is, a prescription for determining the values Y for a given set of values of x . We defined a spatial series recorded in one whole-genome hybridization as one realization of the process. The nature of the data that we register using microarrays as well as the nature of the process itself is likely to be such that the rule(s) generating a sequential series specify the probability distribution of { Y i } and not specific values that are the same at every realization. However, defining such a distribution did not seem feasible. Instead, we determined significant autocorrelation components in the spatial series by assessing the effect of experimental error on the significance of correlations. Following the recording of spatial series in at least three independent biological replicates, we simulated realizations of the process of transcription by resampling relative abundances in a gene-specific manner (see [ 42 ] for details of the bootstrap). For each of the realizations we calculated the ACF at all acceptable lags. We counted the number of times the ACF value appeared to be significant (α = 0.05) across all simulated realizations at each of the lags. The Fourier spectrum was determined using the Lomb algorithm [ 43 ]. To ensure comparability between the wavelet scale and the Fourier period we used the Morlet wavelet as a basis [ 44 ] in Matlab 6.5.1 [ 45 ] or AutoSignal 1.6 [ 46 ]. The significance of the Fourier and wavelet peaks was estimated from peak-type critical limits. The critical limits were generated from Monte Carlo trials with uniformly spaced spatial domain coordinates. The null hypothesis was simulated using white noise as a background distribution. Analysis of gyrase binding on the chromosome Chromatin immunoprecipitation of DNA sequences bound by DNA gyrase and their detection using whole-genome DNA microarrays were adapted from [ 47 ]. Briefly, cells were grown to an OD 600 of 0.5-0.6 in LB or M9 medium and DNA was cross-linked to proteins with formaldehyde at 1% (v/v) final concentration. Following incubation with monoclonal antibodies against the GyrA subunit of DNA gyrase (TopoGEN), the protein-DNA complexes were precipitated using Protein A-agarose beads (Sigma). Ligation-mediated PCR (LM-PCR; adaptor sequences: 5'-GCGGTGACCCGGGAGATCTGAATTC-3' and 5'-GAATTCAGATC-3') was used to amplify DNA following the reversal of cross-links. Sonicated genomic DNA served as a reference in two-color hybridizations, in which both the sample and the reference were labeled in the Klenow reaction following random amplification by LM-PCR. Additional data files Additional data file 1 , available with the online version of this article, consists of a table of the intensities used in the analysis of spatial patterns. The file contains fluorescent intensities recorded in individual channels in two-color microarray hybridizations. The data from replicate arrays are included in the same worksheet, which also contains a brief description of an experiment. The data are also available at [ 19 ]; series accession numbers are GSE1730 and GSE1735. Supplementary Material Additional data file 1 A table of the intensities used in the analysis of spatial patterns Click here for additional data file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC545777.xml

554108

Evaluation of dendritic cells loaded with apoptotic cancer cells or expressing tumour mRNA as potential cancer vaccines against leukemia

Background Leukemia is a clonal disorder characterized by uncontrolled proliferation of haematopoietic cells, and represents the most common form of cancer in children. Advances in therapy for childhood leukemia have relied increasingly on the use of high-dose chemotherapy often combined with stem-cell transplantation. Despite a high success rate and intensification of therapy, children still suffer from relapse and progressive disease resistant to further therapy. Thus, novel forms of therapy are required. Methods This study focuses on dendritic cell (DC) vaccination of childhood leukemia and evaluates the in vitro efficacy of different strategies for antigen loading of professional antigen-presenting cells. We have compared DCs either loaded with apoptotic leukemia cells or transfected with mRNA from the same leukemia cell line, Jurkat E6, for their capacity to induce specific CD4+ and CD8+ T-cell responses. Monocyte-derived DCs from healthy donors were loaded with tumor antigen, matured and co-cultured with autologous T cells. After one week, T-cell responses against antigen-loaded DCs were measured by enzyme-linked immunosorbent spot (ELISPOT) assay. Results DCs loaded with apoptotic Jurkat E6 cells or transfected with Jurkat E6-cell mRNA were both able to elicit specific T-cell responses in vitro. IFNγ-secreting T cells were observed in both the CD4+ and CD8+ subsets. Conclusion The results indicate that loading of DCs with apoptotic leukemia cells or transfection with tumour mRNA represent promising strategies for development of cancer vaccines for treatment of childhood leukemia.

Background Leukemia represents the most common form of cancer in children. There are two main types of childhood leukemia, acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). The success rate in treatment of childhood leukemia has improved continuously over the past decades [ 1 ], and today disease-free survival is 70%–80% for ALL and 40%–60% for AML [ 2 - 4 ]. In the Nordic countries the overall event-free survival in ALL has risen from 57% to75% [ 2 ]. However, in children with high-risk ALL, the progress has only been modest. The relapse rate has decreased in parallel with the improving results, but the prognosis after relapse has not improved. Only 25%–30% of children who relapse will reach and remain in a second remission. Children with AML have a worse prognosis than those with ALL. Event-free survival for AML is below 55%, whereas the cure rate for children with ALL is near 80%. The complete remission rate differs also, with 5%–10% induction failures due to refractory disease and toxicity in AML, compared to 1%–2% in ALL [ 2 ]. Immunotherapy based on vaccination with dendritic cells (DCs) has emerged as an attractive new form of therapy for cancer in general, and DC-based vaccines have already shown promise in follicular non-Hodgkin's lymphoma, and in other hematological malignancies [ 5 - 7 ]. DCs are antigen-presenting cells (APCs) specialized to induce T-cell responses against cells exposing foreign peptides, including tumour-related antigens, in context of MHC molecules [ 8 , 9 ]. DCs reside in tissues in an immature form, where they capture antigens from the environment. After antigen capture, and in response to inflammatory stimuli, DCs mature and migrate to lymph nodes to initiate immunity [ 9 ]. Maturation of DCs is associated with up regulation of the co-stimulatory molecules CD80 and CD86, increased expression of HLA molecules, enhancement of their APC function, and expression of CCR7 chemokine receptors that promote migration to the T-cell area in lymph nodes [ 10 ]. In several animal studies, it has been shown that immunization with cancer-antigen loaded DCs efficiently primes both CD4+ and CD8+ T-cells, resulting in protective immunity against tumours [ 11 - 16 ]. Vaccination with tumor antigen-loaded DCs has been shown to induce both Th and CTL responses, and tumor regression in some patients [ 16 , 17 ]. An important issue in optimizing DC vaccines is the choice of tumour antigen for loading of DCs. Several clinical trials in patients with melanoma have demonstrated that vaccination against a single antigen can induce tumour specific CTLs [ 18 ]. However, for many tumours no specific cancer antigens are known. For such patients, autologous tumor cells or tumor cell lines containing a repertoire of antigens overlapping with the repertoire in the patient's tumor, represents an alternative source of antigens. Effective cross-priming with antigens from tumour cells has been demonstrated with apoptotic cancer cells [ 19 - 21 ]. Transfer of whole tumor mRNA into DCs represents an alternative way of loading DCs. The transfected mRNA can be expressed for a relatively long period of time [ 22 - 24 ] and give rise to specific T-cell responses in vitro and following vaccination of patients [ 25 ]. So far, no studies on the relative efficacy of these two antigen loading methods have been performed. The aim of the present study was to compare DCs either loaded with apoptotic Jurkat E6 cells or transfected with mRNA isolated from Jurkat E6 cells, for their ability to generate T-cell responses against antigens derived from the human T-cell line. The Jurkat leukaemic T-cell line is a reference cell line [ 26 ], and was chosen as a source of antigen in the model experiments described here. We demonstrate that both strategies can be successfully employed to induce T helper and CTL responses against antigens derived from allogeneic leukemic T-cells. Methods Cytokines and chemicals GM-CSF was purchased from Novartis (Basel, Switzerland), IL-4, TNFα, IL1β and IL-6 from CellGenix (Freiburg, Germany), and Prostaglandin E2 (PgE 2 ), IL-7, IL-2 and IL-12 from R&D Systems (Minneapolis, USA). Staurosporin was obtained from (Sigma Aldrich, Saint Louise, Missouri). Preparation of DCs and T cells PBMC from healthy donors (obtained from Buskerud Hospital, Drammen, Norway) were obtained by density gradient centrifugation (Lymphoprep, Nycomed, Norway). Monocyte-derived DCs were generated under serum-free conditions from the adherent fraction of PBMCs cultured in six-well plates at a density of 4 × 10 6 cells/ml for 1.5 h at 37°C in 3 ml CellGro DC medium (CellGenix, Freiburg, Germany). Non-adherent cells were collected and frozen for later use as responder cells. Adherent cells were cultured in 3 ml CellGro DC medium, supplemented with 800 U/ml GM-CSF and 10 ng/ml IL-4 every second day, until day 5 when maturation of DCs was induced by addition of maturation cocktail (10 ng/ml TNF- , 10 ng/ml IL-1 1000 U/ml IL-6 and 1 μg/ml PgE 2 ) for 24 h. Characterization of DC phenotype was done by staining 0,5 × 10 6 cells with fluorochrome-labelled antibodies against the Lin1 panel (CD3, CD14, CD19, CD16, CD20, CD56), HLA-DR, CD1a, CD80, CD83, and CD86 (Becton Dickinson, San Jose, CA), and analyzing by FACSCalibur flow cytometry (Becton Dickinson). The mAb isotypes used were IgG1 FITC, IgG2a PE, IgG1 APC (Becton Dickinson, San Jose, CA). Assessments of apoptosis and phagocytosis of apoptotic cells Jurkat E6 cells obtained from American Type Culture Collection (ATCC) were exposed to 1 μM staurosporin for 3 h to induce apoptosis. Apoptotic cell death was assessed using Annexin V-FLUOS as described by the manufacturer (Boehringer Manheim, Manheim, Germany). For assessments of phagocytosis, Jurkat E6 cells were stained with the green fluorescent dye PKH-67 (Sigma Aldrich) as described in the kit manual, exposed to 1 μM staurosporin for 3 h and incubated with immature DCs at a ratio of 3:1. After 6 h, immature DCs were labelled with a red fluorescent antibody (mAb CD1a-PE). Phagocytosis of apoptotic cells was measured quantitatively by flow cytometry. Similarly, phagocytosis was visualized by confocal laser microscopy (Leica TCS SP, equipped with HeNe and Ar lasers) using apoptotic Jurkat E6 cells pre-stained with PKH-26 red fluorescent dye (Sigma Aldrich) and DCs stained with the green fluorescent dye PKH-67. Preparation of Jurkat E6-cell mRNA and transfection of DCs Jurkat E6 cells were used as a source of tumor material. Total RNA was isolated from 20–25 × 10 6 cells using Trizol Reagent as described by the manufacturer (Invitrogen, Basel, Switzerland). Poly (A) + mRNA was isolated from total RNA using the GenoPrep Direct mRNA kit (GenoVision, Oslo, Norway). Purified mRNA was used fresh or stored at -80°C until use. Transfection of DCs with mRNA was performed as described previously [ 22 ], with minor modifications. Briefly, immature DCs were washed once, resuspended in RPMI-1640 (BIO-Whittaker, Walkersville, MD) and placed on ice. 400 μl (approx. 2 × 10 6 cells) were mixed with mRNA, transferred to a 4-mm-gap cuvette and pulsed with a BTX ECM-830 square-wave electroporator (Genetronics Inc., San Diego, CA) using instrument settings 500 V and 1 ms. Transfected cells were incubated on ice for 30 s followed by addition of 2.0 ml cold CellGro DC medium supplemented with 10 ng/ml IL-4, 800 U/ml GM-CSF and maturation cocktail (see above), and transferred to standard culturing conditions. Transfection with EGFP-pCIpA 102 mRNA (10 μg/400 μl) encoding the green fluorescence protein [ 22 ] was used to verify transfection efficiency. Isolation of T-cell subsets CD4 and CD8 The Negative Isolation Kit (Dynal, Biotech) was used for isolation of CD4 and CD8 T cells according to the manufacturer's protocol. Isolation was performed on day 7 after in vitro priming, before setting up the ELISPOT assay. Induction of primary T-cell responses Mature DCs (0.3 × 10 6 ) expressing Jurkat E6-cell mRNA or loaded with apoptotic Jurkat E6 cells, were co-cultured with 3 × 10 6 autologous non-adherent PBMC for 7 days in 1.0 ml CellGro DC medium without serum, before setting up the ELISPOT assay. The cultures were tested for INF- production in an ELISPOT assay [ 27 ] following restimulation for 24 h with thawed antigen-loaded DC using 0.5 × 10 5 , 1.0 × 10 5 , 2.0 × 10 5 and 4.0 × 10 5 responding cells and 0.5 × 10 4 DCs per well. Mock transfected DCs were used as control. The assay was done in duplicate. Spots were counted manually, and the frequency of reactive T cells was calculated according to the formula: (spots with transfected DC - spots with non transfected DC)/number of T cells added. Results Generation of immature DCs and phagocytosis of apoptotic Jurkat E6 cells Based on previous observations that immature DCs efficiently capture antigens from the environment, we first investigated the ability of immature DCs to similarly phagocytose apoptotic Jurkat E6 cells. Immature DCs were prepared from PBMC in the presence of IL-4 and GM-CSF. To confirm generation of immature DCs, cells were examined by flow cytometry for expression of lineage and differentiation specific markers. The Lin 1 cocktail contains antibodies to CD3, CD14, CD16, CD19, CD20, and CD56 and differentiates DCs from other leukocytes by their lack of staining with Lin1. In contrast, CD1a and HLA-DR are expressed on immature DCs, and maturation is revealed by increased expression of CD83, and the costimulatory molecules CD80 and CD86. As shown in Fig. 1a the generated cells displayed the characteristic phenotype of immature DCs with high expression of CD1a and HLA-DR, and low or no expression of CD80, CD83 and CD86. For induction of apoptosis, Jurkat E6 cells were treated with staurosporin. In an independent series of experiments, optimal early apoptosis was observed after 3 h (data not shown) and early apoptotic cells were accordingly used in all experiments. To verify apoptosis after 3 h exposure to staurosporin, cells were stained with annexin V-FLUOS and examined by flow cytometry. The assessments showed that more than 90% of the cells were annexin-V positive (Fig. 2 ). To study uptake of apoptotic leukemia cells by immature DCs, Jurkat E6 cells were stained with the red fluorescent dye PKH-26 before induction of apoptosis. Immature DCs were stained with the green fluorescent dye PKH-67. Apoptotic Jurkat E6 cells were then co-incubated with immature DCs for various periods of time to determine the optimal conditions for internalization of apoptotic Jurkat E6 cells. We observed that immature DCs ingested apoptotic Jurkat E6 cells within 6 h of co-incubation. Phagocytosed apoptotic Jurkat E6 cells (red stained) were observed inside or in the process of being phagocytozed by immature DCs (green stained) by confocal microscopy (Fig. 3b ). Flow cytometry was used to further determine the efficiency of DC loading with apoptotic leukemic cells. In these experiments, apoptotic Jurkat E6 cells had been pre-stained with the green fluorescent dye PKH-67 and DCs were identified by staining with PE-conjugated anti-CD1a. Highly efficient uptake of apoptotic Jurkat E6 cells was confirmed, since virtually all CD1a positive cells showed green PKH-67 staining (Fig. 3a ). Following antigen loading, DCs were matured in the presence of pro-inflammatory cytokines for 24 h. Assessments by flow cytometry confirmed that this treatment led to up-regulation of CD83, and the co-stimulatory molecules CD80 and CD86, in compliance with a mature DC phenotype (Fig. 1b ). Transfection of immature DC with mRNA from Jurkat E6 cells Transfection of cells with tumor-derived mRNA is an alternative method for loading of DCs with tumor antigens. mRNA from the Jurkat E6 cells was isolated and electroporated into DCs according to previously optimized methods [ 22 ]. Following this protocol, optimal conditions for electroporation were 500 volt and 1 ms when using a 4-mm-gap cuvette. These conditions produced both efficient transfections (142 × background fluorescence; Fig. 4 ) and a survival rate indistinguishable from untransfected cells (data not shown). The transfected DCs were matured as described above and induction of the characteristic phenotype was confirmed by flow cytometry (Fig. 1c ). Analysis of T-cell responses ELISPOT assay of INF-γ producing cells is the method of choice for assessments of T-cell responses against cancer vaccines representing a heterogeneous mixture of antigens. This assay measures in a quantitative way the number of reactive T cells in pre and post-vaccination samples and thus directly relates the effect of vaccination to in-vivo expansion of reactive T cells. Autologous T cells were stimulated with tumour-mRNA transfected DCs and with DCs loaded with apoptotic Jurkat E6 cells. Fig. 5 shows the results from experiments with cells derived from three different donors. In all experiments a specific T-cell response against antigen-loaded DCs as compared to control DCs (mock transfected/non-loaded) could be demonstrated. No clear-cut difference between the two modes of antigen-loading was observed. In all experiments with un-fractionated T cells, we also observed T-cell reactivity against control DCs. This background completely obscured the specific response if in vitro priming was performed in the presence of exogenously added recombinant human IL-2 (results not shown). Antigen loading of DC by mRNA transfection and phagocytosis will introduce the Jurkat antigens into two different antigen processing pathways, cytosolic expression and processing for mRNA-encoded antigens and endosomal processing for phagocytosed apoptotic cells. As a result, one might expect that mRNA loading would preferentially result in activation of specific CD8+ CTLs, while loading with apoptotic cells would mainly result in activation of specific CD4+ Th1 cells. To investigate if this was the case, we separated the responding T-cell population into a CD4+ containing fraction and a CD8+ containing fraction using negative selection with Dynabeads coated with CD8 and CD4 antibodies respectively. The results shown in Fig. 5 demonstrate that a specific Th1 response was obtained in all donors and that both methods of loading resulted in a Th1 response. The frequency of specific Th1 cells varied between donors, with a trend indicating that mRNA loading in general is more efficient than apoptotic cells in priming of a Th1 response. The results depicted in Fig. 5 clearly show that relatively high frequencies of specific CTLs can be generated in all donors and that both methods of antigen-loading result in CTL priming. In donor 2 and 3, mRNA loading was clearly superior to apoptotic cells, indicating that expression of mRNA-encoded antigens more efficiently entered the proteasomal pathway for processing of HLA class I restricted antigens. Discussion Immunotherapy for childhood leukemia has the potential to contribute to long-term control or cure of the disease. Until now immunotherapeutic approaches for leukemia have been limited to trials of cytokine therapy [ 3 ]. Further development of biologically based treatments may prove to be effective in therapy of patients suffering from this disease. Several forms of DC-mediated immunotherapy are currently being investigated using a wide variety of vaccination protocols summarized in [ 28 ]. Two very important issues are the choice of antigen and the method of antigen loading. In the present study we have chosen to use the complex antigen mixture represented by whole tumor cells. Reports comparing the ability of apoptotic and necrotic cells to induce DC maturation [ 29 ] found that incubation of DCs with necrotic, but not apoptotic, tumor cell lines induce maturation. However, other reports concluded that incubation with apoptotic cells is sufficient to induce DC maturation [ 30 - 34 ]. In our study we have used apoptotic cells, and the requirement for DC maturation signals was provided by a standardized maturation cocktail. We accordingly analyzed human monocyte-derived DCs for their ability to: (a) take up apoptotic leukemia cells and express transfected mRNA, (b) express a mature phenotype following tumour-antigen capture and culture in maturation cocktail and (c) prime un-fractionated T cells as well as the CD4+ and CD8+ T-cell subsets. Due to the complexity of the antigens represented by the allogeneic tumor cells, the aim of these model experiments was not to use this allogeneic system to prove that we could elicit tumour specific T-cell responses in this way, but to provide data to demonstrate efficient antigen transfer and compare the relative efficacy of DCs loaded by the two different procedures, in eliciting complex T-cell responses. Our results demonstrate that immature DCs can efficiently take up apoptotic Jurkat E6 cells, and that phagocytosis was mainly confined to the CD1a+ subset of immature DC. Furthermore, support for expression of transfected mRNA derived from the allogeneic leukemia cell line is indirectly provided by its ability to prime T-cell responses specific for transfected cells. We also show that the two different methods of antigen-loading did not result in any apparent differences in the phenotype of the mature DCs. In terms of immune responses both methods of antigen loading produced DCs capable of inducing INF- secreting T cells. However, it appeared that DCs loaded with tumour-mRNA in general were most potent in inducing T-cell responses. We observed that the frequencies of induced INF-γ producing T cells depended on the individual donor, the method of antigen-loading and the subset of T cells studied. Such variations are not surprising, since the model system used employs allogeneic cells and no effort was done to HLA match the blood donors with the Jurkat cell line in these experiments. Since the experimental system is based on the use of an allogeneic cell line, we expect multiple antigens, encoded by a broad array of polymorphisms, including other HLA alleles to be involved. We have therefore taken advantage of the genetic differences between responding cells and the leukaemia cell line by using the combined repertoires of membrane expressed and cross presented allo-antigens as a sensitive readout for immunological response in our experiments. We expected that the two different procedures would provide some differences in loading of HLA molecules with tumour-derived antigens and subsequently in the responding T-cell subsets. According to the current dogma, processed peptides from phagocytosed apoptotic cells would be directed to HLA class I molecules by a process known as cross-presentation and to HLA class II molecules by the classical pathway. Cross-priming of CTL with antitumor activity has been demonstrated with DCs loaded with apoptotic tumour cells [ 19 , 21 , 35 ]. Specifically, Schnurr et al. demonstrated the antigens from apoptotic pancreatic carcinoma cell lines, either in form of whole cells or as released particles, were potent in inducing CTL-cell priming and activation by DC. In addition, Hoffman et al. reported stronger CTL responses with apoptotic tumour cells in a squamous cell carcinoma model. The enhanced CTL activation by antigens from apoptotic cells may be attributed to several mechanisms. After ingestion, most particulated antigens requiring phagocytosis are digested into peptides associating with HLA class-II molecules in the endocytic compartments and are presented to T-helper cells [ 36 ]. Conversely, scavenger receptor-mediated phagocytosis of apoptotic tumour cells allows antigens to gain access to HLA class-I compartments, resulting in cross-presentation of the antigens to CTL [ 20 ]. In addition, enhanced CTL responses to tumours might be mediated by heat shock proteins expressed by stress induced apoptotic tumour cells [ 37 ]. On the basis of this theoretical background and reported observations [ 21 , 31 ] we believe that antigen preparations from apoptotic tumour cells can also represent an alternative in DC-based tumour vaccines. On the other hand, tumour mRNA expressed in DCs would be processed for presentation by HLA class I molecules. In accordance with this, DCs loaded with apoptotic leukaemia cells stimulated both CD4+ and CD8 positive T cells and mRNA loaded DCs were superior in inducing CD8+ T-cell responses [ 38 , 39 ]. Interestingly, mRNA-loaded DCs were also able to induce specific CD4+ T-cell responses in all donors tested, suggesting some leakage of endogenously produced proteins into the lysosomal antigen-processing compartments. Similar results have recently been published by Su et al., who demonstrated a significant Th response against the defined tumour antigen hTERT following in vitro stimulation of un-fractionated T cells with hTERT mRNA transfected DC. The Th response could be further augmented by targeting the antigen to the lysosomal compartment using mRNA encoding a chimeric hTERT/lysosome-associated membrane protein (LAMP-1) protein. The aim of the present study was to determine if loading of DCs with antigens derived from a tumour cell line, either as apoptotic cells or as mRNA would provide a basis for an efficient vaccine, using ELISPOT as a read-out of immune responses. It has been shown that DCs transfected with antigens encoded in tumor mRNA is capable of inducing potent T-cell responses against tumour-specific epitopes [ 40 ]. While protein antigens from tumour lysate are rapidly proteolysed following endocytosis by antigen-presenting cells, model experiments using mRNA encoding a fluorescent protein, EGFP, has shown that protein is still being produced 24 hrs after transfection of DCs, with peak expression after 48 hrs [ 22 ]. Thus, tumor mRNA transfected DCs may not only represent a potent strategy for CTL priming but may also represent a general method for DC-based vaccines. In vaccine preparations using DCs, mRNA is thus preferable to a protein lysate. Similarly, immunization with DCs loaded with mRNA from leukaemia cells could represent a feasible approach in treatment of these cancers. It is now widely accepted that not only CTLs but also CD4 (+) T-helper cells are critical to the generation and maintenance of potent antitumor responses in vivo. In this context, our observation and that of others demonstrating that DCs loaded with mRNA also are equally capable of inducing Th responses strongly argue in favour of this type of vaccination. Our preclinical results further support that vaccination of leukemia patients with tumour-mRNA transfected autologous DCs should be clinically evaluated as therapeutic strategy. Conclusion In our study we demonstrate that both DCs loaded with apoptotic Jurkat E6 cells or transfected with mRNA isolated from Jurkat E6 cells, can induce T-helper and CTL responses against antigens derived from allogeneic leukemic T-cells. We also show that the two different methods of antigen-loading did not result in any apparent differences in the phenotype of the mature DCs. In terms of immune responses both methods of antigen loading produced DCs capable of inducing INF- secreting T cells. However, it appeared that DCs loaded with tumour mRNA in general were most potent in inducing T-cell responses. Competing interests The author(s) declare that they have no competing interests. Authors' contributions SJJ performed preparation of DCs and T cells, assessments of apoptosis and phagocytosis of apoptotic cells, preparation of Jurkat E6-cell mRNA and transfection of DCs, isolation of T-cell subsets CD4 and CD8 and induction of primary T cell responses. SSL did the transfection of DCs with EGFP mRNA and fluorescence microscopy of DCs. RP, FW and GG planned the project. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC554108.xml

543474

A simulation model of Escherichia coli osmoregulatory switch using E-CELL system

Background Bacterial signal transduction mechanism referred to as a "two component regulatory systems" contributes to the overall adaptability of the bacteria by regulating the gene expression. Osmoregulation is one of the well-studied two component regulatory systems comprising of the sensor, EnvZ and the cognate response regulator, OmpR, which together control the expression of OmpC and OmpF porins in response to the osmolyte concentration. Results A quantitative model of the osmoregulatory switch operative in Escherichia coli was constructed by integrating the enzyme rate equations using E-CELL system. Using the substance reactor logic of the E-CELL system, a total of 28 reactions were defined from the injection of osmolyte till the regulated expression of porins by employing the experimental kinetic constants as reported in literature. In the case of low osmolarity, steady state production of OmpF and repression of OmpC was significant. In this model we show that the steady state – production of OmpF is dramatically reduced in the high osmolarity medium. The rate of OmpC production increased after sucrose addition, which is comparable with literature results. The relative porin production seems to be unaltered with changes in cell volume changes, ATP, EnvZ and OmpR at low and high osmolarity conditions. But the reach of saturation was rapid at high and low osmolarity with altered levels of the above components. Conclusions The E-CELL system allows us to perform virtual experiments on the bacterial osmoregulation model. This model does not take into account interaction with other networks in the cell. It suggests that the regulation of OmpF and OmpC is a direct consequence of the level of OmpRP in the cell and is dependent on the way in which OmpRP interacts with ompF and ompC regulatory regions. The preliminary simulation experiment indicates that both reaching steady state expression and saturation is delayed in the case of OmpC compared to OmpF. Experimental analysis will help improve the model. The model captures the basic features of the generally accepted view of EnvZ-OmpR signaling and is a reasonable starting point for building sophisticated models and explaining quantitative features of the system.

Background Among prokaryotes, a remarkable number of cellular functions are controlled by two component regulatory systems [ 1 ]. Dedicated circuits transduce and interpret specific signals such as pH, temperature, osmolarity, light, nutrients, ions, pheromones and toxins to regulate a wide range of processes including motility, virulence, metabolism, the cell cycle, development switches, antibiotic resistance and stress responses in a variety of systems from prokaryotes, archaea and eukaryotes [ 2 ]. Two component systems interact with each other and also with other regulatory systems mediating specific gene expression or cellular locomotion. Such complexes should be analysed quantitatively. Although quantitative data on cellular processes other than metabolism are still relatively sparse, scientists have modeled some systems with considerable success [ 3 ]. For example computer simulations of the chemotaxis two component system have been extensively studied [ 4 ]. Two-component regulatory systems (also called the HAP or His-Asp phosphotransfer mechanism) are widely used signaling machinery of bacterial adaptive responses. The system constitutes of sensor kinases and response regulators that can be phosphorylated and dephosphorylated by sensor kinases [ 5 ] (Figure 1 ). The first component sensor-transmitter that spans the cytoplasmic membrane has two domains a sensory domain and a transmitter domain. The two domains are anchored in the cytoplasmic membrane by two membrane-spanning regions. The transmitter domain is capable of autophosphorylation utilizing ATP. During this reaction a phosphate group (PO 4 ) is transferred to a specific Histidine (His) residue in the protein. The transmitter domain possesses a kinase activity that enables it to transfer the phosphate group to the receiver-regulator, when the environmental signal is detected [ 6 ]. The receiver-regulator that is cytoplasmically localized also consists of two domains. The first domain has a specific Aspartate (Asp) residue that can accept the phosphate group (PO 4 ) from the transmitter domain. The second domain is the regulator that in response to phosphorylation can bind to specific DNA sequences near specific promoters to activate gene expression. These two domains are linked together by a flexible linker [ 7 ]. Osmoregulation is one of the well-studied two component regulatory systems operative in Escherichia coli , playing an important role in regulating the cellular response to different solute concentrations in their environment [ 8 ]. Among such types of osmotic responses, the expression of the major outer membrane proteins, OmpC and OmpF, has been the subject of extensive studies [ 9 ]. Both the OmpC and OmpF proteins form passive diffusion pores in the outer membrane, which facilitates the diffusion of small hydrophilic molecules across the membrane [ 10 ]. EnvZ the sensor kinase serves as a substrate for the phospho-transfer to aspartate-55 of OmpR, the response regulator [ 11 , 12 ]. EnvZ is a bifunctional histidine kinase that exhibits dual opposing functions, both phosphorylation and dephosphorylation of OmpR [ 13 , 14 ]. The sensor kinase is triggered by the osmolyte concentration of the environment and controls the production of phosphorylated regulator OmpR. This leads to the expression of outer membrane proteins OmpC or OmpF [ 15 ]. Phosphorylated OmpR binds to the regulatory sequences upstream of the ompC and ompF promoters [ 16 , 17 ]. OmpR undergoes a conformational change based on phosphorylation [ 18 ] and regulates the expression of the porin genes ompF and ompC in Escherichia coli [ 19 ]. The osmolyte works as the initial control element of the phosphorylation cascade. The primary signal for such a conformational change may be that caused by a change in the physical membrane-tension due to osmotic pressure [ 20 ]. First, the EnvZ-dimer, in the cytoplasmic membrane senses the environmental osmotic stimulus. The N-terminal membrane-spanning and periplasmic domains of the EnvZ-dimer presumably can take on two alternative conformational states ( i.e ., a high osmolarity form and a low osmolarity form) regulated by the osmotic signal and modulates the ratio of the kinase to phosphatase activity of EnvZ [ 21 , 22 ]. Under low osmolyte concentrations, Phosphatase activity of EnvZ predominates the kinase activity resulting in the binding of phosphorylated OmpR to the high affinity promoter of ompF gene triggering OmpF expression. In case of high osmolyte concentration the kinase activity of EnvZ is triggered resulting in the binding of the phosphorylated regulator to the low affinity promoters of ompC gene favouring OmpC expression [ 23 ]. Events associated with high osmolarity In the high osmolarity state, EnvZ actively undergoes autophosphorylation at histidine residue-243 in the C-terminal kinase domain, and then efficiently transfers its phosphoryl group to the N-terminal receiver domain of OmpR at aspartate residue-55 through EnvZOmpR complex formation [ 24 , 25 ]. Upon phosphorylation, OmpR becomes an active dimer that exhibits enhanced DNA-binding ability specific for both the ompC and ompF genes. As the number of phosphorylated OmpR protein molecules increases, two events occur: OmpR binds not only to the high affinity binding sites upstream of the ompF promoter but also to the one low affinity-binding site. Binding to this low affinity site results in repression of ompF gene expression. So OmpF porin protein production is stopped. When OmpR binds to the three low affinity sites upstream of the OmpC promoters ompC gene expression is stimulated, more OmpC porin protein is made and appears in the outer membrane of the cell [ 26 ]. The summary of events is depicted in Figure 2 . Events associated with low osmolarity In the low osmolarity state (Figure 3 ), however, EnvZ exhibits relatively low kinase activity ( i.e ., high phosphatase activity) towards OmpR. Such osmotic modulation of the kinase/phosphatase activity of EnvZ results in the relative amounts of the phosphorylated form of OmpR in cells varying in response to the medium osmolarity. When the medium osmolarity is low, the relative amount of the phosphorylated form of OmpR in cells is relatively small. In this particular situation, the ompF gene is first triggered, because the ompF promoter has relatively high-affinity OmpR-binding sites [ 27 ]. As the medium osmolarity increases, the relative amount of the phosphorylated form of OmpR increases proportionally, this in turn results in preferential activation of the ompC gene with low affinity OmpRP binding sites [ 9 ]. In summary, the relative amount of phosphorylated OmpR protein in the cell determines whether OmpF or OmpC is the predominantly expressed outer membrane porin. The relative amount of phosphorylated OmpR is determined by the perception of osmotic pressure by EnvZ [ 28 ]. Results and discussion Simulation of low osmolarity Reports have proposed that at low osmolarity, OmpRP binds cooperatively to F1, F1F2 and F1F2F3 sites resulting only in OmpF expression (Figure 3 ). Accordingly low osmolarity conditions were simulated assuming that Escherichia coli cells are grown in normal nutrient medium. The first event of start of simulation is the activation of phosphatase activity of EnvZ followed by autophosphorylation of EnvZp by ATP dissociation. Later EnvZpp-OmpR complex formation occurs (Figure 3 ) [ 29 ]. This short lived complex triggers phosphorylated regulator OmpRP. The phosphatase activity carries out the dephosphorylation of the phosphorylated regulator. Thus the concentration of cellular OmpRP is available only for binding of ompF promoters leading to OmpF porin expression. In the case of Low osmolarity steady state production of OmpF and expression of OmpC could be seen initially at the start of simulation. However at the saturation of OmpF, repression of OmpC was significant. The steady state production of OmpF expression and OmpC repression is indicated in Figure 4a . The sucrose levels were maintained as normal (around 150 molecules). OmpF synthesis seen to be triggered at the start of simulation. The entire trend of OmpF synthesis, gradual increase, steady state (at 50 seconds after the start of simulation with about a constant increase of 10 molecules for every 10 seconds) and final saturation (near 90 seconds with an constant value of 262 molecules) are represented graphically. (Figure 4a ). A similar trend is seen with OmpC molecules reaching saturation at 70 seconds. The intermediates of low osmolarity pathway namely promoter binding is indicated in Figure 5a . It could be seen that EnvZpOmpR complex is formed at very basal levels thereby directly having control over OmpC molecules. Additionally only 3.5% of cellular OmpR was phosphorylated at minimal sucrose in low osmolarity conditions validating the model that is operative in low osmolarity. Simulation of high osmolarity For high osmolarity conditions, the model generated incorporates the concentration of sucrose with the assumption that Escherichia coli cells are grown in nutrient broth with 20% additional concentration of sucrose (1.11 M equivalent) [ 30 ]. With this injection stimulus, the kinase activity of EnvZ is enhanced leading to OmpC expression. The first event of start of simulation is the activation of kinase activity of EnvZ followed by autophosphorylation of EnvZk by ATP dissociation. Shortly thereafter, EnvZkOmpR complex level rises and the dissociation of the complex raises the level of the phosphorylated regulator OmpR. This is followed by cooperative binding of OmpR to the ompF promoters and ompC promoters. OmpC molecules start accumulating. In the course of time there is complete dissociation of EnvZkOmpR complex, thereby both sensor and response protein are brought back to the pool (Figure 2 ). In our model sucrose levels turns the EnvZ to take up kinase activity by increasing EnvZk concentration [ 31 ]. In this model we show that the steady state production of OmpF is dramatically reduced in the high osmolarity medium. As the cooperative binding of OmpRP to ompC promoter sites is known to require a higher concentration of OmpRP than ompF promoter sites, this is achieved by reducing the EnvZ phosphatase activity. In this manner, a few fold increase in the OmpRP concentration on the cell is enough to induce OmpC expression and concomitant repression of OmpF expression by OmpRP binding to the F4 repressor site. Recent in vivo studies reveal that F4 site is key factor responsible for OmpF repression [ 32 , 28 ]. The rate of OmpC production increased notably after sucrose addition, which is comparable with literature results. OmpC expression reaching saturation and subsequent OmpF repression is indicated in Figure 4b . In high osmolarity, the increase of sucrose level (to about 1.11 M equivalent) [ 30 ] in silico through the virtual pipette directly favours OmpC production. OmpC production follows the same trend as OmpF in steady state (from 60 seconds to 90 seconds with a constant increase of 13 molecules per second) and saturation reaches by 100 seconds. OmpF repression shows steady state increase of 6 molecules for every 10 seconds and finally at saturation takes up a constant value by 90 seconds (Figure 4b ) but still the events are delayed compared to low osmolarity. Here again the key marker EnvZk is found in significant levels as seen in the graphical representation and so are the promoter-regulator complex (Figure 5b ). Here 10% of cellular OmpR is finally phosphorylated at the end of simulation agreeing with the literature data. Effect of Volume changes over porin production Shrinkage of Escherichia coli is associated with osmotic change [ 33 ]. The effect of volume increase and decrease over the porin production was verified with the simulation model. A volume decrease of 10% and 20% from the specified 10 -15 Litres was incorporated into the simulation model at low and high osmolarity conditions. In both cases the decrease in volume did not affect the relative porin production. Although the reach of saturation was rapid, the levels of porin and their relative ratios were found to be maintained with the same trend. This agrees with the data by Wood [ 34 ] with regard to the phases of the osmotic stress response by Escherichia coli K-12. The simulation model presented is at the first phase of physiological and structural responses triggered by osmotic shift where there is decreased cytoplasmic streaming. Our simulation time also corresponds with the time period of this first phase of the responses in the above literature. Table 2 represents the OmpC and OmpF levels at decreased volume levels. Effect of ATP changes over porin production To address the issue whether ATP levels has any effect over relative porin levels, simulation run was done at high and low levels of ATP. ATP level of 3 and 5 mM has been reported in exponentially growing Escherichia coli cells [ 35 , 36 ]. Simulation was carried out at these two concentrations of ATP. Table 2 indicates relative porin levels at high and low ATP levels. As has been reported earlier, the ATP increase leads to plasmolysis, thereby leading to crowding of molecules [ 34 ]. This does not seem to affect the ratio of porins in the simulated system even though the reach of simulation was rapid. This is possibly due to the robust nature of the system. Porin production in the complete absence of the Sensor, EnvZ Regulation of OmpC and OmpF expression in Escherichia coli in the absence of sensor, EnvZ has been studied [ 17 ]. We have thus examined the steady state production of OmpF and OmpC in the absence of EnvZ and also looked at the rate of production of the porins during osmolytic shift. As reported in vitro , EnvZ is required for the maximal OmpC production and for efficient induction of OmpC at high osmolarity. This is established in the in silico model. Also the lack of EnvZ in the simulation did not affect the OmpF at low osmolyte condition and incomplete OmpF repression could be noticed after osmolyte shift as reported by in vitro studies. The relative levels of porins without EnvZ is cross verified with the data reported by Frost et al. [ 17 ], over steady state production of OmpF and OmpC with minimal sucrose and osmolytic shift conditions (Table 2 ). Effect of Elevated EnvZ levels over porin production Within the context of the EnvZ/OmpR two component system, the mathematical model predicts the OmpF/OmpC transcription to be insensitive to variations in the level of EnvZ and OmpR. By increasing levels of EnvZ upto 10 fold (1000 molecules), the relative porin ratio was found to be constant during simulation, evidently agreeing with the robust nature of the switch as per the mathematical model [ 35 ]. Effect of OmpR levels over porin production As with the case of EnvZ, through in silico model the OmpC/OmpF transcription was found to be independent of the OmpR levels upto 10-fold increase (20000 molecules). In either case both at high and low osmolarity, porins levels were found to be insensitive to elevated or decrease levels of response regulator, OmpR corresponding to the results of mathematical model (Table 2 ). Conclusion Signaling pathways, for example, commonly operate close to points of instability, frequently employing feedback and oscillatory reaction networks that are sensitive to the operation of small number of molecules [ 37 , 38 ]. The model simulated here is clearly a simplified description of the EnvZ/OmpR system. There are a number of aspects of the circuit that have not been included such as EnvZ dimerization, conformational changes of OmpR [ 6 ] or additional enzymatic steps. The simulation is based on the mathematical model of the EnvZ-mediated cycle of phosphorylation and dephosphorylation [ 39 ]. Thus, this model predicts that the regulation of OmpF and OmpC as a direct consequence of the level of OmpR-P in the cell and is dependent on the way in which OmpR-P interacts with sites in the ompF and ompC regulatory regions [ 40 ]. Previously, it was suggested on the basis of a simplified model for the EnvZ-mediated cycle of phosphorylation and dephosphorylation of OmpR that the output of the circuit (the concentration of OmpR-P) should be independent of the concentration of EnvZ and OmpR in the cell [ 41 ]. We have shown porin regulation at high and low osmolyte concentrations where the dual activity of EnvZ is primarily controlled by the concentration of osmolyte stimulus at the start of simulation. The preliminary simulation experiment indicates that both reaching steady state expression and saturation is delayed in the case of OmpC compared to OmpF. The relative porin production seems to be unaltered with changes in cell volume, ATP, EnvZ and OmpR at low and high osmolarity conditions. But the reach of saturation was rapid at high and low osmolarity with altered levels of the above components. Experimental analysis will help improve the model. The model captures the basic features of the generally accepted view of EnvZ-OmpR signaling and is a reasonable starting point for building sophisticated models and explaining quantitative features of the system. At the same time, beyond its applicability to EnvZ-OmpR the model provides an interesting mechanism for achieving robust behavior with a bi-functional enzyme that may be broadly applicable to the other regulatory circuits within cells. Methods The E-CELL Windows version 2.25 was employed for simulation [ 42 , 43 ]. The software was installed with the third party software namely Active Perl, JRE (java runtime environment) and Borland C++ compiler [ 44 ] essential for running simulations. The information defining all the components of the osmoregulatory switch, reactions and appropriate reactors and rate constants and environmental parameters describing volume was incorporated in the rule file. This file was further compiled through Active perl. The order of reaction kinetics, time of simulation and time interval was specified in script file. The reactor is basically a file describing /defining the kinetics of the equation along with rate constants. Reactor file were complied using C++ compiler. Figure 6 summarizes the method of construction of quantitative model. Creation of rule file based on the mathematical model The computational model of osmoregulatory switch is based on the mathematical model by Goulian and Batchelor [ 39 ]. The entire model is described in the rule file. The cell system and cell environment was defined first. Changes in volume could not be incorporated, hence in this in silico approach, volume parameters were assumed constant. Also simulations with other system have assumed volume as a constant parameter irrespective of the system simulated. Simulations could not be defined and shown visually for volume parameter, as E-CELL has no provision for spatial information. Table 3 details the list of substances in osmoregulatory with their respective substance IDs. Reactor Specifications E-Cell is based on an object-oriented modeling theory, structured Substance-Reactor Model (SRM). The simulation models are constructed with three fundamental object classes, Substance, Reactor and System. Substances represent state variables, Reactors represent operations on the state variables, and Systems represent logical and/or physical compartments containing other objects. The distributed package of version employed for carrying out simulation has 18 different classes of standard Reactors, such as for Michaelis-Menten formula and generalized chemical equilibrium [ 43 ]. In the simulation systems, the rate equations of all the reactions are defined. Every reaction follows different kinetics based on the substrate involved and is dependent on reaction type. The reactors employed for osmoregulatory switch includes mass action, Catalysed mass action (specified in Table 4 ) defining all reactions from sucrose injection till the regulated expression of the porins. Reactor is the term employed here, as in E-CELL, to describe the reaction rate. The volume of the system is assumed to be unchanged during simulation using a constant parameter reactor. Molecular binding such as osmolyte interaction and response regulator DNA binding was modeled using Mass Action reactor, which computes velocity as a product of concentration of substrates and a kinetic constant. The expression of porin was modeled on mass action principles with catalyst embedded using a Catalyzed Mass Action reactor. Autophosphorylation of sensor, sensor-regulator complex formation and ATP dissociation was modeled using MichaelisUniUni reactor. Auto dephosphorylation reactions was modeled using MichaelisUniUnireactor. Zeroreactor, which calculates velocity independent of concentration of molecular species, was employed for modeling complex dissociation. A Decay reactor was employed for defining the disintegration or decay of components. Table 4 details the reaction type and reactors with their respective chemical constants employed for constructing the model. The present model is built on the assumption of the in vivo condition considering Escherichia coli cells grown in mid-log phase. Accordingly the levels of OmpR and EnvZ are reported to be 3500 and 100 molecules in cell respectively. OmpR and EnvZ levels were almost the same from cells grown in L-broth medium or in a high osmolarity medium (NB (Nutrient Broth) +20% sucrose) [ 26 ]. The ratio of OmpR to EnvZ is reported to be constant, assuming the cell volume to be 10 -15 liters [ 45 ]. At low osmolarity the phosphorylation of only 3.5%(120 nM or 70 OmpRP molecules/cell) of total OmpR molecules in a cell (2024 molecules OmpR molecules per cell) would be enough to activate the expression of OmpF, whereas at high osmolarity the phosphorylation of about 10%(590 nM or 350 OmpRP molecules /cell) of total OmpR molecules in a cell (3500 molecules per cell) would be sufficient to promote the expression of OmpC and to repress the expression of OmpF [ 46 ](Table 1 ). The majority of OmpR still remains unphosphorylated, as it's pool is very large. It is important to note that the osmoregulation of the OmpF and OmpC gene is finely tuned by having a very large pool of OmpR molecules [ 30 ]. The list of key substances participating in osmoregulatory switch with their initial concentration at the start of simulation is summarized in Table 5 . The data was adapted from Cai and Inouye [ 30 ]. As the data with regard to the number of promoters was not available, different values were taken and checked with the porin production and their relative ratios. The relative ratios were found be unaltered with any promoter levels. Authors contribution KVS was responsible for data collection and analysis. SK conceived of the study, and participated in its design and analysis. Two referees and an advisor of the journal helped to bring this information into the biological context.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC543474.xml

555558

Automatic detection of false annotations via binary property clustering

Background Computational protein annotation methods occasionally introduce errors. False-positive (FP) errors are annotations that are mistakenly associated with a protein. Such false annotations introduce errors that may spread into databases through similarity with other proteins. Generally, methods used to minimize the chance for FPs result in decreased sensitivity or low throughput. We present a novel protein-clustering method that enables automatic separation of FP from true hits. The method quantifies the biological similarity between pairs of proteins by examining each protein's annotations, and then proceeds by clustering sets of proteins that received similar annotation into biological groups. Results Using a test set of all PROSITE signatures that are marked as FPs, we show that the method successfully separates FPs in 69% of the 327 test cases supplied by PROSITE. Furthermore, we constructed an extensive random FP simulation test and show a high degree of success in detecting FP, indicating that the method is not specifically tuned for PROSITE and performs well on larger scales. We also suggest some means of predicting in which cases this approach would be successful. Conclusion Automatic detection of FPs may greatly facilitate the manual validation process and increase annotation sensitivity. With the increasing number of automatic annotations, the tendency of biological properties to be clustered, once a biological similarity measure is introduced, may become exceedingly helpful in the development of such automatic methods.

Background Computational protein annotation is a major goal of bioinformatics and annotation methods are widely used. A wide variety of annotation methods exist, many of which rely on some kind of scoring. Typically, when testing whether a protein should be given a certain annotation, a score threshold is set, and proteins that score higher than the threshold are given the annotation. Obviously, some annotation mistakes may occur. Such mistakes can be divided into false positives (FPs) and false negatives (FNs). FPs (or false hits) are annotations that were mistakenly assigned to a protein (type I error). FNs (or misses) are annotations that should have been assigned to a protein but were not (type II error). Adjustment of score thresholds allows tradeoff between these two types of mistakes. FPs annotations are considered to be of graver consequence than FNs. This is partially due to the fact that introduction of a false positive annotation into a protein database may cause other proteins to become incorrectly annotated on the basis of sequence similarity [ 1 , 2 ]. A systematic evaluation of the source of false annotations that already contaminated current databases was reported [ 3 ]. Several automatic systems such as PEDANT [ 4 ] and GeneQuiz [ 5 ] were introduced with the goal of matching the performance of human experts. Still, over interpretation, FN errors, typographic mistakes and the domain-based transitivity pitfall [ 6 ] limit the use of such fully automatic systems for inferring protein function. Due to the importance of minimizing the amount of false annotations and maintaining highly reliable protein databases, three methods are generally used to avoid false annotations. The first method is manual validation of the annotation of each protein, which creates a serious bottleneck in the addition of new proteins and annotations to the database. The second method is using high score thresholds, thus lowering the rate of FPs but also increasing the rate of FNs (resulting in a loss of sensitivity). The third method is requirement for hits from different detection methods, eliminating advantages that are unique to some methods. Thus it would be beneficial to develop means by which FP annotations could be detected automatically, allowing both high throughput and high sensitivity. Here we present such a method that uses clustering of protein functional groups to separate true positives (TPs) from FPs automatically. Our method is based on the following notions: (a) protein annotations represent biological properties; (b) protein functional groups share specific combinations of biological properties, essentially constituting "property clusters"; (c) if two proteins have very different combinations of annotations, they are unlikely to share a single functional annotation and therefore there is a high chance that one of them was given that annotation incorrectly. These notions are not obvious, but were shown to correctly indicate false annotations in some individual cases tested manually using the graphical annotation-analysis tool of PANDORA [ 7 ]. We aim to generalize these sporadic observations and to address the feasibility of automating the detection of FP. Using these ideas, the method attempts to separate a group of proteins into "property clusters", by introducing a measure that quantifies the similarity between the annotation combinations of two proteins. According to our basic notions, these clusters are likely to be in accordance with false and true hits. We tested our method on the PROSITE protein signature database [ 8 ]. The database consists of 1189 protein signatures (essentially annotations) that were assigned to a protein database. PROSITE annotation of proteins is manually validated, stating for each protein hit whether the annotation is a TP or a FP. Out of this set of 1,189 signatures, we chose a subset of all signatures that have both true and false hits, and this served as our test set. Altogether 327 such signatures were collected and tested. For each of the signatures, the method examined the set of proteins that were assigned the signature. We called the separation successful only if at any step of the clustering process all the TPs were clustered together without any FPs. We applied a stringent scoring, where a partial success is considered failure. Furthermore, we constructed a random FP simulation test in order to provide a more extensive test. In this test, all 5,551 InterPro [ 9 ] annotations were considered. For each InterPro annotation we selected the set of proteins in SwissProt [ 10 ] that were assigned that annotation, and added to that set random proteins, simulating proteins that were assigned the annotation by mistake (FPs). For each annotation we repeated the test 15 times: 5 times with 1 random protein, 5 times with 5 random proteins and 5 times with 10 random proteins. This artificial contamination of the annotation source strives to simulate mistaken annotations that may occur under some automation annotation inference schemes. Results Property-based clustering We begin by describing the method of property-based clustering. Given a set P of all proteins that were given a certain annotation, and that there are both FPs and TPs in P, we would like to separate the set into disjoint subsets, so that one of the subsets will include all TPs and no FPs (leaving one or more subsets containing the FPs). Annotation-based clustering is used to detect these subsets. We define an annotation as a binary property assigned to a protein (each protein may or may not have a "hit"). At the first stage, annotations from GO (Gene Ontology) [ 11 ], InterPro (entries) and SwissProt (keywords) are gathered for all proteins in P. The clustering works in the following way: between each two proteins we define a similarity score that tries to quantify how much do the two proteins have in common from a biological perspective. The score between two proteins p 1 and p 2 is defined as: where A 1 and A 2 are the set of annotations of proteins p 1 and p 2 respectively, i is the current annotation, and f ( i ) is the frequency of i in the database. This score uses the following logic: if two proteins share an annotation, they are biologically similar in some manner. The more annotations these proteins share, the more cause we have to believe that they are similar biologically. However, two proteins sharing an annotation like "Enzyme" (that appears 45,991 times in our database) should receive a worse similarity score than two proteins that share a much uncommon annotation like "Heat Shock Protein" (that appears only 832 times). This is taken into account by using log( f ( i )). Obviously, one could think of different scoring schemes that would quantify this differently. For a specific example of how the score is calculated see Table 1 . The similarity score is calculated between every two proteins in P. Next, we define the similarity score between two clusters as the arithmetic average of scores of all inter-cluster protein pairs: where C 1 and C 2 are clusters of proteins. Starting with clusters of 1 protein each, the method begins by an initial one-step clustering which merges all clusters that have the exact same combination of annotations. Following this the primary clustering commences: At each clustering step the two clusters that have the highest similarity score are merged. At each step the contents of the clusters are evaluated, and if all TP proteins appear in one cluster without any FPs, we say that the clustering process successfully separated the TPs from the FPs. Note that we do not require all the FPs to be grouped into one cluster, due to the fact that they cannot be expected to share biological similarity amongst themselves. PROSITE test Out of 327 sets of proteins that share a PROSITE signature, the method showed successful separation (as defined previously) in 227 sets, i.e. 69% of the cases. The average size of the protein sets was 156.1 and the median 76. Altogether 58,254 proteins were used for this test. The average and median FP rates (FP rate is defined as: FP/(TP+FP)) of the sets were 0.12 and 0.07 respectively. These general statistics about the test set indicate that the sets were large enough and had a high enough amount of TPs and FPs so that the chance of random success would be minimal. In order to demonstrate the method's performance in this test, we provide the following example of testing a single protein set. The set presented here is the set of all 37 proteins that matched the PROSITE "Serum albumin family" signature. Each protein in the set contains an average of 18.2 annotations (obviously not all are relevant). First, the score between every pair of proteins is calculated, based on their mutual annotations. Next, the proteins undergo a preliminary clustering step in which all proteins that have the exact same combination of annotations are merged into clusters. Following this, the proteins are clustered together based on their mutual similarity score. Finally, once the clustering has finished we examine the tree to see if the true positives were separated from the false positives. In the given example, there are 5 proteins that were incorrectly assigned the PROSITE annotation (FPs), and in Figure 1 we see that they are indeed separated from the TP proteins. Random FP simulation test 5,551 sets of proteins were tested 15 times each and showed successful separation in 74% of the cases. Altogether 99,076 proteins were used for this test. This can be subdivided into 78% success for the sets that had 1 random protein added, 74% success for the sets that had 5 random proteins added and 68% for the sets that had 10 random proteins added. The average set size was 78 proteins. The drop in the performance by increasing the level of FPs is due to the fact that there is a higher chance that one of the randomly selected proteins will be biologically similar to the TPs. Since we consider only cases in which all FPs are detected, then there would be a higher chance of failure as the number of randomly-generated FPs increases. While the simulation of FP errors randomly provides endless amounts of test sets, which is a clear advantage over the limited test sets provided by a real database such as PROSITE, the simulation has its own limitations. The hidden assumption made by this approach is that the FP hits are independent of each other. This assumption is not necessarily true: for example, if annotation is done by means of sequence similarity, false hits may be more likely to be biologically similar to each other (e.g. belong to the same family). In fact, in many cases in the PROSITE test we find that the correct separation separates the TP proteins into one cluster and the FP into one or two clusters, suggesting that the FPs share some degree of biological similarity (see "Determination of the correct halting step"). This difference in the way that FP annotations are generated may also account for the difference in success rates between the PROSITE test set and the simulated test set. The way FP annotations are introduced into databases is impossible to model, but the combined success of the method on both a real database test set and on an extensive simulated test set seems promising. A further issue which concerns the simulation method is determining the amount of FPs to add to each set. Here we chose to add 1, 5 or 10 proteins to each set. This does not necessarily reflect the amount of FPs in real databases. Understandably, each database's average FP rate depends on its specific characteristics. However, the PROSITE database's average FP rate of 0.12 (median of 0.08) might give an indication as to what a typical rate is. In comparison, the average FP rate for our random simulation set was 0.11 (median of 0.07), which suggests that our choice was reasonable. Determination of the correct halting step We call a clustering process successful if it managed at any step to separate the false annotations. However, this step must be somehow determined automatically. There are two approaches to this: one is to use an intrinsic parameter of the clustering process that would indicate where the correct halting step is located; the other is selecting a predetermined step of the process. We chose the similarity score at each merging step as an intrinsic process parameter. When plotting the score against the progression of the clustering (Figure 2 ), a knee shape in the plot would indicate a point of stability (biological similarity), suggesting it as a potential halting step. Analysis of the second derivative of this plot allows finding these knee-shaped stability points automatically. Using this method, 56% percent of the correct halting steps in the PROSITE test were correctly predicted. A different approach was to always choose the last step or the last two steps as the correct halting step. This resulted in 45% and 65% correct prediction, respectively. Furthermore, the union of the correct predictions made by both approaches indicates that together they correctly predict the halting step in 79% of the PROSITE test cases. Discussion Prediction of success Interestingly, we found that with certain sets the method tended to be more successful than with other sets, probably indicating that these sets are more coherent biologically. This might suggest exploring an approach in which for each annotation one could predict the level of success provided by this method. Furthermore, we used the InterPro categorization of annotations into types in order to check success in specific annotation types. InterPro divides its annotations into different categories, such as "domain", "repeat" and "family". Understandably, "family" type annotations had a ~30% higher success rate than the other annotation types, primarily due to the fact that the "family" annotations often represent protein sets that are biologically coherent whereas other types such as "repeat" or "domain" annotations are biologically diverse. This result would be expected by a method that performs a clustering based on biological similarity. This indicates that this approach should be aimed primarily at functional family annotations. However, functional families can be defined at different resolutions: an alcohol dehydrogenase belongs to the enzyme family, the dehydrogenase family and the alcohol dehydrogenase family. The test sets of the PROSITE and InterPro databases mainly represent mid-level and low-level annotations, with a typical size of tens or a few hundreds of proteins (see the statistics given previously). In order to further our understanding of the resolution in which this method is successful, we divided the protein groups into size categories and studied the relative amount of success in every size category. Figure 3 shows that as the group size increases, the rate of success decreases. Assuming larger sets represent the higher level annotations of InterPro, this suggests that when the annotations are more general ("higher" in the biological functional hierarchy) they have less in common biologically. Therefore, we would not expect the method to succeed on very general terms such as "enzyme". Sporadic tests of several high level GO annotations suggest that this is indeed the case (data not shown). Annotation source interdependency Because multiple annotation sources were used, concerns arose regarding interdependencies amongst them. For example, InterPro is highly dependent on PROSITE, so proteins that have a PROSITE annotation will very likely be assigned an InterPro annotation as well automatically. In order to minimize this effect, we did not allow the algorithm to use the InterPro annotations that matched the PROSITE annotation which was being tested. Furthermore, in order to increase reliability of the random FP simulation test, all known PROSITE FPs were removed from InterPro prior to the test. Still, there is some concern that the results are partially biased due to annotation source interdependencies. Furthermore, it is difficult to determine whether these dependencies represent true biologically dependent properties, or simply a duplication of the same property in different sources. Keeping this difficulty in mind, our results which show different levels of success for different types of annotations (see "Prediction of success") indicate that the success of the method is more likely due to biological dependency rather than artificial duplication. Sufficient annotation It should be stressed that the clustering process is based on sufficient annotation. Therefore, it may be difficult to apply this method to proteins that are poorly annotated. Still, these cases should be relatively rare: Nearly 77% of the ~1,600,000 proteins in TrEMBL [ 10 ] have at least one annotation by InterPro, and when considering several annotation sources there are on average ~10 annotations per SwissProt protein. Note that the amount and richness of annotation is constantly increasing at a fast rate. Furthermore, the ability to detect false annotations automatically may allow an increase in the sensitivity of current methods, thereby allowing more extensive annotation of proteins. It is worthwhile noting that amongst the 58,254 proteins used in these sets there were 3,587 (6%) proteins annotated by SwissProt as "hypothetical proteins". 18% of the sets that were successfully separated contained such hypothetical proteins, with an average of 8% hypothetical proteins for each such set. These results suggest that the method is capable of handling to some extent hypothetical proteins of unknown function. Another helpful approach to the problem of insufficient annotation could be the introduction of quantitative protein properties that are easily determined and show some correlation with function (i.e. the protein length, its Isoelectric point, etc.) into this method. Preliminary testing showed some positive correlation between protein length and Isoelectric point with function in certain cases (not shown). Conclusion Introduction of FP annotations into protein databases can be harmful. It has been shown that once a mistaken annotation is introduced into a database, it often transfers to other proteins that are sequentially similar causing a propagation of false annotation [ 1 ]. Due to the importance of keeping high-quality databases, either the proteins are manually checked one by one or the annotation detection sensitivity is reduced in order to minimize FPs. The error rate and the limited sensitivity of assigning structural annotations using PSI-BLAST [ 12 ] or SAM-T98 [ 13 ] and methodologies based on HMMs and SVMs had been reported [ 14 ]. Naturally the process of manual validation of the annotation of protein databases is extremely time-consuming and in many cases is subjective to the expert view. Automatic detection of false annotations greatly facilitates the task of manual validation of annotation, and allows using lower thresholds when trying to detect protein signatures, therefore allowing higher method sensitivity. Based on the notion that protein functional groups share specific combinations of annotations, we have introduced a method that by separating a set of proteins into biological "property clusters" shows successful separation of incorrectly annotated proteins from correctly annotated proteins. We test the method both with a manually validated test set and with a randomly constructed test set, and in both cases show a high degree of success. These results suggest that this tendency of certain annotations to appear in groups may be used as a basis of automatic methods that detect FPs. Naturally, different computer learning methods can be used to take advantage of these interdependencies of biological properties (for example see [ 15 ]). Methods Sources We created a database that includes all proteins from SwissProt 40.28 (114,033 proteins) [ 10 ]. The database also included annotation of these proteins by GO[ 11 ], SwissProt and InterPro [ 9 ]. GO terms represent a wide range of biological terms concerning molecular function, cellular localization and biological processes, and span various degrees of specificity: from very general terms to very specific ones. GO terms are assigned to proteins both manually and automatically. InterPro annotations are assigned automatically by sequence and typically represent functional families and domains of no more than a few hundred protein members. SwissProt keywords are assigned manually and cover various biological subjects. Annotation source and the number of annotation for each (in parenthesis) are: SwissProt version 40.28 (865 keywords), InterPro version 5.2 (5,551 entries), GO as of July 2002 (5,229 terms), PROSITE version 17.5 (1,189 signatures). Authors' contributions NK and ML conceived of the study. NK designed the method. NK implemented and developed the method. NK designed the tests and analyzed the results. NK and ML wrote the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC555558.xml

539327

Sorghum Genome Sequencing by Methylation Filtration

Sorghum bicolor is a close relative of maize and is a staple crop in Africa and much of the developing world because of its superior tolerance of arid growth conditions. We have generated sequence from the hypomethylated portion of the sorghum genome by applying methylation filtration (MF) technology. The evidence suggests that 96% of the genes have been sequence tagged, with an average coverage of 65% across their length. Remarkably, this level of gene discovery was accomplished after generating a raw coverage of less than 300 megabases of the 735-megabase genome. MF preferentially captures exons and introns, promoters, microRNAs, and simple sequence repeats, and minimizes interspersed repeats, thus providing a robust view of the functional parts of the genome. The sorghum MF sequence set is beneficial to research on sorghum and is also a powerful resource for comparative genomics among the grasses and across the entire plant kingdom. Thousands of hypothetical gene predictions in rice and Arabidopsis are supported by the sorghum dataset, and genomic similarities highlight evolutionarily conserved regions that will lead to a better understanding of rice and Arabidopsis .

Introduction Sorghum bicolor is a vitally important crop in Africa and much of the developing world. It has a remarkable ability to endure both drought conditions and water-logging and it grows well on marginal lands [ 1 ]. It is the dietary staple of more than 500 million people in more than 30 countries with only rice, wheat, maize, and potatoes feeding more people than sorghum [ 1 ]. Sorghum is in the panicoid grass subfamily and is closely related to maize, millet, and especially sugarcane, and is more distantly related to wheat and rice. Its value as a dietary staple to much of the world and its placement within the grass family make it a valuable target for genome sequencing. Genome sequencing in most plants is difficult because of the size and complexity of the genomes. Plant genomes range in size from 54 megabases (Mb) for Cardamine amara to 124,000 Mb for a lily ( Fritillaria assyriaca ) [ 2 ]. Although they vary drastically in size, the larger genomes do not correspond to proportionally more genes, but instead to repetitive elements that have blossomed in the plant kingdom [ 3 , 4 , 5 , 6 ]. The extremely large genomes of such economically important crops as bread wheat (16,900 Mb), maize (2,600 Mb), soybean (1,100 Mb), and sorghum (735 Mb) [ 2 ] make them difficult to tackle with standard methods of genome sequencing such as clone-by-clone [ 7 ] and whole-genome shotgun [ 8 ]. For example, a whole-genome shotgun project of maize to 8× genome equivalents would require nearly 24 million sequencing reads, and sorghum would require 7.5 million reads. Additionally, the maize and sorghum genomes are more than 75% repetitive [ 9 , 10 ], which would make the final assembly of shotgun sequence extremely difficult [ 11 ]. The large-insert clone-by-clone approach solves some of the difficult assembly problems, but it requires a much larger initial investment in resources and is much more expensive. Furthermore, the highly repetitive large-insert clones would still be difficult to assemble. Evidence has accumulated over the last ten years that many plant genomes are separated into large tracts of methylated repeats and stretches of hypomethylated, low-copy gene–rich space [ 4 , 6 , 12 , 13 , 14 , 15 ]. On the basis of this knowledge of plant genome architecture, two techniques have been developed to isolate the low-copy or hypomethylated regions of the genome for sequencing. The first technology, high C 0 t selection (C 0 t is the product of the DNA concentration [C 0 ] and the reassociation time in seconds [ t ]), allows the separation of low-copy sequences from those of high copy based on annealing rates [ 16 , 17 ]. High C 0 t selection has been used successfully to sequence the low-copy, genic regions of maize [ 18 ] and has been applied to sorghum [ 10 ]. The second technology is methylation filtration (MF), which preferentially clones the hypomethylated fraction of the genome. MF has also been successfully applied in maize to sequence the genic regions [ 18 , 19 , 20 ]. It appears that MF will be a successful strategy across the plant kingdom, as it has been shown to enrich for genes in all plants tested, from monocots to dicots to gymnosperms, and even in nonvascular plants such as moss (P.D. Rabinowicz, unpublished data). We have applied MF technology to generate sequence from the hypomethylated portion of the sorghum genome. Successful sequencing of fewer than 550,000 MF reads revealed that approximately 96% of the gene set of sorghum has been sequence-tagged, with an average coverage of 65% across their length. Because MF targets genomic sequence within and around genes, many important components of the genome are represented, including promoters, microRNAs (miRNAs), introns, simple sequence repeats (SSRs), and potentially active transposable elements. The sorghum gene space is a powerful resource for comparative genomics within the grass family and across the plant kingdom. The MF dataset can be used to confirm hypothetical genes in complete genomes such as rice and Arabidopsis and to identify functional elements conserved across different plant species. Results/Discussion The Size of the Genome Space Sampled by MF To calculate the genome space sampled by MF, two independent methods were used, genome sampling and gene-enrichment. Genome sampling is an empirical calculation based on a modification of the Lander-Waterman equations [ 21 ], as used by Whitelaw and colleagues [ 18 ]. The reduced genome size is calculated based on the size of the sampled space as judged by the number of times that independent reads overlap. Independent reads will overlap more often when sampling a small region versus a larger region; therefore, one can derive an empirical assessment of the size of the region being sampled [ 18 ]. The sampled genome space for the sorghum MF set is 262 Mb. The gene enrichment method works on the assumption that genes are enriched in the MF libraries in proportion to the reduction in genome size. For example, if the genome is reduced by 3-fold, then gene discovery should occur 3-fold faster in MF versus whole-genome shotgun libraries. The extent to which this number agrees with the genome sampling method is the extent to which the genes reside in the sampled space. We calculate gene enrichment because it can be estimated very early in a sequencing project, whereas the genome sampling method requires at least 0.1× coverage of the sampled space to get an accurate estimate (unpublished data). The gene enrichment factor is called filter power (FP); we use FP to derive the sampled genome space by dividing it into the size of the whole genome. We calculated the sorghum FP using a subset of our filtered and unfiltered (UF) sequences compared to a curated database of known genes over a range of BLAST Expect values (E-values) ( Table 1 ). The FP is between 3.0 and 3.8 with a median value of 3.15. By dividing this range of FP values into the 735 Mb sorghum genome, the sampled genome is estimated to be between 193 Mb and 245 Mb, with a median of 233 Mb. The median estimate is somewhat lower than the 262 Mb estimation derived by the genome sampling method. However, the result depends critically on genome size estimates, which for S. bicolor range from 735 Mb to 858 Mb [ 2 ]. If 858 Mb is used, gene enrichment predicts a 272-Mb gene space, which is slightly higher than the 262 Mb obtained by genome sampling, thus bracketing the genome sampling approximation, depending not only on the range of FP, but on the range of genome size estimates. Table 1 Gene Enrichment (or FP) of MF Versus UF Sequences FP was calculated by comparing the MF and UF sequences to a curated set of Arabidopsis proteins, then dividing the proportion of hits in MF by the proportion of hits in UF over a range of BLAST E-values. The median FP is 3.15, with a range of 3.0 to 3.8 Therefore, completely independent estimates of gene space, namely genome sampling and gene enrichment, agree well and are within measurement error. For the purposes of this manuscript, 247 Mb, which is the average of the two methods, will be used as an approximation of the sampled, hypomethylated genome space ( Figure 1 ). The MF dataset consists of a nuclear coverage, after collapsing read pairs, of 285 Mb, which is approximately 1.15× coverage of the sampled space. Figure 1 Genome Reduction MF reduces the sorghum genome by 66% in sampling a hypomethylated space of approximately 247 Mb (green) and filtering out 488 Mb (red) of the 735-Mb sorghum genome. Gene Tagging and Coverage The purpose of a genome reduction method such as MF is to identify genes in a robust and efficient manner. We assessed the efficiency of gene discovery by calculating the percentage of known genes tagged as a function of read number for MF and compared this value to the rate of gene discovery obtained by expressed sequence tags (ESTs) for sorghum ( Figure 2 ). Additionally, we conducted a simulation in Arabidopsis to assess the expected gene identification rate in a completed plant genome where the level of coverage could be controlled precisely in silico (see Expected Gene Tagging, below). The results of these analyses are summarized in Figure 2 . Figure 2 Gene Discovery Rate Gene discovery rates for sorghum MF (blue), sorghum ESTs (pink), and an Arabidopsis simulation (dotted black) are shown. The gene discovery rates for the MF and ESTs were calculated based on matches to a set of 137 genes annotated on sorghum BAC clones versus the number of MF and EST reads. The Arabidopsis simulation was calculated based on the fold-coverage of chromosome 1, which contains 7,520 genes. The fold coverage was converted into read numbers as detailed in the Materials and Methods . To estimate the percentage of genes that have been tagged by MF, we used high-quality sorghum bacterial artificial chromosome (BAC) sequences as a source of gene annotations. At the time of analysis, 14 finished sorghum BACs had been deposited in GenBank ( http://www.ncbi.nlm.nih.gov/ ). Because the GenBank annotations were outdated, we reannotated the BACs through a custom annotation pipeline (see Materials and Methods ). We annotated a total of 148 genes on these BACs, then mapped our MF reads to the BACs using stringent BLAST criteria. Of the 148 genes, the MF reads match 133 (90%) of them, with an average nucleotide coverage of 61%. However, 11 of the 148 annotations are alpha kafirin storage protein genes on BAC AF527808. Ten of them constitute a tandem repeat cluster of nearly identical sequences that could be expected to be methylated [ 22 ] and are therefore not recovered efficiently in a MF library. This is indeed the case, as only two out of the 11, or 18%, are recovered in the MF clones. This is far below the 90% average for the whole set, suggesting that the kafirin genes may be at least partially methylated (see Methylated Gene Recovery, below). If we remove these 11 genes from the analysis, 131 (95.6% [ Figure 2 ]) of the remaining 137 genes are tagged across 65% of their nucleotides. We also removed the kafirin genes from the EST analysis in Figure 2 . In addition to tagging 95.6% of the gene set, a majority of the coding sequence (CDS), upstream, and downstream genomic regions are covered. The average coverage of the CDS regions of all 137 genes is 65%, thus providing a tag across more than half of the gene on average. This coverage is consistent with the 67% nucleotide coverage predicted at 1.15× raw sequence coverage [ 21 ]. Additionally, we calculated the nucleotide coverage 500 basepairs (bp) upstream (5′) and downstream (3′) of the CDS and found 74% and 69% coverage, respectively. The coverage of the 5′ and 3′ regions is higher than expected, which is at least partly due to the close spacing of sorghum genes in this set, with 16/137 (greater than 10%) having 5′ and/or 3′ regions within 1 kb. For comparison, the gene tagging ability of the publicly available sorghum EST sequences was assessed. At the time of analysis, there were 161,766 sorghum ESTs deposited in GenBank. Using criteria of 98% identity over at least 50 bp of the CDS, the sorghum ESTs matched 84/137 (61%) of the annotated BAC genes ( Figure 2 ). Notably, the ESTs did not match any of the 11 kafirin genes. Expected Gene Tagging: An Arabidopsis Simulation If MF faithfully represents the genic region of sorghum and contains the vast majority of the genes, then the rate of gene tagging should produce results that are similar to whole-genome shotgun coverage [ 21 ] at the same level of raw coverage. To test this hypothesis, we simulated a whole-genome shotgun project of the finished Arabidopsis chromosome 1 (see Materials and Methods ). We decided to use Arabidopsis for the simulation because it is finished to high quality, the gene predictions are the most robust of any plant species, and Arabidopsis best represents the size of plant genes, which are much smaller on average than animal genes. The simulation showed that, at 1.1× coverage, 96.4% of the genes are sequence-tagged across 66.8% of their length. These numbers are very similar to the percentages calculated from the MF gene tagging analysis (95.6% of genes covered over 65%). Since the simulation is set to replicate Lander-Waterman whole-genome shotgun conditions, these results mean that MF obeys the mathematics of Lander-Waterman, although it is a highly fragmented sampling space. Furthermore, if the BAC gene set is representative of the genome, this implies that nearly all the genes in the genome are accessible to MF and that all genes are currently covered over an average of 65% of their length. Theoretically, 100% nucleotide coverage will be reached at 6× coverage, which would require less than 2.5 million additional MF reads. Figure 2 shows the comparison of the gene tagging rates for the Arabidopsis simulation, the MF reads, and the sorghum ESTs. Notice that the gene tagging for the sorghum ESTs and MF are more rapid than the Arabidopsis simulation. Rather than reflecting a real difference in ability to tag genes using MF versus whole-genome shotgun, this higher rate likely reflects the larger average gene length for the sorghum CDS annotations (3 kb) versus Arabidopsis (2.3 kb), making gene tagging more rapid in sorghum. Additionally, the sorghum ESTs show the most rapid gene-tagging rate, but begin to level off at 60% gene tagging and are passed by the sorghum MF after 70,000 reads. Methylation of Transposons, Repeats, and Pseudogenes Overall, recognizable repeats constitute 62% of the sorghum genome ( Table 2 , Unfiltered), which is comparable to maize [ 18 , 19 ]. Retrotransposons are the most abundant class of repetitive DNA sequence, occupying about 1/3 of the genome, followed distantly by DNA transposons at 1/20 of the genome ( Table 2 , Unfiltered). MF reduces the recovery of ribosomal genes, centromeric repeats, and retrotransposable elements ( Table 2 , Filtered), so that only 27% of filtered reads match repeats. These results can be described in terms of the total fraction of repeats ( R/N , where R is the total length of repeats in the genome and N is the size of the genome), the unmethylated fraction of repeats ( r/UM , where r is total length of repeatsin the unmethylated fraction and UM is the size of the unmethylated genome), and the filter power (FP) ( N/UM ) according to Palmer and others [ 19 ]. Given a FP of 3.15 ( N/UM ), we can calculate the proportion of unmethylated repeats ( r/R ) as ( [r/UM]/[R/N] )/( N/UM ), or approximately 10%. This is consistent with maize [ 19 ], and indicates that a substantial portion of sorghum transposons, especially DNA transposons, are unmethylated and may be capable of transposition. For example, the active sorghum transposon Candystripe1 ( Cs1 ) [ 23 ] is represented in our dataset across 23% of its length (unpublished data). The lower-than-average percent coverage (23% versus 66%) may be due to some methylation within the element, as has been reported for several maize transposons [ 12 ]. Additionally, Cs1 is known to have a low copy number (less than 10) in sorghum, and the redundancy of coverage across the 23% represented suggests that MF is sampling from a single element (unpublished data). Table 2 Repeat Analysis for MF Versus UF Reads a Note that the MF sequences sample approximately 1/3 of the genome, so this percentage of repeats reflects 1/3 of the total genomic content MITES, miniature inverted terminal repeat elements The majority of methylation in plants occurs at the canonical sites CG and CNG (where N is any nucleotide) [ 24 , 25 , 26 , 27 ]. MF uses in vivo restriction via modified cytosine restriction, subunits BC (mcrBC) at the recognition site (A/G) methylated cytosine (mC). The observed versus expected occurrences of mcrBC sites, along with those sites that overlap the canonical methylation sites of CG and CNG, are shown for retrotransposons and genic sequences in Figure 3 A and 3 B, respectively. Although the mcrBC half-sites ([A/G] C) occur as expected in MF and UF retrotransposons and genes, the sites that overlap canonical methylation sites are significantly reduced in MF versus UF retrotransposons, but not in genic sequence, where, in fact, they occur more frequently in MF than UF ( Figure 3 ). It has been shown previously that CG and CNG nucleotides are suppressed in MF repetitive elements [ 19 ], presumably because mCs have been converted over time to thymine by deamination [ 28 ]. Our results suggest that such conversion has occurred in transposon sequences, but not in genes, consistent with their differential methylation. Figure 3 CG and CNG Suppression in MF versus UF Sequences Sequences were analyzed for their mcrBC half-sites, those that overlap CG dinucleotides, and those that overlap CNG trinucleotides. The ratio of observed to expected sites is graphed for filtered (hatched) and unfiltered (white) for retrotransposons (A) and CDSs (B). The increased frequency of CG and CNG nucleotides in genic sequences recovered by MF versus UF ( Figure 3 B) suggests that CDS derived from MF and UF are different. One source of this difference may be the presence of pseudogenes. In plants, most pseudogenes are marked by small insertions and deletions, resulting in frame shift(s) of the coding region, but are otherwise indistinguishable from functional genes [ 29 ]. Pseudogenes are likely targets of silencing and are thus probably methylated, excluding them from MF sequences. To test if pseudogenes are more abundant in the UF dataset, sequences from both UF and MF that matched Arabidopsis proteins, and are therefore considered genes, were compared to a database of all plant proteins using BLASTX. Sequences with more than one high-scoring segment pair and with an E-value of 1 × 10 −20 or less were analyzed for the presence of a frame shift. The rate of potential frame shifts for UF is 103/530 (19.4%) versus 1,599/17,103 (9.35%) for MF, indicating that pseudogenes are recovered at a higher rate in UF (comparable to the rate of retrotransposons) and are therefore most likely methylated. Methylated Gene Recovery Comparison with the BAC sequences revealed that a small number of genes were not represented in the sorghum MF reads. Two explanations were considered: First, these genes may have been missed by chance, as only 97% of sorghum genes were expected to be sampled by this depth of coverage. Second, these genes might be methylated. Two examples were chosen for further analysis: the teosinte branched2 gene ( tb2 ), which was recovered in our dataset, and the kafirin storage protein gene cluster ( Figure 4 ). The kafirin gene cluster was chosen because it is underrepresented in the MF sequences and could be methylated since it is a tandem repeat cluster [ 22 ]. We used a real-time PCR technology to assess DNA methylation (see Materials and Methods ). As expected, methylation analysis of tb2 (on BAC AF466204) indicates that it is unmethylated ( Figure 4 A and 4 C). Figure 4 Methylation Status of tb2 and Kafirin Cluster (A and B) Restriction maps of the tb2 gene (A) and the kafirin consensus sequences (B) are shown. The relevant restriction sites are indicated vertically and the numbers indicate the distances scale in basepairs. Each CDS is depicted as a blue-shaded arrow, and the region assayed is indicated by a black bar. The circles depict sites that are not present in every kafirin gene, and the color represents the number of genes that do not share the site. The orange circle (5′-most HhaI site) is a site conserved in nine of 11 kafirin genes, and the red circle (3′-most PstI site) is a site present in ten of the 11. (C) Results from a representative methylation analysis of tb2; the inset depicts the template dilution standard curve used to set the threshold for the experiment. Each experiment was performed three times with four on-board replicates per assay point. The results for each of the four differentially treated reactions are depicted with different colors. Red, mock-treated; blue, mcrBC-digested; orange, HhaI-digested; and green, HhaI + mcrBC double-digest. The inset shows the standard dilution control with two replicates at each dilution. The control was used to set the threshold for detection. The specificity of each reaction was confirmed using melt-curve analysis. (D) Results from a representative methylation analysis of the 11 kafirin genes. The results for each of the six differentially treated reactions are the same as in (C), with the following additional digests: pink, PstI-digested; light blue, PstI + mcrBC double-digest. Notice that the mcrBC with and without PstI yields the same Ct, while HhaI + mcrBC (green) yields a higher Ct on average; suggesting additional cleavage. For the kafirin gene cluster, only two of 11 genes from BAC clone AF527808 were represented in the MF dataset, suggesting that most or all of them may be methylated. Ten of the genes are tandemly arrayed in a cluster and share an average of 99.1% sequence identity, while the eleventh gene is located 45 kb away and is more diverged (76.2% identity on average). A 247-bp region was selected for PCR close to the 5′ end because of its near identity across all 11 genes and because of the high CG and CNG content ( Figure 4 B). The methylation results are depicted in Figure 4 D. PstI sites are methylated (at CNG), since the PstI-treated sample ( Figure 4 D, pink) has the same cycle threshold (Ct) as the mock-treated sample ( Figure 4 D, red). This result is supported by the mcrBC digested sample, which has a significantly higher Ct ( Figure 4 D, dark blue) than the mock-treated DNA control. All, or almost all, of the PstI sites are methylated, because the double PstI +McrBC digest ( Figure 4 D, light blue) has the same Ct as mcrBC alone ( Figure 4 D, dark blue). These results indicate that every gene has CNG methylation covering these sites. As for CG methylation, the HhaI-digested (orange) sample has the same Ct as the mock-treated control (red); however, the Ct of the HhaI + McrBC double digest (green) is 2.46 cycles greater than the mcrBC alone (dark blue), indicating that some HhaI sites must not be modified. A cycle threshold difference of 2.46 indicates that there is 2 2.46 , or approximately 5.5-fold, less DNA in the HhaI + mcrBC double-digested sample. This suggests that two out of the 11 kafirin genes have some unmethylated HhaI sites. To determine which kafirin genes might be unmethylated, we sequenced the kafirin PCR products from mcrBC treated and untreated genomic DNA (gDNA). 130 sequences from mcrBC-treated DNA and 126 sequences from the mock-treated sample were analyzed. The kafirin genes fall into “subfamilies” based on six polymorphisms within this highly conserved genomic region (see Materials and Methods ). Each of these subfamilies was represented among the sequenced clones, including the orphaned kafirin gene outside the tandem array, indicating that none was completely removed as a consequence of mcrBC treatment. Thus, it is likely that all the kafirin genes contain some level of methylation, and that the genes are displaying nonuniform CG methylation randomly, perhaps on a per-cell basis, across all 11 genes. Drought Resistance Genes In order to assess how useful the current low level (approximately 1×) coverage of the gene space is for answering important comparative genomics questions, we chose to analyze genes related to drought resistance. Sorghum's ability to grow in arid conditions makes it an attractive source of genes to enhance drought resistance in other grasses. Part of the drought-responsive pathway in plants involves the activation of dehydration-responsive element binding protein (DREB) transcription factors belonging to the APETALA2 (AP2) family. The overexpression of DREB1-encoding genes can promote drought, freezing, and salinity tolerance in transgenic plants [ 30 ]. A screen of the sorghum MF dataset reveals five full-length DREB1-like proteins, based on conservation of the AP2 domain and a conserved C-terminal LWSY motif (see Materials and Methods ). A phylogenetic tree constructed from the AP2 domains of the Arabidopsis , rice, and sorghum DREB1-encoding genes suggests that sorghum has expanded the DREB1 family and that SbDREB1–1 and SbDREB1–2 are the closest orthologs to the Arabidopsis DREB1 family ( Figure 5 ). This analysis also suggests that the rice gene OsDREB1D may not belong to the DREB1 family, a hypothesis supported by the fact that OsDREB1D does not contain the conserved LWSY motif and its expression was not detected under drought, freezing, or salt-stress conditions [ 31 ]. An expansion of the DREB1 family in sorghum may contribute to the plant's enhanced drought resistance. Certainly the identification of other sorghum genes involved in the drought response regulatory pathway is now possible. This analysis highlights the utility of this dataset in answering fundamental comparative biology questions even at such a low level of gene space coverage. Figure 5 Phylogenetic Comparison of Sorghum DREB1 Genes A phylogenetic tree comparing the AP2 domain of the sorghum DREB1 genes to those of Arabidopsis and rice was constructed using CLUSTALX [ 61 ]. The genes encoding proteins from Arabidopsis are DREB1A, DREB1B, and DREB1C . Rice genes are OsDREB1A, OsDREB1B, OsDREB1C (nucleotides 142,337–142,981), and OsDREB1D (nucleotides 1,489–2,250). AP2 domains from other Arabidopsis proteins are also included: APETALA2 (R2 domain), AtERF-1, LEAFY PETIOLE, and TINY . Global Comparisons to Rice and Arabidopsis In order to assess the utility of the sorghum MF set for cross-genome annotations, we compared the annotation of rice by sorghum MF versus the complete gene set in Arabidopsis . The rice genes were downloaded from The Institute for Genomic Research (TIGR) and contain the genomic sequence of gene predictions, which includes exons and introns. The rice set contains 57,535 genes that we categorized into known (23,115), hypothetical (21,438), and repetitive (12,982), based on the annotation (see Materials and Methods ). The rice sequence was used as the query in searches of sorghum MF and Arabidopsis proteins. A rice gene was considered supported if it had a best match better than or equal to a BLAST E-value of 1 × 10 −8 . Of the rice gene set, 46,450 (81%) had a match to sorghum MF, while only 38,462 (67%) matched Arabidopsis . The matches can be further broken down by category, with 22,282 (96%) of known rice genes, 13,262 (62%) of hypothetical rice genes, and 10,906 (84%) rice repeats matched by sorghum MF. In comparison, Arabidopsis annotated 20,827 (90%) known, 7,850 (37%) hypothetical, and 9,785 (75%) repeats. Thus, the 1.15× coverage of the closely related sorghum gene space does a much better job of providing supporting evidence for gene predictions in rice than does Arabidopsis . Interestingly, the number of hypothetical genes matched by sorghum MF is almost 2-fold higher than that annotated by Arabidopsis . This may indicate a higher proportion of grass-specific genes in the hypothetical predictions. To understand how well cross-species gene annotation is accomplished in a low-redundancy MF versus a nearly complete genome, we compared the annotation of Arabidopsis by sorghum MF to that by rice. Such a comparison provides a good test of annotation capacity without being complicated by different evolutionary distances, since Arabidopsis, being dicotyledonous, is expected to be the same evolutionary distance from both sorghum and rice. An Arabidopsis protein was considered supported if it had a BLAST match less than or equal to an E-value of 1 × 10 −8 ( Figure 6 ). In this analysis, 19,700 (84%) of the known and 1,664 (38%) of the hypothetical proteins had a match to sorghum MF, whereas 21,093 (90%) of the known and 1,979 (45%) of the hypothetical proteins had a match to rice. This indicates, as expected, that a complete monocot genome is a better tool for annotating a dicot than is a partial genome; however, the difference is not that big, suggesting that a low level, cost-effective skim of many different genomes for comparative genomics may be more economical than complete sequencing. Figure 6 Annotation of Arabidopsis by Sorghum MF Versus Rice Gene Sequences Shown are the number of Arabidopsis proteins that are matched in a TBLASTN comparison to the sorghum MF set (blue) versus the rice gene sequences (yellow). The Arabidopsis proteins, after having known repetitive elements removed (see Materials and Methods ), have been categorized as either hypothetical or known based on the definition line. Arabidopsis proteins were considered supported if they matched with an E-value less than or equal to 1 × 10 −8 . Sb, S. bicolor MF set; Osj:seq, Oryza sativa japonica gene sequences. Interestingly, although the rice sequences match more Arabidopsis proteins than sorghum, the set is not completely overlapping, and sorghum matches 247 proteins that are not matched by the rice sequences. Since we used rice gene predictions as our database for comparison, it is likely that some of the Arabidopsis proteins are in the genome but are not annotated as genes. To address this possibility, we compared the 247 Arabidopsis proteins to the entire rice genome ( Oryza sativa japonica ) and found that 59 did indeed match to the bare gDNA versus the annotations, and therefore were not unique to the sorghum- Arabidopsis genomes. That left 188 proteins that may be conserved in sorghum and Arabidopsis, but not in rice. The O. s. japonica genome was sequenced by the BAC-by-BAC method [ 32 ], and it is likely that some regions are not represented in the BAC clones. Therefore, we compared these 188 to the O. s. indica genome, which was sequenced by whole-genome shotgun [ 33 ] and would have different biases than BAC-by-BAC. Again, a proportion (61) of these were found in the genome under our BLAST criteria, leaving 127 Arabidopsis proteins that are supported by sorghum but either missing from or significantly diverged in the current versions of the O. s. japonica and O. s. indica genomes ( Table S1 ). Laboratory experiments will be needed to confirm that these are truly missing from rice; if they are missing, they represent an interesting set of genes that could highlight previously unknown shared features between sorghum and Arabidopsis to the exclusion of rice. For example, the myb-related protein CAPRICE, a gene involved in root-hair cell development [ 34 , 35 ], was in this set, which may indicate a previously unknown conserved root development pathway in sorghum and Arabidopsis to the exclusion of rice. MiRNAs MiRNAs are a class of small RNAs that are important in gene regulation through recognition and cleavage of target mRNA. They are short sequences, usually 18–24 nucleotides in length, that match target genes and gene families, although usually imperfectly. Regulation is achieved through cleavage by the RNAi silencing complex. They are encoded by hairpin precursors that are processed in at least two steps by RNase III-domain ribonucleases related to Dicer . MiRNAs have been found in all eukaryotes surveyed and seem to be well conserved between plant species [ 36 , 37 , 38 ]. We downloaded 122 and 92 known rice and Arabidopsis miRNAs, respectively [ 39 ], and used them in a BLAST search against the sorghum MF set. Of these, 91 (75%) of the rice miRNAs and 44 (48%) of the Arabidopsis miRNAs had exact matches in the sorghum MF set ( Table 3 ). For comparison, the miRNAs were searched against the completed rice genome, sorghum ESTs, and maize MF + HC (high C 0 t ) assemblies, with 121, 16, and 88 of the rice miRNAs and 52, 10, and 46 of the Arabidopsis miRNAs matching, respectively. Table 3 MiRNA Content in Sorghum, Rice, and Maize a Although the 122 miRNAs were reported by Jones-Rhoades and colleagues [ 39 ] in the O. sativa japonica genome, we were not able to find a perfect match for MIR395f, although there are several nearly identical matches To ensure that these were authentic matches and not just due to chance, we performed a test with shuffled miRNA sequences, maintaining the nucleotide composition (see Materials and Methods ). None of the shuffled sequences matched any of the databases, indicating that the matches are authentic and not due to the small size or a biased nucleotide composition of the miRNAs. Additionally, precursor sequences surrounding these miRNAs could form hairpins ( Figure 7 and unpublished data), and were also matched by rice gDNA, indicating they are likely to encode the corresponding miRNA. Figure 7 Secondary Structure of Predicted MiRNAs Predicted hairpin secondary structure of miRNA MIR156a from rice and the newly discovered ortholog from sorghum. The 21-nucleotide MIR156a sequence is highlighted in red. We do not know a priori how many of the rice miRNAs would be expected to be conserved in sorghum, but we can assume that most, if not all, of the miRNAs conserved between Arabidopsis and rice would also be conserved between Arabidopsis and sorghum. Therefore, given that the rice genome is nearly complete, we expect to find the same 52 Arabidopsis miRNAs in sorghum, and we have identified 44 (85%). The eight that are missing may be present in the data but not identified because of sequencing errors; not yet sampled, as we expect only approximately 66% of the nucleotides to be present at this level of coverage; or some of these eight may represent miRNAs conserved in rice but lost in sorghum. Simple Sequence Repeats SSRs are stretches of DNA with simple sequence pattern repetitions, usually in the form of di-, tri-, or tetra-nucleotide expansions such as (CA)n, (CAG)n, or ( GATA)n. These stretches of DNA are useful for genetic marker analysis, because they are unstable and often are polymorphic between closely related individuals [ 40 , 41 ]. Overall, SSRs are enriched in MF sorghum sequences, 22,445 of 417,113 (5.4%), compared to UF, 335 of 17,276 (1.9%), indicating that most SSRs are unmethylated. GC-rich trinucleotide repeat (TNR) SSRs in plants have been shown to be preferentially associated with coding regions [ 42 , 43 ]. We observe an increase in the proportion of GC-rich TNRs to total TNRs in MF sequences, 6,464 of 8,957 (72%), compared with whole-genome shotgun, 63 of 129 (49%). This observation suggests that this collection of sorghum sequences is laden with new and publicly available molecular breeding and genetic mapping tools. The Sorghum Genome and Comparative Genomics The sequence of the sorghum gene space provides an excellent tool for comparative genomics [ 44 ]. Unlike maize, which it otherwise resembles, sorghum has not undergone recent genome duplications, although there is evidence for ancient duplications in most cereal genomes [ 45 ]. For this reason, sorghum and rice share a greater degree of colinearity than maize and rice [ 40 ], potentially facilitating mapping of quantitative traits across these three genomes, including drought resistance [ 46 ]. Sorghum is also a close relative of sugarcane ( Saccharum spp.), whose large and variable chromosome content makes genome sequencing impractical. The availability of a large number of sugarcane EST sequences [ 47 ] will enable comparison of these genomes to identify genes of potential agronomic value in this species as well. Such comparisons will extend even to the large collection of microsatellite SSR markers reported [ 41 ]. The sequence reported here is an important first step for these comparisons. The sorghum gene set present in the MF data is very nearly complete, as illustrated by the ability to annotate Arabidopsis nearly as well as the completed rice genome and by the ability to identify 95% of the genes from finished sorghum BACs. This was achieved with a minimal sequencing effort, which brings within reach the prospect of sequencing multiple strains of the same species. Such a feat is of critical importance in maize, in which inbred lines differ substantially in gene order and content [ 48 , 49 ]. Sequencing Large Plant Genomes Using MF A disadvantage of gene enrichment strategies, whether they are EST sequencing, high C 0 t selection, or MF, is that the recovered fragments are not positioned on the genome. Mapping has to be accomplished by either mapping the reads to a physical or genetic map or by combining the gene enrichment with an anchored clone map. MF reads are enriched for SSRs, which make good genetic markers and allow some reads to be placed on a genetic map. If a framework physical map of fingerprinted BAC clones exists, then MF can be easily integrated onto the physical map in three ways: PCR mapping to BAC pools, hybridization to BAC filters, and/or by sequence integration. Sequence integration can be accomplished using either BAC end sequence or shotgun sequence from a representative tiling path of the BAC contigs. While there is no whole genome BAC map of sorghum yet available, a robust map is almost complete in maize [ 50 , 51 ]. It is estimated that a BAC tile of maize will consist of approximately 18,000 BAC clones. Skim-sequencing from these clones at approximately 1× coverage, combined with a deep coverage through gene enrichment, are predicted to generate a high-quality sequence map for a fraction of the cost of whole genome sequencing [ 48 ]. BAC sequencing projects are ongoing for sorghum [ 40 ], which can use the MF reads in much the same way to enhance the BAC shotgun sequence and speed the completion of the genome. Materials and Methods MF library construction Seeds of S. bicolor ATX623, kindly provided by J. Osborne (NC+ Hybrids, Colwich, Kansas, United States), were germinated and grown in soil under growth chamber conditions. Then gDNA was purified from isolated nuclei of 1-mo-old leaves as described [ 52 ], except that OptiPrep (Axis-Shield PoC, Oslo, Norway) was used. Shearing of nuclear DNA was performed using either a nebulizer (Cis-Us, Bedford, Massachusetts, United States) or Hydroshear (GeneMachines, San Carlos, California, United States). Sheared fragments were end-repaired using a variety of enzymes including mungbean nuclease, T4 DNA polymerase, Klenow fragment, and T4 polynucleotide kinase. End-repaired fragments were size-selected on an agarose gel and DNA fragments ranging from 0.7 to 1.5 kb were extracted and ligated to dephosphorylated, HincII-digested pBC SK– vector (Stratagene, La Jolla, California, United States) which was used to construct both MF (GeneThresher technology; Orion Genomics, Saint Louis, Missori, United States) and UF libraries. Ligation reactions were transformed into mcrBC+ and mcrBC– strains of E. coli for generation of MF and UF libraries respectively. Recombinant clones were picked using Genetix Q-bot robot (Research Genetics, Carlsbad, California, United States) and stored individually in 384-well microtiter plates. Sequence data Two sources of MF sequencing reads were used. Out of 604,641 attempts at Orion Genomics, 532,150 were successful (accession numbers CL147592–CL197752 and CW020594–CW502582), 514,983 of which are considered of nuclear origin based on comparison with chloroplast, mitochondrial, viral, and bacterial databases. Additionally, we have included 36,825 sorghum MF reads previously generated at Cold Spring Harbor Laboratories (Cold Spring Harbor, New York, United States) (accession numbers CC058553–CC059980, BZ329127–BZ342789, BZ342901–BZ352342, BZ365856–BZ368372, BZ369686–BZ370012, BZ421595–BZ424357, BZ625682–BZ629992, and BZ779555–BZ781928). The sorghum UF sequences also came from two sources: Orion Genomics (1,819 reads) (accession numbers CW512190–CW514008) and the University of Oklahoma (15,889 reads) (NCBI TraceDB accession numbers TI566112507–TI566128395). The average read lengths were 600 bp and 550 bp for each class of reads, giving a total, raw nuclear dataset of approximately 330 Mb (MF) and 10.5 Mb (UF). The MF dataset was further collapsed by assembling overlapping read pairs to generate a set of independent sampling events comprising approximately 285 Mb. Database curation and FP calculation We have done a first pass definition-line curation of publicly available sequence databases to eliminate obvious transposon sequences that would hamper subsequent analyses by virtue of inflating the true “gene” content of the given database. The Arabidopsis protein set, which was used for the gene enrichment calculations and assessment of cross-genome annotation potential, was downloaded from the NCBI ( ftp://ftp.ncbi.nih.gov/genomes/Arabidopsis_thaliana/CHR_*/*.faa ). The files were dated 23 May 2003 and contained 28,581 sequences (12,112,846 total letters). Repeats were removed from this dataset if the definition line meets both of the following two criteria: (1) Matched the case-insensitive regular expression “/retro|mutator|transpos|reverse transcriptase|polyprotein|\bgag\b|BARE-1|athila/”, and (2) did not match “/\[.*retro.*\]|leucine|WD-repeat|WD repeat|WD40|WD-40| ankyrin|telomere|arm repeat|PPR-repeat|armadillo|tetratricopeptide|TPR-repeat|TPR repeat|Kelch|pentapeptide|C-repeat/”. This second step was used to replace falsely identified nonrepetitive elements. Removing repeats reduced the database size by 640 sequences to 27,941, which included 4,412 sequences identified as hypothetical by matching the definition line to the case-insensitive regular expression “/hypothetical/.” The rice sequence set was downloaded from TIGR ( ftp://ftp.tigr.org/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_2.0/all_chrs/all.seq ). The file was dated 30 April 2004 and contained 57,535 sequences (155,419,428 total bases). The rice sequence set contains the genomic regions for all predicted rice genes, which includes exons, introns, and untranslated regions where good evidence is provided. No sequences were removed from this database, but they were classified as “repeats,” “hypothetical,” and “known” by the following criteria. (1) Sequences were classified as repeats if the definition line matched a case-insensitive regular expression “/retro|transpos|reverse transcriptase|gag|polyprotein|mutator|maggy|rire|gypsy|copia|bare-1/”; (2) the sequences were classified as hypothetical if the definition line matched a case-insensitive regular expression “/hypothetical/”; and (3) the remaining sequences were classified as known. In total, there were 13,008 repeats, 21,441 hypotheticals, and 25,263 known proteins. The rice chromosomal genomic sequences were used for miRNA identification ( ftp://ftp.tigr.org/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_1.0/all_chrs/all.con ) and dated 05 September 2003. It contains 12 chromosomes, with sequences comprising 358,546,960 bases. Sorghum ESTs were download from the NCBI ( ftp://ftp.ncbi.nih.gov/genbank/*.seq.gz and ftp://ftp.ncbi.nih.gov/genbank/daily-nc/nc*.flat.gz , which was last dated 20 October 2003, and contains 161,766 sequences (83,411,684 total bases). No sequences were removed from this database. Gene enrichment was calculated by comparing the rate of gene discovery between MF and UF sequences. To ensure high quality, unique sampling events, reads were chosen that contained at least 100 contiguous Phred Q20 bases and only one read per clone. Detection of genes was accomplished by an NCBI-BLASTX search (parameters: -e 0.01; -b 5; -v 5) of the curated Arabidopsis protein database (see Materials and Methods ). Aside from the curation of the Arabidopsis database to remove repetitive elements, matches to proteins annotated as hypothetical were not counted. Hypothetical genes are often false gene predictions or unknown repetitive elements. In order to calculate a gene enrichment factor, or FP, the proportion of matches from MF sequences are compared to the proportion of matches in UF sequences over a range of E-values from 1 × 10 −5 to 1 × 10 −20 , such that all matches better than the given E-value are tabulated ( Table 1 ). For sorghum, the genome size is estimated at 735 Mb [ 2 ]. Dividing the genome size by the median 3.15 FP provides an estimate of a 233 Mb sampled space. BAC annotation There were 14 finished BAC clones at the time of analysis, with the following accession numbers (and GenInfo identifiers). AC120496.1 (GI:20486389), AF010283.1 (GI:2735839), AF061282.1 (GI:4539654), AF114171.1 (GI:4680196), AF124045.1 (GI:5410347), AF369906.1 (GI:19851516), AF466199.1 (GI:18390096), AF466200.1 (GI:18481699), AF466201.1 (GI:18483227), AF466204.1 (GI:18568251), AF503433.1 (GI:21326110), AF527807.1 (GI:22208458), AF527808.1 (GI:22208471), and AF527809.1 (GI:22208503). The BACs were manually annotated, then reads were mapped to the BACs by BLAST to determine the locations of hits relative to the genes. We analyzed the BACs with several computational tools in addition to manual editing. Repetitive elements were identified using RepeatMasker [ 53 ] with the MaskerAid speed enhancement [ 54 ] and the TIGR cereal repeat database. The TIGR cereal repeat database, dated 11 July 2003, was downloaded ( ftp://ftp.tigr.org/pub/data/TIGR_Plant_Repeats/ ) and contained 11,043 repeat entries. RepeatMasker was run with the following parameters: “-s; -w; -no_is; -nolow; -lib cereal_repbase.lib”. RepeatMasker parameters also included “-xsmall” to mask in lowercase and “-w” to use the MaskerAid [ 54 ] enhancement. To look for known protein-coding genes, we searched each repeat-masked BAC against all plant proteins with WU-BLASTX 2.0MP-Washu (23 May 2003) [ 55 , 56 ] using a serial strategy [ 57 ]. The first search used the parameters W=5; V=0; E=1e-5; X=10; nogap; kap; altscore=“* any na”; altscore=“any * na”; wordmask=seg; lcmask. The second search used default parameters. To look for transcript similarities, we searched all plant transcripts with WU-BLASTN using a serial strategy with the following first-round parameters: W=12; V=0; X=7. In the second round we used the parameter: W=9. Both BLASTN searches had these additional parameters: wordmask=seg; lcmask; M=1; N =–1; Q=3; R=3; kap; E=1e-10; hspmax=0. To look for potentially novel genes, we used Fgenesh ( http://www.softberry.com/berry.phtml ) with monocot parameters, Genscan [ 58 ] with Arabidopsis parameters, and SNAP [ 59 ] with Arabidopsis parameters. In order to annotate the locations of genes in each BAC, we loaded all the computational results into the ACEDB viewer ( http://www.acedb.org ) and edited gene structures by hand. One of the challenges was how to determine when the tools had identified pseudogenes. These are often marked by adjacent repeats, BLASTX alignments containing stop codons, or gene predictions with tiny introns to compensate for frame-shifts. Another challenge was how to use cross-species alignments. Alignments that are nearly identical to genomic sequence are useful for delimiting exon boundaries, but inexact matches pose problems because alignments may terminate because of real exon boundaries or differences between the sequences. Since most of the alignments were from plants other than S. bicolor, we did not employ any programs that align a protein or transcript directly to a genome. Instead, we assigned the position of the splice sites in part by consulting exon predictions, since gene finding algorithms contain probabilistic models of splice sites. We did not report any raw gene predictions. However, some genes do contain exons with no overlapping evidence and are included in the gene structure because they complete an otherwise incomplete gene structure and in some cases are necessary to maintain the reading frame. The BAC annotations are available in GFF format with the supplemental online data. The sorghum MF sequences were compared to the collection of 14 sorghum BACs using NCBI-BLASTN (parameters: -p blastn; -F ‘m D'; -e 0.01; -b 14; -v 14). A sequence was considered mapped to a BAC if the match was over 90% of the read with 98% identity or higher. A single read was mapped to only one location. A gene was considered tagged if one or more MF sequence(s) overlapped the CDS region by 50 bases or more. The set of S. bicolor ESTs were mapped to the BACs using the same BLAST parameters, but a gene was considered tagged using less stringent criteria, since genomic introns will not align. A gene is tagged by an EST if it aligns at 98% identity over at least 50 bp, but there was no requirement for the percentage of the EST that needed to be aligned. Arabidopsis simulation A computational simulation of shotgun sequencing Arabidopsis chromosome 1 was compared to the empirical gene tagging results in sorghum. The sequence and annotation of Arabidopsis chromosome 1 was downloaded from TIGR ( ftp://ftp.tigr.org/pub/data/a_thaliana/ath1 ) on 20 February 2004. The chromosome is approximately 30 Mb long with 7,520 genes annotated. The median gene size is 1,960 bp. Computationally generated “reads” of 700 bp in length were created from the chromosome for different levels of raw coverage from 0.5× up to 3.5×. The reads were then mapped back to the chromosome annotation to determine the percentage of the 7,520 genes that were tagged at each level of raw coverage (results shown in Figure 2 ). The percent gene tagging was calculated on a fold-coverage basis (e.g., 0.5×, 1.0×, etc.), so in order to convert it to a meaningful read number basis for Figure 2 , we converted the fold-coverage to a number of reads by using the estimated genome space (247 Mb) divided by the average sorghum read size (604 bp), resulting in approximately 409,000 reads per 1× coverage. MiRNA analysis The A. thaliana and O. sativa miRNAs were downloaded from the supplementary online material for Jones-Rhoades and colleagues [ 39 ]. This dataset contains 122 and 92 computationally predicted and experimentally confirmed miRNAs for Oryza sativa and Arabidopsis thaliana , respectively. The miRNAs are grouped into 18 and 22 families for rice and Arabidopsis , respectively. These sequences were used in a WU-BLASTN [ 55 ] search of the MF sorghum set (parameters: -W 18; -M 1; -N −4; -Q 1; -R 1; -wordmask=seg; -warnings). A match was scored if the miRNA matched at 100% identity over its complete length. The same parameters were used for the rice genome, sorghum ESTs, and maize MF + HC databases. The maize MF + HC database is release 4.0 of the Zea mays MF and HC combined assembly from TIGR ( http://www.tigr.org/tdb/tgi/maize/ ). In order to test the specificity of these miRNA matches, we generated shuffled sequences for the 122 rice and 92 Arabidopsis miRNAs. The shuffling maintains the nucleotide composition of each while scrambling the order [ 60 ]. The shuffled sequences were used in WU-BLAST searches against all the databases with the same parameters as above. None of the shuffled sequences had an identical match to any database. These results indicate that the miRNAs are not matching simply because of their small size and nucleotide composition, but probably represent authentic evolutionarily conserved units. Comparison with rice and Arabidopsis . The rice sequences were compared to the sorghum MF using NCBI-BLASTN with the rice sequences as the query and the sorghum MF reads as the database (parameters: -p blastn; -b 1500; -v 1500; -r 1; -q -1; -G 2; -E 1; -F ‘mD'; -e 1e-5). We counted a rice gene as hit if the E-value was less than or equal to 1 × 10 −8 , which corresponds to a bit score of approximately 61. The rice hits were then counted and categorized. To assess how well rice is annotated by a dicot, the rice sequences were also searched against the Arabidopsis protein set using NCBI-BLASTX (parameters: -p blastx; -e 1e-5; -F ‘mS'). We counted a rice gene as hit if the E-value was less than or equal to 1 × 10 −8 , which corresponds to a bit score of approximately 51. The rice genes hit were counted and categorized. The Arabidopsis protein set was compared the sorghum MF dataset using NCBI-TBLASTN (parameters: -p tblastn; -e 1e-5; -F ‘mS'). We counted an Arabidopsis protein as hit if the E-value was less than or equal to 1 × 10 −8 , which corresponded to a bit score of approximately 57. The Arabidopsis hits were then counted and categorized as shown in Figure 6 . The Arabidopsis protein set was also compared to the rice gene sequence dataset using NCBI-TBLASTN (parameters: -p tblastn; -e 1e-5; -F ‘mS'). We counted an Arabidopsis protein as hit if the E-value was less than or equal to 1 × 10 −8 , which corresponded to a bit score of at least 57. The Arabidopsis hits were then counted and categorized as shown in Figure 6 . There were 247 Arabidopsis proteins that were annotated by sorghum MF but not rice sequence. These 247 proteins were then compared to the entire rice genome using NCBI-TBLASTN (parameters: -p tblastn; -e 1e-5; -F ‘mS'). From that set of 247 we removed any Arabidopsis proteins if the E-value was less than or equal to 1 × 10 −8 , which corresponded to a bit score of at least 59. The remaining 188 proteins were then compared to the entire genome from the O. s. indica cultivar of rice [ 33 ] using NCBI-TBLASTN (parameters: -p tblastn; -e 1e-5; -F ‘mS'). From that set of 188 we removed any Arabidopsis proteins if the E-value was less than or equal to 1 × 10 −8 , which corresponded to a bit score of approximately 60. The resulting set contained 127 Arabidopsis proteins that were supported by sorghum MF but not found in the rice genomes. Methylation analysis Methylation was assessed using MethylScreen analysis, which is a real-time PCR technique that reports DNA methylation occupancy information for genomic markers through enzymatic interrogation. MethylScreen analysis compares the cycle thresholds (Cts) of gDNA that has been subjected to various treatments and infers 5′ methylated cytosine (5 mC) occupancy through the changes in Ct mediated by the treatments. The Ct of any locus is a function of the number of copies present within the assay tube. MethylScreen analysis relies upon the simple formula that total gene copies = number of methylated copies + number of unmethylated copies in every sample. Typical sample assays utilize four sample subportions, the first portion of a sample is mock-digested, reporting total copies present. A second (and equal) portion is treated with a methylation-sensitive restriction enzyme (MSRE), reporting the number of gene copies that are methylated. The third portion is digested with a methylation-dependent restriction enzyme (MDRE) such as mcrBC, reporting the number of copies that are unmethylated. The fourth reaction is doubly-digested with both the MSRE and MDRE. When working with relatively pure samples, methylated loci have a Ct from MSRE that is the same as the untreated control, and the Ct obtained from the MDRE is greater. Conversely, unmethylated loci have a MDRE Ct that is identical to untreated and a greater Ct in the MSRE reaction. The gene tb2 targeted a 263 bp region from the 5′ end of the gene for the assay. There are four HhaI restriction sites and more than 25 possible mcrBC half-sites (5′-RC-3′) in this region ( Figure 4 A). We developed a SYBR green real-time PCR assay using the Dynamo Kit from MJ Research (Boston, Massachusetts, United States). The forward primer used was 5′- GCCGCCGCCGACGCCAGCTTTCAC-3′, and the reverse primer was 5′- ATCCCGGGCGCGGTGCATATCTTGCTGTG-3′. The cycling parameters were 95 °C for 3 min, followed by 50 cycles of two-step PCR: 95 °C for 30 s and 70 °C for 30 s. We utilized both a low-temperature (70 °C) and a high-temperature (82 °C) plate read. 2 μg of gDNA was added to a 200-μl reaction cocktail for digestion using the conditions specified by NEB (Beverly, Massachusetts, United States). Half of the sample was digested with 40 U of HhaI overnight, while the other half remained mock-digested. Both “digests” were subsequently split in two, and to each new digestion, cocktails with NEB2, BSA, and 2×GTP were prepared using a final volume of 100 μl. 40 U of mcrBC was added to one of the mock-digested samples and to one of the HhaI-digested samples. All four reactions were incubated overnight at 37 °C. The PCR assays utilized approximately 40 ng from each of the digests. All amplifications were performed in quadruplicate. A standard dilution curve of S. bicolor gDNA in 1× NEB2 was used to ensure linearity of the system. All reactions were verified using melt-curve analysis. Three replicate analyses were performed (digestions and cycling). Each of the 11 genes was broken into approximately 1.5-kb pieces, which were aligned to create a consensus kafirin assembly ( Figure 4 B). The consensus kafirin sequence was examined and a 247-bp region was selected. The forward primer was 5′- CTCCTTGCGCTCC TTGCTCTTTC-3′, (where GCGC is a HhaI restriction site) and the reverse primer was 5′- GCTGCGCTGCGATGGTCTGT-3′. We used the same SYBR green real-time PCR assay with the Dynamo Kit (MJ Research), as mentioned above for the tb2 gene. Cycling parameters were 95 °C for 3 min, followed by 50 cycles of two-step PCR: 95 °C for 30 s and 56 °C for 30 s. We utilized both a low-temperature (70 °C) and a high-temperature (82 °C) plate read. The input of gDNA was cut to 10 ng per reaction. All amplifications were performed in quadruplicate. Three replicate analyses were performed (digestions and cycling). The threshold was set using a template dilution standard control. For the kafirin genes, the average difference in Ct between the mcrBC single and the HhaI + McrBC double-digests is 2.46 cycles (22.08 ± 0.34 Hha I + McrBC - 19.62 ± 0.19 McrBC). PCR products from the kafirin cycling reactions were cloned using the topoisomerase-assisted method (Invitrogen, Carlsbad, California, United States). Libraries of insert-bearing clones were generated using standard techniques. From each library, 200 lacZ-negative clones were selected for characterization. The clones were sequenced with a single read using the M13 priming site on the pCR2.1 plasmid (Invitrogen). All seven subfamilies were discovered from both the treated and untreated genomic samples (unpublished data), indicating that all 11 genes were amplified and recoverable, even in the mcrBC-digested fractions. Identification of DREB1 orthologs in the sorghum dataset The five Arabidopsis DREB genes DREB1A, DREB1B, DREB1C, DREB2A, and DREB2B were used in a TBLASTN search of an assembly the sorghum dataset using WU-BLAST (parameters: E =e-5; matrix=BLOSUM80; topcomboN=1; wordmask=seg+xnu). Matches to the sorghum assembly with an E-value of 1 × 10 −8 or less were analyzed with FGENESH (monocot) to select assemblies with a full-length protein. Out of 67 full length proteins identified in this manner, five sorghum proteins were identified as DREB1 genes based on conservation of the AP2 domain and a conserved C-terminal motif, LWSY [ 31 ]. Supporting Information Table S1 Arabidopsis Proteins with Homologs in Sorghum but Not Rice Shown is a list of 127 Arabidopsis proteins that have matches to the sorghum MF set at a TBLASTN E-value less than or equal to 1 × 10 −8 , but are not found in the O. s. japonica or O. s. indica genomes at the same cutoff. (120 KB DOC). Click here for additional data file. Accession Numbers The sorghum MF sequence set is deposited in the Genome Survey Sequence division of GenBank ( http://www.ncbi.nlm.nih.gov/ ). On 6 January 2004, Orion deposited 50,161 of the sequences under accession numbers CL147592–CL197752. The 36,825 Cold Spring Harbor Laboratories MF sequences were previously deposited in GenBank under the accession numbers CC058553–CC059980, BZ329127–BZ342789, BZ342901–BZ352342, BZ365856–BZ368372, BZ369686–BZ370012, BZ421595–BZ424357, BZ625682–BZ629992, and BZ779555–BZ78192. The remaining 481,989 MF sequences from Orion Genomics are deposited in GenBank under accession numbers CL147592–CL197752 and CW020594–CW502582. The Orion UF sequences are deposited in GenBank's Genome Survey Sequence under accession numbers CW512190–CW514008. The University of Oklahoma UF sequences are deposited in the NCBI trace archive under accession numbers TI566112507–TI566128395. GenBank accession numbers for other genes discussed in this paper are sorghum Cs1 (AF206660); Arabidopsis DREB genes DREB1A (Q9M0L0), DREB1B (P93835), DREB1C (Q9SYS6), DREB2A (O82132), and DREB2B (O82133). Genbank accession numbers for BAC clones (with GenInfo identifiers) are AC120496.1 (GI:20486389), AF010283.1 (GI:2735839), AF061282.1 (GI:4539654), AF114171.1 (GI:4680196), AF124045.1 (GI:5410347), AF369906.1 (GI:19851516), AF466199.1 (GI:18390096), AF466200.1 (GI:18481699), AF466201.1 (GI:18483227), AF466204.1 (GI:18568251), AF503433.1 (GI:21326110), AF527807.1 (GI:22208458), AF527808.1 (GI:22208471), and AF527809.1 (GI:22208503). The GenBank accession number for the protein CAPRICE is NP_182164. Accession numbers for genes used in phylogenetic analysis of sorghum DREB are as follows. Rice genes are OsDREB1A (AF300970), OsDREB1B (AF300972), OsDREB1C (AP001168, nucleotides 142337–142981), and OsDREB1D (AB023482, nucleotides1489–2250); AP2 domains from other Arabidopsis proteins are also included: APETALA2 (R2 domain, accession number P47927), AtERF-1 (BAA32418), LEAFY PETIOLE (AAF32292), and TINY (Q39127).

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539327.xml

374242

Artificial Prions Created from Portable Control Elements

For decades, scientists accepted that the nucleic acids, DNA and RNA, packed with thousands of protein-coding genes, were the sole purveyors of genetic information; all inherited traits, from eye color to shoe size, must be stored and expressed through nucleic acid mechanisms. But prions are an exception. These misshapen proteins are capable of growing, replicating, and infecting other cells—that is, they are heritable. And all without a scrap of DNA. Most famous as the culprit behind bovine spongiform encephalopathy, or mad cow disease, prions also occur naturally in some organisms and may play important roles in their growth and development. Prion-forming proteins normally exist as benign cellular components, such as enzymes or receptors. But they possess the innate ability to alter their three-dimensional structure, or fold, which changes their function and makes them almost impossible to destroy. Like other misfolded proteins, such as those responsible for Alzheimer's and Huntington's diseases, prions pack together and form aggregates. But what distinguishes prions from simple protein aggregates is their exponential growth and amplification, which allows them to infect new host cells. Prions grow by inducing normal proteins to alter their shape and adhere to an initial aggregate “seed.” These growing masses are then thought to divide with the help of “chaperones,” cellular proteins that aid in protein folding and transport, resulting in smaller prion particles called propagons. The propagons are then distributed to both mother and daughter cells during division, thereby infecting the next generation of cells. Though this theory of the prion life cycle was proposed a few years ago, scientists are still working out the underlying molecular mechanisms As they report in this issue, Lev Osherovich and colleagues dissected yeast prions and found that growth and heritability are controlled by two independent and “portable” sequences. Furthermore, the heritability element seems to be the only thing that keeps slow growing protein aggregates from becoming infectious prions. Previous research showed that one end of the yeast protein, Sup35p, is critical for turning this normal housekeeping enzyme into a prion. The “prion-forming domain” of Sup35p consists of two segments: one stretch rich in the amino acids glutamine and asparagine and another made up of several, small series of amino acids, called oligopeptides. Osherovich and colleagues had earlier found another yeast protein, New1p, which had similar segments, though in reverse order. To study the function of these sequences, the team constructed several strains of yeast, each with a small part of the prion-forming domain missing. By watching the behavior of these modified proteins, each fused to a green fluorescent protein for easy observation, the authors could infer the roles of the deleted segment. For both Sup35p and New1p, the authors found that the area rich in glutamine and asparagine was responsible for the aggregation and growth of prions—acting like a patch of Velcro that locks the misshapen proteins together. While this had been suggested by previous research, the authors also found that this sticky sequence only adheres to proteins that mirror its own pattern of amino acids, thereby explaining why prions from different species don't often interact, a phenomenon called the species barrier. The stretch of oligopeptide repeats in Sup35p and New1p, however, was required for the inheritance of prions—the proper division of prion masses and subsequent distribution of propagons during cell division. The authors suggest that oligopeptide repeats function as a secure binding location for the chaperone proteins, which are necessary for heritability, and thus infectiousness, of prions. Their results also help to explain why stable inheritance of prions is rare; while many proteins have stretches of amino acids similar to the described aggregation sequence, few also contain sequences like oligopeptide repeats that permit inheritance. Though both the aggregation sequence and the oligopeptide repeats are required for prion growth and infection, the segments seemed to function completely separately, allowing the authors to create a synthetic prion-forming domain by combining the aggregation element of New1p with the Sup35p replication/heritability element. This artificial prion acted like New1p, again showing that it is the sticky, aggregation element that specifies which proteins will be added to the growing prion mass. Osherovich and colleagues then went on to create another artificial prion by fusing the oligopeptide repeats to an expanded polyglutamine tract, the type of aggregation sequence responsible for the toxic buildup of brain proteins in Huntington's disease. With this simple addition, the slow growing aggregate was transformed into a heritable, infectious prion. By creating artificial hybrid prions, Osherovich and colleagues showed that the two discrete elements of prion-forming domains are portable and work together regardless of their origins. The authors suggest that other artificial prions could be used as a model system to study different types of aggregation sequences, such as those found in the human prion protein responsible for Creutzfeldt-Jakob's disease or the misshapen plaques of proteins that contribute to Alzheimer's disease.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC374242.xml

555564

Regulation of prokineticin 2 expression by light and the circadian clock

Background The suprachiasmatic nucleus (SCN) contains the master circadian clock that regulates daily rhythms of many physiological and behavioural processes in mammals. Previously we have shown that prokineticin 2 ( PK2 ) is a clock-controlled gene that may function as a critical SCN output molecule responsible for circadian locomotor rhythms. As light is the principal zeitgeber that entrains the circadian oscillator, and PK2 expression is responsive to nocturnal light pulses, we further investigated the effects of light on the molecular rhythm of PK2 in the SCN. In particular, we examined how PK2 responds to shifts of light/dark cycles and changes in photoperiod. We also investigated which photoreceptors are responsible for the light-induced PK2 expression in the SCN. To determine whether light requires an intact functional circadian pacemaker to regulate PK2 , we examined PK2 expression in cryptochrome1,2-deficient ( Cry1-/-Cry2-/- ) mice that lack functional circadian clock under normal light/dark cycles and constant darkness. Results Upon abrupt shifts of the light/dark cycle, PK2 expression exhibits transients in response to phase advances but rapidly entrains to phase delays. Photoperiod studies indicate that PK2 responds differentially to changes in light period. Although the phase of PK2 expression expands as the light period increases, decreasing light period does not further condense the phase of PK2 expression. Genetic knockout studies revealed that functional melanopsin and rod-cone photoreceptive systems are required for the light-inducibility of PK2 . In Cry1-/-Cry2-/- mice that lack a functional circadian clock, a low amplitude PK2 rhythm is detected under light/dark conditions, but not in constant darkness. This suggests that light can directly regulate PK2 expression in the SCN. Conclusion These data demonstrate that the molecular rhythm of PK2 in the SCN is regulated by both the circadian clock and light. PK2 is predominantly controlled by the endogenous circadian clock, while light plays a modulatory role. The Cry1-/-Cry2-/- mice studies reveal a light-driven PK2 rhythm, indicating that light can induce PK2 expression independent of the circadian oscillator. The light inducibility of PK2 suggests that in addition to its role in clock-driven rhythms of locomotor behaviour, PK2 may also participate in the photic entrainment of circadian locomotor rhythms.

Background Light is the principal zeitgeber that entrains circadian rhythms of physiology and behaviour [ 1 , 2 ]. The major light input pathway to the suprachiasmatic nucleus (SCN) is the retinohypothalamic tract [ 3 ], which arises from a population of retinal ganglion cells [ 4 ]. Recent studies have demonstrated that melanopsin-containing retinal ganglion cells, rods, and cones all convey photic information to the SCN, and mice lacking these photoreceptive systems cannot be entrained by light [ 5 - 11 ]. Excellent progress has been made in the understanding of circadian photic entrainment [ 12 - 15 ]. This includes light-induced transcriptional activation of core clock genes in the SCN, such as Per1 and Per2 , as well as immediate-early gene c-fos . Exposure to light pulses at night induces expression of these genes in the SCN, and this light induction mechanism has been suggested as a critical pathway for the resetting of circadian clock in response to changes in light/dark conditions [ 16 - 19 ]. Intercellular signalling mechanisms between SCN neurons are also important in circadian photic entrainment, as mice with mutation in a neuropeptide receptor for VIP (Vasoactive Intestinal Peptide) and PACAP (Pituitary Adenylate Cyclase Activating Peptide) are unable to sustain normal circadian behaviour and exhibit loss of sensitivity to light [ 20 ]. In addition to the effect of light on circadian entrainment, light also has a direct effect on physiology and behaviour, generally termed as "masking" [ 21 , 22 ]. For instance, light pulses given at night acutely suppress the locomotor behaviour of nocturnal rodents [ 21 , 22 ], and this can occur without functional clockwork [ 23 - 27 ]. Masking may account for the maintenance under normal light/dark conditions of wheel-running rhythms in cryptochrome-deficient ( Cry1-/-Cry2-/- ) mice, which are behaviourally arrhythmic under constant darkness. The contribution of masking to normal locomotor activity rhythms is unclear, as is the participation of the SCN in masking effects of light. Vitaterna et al (1999) first observed a light-driven Per2 rhythm in the SCN in Cry1-/-Cry2-/- mice, and have suggested that the light-driven molecular rhythm in the SCN may be related to the preservation of their locomotor rhythm [ 25 ]. We previously found that prokineticin 2 ( PK2 ) is a first order clock-controlled gene, whose expression in the SCN is regulated by CLOCK and BMAL1 acting on the E-boxes in the gene's promoter [ 28 ]. We have also demonstrated that PK2 may function as a SCN output molecule that transmits circadian locomotor rhythm via activation of a G protein-coupled receptor [ 28 , 29 ]. Interestingly, we also observed that PK2 expression is rapidly induced by light pulses administered at night [ 28 ], a characteristic that is usually seen with core clockwork genes but not clock-controlled genes. Here we further investigated the light regulation of the rhythm of PK2 expression in the SCN. In particular, we investigated the photoreceptive mechanisms responsible for the light-induced PK2 expression in the SCN. Utilizing Cry1-/-Cry2-/- mice, we also determined whether light can drive PK2 expression in the SCN independent of a functional circadian clock. Results PK2 responds differentially to the delay and advance of light/dark cycles We first examined the effects of abrupt shifts of light/dark cycles on PK2 mRNA rhythm in the SCN. Animals were first entrained for two weeks under 12 hour light: 12 hour dark (LD), then subjected to either a 6 hour delay (6hrD) shift or 6 hour advance (6hrA) shift of light/dark cycles. We measured PK2 mRNA in the SCN of these animals to examine how quickly the PK2 mRNA rhythm re-entrains to the shifted light/dark cycles. Under LD, PK2 mRNA peaks during the day and remains low or undetectable during the night. During the first cycle of the delayed shift (6hrD), the PK2 mRNA rhythm responds quickly: the rising phase of PK2 expression adjusts rapidly to the delayed light/dark cycles, while the falling phase still resembles that of the unshifted light/dark cycles (Figure 1A ). In contrast, the PK2 mRNA rhythm responds very little to a 6 hour advance shift (6hrA). During the first cycle of the advance shift, the PK2 oscillation pattern remains similar to that of the unshifted LD (Figure 1B ). These changes in PK2 expression during 6hrD or 6hrA shift indicate that the endogenous circadian clock exerts dominant control over the PK2 rhythm, as PK2 expression cannot respond immediately and completely to the shifts of light/dark cycles. As it normally takes about 1–2 days for locomotor rhythms to stably entrain to phase delays and about 5–6 days to entrain to phase advances [ 30 , 31 ], we next examined the timecourse of shifts of the PK2 rhythm to 6 hour phase advances and delays. Consistent with the animal's locomotor behaviour, the PK2 mRNA rhythm reaches stable phase within 2 days of 6hrD shift (Figure 1C ). In contrast, only the rise of PK2 reaches stable phase within 2 days of 6hrA shift, while the fall of PK2 takes longer (Figure 1D ). Thus, we further examined whether the PK2 rhythm is stably entrained after 6 days of 6hrA shift. As expected, the PK2 rhythm is completely entrained to 6hrA shift after 6 days (Figure 1D ). Together, the differential responses of PK2 rhythm to a 6hrD or 6hrA shift indicate that the endogenous circadian clock predominantly controls PK2 rhythm, as circadian oscillators typically show rapid phase delays but advance with transients [ 31 , 32 ]. The entrainment patterns of PK2 during phase shifts are consistent with behavioural studies in animals and human subjects [ 30 , 31 ]. PK2 rhythm is entrained by different photoperiods We next examined the effect of photoperiod on the PK2 molecular rhythm in the SCN. PK2 mRNA was measured in the SCN of mice entrained under different photoperiods: 8 hour light: 16 hour dark (8L:16D), 16 hour light: 8 hour dark (16L:8D), or 20 hour light: 4 hour dark (20L:4D). During 12L:12D, PK2 mRNA is highly expressed during the 12 hour light phase with peak level at ZT4 (Figure 1A , Figure 3A ). Under 16L:8D, PK2 mRNA expands to the entire 16 hour light phase and is essentially undetectable during the 8 hour dark period (Figure 2B ). However, the expression of PK2 mRNA is not confined to the light phase of the shorter photoperiod (8L:16D), as PK2 mRNA rises before lights on and persists after lights off (Figure 2A ). The temporal profile of PK2 mRNA under this short photoperiod (8L:16D) is very similar to that observed under 12L:12D (Figure 1A , Figure 3A ) or constant darkness (2DD) [ 28 ]. Thus, although light can induce PK2 mRNA and expand the duration of PK2 expression, the phase angle of PK2 expression is determined by the circadian clock, and its duration cannot be further compressed under shorter photoperiods. Interestingly, the peak of PK2 mRNA expression was significantly higher in long days (16L:8D) than in shorter days (8L:16D) (Figure 2A–B ), further indicate the enhancing effect of light on PK2 expression. However, a significant reduction in the PK2 peak level is observed under a very long photoperiod (20L: 4D) (Figure 2C ). We also noticed that under 20L:4D, PK2 mRNA is further expanded and becomes detectable even in dark phase (Figure 2C ). Under this long photoperiod (20L:4D), the difference between the peak and basal level of PK2 is only about 4 fold (Figure 2C ). As it has been reported that the rhythms of mPer1 and mPer2 mRNAs in the SCN are also entrained with different phase angles under a variety of photoperiods [ 33 - 35 ], we have also examined Per1 and Per2 rhythm in our photoperiod studies (see Additional file 1 ). The Per1 and Per2 rhythm we observed under these photoperiods are consistent with previous findings [ 35 ]. Taken together, these results indicate that changes in photoperiod alter PK2 rhythm in the SCN, and the amplitudes of PK2 mRNA oscillation are greatly reduced in very long photoperiods. Light inducibility of PK2 is eliminated in mice that lack melanopsin, rod and cone phototransduction system ( Opn4-/-, Gnat1-/- Cnga3-/- mice) As melanopsin has been implicated in circadian photoreception [ 5 - 11 ], we examined whether the PK2 molecular rhythm is normally entrained in melanopsin-deficient ( Opn4-/- ) mice. Figure 3 shows that the oscillation profile of PK2 in the SCN of Opn4-/- mice is essentially identical to that observed in the wild type mice under LD. This normal temporal profile of PK2 mRNA corresponds with the normal locomotor rhythm of Opn4-/- mice under light/dark conditions [ 7 , 8 ]. As Opn4-/- mice display attenuated phase resetting in response to light pulses and exhibit impaired light masking responses to bright light [ 36 ], we also examined whether light inducibility of PK2 is blunted in Opn4-/- mice. Figure 3B shows that light pulse-induced PK2 in the SCN of Opn4-/- mice was significantly reduced by about 50% and 60%, one and two hours after the light pulse, respectively. The Opn4-/- light pulse studies show that a residual PK2 expression is still present after a light pulse, suggesting that without melanopsin, other phototransduction system can still transmit light information to induce PK2 expression. Thus, we decided to examine the light inducibility of PK2 in triple knockout mice lacking melanopsin, rod and cone phototransduction systems ( Opn4-/- Gnat1-/- Cnga3-/- mice), as these animals free run under light dark conditions (LD) and lack masking responses to light [ 10 ]. Figure 3C shows that the light pulse-induced PK2 in the SCN is completely eliminated in these triple knockout mice, consistent with their malfunctioning photoentrainment systems and their lack of masking responses to light [ 10 ]. In addition, we also observed that PK2 mRNA followed the free-running locomotor rhythms in these triple knockout mice (Figure 3D ), with high levels of PK2 during the inactive phase (CT3) and low levels during active phase (CT15). Together, these results suggest that melanopsin contributes to the light inducibility of PK2 , and intact melanopsin with functional rod/cone phototransduction systems are required for the light inducibility of PK2 . A low amplitude PK2 rhythm is preserved in cryptochrome-deficient ( Cry1-/-Cry2-/- ) mice under light/dark conditions Previous studies have shown that the light-regulated Per2 rhythm is maintained in the SCN of cryptochrome-deficient ( Cry1-/-Cry2-/- ) mice that lack functional circadian clock [ 25 , 37 ]. In order to determine whether the regulation of PK2 , Per1 , Per2 and Bmal1 expression by light requires an intact circadian pacemaker, we systematically assessed the temporal mRNA profiles of clockwork genes in Cry1-/-Cry2-/- mice under both light/dark (LD) and constant dark (DD) conditions. Figure 4 shows that the molecular rhythm of Per2 remained largely intact in Cry1-/-Cry2-/- mice entrained under12L:12D, with levels about 4-fold higher during the light phase than the dark phase. This amplitude of the Per2 oscillation profile was similar to that observed in wild type mice [ 18 , 38 ]. A low amplitude Per1 rhythm in Cry1-/-Cry2-/ mice was also apparent under LD, but not DD (Figure 4B ). We further detected a light-driven Bmal1 rhythm in the SCN of Cry1-/-Cry2-/- mice under LD, but not DD (Figure 4C ). Interestingly, this Bmal1 rhythm in Cry1-/-Cry2-/- mice peaked during light phase, opposite from the Bmal1 rhythm in wild type mice and in phase with Per1 [ 39 , 40 ]. As it has been suggested that PER2 can positively regulate Bmal1 expression via inhibition of the orphan nuclear receptor REV-ERBα [ 41 , 42 ], it is possible that this Bmal1 rhythm is secondary to the light-driven Per2 rhythm. Further studies are required to clarify this observation. We also examined the molecular rhythm of PK2 in Cry1-/-Cry2-/- mice. Figure 4D shows that PK2 mRNA rhythm in the SCN of Cry1-/-Cry2-/- mice was apparent under LD, with the presence of a low level PK2 during light phase and absence of PK2 during dark phase (see Additional file 2 ). Similar to wild type mice, the peak level of this low amplitude PK2 rhythm was around ZT4, although its peak was only about 8% of that observed in wild type mice (Figure. 4D , Figure 1A , Figure 3A ). No PK2 rhythm was evident when Cry1-/-Cry2-/- mice were placed under DD (Figure 4D ). Furthermore, the inducibility of PK2 to nocturnal light pulses is also maintained in Cry1-/-Cry2-/- mice. PK2 mRNA increased one and two hours after a brief light pulse at ZT14 (Figure 4E ). Nevertheless, light-induced PK2 was still detected in Per1,2,3-/- mice and Clk-/- mice that lack functional circadian clock (Cheng, Weaver & Zhou, unpublished observations). As PK2 remains responsive to light in these clock mutant mice that lack functional circadian clock, it is likely that the low amplitude PK2 rhythm in Cry1-/-Cry2-/- mice under LD is directly driven by light. In order to test whether this light-driven PK2 rhythm may be related to the maintenance of behavioural rhythms observed in Cry1-/-Cry2-/- mice under LD, we studied the responses of Cry1-/-Cry2-/- mice to a 6 hour advance of lighting schedule. In contrast to the transients of entrainment of locomotor rhythms in wild type mice (which takes about 4-5 days to re-entrain to phase advance), the locomotor activity of Cry1-/-Cry2-/- mice adjusted rapidly to 6 hr advance (Figure 4F ). Such a rapid response is characteristic of masking. A correlative rapid adjustment of PK2 was also observed in the SCN of Cry1-/-Cry2-/- mice (Figure 4G ). As Cry1-/-Cry2-/- mice lack functional circadian clock and their locomotor behaviour and PK2 expression patterns are completely light driven, our results suggest that this low amplitude, light-driven rhythm of PK2 may contribute to or underlie the masking of locomotor behaviour in these animals. Discussion Our studies indicate that the molecular rhythm of PK2 in the SCN is predominantly controlled by the circadian clock, with light playing a modulatory role. Abrupt shifts of light/dark cycles significantly altered the phase of the PK2 rhythm. While PK2 expression re-entrained rapidly to phase delays, it takes several cycles of transients for PK2 to be stably entrained to phase advances (Figure 1 ). The rate of re-entrainment of PK2 molecular rhythms to these shifts is consistent with that of behavioural adaptation of animals and human subjects [ 30 , 31 ]. Our photoperiod studies indicate that PK2 expression in the SCN responds differentially to changes in photoperiod length (Figure 2 ). Although increasing light period can induce PK2 expression and expand the duration of PK2 rhythm (Figure 2B ), shortening of the light period does not lead to corresponding reduction of the duration of PK2 expression (Figure 2A ). It appears that a minimal duration of PK2 expression is maintained under short photoperiod (Figure 2A ) and constant darkness [ 28 ], which further indicate the dominant control of PK2 expression by the circadian clock. Interestingly, the amplitude of the PK2 oscillation was greatly reduced under very long photoperiod (20L:4D) (Figure 2C ). As the amplitude of both Per1 and Per2 rhythms were also reduced during 20L:4D (see Additional file 1 ), it is likely that these depressed rhythms of clockwork genes may contribute to the depressed PK2 rhythm observed. Whether reduction in the amplitude of expression in any of these genes is related to arrhythmicity in LL deserves further examination. Our studies with Cry1-/-Cry2-/- mice revealed the presence of a light-driven PK2 molecular rhythm in the SCN under LD, indicating that light can drive PK2 rhythm independent of functional circadian clock. Interestingly, the molecular rhythms of some clockwork genes such as Per2 , Per1 , and Bmal1 were also partially maintained in the SCN of Cry1-/-Cry2-/- mice under LD (Figure 4 ). Thus, light-driven molecular oscillations of clockwork or clock-controlled output genes exist in the absence of functional circadian clock. Vitaterna et al (1999) first noticed such light-regulated Per2 molecular rhythm in the SCN of Cry1-/-Cry2-/- mice, and suggested the term of "light-driving" effect [ 25 ]. As Cry1-/-Cry2-/- mice lack functional circadian clocks and their locomotor behaviour remains rhythmic under LD, but not under DD conditions [ 24 , 25 ], it is likely that these light-driven molecular rhythms may drive the locomotor rhythms in these animals. As we have previously shown that PK2 may be a critical output molecule responsible for circadian locomotor rhythms, the presence of this light-driven PK2 rhythm in Cry1-/-Cry2-/- mice may thus contribute to or underlie masking as well as the free running behavioural rhythms in these animals. It is well established that an intact SCN is necessary for the preservation of free running locomotor rhythms [ 43 ]. The role of the SCN in masking of locomotor activity by light is controversial, with similar studies having produced contradictory results [ 23 , 44 ]. Thus, it is possible that there might be common signal molecule(s) that mediate(s) the light-masking and the circadian clock-controlled locomotor behaviour. Construction of PK2 -deficient mice will be necessary to resolve the exact role of PK2 in the light-driven locomotor rhythms. The light inducibility of PK2 in the SCN is an unusual characteristic for a clock-controlled gene. Our results demonstrate that melanopsin-positive retinal ganglion cells, in conjunction with rods and cones, are responsible for the light-inducibility of PK2 (Figure 3 ). The same photoreceptive system has been shown responsible for the entrainment of locomotor rhythm [ 5 - 11 ]. The light inducibility of PK2 may be related to the presence of a putative cyclic AMP response element (CRE) in the promoter of the PK2 gene [ 28 ]. CRE-dependent activation is critical for light-induced gene expression in the SCN [ 45 - 48 ]. The reduced light inducibility of PK2 in mutant mice that lack functional clock may indicate that CRE-dependent pathway and CLK/BMAL1 transcriptional factors may interact in the light-induced PK2 expression in the SCN. Accumulative data have implicated the photic regulation of the transcription of clock genes such as Per1 and Per2 in the entrainment of behavioural rhythms [ 30 , 34 ]. The phase of the core SCN clock gene expression determines the timing of clock-controlled SCN output signals that ultimately regulate physiology and behaviour. Unlike the Per1 promoter, whose activation in the SCN shifts rapidly when the LD cycle is advanced [ 31 ], PK2 exhibits transients during phase advance, more similar to those of Cry1 and Cry2 [ 30 , 31 ]. This is consistent with the role for PK2 as a clock-controlled gene and thus is downstream from the light-regulated expression of Per1 or Per2 . The presence of E box motifs in the PK2 promoter suggests that light-regulated Per1 (and perhaps Per2 ) expression can influence PK2 expression. However, the light inducibility of PK2 indicates that PK2 may have a more direct and central role in entrainment in addition to its putative role as an SCN output signal. In other words, whether PK2 functions completely outside the central circadian loops or partly within them has yet to be determined. It is well established that the activation of glutamate receptor and its downstream actions are critical for the retinohypothalamic inputs of light to the SCN [ 49 ]. As receptor for PK2 is highly expressed in the SCN [ 28 ] and activation of the PK2 receptor triggers similar signalling pathways as that of glutamate receptors [ 29 ], it is possible that the circadian clock and/or light-driven PK2 may feed back to the core circadian loops in the SCN. In addition, PK2 has recently been shown to excite neurons that express PK2 receptor [ 50 ], further suggesting that PK2 may activate the firing of SCN neurons, and thus possibly participate in the synchronization of the circadian clock. Thus, the light inducibility of PK2 may be relevant to both the phase resetting of the core circadian loops and critical SCN output signals. Conclusion Our studies demonstrate that PK2 is predominantly driven by the circadian clock, as PK2 expression exhibits circadian transients in response to phase advances. Furthermore, shortening of the light period does not result in corresponding reduction of the phase of PK2 rhythm, also consistent with the dominant control from the circadian clock on PK2 expression. However, light also modulates PK2 rhythm. Nocturnal light pulses can directly induce PK2 expression in the SCN. Studies with Cry1-/-Cry2-/- mice revealed that light can drive a low amplitude PK2 molecular rhythm in the SCN in the absence of functional circadian oscillators. These studies demonstrate that PK2 molecular rhythm in the SCN is controlled by dual mechanisms: dominantly by the circadian transcriptional loops but also directly by light. The light inducibility of PK2 in the SCN suggest that in addition to PK2's role as a SCN output signal, PK2 may also participate in the photic entrainment of circadian clock and perhaps in masking. Methods Experiments of light/dark cycle shifts Male adult C57BL/6 mice (Taconic Farms, New York) were entrained under 12 hour light: 12 hour dark (12L:12D, lights on at 0700 h) cycle for two weeks with food and water available ad libitum. Light phase was either delayed by 6 hours (lights on at 1300 h) or advanced by 6 hours (lights on at 0100 h) and samples were taken every three hours for the 24 hour period (Zeitgeber time, ZT, ZT1-22). To examine PK2 expression two days after the shift, animals were placed in two additional light/dark cycles and brain samples were collected. All animal procedures were approved by the Institutional Animal Care and Use Committee and consistent with Federal guidelines. In situ hybridization was used in all studies to examine PK2 mRNA expression in the SCN [ 28 ]. Antisense and sense riboprobes containing the coding region of mouse PK2 (accession number AF487280 1-528 nt), mouse Per1 (accession number AF022992 340-761nt), mouse Per2 (accession number AF035830 9-489 nt) and mouse Bmal1 (accession number AB015203 864-1362 nt) were generated. Photoperiod studies Animals were initially entrained under 12L:12D for one week, followed by placement in different photoperiods (light intensity ~400 lux) for three to four weeks: 8 hour light:16 hour dark (8L:16D, lights on at 0900 h, lights off at 1700 h), 16 hour light: 8 hour dark (16L:8D, lights on at 0500 h, lights off at 2100 h). For the 20 hour light: 4 hour dark (20L:4D, lights on at 0300 h, lights off at 2300 h), animals were first placed in 14L:10D for one week, transferred to 16L:8D for another week, followed by two weeks in 20L:4D. All brain samples were taken every two hours throughout the 24 hour cycle, except the first and the last two time points which were sampled every three hours. Studies of melanopsin-deficient mice and mice that lack melanopsin, rods and cones Wild type and melanopsin-deficient ( Opn4-/- ) mice (on C57BL/6:129 hybrid background) [ 5 ] were entrained to 12L:12D and sampled every three hours for the 24 hour period (ZT1-22). For light pulse studies, wild type, Opn4-/- mice and triple knockouts ( Opn4-/- Gnat1-/- Cnga3-/- mice) that lack melanopsin, rod and cone phototransduction systems were used [ 10 ]. Animals received a 15 min light pulse (~200 lux) at ZT14 and brains were sampled one or two hours after light pulse. Dark control animals did not receive a light pulse. Studies of cryptochrome-deficient ( Cry1-/-Cry2-/- ) mice Cryptochrome-deficient ( Cry1-/-Cry2-/- ) mice on a C57BL/6:129 hybrid background were kindly provided by Dr. Aziz Sancar (University of North Carolina at Chapel Hill). Wild type and Cry1-/-Cry2-/- mice were entrained to 12L:12D and sampled every three hours for the 24 hour period (ZT1-22). A second group of Cry1-/-Cry2-/- mice were placed into two days of constant darkness (2DD) (Circadian time, CT, CT1-22). The mRNA levels of PK2 , Per2 , Per1 and Bmal1 were measured in the SCN. For light pulse experiments, Cry1-/-Cry2-/- mice received a 15 min light pulse (~400 lux) at ZT14, and sampled one or two hours after light pulse. Dark control Cry1-/-Cry2-/- mice did not receive a light pulse. For the shifting experiments, wildtype and Cry1-/-Cry2-/- mice were initially entrained under 12L:12D, then subjected to an acute 6 hour advance of lighting schedule. Running-wheel activities of these mice were monitored 10 days before and 10 days after the 6 hour advance shift. The 6 hour phase advance was then repeated and brains were collected at ZT4 and ZT16 on the day of the shift. Authors' contributions ELB sampled the Cry1-/-Cry2-/- mice and performed behavior experiments on Cry1-/-Cry2-/- mice. SH sampled the melanopsin-deficient mice and triple knockout mice. MYC performed the tissue sectioning, in situ hybridizations and all quantitative analyses. MYC, ELB and QYZ drafted the manuscript. ELB, SH, MYC and QYZ designed the studies. All authors read and approved the final manuscript. Supplementary Material Additional File 1 Effect of different photoperiods on molecular rhythms in the SCN. Temporal profiles of Per1 (a) and Per2 (b) mRNA under 8L:16D, 16L:8D, 20L:4D. Open and filled bars indicate light and dark periods, respectively. The zeitgeber time (ZT) on the x-axis reflects the timescale for each photoperiod. Each value represents the mean ± SEM of 3–4 animals. Click here for file Additional File 2 PK2 mRNA expression in Cry1-/-Cry2-/- and wildtype mice. Representative autoradiograms of PK2 mRNA in the SCN of Cry1-/-Cry2-/- mice (Cry) and wild type mice (WT) under LD (ZT1-22) are shown (top and bottom row, respectively). Scale bar = 1 mm. Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC555564.xml

548521

Generation of human antibody fragments against Streptococcus mutans using a phage display chain shuffling approach

Background Common oral diseases and dental caries can be prevented effectively by passive immunization. In humans, passive immunotherapy may require the use of humanized or human antibodies to prevent adverse immune responses against murine epitopes. Therefore we generated human single chain and diabody antibody derivatives based on the binding characteristics of the murine monoclonal antibody Guy's 13. The murine form of this antibody has been used successfully to prevent Streptococcus mutans colonization and the development of dental caries in non-human primates, and to prevent bacterial colonization in human clinical trials. Results The antibody derivatives were generated using a chain-shuffling approach based on human antibody variable gene phage-display libraries. Like the parent antibody, these derivatives bound specifically to SAI/II, the surface adhesin of the oral pathogen S. mutans . Conclusions Humanization of murine antibodies can be easily achieved using phage display libraries. The human antibody fragments bind the antigen as well as the causative agent of dental caries. In addition the human diabody derivative is capable of aggregating S. mutans in vitro , making it a useful candidate passive immunotherapeutic agent for oral diseases.

Background Dental caries is one of the most common infectious diseases of humans. The main causative agent is a group of streptococcal species collectively described as the mutans streptococci [ 1 ]. Streptococcus mutans has been identified as the major etiological agent of the disease. Unlike many other diseases, dental caries is as prevalent in the West as it is in developing countries, and therefore attracts significant interest from medical and dental authorities as well as pharmaceutical companies. The first step in the initiation of infection is the attachment of the bacterium to a specific receptor, and this is an ideal point for intervention. Two groups of proteins from mutans streptococci represent primary candidates for a human caries vaccine: i) glucosyltransferase enzymes, which synthesize adhesive glycans and allow microbial accumulation, and ii) cell surface fibrillar proteins that mediate adherence to the salivary pellicle [ 2 ]. The bacterial adhesin SAI/II [ 3 ], a surface-displayed protein with a molecular mass of 190 kDa, plays an important role in the initial attachment of S. mutans to the tooth surface. Antibodies recognizing this protein prevent colonization of the buccal cavity by the bacterium and could be developed as a vaccine against dental caries. The most suitable vaccination strategy would be passive immunization, in which monoclonal antibodies or fragments thereof are applied to the tooth surface e.g. using toothpaste, mouthwash or chewing gum. This would make active immunization with the S. mutans adhesin unnecessary. The murine monoclonal antibody Guy's 13 [ 4 ] which specifically recognizes the SAI/II protein of S. mutans and Streptococcus sobrinus has been used successfully to prevent S. mutans colonization and the development of dental caries in non-human primates [ 5 ]. The antibody also prevented bacterial colonization in human clinical trials [ 6 , 7 ]. However, like other murine antibodies, a major limitation in clinical applications may be the human anti-mouse antibody response (HAMA), which can increase the rate of clearance and initiate allergic reactions [ 8 ]. The problems associated with murine antibodies can be overcome by replacing murine sequences with their human counterparts, e.g. by chimerization [ 9 ], CDR grafting [ 10 ] and guided selection using phage display technology [ 11 ]. Furthermore, the use of antibody fragments rather than whole antibodies also removes some of the constant regions that may provoke an immune response. There has been a growing interest in the use of single-chain fragment variable (scFv) antibodies, in which the variable regions of the heavy and light chains are combined in the same polypeptide chain (Huston, 1988 #2785). The advantages of such derivatives are that they can be expressed as single transgenes in various hosts, they fold spontaneously to adopt the correct tertiary structure, and their small size facilitates tissue penetration. The scFv has the heavy and light chain variable regions joined by a flexible peptide linker allowing the two domains to interact, forming a univalent antibody. Alternatively, diabodies have the same structure but the two domains are joined by a shorter, less-flexible linker, forcing dimerization and the formation of divalent antibodies (Holliger, 1993 #3498). We have generated human derivatives of the murine Guy's 13 antibody using a chain-shuffling approach based on human antibody variable gene phage-display libraries. We have taken the variable gene regions of the original Guy's 13 monoclonal antibody and created human scFv and diabody derivatives by chain shuffling in human phage-display libraries. Firstly, the heavy chain variable gene of the Guy's 13 construct was introduced into a naïve human light chain phage display library to select human light chains that, in combination with the murine heavy chain, showed binding specificity for the SAI/II antigen. Once such chimeric antibodies had been selected, the murine heavy chain gene was replaced with the human counterpart, by introducing the selected human light chain genes into a human heavy chain phage display library. The resulting clones were expressed in bacteria and tested for specificity in ELISAs using both SAI/II antigen and whole S. mutans . The stepwise procedure for generating human antibody chains allows the advantages of scFv and diabody antibody fragments to be exploited without suffering the negative effects of non-human antibodies in a clinical setting. The human antibody fragments were expressed in bacteria as scFv and diabody derivatives and used to aggregate S. mutans in vitro . The diabodies were able to aggregate the bacteria and therefore have the potential to be developed as therapeutic agents to treat and/or prevent dental caries. Results Human recombinant scFv antibodies against SAI/II Human scFv antibody fragments based on the murine monoclonal antibody Guy's 13 were constructed using two consecutive rounds of variable-domain shuffling and phage-library selection (Figure 1 ). First, a chimeric scFv was generated by amplifying the murine Guy's 13 heavy chain variable region, and inserting it into a human light chain variable region phage display library. The resulting phage display library had a complexity of 5 × 10 5 . Single chain Fv antibody fragments with appropriate binding activities were selected on purified, immobilized SAI/II antigen. Three rounds of selection were carried out and unique candidate antibodies were identified by ELISA (Figure 2 ). Subsequent sequencing yielded five antibody fragments (chimscFvA1, chimscFvA6, chimscFvA9, chimscFvB4, and chimscFvG4). Sequencing of the human variable genes showed that two of the clones chimscFvA6, chimscFvB4 belonged to family Vκ1, clone chimscFvA6 was homologous to HK137 and chimscFvB4 was homologous to the L12 germline gene family. ChimscFvA9 belonged to family Vκ4 DPκ24. ChimscFvA1 and chimscFvG4 belonged to family Vλ3 DPL16) (data not shown). Inhibition ELISA showed that the binding of all six chimeric scFvs to SAI/II could be inhibited by the murine monoclonal antibody Guy's 13. The binding of chimeric scFvs A6, A9, B4 and C6 was inhibited by approximately 80%, suggesting that epitope recognition was maintained (Figure 3 ). The binding of the chimeric scFvs A1 and G4 was only inhibited by approximately 30%, suggesting that these antibodies recognized a different epitope. The selected human V L genes were introduced into a human V H library (complexity 8 × 10 8 ) and a combinatorial library with a complexity of 1 × 10 6 was established. Three rounds of selection were carried out in solution using SAI/II antigen coupled to paramagnetic beads. Eleven human scFvs were identified by ELISA (data not shown). Subsequent sequence analysis identified three human scFvs: clones huscFv B10, huscFv D12 and huscFv H6. Table 1 shows the amino acid sequences of the human scFv antibody fragments. The human V L domain in chimeric scFv A6 (Vκ1 HK137) was selected in combination with two different human variable heavy chains, giving human scFvs B10 and H6, respectively. The V H domain of human scFv B10 is homologous to V H 1 family DP10, and the V H domain of human scFv H6 is homologous to V H 3 family DP35. The human V L domain in chimeric scFv B4 (Vκ1 L12) was selected in combination with one human variable heavy chain giving the human scFv D12. The V H domain of human scFv D12 is homologous to VH5 family DP73. Figure 4 shows the binding of the three human scFvs to the SAI/II antigen and the pathogenic bacterium S. mutans . Inhibition ELISA showed that the binding of all three human scFvs to SAI/II was inhibited by Guy's 13, suggesting that epitope recognition was maintained. Generation of human diabodies and agglutination of S. mutans Recombinant antibody fragments can be engineered to assemble into stable multimeric oligomers of high binding avidity and specificity [ 12 ]. A scFv molecule joined by a linker of 3–12 residues cannot fold into a functional Fv domain and instead associates with a second scFv molecule to form a bivalent dimer (diabody, approx. 60 kDa). For the cross-linking of cell surface antigens, at least two binding domains are necessary. The diabody is the smallest bivalent antibody molecule that can fulfill this prerequisite. Through reduced off-rates, which result from multiple binding to two target antigens and to rebinding when one Fv dissociates, the diabody is suitable to facilitate specific agglutination of bacteria. We constructed human diabodies by isolating the variable heavy and light chain genes from human scFvs B10, D12 and H6 and murine scFv Guy's 13, amplified by PCR (Table 2 ) and inserting them in two consecutive steps into the vector pHenIXdia, containing a 10 amino acid residue linker. The integrity of the clones was confirmed by sequencing and the binding activity was demonstrated by ELISA using both the SAI/II antigen and S. mutans cells (Figure 4 ). Ma et al. [ 7 ] reported that bivalent binding of the murine Guy's 13 is required for protection against dental caries, since the F(ab')2 derivative was protective but not the monovalent Fab fragment. S. mutans was aggregated in a dose dependent manner when grown in the presence of murine diabody Guy's 13 and human diabody D12 (Figure 5A and 5B ). Discussion Antibodies recognizing and neutralizing the oral pathogen S. mutans provide a novel approach for the control and prevention of dental caries. A monoclonal antibody that binds specifically to the SAI/II surface adhesin of S. mutans was isolated by Smith et al. (20) and has been expressed in plants as a secretory IgA (sIgA) [ 13 ]. In ongoing Phase II clinical trials, this recombinant antibody has been shown to prevent recolonization of the mouth by S. mutans when coated onto the teeth and gums following eradication of the bacteria. The sIgA is probably the most appropriate format for the topical application of antibodies that inhibit the colonization of the tooth surface by S. mutans because this is the predominant form of antibody naturally found in the saliva. However, each sIgA comprises ten polypeptide chains of four different types making it difficult to produce on a large scale in conventional production systems. The more convenient diabody antibody derivatives, which can be expressed in large quantities in microbial culture systems, may be more suitable for the type of production scales that would be required for the routine control of dental caries using this strategy. As a first step in this direction, we have taken the variable gene regions of the original Guy's 13 monoclonal antibody and created human scFv and diabody derivatives by chain shuffling in human phage-display libraries. The heavy chain variable gene of the Guy's 13 construct was introduced into a naïve human light chain phage display library to select human light chains that, in combination with the murine heavy chain, showed binding specificity for the SAI/II antigen. Once such chimeric antibodies had been selected, the murine heavy chain gene was replaced with the human counterpart, by introducing the selected human light chain genes into a human heavy chain phage display library. The resulting clones were expressed in bacteria and tested for specificity in ELISAs using both SAI/II antigen and whole S. mutans . The stepwise procedure for generating human antibody chains allows the advantages of scFv and diabody antibody fragments to be exploited without suffering the negative effects of non-human antibodies in a clinical setting. Small scFv and diabody antibody fragments are easy to express in large quantities, they penetrate tissues easily and they lack the constant domains that promote often-unwanted and usually superfluous effector functions. However, where such antibodies are murine in origin, they can provoke an immune reaction in the human host, leading to rapid clearance and poor efficacy during long-term treatment. Since dental caries tends to be chronic rather than acute, murine antibodies would be of little benefit to patients in the long term. Two drawbacks of scFvs compared to the ideal sIgA format are monovalency and instability. ScFvs are monovalent because the heavy and light chains are joined by a flexible peptide linker, which allows the two domains to fold and interact with each other. We have addressed this problem by converting the scFv antibodies into diabodies, which is achieved by shortening the linking peptide and forcing the heavy and light chain variable domains to seek interaction partners as part of a dimer. As a consequence of this interaction, the diabody is bivalent like the parent immunoglobulin, and therefore has increased binding avidity. We showed that the bivalent binding of the diabody antibody constructs leads to agglutination of S. mutans . The problem of decreased stability may be more difficult to address, because the efficacy of scFv and diabody molecules used to treat dental caries will depend largely on their persistence and effective concentration. Secretory IgAs include a secretory component, which protects the antibody from proteolytic degradation in the saliva. One possible solution is to include this secretory component in any scFv or diabody format through the use of further gene fusion strategies. However, it is envisaged that antibodies for the prevention of dental caries will be administered in the form of toothpaste or mouthwash, or perhaps chewing gum, which will allow the treatment to be refreshed at regular intervals. Conclusions We have shown that the humanization of a murine monoclonal antibody can be easily achieved using a chain-shuffling approach based on scFv antibody phage display libraries. The human antibody derivatives of the murine Guy's 13 antibody can be expressed and isolated from bacteria, recognize SAI/II and S. mutans with great specificity, and can successfully aggregate S. mutans cells in the dimeric form in a dose dependent manor. These recombinant therapeutic proteins therefore represent the first step towards an inexpensive and convenient general treatment for dental caries. Methods Propagation of S. mutans S. mutans 20523 serological group c was purchased from DSMZ (Braunschweig, Germany) and grown in an S2 containment laboratory in trypticase soy yeast extract medium (30 g l -1 trypticase soy broth, 3 g l -1 yeast extract, pH 7.0–7.2) at 37°C for 2 d prior to use. Cloning the S. mutans spa P gene encoding surface antigen SAI/II Nucleotides 214–3048 of the spa P gene [ 14 ], which encodes the SAI/II antigen, were removed from pUC18 as a Sfi I/ Not I fragment and inserted into the bacterial expression vectors pCantab5E (Pharmacia) and pSin1 [ 15 ] which had been digested with the same enzymes followed by transformation into E. coli TG1. The pCantab5E vector contained an additional sequence encoding the E-tag, facilitating the detection of expressed proteins using the monoclonal antibody 5E (Pharmacia). The pSin1 vector similarly contained sequences encoding a MYC-tag, facilitating detection with the murine monoclonal antibody 9E10 (ATCC CRL 1729), and a His 6 tag, allowing purification of expressed proteins by immobilized metal-chelate affinity chromatography (IMAC) and detection using a murine Penta-HIS antibody (Qiagen). SAI/II expressed using pSin1 was used for the selection of antibodies from phage-display libraries. SAI/II expressed in pCantab5E was used for enzyme-linked immunosorbent assays (ELISAs). Coating paramagnetic beads with SAI/II For the selection of phage-display antibodies, 250 μl of phosphate-buffered saline (PBS)-washed Dynabeads (Dynal Biotech GmbH) was resuspended in 500 μl 0.1 M phosphate buffer (pH 7.4) and mixed gently for 2 min. The beads were collected with a magnet, the supernatant discarded and the beads resuspended in 250 μl of the same buffer, followed by the addition of 500 μl SAI/II antigen (1 mg/ml). After incubation for 16 h at 37°C with slow tilt rotation, the beads were collected with a magnet and the supernatant was discarded. The coated beads were washed four times, twice with 0.13 M NaCl, 1% milk powder in 0.01 M phosphate buffer (pH 7.4) for 5 min at 4°C, once with 0.2 M Tris-HCl (pH 8.5) for 4 h at 37°C and again in the same buffer for 5 min at 4°C. Cloning scFv Guy's 13 in pSin1 The variable region genes of the heavy and light chain of the murine monoclonal antibody Guy's 13 were amplified using oligonucleotide primers LMB3 (5' CAG GAA ACA GCT ATG AC 3') and fdSeq 1 (5' GAA TTT TCT GTA TG/AG GG 3') followed by digestion with Sfi I and Not I. The products were inserted into the phagemid vector pSin1, which had been treated with the same enzymes, and the recombinant vector was introduced into E. coli strain TG1. Construction and selection of human SAI/II-specific scFv antibodies Figure 1 shows the schematic scheme of the construction and selection of human SAI/II specific scFv antibody approach. The variable heavy chain antibody domain of the murine antibody Guy's 13 was cloned as an Sfi I/ Sal I fragment in the bacterial expression vector pHenIX containing a light-chain antibody phage-display library derived from naïve human peripheral blood lymphocytes (8 × 10 8 ; R. Finnern, unpublished data). This vector is based on the phagemid vector pHen1 [ 16 ] designed to express antibody fragments as N-terminal fusions with the minor coat protein of filamentous bacteriophage M13. An amber stop codon between the two fusion partners allows the expression of both soluble antibody fragments and phage particles displaying recombinant antibodies. The recombinant vectors were introduced into E. coli strain TG1. Three rounds of selection were carried out using immobilized SAI/II antigen as described by Marks et al. [ 17 ]. Elution was achieved using the monoclonal antibody Guy's 13 to select binders recognizing the same epitope. The expression of soluble scFvs was performed as described by Marks et al. [ 17 ] and scFvs specific for the SAI/II antigen were identified by ELISA using SAI/II antigen. The selected variable antibody domain genes of the shuffled human light chains were cloned as Apa LI and Not I fragments in pHenIX containing a human variable heavy chain library (8 × 10 8 ; R. Finnern, unpublished data) and introduced into E. coli TG1. This was achieved by PCR amplification of the human light chain genes using primers Vκ4 Apa LI (5'-TGAGCACACAGTGCACTCGACATCGTGATGACCCAGTCTCC-3'), Vκ1 Apa LI (5'-TGAGCACACAGTGCACTCGACATCCAGATGACCCAGTCTCC-3') and Jκ1 Not I (5'-GAGTCATTCTCGACTTGCGGCCGCACGTTTGATC/TTCCAC/GCTTGGTCCC-3'). Three rounds of selection were carried out using SAI/II antigen immobilized on Dynabeads. Briefly, 150 μg of SAI/II-coated beads was blocked for 1 h with 2 ml 2% milk powder. The beads were collected with a magnet, washed in PBS and incubated with the antibody phage display library for 1 h on a turntable. The beads were washed 15 times with PBS containing 0.05% Tween 20 and 15 times with PBS to remove unbound phage. Bound phage were eluted with 100 μl 100 mM triethanolamine for 10 min on a turntable followed by neutralization in 200 μl 1 M Tris-HCl (pH 8.0). Eluted phage were used to infect exponentially growing E. coli TG1 and grown overnight at 30°C on TYE plates containing 100 μg ml -1 ampicillin, 1% glucose. Selection, phage rescue and induction of soluble scFv expression were carried out as described by Marks et al. [ 17 ]. Antigen-specific human scFvs were identified by ELISA using the SAI/II antigen. Propagation of phage display antibody libraries One litre of 2× TY (supplemented with 100 μg ml -1 ampicillin, 1% glucose) was inoculated with an aliquot of the phage antibody library glycerol stock. The rescue and induction of the phage was carried out essentially as described in Marks et al. [ 17 ]. Phagemid rescue was carried out by the addition of 10 10 units of helper phage VCSM13 (Pharmacia) to the growing phage antibody library. The culture medium was changed to 2× TY containing 100 μg ml -1 ampicillin and 25 μg ml -1 kanamycin and incubated on an orbital shaker overnight at 30°C and 250 rpm. Phage were purified twice by polyethylene glycol (PEG) precipitation (20% PEG, 2.5 M NaCl) and resuspended in a final volume of 2 ml PBS. The phage were stored at 4°C until further use. DNA sequencing The number of unique clones was determined by PCR amplification of the recombinant antibody inserts using primers LMB3 (5' CAG GAA ACA GCT ATG AC 3') and fdSeq 1 (5' GAA TTT TCT GTA TG/AG GG 3') followed by digestion with the restriction enzyme Bst NI (New England Biolabs). The variable antibody genes from two clones of each restriction pattern were analyzed by PCR cycle sequencing using infrared-labeled primers according to the manufacturer's instructions (Licor). Sequencing reactions were carried out on a Licor automated DNA sequencer (4000 L) and the sequences were analyzed using Sequencher 3.1 (Gene Codes Corporation). The sequences of the V H and V L genes were compared with the germline sequences in the V-BASE database (Tomlinson et al., MRC Centre for Protein Engineering, Cambridge, UK). Construction of diabodies The construction of diabodies [ 18 ] was carried out by PCR amplification of the variable heavy and light chain antibody regions of the human scFv clones and subcloning these in vector pHenIXdia. The diabody constructs consisted of the variable heavy and light chain antibody domains linked by a ten-amino-acid linker (TGGGGSSSAL), forcing the expressed domains to attach to a complementary chain in solution to create two antigen-binding sites. The primers used for the construction of the diabody antibody format are listed in Table 1 . Upscaled production of recombinant proteins Recombinant proteins were recovered from the bacterial periplasm following induction with 0.5 mM final concentration of isopropyl-β-D-galactopyranoside (IPTG) for 3–4 h at 30°C [ 19 ]. After centrifugation (4000 × g, 4°C, 30 min), the pellet was resuspended in 10 ml 30 mM Tris-HCl (pH 8.2) containing 20% sucrose, 1 mM ethylenediaminetretaacetic acid (EDTA), incubated on ice for 15 min and centrifuged as above. The pellet was resuspended in 10 ml 5 mM MgSO 4 , 1 mM EDTA and incubated for 15 minutes on ice before a final centrifugation step as above. Both supernatants were pooled, dialyzed against PBS and stored at 4°C. Recombinant proteins were also expressed in the periplasm under osmotic stress in the presence of compatible solutes as described by Barth et al. [ 20 ]. Briefly, bacteria were grown overnight at 26°C in Terrific Broth (TB) (12 g l -1 bacto-tryptone, 24 g l -1 bacto-yeast-extract, 4 ml l -1 glycerol) containing 100 μg ml -1 ampicillin and 0.5 mM ZnCl 2 . The culture was diluted 30-fold in 200 ml of the same medium. When the OD 600 nm of the culture reached 2.0, it was supplemented with 0.5 M sorbitol, 4% NaCl, 40 mM glycine betaine and incubated at 26°C for an additional 30–60 min. Expression was induced with 1 mM final concentration IPTG and growth for 6 h at 26°C. Cells were harvested by centrifugation at 30,000 × g for 10 min. The recombinant antibody fragments were isolated from the periplasmic space as described above. The periplasmic and osmotic shock fractions were pooled and dialyzed against PBS. Phenylmethylsulfonylfluoride (PMSF) was added to a final concentration of 1 mM. Purification of recombinant proteins The human scFv and diabody antibody fragments were purified by IMAC using the His 6 tag as described by Griffiths et al. [ 21 ]. Briefly, 10 ml columns (BioRad Polyprep chromatography columns) were packed with 500 μl Ni-NTA resin (Qiagen) and washed with five column volumes of PBS prior to loading with the recombinant proteins. The columns were washed with 10 column volumes of PBS containing 10 mM imidazol. Bound proteins were eluted with 250 mM imidazol and collected in 1-ml fractions. Protein concentrations were determined by spectrophotometry assuming that A 280 nm = 1 corresponds to a scFv or diabody concentration of 0.7 mg ml -1 . Gel filtration was used for further purification. A Sephadex 200 column (Pharmacia) was equilibrated with PBS. ScFv or diabody antibody fragments were loaded and run at 1 ml min -1 . Aprotinin (6500 Da), cytochrome C (12,400 Da), carbonic anhydrase (29,000 Da), bovine serum albumin (BSA) (66,000 Da) and Dextran Blue (2,000,000 Da) were used as molecular weight standards (Fluka). Enzyme linked immunosorbent assay (ELISA) S. mutans , SAI/II antigen or BSA were coated on ELISA plates (Nunc) at a concentration of 1–10 μg per well in PBS overnight at 4°C. The plates were washed three times with PBS and blocked with 2% milk powder in PBS for 2 h at room temperature. The scFvs were tested either at a concentration of 1 μg per well or 100 μl per well of overnight-induced culture. Recombinant antibodies containing the MYC tag were detected with the murine 9E10 monoclonal antibody (ATCC CRL1729). Antibodies containing a His 6 tag were detected using the murine anti-Penta-HIS antibody (Qiagen). The murine antibodies were detected with a goat-anti-mouse (Fc-specific) peroxidase-labeled antibody. The assays were developed with 3,3',5,5'-tetramethylbenzidine (TMB) (Sigma). Reactions were stopped by the addition of H 2 SO 4 after 20 min and readings taken at OD 450 nm . Between every incubation step, the plates were washed three times with PBS containing 0.05% Tween 20 and three times with PBS. Inhibition ELISA was carried out by binding of chimeric scFv to SAI/II and replacement with an excess of mAb Guy13. Bound scFv were detected as described above using the murine 9E10 monoclonal antibody detecting the MYC-tag. Agglutination of S. mutans Cultured S. mutans was divided into 20-μl aliquots and incubated with serial dilutions of bacterially expressed recombinant antibodies dia mGuy13 and diaD12, respectively for 2 d at 4°C or 1 h at 37°C on Lab-Tek II chamber slides (Nalge Nunc International). As negative control an unrelated human diabody was used. Excess medium was discarded and the cells were air-dried. The bacteria were counterstained with Gram solution (Diagnostica Merck). The slides were mounted with Immunofluor medium (ICN Biomedicals Inc) and photographed with a Zeiss Axioskob microscope. Authors contributions MBK generated the diabody formats and performed all in vitro assays. MH did the upscaled expression of the antibodies. HS assisted in the analysis of data. JKCM assisted in the analysis and interpretation of the data especially in respect to the data obtained with the murine mAb Guy13. SB assisted in the interpretation of the data. RF assisted in the conception and revision of the project. RF* generated the human phage display libraries, developed the concept and supervised the project

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548521.xml

543448

A model for analysis, systemic planning and strategic synthesis for health science teaching in the Democratic Republic of the Congo: a vision for action

Background The problem of training human resources in health is a real concern in public health in Central Africa. What can be changed in order to train more competent health professionals? This is of utmost importance in primary health care. Methods Taking into account the level of training of secondary-level nurses in the Democratic Republic of the Congo (DRC), a systemic approach, based on the PRECEDE PROCEED model of analysis, led to a better understanding of the educational determinants and of the factors favourable to a better match between training in health sciences and the expected competences of the health professionals. This article must be read on two complementary levels: one reading, focused on the methodological process, should allow our findings to be transferred to other problems (adaptation of a health promotion model to the educational sphere). The other reading, revolving around the specific theme and results, should provide a frame of reference and specific avenues for action to improve human resources in the health field (using the results of its application in health science teaching in the DRC). Results The results show that it is important to start this training with a global and integrated approach shared by all the actors. The strategies of action entail the need for an approach taking into account all the aspects, i.e. sociological, educational, medical and public health. Conclusions The analysis of the results shows that one cannot bring any change without integrated strategies of action and a multidisciplinary approach that includes all the complex determinants of health behaviour, and to do it within the organization of local structures and institutions in the ministry of health in the DRC.

Background A partnership of the Ministry of Health of the Democratic Republic of the Congo (DRC) – more specifically, the directorate that is in charge of health science education – the French-speaking community of Belgium and various education and training associations made it possible to set up and carry out a teaching innovation project to bolster human resources in the health sector. One of the major public health challenges in Africa is to find efficient ways to enhance human resources in the health sector. The goal of the medical policy in the DRC is to promote the health of the population by providing high-quality health care that is complete and integrated and in which the community participates, within the general context of the fight against poverty [ 1 ]. With this intention, the Ministry of Health defined six strategic axes to support the reinforcement of primary health care: • restructuring the health system according to political, legislative and administrative orientations as well as updating standards of services; • increasing the availability of resources by implementing an adequate administrative process; • establishing an integrated system of preventive and curative care and health promotion for the target groups; • strengthening the programmes of support to health activities; • coordinating, promoting intrasectoral and intersectoral collaboration and partnership for health; • promoting a suitable environmental framework for health. The Ministry attaches priority importance to delivering high-quality care and health services and by: reaffirming the strategy of primary health care (PHC) as a fundamental option of the national policy on health; reaffirming the health zones' or districts' achieving a minimum package of activities as an operational unit; and regular procurement of essential drugs, including biological products and other laboratory reagents. Training of nurses in the DRC is organized in all provinces of the country and conducted through technical medical institutes (ITM), medical educational institutes (IEM) for the secondary level and higher institutes of medical technology (ISTM) for the higher level. In 1998 there were 308 ITM and ISTM in the country; to date there are 254 schools of nurses (male and female) at the secondary level in the country recognized by the DRC government. According to the type of management, these institutions are classified as public, officially agreed and private schools. However, the autonomy of management for all these schools is extensive, given the near inexistence of government subsidy. The infrastructure and quality of training differ widely from one ITM to another. As a rule, the solutions that one observes in public health are located in the area of further training for health professionals. To the extent that further training is important, it is disappointing to ascertain the low yields that these various training courses have in improving the quality of health care and services [ 2 ]. There is a lack of prior analysis of the main training needs when it comes to developing abilities and independence, a lack of these training programmes' integration into existing structures, and a lack of consistency between these schemes and a complex environment composed of many interacting players. Moreover, it is difficult, for many different reasons, to escape from the many vertical programmes (more than 40 in the DRC) that impose their own training modules on target audiences that have been set beforehand on their level. When it comes to basic training, the young nurses who have just graduated from secondary school and make up the critical mass of health professionals in primary health care are commonly required by specific private or church employers to enrol in a full year of training after their academic studies in order to try to fill the gaps between their basic training and health professionals' actual training needs. Confusion on the part of the ministries and other forces involved is attested to by the absence of vision and lack of expertise in educational research and the lack of reforms of the educational curricula in order to keep pace with the fundamental changes that have resulted from the decentralization of primary health care. In addition, several other problems make this succinct analysis more complex. The unemployment rate for the country's nursing school graduates is extremely high, even for those who graduate from the best schools in Kinshasa, the capital. The situation is compounded by the almost total absence of quality management mechanisms, especially when it comes to taking stock of the health workforce that exists. The situation is a complex one in which the expected changes are not clearly discerned. The problem of health sector human resources is vast. The context that interests us in this article is that of the human resources who are in the front line when it comes to grappling with the various communities' health problems. These are the nurses who have completed (technical) secondary school courses. They are the main primary health care professionals in Central Africa, especially in the DRC [ 3 ]. This choice limits the initial problem to a specific target population. However, the approach that we envisioned can be transferred quite easily to the other sectors concerned by health manpower management (mainly registered nurses, doctors, laboratory technicians and other health professionals). The question of research is at three levels: on the theoretical level, this article proposes importing a theoretical model from one field to another; on the methodological level, it uses the action research-like mode of data collection to better establish results; on the empirical level, in the DRC research is unusual ground from which to introduce an innovation. This qualitative research pursues two objectives: to present a methodology (adaptation of a health promotion model to the educational sphere) and to study the results of its application (health science teaching in the DRC). The qualitative hypothesis that subtends this thinking is that using an analytical, systemic planning and strategic synthesis model based on a systemic and participatory approach on various strategic and operational levels will procure the necessary vision for changing the basic education and training of nurses in (technical) secondary schools in the DRC. There is a lack of literature on experiments and experiences that use analytical or planning methods to understand complex social realities and consequently the adoption of strategic plans of action that should result from such experiences. It is thus important for all the players in the process to use the outcomes of the various stages of analysis and planning to produce an appropriate and adapted logical framework. It is necessary, however, to be able to set down on paper the methodology that is used to be able to construct a model that by the end of the process appears obvious for the actors involved, i.e. teachers, internship supervisors and school management, personnel within the ministry's Directorate for Health Education and project officers. Methodology Generally speaking, three relatively distinct stages were necessary, as follows: 1st methodological stage: set the strategic, managerial and operational levels In the DRC, the Ministry of Health and more specifically its directorate in charge of health science education, is responsible for the basic training of secondary school nursing students and further training of health professionals (Figure 1 ). Figure 1 Organizational structure of the Ministry of Health of the Democratic Republic of the Congo The situation sometimes varies in neighbouring contexts. Thus in Rwanda, for example, the Ministry of Education is in charge of basic education and training in the technical schools and for the medical professions. In France and Belgium, the general choice was to have the Ministry of Education responsible for most of the training and education of health professionals. We shall not discuss in this article the relevance of the place of oversight for this type of brief for training health professionals. We shall limit our remarks to the need to choose the best place for systemic planning and strategic vision in the existing context. So it is that in the 6th Directorate of the Ministry of Health of the DRC, which is in charge of paramedical secondary education, the need was perceived to develop systemic planning tools that would give a comprehensive, consistent vision of the sector's needs. While the 6th Directorate is indeed the strategic level, there was early involvement of an operational level, in the form of a sample of schools and teachers, and creation of a management unit to guide the implementation of the plans by the 6th Directorate and teachers from the grassroots. It is important to remember that we are talking about systemic and operational planning, not just strategic planning [ 4 ]. For the strategic level, it is thus necessary to determine the organizational level that is close enough to operations on the ground, yet at the same time is independent enough to take specific, actual strategic directions. 2nd methodological stage: Choosing an analytical, planning and strategic synthesis model that fosters a systemic vision This article follows on from the given that organizations and human beings are complex, and one way to have public health actions that heed this complexity is to use a systemic approach to analyse them [ 5 ]. Various models for a systemic approach exist. The approach that we chose to develop a logical framework for analysing, planning, and synthesising the work of the ministry's directorate in charge of health science education in the DRC is Green and Kreuter's PRECEDE PROCEED model [ 6 , 7 ]. The PRECEDE model emanates from a conceptual synthesis of the founding elements and roots of what would become health promotion. The PRECEDE PROCEED planning model is welcome because of its multidisciplinary approach, based on the fields of epidemiology, social sciences, behavioural sciences, education and health administration. In a nutshell, the fundamental principles that gave rise to this approach come from the multifactoral nature of all problems. Once this has been posited, all efforts made to act upon behaviour, the environment and social factors must necessarily be multidimensional and multisectoral. The acronym PRECEDE means "Predisposing, Reinforcing and Enabling Constructs in Educational/Environment Diagnosis and Evaluation", while the acronym PROCEED means "Policy, Regulatory and Organizational Constructs in Educational and Environmental Development". The PRECEDE-PROCEED model emphasizes planning interventions by focusing on the expected outcomes of actions based on epidemiological, social, behavioural, environmental, educational, organizational, administrative and political diagnoses of a socio-health and/or educational situation. The stages in the construction of a systemic model for analysing the problem that interests us – health science teaching – are adapted as the process unfolds. One of this method's great potentials is its great flexibility, or its ability to adapt to the specific analyses' needs. A systemic analysis and planning model is built dynamically, in a process that calls for continuous assessment. The model that the ministry's 6th Directorate came up with, and that is presented below, must change with changing knowledge in the area. 3rd methodological stage: allowing a participatory approach to using the model It is important to stress the qualitative process of continual exchanges and constant observations among the players (teachers, internship supervisors, school management, ministry officers and donors) that made it possible to fill the gaps in the information-gathering process. All these elements are much more difficult to organize in one well-defined stage, but are indeed part of a process that stresses the participatory approach and comes under the third strand of the methodology being discussed. The development of the first model proceeded during the workshop held in Kinshasa at the starting of the project, with the participation of personnel from three pilot schools and the Ministry of Health in October 2002. The three-day workshop, with 40 participants from various institutional levels, permitted the establishment in December 2003 of strategic orientations and guidelines for the continued broadening of the programme. All PRECEDE PROCEED models are built upon the players' actual experiences of the problem to solve. The clarification of the problem itself, which is part of the epidemiological and social diagnosis, comes out of a debate that must be conducted with all the parties concerned. This problem will gradually become more and more clear as its statement shuttles back and forth among the parties until it eventually satisfies the strategic and institutional level that is in charge of the programme and that the problem concerns directly. If, thanks to a resolutely participatory approach, all of the players adopt the use of a systemic approach on a real strategic, managerial and operational level, it will become a solid tool for the entire teaching body concerned. Results and discussion The presentation of results is at two levels: PRECEDE results and PROCEED results. The first are mainly descriptive (to tell and analyse the facts). PROCEED results are more normative, leading to certain recommendations for practitioners and other actors. PRECEDE results To facilitate presentation of the results and understanding of this coherent, overall vision of the interrelated elements, it was considered pertinent to retranscribe the full model as it exists for the 6th Directorate of the DRC's Ministry of Health. To structure the results' presentation, we shall follow the order in which the model's construction progressed. The table must be read from right to left, starting with the epidemiological and social diagnosis, then going on to the behavioural diagnosis and from there to the analysis of the educational and environmental determinants of these behaviours, and then to end with the analysis of the institutional diagnosis (see Additional file 1 ). Epidemiological and social diagnosis In the Ministry of Health, all the players are concerned by the mortality and morbidity indicators in the country. For health professionals, the lack of quality of the service provision and care provided by their health system is an obvious cause of the people's lack of confidence in their health system [ 8 ]. However, to produce a verifiable systemic analysis and then effective strategic synthesis within the directorate that interests us, the problem of the directorate in charge of health science education must first be clarified in connection with this broader problem. Thus the mismatch between health science teaching and the competence that health professionals are expected to have was seen as connected to the lack of quality in health care and services. All the players on their various organizational levels – ministry staff, teachers, basic supervisors, donors and project officers – took this diagnosis on board as a major concern. Behavioural diagnosis Who are these players and what behaviour can explain, through a direct link, the diagnosis of inadequacy? In answering these questions with the players themselves, we discover that there are groups of players that are never clearly identified yet are clearly related to this problem of inadequacy. This is the case, for example, of the donors and nongovernmental organizations (NGOs). Revealing all the groups of players makes it possible to see more easily why importance should be given to a multisectoral approach, especially one that covers teachers and medical and paramedical professionals. If the population is considered a group of players that is separate from the problem at hand, it will not be possible to take it into account in setting up action strategies, to the extent that the aim of such work is to better define people's expectations in terms of the quality of care and arrive at a better understanding of their behaviour. During the discussions, the teachers felt that priority had to be given to separating the group of teachers from that of intern supervisors in order to better highlight the particularities and role of the field training. The school managements revealed their specific role in this problem of mismatches. Indeed, the teachers' and supervisors' behaviour is strongly linked to their own behaviour in dealing with changes [ 9 ]. We have presented one or the other behaviour for each of these groups of players as examples only. Environmental diagnosis This diagnosis allows for the factors that are linked to the environment and are direct causes of the epidemiological and social diagnosis. In a context such as that of the DRC, geopolitical and socioeconomic factors head the list, along with the health structures' inaccessibility. To take a more constructive approach without denying reality, it is necessary to focus the analysis of this diagnosis on the more targeted problem of the inadequacies in the training sector. This reveals variables that are more controllable for the levels that are concerned and that everyone agrees are connected to the problem. These are: the learning environment, teaching environment, class hours that facilitate or hamper certain types of learning, etc. Educational and motivational diagnosis The educational diagnosis enables one to home in on the educational and motivational determinants, which must not be overlooked when one goes on to an interventional phase. To the extent that the systemic approach gives significance to each group of players (teachers, learners and others) as well, as is the case in the DRC, it is fully possible to set up a frame of analysis, assessment and action-research that presents the variables and determinants in a PRECEDE model that are specific to each specific group (action-research framework). This is what was done in the DRC to follow the changes in teaching practices, in conjunction with each intervention that was identified, that were made in the specific group of teachers. The results show that it is relatively easy to separate the educational determinants from each other in order to facilitate subsequent reflection about the strategic action to take. The predisposing factors that concern knowledge, experience, attitudes, perceptions and representations have a key place in relation to the behaviours of the players of interest to the Directorate for Education. This construction shows clearly that the training given is usually concerned with knowledge only and generally does not make use of the learners' life experiences. The other important result is to be able to visualize the place of representations in a conceptual framework that will likewise be used for the action. For example, there are the various representations of learning theories when it comes to teaching methods or unfounded beliefs about the quality of care. Specific models exist that enable one to delve much deeper into perceptions and beliefs [ 10 , 11 ]. These are complementary research models. When we are seeking to develop a tool that can be used to construct an operational model for strategic choices of action to take on a high institutional level, the possibility of providing this place for representations and beliefs is already vital and elucidating. The enabling factors in terms of actual competences (skilfulness, know-how and behaviour) are too often disregarded and underestimated in interventions. Incorporating them in this model thus enables the directorate in charge of this branch of education to check to what extent the projects, programmes and other support measures consider this priority strand in terms of development independence. The reinforcing factors, which are sometimes also referred to as facilitating factors, are the determinants that act upon the positive feedback loops. The importance that all the players give to this type of variable in constructing the model confirmed the need that the directorate had already felt to find means to set up long-range monitoring mechanisms for the various activities engendered by the programme or by some more specific projects. The model contains a certain number of variables. It is clear that it can be enriched in the course of the process through its use and the players' better discovery and gradual appropriation of its features. Institutional diagnosis In terms of results, the institutional diagnosis requires analysing the situation at the organizational level that corresponds to the level of the model's application. This is a national health science education programme under the Ministry of Health. As such, the institutional diagnosis stresses essential strategic variables if one wants to work on a well-knit, comprehensive set of changes. It thus entails the need to analyse the institutional standards when it comes to inspections and assessment, but also those governing health system management and health sector human resource quality management (for example, the existence or lack of a Nursing Board). This is also the level on which we shall discuss how the programme dovetails with other variables and determinants. PROCEED results Before strategic thinking can be put into place based on this situational analysis, it is possible to go on to a more dynamic reading of the relations between determinants and variables. So it is that the DRC's Ministry of Health directorate in charge of health science education foresees a certain number of strategic axes for action. The aim of the action is the problem's translation into an objective form. The directorate thus considered its main goal to be to improve the match between what is learnt in schools and health professionals' needs and the population's expectations. To better understand how the reading of the conceptual model brought us to the action strategies, it is useful to stress an intermediate step that is summarized in Figure 2 . Figure 2 Visual summary of considered actors A natural adaptation of the PRECEDE model was to define the groups of players by their behaviour. In this way, we obtained a better picture of the division of responsibilities to achieve a common goal and evidence of the need for interdisciplinary work [ 12 ]. A comprehensive reading of the PRECEDE model points us towards a strategic choice that integrates an institutional and educational approach from education with an epidemiological and social approach from health and welfare. The players in their respective environments are located between the two. This model shows the need to find a common thread between education and health needs that allows for the place and role of each group of players in their context. The results bring to the fore a number of behaviours that attest to a lack of independence, absenteeism, lack of collaboration, failure to connect theory and practice, a lack of communication, ignorance of the teacher's role, etc., depending on the group. Examination of these results prompts us to stress the importance that must be given to the learning environment and, when it comes to action strategies, the importance to give to a learning environment that is in tune with the strategic axes that are selected, in this part just as in the other parts of the situation's analysis. Given this finding and the need to link the educational and institutional diagnosis with the health problem (seen as an appropriate form for education), one proposed strategic hypothesis is to favour learning techniques that are based on active teaching methods [ 13 ]. In going consistently through the various diagnoses and organizational levels, this choice led the education directorate to think about changing its programmes and standards so as to base them on novel teaching concepts such as skills-based learning [ 14 ] and setting a skills reference framework on the basis of in-depth research done with the entire set of clearly identified target populations. The reading of these results in terms of strategic action reveals the need to bolster the analytical and planning process that already exists within the directorate to pay more attention to the educational and environmental determinants for all the target populations concerned. To our mind, the success of the expected changes in terms of narrowing the gap between the "supply" and the "demand" hinges on this. The following diagram summarizes the strategic axes that the Directorate for Health Science Education chose to achieve this objective on the basis of the PRECEDE analysis (Figure 3 ). Figure 3 Strategic axes for action This figure reveals four axes to be reinforced: • to reinforce communication and coordination in conjunction with the other reinforcing factors: the pilot schools' teaching method committees, teaching monitoring and feedback, the setting-up of networks, etc.; • to develop methods to enhance the learner's autonomy: active teaching, constructivist approach, interdisciplinary, critical spirit, etc.; • to foster a learning environment that enables the learners to acquire knowledge: library, teaching materials, computer learning, computerized documentation centre, etc.; • to provide institutional and structural support: standards and curricula in tune with teaching and organizational innovations and skills targets that fit health professionals' needs and meet the community's expectations. The discussion will take place on two levels – the operational and the conceptual. On the operational level, we feel it is interesting not to dwell on the presentation of the model as it could have been applied, but on its actual application. The results are presented so as to allow the reader to understand how to organize the problems that are felt to exist in health science education in the DRC. Even if the Directorate for Health Science Education is well aware of its problems, the systemic modelling of the interconnected variables and populations seems to give it a conceptual and operational tool that is useful on various levels, as follows: • tool for dynamic analysis of the situation with regular updates; • tool for systemic planning that also enables the directorate to put forward arguments in dealing with donors and NGOs in the sector; • assessment tool that gives more importance to assessment criteria such as cohesiveness, consistency, relevance, appropriation, and comprehensiveness, i.e., process criteria; • research and evaluation tool that can also promote a more quantitative approach to analysing the relations between variables and various diagnoses or within the same diagnosis; • a dialogue-enhancing tool, for it gives the groups of players involved a vision of the planned change and a common objective. To sum up, this is a tool that provides a certain guarantee that the strategy development process is informed, meets the needs and is complete [ 15 ]. The list of these advantages is obviously based on some baseline conditions: a participatory process in which the model is developed and operates and the appropriation of the concepts that subtend the model [ 16 ]. Even though it was more difficult to describe how the intervention strategies are set, based on the construction of this model, we should like to stress that a complete analysis of the situation that is based on this systemic approach usually reveals the relevant strategic axes on its own and despite the protagonists' limited ability to synthesize the situation. On the conceptual level, the discussion will revolve around Figure 4 . Figure 4 Multi-field vision of the change We observed through the PRECEDE analysis and then the PROCEED strategic reflection phase that many disciplines converged in order to lead us to this hypothesis and a common objective of needs-matching. Indeed, when action is carried out it will be a matter of achieving a gradual advance that occurs along the (horizontal and vertical) strategic axes defined earlier in this article. Moreover, we are confronted with strategic choices that involve at least three dimensions: a public health approach, an education approach and a sociological approach. These three dimensions are part of the data collection process's success, as well as the success of the strategic choices that follow. This reinforces the fact that the PRECEDE PROCEED model comes from the development of an approach aimed at meeting the need for education and health promotion tools and methods. So it is that we see numerous applications of this process in technical health education establishments that spring from a true systemic analysis of the problems with full mastery of a structuring capacity, unlike some other models such as causal analysis (17). Similarly, we can consider that defining a problem at an institutional and organizational level also requires the identification and involvement of all the parties concerned. We can also consider that the tools that help to understand the relations between elements and insist on a better search for behavioural determinants are prerequisites for organizational learning that has groups of players interact with each other. This is all the more true if the change that is ultimately expected (a match between two sides of the equation) is contingent on changes in the players' behaviour and practices, as is the case of health education. In terms of limits of this research, it targets the analysis of an inadequacy within human resources' management in health, which is that of training of nurses from professional technical levels. Other levels of inadequacies are worthy to be analysed in a complementary way relating to other health professionals, the sectors of health and education planning. The Green model is complementary to the use of methodological dynamic references much as the management of the project cycle focuses on managing interventions or projects whose aim is to contribute to changing a situation from unsatisfactory to satisfactory. Its use within the framework of the project could obtain more means while enabling developments relating to action research. In this context, the contribution from other disciplines, such as psychology, could be reinforced. Conclusions With regard to the three levels of starting research – the theoretical, methodological and empirical – PRECEDE PROCEED analysis is a model that can be applied to varied situations and problems, although it must be used participatively and proactively in order to enhance its utility in specific circumstances as a personal transfer tool. On the empirical level, the will of all actors – and the Ministry of Health in particular – to have a clear vision of the projected change and manner of reaching that point, while integrating the complexity, was the element carrying the process. We advance the hypothesis that L. Green's systemic approach may become one of a set of active methods, such as problem-based learning, cooperative learning, or even project-based learning, to transfer to learners in nursing schools and sections in the DRC. Indeed, the ability to analyse and synthesize, but also to carry out education and health promotion actions, is essential. List of abbreviations DRC: Democratic Republic of Congo PRECEDE: Predisposing, Reinforcing and Enabling Constructs in Educational/Environment Diagnosis and Evaluation PROCEED: Policy, Regulatory and Organizational Constructs in Educational and Environmental Development Competing interests The authors declare that they have no competing interests. Authors' contributions FP is responsible for this research. She initiated the project in DRC. She is a specialist in public health and pedagogy. She participated in the design and coordination of the study and drafted the manuscript. JK and DB are two teachers in charge of the reform of the nursing programme in DRC. They set up the collection of data for this study and finalized the analysis. YC and AL participated in the design of the study and the adaptation of de Green's model in the field of nursing training. YC participated in writing the manuscript. MG specializes in management and pedagogy. She conceived the study with FP and participated in its design. DP took part in the elaboration of the methodology. She brought an expertise in health promotion and wrote part of the manuscript. All authors read and approved the final manuscript. Supplementary Material Additional File 1 Table 1. PRECEDE model for health science teaching in the DRC oversized table Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC543448.xml

423155

Escape Velocity: Why the Prospect of Extreme Human Life Extension Matters Now

Should we be considering the social and economic ramifications of a society where life-span could be limitless?

The biogerontologist David Sinclair and the bioethicist Leon Kass recently locked horns in a radio debate ( http://www.theconnection.org/shows/2004/01/20040106_b_main.asp ) on human life extension that was remarkable for one thing: on the key issue, Kass was right and Sinclair wrong. Sinclair suggested, as have other experts, including his mentor Lenny Guarente and the National Institute on Aging advisory council member Elizabeth Blackburn, that Kass and other bioconservatives are creating a false alarm about life extension, because only a modest (say, 30%) increase in human life span is achievable by biomedical intervention, whereas Kass's apprehensions concern extreme or indefinite life extension. Kass retorted that science isn't like that: modest success tends to place the bit between our teeth and can often result in advances far exceeding our expectations. Coping with Methuselah consists of seven essays, mostly on the economics of life extension but also including one essay surveying the biology of aging and one on the ethics of life extension. The economic issues addressed are wide ranging, including detailed analysis of the balance between wealth creation by the employed and wealth consumption in pensions and health care; most chapters focus on the United States, but the closing chapter discusses these issues in a global context. Each essay is followed by a short commentary by another distinguished author. Within their own scope, all of these contributions are highly informative and rigorous. Dishearteningly, however, all echo Sinclair's views about the limited prospects for life extension in the coming decades. In my opinion, they make three distinct oversights. The first concerns current science. Sinclair and several other prominent gerontologists are presently seeking human therapies based on the long-standing observation that lifelong restriction of caloric intake considerably extends both the healthy and total life span of nearly all species in which it has been tried, including rodents and dogs. Drugs that elicit the gene expression changes that result from caloric restriction might, these workers assert, extend human life span by something approaching the same proportion as seen in rodents—20% is often predicted—without impacting quality of life, and even when administered starting in middle age. They assiduously stress, however, that anything beyond this degree of life extension is inconceivable. I agree with these predictions in two respects: that the degree of life extension achieved by first-generation drugs of this sort may well approach the (currently unknown) amount elicitable by caloric restriction itself in humans, and that it is unlikely to be much exceeded by later drugs that work the same way. In two other ways, however, I claim they are incorrect. The first error is the assumption of proportionality: I have recently argued ( de Grey 2004 ), from evolutionary considerations, that longer-lived species will show a smaller maximal proportional life-span extension in response to starvation, probably not much more than the same absolute increase seen in shorter-lived species. The second error is the assertion that no other type of intervention can do better. In concert with other colleagues whose areas of expertise span the relevant fields, I have described ( de Grey et al. 2002 , 2004 ) a strategy built around the actual repair (not just retardation of accumulation) of age-related molecular and cellular damage—consisting of just seven major categories of ‘rejuvenation therapy’ ( Table 1 )—that appears technically feasible and, by its nature, is indefinitely extensible to greater life spans without recourse to further conceptual breakthroughs. Table 1 Strategies for Engineered Negligible Senescence The seven categories of accumulating molecular or cellular side effects of metabolism whose possible contribution to age-related mammalian physical or cognitive decline is an established school of thought within contemporary biogerontology, and the foreseeable therapies that can repair or obviate them The second oversight made both by the contributors to Coping with Methuselah and by other commentators is demographic. Life expectancy is typically defined in terms of what demographers call a period survival curve, which is a purely artificial construction derived from the proportions of those of each age at the start of a given year who die during that year. The ‘life expectancy’ of the ‘population’ thus described is that of a hypothetical population whose members live all their lives with the mortality risk at each age that the real people of that age experienced in the year of interest. The remaining life expectancy of someone aged N in that year is more than this life expectancy minus N for two reasons: one mathematical (what one actually wants, roughly, is the age to which the probability of survival is half that of survival to N ) and one biomedical (mortality rates at each age, especially advanced ages, tend to fall with time). My spirits briefly rose on reading Aaron and Harris's explicit statement (p. 69) of the latter reason. Unfortunately, they didn't discuss what would happen if age-specific mortality rates fell by more than 2% per year. An interesting scenario was thus unexplored: that in which mortality rates fall so fast that people's remaining (not merely total) life expectancy increases with time. Is this unimaginably fast? Not at all: it is simply the ratio of the mortality rates at consecutive ages (in the same year) in the age range where most people die, which is only about 10% per year. I term this rate of reduction of age-specific mortality risk ‘actuarial escape velocity’ (AEV), because an individual's remaining life expectancy is affected by aging and by improvements in life-extending therapy in a way qualitatively very similar to how the remaining life expectancy of someone jumping off a cliff is affected by, respectively, gravity and upward jet propulsion ( Figure 1 ). Figure 1 Physical and Actuarial Escape Velocities Remaining life expectancy follows a similar trajectory whether one walks off a cliff or merely ages: the time scales differ, but one's prognosis worsens with time. Mild mitigation of this (whether by jet propulsion or by rejuvenation therapies) merely postpones the outcome, but sufficiently aggressive intervention overcomes the force of gravity or frailty and increasingly distances the individual from a sticky end. Numbers denote plausible ages, at the time first-generation rejuvenation therapies arrive, of people following the respective trajectories. The escape velocity cusp is closer than you might guess. Since we are already so long lived, even a 30% increase in healthy life span will give the first beneficiaries of rejuvenation therapies another 20 years—an eternity in science—to benefit from second-generation therapies that would give another 30%, and so on ad infinitum. Thus, if first-generation rejuvenation therapies were universally available and this progress in developing rejuvenation therapy could be indefinitely maintained, these advances would put us beyond AEV. Universal availability might be thought economically and sociopolitically implausible (though that conclusion may be premature, as I will summarise below), so it's worth considering the same question in terms of life-span potential (the life span of the luckiest people). Figure 1 again illustrates this: those who get first-generation therapies only just in time will in fact be unlikely to live more than 20–30 years more than their parents, because they will spend many frail years with a short remaining life expectancy (i.e., a high risk of imminent death), whereas those only a little younger will never get that frail and will spend rather few years even in biological middle age. Quantitatively, what this means is that if a 10% per year decline of mortality rates at all ages is achieved and sustained indefinitely, then the first 1000-year-old is probably only 5–10 years younger than the first 150-year-old. The third oversight that I observe in contemporary commentaries on life extension, among which Coping with Methuselah is representative, is the most significant because of its urgency. First-generation rejuvenation therapies, whenever they arrive, will surely build on a string of prior laboratory achievements. Those achievements, it seems to me, will have progressively worn down humanity's evidently desperate determination to close its eyes to the prospect of defeating its foremost remaining scourge anytime soon. The problem (if we can call it that) is that this wearing-down may have been completed long before the rejuvenation therapies arrive. There will come an advance—probably a single laboratory result—that breaks the camel's back and forces society to abandon that denial: to accept that the risk of getting one's hopes up and seeing them dashed is now outweighed by the risk of missing the AEV boat by inaction. What will that result be? I think a conservative guess is a trebling of the remaining life span of mice of a long-lived strain that have reached two-thirds of their normal life span before treatment begins. This would possess what I claim are the key necessary features: a big life extension, in something furry and not congenitally sick, from treatment begun in middle age. It is the prospect of AEV, of course, that makes this juncture so pivotal. It seems quite certain to me that the announcement of such mice will cause huge, essentially immediate, society-wide changes in lifestyle and expenditure choices—in a word, pandemonium—resulting from the anticipation that extreme human life extension might arrive soon enough to benefit people already alive. We will probably not have effective rejuvenation therapies for humans for at least 25 years, and it could certainly be 100 years. But given the present status of the therapies listed in Table 1 , we have, in my view, a high probability of reaching the mouse life extension milestone just described (which I call ‘robust mouse rejuvenation’) within just ten years, given adequate and focused funding (perhaps $100 million per year). And nobody in Coping with Methuselah said so. This timeframe could be way off, of course, but as Wade notes (p. 57), big advances often occur much sooner than most experts expect. Even the most obvious of these lifestyle changes—greater expenditure on traditional medical care, avoidance of socially vital but risky professions—could severely destabilise the global economy; those better versed in economics and sociology than I would doubtless be even more pessimistic about our ability to negotiate this period smoothly. Overpopulation, probably the most frequently cited drawback of curing aging, could not result for many decades, but the same cannot be said for breadth of access irrespective of ability to pay: in a post-9/11 world, restricted availability of rejuvenation therapies resembling that seen today with AIDS drugs would invite violence on a scale that, shall we say, might be worth trying to avoid. Am I, then, resigned to a future in which countless millions are denied many decades of life by our studied reluctance to plan ahead today? Not quite. The way out is pointed to in Lee and Tuljapurkar's (1997) graph of the average wealth consumed and generated by an individual as a function of age, reproduced in Coping with Methuselah (p. 143). Once AEV is achieved, there will be no going back: rejuvenation research will be intense forever thereafter and will anticipate and remedy the life-threatening degenerative changes appearing at newly achieved ages with ever-increasing efficacy and lead time. This will bring about the greatest economic change of all in society: the elimination of retirement benefits. Retirement benefits are for frail people, and there won't be any frail people. The graph just mentioned amply illustrates how much wealth will be released by this. My hope, therefore, is that once policy makers begin to realise what's coming they will factor in this eventual windfall and allocate sufficient short-term resources to make the period of limited availability of rejuvenation therapies brief enough to prevent mayhem. This will, however, be possible only if such resources begin to be set aside long enough in advance—and we don't know how long we have.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC423155.xml

529322

A Neuron Survival Protein May Give Directions, Too

Anyone who's ever had a physical exam knows the monosynaptic stretch reflex arc. When the doctor taps your patellar tendon with a hammer, your quadriceps muscle briefly stretches and your knee responds with a quick kick. This involuntarily reflex is mediated by muscle spindles, specialized muscle structures containing both proprioceptive (sensory) and motor neurons. Proprioceptive neurons send information about how much the muscle is stretched to the spinal cord, and motor neurons emanating from the spinal cord tell the muscle to contract, which corrects the stretch. This reflex circuit is established in the developing embryo, when neurons migrate through and around developing tissues and send their axons to their signaling targets. Many neurons will successfully extend their axons into a target—which might be a muscle, another neuron, or some other tissue—but not all survive the process. Whether a neuron lives or dies depends on a family of growth factor proteins called neurotrophins. If a sensory neuron doesn't get enough neurotrophin-3 (NT-3), it will die. Proper development of the sensory/motor circuit also depends on NT-3, which is expressed in limb buds, muscle spindles, and the ventral spinal cord: muscle spindle development depends on sensory axons, and motor neuron connections depend on developing limb buds. It has not been clear, however, whether NT-3 simply ensures the survival of proprioceptive neurons or whether it also helps establish the proprioceptive reflex arc. In a new study, Reha Erzurumlu and colleagues demonstrate a clear role for NT-3 in axon guidance. Developing nerves in the mouse spinal cord Attempts to investigate the axon guidance theory have been difficult since sensory neurons die without NT-3. To circumvent this problem, the authors developed a “double knockout” mouse model that deletes both the NT-3 gene and the apoptosis-promoting gene Bax , which activates the cell death pathway for sensory neurons. This system removes NT-3 signaling without killing the sensory neurons, so the researchers can investigate any effects NT-3 may have on axon behavior. Erzurumlu and colleagues showed that sensory neurons, in the absence of NT-3 signaling, project to the right places but never reach their final destination. In normal development, sensory neuron axons travel into the ventral spinal cord and form synapses with motor neuron dendrites in the ventral horn to establish the reflex arc circuit. In the double knockout mice, sensory neurons manage to extend into the spinal cord, but then get lost; they can't find the ventral horn, so they never form a synapse with the motor neuron dendrites. The failure to establish connections between the sensory axons and motor neurons in mice lacking NT-3, the authors argue, indicates that NT-3 is required for proper axon targeting. Similarly, deprived of NT-3 signaling, sensory neuron axons fail to reach their ultimate targets in peripheral muscle. They project down toward the muscle but don't recognize the muscle and thus cannot enter or innervate it. Consequently, the muscles' sense organs, the muscle spindles, cannot differentiate. The authors argue that these findings, along with the results of tissue culture experiments, show that NT-3 acts as a short-distance cue for proprioceptive axons, which travel in the right direction but ultimately lose their way without NT-3. Altogether these results show that proprioceptive axons require NT-3 not just for survival, but to reach their proper endpoints in the peripheral and central nervous system. NT-3 also helps proprioceptive axons initiate muscle innervation and spindle differentiation. Researchers developing therapies to treat neurodegenerative injuries have increasingly focused their attentions on growth factors like NT-3. By identifying the molecules and mechanisms that establish connections between sensory and motor neurons during development, it may be possible to engage similar processes to attenuate neurodegeneration and even repair damaged nerves.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC529322.xml

526777

Toxoplasma and coxiella infection and psychiatric morbidity: A retrospective cohort analysis

Background It has been suggested that infection with Toxoplasma gondii is associated with slow reaction and poor concentration, whilst infection with Coxiella burnetii may lead to persistent symptoms of fatigue. Methods 425 farmers completed the Revised Clinical Interview Schedule (CIS-R) by computer between March and July 1999 to assess psychiatric morbidity. Samples of venous blood had been previously collected and seroprevalence of T. gondii and C. burnetii was assessed. Results 45% of the cohort were seropositive for T. gondii and 31% were positive for C. burnetii . Infection with either agent was not associated with symptoms reflecting clinically relevant levels of concentration difficulties, fatigue, depression, depressive ideas or overall psychiatric morbidity. Conclusions We do not provide any evidence that infection with Toxoplasma gondii or Coxiella burnetii is associated with neuropsychiatric morbidity, in particular with symptoms of poor concentration or fatigue. However, this is a relatively healthy cohort with few individuals reporting neuropsychiatric morbidity and therefore the statistical power to test the study hypotheses is limited.

Background It has been suggested that infections with the zoonoses Toxoplasma gondii and Coxiella burnetii may lead to long term neuropsychiatric morbidity. More specifically, T. gondii infection is hypothesised to be associated with slow reaction and poor concentration [ 1 , 2 ] whilst C. burnetii infection is reported to be associated with persistent symptoms of fatigue for up to ten years following exposure [ 3 - 5 ]. Exposure to the parasitic protozoon T. gondii is common in the UK population as a whole (40% to 50%) whereas exposure to the rickettsia-like C. burnetii leading to Q fever is relatively rare in urban populations [ 6 - 8 ]. We have examined data previously collected from a farm-based occupational cohort recruited in three areas of England to test these hypotheses. Furthermore, given the high risk of suicide amongst farmers as an occupational group[ 9 ], we also investigated associations between infection with either organism and symptoms of depression or depressive ideas. Statistical associations were examined between measures of lifetime exposure to T. gondii and C. burnetii and current psychiatric morbidity measured by the Revised Clinical Interview Schedule (CIS-R). Methods Sample A representative cohort of 606 farmers, farmworkers and family members has been recruited since 1991 in three areas of England to investigate occupational risk factors for zoonoses[ 10 ]. A random sample of farmers was drawn from the Ministry of Agriculture, Fisheries and Food June Agricultural Census lists of agricultural holdings[ 11 ], and each farmer could then nominate a further adult on the same farm holding (usually his wife). Seventy-seven per cent of the cohort were still enrolled in May 1998 and of these, 425 (91%) completed the CIS-R by computer between March and July 1999[ 12 ]. Psychiatric morbidity data The computer-administered version of the CIS-R was used to assess the prevalence of symptoms of neurotic psychopathology in the week prior to interview[ 13 ]. The CIS-R is made up of fourteen sections, each covering a particular area of neurotic symptoms. For this study we utilized data from the sections relating to fatigue, concentration difficulties, depression and depressive ideas as assessment of psychiatric outcome. Individual symptoms are regarded as clinically relevant if they have a score of two or more (range zero to four, or five for section on depressive ideas). Summed scores from all fourteen sections range from zero to fifty seven, the overall threshold for clinically significant psychiatric morbidity is twelve. The time taken to complete the questionnaire ranged from ten to thirty minutes, due to the filtering nature of the questions. As reported previously, the prevalence of clinically relevant levels of each of these neurotic symptoms was relatively low[ 9 ]. Fifteen percent of farmers reached the threshold for a clinically relevant level of symptoms of fatigue (n = 62). Approximately 5% of the farmers reported symptoms of either concentration difficulties (n = 22) or depressive ideas (n = 23), and 4% reported symptoms of depression (n = 18). Approximately 6% of farmers reported significant general psychiatric morbidity (n = 25). Seroprevalence data At enrolment, 10 ml of venous blood was taken from all subjects. Samples were screened for Toxoplasma gondii at Hereford PHL using the Eiken latex agglutination test (cut off titre 1/32). Seroconverters were confirmed by PHLS Toxoplasma Reference Laboratory at Swansea PHL using the dye test and by IgM ELISA. Serum IgG specific antibody levels for Coxiella burnetii phase II antigen were estimated at Bristol Public Health Laboratory (PHL) using an indirect immunofluorescence test. Serum with a reciprocal titre of 32 or more were taken as positive. Seroprevalence data for T. gondii and C. burnetii were available for 370 and 422 individuals respectively. Statistical analysis All analyses allowed for clustering by farm holding in the survey data (Stata Version 6.0, StataCorp, College Station, TX, USA). Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated using logistic regression and were adjusted for sex and age in ten year bands. Results In total, 45% (95% CI 40% to 50%) of the cohort were seropositive for T. gondii and 31% (95% CI 26% to 36%) were positive for C. burnetii . Forty six individuals were seropostivie for both. Exposure to either infection did not differ significantly between men and women ( T. gondii seroprevalence 47% vs. 41%; C. burnetii seroprevalence 32% vs. 28%). T. gondii seroprevalence increased significantly with age (<30 years 9%, 30–39 years 29%, 40–49 years 33%, 50–59 years 54%, 60–69 years 57%, 70+ years 56%; χ 2 = 25.8, df = 5, P = 0.0002). C. burnetii seroprevalence was not associated with age (<30 years 38%, 30–39 years 28%, 40–49 years 36%, 50–59 years 27%, 60–69 years 29%, 70+ years 36%). Those seropositive for T. gondii were not more likely to report symptoms reflecting clinically relevant levels of fatigue, concentration difficulties, depression, depressive ideas or overall psychiatric morbidity than those who were seronegative (Table 1 ). Similarly, evidence of infection by C. burnetii was not significantly associated with reporting of clinically relevant levels of any of these symptoms. Indeed, although based on very limited numbers, a greater percentage of farmers who were seronegative rather than seropositive for C. burnetii reported psychiatric symptoms. Furthermore, neither infection was associated with an increased risk of any studied psychiatric outcome after adjusting for age and sex (Table 2 ). Table 1 Prevalence of self-reported psychiatric symptoms in relation to infection by T. gondii and C. burnetii T. gondii C. burnetii Seropositive Seronegative Seropositive Seronegative N = 166 N = 204 N = 130 N = 292 N (%) Fatigue 27 (16.3) 33 (16.2) 18 (13.9) 43 (14.7) Concentration difficulties 9 (5.4) 13 (6.4) 3 (2.3) 19 (6.5) Depression 7 (4.2) 11 (5.4) 3 (2.3) 15 (5.1) Depressive ideas 10 (6.0) 13 (6.4) 5 (3.9) 18 (6.2) Psychiatric morbidity 10 (6.0) 15 (7.4) 6 (4.6) 19 (6.5) Table 2 Odds ratios for self-reported psychiatric symptoms in relation to T. gondii and C. burnetii T. gondii C. burnetii OR (95% CI) OR (95% CI) Fatigue (n = 62) Seronegative 1.00 1.00 Seropositive 1.18 (0.66–2.14) 0.92 (0.52–1.64) Concentration difficulties (n = 22) Seronegative 1.00 1.00 Seropositive 0.97 (0.41–2.28) 0.32 (0.10–1.06) Depression (n = 18) Seronegative 1.00 1.00 Seropositive 0.81 (0.29–2.30) 0.47 (0.13–1.67) Depressive ideas (n = 23) Seronegative 1.00 1.00 Seropositive 1.11 (0.44–2.77) 0.62 (0.23–1.69) Psychiatric morbidity(n = 25) Seronegative 1.00 1.00 Seropositive 0.88 (0.38–2.04) 0.71 (0.28–1.84) Odds ratios account for sample clustering by farm holding Odds ratios were adjusted for sex and age in 10-year bands (<30, 30–39, 40–49, 50–59, 60–69, 70+) Discussion This study does not provide evidence to support the hypothesis that infection by Toxoplasma gondii is associated with difficulties in concentration. Given the relatively common exposure to latent toxplasmosis in both the general population and in certain occupational cohorts, even a relatively weak association with neuropsychiatric outcome might be of potential public health interest. However, no strong evidence to support such an association has been reported to date. Havlicek and colleagues [ 1 ] reported significantly longer reaction times for completion of a computerised version of a psychomotor test amongst 60 Toxoplasma -positive individuals compared to 56 Toxoplasma -negative individuals, although the difference in reaction time was only up to 17 miliseconds. The authors interpreted this possible behavioural change as a manipulation activity to promote transmission of the parasite; delayed reaction times in rodents could increase the chance of transmission into a definitive host such as the cat. Flegr and collagues [ 2 ] investigated this hypothesis further by comparing the seroprevalence of latent toxoplasmosis in 146 subjects involved in traffic accidents (assumed to have delayed reaction) and 446 members of the general public in the same geographical area. Subjects with latent toxoplasmosis were 2.65 times more likely (95% CI 1.76–4.01) to be identified as having a traffic accident than those who were seronegative. However the study was limited by the sampling of both cases and controls relying on availability of archived blood samples for serological testing of toxoplasmosis thus increasing the possibility of selection bias, together with inadequate consideration of possible confounders. Furthermore there is no evidence from this study to support the hypothesis that infection by Coxiella burnetii is associated with symptoms of fatigue. Wildman and colleagues [ 5 ] have previously reported that symptoms of fatigue were more common in 77 individuals assessed ten years after exposure to Q fever in a UK outbreak than in matched unexposed controls (65% vs 35% P < 0.001). Their study benefited from a comprehensive assessment of fatigue, a matched design to control for age, sex and smoking status, and a relatively high response rate amongst those exposed to Q fever (84%). However only 36% of the matched controls participated in the study, suggesting a possible selection bias. Finally due to the nature of self-report questionnaires and knowledge of study hypotheses the possibility of response bias could not be ruled out. Strengths and limitations Exposure to both T. gondii and C. burnetii was common in this occupational cohort of farmworkers which is not surprising. Indeed the exposure to toxoplasma is common in the UK population as a whole (40% to 50%)[ 7 ], whereas occupational exposure is more important in the epidemiology of Q fever, with prevalence amongst farmworkers being two to three times higher than a comparison group of ambulance and police workers[ 8 ]. This survey benefited from using the CIS-R as a standardised assessment suitable for lay interviewers in assessing minor psychiatric disorder in an occupational setting. The computerised CIS-R assessment provides an easy, quick, inexpensive yet thorough assessment which is acceptable to interviewees and also helps to eliminate observer bias. The prevalence of psychiatric symptoms was lower than expected in this cohort of farm workers[ 10 ], indicating that this is a relatively healthy occupational cohort with only very few individuals reporting clinically relevant levels of symptoms. Therefore unfortunately the statistical analyses are limited in power to test the study hypotheses. Although the response rate for the mental health survey was relatively high, it is possible that those subjects previously exposed to T. gondii or C. burnetii and who subsequently developed symptoms might have been more likely to drop out of the cohort either before recruitment or before this more recent survey ('healthy worker effect'). If this were the case we might underestimate the strength of association between exposure and psychiatric outcome. Finally we have no indicator of duration of infection for those individuals who are currently seropositive for either infectious agent which might affect the observed association if a long lag time is required between exposure and presentation of symptoms. Conclusions We do not provide any evidence that infection with Toxoplasma gondii or Coxiella burnetii is associated with neuropsychiatric morbidity, in particular with symptoms of poor concentration or fatigue. List of abbreviations T. gondii – Toxoplasma gondii C. burnetii – Coxiella burnetii CIS-R – Revised Clinical Interview Schedule OR – odds ratio 95% CI – ninety five percent confidence interval Competing interests The authors declare that they have no competing interests. Authors' contributions HT completed the statistical analyses and drafted the manuscript, DT and RS conceived of the PHLS Farm Cohort study and participated in its design and coordination, GL developed the psychiatric assessments and advised in their use, AS advised on the study hypotheses under investigation. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC526777.xml

548509

Injury morbidity in an urban and a rural area in Tanzania: an epidemiological survey

Background Injuries are becoming a major health problem in developing countries. Few population based studies have been carried out in African countries. We examined the pattern of nonfatal injuries and associated risk factors in an urban and rural setting of Tanzania. Methods A population-based household survey was conducted in 2002. Participants were selected by cluster sampling. A total of 8,188 urban and 7,035 rural residents of all ages participated in the survey. All injuries reported among all household members in the year preceding the interview and resulting in one or more days of restricted activity were included in the analyis. Results A total of 206 (2.5%) and 303 (4.3%) persons reported to have been injured in the urban and rural area respectively. Although the overall incidence was higher in the rural area, the incidence of major injuries (≥ 30 disability days) was similar in both areas. Males were at a higher risk of having an injury than females. Rural residents were more likely to experience injuries due to falls (OR = 1.6; 95% CI = 1.1 – 2.3) and cuts (OR = 4.3; 95% CI = 3.0 – 6.2) but had a lower risk of transport injuries. The most common causes of injury in the urban area were transport injuries and falls. In the rural area, cuts and stabs, of which two thirds were related to agriculture, formed the most common cause. Age was an important risk factor for certain types of injuries. Poverty levels were not significantly associated with experiencing a nonfatal injury. Conclusion The patterns of injury differ in urban and rural areas partly as a reflection of livelihoods and infrastructure. Rural residents are at a higher overall injury risk than urban residents. This may be important in the development of injury prevention strategies.

Background Injuries have been recognised as a major public health problem in both developed and less developed countries [ 1 ]. It is generally acknowledged that this problem is growing rapidly in sub-Saharan Africa [ 2 ]. Studies in Tanzania show that injuries are an important cause of death among adults [ 3 ], and accounted for 12% of all admissions at the national hospital in the country's largest city [ 4 ]. Studies on the magnitude of injuries and the groups at risk, have been conducted world-wide, and especially in developed countries. Hospital based studies, which are commonly reported from developing countries, presumably provide a representative picture of the prevalence and incidence of serious injury, but only a partial picture of the circumstances in which injuries occur. Given the limited access to hospital care in poor countries, however, data based on health facility data are not likely to be representative. In contrast, population-based studies are costly and rarely carried out particularly on topics such as injury, which are not high on the public health agenda in developing countries at present. Several studies have been conducted in high-income countries to examine factors associated with injury morbidity [ 5 - 7 ]. In developing countries, a number of population-based studies on nonfatal injuries have been done [ 8 - 14 ]. In order to understand the circumstances and risk factors associated with nonfatal injuries, we conducted a community-based study in an urban and a rural location of Tanzania. We describe injury patterns in both settings. We also investigate demographic and socioeconomic factors associated with nonfatal injuries. Methods Study area The survey was conducted in Dar es Salaam city (an urban area) and Hai District (a rural area). These areas are part of a health and demographic surveillance system carried out by the Adult Morbidity and Mortality Project (AMMP) in six districts in Tanzania from 1992 to 2004. One aim of the project was to measure rates and causes of morbidity and mortality. At the time this investigation was carried out, the areas were being prospectively monitored through repeated censuses to ascertain the resident population at risk. Deaths were recorded through an active reporting system and probable cause of death determined by a validated verbal autopsy [ 15 , 16 ]. Dar es Salaam city lies on the east coast of Tanzania. The AMMP demographic surveillance population was situated in two of the city's three municipalities. These areas contained 8 'branches' (an urban administrative unit) and covered 63,330 persons in 15,269 households living in urban and peri-urban neighbourhoods. Hai District lies on the South-Western slopes of Mount Kilimanjaro in Northern Tanzania. The AMMP demographic surveillance area in Hai covered 51 out of 61 villages in the district and around 62% of the total district population (159,906 persons in 40,238 households). Agriculture, livestock keeping and commercial mining are the main economic activities there. Details of the study population have been described elsewhere [ 15 , 16 ]. Sampling procedure In the urban surveillance area, an initial cluster sample of 500 households was randomly selected. Because initial data collection yielded fewer injuries than expected in the first two enumeration areas, the sample size was increased to 2,000 households with the difference made up of households selected at random from the remaining six surveillance branches. Thus, the final sample under-represented the first two branches. Information was sought on all individuals residing in the selected households. A two-stage cluster sampling method was adopted in selecting the rural sample. In the first stage, using existing AMMP data on mortality and poverty, six out of 51 villages were selected to represent different levels of socio-economic status and injury mortality. A random sample of 2,000 households was obtained from the selected villages in the second stage. All individuals in the selected households were included in the survey. Ethical clearance and informed consent Ethical clearance for this study was given by the Tanzania Commission for Science and Technology and the Regional Committee for Medical Research Ethics in Norway. Informed verbal consent was sought from each family. For children below age 15, parents or guardians were interviewed; for adolescents aged 15 to 18, consent was obtained from both the parent and the child. Data collection The survey tool was translated into Swahili, back translated into English, and pre-tested before use in the field. Two questionnaires were used in the study. Questionnaire 1 was a screening form used to identify whether a household member had an injury in the past one year that resulted in losing one or more days of 'normal' activity (e.g. not being able to work or go to school). The head of household or any other responsible person was interviewed to obtain information about the household members. Variables included were age, sex, relationship to head of household, level of education, religion, marital status and occupation. Questionnaire 2 was used to record the circumstances in which the injury occurred. Some of the variables included were: month and year when the injury occurred; cause of the injury; place of occurrence; activity at time of injury; length of disability; and health facility use. Efforts were made to interview the injured person if an adult, otherwise we interviewed an informed member of the injured person's household. The number of days with restricted activity (disability days) was considered as a measure of severity of injury. Poverty was assessed at the household level using data from the 2000–2001 National Household Budget Survey and variables from AMMP data. The measure of poverty used was a predicted value of monthly consumption expenditure per adult equivalent for each household included in the study [ 17 , 18 ]. Statistical analysis Data analysis was done using STATA (version 7). Bivariate analyses were performed by cross tabulations and the chi-squared test was used to test for homogeneity. In the case of multiple injuries, the most recent injury episode was considered in all the analysis. Multiple logistic regression was used to examine the influence of socio-demographic and socio-economic factors on the risk of being injured, controlling for potential confounding variables. Odds ratios are reported with 95% confidence intervals. Tests for trend in associations over groups defined by other factors were performed where appropriate. Preliminary analysis indicated that a long recall period underestimated annual injury rates, with the effect being greater for injuries resulting in <30 disability days while the rates for injuries resulting in 30 or more disability days were quite stable[ 19 ]. We therefore categorized severity of injury as 'minor' if resulting in less than 30 days of lost activity and 'major' if resulting in 30 or more days of lost activity. This kind of categorization has also been used in the Ghana study [ 13 ]. The category for major injuries is less likely to include actual minor injuries, and therefore constitutes a well-defined small group. However, the category for minor injuries might include some injuries that were actually severe. About 37 (7.3%) individuals who sustained an injury reported between 15 and 21 disability days, with only 3 reporting 22 to 29 days of restricted activity. Poverty quintiles were classified as most poor, very poor, poor, less poor, or least poor in terms of socioeconomic status. Adjustment for clustering was performed with standard STATA commands for analysis of survey data. Results Data were gathered on a total of 15,223 individuals residing in 3,653 households. The urban sample included 8,188 individuals while the rural sample included 7,035 persons. The response rates were 89% and 92% for the urban and rural areas respectively. The rural sample had a larger proportion of individuals aged 44 years and above (24%) compared to the urban area (11%). This was comparable to national averages as reported by the 2002 national census in Hai (17%) and Dar es Salaam (10%)[ 20 ]. Educational status was higher in the urban area. Of the total sample, 509 persons reported to have sustained an injury in the past one year preceding the survey, representing an injury incidence of 32.7 per 1,000 persons per year (95% CI= 29.9 – 35.7). The incidence for all, minor and major injuries was 24.5, 16.4 and 8.1 per 1,000 persons per year for Dar es Salaam, and 42.5, 32.8 and 9.7 for Hai district. The mean age of the injured was 27.6 years (standard deviation 20) and 62% were males. Almost all injuries were unintentional (96%). On average, 14 days of normal activity were lost per person because of an injury. In Dar es Salaam, the most common cause of injury reported in both males and females was transport injuries, followed by falls and cuts (Table 1 ). In Hai, cuts ranked first, followed by falls and transport injuries. The proportion of individuals who sustained transport injuries in the urban area was four times higher than in the rural area (33.0% vs 7.6% respectively; p < 0.001). Cuts and stabs accounted for 49% of the injuries in Hai compared to only 18% in Dar es Salaam (p < 0.001). There was no statistical difference in the distribution of injury categories between males and females in the urban area (p = 0.37). In the rural area, males and females differed with respect to the most common causes of injuries (p < 0.001), with transport injuries experienced almost exclusively by males, and cuts being more frequent in females. Table 1 Cause of nonfatal injury by sex in the urban and rural areas Cause of injury Total Males Females No. % No. % No. % Urban (Dar es Salaam) Total 206 100 136 100 70 100 Transport injuries 68 33.0 43 31.6 25 35.7 Falls 56 27.2 35 25.7 21 30.0 Cuts/stabs 38 18.5 26 19.1 12 17.1 Burn 12 5.8 6 4.4 6 8.5 Struck by object 12 5.8 10 7.4 2 2.9 Animal bites 4 1.9 3 2.2 1 1.4 Assault 7 3.4 4 2.9 3 4.3 Other 9 4.4 9 6.7 0 0 Rural (Hai) Total 303 100 177 100 126 100 Transport injuries 23 7.6 22 12.4 1 0.8 Falls 83 27.4 49 27.7 34 26.9 Cuts/stabs 149 49.2 77 43.5 72 57.1 Burns 18 5.9 8 4.5 10 7.9 Struck by object 11 3.6 11 6.2 0 0.0 Animal bites 6 1.9 4 2.3 2 1.6 Assault 3 0.9 3 1.7 0 0 Other 10 3.3 9 5.1 1 0.8 As expected, the cause of injury varied by age (Figure 1 ). In the urban area, transport injuries were most common among adults aged 15 years and above while burns were common among children under 5 years. Cuts and stabs ranked second as a cause of injury among the 5 to 14 year olds. In Hai, cuts were the commonest cause of injury in all age groups except among 0–4 year olds where burns and falls were most frequent. Figure 1 1a: Cause of injury by age in Dar es Salaam (urban area) 1b: Cause of injury by age in Hai (rural area) Major injuries accounted for 33% and 23% of all injuries in Dar es Salaam and Hai respectively (Table 2 ). The percentage of transport injuries resulting in a major injury was 41% and 30% in the urban and rural areas respectively. Of cuts or stabs in the rural area, 13% were major whilst in the urban area, 21% of the cuts were major. More than one third of the falls were categorised as major injuries in both areas. Table 2 Major injuries as a percentage of all injuries (≥ 30 disability days) by cause and sex in the urban and rural areas Cause of injury No. of all injuries Both sexes Males Females Percent of major injuries Dar es Salaam (Urban) Total 206 33.0 28.7 41.4 Transport injuries 68 41.2 34.9 52.0 Falls 56 35.7 28.6 47.6 Cuts/stabs 38 21.1 23.1 16.7 Burn 12 33.3 16.7 50.0 Struck by object 12 33.3 30.0 50.0 Other 20 20.0 30.8 0 Hai (Rural) Total 303 22.8 23.7 21.4 Transport injuries 23 30.4 27.3 100.0 Falls 83 38.6 36.7 41.2 Cuts/stabs 149 12.8 13.0 12.5 Burns 18 16.7 25.0 10.0 Struck by object 11 45.5 45.5 0 Other 19 15.8 10.0 28.6 After controlling for age, sex and education, persons residing in Hai were 1.7 times as likely to have had an injury in the past one year as compared to those residing in Dar es Salaam (Table 3 ). Males had a higher risk of being injured than females. Those with primary education only had an increased risk of having an injury compared to their counterparts who had no formal education. Children aged 5 to 14 had slightly higher odds of sustaining a minor injury compared to adults, while for major injuries adults aged 45 years and above were at an increased risk (p < 0.01 comparing trends with age for minor and major injuries). After adjusting for age, sex, education and area, there was no significant association between poverty and risk of nonfatal injury. Male sex turned out to be the only significant risk factor for major injuries. Table 3 Adjusted odds ratios (OR) for all, minor (<30 disability days) and major (≥ 30 disability days) injuries by demographic and socio-economic factors Factors Total All injuries Major injuries Minor injuries Sample No. OR a (95% CI) No. OR a (95% CI) No. OR a (95% CI) Area Dar es Salaam (Urban) 8188 206 1.0 68 1.0 138 1.0 Hai (Rural) 7035 303 1.66 (1.37 – 2.02) 69 1.09 (0.75 – 1.58) 234 1.94 (1.54 – 2.44) p < 0.001 p = 0.65 p < 0.001 Age* 0–4 1720 49 1.28 (0.83 – 1.99) 13 0.85 (0.38 – 1.91) 36 1.53 (0.91 – 2.57) 5–14 3711 140 1.23 (0.98 – 1.56) 29 0.88 (0.56 – 1.40) 111 1.37 (1.04 – 1.79) 15–44 7140 212 1.0 59 1.0 153 1.0 45+ 2651 108 1.25 (0.97 – 1.59) 36 1.57 (0.99 – 2.47) 72 1.12 (0.84 – 1.50) p = 0.20 p = 0.10 p = 0.11 Sex Females 7844 196 1.0 56 1.0 140 1.0 Males 7379 313 1.75 (1.46 – 2.12) 81 1.57 (1.11 – 2.21) 232 1.81 (1.46 – 2.25) p < 0.001 p < 0.01 p < 0.001 Education** None 3546 97 1.0 30 1.0 67 1.0 Primary 9674 362 1.50 (1.09 – 2.06) 92 1.02 (0.59 – 1.74) 270 1.77 (1.21 – 2.59) Secondary+ 2001 50 1.18 (0.77 – 1.82) 15 0.79 (0.37 – 1.71) 35 1.41 (0.84 – 2.36) p = 0.01 p = 0.67 p < 0.01 Poverty quintiles (n = 12320)*** 1 (Most poor) 2471 95 1.13 (0.82 – 1.56) 22 1.05 (0.56 – 1.96) 73 1.15 (0.80 – 1.66) 2 2462 78 0.92 (0.66 – 1.29) 21 1.01 (0.52 – 1.97) 57 0.89 (0.61 – 1.32) 3 2466 81 0.96 (0.69 – 1.32) 23 1.11 (0.61 – 2.01) 58 0.91 (0.62 – 1.32) 4 2462 72 0.86 (0.61 – 1.22) 25 1.24 (0.67 – 2.29) 47 0.74 (0.49 – 1.12) 5 (Least poor) 2459 83 1.0 20 1.0 63 1.0 p = 0.49 p = 0.96 p = 0.19 * 1 missing subject, ** 2 missing subjects a Adjusted odds ratios from logistic regression models that included all variables in table except poverty *** Adjusted for age, sex, area and education p-value for test of heterogeneity We investigated associations between different factors and some of the most frequent causes of injury (Table 4 ). Rural residents were significantly less likely to have transport injuries compared to urban dwellers (OR = 0.39; 95% CI 0.23 – 0.66). Children aged between 5 and 14 years were less likely to sustain transport injuries compared to adults aged 15–44 years. Males had a significantly increased risk of having transport injuries compared to females. However, household consumption expenditure was not associated with risk of transport injuries. In the rural area, the commonest type of transport involved was bicycle (52%) while in the urban area cars and trucks (46%) and commercial buses (22%) were frequently involved (Table 5 ). In Dar es Salaam, 41% of those involved in transport injuries were pedestrians who were struck by motor vehicles or bicycles, whereas in Hai, the largest proportion of those with a transport injury were vehicle occupants (52%) followed by cyclists (35%). Table 4 Adjusted odds ratios (OR) for transport injuries, falls and cuts or stabs by demographic and socio-economic factors Factors Total Sample Transport injuries Falls Cuts/stabs No. OR (95% CI) a No. OR (95% CI) a No. OR (95% CI) a Area Dar es Salaam (Urban) 8188 68 1.0 56 1.0 38 1.0 Hai (Rural) 7035 23 0.39 (0.23 – 0.66) 83 1.56 (1.09 – 2.25) 149 4.27 (2.96 – 6.15) p < 0.001 p = 0.01 p < 0.001 Age* 0–4 1720 5 0.52 (0.14 – 1.97) 15 2.21 (0.98 – 4.98) 5 0.35 (0.12 – 0.99) 5–14 3711 8 0.28 (0.13 – 0.63) 51 2.44 (1.58 – 3.79) 54 1.12 (0.77 – 1.62) 15–44 7140 60 1.0 41 1.0 80 1.0 45+ 2651 18 1.01 (0.56 – 1.83) 32 1.98 (1.22 – 3.21) 48 1.19 (0.82 – 1.74) p < 0.01 p < 0.001 p = 0.07 Sex Females 7844 26 1.0 55 1.0 84 1.0 Males 7379 65 2.66 (1.64 – 4.29) 84 1.62 (1.14 – 2.30) 103 1.38 (1.03 – 1.85) p < 0.001 p < 0.01 p = 0.03 Education** None 3546 10 1.0 30 1.0 23 1.0 Primary 9674 67 1.76 (0.63 – 4.93) 96 1.56 (0.90 – 2.69) 151 1.61 (0.98 – 2.66) Secondary+ 2001 14 1.10 (0.34 – 3.56) 13 1.50 (0.67 – 3.38) 13 1.03 (0.48 – 2.21) p = 0.15 p = 0.27 p = 0.05 Poverty quintiles (n = 12320)*** 1 (Most poor) 2471 15 1.18 (0.56 – 2.49) 19 0.67 (0.36 – 1.22) 41 1.36 (0.81 – 2.28) 2 2462 12 0.90 (0.41 – 2.01) 24 0.85 (0.48 – 1.51) 35 1.14 (0.68 – 1.91) 3 2466 12 0.93 (0.43 – 2.01) 24 0.84 (0.48 – 1.49) 27 0.87 (0.51 – 1.48) 4 2462 13 0.99 (0.47 – 2.11) 21 0.76 (0.41 – 1.39) 26 0.86 (0.50 – 1.50) 5 (Least poor) 2459 14 1.0 27 1.0 30 1.0 p = 0.96 p = 0.74 p = 0.31 * 1 missing subject, ** 2 missing subjects a Adjusted odds ratios from logistic regression models that included all variables in table except poverty *** Adjusted for age, sex, area and education p-value for test of heterogeneity Table 5 Vehicles involved in crashes causing traffic injuries in the urban and rural area Type of vehicle Dar es Salaam (n = 68) Hai (n = 23) % % Car/truck 46 26 Bus 22 9 Bicycle 16 52 Motorcycle 13 4 Train 3 - Cart - 9 Our results show that rural residents were 1.6 times as likely as urban dwellers to experience a fall resulting in an injury. Falls were more likely in children aged below 15 years and adults 45 years and above (Table 4 ). They were reported to occur mainly in and around homes in the urban area (Table 6 ). In the rural area, outside home, on the roads and farms were reported to be the most frequent places of occurrence for falls. Table 6 Place of injury by area and cause Place of injury All injuries Falls Cuts Urban (n = 206) Rural (n = 303) Urban (n = 56) Rural (n = 83) Urban (n = 38) Rural (n = 149) % % % % % % Home Inside 16.9 12.5 23.2 6.0 13.2 8.1 Outside 24.3 24.8 41.1 31.3 39.5 24.8 Workplace/factory 8.3 2.3 3.6 3.6 23.7 1.3 Farm 0 33.3 0 22.9 0 53.0 On the road 38.8 19.5 5.4 25.3 15.8 6.7 Recreation area including sports 9.7 1.9 21.4 2.4 7.9 1.3 School 1.9 4.6 5.4 8.4 0 4.0 Table 4 shows that rural inhabitants had a four fold risk of experiencing injuries due to cuts or stabs compared to urban residents (OR = 4.27; 95% CI = 2.96 – 6.15). It was noted that 68% (102/149) of injuries due to cuts in the rural area were related to agricultural activities, of which 81% occurred in adults aged 15 years and above. The farm and outside homes were where the injuries occurred most (Table 6 ). In the rural area, about 50% of children aged 5–14 years were injured when working on farms or around their homes and one third of the injuries were related to play. In the urban area, most (70%) of the children aged 5–14 years were injured while playing. Burns accounted for about 6% (30/509) of all injuries in both areas. Children aged less than five years were 8 times as likely to sustain injuries due to burns as adults aged 15 to 44 years (OR = 8.58; 95% CI = 1.73 – 42.5). Although the magnitude of association appears to be large, the estimate was based on small numbers (16 and 5 injuries respectively; table not shown). Discussion In this study, we found major differences between urban and rural residents with respect to cause and severity of injury and the circumstances in which they occurred. This has great implications in setting priorities when planning for intervention strategies. Transport injuries formed the most common injury category in the urban area. The low risk of transport injury in the rural areas is probably a reflection of the relatively lower level of motorization in this mainly agricultural area. However, they will often have more serious consequences than other types of injury. A number of studies have reported similar findings. In a study from Pakistan, farmers were found to be at a lower risk of traffic injury than labourers and vendors [ 11 ]. A study from Bangladesh found a low incidence of traffic injury in a rural population [ 14 ]. Our data revealed that in the rural area, bicycle injuries predominated while in the urban area motorized vehicles accounted for a large proportion of transport injuries. Bicycles play a very important role in rural areas of Tanzania as a means of transport. In the urban area, most of the transport injury victims were passengers on public transport and pedestrians. Previous studies from developing countries have also reported the dominance of pedestrians, passengers of commercial vehicles and cyclists as vulnerable road users to transport injuries [ 10 , 11 , 21 ]. A hospital-based study conducted in an urban area in Tanzania reported that 42% of those subjected to transport injuries were pedestrians [ 4 ]. Strategies for prevention of transport-related injuries should take into account the local patterns. Cuts and stabs by instruments such as axes and machetes constituted the most frequent injury category in the rural area, which is due to the fact that rural residents engage in agricultural activities using unprotected equipment. Cuts and stabs also contributed significantly among children aged 5 to 14 years with farm work being the common activity in the rural setting while play was the main contributing factor in the urban area. As emphasized in other studies, there is a need for safe space for play among children. In addition, the issue of a working child in developing countries like Tanzania needs to be addressed. Falls were also a significant contributor among young children and older adults. A better understanding of the circumstances in which falls occur would assist in planning for fall-prevention programmes in Tanzania. Injuries due to accidental poisoning were infrequently reported in both areas and hence grouped under 'others'. It is possible that people were not comfortable reporting such events. In addition, near drowning did not feature among the causes of nonfatal injuries although the urban area has a port. Except for transport injuries which were the commonest cause of fatal and nonfatal injuries in the same surveillance areas, the distribution of nonfatal injuries differed essentially from those of fatal injuries reported for 1992–1998 in the same surveillance areas [ 22 ]. In addition to transport injuries, the commonest causes of injury death were suicide, assault, accidental poisoning and drowning while the main causes of nonfatal injuries were falls, cuts and burns in these settings. A significant risk factor for injury turned out to be the place of residence. The likelihood of self reported injury was 66% higher for rural residents compared to urban residents. Similar findings have been reported from other studies [ 13 , 23 ]. However, a study from Uganda found a high incidence of injury in the urban setting [ 12 ]. For severe injuries, we observed similar rates in the urban and rural areas. The main explanation is that transport injuries are more common and more severe in the urban areas whereas cuts and stabs are less common but more severe than in the rural area. Age is an important risk factor for many injuries but its influence varies between specific injury groups. Our findings show that adults aged 15–44 years are at a high risk of transport injuries. This has great economic impact since these are people in their most productive years and the injuries impose a considerable burden on their families and the society as a whole. Children below 15 years were at greater risk of injuries due to falls. This may be due to high risk environments such as lack of proper play facilities. This is in keeping with studies from other developing countries whereby falls among children have been reported to be a common cause of injury seen in hospitals [ 24 ]. As expected, males were found to have an increased overall risk of injury which was more pronounced for transport injuries. A possible explanation may be that men spend more time on the roads than females and therefore they are more prone to high risk behaviours or unsafe road practices [ 25 ]. Socio-economic status has been documented to be an important determinant of injury, although the effect depends on the socio-economic indicator considered, the cause and severity of injury [ 7 ]. We found no significant relationship between income poverty and nonfatal injuries. This is consistent with findings from other studies [ 26 , 27 ]. Persons with primary education were at a greater risk of injuries than those with no formal education. This finding conforms to results from a previous study done in India [ 28 ]. The findings of this study are subject to a number of limitations. The information is based on self-reported data elicited through interviews, which is subject to recall bias [ 19 , 29 ]. A 12 month recall period was used in this study in order to include as many injuries as possible. Although long recall periods underestimate injury rates, they can be useful in investigating associations between injuries and different risk factors. Initial multiple logistic regression analysis was done using injuries reported in the three months prior to the interview only. We found that the pattern of the associations was similar to that based on all injuries except for area of residence where the size of the effect was stronger when a short recall period was used (results not shown). Other studies have demonstrated that relative risks are not affected when a long recall period is employed [ 30 ]. In a previous report, we found that memory decay was greater in the rural area than in the urban area [ 19 ]. Therefore, the actual difference in rates between the urban and rural area may have been underestimated. Intentional injuries such as assaults and domestic violence are probably underreported since they would not be adequately captured in such a survey. This may lead to injury rates being underestimated. In this study, a clinical injury severity assessment was not possible. Disability days were used instead as a measure of severity of injury. One should be careful in generalizing the findings to other urban and rural settings of the country. Dar es Salaam is in many ways different from other urban areas of Tanzania and Hai is a relatively wealthy rural area. Furthermore, it might be difficult to generalize the findings to the surveillance areas due to the selection process that was employed. However, from our knowledge of the areas, we see no obvious reasons indicating that our samples should be very different from the urban and rural surveillance areas. Despite its limitations, this study has generated information that could be useful for targeted prevention at the local level. Conclusions This study, the first of its kind in Tanzania, describes the patterns of nonfatal injuries and associated socio-demographic and socio-economic factors. It has attempted to identify specific groups of individuals as having a greater risk of experiencing certain types of injuries. This information is important for raising the level of awareness among policy makers and the public in general since the problem of injury receives little attention in most of the developing world including Tanzania. It is also useful in setting priorities for cause-specific prevention strategies. More detailed qualitative studies are required, however, on sensitive events such as assault. A nationally representative sample is also essential to measure the health burden due to nonfatal injuries. Competing interests The author(s) declare that they have no competing interests. Authors' contributions CM designed and conducted the study, performed statistical analysis, wrote the initial draft and revisions of the manuscript after consultation with other authors. IH, AN and GK participated in the design of study and revision of the manuscript. PS and YH participated in study design and co-ordination, and in revision of manuscript. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548509.xml

212689

Genetically Modified Corn— Environmental Benefits and Risks

To plant or not to plant. A discussion of the environmental benefits and risks of genetically modified crops

Corn is one of humankind's earliest innovations. It was domesticated 10,000 years ago when humans learned to cross-pollinate plants and slowly turned a scraggly nondescript grass called teosinte into plump, productive modern corn ( Figure 1 ). As needs change, so does plant breeding. Today, while biotech super-giants manipulate corn genetics to satisfy farmer desires and a global market, indigenous Mexican farmers do so to fulfill individual needs. Although the tools differ, the goal remains the same—to cultivate desirable traits. Figure 1 Crossing for Kernels Over time, selective breeding modifies teosinte's few fruitcases (left) into modern corn's rows of exposed kernels (right). (Photo courtesy of John Doebley.). Plant breeding was once restricted to sexually compatible plants, and generations of offspring were selectively bred to create unique varieties. In fact, corn, along with rice and wheat—today's global crop staples—would not exist without such techniques. With the goal of ever-widening the pool of genetic diversity, conventional plant breeding has gotten more technologically savvy in recent years. For example, realizing that natural mutants often introduce valuable traits, scientists turned to chemicals and irradiation to speed the creation of mutants. From test-tube plants derived from sexually incompatible crosses to the use of molecular genetic markers to identify interesting hereditary traits, the divide between engineering and genetics was narrowing long before kingdom boundaries were crossed. But when geneticists began to explore microorganisms for traits of interest—such as Bacillus thuringiensis (Bt) genes that produce a protein lethal to some crop pests—they triggered an uproar over ethical, scientific, and environmental concerns that continues today. (See Box 1 .) Box 1. Bt Technology Bacillus thuringiensis , a soil bacterium, produces several crystal (Cry) protein toxins that destroy the gut of invading pests, such as larval caterpillars. So far, over 50 cry genes have been identified and found to affect insect orders differently. Considered safe to humans, mammals, and most insects, Bt has been a popular pesticidal spray since the 1960s because it had little chance of unintended effects. Engineering the gene into corn, however, caused an unexpected public backlash. “We thought it was going to be the greatest thing since sliced bread,” says Guy Cardineau, agricultural biotechnologist at Arizona State University. “Here's a way to withstand insect pressure, eliminate the use of pesticides, and Bt spray was widely used in organic agriculture,” he adds. The Bt wrangle illustrates how differently a product and a process can be regarded. After the expensive development process, today's concern is that broad-scale planting of Bt corn will render the toxin ineffective over time. Pests can gradually build resistance to any pesticide, and so the United States Environmental Protection Agency (EPA) requires that 20% of Bt field areas be planted to non-Bt corn to avoid such pressures. But humans have to follow the rules. A recent report from the Center for Science in the Public Interest shows that almost 20% of farmers in the United States Corn Belt are violating EPA standards by overplanting Bt corn, causing some to question the regulations and enforcement that will be necessary for certain GM crops. Despite such discord, genetically modified (GM) crops have the fastest adoption rate of any new technology in global agriculture simply because farmers benefit directly from higher yields and lowered production costs. (See Table 1 .) To date, the two most prevalent GM crops traits are Btderived insect resistance and herbicide resistance. Table 1 Worldwide production of GM crops Four crops account for most GM plantings: herbicide-tolerant soybeans (62%), insect-resistant corn (12.4%), insect-resistant cotton (6.8%), and canola (3%). Source: Summary Report on the Global Status of GM Crops by the International Service for the Acquisition of Agri-Biotech Applications (2002) Since 1987, over 9,000 United States Animal and Plant Health Inspection Service (APHIS) permits have been issued to field-test GM crops. According to APHIS, corn is the most tested plant. The International Service for the Acquisition of Agri-Biotech Applications confirms that biotech corn is the second-most common GM crop (after soybean), with 12.4 million hectares planted in 2002. GM corn starch and soybean lecithin are just two of the ingredients already found in 70% of the processed food supply. With future incarnations on the horizon, GM corn remains a lightening rod for debate. Embroiled in numerous controversies, corn has become biotech's boon and bane. Benefits Emerging As Danforth Center President Roger Beachy, the first to develop a virus-resistant tomato, describes it, the first-generation GM crops were intended to help farmers reduce not only the impact of pests, but also the use of agrochemicals in modern crop production–a legacy of the Green Revolution. After a decade of cultivation, environmental benefits are emerging. Bt corn reduces the need for pesticides, and while the primary benefit comes largely during a heavy corn-borer infestation, an unpredictable event, a secondary effect is that beneficial insects fare much better under these conditions. The numbers are particularly impressive for Bt cotton: the spraying of almost 2 million pounds of pesticides—roughly 50% of previous usage—has been spared since the large-scale adoption of Bt cotton. According to Leonard Gianessi, senior research associate at the National Center for Food and Agricultural Policy, farmers who adopt GM crops make more money in tougher times. Indeed, insect- and virus-resistance traits have already saved several industries. Bt cotton is credited with reviving the Alabama cotton industry, hard hit by uncontrollable bollworm infestations. Likewise, genetically engineered papaya, made resistant to the papaya ringspot virus, salvaged Hawaii's fifth largest crop industry. Herbicide-resistant crops engendered a different reception. While GM critics acknowledge that the use of a more benign herbicide, called by its trade name Roundup, can have environmental benefits, the creation of a market monopoly is a key criticism. However, the increased planting of herbicide-resistant soybeans is an integral, but not sole, factor in the increased adoption of no-till farming— a strategy that reduces soil erosion. Surprise benefits have also occurred. According to the recent International Council for Science (ICSU) review of GM crops, disease-resistant corn crops may have lower levels of mycotoxins, potentially carcinogenic compounds to humans. They result from fungal activity in insect-infested corn crops. With fewer insect holes in plant tissue, associated fungi are not able to invade and produce toxins. While there is a growing amount of data documenting the intended environmental benefits of GM crops, the potential risks are more elusive. Risky Business After seven years of GM crop production and no apparent health effects, potential environmental risks—particularly gene flow into other species—have eclipsed food safety as a primary concern. As pollen and seeds move in the environment, they can transmit genetic traits to nearby crops or wild relatives. Many self-pollinating crops, such as wheat, barley, and potatoes, have a low frequency of gene flow, but the more promiscuous, such as sugar beets and corn, merit greater concern. Determining where genes flow is a thriving research avenue, but the real question becomes “so what?” The risks associated with gene flow—such as creating weeds from introduced traits, reducing biodiversity, or harming nontarget species—are similar to those from conventionally bred crops. “I wouldn't dismiss the ecological concerns out of hand, but I think they can be exaggerated,” says Gabrielle Persley, the ICSU report author. There are few instances of crop plants becoming weeds. Bred so intensely for hundreds of years, most crops cannot survive without human intervention. Increased weediness could be conveyed, however, if the plants are more fit or able to out-compete other crop species by producing more seed, by dispersing pollen or seed further, or by growing more vigorously in a specific environment. In fact, transgenic sunflowers can produce over 50% more seed than traditional varieties. Additionally, recent work shows that, compared to pollen, seeds are more likely to spread GM sugar beet genes into wild relatives in western Europe. Norman Ellstrand, plant geneticist at the University of California at Riverside, has shown that gene flow from many conventionally bred crops increases the weediness of nearby wild relatives. For many domesticated crops, wild varieties do not exist in current areas of cultivation. Nevertheless, regions where crop species originated are particularly vulnerable to transgenic gene flow into local varieties, or landraces. Some fear that transgenic varieties with a competitive advantage might gradually displace valuable genetic diversity. For these reasons, transgenic corn is prohibited in Mexico, home to over 100 unique varieties. Despite the ban, transgenes have been found in Mexican corn. “We have in several instances confirmed that there are transgenes in landraces of maize in Oaxaca,” says Ariel Alvarez-Morales, plant geneticist at the Mexican Center for Research and Advanced Studies (CINVESTAV) in Irapuato. The ramifications of this will not be known for some time, but Luis Herrera-Estrella, CINVESTAV's Director of Plant Biotechnology, is convinced that these single gene traits will be of little consequence to native Mexican varieties. “If Bt genes give an advantage to the farmer, they will keep growing it. In that case it will not be bad,” he says of dynamically changing landraces. “Gene flow has been occurring for 50 years from commercial lines to landraces.” While admitting this, Ellstrand points out that “if there are genes that you don't want to get into landraces—this shows how easily they can get there.” (See Box 2 .) Box 2. Pharma Corn “The gene flow risk that keeps me awake at night is the possibility of hybridization between crops engineered to manufacture poisons and related crops intended for human consumption,” says plant geneticist Norman Ellstrand. Indeed, this application of GM crops seeks to turn corn into cost-effective pharmaceutical factories and may bear the mark of unacceptable risk. It is currently the subject of intense debate. An open-pollinated crop, corn is known for its promiscuity—making it more prone to gene flow risks than other crops. Genetic contamination takes on a whole new meaning when the escapable trait could produce proteins to treat diabetes or a hepatitis B vaccine. Given that pharma corn demands multiple safety measures—including production in remote areas, separate farm equipment, delayed planting to offset pollination—many ask, “Why use corn?” “We know so much about corn genetics,” explains agricultural biotechnologist Guy Cardineau, “and it naturally lends itself to production with kernel packets of protein that can be stored indefinitely.” A number of scientists and United States food makers are not yet convinced that the benefits outweigh the risks and have joined environmental groups in questioning the use of pharma corn. Over 130 acres of pharma crop field-tests were planted in 2002. Several products have moved on to clinical trials. Aware of concerns, the members of the influential Biotechnology Industry Organization decided last December to overturn its initial decision to remove pharma crops from the United States Corn Belt states and voluntarily police their use. A Colorado trial of corn producing a protein to treat cystic fibrosis recently began. Indeed, unintended impacts are a primary concern. The potential risk to nontarget organisms took center stage when a 1999 paper in Nature suggested monarch butterfly populations might be adversely affected by Bt transgenes. Corrected by subsequent publications, the field experiments did not support original laboratory results. But effects on other nontarget organisms, such as soil microbes, remain a concern. When microbial genetics research uncovered how genes could be transferred between species in ways other than reproduction, so-called horizontal gene transfer, it not only explained why microorganisms were so diverse, but that microbes could potentially be endowed with GM plant DNA found in the soil. “Although a theoretical possibility, there is no evidence that it happens at any degree of frequency to be meaningful,” says Persley. Opinions differ on this, however, and seem to follow the United States–European Union divide over the use of GM crops. Kaare Nielsen, microbial geneticist at Norway's University of Tromsø, is one of the few scientists to find examples of horizontal gene transfer. “There are actually very few studies and most of the ones conducted have been on first-generation plants,” Nielsen explains. Given that plant DNA can last in soil for over two years, Nielsen does not believe the possibility can be dismissed and argues that long-term studies are necessary. Work continues in this area in Europe. The lack of baseline ecological data—even agreeing on what an appropriate baseline is—presents a substantial knowledge gap to environmental impact assessments. Scientists, including Nielsen, wonder whether there could be unexpected risk factors. Allison Snow, weed expert at Ohio State University, agrees with what many feel is the most important risk—the inability to anticipate all the effects. “Do we know all of the right questions we should be asking?” she wonders, adding, “Genes are complicated and can interact.” For these reasons, identifying factors that regulate weed and pest populations and determining how microbial community changes affect larger ecosystems are important areas of research. Do Risks Differ for Developing Nations? To two academicians that kindled the biotech revolution, the real GM risks lie in how science is misinterpreted and misused. In fact, much of the currently conducted basic research is not likely to be applied in the near future. Public concerns coupled with corporate consolidation created huge roadblocks, especially in getting the technology to developing nations. While Beachy blames the skyrocketing regulatory costs that “are due to regulators who have not put into context this technology and its relative safety,” Richard Jefferson, chairman and chief executive officer of the Center for the Application of Molecular Biology to International Agriculture in Australia, fears that innovation has been stifled by corporate short-sightedness. “The biggest risk is that [biotechnology] maintains itself as a monolithic, expensive industry with untenable entry barriers for smaller enterprises,” he says. Indeed, when does the risk of not using available technology factor into the debate? (See Box 3 .) Many scientists argue that genetic modification can help to ensure food security in developing countries, specifically in Africa. While more than 25% of the 2002 global biotech acreage was grown in countries such as Argentina, China, and India, there is little applied research on crops relevant to famine-prone African countries. Box 3. Golden Rice Current regulatory constraints have a choke-hold on innovations for genetic modifications that seek to improve subsistence crops, such as rice. Golden rice, yellowed in appearance because it is infused with the vitamin A precursor beta-carotene, could save thousands of malnourished people each year from blindness and the other vitamin A–deficiency diseases prevalent in Southeast Asia. Intellectual property issues and opposition from anti-GM activists have confounded the development for years. Faced with patent issues and regulatory hurdles and costs, developer and academic researcher Ingo Potrykus formed an alliance with Syngenta (then AstraZeneca Corporation) to allow the free licensing of the patents to public research institutions for humanitarian use. In addition, farmers making less than US$10,000 will receive free golden rice seed. After over a decade of work, golden rice is still not on the market. The retired Potrykus is determined to bring this technology to farmers once it passes regulatory field testing—an additional four-year delay that he feels is scientifically unnecessary. “Nobody can construct even a hypothetical risk to the environment from golden rice,” he says, adding that nutritional risks are nonexistent as well. He acknowledges, however, that field tests will be beneficial for acceptance of this and future bio-fortified products. “I have experienced so much support in these countries, I don't think it [the anti-GM lobby] will be able to stop this project once it passes regulation,” he says. “Food security is not going to come from crops being marketed outside Africa, like wheat or rice,” says John Kilama, Uganda native and president of the Global Bioscience Development Institute. He points out that of traditional staple crops such as cow peas and millet, only cassava has merited some publicly-funded research. Beachy estimates that it takes between US$5 million and US$10 million to bring a GM crop to market. Given regulatory costs, neither industry nor universities can afford to develop products that do not have mass appeal. “If the crop is not worth that much to the seed market, it's likely that we'll never see the varieties because of the cost of regulation,” he says. To ensure a return on research investments, with the regulatory costs often the biggest ticket item, developing blockbuster traits is a priority. “Given the diversity of environments and cropping systems, there are not many more blockbusters such as Roundup resistance in the wings,” says Jefferson. The alternative, he adds, is to make it cheaper to innovate local varieties in ways that are likely to gain public acceptance. (See Box 4 .) Box 4. Apomixis One way to minimize the problems associated with gene flow is to introduce sterility, such that pollen cannot transmit information. Richard Jefferson has high hopes for an accessible, cheap way for farmers to produce genetically superior seeds, called apomixis. But similar concepts have been floated before. The controversial terminator technology prevented gene flow, but it also outraged activists because it kept farmers from reusing seed. Unlike terminator, apomixis is “germinator” technology—avoiding fertilization altogether by producing seeds without pollination. In effect, seeds can be natural clones of the mother, instead of a genetic exchange between mother and father. Therefore, hybrid quality can be maintained as farmers use seed year after year. Although apomixis occurs naturally in about 400 plant species, Jefferson believes that it can be successfully developed as a useful trait in other crop plants. To ensure its widespread availability, Jefferson and collaborators pledged not to create restrictive patent rights that could block the development of apomixis. “The Green Revolution largely bypassed Africa,” says Josette Lewis, biotechnology advisor for the United States Agency for International Development. Given monetary constraints that prevent access to many biotechnologies, many scientists worry that the Gene Revolution might as well. Looming trade issues coupled with food insecurity shape the debate in Africa. Caught between the United States and European Union trade disputes, sub-Saharan countries are eager to use any technology that will enhance production without jeopardizing trade. Increasingly, industry is responding to the developing nations' needs. Both newly formed, the industry-focused African Agricultural Technology Foundation and the Public-Sector Intellectual Property Resource for Agriculture are attempting to ease cost restrictions and encourage access to current and future patents. By entering into such agreements, industries will be able to protect patent rights and commercially important markets. Such partnerships are already working. The Syngenta Foundation for Sustainable Agriculture is working together with the International Maize and Wheat Improvement Center (CIMMYT) and the Kenyan Agricultural Research Institute to overcome corn stemborer infestations in Kenya ( Figure 2 ). “CIMMYT hopes to have a handful of local Bt corn varieties in farmers' fields by 2008,” says the admittedly ambitious Dave Hoisington, director of CIMMYT's Applied Biotechnology Center. Collaborations between public and private sectors may be the only way to navigate thorny patent issues and research crop varieties of interest to developing countries. Figure 2 Biotech Bridge to Africa In an effort to reduce corn stem-borer infestations, corporate and public researchers partner to develop local Bt corn varieties suitable for Kenya. (Photo courtesy of Dave Hoisington/CIMMYT.). Conclusion “Agricultural biotechnology is here to stay” read a recent opinion piece by Gianessi. No doubt he is correct. As genetic engineering continues to evolve, transgenic methods will become just one of many tools. In fact, some researchers are currently focusing their work on manipulating an organism's own genetic code to achieve desired traits. Scientific inquiry will continue to weigh the risks and benefits of such technologies, realizing that there may never be enough evidence to ensure zero risk. Only with data will tolerable levels of environmental risks be determined—case by case. Indeed, the level of risks and benefits may differ for developing nations, where decisions must be made in the face of food security concerns. To many scientists, the risks associated with forgoing genetic engineering far surpass any environmental risk associated with its use and further development. However, all stakeholders must have access to the tools in order to realize future benefits. In the quest to feed the world, a few things are clear. Public outcries will continue to vet the need and use of genetic engineering. Private organizations will necessarily focus on research for profit, while exploring collaborative prospects. Public initiatives, however, will provide the critical bridge between science and global food security. Although genetic engineering cannot be summarily accepted or rejected, any lack of scientific risk now doesn't negate future concerns. And, no matter what direction future research takes, corn will continue to be a bellwether crop.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC212689.xml

544864

Desmoplastic Infantile Ganglioglioma: cytologic findings and differential diagnosis on aspiration material

Background Desmoplastic infantile ganglioglioma (DIG) is a rare WHO Grade I tumor of infancy that is characterized by large volume, superficial location, invariable supratentoriality, fronto-parietal lobe predilection and morphologically, by an admixture of astroglial and neuroepithelial elements in a desmoplastic milieu. With over 50 cases described, the histologic and radiographic spectrum of DIG has been well-characterized. The superficial location of DIGs may render them greatly amenable to preoperative assessment utilizing aspiration cytology; however, the cytologic features of this rare tumor have only been reported once previously. Case Presentation We present herein cytomorphologic findings from the intraoperative aspiration of a typical case of DIG diagnosed in a 1-year-old male. As evaluated on a single liquid-based preparation, the specimen showed low cellularity and was comprised predominantly of a population of dispersed (occasionally clustered) large neuronal cells with eccentrically located hyperchromatic nuclei (which were occasionally binucleated) and abundant unipolar cytoplasm. Rare smaller astroglial cells were intermixed. Despite the tumor's characteristic desmoplastic histologic appearance, no stromal fragments were identified on the aspiration material. Conclusions A differential diagnosis is presented and analyzed in detail and it is concluded that when these large neuronal cells are encountered in an aspirate of a brain mass in a child, a combination of clinical, radiologic and immunohistochemical parameters can eliminate most of the differential possibilities.

Background The clinicopathologic features of 11 examples of a distinctive pediatric tumor designated desmoplastic supratentorial neuroepithelial tumors of infancy (also known as desmoplastic infantile ganglioglioma, [DIG]) were originally described by Vandenberg et al in 1987 [ 1 ]. Since that seminal report, at least 40 additional cases have been described, such that the clinical, radiologic and histopathologic features of this tumor are now well-defined. An uncommon tumor that constituted less than 0.04% of all central nervous system (CNS) tumors in one series [ 2 ], DIG is classified as a Grade 1 tumor in the World Health Organization (WHO) classification of CNS tumors [ 3 ]. They most commonly occur in children less than 18 months of age [ 1 ] who typically present with symptoms related to an intracranial mass effect [ 3 ]. DIGs are generally of large size, are solid to cystic, show a predilection for the frontal and parietal cerebral lobes, and are typically superficially located with at least focal attachment to the overlying dura [ 1 - 3 ]. The superficial location of DIGs may render them greatly amenable to preoperative assessment utilizing aspiration cytology. However, there is a dearth of information on the cytomorphologic features of these tumors [ 4 ]. To contribute information of possible utility in their pre-operative or intra-operative assessment, we report herein cytomorphologic features associated with a typical case of DIG. Case Presentation A one-year-old boy was noted to have a striking increase in head circumference as compared to a previous measurement. Neurological examination and developmental status were normal at that point. Within 2 weeks, the patient deteriorated rapidly, with poor mobilization, feeding and verbalization. He was brought to the emergency room where an emergent computed tomographic scan showed a large left hemispheric cerebral mass (Figure 1-1 ) with an underlying cystic component and a more superficial area of bright enhancement; the rest of the brain showed massive edema. He was emergently admitted and within 24 hours, a gross resection of the tumor was carried out. At surgery, following a parietal craniotomy, 50–60 cc of straw colored fluid was aspirated from the cystic component through a taut dura. After the excision of the dura, the bright area of enhancement previously noted was an area of tumor attachment to the dura in the parietal region. Otherwise, the tumor showed a well-demarcated interface with the subjacent normal brain parenchyma and a complete gross resection was achieved. A follow-up magnetic resonance image at 12 months post-surgery showed no evidence of tumor recurrence. Functionally, the patient was felt to have a mild right hemiparesis and some probable language delay, but otherwise showed no neurological deficits. Figure 1 Radiologic, cytologic and morphologic appearance of the tumor. 1 : This computed-tomographic scan of the patient's cerebral mass shows a large cystic mass with peripheral enhancement at the solid portion which attached to the overlying dura; 2 : In addition to scattered individual cells, variably sized clusters of neuronal cells were identified, all composed of cells with eccentrically located, occasionally binucleated hyperchromatic nuclei and abundant unipolar cytoplasm [original magnifications ×400]; 3 : Occasional neuronal cells were binucleated ( 3a ) while others showed bland nuclear features ( 3b ) [original magnifications ×400]; 4 : Scattered astroglial cells with more convoluted nuclear contours and less cytoplasm were also present. [original magnifications ×400]; 5 : Typical histologic appearance of desmoplastic infantile ganglioglioma, showing scattered ganglion cells in a desmoplastic and fibroblastic, vaguely storiform background (original magnification ×200, inset ×400) Materials and Methods For cytology, a slide was prepared from 50–60 cc of straw-colored fluid utilizing the ThinPrep ® 2000 Automated Slide Processor (Cytyc, Boxborough, MA) according to the manufacturer's instructions. For the tumor specimen, approximately 10 × 6.5 cm of fragmented gray and white cerebral tissue was received and entirely processed routinely: tissue sections were fixed in 10% neutral buffered formalin, processed, embedded in paraffin, sectioned to 4 μ-thick sections and stained with hematoxylin and eosin, Nissl stain and reticulin. The immunohistochemical profile of the tumor was evaluated on 4 μ thick, formalin-fixed, deparaffinized sections using a DAKO Autostainer (Carpinteria, CA, USA) based on the avidin-biotin-peroxidase complex with antibodies ki-67 (dilution 1:320, DakoCytomation Corp, Carpinteria, CA), synaptophysin (dilution 1:600, DakoCytomation) and glial fibrillary acid protein [GFAP] (dilution 1:10, DakoCytomation). Pathologic findings As evaluated on a single liquid-based preparation, the specimen showed low cellularity and was comprised predominantly of a population of dispersed (occasionally clustered) large neuronal cells (~70 μm diameter each) with round eccentrically placed uniform hyperchromatic nuclei, undulating and slightly convoluted nuclear membranes, and abundant unipolar granular cytoplasm (ganglion cells) (figures 1-2 and 1-3 ). Occasional cells were binucleated (figure 1-3a ). A spectrum in the degree of nuclear membrane irregularities was noted, with most cells displaying irregular features as described above, while other cell showed bland nuclear features (figure 1-3b ). However, all displayed nuclear polarity to the cytoplasm. Rare smaller cells interpreted as astroglial cells were interspersed between the larger cells. The latter cells showed nuclear hyperchromasia, more prominent irregularities in their nuclear membranes and a smaller cytoplasmic rim. (figure 1-4 ). Overall, significant proportions of both cellular populations showed varying degrees of degenerative changes manifested as lack of clear delineation of nuclear and cytoplasmic borders and a loss of nuclear detail. Several clusters were composed of large cells, and in these clusters, constituent large cells showed less cytoplasm but retained a unipolarity in relation to the nuclei and their nuclear features were identical to those of the more predominant population of large cells. Additionally, scattered foamy histiocytes were present. A finely granular background material consistent with necrosis was present, but there was no distinct neurofibrillary material. Vascular structures or stroma were not present. Histologically, the tumor was partially attached to the dura and was present in the subarachnoid space. The bulk of the specimen was a variably cellular desmoplastic component whose predominant constituent cells were elongated spindle cells arranged in a reticulin-rich, storiform pattern (figure 1-5 ). At higher magnification, ganglion-type cells (Nissl stain positive) with 1–4 round nuclei, prominent nucleoli, and abundant unipolar cytoplasm were present (Figure 1-5 , inset). Immature or abortive ganglion cells with enlarged single nuclei and markedly irregular nuclear membranes were rare but identifiable morphologically. Less "differentiated" aggregates of cells with hyperchromatic nuclei and minimal cytoplasm, as has been well-described in DIGs [ 1 - 3 ], were present. Mitotic figures were rare and small foci of necrosis were limited to the less differentiated component. Immunohistochemical stains for synaptophysin was positive in the ganglion cells only. GFAP was positive in astroglial elements within the desmoplastic regions; the latter was negative for synaptophysin. The ki-67 labelling index was 11.3% (evaluated in the area of greatest density of positive staining cells). Discussion The clinical presentation, radiographic appearance and histopathologic features of this case are entirely consistent with those described for desmoplastic infantile ganglioglioma [ 1 - 3 ]. There has been a significant evolution in the understanding of this rare tumor since its original description in 1987 [ 1 ]. DIGs are typically large supratentorial tumors that, at least as observed radiographically in one patient, are initially solid then become cystic [ 5 ]. Although this tumor is considered a grade 1 tumor based on the histopathologic features of cases described prior to the publication of the WHO monograph in 2000, at least one report has since documented anaplastic features in a case of DIG, which was ultimately fatal [ 6 ]. However, follow-up has generally been favorable following complete resection in the reported cases of DIG, with a median post-surgical interval of 8.7 years without metastases or recurrence in one series of 14 patients [ 2 ]. Additionally, in some cases, spontaneous regression of tumor following subtotal tumor resection has been documented [ 7 , 8 ]. The distinctive clinical features of DIG, being a typically superficially located tumor occurring in young children (with potentially unclosed fontanelles), may render them particularly amenable to pre-operative assessment using aspiration cytology. In addition, familiarization of practitioners with the cytopathologic features of DIG may be useful because 1) With the aforementioned cases of DIG regressing after subtotal resection [ 7 , 8 ], it might be unnecessary to aggressively resect these tumors to negative margins, and a preoperative aspiration diagnosis of DIG will be helpful in the neurosurgical planning and 2) In their intra-operative assessment, imprint cytopathology may potentially be more diagnostic than histopathology. However, to our knowledge, the cytologic features of DIG have been documented only once previously [ 4 ]. In that report, Hasegawa et al [ 4 ] reported aspiration and imprint cytology findings in two cases of DIG. Two distinct cellular populations were identified, a predominant population of small to intermediate sized astroglial cells and "a few" large cells with round nuclei, prominent nucleoli and profuse cytoplasm that was unipolar to the nuclei in all their illustrations. In the current case, the reverse was found, with the predominant cells being an identical population of large cells with round nuclei, prominent nucleoli and abundant unipolar cytoplasm, and only rare unequivocal astroglial cells. Most of the analysis of the aforementioned report [ 4 ] was on the imprint smears, and although a mixture of small and large cells were also identified on the aspiration smear, there was no stated assessment or low-power illustration of the relative ratio of small to large cells on the latter. In the current case, a cell-block for immunohistochemical confirmation of the nature of two-cell population was unavailable; however, the larger cells were positive for neurofilament (confirming their neuronal nature) while the smaller cells were positive for GFAP (confirming their astroglial nature) in the report of Hasegawa et al [ 4 ]. The finding of large cells with eccentrically located nuclei and abundant unipolar cytoplasm in an aspiration specimen of a cerebral mass occurring in a young person should generate a differential diagnosis that includes DIG, atypical teratoid/rhabdoid tumor (AT/RT), dysembroplastic neuroepithelial tumor (DNT), ganglioglioma, supratentorial primitive neuroectodermal tumour (PNET) with ganglionic differentiation (ganglioneuroblastoma), anaplastic large cell lymphoma and pleomorphic xanthoastrocytoma. In our opinion, clinical features as well as immunohistochemical analysis can significantly help reduce the likelihood for most of the aforementioned entities. The distinction of DIG from AT/RT is probably of the greatest prognostic significance, since in contrast to DIG, AT/RT is a highly malignant tumor that is almost uniformly fatal [ 9 ]. Although AT/RT occurs in infants or young children, most cases occur in the posterior fossa, in contrast to DIGs, which are invariably supratentorial [ 1 - 3 ]. Radiographically, AT/RT are typically not distinctly cystic, although necrosis may impart an irregularly cystic appearance. Immunohistochemically, the rhabdoid cells of AT/RT co-express vimentin and epithelial membrane antigen [ 10 ], in contrast to the ganglion cells of DIG. However, rhabdoid cells may rarely express neurofilament, an immunophenotypic overlap with DIG. Morphologically, the distinct cytoplasmic borders, "inclusion-like" cytoplasmic globule and overall dense eosinophilia of the cytoplasm of rhabdoid cells, in conjunction with the aforementioned clinicopathologic parameters, may help in their distinction from ganglion cells of DIG [ 9 , 10 ]. The separation of DIG from other tumors containing true ganglion cells based on cytomorphology alone would probably pose the greatest difficulty. These tumors include DNT, ganglioglioma, and supratentorial PNET with ganglionic differentiation; all have a predilection for, or at least may potentially occur in children. In addition to ganglion cells, cytomorphologic features of DNT include oligodendroglial-like cells arranged in lobules and neurons in abundant extracellular mucin or neurofibrillary material [ 11 , 12 ]; these findings were neither identified in the 2 cases of DIG reported by Hasegawa et al [ 4 ], nor the current case. The distinction of gangliogliomas form DIG based on cytomorphology alone may be impossible even in the presence of a significant stromal component on the aspirate. Gangliogliomas, like DIG may be solid to cystic and show desmoplasia, although in contrast to DIG, they have a predilection for the temporal lobe and most commonly occur in an age group slightly older than is typical for DIG [ 13 ]. However, it should be noted that conventional gangliogliomas and DIG may exist on a morphologic spectrum, and a case with morphologic features of both entities has been described [ 14 ]. Supratentorial PNET with ganglionic differentiation (ganglioneuroblastomas) also show significant clinicopathologic overlap with DIG, as they are supratentorial, occur in young children, may be cystic and may show desmoplasia [ 15 ]. Cerebral ganglioneuroblastomas are extremely rare, and in the absence of treatment-related cytodifferentiation, will show a significant neuronal component of smaller cells. However, the cytomorphologic features of pure ganglioneuroblastoma have not been well-characterized. Other less likely differential considerations include pleomorphic xanthoastrocytoma and anaplastic large cell lymphoma (ALCL), both of which may contain large cells with eccentrically located nuclei and abundant cytoplasm. The temporal lobe predilection, lack of neuronal differentiation, presence of xanthomatous cells, smaller tumor size and older age of patients with pleomorphic xanthoastrocytoma should permit an easy distinction of the large cells in this tumor from those of DIG. Primary brain ALCL is exceedingly rare and generally occurs in older individuals, with a mean age of 29 years in one series [ 16 ]. Immunoreactivity for CD30, ALK and CD45 in ALCL and absence of similar immunoreactivity in DIG should facilitate a distinction in rare cases that occur in very young children. When the cytomorphologic findings of DIG are described in more cases, it is likely that the cytologic spectrum will mirror the histologic heterogeneity of this tumor. For example, astroglial cells predominated in the two cases of Hasegawa et al [ 4 ] while the ganglion cells predominated in ours. In addition, it is conceivable that an aspirate would only capture the immature neuroepithelial cells which frequently characterizes DIG. Nonetheless, it is concluded that when the large neuronal cells are encountered in an aspirate of a brain mass in a child, a combination of clinical, radiologic and immunohistochemical parameters can eliminate most of the differential possibilities. Competing interests The author(s) declare that they have no competing interests. Authors' contributions All authors made substantial contributions to the intellectual content and/or presentation of the manuscript. ITO (cytologist), diagnosed the cytopathological aspects of the case and co-supervised the entire project. JHK (neuropathologist), diagnosed the histological aspects of the case and co-supervised the project. OF wrote the initial version of the manuscript. MRM, DH and AWZ collected pathological, clinical and/or photographical information and revised the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC544864.xml

423141

α-Actinin-4-Mediated FSGS: An Inherited Kidney Disease Caused by an Aggregated and Rapidly Degraded Cytoskeletal Protein

Focal segmental glomerulosclerosis (FSGS) is a common pattern of renal injury, seen as both a primary disorder and as a consequence of underlying insults such as diabetes, HIV infection, and hypertension. Point mutations in theα-actinin-4 gene ACTN4 cause an autosomal dominant form of human FSGS. We characterized the biological effect of these mutations by biochemical assays, cell-based studies, and the development of a new mouse model. We found that a fraction of the mutant protein forms large aggregates with a high sedimentation coefficient. Localization of mutant α-actinin-4 in transfected and injected cells, as well as in situ glomeruli, showed aggregates of the mutant protein. Video microscopy showed the mutant α-actinin-4 to be markedly less dynamic than the wild-type protein. We developed a “knockin” mouse model by replacing Actn4 with a copy of the gene bearing an FSGS-associated point mutation. We used cells from these mice to show increased degradation of mutant α-actinin-4, mediated, at least in part, by the ubiquitin–proteasome pathway. We correlate these findings with studies of α-actinin-4 expression in human samples. “Knockin” mice with a disease-associated Actn4 mutation develop a phenotype similar to that observed in humans. Comparison of the phenotype in wild-type, heterozygous, and homozygous Actn4 “knockin” and “knockout” mice, together with our in vitro data, suggests that the phenotypes in mice and humans involve both gain-of-function and loss-of-function mechanisms.

Introduction In humans, ACTN4 mutations cause a form of focal segmental glomerulosclerosis (FSGS) ( Kaplan et al. 2000 ). This lesion, which describes a pattern of injury characterized by regions of sclerosis in some renal glomeruli, is a common finding in kidney disease from a wide range of primary disorders, including HIV infection, diabetes, and hypertension ( Ichikawa and Fogo 1996 ; Somlo and Mundel 2000 ). The four α-actinin genes encode highly homologous proteins that normally form approximately 100 kDa head-to-tail homodimers. While the best-defined function of α-actinin-4 is to cross-link and bundle actin filaments, α-actinins have been found to interact with a large and diverse set of other proteins ( Honda et al. 1998 ; Takada and Beggs 2002 ). α-Actinin-2 and α-actinin-3 are located predominantly in the sarcomere, while α-actinin-1 and α-actinin-4 are widely expressed. In the human kidney, only α-actinin-4 expression is detected ( Kaplan et al. 2000 ). Human ACTN4 -associated FSGS is inherited in an autosomal dominant pattern. By contrast, mice homozoygous for Actn4 null alleles have glomerular disease, while heterozygous Actn4 null mice have no readily apparent phenotype ( Kos et al. 2003 ). Transgenic mice harboring a mutant Actn4 targeted to the glomerular podocyte have an FSGS-like lesion, although it is not clear whether this is due to the dominant effect of the mutant Actn4 per se or to dysregulated Actn4 expression in this cell ( Michaud et al. 2003 ). We previously showed that FSGS-associated ACTN4 mutations increase the binding of α-actinin-4 to F-actin ( Kaplan et al. 2000 ). This was confirmed independently by different methodology ( Michaud et al. 2003 ). However, the relationship between altered actin binding and disease is unclear. Of particular interest is whether the human disease is due to a gain-of-function effect of mutations on protein function or due to a partial loss of normal function. We therefore performed a series of experiments to help us understand the biological consequences of mutations in ACTN4 . We demonstrate here that mutant α-actinins exhibit altered structural characteristics, localize abnormally, and have significantly diminished half-life. By developing a mouse model harboring a disease-associated point mutation, we confirm the pathologic effect of this mutation on glomerular function. Our results suggest that the major effects of Actn4 mutations are protein misfolding and accelerated degradation, leading to loss of normal α-actinin-4 function, α-actinin-4 aggregation, and progressive kidney disease. Results Mutant α-Actinin-4 Conformation In overlay assays (performed as per Chan et al. 1998 ; data not shown), mutant α-actinin-4 is able to bind both mutant and wild-type α-actinin-4. Since mutant α-actinin-4 polypeptides are able to interact, we used sucrose gradient assays to examine whether the mutant α-actinin-4 dimerizes normally. We produced radiolabeled in vitro translated wild-type and mutant α-actinin-4 (K228E, T232I, R235P) and subjected these proteins to centrifugation in 5%–20% sucrose gradients. We found that all of the wild-type α-actinin-4 eluted as a single peak, as expected. By contrast, approximately 80% mutant α-actinin-4 eluted at the same peak as in wild-type, while about 20% of the mutant protein eluted much more quickly and peaked in the first fraction. We then used a 10%–40% sucrose gradient in an attempt to better characterize this small peak. Again, in multiple experiments, about 20% of the mutant α-actinins eluted in the first fractions. α-Actinin has a sedimentation coefficient of 6.4 ( Feramisco and Burridge 1980 ), and both wild-type and the majority (approximately 80%) of mutant actinins migrated in this position. However, a significant fraction of mutant α-actinin-4 had a sedimentation coefficient equal to or greater than 11.3, the sedimentation coefficient of catalase ( Figure 1 ). This extremely high sedimentation rate suggests either the presence of large α-actinin-4 multimers or the existence of large (and perhaps insoluble) aggregates. When these experiments were performed in the presence of a large excess of cold α-actinin, this rapidly sedimenting fraction of mutant α-actinin was unchanged, showing that the abnormally conformed mutant α-actinin-4 could not be shifted back into normally sedimenting mutant/wild-type heterodimers ( Figure 1 ). Figure 1 Mutant α-Actinin-4 Sediments Abnormally (A) Pattern of sedimentation of α-actinin-4 in a 10%–40% sucrose gradient. Results shown are for wild-type α-actinin-4, K228E α-actinin-4, and K228E α-actinin-4 after addition of excess cold (wild-type) α-actinin. A fraction of the mutant α-actinin-4 sedimented at least as quickly than the highest molecular weight marker, catalase, which has a sedimentation coefficient of 11.3. This was observed with all of the other mutants tested as well (data not shown), but never with labeled wild-type α-actinin-4. Addition of cold α-actinin did not alter the sedimentation pattern seen with the mutant form of α-actinin-4. (B) Results illustrated graphically. Units are arbitrary. α-Actinin-4 Behavior in Cells We performed both transfection and nuclear injection studies in a conditionally immortalized podocyte cell line to look at the effect of disease-associated mutations on α-actinin-4 localization. Irrespective of the method used to express the mutant α-actinins in cells, we found altered localization of the mutants. Consistent with the altered sedimentation observed in vitro, mutant α-actinin-4 formed localized aggregates when expressed in cells. We used video microscopy to view the fate of the green fluorescent protein (GFP)–α-actinin-4 after nuclear injection of the cDNA. Similar results were observed in three independent sets of experiments. In each experiment, 15–35 cells were microinjected in the nucleus with plasmid DNA; of these, five to ten cells showed signal at 4–6 h after injection. Consistently, the mutants behaved abnormally, were unevenly distributed in the cell cytoplasm, and were much less dynamic compared with the wild-type proteins ( Figure 2 A; see also Videos S1–S3 ). These findings were consistent with the indirect immunofluorescence (IF) studies performed in Actn4 K228/K228 mice (see below). Figure 2 Mutant α-Actinin-4 Behavior in Cells (A) Mutant and wild-type α-actinin-4 show different localization and dynamics when expressed in a conditionally immortalized differentiated mouse podocyte cell line. Differentiated podocytes were injected in the nucleus with equal concentrations of expression plasmid for GFP fusions of mutant and wild-type actinins. At 2–4 h after injections, cells were imaged and both phase and fluorescence images recorded as described in the Materials and Methods . To illustrate changes in distribution of the fluorescence signal, three fluorescence images each 1 min apart were overlaid as red, green, and blue panes. Areas of fluorescence that were the same in all panes show as white, while dynamic areas are indicated by the color. The top panel indicates the initial phase and overlain dynamic fluorescence images of wild-type α-actinin-4, while the bottom two panels illustrate characteristic results for mutants K228E and T232I at 3 min time intervals. ( See Videos S1–S3 .) (B) Transfections in podocytes derived from mutant and wild-type mice. When transfected into conditionally immortalized podocytes of all three α-actinin-4 genotypes (+/+, K228E/+, or K228E/K228E), wild-type GFP–α-actinin-4 shows diffuse cytoskeletal localization. Mutant GFP–α-actinin-4 shows a similar alteration in localization when expressed in these three cells types. We transfected conditionally immortalized podocytes derived from Actn4 +/+ , Actn4 K228E/+ , and Actn4 K228E/K228E mice with either wild-type GFP–α-actinin-4 or K228E mutant GFP–α-actinin-4 ( Figure 2 B). The wild-type GFP–α-actinin-4 showed diffuse cytoplasmic localization in podocytes of all three genotypes. By contrast, mutant α-actinin-4 showed a similar aggregated appearance in all three cell types. This suggests the absence of a strong dominant effect of the mutant protein on the wild-type, as the endogenous K228E actinin does not alter the localization of the wild-type GFP-tagged protein. We developed α-actinin-4 “knockin” mice using the methods of homologous recombination in embryonic stem cells. Previously we developed an Actn4 “knockout” mouse ( Kos et al. 2003 ). As indicated schematically in Figure 3 A, mating these mice with germline Cre transgenic mice produced offspring in which an intronic loxP -flanked neomycin resistance cassette had been excised. We bred these heterozygous mice ( Actn4 K228E/+ ) to generate litters with wild-type, heterozygous, and Actn4 K228E/K228E mice. We genotyped mice by testing for the presence or absence of an engineered silent EarI site as described previously ( Kos et al. 2003 ). The homozygous K228E mice, in contrast to the Actn4 -deficient model, demonstrated normal levels of Actn4 mRNA expression ( Figure 3 B). RT-PCR and sequencing of the transcript confirmed that the K228E allele was in fact expressed in the homozoygous mice. However, in multiple tissues tested (kidney, lung, leukocytes, brain), as well as in fibroblasts derived from these mice, α-actinin-4 protein expression was markedly reduced, with an approximately 90% reduction in protein expression in Actn4 K228E/K228E homozygotes and an approximately 50% reduction in Actn4 K228E/+ heterozygotes ( Figure 3 D). Lymphocytes from a human subject heterozygous for one ACTN4 K2228E mutation similarly had an approximately 50% reduction in the amount of α-actinin-4 compared with related and unrelated controls ( Figure 3 E). Figure 3 “Knockin” Mouse Model (A) Targeting construct. As we have described elsewhere ( Kos et al. 2003 ), targeting initially resulted in a “knockout” allele, due to disruption of normal transcription, presumably by the intronic loxP -flanked neomycin resistance cassette. After breeding to Cre transgenic mice, this neomycin cassette was excised, as illustrated. (B) Northern blot analysis using kidney total RNA illustrates that the expression of the Actn4 transcript in K228E/K228E is similar to the expression in wild-type mice. (C) RT-PCR and sequencing of the relevant portion of Actn4 exon 8 confirms that the transcript in mice homozygous for the targeted allele contains the desired point mutation (top, wild-type; bottom, targeted). (D) Western blot showing markedly decreased expression of α-actinin-4 protein in K228E/K228E homozygous mice and moderately decreased expression in heterozygotes. Shown are blots using protein from cultured fibroblasts. Results were similar using protein extracted from lung, brain, liver, and kidney (data not shown). β-actin control is shown below. (E) Western blot showing expression of α-actinin-4 in lymphocytes from a human K228E/+ heterozygote ( Kaplan et al. 2000 ) and three wild-type controls (two related, one unrelated). β-actin control is shown below. α-Actinin-4 Degradation We observed decreased mutant α-actinin expression in an immortalized knockin mouse fibroblast homozygous for the Actn4 K228E mutation, compared with the wild-type fibroblast ( Figure 3 D). To help determine the fate of the mutant α-actinin, we performed pulse and pulse–chase studies. We labeled Actn4 K228E/K228E fibroblasts, Actn4 K228E/+ fibroblasts, and wild-type fibroblasts with [ 35 S]methionine for different timepoints (pulse) following incubation in methionine-deficient medium. In order to trace the newly synthesized 35 S-labeled α-actinins, we used an α-actinin-4-specific antibody to immunoprecipitate α-actinin-4 from the cell lysates. (We detected no α-actinin-4 in the cell pellets.) As shown in Figure 4 A, we found that at any timepoint, there was less mutant than wild-type α-actinin-4 synthesized. However, as shown in Figure 4 B, the rates of increase in labeled α-actinin-4 were similar, suggesting that the low expression level of mutant α-actinin-4 is not due to a defect in synthesis. We then labeled the mutant and wild-type fibroblasts with [ 35 S]methionine for 3 hours (pulse), following incubation in methionine-deficient media, and then incubated the cells in media containing excess cold methionine (chase) for different timepoints to follow the degradation of the newly synthesized α-actinin-4. As shown in Figure 4 C and 4 D, mutant α-actinin-4 degraded at a much faster rate than did the wild-type protein. The estimated half-life of mutant α-actinin-4 is about 15 h, while the half-life of the wild-type α-actinin-4 is much greater than 30 h. Three replicate experiments gave the same results. The rapid degradation of the K228E mutant α-actinin-4 is reversed by treatment with lactacystin ( Figure 4 E), indicating that this mutant form of α-actinin-4 is degraded through the ubiquitin–proteasome pathway. Figure 4 α-Actinin-4 Synthesis and Degradation (A and B) Synthesis of α-actinin-4 by wild-type and K228E/K228E fibroblasts. The rate of increase in the accumulation of mutant and wild-type α-actinin-4 is similar, as indicated by the superimposed shapes of the synthesis curves. (C and D) Pulse–chase experiments showing degradation of α-actinin-4 in K228E/K228E cells. Half-life of wild-type α-actinin-4 is greater than 30 h. Half-life of mutant α-actinin-4 is approximately 15 h. (E) Half-life of K228E mutant α-actinin-4 is restored to near-normal levels by the addition of lactacystin. Shown is labeled α-actinin-4 levels, expressed as a percentage of α-actinin-4 at time 0 and at 16 h and in the presence of 2.5 μM lactacystin in DMSO or in DMSO alone. In Vivo Phenotype We performed standard histologic analyses of kidneys from Actn4 K228/K228E mice, as well as Actn4 K228/+ and wild-type littermates. In Actn4 K228/K228E mice as old as 13 mo, we saw no abnormalities by light microscopy with periodic acid–Schiff (PAS) and hematoxylin-and-eosin (H & E) staining. All of the 11 Actn4 K228/K228E kidneys examined by electron microscopy had abnormalities in podocyte structure. Typically, these consisted of focal areas of foot process effacement ( Figure 5 A). By contrast, we found mild abnormalities in one of 13 wild-type and one of nine Actn4 K228E/+ littermates. Mice were typically genotyped at or shortly before the time of weaning (at approximately 3 wk of age). Only 10% (23 of 231) of the offspring of crosses between heterozygous mice were homozygous for the K228E change, suggesting increased peri- or neonatal lethality in the homozygous mice, similar to what we have observed in Actn4 null mice ( Kos et al. 2003 ). In Actn4 K228/K228E mice, we frequently observed the appearance of abnormal electron-dense structures in the podocyte cell bodies ( Figure 5 D). Figure 5 In Vivo Phenotype Electron micrographs from Actn4 wild-type (A) and Actn4 K228E/K228E mice (B–D). As shown, Actn4 K228E/K228E mice were found to have abnormalities that were typically focal, with some areas of podocyte foot process effacement (B), as well as areas that appeared essentially normal (C). Bottom image ([D] using tannic acid counterstaining) illustrates electron-dense deposits observed in several podocyte cell bodies in Actn4 K228E/K228E mice. No such deposits were observed in wild-type or heterozygous mice. We measured urine protein excretion in mice of five different genotypes: wild-type ( Actn4 +/+ ), heterozyogtes for either a null or K228E allele ( Actn4 +/– and Actn4 K228E/+ ), and homozoygotes for either a null or K228E allele ( Actn4 –/– and Actn4 K228E/K228E ). Results were quite variable within each genotypic group of mice (likely reflecting differences in age and genetic background). However, the overall pattern of protein excretion was similar in the Actn4 +/+ , Actn4 +/– , and Actn4 K228E/+ mice, while both groups of homozygous mutant mice ( Actn4 –/– and Actn4 K228E/K228E ) had significantly greater—and similar—degrees of proteinuria ( Figure 6 B). We did not see significant differences in serum creatinine levels or blood urea nitrogen levels (BUN) between mice of the three genotypes ( Figure 6 A), nor did we identify any single mutant mouse with significant BUN or creatinine elevations. Figure 6 Biochemical Characteristics of Mutant Mice (A) Average BUN and creatinine levels in Actn4 K228E/– ( n = 12), Actn4 +/+ ( n = 8), and Actn4 K228E/K228E ( n = 12) mice at the time of sacrifice. Differences were not statistically significant. Error bars show standard deviation. (B) Summary of proteinuria in wild-type, Actn4 +/– , Actn4 K228E/– , Actn4 –/– , and Actn4 K228E/K228E mice, measured by albumin dipstick. Distribution of measurements are illustrated graphically for each genotype. To determine whether the K228E point mutation altered α-actinin-4 localization in vivo, we performed indirect IF studies. As shown in Figure 6 A, α-actinin-4 appears to be mislocalized and aggregated in Actn4 K228E/K228E kidneys. By contrast, we see no difference in the expression of slit-diaphragm proteins ZO-1 and nephrin. The merged ZO-1/Actn4 images show overlapping patterns of expression in the wild-type and Actn4 K228E/+ mice, but clearly distinct expression patterns in the Actn4 K228E/K228E mice. We examined α-actinin-4 localization in a human ACTN4 K228E/+ kidney biopsy sample ( Figure 6 B). Although we are cautious in our interpretation, given the availability of only one biopsy sample, there appears to be a more punctate appearance to the α-actinin-4 distribution, consistent with the existence of some α-actinin-4 aggregates in human heterozygotes. Discussion We have previously shown that dominantly inherited point mutations in the α-actinin-4 gene ACTN4 cause a form of human glomerular disease ( Kaplan et al. 2000 ). We have also shown that mice lacking α-actinin-4 expression develop a severe glomerular lesion ( Kos et al. 2003 ). Here we have further explored the biochemical and cell biologic alterations caused by disease-associated α-actinin-4 mutations. Human α-actinin-4-associated FSGS is characterized by a dominant pattern of inheritance. Affected individuals typically develop disease in adulthood. Some individuals develop progressive renal failure, others develop moderate proteinuria, while a few show no evidence of kidney dysfunction well into adulthood. This contrasts with other recently elucidated inherited disorders of the podocyte caused by mutations in the slit-diaphragm proteins nephrin and podocin, where disease typically presents in the neonatal period or in childhood and follows a recessive pattern of inheritance ( Kestila et al. 1998 ; Boute et al. 2000 ). Mice lacking slit-diaphragm proteins CD2AP and Neph-1 similarly present with very early-onset nephrosis (Shih et al. 1998; Donoviel et al. 2001 ). Furthermore, in contrast to the typically diffuse podocyte foot process effacement observed in kidney biopsies from individuals with these recessive forms of disease, individuals with ACTN4 -associated FSGS have focal podocyte abnormalities, nonnephrotic levels of proteinuria, and slowly progressive adult-onset disease leading to significant (and often end-stage) renal failure in adulthood. These phenotypic differences themselves suggest a different mechanism of disease from what is observed with slit-diaphragm protein defects. Our earlier experiments suggested that mutant α-actinin-4 binds actin filaments more strongly than wild-type α-actinin-4 in vitro. However, this may reflect a propensity toward oligomerization rather than increased F-actin binding per se. The finding that the mutant α-actinin-4 forms aggregates with greatly decreased half-life suggests two possible models to explain the human (and mouse) disease. One model would explain the development of podocyte damage as a direct effect of protein aggregation and the toxic effects of such aggregation, as is observed in several degenerative neurologic conditions such as Alzheimer and Parkinson diseases ( Horwich 2002 ). The second model explains the disease as a loss-of-function disease, reflecting the increased rate of mutant α-actinin-4 degradation. We do not regard these models as mutually exclusive. In fact, we believe it likely that the development of disease may involve both of these mechanisms. It is interesting to note that α-actinin-4-mediated kidney disease bears some similarities to the adult-onset neurodegenerative condition Huntington disease. In Huntington disease, dominant mutations that lead to expanded polyglutamine tracts cause neurodegeneration. The mutant huntingtin protein is misfolded and forms aggregates that are thought to have dominant, toxic effects on neuron function ( Bucciantini et al. 2002 ; Bates 2003 ). These proteins also play critical and nonredundant roles in the relevant organs: analogous to what we have observed in α-actinin-4-deficient mice, mice with reduced huntingtin expression show abnormal brain development and perinatal lethality ( White et al. 1997 ). In contrast to humans, mice heterozygous for α-actinin-4 mutations do not have overt disease. We suspect that in humans, over a timespan of several decades, the combination of decreased α-actinin-4 expression and the formation of aggregates ultimately proves toxic. We suggest that the relatively short life of mice compared with that of humans may be the major explanation of this difference. Not all humans carrying disease-associated mutations develop clinical disease ( Kaplan et al. 2000 ). Disease, in both human and murine heterozygotes, likely requires a “second hit,” either in the strict genetic sense of a second mutation in the relevant cell type or a “physiologic hit,” such as elevated blood pressure, renal toxin exposure, or vascular disease, to name three examples. We suspect that renal stresses will uncover deleterious renal phenotypes in heterozygous mice, similar to the disease observed in humans. With aging and gradual accumulation of aggregates in the terminally differentiated podocyte, we believe that humans with ACTN4 mutations become increasingly susceptible to minor insults. We note also that mice, unlike humans, express α-actinin-1 in podocytes ( Kos et al. 2003 ). This may help stabilize mutant α-actinin-4 or may produce more redundancy, giving the mouse glomerulus greater protection to perturbations in this pathway. We note that, as shown in Figure 7 , the appearance of α-actinin-4 is more punctate in the kidney from a K228E heterozygous individual than a control, consistent with the existence of aggregates in heterozygous humans. This effect, however, is subtle, and is consistent with the lack of any detectable renal phenotype in several humans who carry disease-associated ACTN4 mutations ( Kaplan et al. 2000 ). Although the number of ACTN4 mutations we have found is small, we have not detected any human disease-associated ACTN4 mutations to date that lead to premature termination, suggesting that simple haploinsufficiency may not by itself cause disease (our unpublished data). Figure 7 In Situ Protein Localization (A) IF studies of glomerular protein expression in Actn4 +/+ , Actn 4 K228E/+ , and Actn4 K228E/K228E mice. As indicated, expression of α-actinin-4, ZO-1, and nephrin is shown, as is a merged image of α-actinin-4 and ZO-1 expression. (B) Glomerular expression of α-actinin-4 in normal human kidney and in an individual heterozygous for a K228E mutation. As indicated in Figure 5 and in the text, heterozygous K228E Actn4 knockin mice have no clear phenotype either by histologic analysis at the light and electron microscopic levels or by analysis of urine protein and serum creatinine. Even at more advanced ages, the Actn4 K228E/+ mice appear normal. By contrast, we observe clear glomerular phenotypes in both Actn4 –/– and Actn4 K228E/K228E mice. Our genetic observations are consistent with our biochemical observations. Specifically, we believe that α-actinin-4 mutations lead to a reduction in normal α-actinin-4 activity and to protein aggregation and that glomerular phenotypes reflect both loss of normal α-actinin-4 function and toxic effects of aggregated α-actinin-4. While the disease observed in homozygous Actn4 K228E/K228E mice may primarily reflect loss of function resulting from rapid α-actinin-4 degradation, heterozygous humans may show slow development of podocyte damage from the effects of α-actinin-4 aggregation. Is alteration of α-actinin-4 expression or conformation a cause or a mediator of secondary forms of kidney disease? These seem plausible hypotheses given the data presented here, together with results from other investigators showing alterations in α-actinin-4 levels in association with proteinuria in certain animal models ( Shirato et al. 1996 ; Smoyer et al. 1997 ). This suggests that physiologic processes that alter the expression, conformation, or both of this (and other) cytoskeletal proteins, either at the protein or the transcript level, might be amenable to interventions that restore normal patterns of protein expression. Materials and Methods Cell and cell culture Mouse podocytes were kindly provided by Dr. Peter Mundel (Albert Einstein College of Medicine, Bronx, New York, United States) and cultured as described previously ( Mundel et al. 1997 ). In brief, undifferentiated podocytes were cultured in RPMI-1640 (Cellgro, CellGenix, Freiburg, Germany) medium containing 10% fetal calf serum (FCS) and 20 U/ml γ-interferon (γ-IFN) at 33°C and 5% CO 2 . Differentiated podocytes were cultured in the medium containing no γ-IFN at 37°C. Additional conditionally immortalized podocytes from “knockin” litters were generated exactly as described previously ( Mundel et al. 1997 ). Fibroblasts were derived from lungs dissected from newly sacrificed adult mouse littermates of different genotypes. Cells were propagated in culture and immortalized by transfection with an SV-40 large T-antigen expression plasmid using Fugene 6 transfection reagent (Roche, Basel, Switzerland). Protein extraction Fibroblasts were allowed to grow to confluence, then scraped off the tissue culture plate in the presence of cold phosphate-buffered saline (PBS) and spun at 3,000 rpm at 4°C for 10 min. Lymphocytes were isolated from whole blood with Histopaque-1077 solution (Sigma, St. Louis, Missouri, United States) following the manufacturer's instructions. The pellets were resuspended in ice-cold lysis buffer (150 mM NaCl, 50 mM Tris [pH 8.0], 1% Triton X-100, 1 mM Na-orthovanadate, 4 μM microcystin, and protease inhibitor). We collected the supernatant and estimated the protein concentration using either the Bradford method or equalizing the protein concentration in different lysates by Western blot using β-actin as a standard. Transient transfection and Immunocytochemistry We mutated a wild-type pBluescript-GFP-ACTN4 clone using a QuickChange kit (Stratagene, La Jolla, California, United States) to create clones harboring each of three disease-associated mutations (K228E, T232I, S235P) ( Kaplan et al. 2000 ). These mutant and wild-type α-actinin-4 clones were transfected into podocytes using Fugene 6 (Roche). Cells 60 h after transfection were fixed in 2% paraformaldehyde and 4% sucrose in PBS for 5 min and then permeabilized in 0.3% Triton for 5 min. Fixed cells were blocked with 2% FCS, 2% BSA, 0.2% fish gelatin in PBS for 60 min and incubated with anti-α-actinin-4 antibody, and rabbit antigen–antibody complexes were visualized with fluorochrome-conjugated secondary antibodies. Sucrose gradient studies Sucrose gradients of 5%–20% and 10%–40% were made using a buffer containing 0.02 M Tris–HCl (pH 7.5), 0.15M NaCl, 0.1 mM EDTA, and 0.2 mM of DTT and were internally calibrated with BSA, carbonic anhydrase, and catalase. In vitro translated and radiolabeled wild-type and mutant α-actinin-4 (K228E, T232I, S235P) were loaded onto the gradient, centrifuged at 40,000 rpm for 15 h at 4°C, and eluted into 0.2 ml fractions. Eluates were analyzed by SDS-PAGE and autoradiography. Pulse–chase studies Immortalized mouse fibroblasts were incubated in methionine-free MEM medium containing 10% FCS (dialyzed overnight using 12K-14K SPECTRA/POR porous membrane in normal saline at 4°C) and 25 mM HEPES for 20 min at 37°C. [ 35 S]Methionine was added to a final concentration of 0.1 mCi to pulse the cells. Cells were pulse labeled for 0, 15, 30, 60, 120, 180, and 240 min. To study the degradation of α-actinin-4, cells were pulsed for 3 h and then chased for 0, 3, 6, 12, 20, 24, and 30 h with excess cold methionine. We used anti-α-actinin-4 antibody recognizing the N-terminus to precipitate α-actinin-4 from the cell lysates. Protein A–sepharose beads (Pierce Biotechnology, Rockford, Illinois, United States) were preincubated with the anti-α-actinin-4 antibody for 2 h at 4°C and then incubated with the lysates overnight at 4°C. The beads were washed with lysis buffer and resuspended in SDS-PAGE loading buffer. Samples were resolved on a 10% polyacrylamide gel and visualized by exposure to radiographic film. For lactacystin treatment, cells were first pulsed for 3 h as above, followed by addition of 2.5 μM lactacystin dissolved in DMSO (A. G. Scientific, San Diego, California, United States) or 0.125% DMSO alone with cold methionine. Nuclear DNA injection and imaging For injection and imaging, cells were cultured on MatTek Corporation (Ashland, Massachusetts, United States) 35 mm coverslip dishes in F12 media (BioFluids, BioSource International, Carmarillo, California, United States) without phenol red and supplemented with 10 mM HEPES and antibiotics. Plasmid DNA at 0.5–2.0 ng/nl was injected in cell nuclei using a Narishige (Lake Forest, California, United States) IM-200 picoliter pressure injection system. OD microcapillary glass pipettes (1.0 mm) were pulled to a fine tip using a Narishige PB-7 needle puller. Cells were maintained at 37°C using a modified Harvard Apparatus (Hopkinton, Massachusetts, United States) microscope incubator mounted on a Nikon (Tokyo, Japan) Diaphot 300 inverted microscope. Images were collected using a Princeton Instruments MicroMax 1300Y cooled CCD camera (Roper Scientific, Tucson, Arizona, United States). Excitation and emission wavelengths were controlled using dichroic and bandbass filters from Omega Optical (Brattleboro, Vermont, United States) and a Sutter Instrument (Novato, California, United States) Lambda 10–2 filter wheel image acquisition. Device control and postacquisition processing were done with Isee Imaging Software (Inovision Corporation Raleigh, North Carolina, United States). Actinin dynamics display To display the changes in GFP–actinin distribution over time, images collected at 1 min intervals were used sequentially as red, green, and blue channels of a RGB composite image. Areas of signal that did not change have equal representation in each of the channels and generate a white signal in the final image. Areas of signal that did change on a minute-to-minute basis are indicated by either a red, green, or blue hue. For example, a region rich in red would indicate an signal present in the first image, but absent in the two sequential images, indicative of a withdraw or loss of signal in that region. Generation of K228E mutant mice Previously, we described the development of a mouse model lacking detectable Actn4 expression ( Kos et al. 2003 ). These mice harbor a germline mutation in exon 8 of Actn4 encoding a K228E substitution, as well as an intronic loxP -flanked neomycin resistance cassette. We bred these mice to transgenic mice with germline expression of Cre recombinase (129-TgN(Prm-Cre)58Og; Jackson Laboratory, Bar Harbor, Maine, United States). We verified excision of the neomycin resistance cassette by PCR. Heterozygous mice were crossed to obtain mice homozygous for the K228E substitution. Mice were genotyped for the K228E as described previously ( Kos et al. 2003 ). We used Trizol reagent to extract RNA from kidneys for RT-PCR and for Northern blot analysis. Mouse phenotyping Freshly harvested kidneys were fixed in Bouin's solution. H & E and PAS staining were performed using standard methodology. Electron microscopy was performed after fixation in Karnovsky's media using standard diagnostic protocols. For the electron micrographs, all of the glomeruli imaged were from as deep into the renal cortex as possible. Indirect IF studies were performed using standard methods ( Kos et al. 2003 ). Urine microalbumin was assessed by a reader blinded to mouse genotype using albumin dipsticks (Albustix, Bayer, Leverkusen, Germany). BUN and creatinine measurements were performed at the clinical chemistry laboratory at Brigham and Women's Hospital. Supporting Information Video S1 Podocytes Following Nuclear Injection of Wild-Type GFP-α-Actinin-4 cDNA (14.8 MB MOV). Click here for additional data file. Video S2 Podocytes Following Nuclear Injection of K228E GFP-α-Actinin-4 cDNA (11.8 MB MOV). Click here for additional data file. Video S3 Podocytes Following Nuclear Injection of T232I GFP-α-Actinin-4 cDNA (10.9 MB MOV). Click here for additional data file. Accession Numbers GenBank accession numbers for genes and proteins discussed in this paper are NM_004924 and NP_004915 (human ACTN4 ; also LocusLink ID 81) and NM_021895 and NP_068695 (mouse Actn4 ; also LocusLink ID 60595). OMIM numbers are 603278 ( FSGS-1 ) and 604638 ( ACTN4 ).

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC423141.xml

535537

Role of a critical visceral adipose tissue threshold (CVATT) in metabolic syndrome: implications for controlling dietary carbohydrates: a review

There are likely many scenarios and pathways that can lead to metabolic syndrome. This paper reviews mechanisms by which the accumulation of visceral adipose tissue (VAT) may contribute to the metabolic syndrome, and explores the paradigm of a critical VAT threshold (CVATT). Exceeding the CVATT may result in a number of metabolic disturbances such as insulin resistance to glucose uptake by cells. Metabolic profiles of patients with visceral obesity may substantially improve after only modest weight loss. This could reflect a significant reduction in the amount of VAT relative to peripheral or subcutaneous fat depots, thereby maintaining VAT below the CVATT. The CVATT may be unique for each individual. This may help explain the phenomena of apparently lean individuals with metabolic syndrome, the so-called metabolically normal weight (MONW), as well as the obese with normal metabolic profiles, i.e., metabolically normal obese (MNO), and those who are "fit and fat." The concept of CVATT may have implications for prevention and treatment of metabolic syndrome, which may include controlling dietary carbohydrates. The identification of the CVATT is admittedly difficult and its anatomical boundaries are not well-defined. Thus, the CVATT will continue to be a work in progress.

Introduction Arguably, the major pathogenic factor in the metabolic syndrome is central obesity [ 1 , 2 ]. While abdominal obesity is determined by the accumulation of both subcutaneous adipose tissue (SCAT) and visceral adipose tissue (VAT), the excess accumulation of VAT appears to play a more significant pathogenic role. VAT depots, located in the body cavity beneath the abdominal muscles, are composed of the greater and lesser omentum (peritoneum that is attached to the stomach and links it with other abdominal organs) and the mesenteric fat. A lesser amount of VAT is located retroperitoneally. In general, VAT accounts for up to 20 percent of total fat in men and 5–8 percent in women. The abdominal SCAT is located immediately beneath the skin and on top of the abdominal musculature. The predominance of lower body fat is SCAT, most of which is stored in the femoral and gluteal regions [ 3 - 5 ]. Abdominal obesity can reflect a predominance of flabby SCAT; a firm, only modestly enlarged waist line resulting from deep VAT pushing the abdominal musculature outward; or a combination of enlarged SCAT and VAT depots. With the advent of more precise imaging techniques, e.g., magnetic resonance imaging (MRI) [ 6 ], computed tomography (CT) [ 7 ], and ultrasound [ 8 ], it has become evident that the accumulation of VAT not only accompanies but antedates the onset of the components of the metabolic syndrome and related disorders, e.g., insulin resistance, hypertension [ 9 ], glucose intolerance [ 10 ], type 2 diabetes, and coronary heart disease [ 11 ]. To date, it has not yet been established that insulin resistance, i.e., resistance of cells to insulin's effects, is responsible for the onset of the multiple risk factors associated with insulin resistance syndrome and subsequent development of atherosclerosis and cardiac events [ 12 ]. In fact, National Cholesterol Education Program Adult Treatment Panel (ATP III) criteria for Metabolic Syndrome have been found to have a low sensitivity for predicting insulin resistance [ 13 - 15 ] and may be better thought of as predictors for cardiovascular risk [ 16 ]. In a recent study of a large number of apparently healthy men and women of varying age, VAT area was significantly associated with all of the metabolic syndrome criteria as defined by the NCEP ATP III. This was independent of insulin sensitivity and SCAT area. Insulin sensitivity was found to be independently associated with the criteria for HDL cholesterol, triglycerides (TGs), and fasting plasma glucose (FPG). SCAT area was independently correlated with only waist circumference after adjusting for VAT area and insulin sensitivity [ 11 ]. In addition, the study results showed that clinical assessments of increased waist size and TG levels are strongly associated with decreased insulin sensitivity and increased VAT in individuals with fasting FPG <6.4 mmol/L [ 11 ]. The term "metabolic syndrome" is now preferable to "insulin resistance syndrome," and has a prevalence of 25 percent in U.S. individuals age >20, rising to >40 percent by age 60 [ 17 ]. The importance of central obesity is well-recognized in the definitions of metabolic syndrome [ 18 ] per the American College of Endocrinology, [ 19 , 20 ] National Cholesterol Education Program Adult Treatment Panel (ATP III), [ 21 ] European Group for the Study of Insulin Resistance, [ 22 ] and World Health Organization (WHO) [ 23 ]. However, even apparently lean individuals with normal BMIs can have a significant accumulation of VAT with increased risk factors for cardiovascular disease and diabetes (metabolically obese normal weight; MONW) [ 24 - 26 ]. Meanwhile, obese individuals with large BMIs but relatively little VAT can present with normal metabolic profiles and a paucity of risk factors for metabolic syndrome, cardiovascular disease, and diabetes, i.e., the metabolically normal obese; MNO) [ 5 , 27 ]. Ectopic Fat Storage Syndrome The ectopic fat storage syndrome hypothesis suggests that as adipocytes hypertrophy and reach their capacity for storing more fat, then additional fat from excess dietary lipids or calories is deferred to non-adipose tissues intracellularly, e.g. liver, skeletal muscle, heart, and the beta cells of the pancreas where they can exert toxic effects and dysfunction [ 7 ]. This "lipotoxicity" may also be exacerbated by impaired oxidation of fat within tissues [ 7 , 28 - 30 ]. Furthermore, adipose tissue is a major endocrine organ that secretes numerous polypeptide hormones and cytokines that are proinflammatory and proatherogenic. These play a major role in affecting insulin action in skeletal muscle and creating a low-grade state of inflammation and endothelial dysfunction [ 31 ]. Compared to SCAT, VAT has been correlated more with endothelial dysfunction [ 32 , 33 ]. CVATT – A Working Hypothesis It must be emphasized that the current proposal is a working hypothesis. Figure 1 describes a critical VAT threshold (CVATT) which is unique for a given individual and represents a range for the accumulation of a critical mass of VAT (CVATT) that when achieved, leads to the development of metabolic syndrome. Note that insulin sensitivity is important for weight gain [ 34 ] and accumulation of VAT, and investigators have proposed that insulin resistance may actually, to a certain extent, be beneficial by protecting cells with already impaired fatty acid oxidation. Once the CVATT is reached, insulin resistance (IR) occurs, which may be protective initially [ 29 , 35 - 37 ]. In addition to protecting against further weight and fat gain [ 34 , 38 - 41 ], insulin resistance prevents glucose and more fat from entering the cell and becoming preferentially oxidized. Hence, insulin resistance also allows intracellular fat already present within the cell to become oxidized rather than cause further damage through "lipotoxicity [ 29 , 30 , 40 , 42 , 43 ]." Figure 1 Critical Visceral Adipose Tissue Threshold (CVATT). According to the hypothesis, there is an individual range for accumulating a critical amount of visceral adipose tissue (VAT). Insulin sensitivity is important for weight gain and accumulation of VAT. Once the critical VAT threshold (CVATT) is reached, insulin resistance occurs, which may be protective initially and impair further weight and fat gain. Continuation of VAT accumulation can lead to metabolic syndrome. However, only a modest weight loss (5–10 percent) with accompanying VAT loss can reverse the process. Implications of VAT loss It is encouraging that only a modest loss of 5–10 percent of body weight in obese patients is associated with preferential mobilization of VAT compared to SCAT, leading to simultaneous improvement in all metabolic markers of CHD risk. Such modest weight loss can prevent and reverse type 2 diabetes [ 44 - 48 ], and sustained weight loss in obese women results in a reduction in elevated inflammatory cytokine levels and an amelioration of endothelial dysfunction [ 49 , 50 ]. Surgical removal of VAT may reduce insulin resistance and plasma insulin levels [ 51 , 52 ], while liposuction of SCAT does not confer metabolic benefits [ 53 ]. Weight loss usually leads to VAT reduction as well as reduction of depots of fat in non-adipose organs, thereby improving insulin sensitivity [ 48 ]. However, once individuals improve insulin sensitivity by losing weight and crossing beneath their CVATT [ 48 ], they may now be more susceptible to weight gain and struggle to maintain this new state. With total weight loss, those with greater amounts of VAT initially lose more VAT, and VAT is more sensitive to weight reduction because the VAT adipocyte is more metabolically active and sensitive to lipolysis [ 5 , 54 ]. After the initial weight loss, further dietary restriction may lead to an overall reduction in body fat, rather than specific loss from a particular site. The metabolic improvements observed with only modest reductions in total weight underscore the importance of VAT in insulin resistance and metabolic abnormalities [ 48 , 55 ]. Once the individual has lost a significant amount of VAT and is now below his CVATT, improvement in insulin sensitivity does not bear a linear relationship to the magnitude of weight loss [ 48 ]. The identification of the CVATT is admittedly difficult and its anatomical boundaries are not well-defined. Thus, the CVATT will continue to be a work in progress. While there are numerous studies linking VAT quantity to insulin resistance and metabolic syndrome, this does not necessarily prove that VAT is the cause. However, there are a number of plausible mechanisms linking VAT to the metabolic syndrome. Adipose tissue as an endocrine organ Adipokines Once thought to be an inert energy storage depot, adipose tissue is now known to be a critical endocrine organ. The term "adipocytokines" or "adipokines" has been used to describe the numerous adipocyte secretory products which include: adiponectin, adipsin, estrogen, angiotensin II, angiotensinogen, leptin, plasminogen activator I (PAI-1), agouti protein, resistin [ 56 ], acylation stimulating protein (ASP), bone morphogenic protein (BMP), prostaglandins, IGF-1, and various IGF binding proteins, tumor necrosis factor alpha (TNFα), interleukins (ILs), transforming growth factor (TGF)-B [ 57 ], and fibroblasts, as well as FFAs themselves. Adipokines such as IL-6 and PAI-1 are more highly secreted by VAT than abdominal SCAT [ 58 , 59 ], while leptin is more highly secreted by SCAT [ 60 ]. Adipokines from VAT can be delivered via the portal system directly to the liver where they can affect hepatic, and ultimately systemic, inflammation. In an ex vivo study, VAT released greater amount of IL-6 and PAI-1 compared with abdominal SCAT [ 58 , 61 ]. Adiponectin has many beneficial vascular and metabolic effects, e.g., it serves as an anti-inflammatory molecule for vascular walls as well as adipose tissue, inhibits vascular smooth muscle proliferation, protects endothelium from macrophage adhesion and macrophage-induced injury [ 62 ], may increase fatty acid oxidation in peripheral tissues [ 63 ], protects against ectopic fat storage and has been linked with insulin sensitivity [ 64 , 65 ]. Ironically, although produced by adipose tissue, adiponectin levels are lowered with greater degrees of obesity and with overfeeding. Decreased concentrations of adiponectin are associated with type 2 diabetes, hypertension, elevated glucose levels, insulin and TGs, and cardiovascular disease (CVD). It has been suggested that adiponectin is under feedback inhibition in obesity and reduced in patients with metabolic syndrome [ 66 ]. Adiponectin mRNA and protein levels have been found to be reduced in omental VAT compared with SCAT [ 67 ], and VAT may also produce an as-yet-identified factor that destabilizes adiponectin mRNA [ 66 , 68 ]. The strong inverse correlation between serum adiponectin levels and VAT mass may in part explain the link between VAT and metabolic syndrome [ 66 ]. Over 90 percent of the adipokines released by adipose tissue, except for adiponectin and leptin, could be attributed to non-fat cells, e.g., macrophages, retained in the adipose tissue matrix [ 61 ]. Implications of fat mass expansion Fat mass can expand in one of two ways: individual adipocytes can increase in volume or they can increase in number as more are derived from preadipocytes. As adipocytes grow larger, they become dysfunctional. The total number of adipocytes is increased with increasing fat mass, but it is the increased number and percentage of large adipocytes, compared to the smaller ones, that may partially account for the inability of adipose tissue to function properly [ 69 ]. While the smaller adipose cells tend to be more insulin sensitive, large adipocytes become insulin resistant and contribute more to the metabolic problems associated with obesity [ 69 ]. Preadipocytes from the SCAT depots have a greater differentiation capacity than those from the VAT depots [ 70 , 71 ]. The differentiation of preadipocytes into lipid-storing adipocytes is regulated in part by the nuclear hormone receptor, peroxisome proliferators activated receptor (PPAR). Activation of this receptor by natural ligands, such as prostaglandin metabolites, or synthetic ligands, such as thiazolidinediones (TZDs), leads to stimulation of the differentiation pathway [ 71 ]. This increases the number of smaller adipocytes in SCAT with a high avidity for FA and TG uptake. These increased adipose stores made up of new, smaller, more insulin sensitive adipocytes act as a sink or powerful 'buffers,' avidly absorbing circulating fatty acids and triglycerides in the postprandial period. This prevents their diversion to non-adipose tissues, thereby protecting against ectopic fat syndrome and metabolic syndrome. It has been proposed that an inability to differentiate new adipocytes to accommodate and store excess energy, underlies the development of type 2 diabetes [ 72 , 73 ]. A thiazolidinedione (TZD) paradox TZDs can increase the number of new fat cells, and because obesity is a major cause of insulin resistance, this represents an apparent paradox. Ex-vivo studies of human preadipocytes from SCAT and VAT depots have demonstrated that TZD-stimulated differentiation is much greater in SCAT than VAT preadipocytes [ 71 ]. Since TZDs selectively promote adipogenesis in SCAT and not VAT, this would encourage the redistribution of body fat away from "harmful" VAT sites and toward "safer" SCAT ones [ 74 - 76 ]. Thus, in this way, TZDs could allow for pushing the patient to below his CVATT. Paradoxically, the TZDs can lead to weight gain while improving insulin sensitivity as the new SCAT adipocytes continue to trap FA and as fat storage continues, eventually the new adipocytes will enlarge, become less insulin sensitive, and ultimately contribute to insulin resistance [ 77 ]. TZDs may also exert anti-inflammatory effects on adipocytes by reducing the production of serum amyloid A (SAA) and preventing the TNFα-mediated expression of adiponectin production [ 69 ]. Adipose macrophages Macrophages increase their accumulation within fat depots in direct proportion to increases in adipose tissue and adipocyte size. The increased macrophage activity observed in the adipose tissue of the obese may reflect a combination of conversion of local preadipocytes to macrophages and activation and recruitment of resident macrophages and circulating monocytes. This seems to occur after the onset of adiposity but prior to insulin resistance, and supports the notion that pathophysiological consequences of obesity involve macrophages and inflammation that contribute to insulin resistance and metabolic syndrome [ 78 , 79 ]. Evidence suggests that macrophages and adipocytes not only express overlapping sets of genes and serve similar functions, but also commingle in the same part of the body – the fat tissue [ 80 ]. VAT Versus SCAT (See figure 2 ) Figure 2 VAT versus SCAT VAT There are numerous inherent differences between VAT and SCAT. VAT is a major predictor for insulin resistance [ 81 ] and metabolic syndrome [ 11 ]. Compared to SCAT, VAT adipocytes have a higher rate of lipolysis, which is more readily stimulated by catecholamines and less readily suppressed by insulin [ 82 ]. VAT also produces more IL-6 and plasminogen activator inhibitor-1 (PAI-1) [ 81 ]. The "Portal Theory" suggests that insulin resistance and many of its related features could arise from VAT delivering free fatty acids (FFAs) at a high rate to the liver via the portal vein into which VAT directly drains [ 83 , 84 ]. This, in turn, would increase hepatic glucose production, reduce hepatic insulin clearance, and ultimately lead to insulin resistance, hyperinsulinemia, hyperglycemia as well as non-alcoholic fatty liver disease (NAFLD). FFA flux could also lead to enhanced production of triglycerides (TGs) and apolipoprotein B-rich lipoproteins, which are features of the insulin resistance syndrome [ 55 , 85 ]. Delivery of VAT derived pro-inflammatory cytokines may contribute to hepatic pathology such as non-alcoholic steatohepatitis (NASH). VAT also releases a large amount of glycerol which enters the liver where it can be converted to glucose, thereby contributing to hyperglycemia [ 86 ]. It is likely that the relationship observed between VAT and metabolic complications may not exclusively result from FFA flux from VAT into the portal vein and the portal theory does not adequately hold up as the sole explanation of the role of VAT in metabolic syndrome [ 7 ]. VAT has twice the glucose uptake rate as SCAT Recently, omental VAT cells have been shown to have an approximately two-fold higher rate of insulin-stimulated glucose uptake compared with SCAT adipocytes, and this could be explained by a higher GLUT-4 expression [ 87 ]. Perhaps in situations with a high intake of dietary glycemic load, a higher rate of glucose uptake and subsequently lipogenesis might be one mechanism by which TGs are stored preferentially in the VAT depot. VAT is highly lipolytic and resistant to insulin's lipogenic effects yet apparently can remain insulin sensitive to glucose uptake. This efficiency in glucose uptake may reflect VAT's ability to accumulate and maintain its activity. Enhanced glucose utilization in VAT would be accompanied by less lipid oxidation, which would indirectly promote TG storage [ 87 ]. Testosterone VAT has a high density of androgen receptors and testosterone which can amplify its own effect by up-regulation of androgen receptors, inhibiting the expression of lipoprotein lipase (LPL) and FA uptake [ 5 , 88 ]. In men, VAT is strongly negatively correlated with plasma total and free testosterone and sex-hormone binding globulin (SHBG) concentrations. Thus, in young men whose plasma total testosterone and free testosterone are high, the amount of VAT is low. As men age, exceed their 20s, and reach middle age, their total and free testosterone decline, more fat is deposited in VAT stores, they often develop the "pot belly," and their risk for CHD increases [ 5 , 89 ]. The effects of testosterone on insulin resistance and metabolic syndrome risk factors are opposite in men and women [ 5 , 88 , 90 , 91 ]. Testosterone production often declines in women as they age, but VAT obesity in women is associated with elevated levels of total testosterone, free testosterone., and SHBG [ 92 ]. Hyperandrogenicity can also occur in polycystic ovary syndrome, where hyperinsulinemia can stimulate ovarian androgen production and suppress serum SHBG [ 88 , 93 ]. While weight loss in both sexes has been consistently shown to reverse the abnormalities in testosterone levels [ 94 - 97 ], a number of placebo controlled studies have consistently demonstrated that administering testosterone to obese men resulted in a significant reduction in VAT. This occurred without significantly altering amounts of total body fat or lean body mass [ 89 , 98 - 100 ]. However, the use of testosterone for VAT obesity is left open to debate [ 90 ]. 11-β-Hydroxydehydrogenase1 (11-β HSD1) Patients with type 2 diabetes and metabolic syndrome often appear Cushingoid, yet they invariably do not have elevated plasma cortisol [ 101 ]. Compared to SCAT, VAT has more glucocorticoid receptors [ 88 ]. The enzyme 11-β hydroxysteroid dehydrogenase type 1 (11-β HSD1) converts inactive cortisone to the active compound cortisol, and, if overexpressed, may cause increases in local cortisol concentrations [ 101 ]. Local production of active cortisol from inactive cortisone driven by 11-β-HSD-1 activity is very high in VAT and barely detectable in SCAT. Therefore it is likely that the VAT depot actively contributes to the production of high local concentrations of cortisol, which might not be reflected by plasma levels. These, in turn, might contribute to an increase in VAT accumulation [ 102 ]. 11-βHSD1 inhibition holds promise as a therapeutic target for VAT-associated metabolic syndrome [ 103 , 104 ]. VAT and impaired skeletal muscle oxidation The amount of fat deposited within skeletal muscle (intramyocellular lipid – IMCL) and the ability of muscle to oxidize fat are important determinants of weight gain,[ 105 ] weight regain following weight loss [ 106 ], and the development of insulin resistance syndrome [ 107 ]. IMCL and the VAT depot might not be independent from each other. Furthermore, the relationship between IMCL and insulin sensitivity is independent of percent total body fat and SCAT but not of VAT [ 108 ]. In individuals with type 2 diabetes, among the depots of regional and overall adiposity, VAT was the depot of adipose tissue that was most strongly related to skeletal muscle insulin resistance [ 109 ]. Colberg et al studied the fasting patterns of skeletal muscle fatty acid uptake and oxidation in healthy, lean and obese premenopausal women who had a cross-sectional VAT area over a range from 18 to 180 cm 2 and BMIs from 19 to 39 kg/m 2 [ 110 ]. The researchers found that insulin sensitivity as well as postabsorptive rates of FFA utilization or oxidation by muscle were diminished in relation to VAT. Women with increased VAT did not have lower plasma FFA levels or lower rates for appearance of FFA, yet they had an impaired or reduced uptake of plasma FFA by the skeletal muscle in the leg [ 111 ]. Together, this supports a role for VAT, IMCL lipid deposition, and perhaps impaired oxidation of nonadipose tissue lipid in insulin resistance and metabolic syndrome. VAT may influence central SCAT Mauriege et al found that adrenoreceptor sensitivity was increased in SCAT cells of individuals who have a higher VAT accumulation compared to those with a low VAT deposition [ 112 ]. SCAT adipocytes from women with visceral obesity exhibit higher lipolysis rates in vitro than those obtained from women with little VAT [ 113 ]. Mauriege et al also demonstrated that among men with high levels of VAT, SCAT adipocytes are more sensitive to β-adrenergic lipolysis which may further exacerbate an impaired insulin action, a potentially important factor in the etiology of metabolic syndrome associated with visceral obesity [ 112 ]. Moreover, an increased truncal SCAT mass and an increased amount of VAT mass can independently predict insulin resistance [ 114 ]. Together, these findings support that VAT may enhance central SCAT lipolysis and accelerate release of peripheral FFAs. The PPARs are important transcription factors that play an important role in the induction of adipose-specific genes, the proliferation and differentiation of adipocytes, and the development of mature adipose tissue. A number of transcription factors are involved, including PPARγs. Giusti et al suggest that in VAT, the expression of PPARγ2 is controlled by local transcription factors (RXRα, αSREBP1, and SREBP1c) promoting fat storage in adipocytes. Given that the fat storage capacity is limited in VAT, RXRα induces the expression of PPARγ2 in SCAT to increase its overall capacity [ 115 ]. These data also suggest that the signal to promote fat storage may occur in VAT and that other metabolic and hormonal factors are involved in the control and modulation of adipogenesis in visceral fat [ 115 ]. Perhaps the above can be explained as follows. SCAT cells may act as a buffer or sink for circulating FAs and TGs but once they reach their capacity they lose their protective benefits. Initially, VAT may influence SCAT to expand and act as a buffer. However, once the critical VAT threshold (CVATT) is achieved and metabolic syndrome has begun to develop, then VAT may influence central SCAT to become more VAT-like, i.e., more lipolytic and less sensitive to insulin's adipogenic or lipid storing effects. SCAT Greater preadipocyte differentiation and protection As discussed earlier, preadipocytes from SCAT depots have a greater capacity than VAT to differentiate into numerous, small, insulin-sensitive, adipocytes [ 70 , 71 ]. These lipid-storing cells act as a buffer or sink for circulating FAs and TGs, thereby preventing their deposition in non-adipose tissues, e.g., skeletal muscle, pancreas, and liver, where they could contribute to lipotoxicity, apoptosis, and insulin resistance [ 73 , 116 ]. Does SCAT replenish VAT? In defending the role of VAT accumulation in individuals with metabolic syndrome, we must postulate a high rate of lipid turnover, with high rates of lipolysis at certain times matched by high rates of lipid deposition at other times. Otherwise, as Frayn points out, the hyperlipolytic VAT would ultimately disappear [ 117 ]. He also suggests that if SCAT were to become insulin resistant, and therefore resistant to fat storage, then fat might tend to be deposited in VAT depots. Another possibility is that the usually larger SCAT depot has a greater potential to contribute to insulin resistance through release of FFA into the systemic circulation. However, this would not adequately explain the subset of individuals who demonstrate metabolic profiles consistent with insulin resistance but are in fact lean, healthy-appearing with normal BMIs, excess VAT, little SCAT, and are referred to as "metabolically obese, normal weight (MONW) [ 26 ]. As described above, perhaps once VAT expands and SCAT depots reach their capacity for storing FAs, then do SCAT adipocytes become insulin resistant, release FFAs, and contribute to systemic insulin resistance and metabolic syndrome. While some studies cast doubt on the portal theory and its implications for VAT's direct delivery of FFA to the liver [ 118 , 119 ], they leave open other mechanisms via which VAT could induce insulin resistance and other metabolic disturbances, e.g., by producing proinflammatory cytokines which could be directly delivered to the liver where they can potentially affect hepatic metabolism [ 117 ]. These will be discussed below. Peripheral fat mass may protect against atherosclerosis and metabolic syndrome If trunk fat is taken into account, accumulation of fat in the hips and legs is an independent predictor of lower cardiovascular and diabetes-related mortality, and it seems to protect against impaired glucose metabolism, especially in women [ 120 - 124 ]. In a study of 1,356 women ages 60–85, those with excessive peripheral fat had less atherosclerosis (determined by aortic calcification scores), and the quartile with both the highest amount of central fat and peripheral fat seemed to be partially protected by the high percentage of peripheral fat mass as reflected in a number of measured risk factors [ 121 ]. These findings corroborate similar findings by the same group who followed 316 postmenopausal women for 7.7 years and monitored progression of aortic calcifications [ 120 ]. In yet another study, Tanko et al demonstrated that peripheral fat mass (SCAT) in generally obese, post-menopausal women is associated with increased adiponectin and higher insulin sensitivity [ 125 ]. Together, these support protective roles for peripheral fat. In addition to fat trapping, these might include possible influences on adipokines, e.g., they might contribute to an increase in adiponectin, which could improve FA oxidation [ 126 ]. One must interpret these results with caution because the measuring technique of dual-energy X-ray absorptiometry (DXA) does not allow separate quantification of intermuscular and subcutaneous fat in the arms and legs as well as SCAT in the trunk [ 121 ]. While VAT is a major predictor of insulin sensitivity in overweight and lean individuals [ 114 , 127 ], others have found abdominal SCAT to contribute to insulin resistance independently of VAT [ 128 , 129 ]. An example of metabolically innocent obesity When there is an inability to store fat, due to lipodystrophy, the adipocytes' storage capacity is exceeded and lipids accumulate and cause lipotoxicity in liver, muscle, and other organ tissues [ 7 ]. A counterpart of lipodystrophy may be illustrated by patients with multiple symmetric lipomatosis (MSL), a condition characterized by regional excess of subcutaneous adipose tissue. These patients have higher adiponectin levels, a high degree of insulin sensitivity and glucose tolerance, very low lipid levels in liver and muscle cells, and markedly little VAT [ 130 ]. In this case, SCAT may be protective and beneficial. This may be analogous to thiazolidinedione action, which also promotes SCAT deposition while improving insulin sensitivity and glucose tolerance [ 74 , 75 ]. Estrogen Estrogen promotes the accumulation of peripheral gluteo-femoral SCAT, which may be protective [ 131 ]. The abundant presence of peripheral fat mass in generally obese women is associated with increased plasma adiponectin, and the loss of estrogen with menopause is associated with an increase in central fat [ 132 ]. This accounts for the progression in many overweight women after menopause from a predominantly pear-shape or "gynoid" habitus to the apple or "android" shape. Contrary to popular belief, menopause does not seem to independently cause a gain in total body weight; the increases in BMI that often accompany menopause are usually consistent with normal aging [ 133 ]. However, even without weight gain, body fat distribution changes; postmenopausal obese women tend to accumulate abdominal fat along with deterioration of risk factors, even if total body weight and BMI do not change during menopause transition. After menopause, when ovarian function declines, adipocytes become the primary source of endogenous estrogens [ 134 ], and compared to "gynoid" or pear-shaped women, those with central obesity (apple- or "android-" shaped) have lower plasma SHBG and higher estradiol [ 125 , 135 ]. This suggests regional differences in the enzymatic conversion of steroid hormones in VAT versus SCAT [ 125 , 136 - 138 ]. In ovarian hormone-deficient women, SCAT adipocyte size, lipoprotein lipase (LPL) activity, and basal lipolysis were not found to be significantly greater compared to regularly cycling premenopausal women. However, in the ovarian hormone-deficient women, omental (VAT) adipocyte size was significantly higher, and the omental/SCAT LPL activity ratio and VAT lipolysis were also significantly higher [ 139 ]. For a given amount of total body fat, men have been found to have about twice the amount of VAT than what is found in premenopausal women but this may change after menopause when VAT storage becomes predominant [ 140 , 141 ]. Along with an increase in VAT, a decline in estrogen is also associated with reduced lean body mass as well as other features of the metabolic syndrome including: dyslipidemia with elevation in Lp(a), triglycerides, and an increase in small, dense, LDL particles. Estrogen deficiency also may influence cardiac risk by its effects on the insulin action, the arterial wall, and fibrinolysis. Park et al showed that postmenopausal women lost less VAT compared with the premenopausal women during a weight reduction program (10.5 percent vs. 25.7 percent respectively) [ 142 ]. The reasons behind this are presently unclear. As mentioned above, in menopause, adipocytes are primary sources of endogenous estrogens in women [ 125 , 134 ], and estrogens are known inhibitors of IL-6 secretion [ 143 ]. It is worth noting that the relationship between BMI and serum IL-6 was observed only in postmenopausal women, and this relationship was lost among those women receiving hormone replacement [ 144 ]. Adipose tissue-derived estrogens in postmenopausal women would not be sufficient to reduce IL-6 in a similar way as endogenous estrogens do in premenopausal women [ 145 ]. Perhaps in premenopausal women, endogenous estrogen from the ovaries helps keep VAT volume relatively low and is thereby protective. Estrogen by itself seems to protect postmenopausal women receiving replacement therapy from VAT accumulation, and in women with type 2 diabetes, estrogen replacement may protect against the risk of cardiac events [ 146 , 147 ]. Compared to men of similar age, premenopausal women appear to be significantly protected from CHD. However, by age 70 the incidence of CHD is equal in men and women, suggesting that estrogen deficiency causes a rapid acceleration in CHD risk [ 133 ]. Yet, in elderly, postmenopausal women, Tanko et al showed that those women with higher amounts of central versus peripheral obesity had significantly higher levels of estradiol and lower adiponectin. This suggests that prolonged and increased exposure of SCAT cells to estradiol may eliminate the protective effect of SCAT by affecting SCAT's ability to release adiponectin thereby promoting the atherogenic effects of IL-6 [ 125 ]. Perhaps future research will help clarify whether central obesity has any implication for increased susceptibility to the adverse cardiovascular effects of hormone replacement therapy (HRT) in diabetic patients early after initiation of therapy [ 125 ]. Obesity, particularly visceral obesity, as well as insulin resistance and hyperinsulinemia are associated with breast cancer [ 148 ]. Insulin may increase estrogen action by increasing bioavailable estrogen due to a decrease in sex hormone-binding globulin, by influencing estrogen receptors, and by increasing aromatization of androgen to estrogen at the tissue level, a phenomenon which has been demonstrated in breast tissue. Estrogen upregulates the IGF-1 receptor and IGFBP-1 and -2 and may directly activate the IGF-1 receptor, thereby increasing insulin signaling [ 149 ]. Around 1900, most women died soon after menopause. The average lifespan of persons in the United States has since lengthened by greater than 30 years [ 150 ], which means that women, and men, too, are now spending 30 or more years with hormonal and physiological states that society and medicine has not had to deal with previously. These, combined with significant dietary and lifestyle changes since 1900, must be considered as critical contributing factors to the world's current epidemic of metabolic syndrome. Lipotoxicity Model Overnutrition, lipotoxicity, leptin, and the metabolic syndrome When one consumes too many calories, especially in the form of excessive carbohydrates, the liver converts excess glucose to fatty acids. First, glucose that is not oxidized or stored as glycogen is metabolized to acetyl CoA, which then enters the lipogenic pathway. Acetyl CoA is catalyzed to form malonyl CoA, which in turn inhibits carnitine palmitoyl transferase 1 (CPT-1, the enzyme responsible for fatty acid transport into the mitochondria) [ 42 ]. The net effect is that malonyl CoA (from excess carbohydrates, glucose, and insulin) reduces the oxidation of FAs [ 151 ]. This results in increased accumulation of intracellular fat in the form of long chain fatty acids and their derivatives, e.g., TGs and ceramide [ 28 , 29 ]. Cellular TG accumulation is not initially toxic and may actually be protective by diverting excess FAs from pathways that lead to cytotoxicity [ 152 ]. While glucose is being preferentially utilized, the FAs are metabolized by pathways other than their preferred β oxidation, leading to toxic products, e.g., ceramide, which may cause apoptosis and lipotoxicity [ 28 , 30 , 153 ]. The subsequent development of the cell's resistance to insulin-mediated glucose uptake, which prevents further influx of glucose, may be viewed as being protective in that it limits the amount of intracellular glucose to be preferentially metabolized over the β oxidation of intracellular FAs [ 29 , 37 , 154 ]. The cell can be insulin resistant with respect to glucose uptake and metabolism but remain sensitive to insulin's lipogenic effects and the de novo synthesis of fat. Overconsumption of calories, especially in the form of carbohydrates, also stimulates hyperinsulinemia that can then upregulate SREBP-1c and increase de novo lipogenesis [ 43 ]. Leptin protects against lipotoxicity Leptin The first adipocyte-specific hormone to be characterized, leptin is produced predominantly by SCAT adipocytes compared to VAT. Females produce leptin at about twice the rate in males [ 155 ], and leptin secretion increases with enlarged adipocyte cell size. Circulating leptin rises by 40 percent after acute overfeeding and more than three-fold after chronic overfeeding, whereas fasting is associated with decreased leptin levels [ 156 ]. Dietary carbohydrates may influence leptin action The increase in leptin concentration after meals is not simply a result of a caloric load, but is in response to a signal that is not present following a fat load without carbohydrate [ 157 ]. Leptin circulates in a free form and is also bound to a soluble leptin receptor – sOBR, which is positively associated with energy intake from carbohydrates and negatively associated with energy intake from dietary fat [ 158 ]. Excess caloric consumption and fat deposition results in newly synthesized FAs that are transported as VLDLs and stored as TG in adipocytes. Initially, these expanding adipocytes secrete leptin in proportion to their growing fat accumulation. Leptin also crosses the blood brain barrier, stimulates its receptor in the hypothalamus, and causes the release of neuropeptide-Y (NP-Y), which reduces feeding behavior [ 85 ]. This, in turn, suppresses appetite and stimulates thyroid function. Leptin affects peripheral tissues, and is a determinant of insulin sensitivity. The ensuing hyperleptinemia increases fat oxidation in skeletal muscle [ 159 - 161 ], and also keeps de novo lipogenesis in check by lowering the involved transcription factor, i.e., SREBP-1c mRNA (sterol regulatory element binding protein 1c mRNA) [ 43 ]. It promotes cholesterol ester synthesis in macrophages in a hyperglycemic environment, an important process in the formation of foam cells in atherosclerosis which may suggest a protective role of relative leptin resistance [ 162 ]. Leptin also possibly increases sympathetic nervous system (SNS) activity with subsequent decreased FFA oxidation and thermogenesis [ 163 ]. All of these effects of leptin tend to limit further weight gain. Leptin resistance As the process progresses, inefficient leptin action can lead to the opposite of leptin's protective effects, e.g., hyperphagia, decreased fat oxidation, increased tissue TG levels, insulin resistance, and overweight. Subsequently, plasma leptin levels rise. The majority of obese individuals with high leptin levels show a leptin insensitivity or "resistance [ 164 ]," which occurs at the leptin receptor level. In animal models, leptin-resistance and leptin-deficiency increases, and upregulates the hepatic expression of SREBP-1c mRNA, which may stimulate an increase in fat production via de novo lipogenesis. Together, all of these features suggest a state of "leptin resistance" which may ultimately lead to obesity and metabolic syndrome [ 29 , 165 ]. It is quite possible that hyperleptinemia in diet-induced obesity serves to protect nonadipose tissues (e.g. muscles, liver, pancreatic β cells, and myocardium) from the toxic effects resulting from the spillover of full adipose stores and subsequent ectopic deposition of FFAs. In defense of this paradigm, Unger points out that normally rats can tolerate a 60 percent fat diet because 96 percent of the surplus fat is stored in an enlarging adipose tissue mass, in which leptin gene expression increases proportionally [ 166 ]. However, when leptin is congenitally absent or inactive, or ineffective due to resistance, even on a normal or low-fat diet, excess dietary fat is deposited in nonadipose tissues. This causes dysfunction (lipotoxicity), and possible cell death (lipoaptosis) [ 29 ]. Acquired leptin resistance occurs in aging, obesity, Cushing's syndrome, and acquired lipodystrophy, a condition associated with protease inhibitor therapy of AIDS. Preliminary evidence suggests that patients with these conditions have increased ectopic fat, i.e., lipid deposition in non-adipose tissues [ 29 ]. Role of triglycerides in leptin resistance The relation between cerebrospinal fluid and serum levels of leptin in obese humans suggests that defective blood brain barrier (BBB) transport accounts for a great deal of leptin resistance in the CNS. Banks et al showed in mice that serum TGs directly inhibit the transport of leptin across the BBB and so could be a major cause of leptin resistance across the central nervous system (CNS). Thus they suggest that serum TGs are likely a major cause of the leptin resistance seen in both obesity and starvation [ 167 ]. This hypothesis explains why lowering TGs may be therapeutically useful in enhancing the effects of leptin. Implications for VAT in relative hypoleptinemia and metabolic syndrome Compared to VAT, SCAT is the predominant source of leptin [ 60 ], yet patients with VAT obesity may tend to have higher leptin levels than normal, lean individuals but lower than those with predominantly SCAT or subcutaneous obesity [ 29 ]. This suggests that the hyperleptinemia of predominantly VAT obesity is not high enough to overcome a leptin resistance due to the accumulation of ectopic fat in nonadipose tissues, which leads to lipotoxicity and ultimately the metabolic syndrome [ 29 ]. Lipodystrophies – A paradigm for the roles of fat depots and insufficient leptin action in metabolic syndrome A number of clinical states exhibit evidence of leptin insufficiency, either leptin deficiency or resistance, and they all have in common the metabolic syndrome. These include rare genetic diseases known as lipodystrophies, which are characterized by a redistribution of fat. Ironically, in the more severe cases, e.g., congenital generalized lipoatrophy, near-complete fat loss is associated with severe insulin resistance, fatty liver, and classic features of the metabolic syndrome. There is hyperleptinemia along with hyperphagia and a predominance of intra-muscular fat [ 168 ]. Dunnigan-type familial partial lipodystrophy is a rare autosomal dominant condition characterized by markedly reduced plasma leptin levels along with gradual loss of SCAT from the extremities, trunk, and gluteal region, commencing at the time of puberty, as well as hyperinsulinemia, glucose intolerance, dyslipidemia (high TGs with low HDL), and diabetes [ 169 , 170 ]. These individuals do maintain central obesity and VAT [ 169 ], which supports a relatively protective role for SCAT and implicates VAT as being more pathogenic. The aforementioned potential role of TGs in leptin resistance may have implications for patients with lipodystrophy and lipoatrophy who have little or no fat mass, and as a result, have very little or no leptin. They also have severe hypertriglyceridemia that is reversed by treatment with leptin [ 168 , 171 ]. The elevated plasma level of TGs in these patients is likely inducing leptin resistance that is preventing the leptin from inducing TGs to be used as an energy source. Thus the TGs in these patients are not oxidized, and they are unable to settle into fat stores that would normally act as a TG sink and prevent their diversion to non-adipose tissues where they contribute to lipotoxicity and insulin resistance. Fat depots can protect against lipotoxicity Fat provides leptin and adiponectin Transplantation of adipose tissue grafts in animal models of congenital lipoatrophy reverses the signs of the metabolic syndrome in a dose-dependent fashion [ 172 ]. Furthermore, leptin treatment in humans and animals with lipodystrophies also reverses fatty liver and insulin resistance. However, transplantation of ob/ob adipose tissue (which does not produce leptin) in lipodystrophic rats does not reverse diabetes [ 173 ] nor is it beneficial to inject leptin in obese humans with leptin resistance [ 18 ]. These support the notion that insufficient leptin action may be a cause of metabolic syndrome, and that adequate leptin derived from SCAT is protective. Like leptin, adiponectin secretion increases early on in obesity and plays a role in reducing the expression of lipogenic enzymes and increases FA oxidation in peripheral tissues thus limiting ectopic fat accumulation [ 174 ]. The fact that adiponectin is secreted initially by fat but levels are reduced as fat depots increase, may help resolve the paradox of both lipodystrophy and obesity both being insulin-resistant states [ 73 ]. Critical Visceral Adipose Tissue Threshold (CVATT) – Individual Variation The CVATT has tremendous individual variation; thus a relatively "thin" individual (with a normal BMI) and an excess of VAT for him, may be metabolically obese, normal weight (MONW) [ 26 ]. Meanwhile, another individual with a large "pot belly" may have a great capacity to store fat as SCAT with relatively little VAT or he may have a high threshold for VAT. This may explain the finding that while some individuals weighing even up to 200 kg do not show any signs of type 2 diabetes or dyslipidemia, while in others, diabetes or dyslipidemia either develop or deteriorate with an increase in body weight of only one kg [ 175 ] – perhaps just enough to exceed the CVATT. A number of studies have looked at a possible CVATT [ 176 - 182 ]. Using CT scans to measure VAT volume, Williams et al found that a value of above 110 cm 2 was associated with an increased risk of CHD in pre-and postmenopausal women [ 177 ]. Similarly, Despres and Lamarche observed a VAT cutoff of 100 cm 2 was associated with increased CHD risk in young adult men and premenopausal women (mostly of French Canadian descent) [ 179 ], and a cutoff range of 100–110 cm 2 has also been observed by others [ 176 , 181 ]. Other studies have suggested thresholds of > 130 cm 2 for metabolic deterioration [ 183 , 184 ]. De Nino et al found that insulin resistance did not appear until women were older than 60 years and had accumulated levels of VAT that approximated the levels seen in men, suggesting a possible threshold effect of VAT on insulin resistance [ 185 ]. As discussed below with MONW, genetic and ethnic factors play a role. For example, in nonobese and obese Japanese males and females, fat areas at the umbilicus as determined by CT had threshold values for metabolic syndrome with only > 100 cm 2 for men and > 90 cm 2 for women [ 186 ]. Brochu et al were unable to demonstrate that obese postmenopausal women who reduced their weight and attained a level of VAT below 110 cm 2 would show greater improvement in their metabolic profile compared to those who also lost weight but remained above the 110 cm 2 VAT threshold [ 178 ]. However, there were only 25 total subjects and the women had relatively normal metabolic profiles at baseline. Perhaps due to the relatively small number of subjects, only five lost less than 20 percent of their baseline VAT value. Thus it is unclear whether even smaller losses of VAT than those observed improved metabolic outcomes. The researchers did find larger losses of VAT and a greater improvement in insulin sensitivity in those who attained a VAT level < 110 cm 2 [ 178 ]. It should also be noted that in postmenopausal women, peripheral SCAT may be protective, even in the face of large amounts of VAT, and this needs to be accounted for [ 120 , 121 , 125 ]. While studying obese Japanese women, Tanaka et al recently validated the 100 cm 2 CVATT but their longitudinal data from both pre- and posttreatment suggest that these women should reduce their VAT area to <60 cm 2 through weight reduction to improve CHD risk factors [ 181 ]. Metabolically obese normal weight (MONW) VAT accumulation contributes to metabolic risk factors in nonobese individuals [ 187 , 188 ]. Ruderman et al have shown that normal weight individuals may also have insulin resistance and the disorders of the metabolic syndrome [ 26 ]. They designated such individuals as "metabolically obese normal weight – MONW [ 189 , 190 ]." MONW subjects (BMI < 25 kg/m2) have been characterized by an excess of VAT area (> 100 cm 2 by abdominal CT), insulin resistance, and hyperinsulinemia [ 24 - 26 ]. As pointed out earlier, the development of insulin resistance may limit further weight gain [ 34 , 38 - 41 , 191 ]. A rapid and early development of insulin resistance prior to significant weight gain would explain that a significant number of the normal-weight population have insulin resistance [ 26 ]. The prevalence of MONW could be as high as 13 – 18 percent [ 26 , 192 , 193 ]. MONW and low birthweight Both low birthweight (LBW) [ 194 ] and lowest weight at one year of age have been linked to, VAT accumulation[ 195 ] insulin resistance and cardiovascular risk factors in middle-aged and elderly individuals, many of whom could be classified as MONW with metabolic syndrome. By middle age, many LBW subjects have BMIs less than 24–26 kg/m2 and would be classified as MONW. While some data suggest that LBW babies have central adiposity in middle age, definitive measurements of VAT in these individuals are still lacking [ 26 ]. Ethnicity and MONW One should consider ethnic differences when attempting to identify MONW subjects. Lean appearing individuals, especially in certain ethnic groups such as the Japanese, may have significant amounts of VAT that surpass their CVATT but appear with what, for the general population, would be considered a normal BMI and waist circumference [ 196 ]. For example, nonobese Japanese (BMI<25) with increased VAT areas (100–110 cm 2 ) fulfill the criteria for MONW [ 25 , 181 ]. In another study, relatively lean Japanese patients with newly diagnosed type 2 diabetes had increased VAT. Through diet and without medication for three months, the amount of VAT in these patients became comparable to that in normal-weight control subjects. Therefore, a three-month dietary treatment regimen with small to moderate weight loss was very effective in decreasing excess VAT in this population [ 197 ]. This illustrates the importance of early recognition of an individual's approaching or exceeding his CVATT. Park et al were among the first to demonstrate that healthy, non-obese Asian American women may have higher amounts of VAT, and that normative values or standards for VAT derived from Caucasians may not be applicable to Asians [ 196 ]. On the other side of the spectrum, a 10-year prospective study studied increased BMI in Micronesian Nauruans (an ethnic group from the central Pacific Ocean with rapidly increase in prevalence of obesity) and Melanesian- and Indian-Fijians. Overall, there was little evidence to suggest that obesity was a risk factor for total or cardiovascular mortality in these populations [ 198 ]. Metabolically normal obese (MNO) McGarry found that one of his most obese patients in a series (BMI 32.8 kg/m2) was one of the most insulin-sensitive but had one of the lowest values for intramyocellular lipid (IMCL). Conversely, another subject, with a BMI of only 18.9 kg/m2, proved to be highly insulin-resistant but had a large amount of IMCL. This supports that insulin sensitivity appears to correlate more with where the fat is located rather than the total amount in the body [ 42 ]. This has implications for the phenomena of the metabolically obese normal weight (MONW) and the metabolically normal obese (MNO) individuals. Like some of the Micronesian Nauruans and Indian-Fijians above, there are individuals who are obese and who nevertheless are metabolically normal – "metabolically normal obese; MNO." Unlike their MONW counterparts, MNO individuals have very little VAT accumulation. They often share an onset of obesity early in childhood, normal VAT, lower TGs, and increased HDL. The actively competitive Japanese wrestlers maintain their gross obesity by consuming a 5,000 to 6,000 calorie diet. They are MNO, and their VAT is normal in amount, i.e., they have excessive amounts of SCAT [ 91 ]. On retirement, when they discontinue their rigorous training regimen, they markedly develop increased insulin resistance and metabolic syndrome. It is likely that that their VAT increases concomitantly [ 26 , 27 ] and exceeds their CVATT. Data from the European Group for the Study of Insulin Resistance (1146 hyperinslinemic/euglycemic clamp studies from 20 clinical centers in Europe) showed that in "simple" obesity, insulin resistance is not as prevalent as previously thought [ 199 ]. MNO could account for as much as 20 percent of the obese population [ 193 ]. In another study using HOMA to determine insulin resistance, Bonora et al showed that 11 percent of the entire group of overweight individuals fit the criteria of MNO [ 200 ]. Brochu et al extensively studied 43 sedentary, obese, postmenopausal women and found that 17 were MNO, while 26 had reduced insulin sensitivity (estimated by clamp) [ 201 ]. The two groups were similar in total body fat mass, SCAT amount, as well as waist circumference, and total daily energy expenditure. However, lean body mass was significantly greater in the metabolically abnormal subjects. Unlike SCAT, VAT measured by CT was inversely related to the insulin sensitivity and to a classification of MNO. In fact, despite similar levels of total body fatness, MNO individuals showed 49 percent less VAT as measured by CT. However, the level of VAT was still significant. Furthermore, using doubly labeled water and indirect calorimetry, Brochu et al were unable to demonstrate a meaningful difference between resting metabolic rate and daily physical energy expenditure between MNO and obese individuals at risk [ 201 ]. MNO and childhood obesity Several investigators have found that there has been a positive association between insulin sensitivity and duration of obesity, i.e., those who are obese since childhood are more likely to remain insulin sensitive. In one study 48 percent of the MNO women presented with a history of an earlier age-related onset of obesity (between 13 and 19 years of age) and less VAT compared with 29 percent of the metabolically abnormal obese [ 201 ]. Insulin sensitivity seems to be dependent upon adipose cell size; as adipocytes within tissue grow larger, they become more insulin resistant [ 202 ]. Normal-sized, more insulin-sensitive adipocytes have been associated with early onset of obesity [ 203 ]. Perhaps today we are beginning to see that with the marked increase in overfeeding and extent of obesity at younger ages, hypertrophy of fat cells may occur earlier and hence metabolic syndrome is now occurring with greater frequency in children. Puberty and VAT During puberty, a certain degree of insulin resistance is normal, and children who are more insulin resistant have decreased SCAT fat gain [ 204 ]. Early in the development of juvenile obesity, increased VAT, hyperinsulinemia, and insulin resistance are closely linked [ 205 ]. Adrenal androgens are elevated in obese children and have been associated with early pubertal development in these children[ 206 , 207 ] Sex differences in VAT begin to emerge during puberty, with boys having more VAT than girls. Some studies suggest that the rate of VAT accumulation can be slowed in children by using exercise interventions [ 208 , 209 ]. Fit and fat VAT is strongly associated with fitness even within individuals of the same weight. This is illustrated by the earlier mentioned example of the active Sumo wrestler in his prime who has relatively little VAT [ 91 ]. Regular exercise can selectively reduce VAT with minimal change in weight [ 210 - 212 ]. This could especially add to the frustration level of the middle-aged or post-menopausal woman who regularly exercises moderately without inducing measurable reduction in body weight or fatness. She may still benefit from reducing her VAT or attenuating the gain of VAT "normally" experienced by sedentary women as they age. It should be emphasized that the lower VAT level associated with increased fitness is modest but nonetheless clinically important. Reduced morbidity is likely explained by factors in addition to a reduced VAT, and VAT likely explains morbidity independent of fitness [ 213 ]. Sumo wrestlers tend to have most of their central adiposity stored subcutaneously (as SCAT), and, perhaps a shift toward more VAT accompanies their contracting metabolic syndrome upon their retirement – with premature death to follow [ 26 , 91 ]. This may also explain the body of work showing that overweight or "fat" individuals who are fit (according to cardiorespiratory testing on a treadmill) are at less risk for a cardiac event or developing type 2 diabetes than a "leaner" individual who is unfit [ 214 , 215 ]. Thus, the former could be considered "fit and fat." High levels of cardiorespiratory fitness (CRF) reduce CRP and the rate of cardiovascular morbidity and mortality, independent of obesity [ 216 ]. CRF is also associated with lower abdominal fat independent of BMI, and for a given BMI or waist circumference (WC), individuals with moderate CRF had lower levels of total fat mass and abdominal SCAT and VAT than individuals with low CRF for a given BMI or WC value [ 213 , 217 ]. Low CRF is an independent risk factor for mortality in healthy-appearing and diseased populations, and is associated with elevated CRP and reduced fasting glucose control in women with type 2 diabetes [ 218 ]. It is likely that compared to the fit and fat, the unfit and lean-appearing individual may have greater amounts of "hidden" VAT. Effects of exercise In obese patients, increasing physical activity can enhance fat oxidation, reduce IMCL and improve insulin sensitivity [ 219 ]. Exercise training may reduce waist size, independent of changes in BMI, and exercise without weight loss is effective in reducing VAT and preventing further increases in obesity [ 213 , 220 ]. Ross et al showed that either modality, caloric restriction alone or daily exercise without calorie restriction, is an effective strategy for reducing obesity in moderately obese men. Their findings also suggest that exercise without weight loss is a useful method for reducing VAT and preventing further increases in obesity [ 220 ]. Irwin et al studied 168 overweight, postmenopausal, previously sedentary women in a randomly controlled trial of exercise versus no exercise. While the body weight lost at 12 months among the exercisers was modest, the amount of intra-abdominal fat lost was considerable (8.5 g/ cm 2 ) and was dose-dependent. The women who exercised for approximately 200 min/wk lost 4.2 percent of total body fat and 6.9 percent of VAT without reducing their energy intake [ 212 ]. Exercise may counteract the abnormal metabolic profiles associated with abdominal obesity by reducing VAT along with other independent mechanisms. It promotes adaptive responses including those causing muscles to increase their use of lipid stores rather than relying primarily on carbohydrate reserves. Even a single bout of exercise can reduce triglyceride levels, increase HDL levels, reduce resting blood pressure, increase glucose tolerance, and reduce insulin resistance [ 221 ]. While evidence supports that CRF may be associated with a lower VAT, this is certainly not proven. However, study results suggest that individuals with moderate to high CRF levels have lower WC than men with low CRF independent of BMI [ 213 , 222 ]. Data support that the substantial reductions in health risk often associated with modest weight loss (<10 percent) may be mediated in part by a preferential reduction in VAT [ 48 , 216 , 217 , 220 , 223 ]. This is reinforced by the finding that reductions in VAT alone were related to improvements in glucose tolerance and insulin sensitivity [ 220 , 224 ]. Therefore, it would seem reasonable to infer that the combination of high CRF and low abdominal fat, especially VAT, would be associated with reductions in metabolic risk compared with those with the same BMI, but low CRF and high VAT [ 213 ]. Adding resistance training to aerobic exercise may add to an improvement in insulin sensitivity related to a loss of VAT and an increase in muscle density [ 225 , 226 ]. Surgical Interventions Shed Light on Pathophysiology Surgical removal of VAT in animals and humans dramatically improves insulin resistance and diabetes. In a Swedish, single-center, randomized and controlled pilot trial of 50 severely obese adults, Thorne et al compared 25 patients who underwent adjustable gastric banding (AGB) alone with AGB plus surgical removal of the total greater omentum. At two-year follow-up there were no statistical differences between groups with regard to weight loss, changes in WHR or sagittal diameter. However, the improvements in oral glucose tolerance insulin sensitivity and fasting plasma glucose and insulin were 2–3 times greater in omentectomized as compared to control subjects, which was statistically independent of the loss in BMI [ 52 ]. More recently, this has led to a study of another experimental procedure performed by surgeons at Boston's Beth Israel Deaconess Medical Center working in conjunction with Joslin Diabetes Center. Using a two-hour laparoscopic procedure that involves extracting strips of only the omentum through tiny incisions, this will be the first study to examine the possible health benefits of removing only the omentum [ 227 ]. Recently Klein et al. demonstrated that liposuction conferred no benefits with regard to metabolic profile [ 53 ]. Furthermore, Weber et al showed that after 3 months, animals that had lipectomy of > 50 percent of SCAT had more intra-abdominal VAT as percentage of total body fat, higher insulinemic index, a strong trend toward increased liver fat content, and markedly elevated serum TGs compared with animals that had undergone a sham operation and received either a high- or low-fat diet [ 228 ]. Together with the findings above, these support a pathologic role for VAT and a possible protective role for SCAT. Removing SCAT might actually increase risk as one removes a buffer or sink for peripheral TGs [ 228 ]. Environmental Considerations Organochlorines, adipose tissue, and energy balance Since our genes have not changed significantly in the past 10,000 years, the rise in obesity can be attributed to the environment, including what we are exposed to in the way of food as well as the level of physical activity. While the main focus has been on diet and activity, what may be overlooked is the tremendous increase in exposure to synthetic organic and inorganic chemicals, which can damage many of the mechanisms involved in weight control. Most of us have been exposed to organochlorines found in pesticides, dyes, solvents, etc., and we contain residues in our adipose tissue, where they are preferentially stored. Thus, the obese tend to have increased organochlorine concentrations compared to lean individuals [ 229 ]. During body weight loss, a decrease in fat mass results in lipid mobilization, and organochlorine concentrations increase both in plasma and remaining adipose tissue. Even after adjustment for weight loss, the related increase in organochlorine concentration has been correlated with decreases in triidothyronine (T3) concentration and resting metabolic rate [ 230 ]. This is also associated with a reduction in activity of the skeletal muscle oxidative enzymes that normally are involved in fat oxidation [ 231 ]. The net effect could prevent further weight gain and might even encourage weight regain beyond the initial baseline [ 232 ], which could contribute to VAT. Implications of Controlling Dietary Carbohydrates Reduced fat oxidation and carbohydrates Frisancho points out that an important contributing factor for obesity in modern as well as developing nations is a reduced fat oxidation and increased metabolism of carbohydrate. This has been brought about by a shift toward the body's preference toward oxidizing carbohydrate rather than fat – resulting in an increased deposition of body fat. In developing nations, obesity can co-exist with developmental undernutrition, which can result in obesity with short stature [ 233 ]. A solution to reducing the ectopic fat, as well as VAT, burden would be to enhance its oxidation in nonadipose tissues, e.g., liver, pancreas, and skeletal muscle. This will push the system toward below the CVATT and improve insulin sensitivity. In their review, Westman et al cite many studies that have consistently shown that low-carbohydrate/high-fat diets consumed for more than seven days induce powerful metabolic adaptations to enhance fat oxidation [ 37 ]. Such diets will reduce muscle glycogen content and carbohydrate oxidation, even in well-trained athletes who already demonstrate increased oxidation [ 37 , 154 ]. The authors' paradigm suggests that, under these conditions, insulin resistance could improve by reducing glucose appearance and cellular influx, resulting in a preferential fat oxidation and protection against lipotoxicity. In an elegant study, Bisschop et al support this by showing that high-fat, low-carbohydrate diets do not affect the action of insulin on total glucose disposal but decrease basal endogenous glucose production and improve insulin-stimulated nonoxidative glucose disposal [ 234 ]. Sharman et al demonstrated short term improvements of a ketogenic diet on lipids in normal weight men. These benefits occurred without total weight loss but there was evidence of a change in body composition toward more lean body mass [ 235 ]. One would also expect a reduction in VAT as he moves to the left or below his CVATT (See figure 1 ). Weight loss does not appear to be necessary to reduce mortality rates in overweight or obese men who increase their aerobic fitness or level of physical activity [ 224 ]. Similarly, in overweight, postmenopausal women, exercise may lead to improved metabolic profiles and VAT loss without total weight loss [ 212 ]. Dietary carbohydrate and VAT Optimizing macronutrients and food preparation can have beneficial effects in those with visceral obesity. A number of recent reviews support the metabolic benefits of controlling glycemic index (GI) [ 236 ] and glycemic load (GL) [ 237 ]. In a 12-month pilot study of teens, compared to a conventional diet, a lower GI diet led to greater total weight and fat loss without regain from months 6–12. While insulin resistance as measured by HOMA increased in the conventional group (possibly in part attributable to puberty), the lower GI group showed no change [ 238 ]. Recently, Silvestre et al showed that compared to an energy-restricted low-fat diet, a short-term, very low-carbohydrate diet was associated with greater weight and fat loss with an apparent preferential loss of central fat [ 239 ]. VAT cells have a two-fold higher glucose uptake rate compared with SCAT cells [ 87 ]. It may follow that reducing glucose exposure by reducing glycemic load may reduce the supply of glucose to the VAT depot and possibly impair its accumulation. Glucose raises insulin concentration, which can stimulate 11-β-HSD1, increase active cortisol in VAT, and enhance VAT accumulation [ 102 ]. Feeding rats a high-GI starch diet over five weeks resulted in higher VAT and larger adipocyte volume than did feeding low-GI starch ad libitum. Replacing this with a low-GI starch diet increased insulin -stimulated glucose oxidation, decreased glucose incorporation into total lipids and decreased VAT adipocyte diameter [ 240 , 241 ]. Together, these add to the evidence supporting the benefits of lowering GI to reduce and maintain lower volumes of VAT. Feeding rats a high sucrose diet increases both VAT and muscle insulin resistance [ 242 ]. Keno et al. demonstrated in rats that a high sucrose diet compared to a lab chow diet led to a significantly greater fat cell volume in VAT depots [ 243 ]. Although fat cell number did not change, the ratio of VAT weight to SCAT weight was also significantly increased in the rats fed a high sucrose diet, providing further evidence for controlling the dietary GI and GL. A number of studies have demonstrated an association between glycemic load (GL) and levels of CRP [ 244 , 245 ], which is a powerful predictor for diabetes and CHD, and is positively associated with both insulin resistance and the prevalence of the metabolic syndrome [ 246 ]. O'Brien et al showed that compared to a high carbohydrate diet, a low carbohydrate diet reduced SAA and CRP, both markers of inflammation and risk factors for metabolic syndrome [ 247 ]. Relative to fat (cream) and protein (casein), a glucose challenge elicits the greatest production of radical oxygen species (ROS) by polymorphonuclear and mononuclear white cells [ 248 , 249 ]. Chronic carbohydrate ingestion with a high GL diet can lead to hyperinsulinemia, as well as hypertrophy, functional dysregulation, and overresponsiveness of the pancreatic β cell and hepatic production of newly synthesized fatty acids via de novo lipogenesis [ 43 ]. A Johns Hopkins study examined intra-operative liver biopsies obtained from 74 consecutive morbidly obese patients undergoing bariatric surgery. Compared with patients with the lowest carbohydrate intake [ 246 ], a high-carbohydrate diet was associated with an odds ratio of 7.0 for liver inflammation. A high fat diet appeared to be protective, with those in the highest fat intake group having an OR of 0.17 [ 250 ]. This is consistent with the findings of others who found that dietary fat explained only two percent of the variance in general adiposity and dietary fat appears to play only a minor role in determining general adiposity and is not related to VAT when measured in cross-sectional studies [ 251 ]. Apparently, GL may be more significant in this regard. Compared to SCAT, VAT (both adipose and non-adipose cells within VAT) is associated more with PAI-1 – a powerful risk factor for CHD [ 58 , 252 ]. In patients with type 2 diabetes, a simple and modest lowering of the GI compared to an otherwise similar diet led to dramatic changes: a normalized PAI-1 activity (-54 percent, P < 0.001) as well as lowering of both blood glucose and plasma insulin concentrations by 30 percent, and a 29 percent decrease in LDL-C [ 253 ]. All subjects began with a BMI < 27, and there was only a slight but similar weight loss in both groups over the 24 days. The results support the potential benefit of lowering dietary GI in patients with metabolic syndrome, especially those with VAT and elevated PAI-1. This is also supported by the observation of hyperglycemia induces PAI-1 gene expression in adipose tissue of rats [ 254 ]. Esposito et al demonstrated in both diabetics and non-diabetics that after consuming a high carbohydrate high-fiber meal, IL-18 (a potent pro-inflammatory cytokine) concentrations increased [ 49 ]. Adiponectin concentrations decreased after the high-carbohydrate, low-fiber meal in diabetic patients. The fiber content of complex carbohydrates seemed to affect circulating IL-18 and adiponectin concentrations in response to the same carbohydrate load. The pizza that was made with whole flour and was rich in fiber was associated with reduce serum IL-18 concentrations and unchanged serum adiponectin concentrations. Meanwhile, the pizza prepared with refined flour and was low in fiber raised circulating IL-18 concentrations. Serum glucose and TG concentrations were not significantly different between the two types of pizza. The study did not completely resolve the mechanism by which the fiber content of meals influences IL-18 and adiponectin. However, it appears that while the GL of each pizza was the same, the GI of the whole wheat pizza would be much less and may be more beneficial. Recently, dietary TGs have been demonstrated to contribute to CNS leptin resistance by impairing the transport of leptin across the blood brain barrier where it would usually stimulate the release of neuropeptide-Y and reduce feeding behavior [ 167 ]. Reducing dietary carbohydrates lowers serum TGs, which theoretically should protect against this form of leptin resistance [ 167 ]. Dietary influences on leptin action Leptin may enhance fatty acid oxidation and protects against fat deposition and lipotoxicity. As mentioned earlier, normally, rats can tolerate a 60 percent fat diet because 96 percent of the surplus fat is stored in an enlarging adipose tissue mass, in which leptin gene expression increases proportionally [ 166 ]. However, when leptin is congenitally absent or inactive, or ineffective due to resistance, even on a normal or low-fat diet, unutilized dietary fat is deposited in nonadipose tissues, causing dysfunction (lipotoxicity), and possible cell death (lipoaptosis) [ 29 ]. Acute overfeeding can cause circulating leptin levels to rise by 40 percent and more than three-fold after chronic overfeeding, whereas fasting is associated with a decreased leptin levels. This may suggest that overfeeding leads to leptin resistance. Dietary carbohydrates may influence leptin action. Some investigators have suggested that the increase in plasma leptin concentration observed after meals is not simply a result of an energy load but is in response to a signal that is not present following a fat load without carbohydrate [ 157 ]. SCAT-derived leptin (which circulates in a free form and is bound to a soluble leptin receptor – sOB-R) plays a key role in regulating energy homeostasis and metabolism, sOB-R is positively associated with energy intake from carbohydrates and negatively associated with energy intake from dietary fat [ 158 ]. While this suggests that dietary fat and carbohydrates regulate free leptin levels, the implications of this are not yet completely clear. Stress There is an association with lifestyle, worry, cortisol levels, and abdominal girth. Those who were found to have the highest levels of chronic stress had the highest levels of cortisol and VAT [ 255 - 257 ]. This is supported by evidence that a number of medications, including prednisone, may cause an excess of cortisol and insulin resistance. Taken orally, cortisol raises blood pressure, and it has been shown to impair brachial artery blood flow in response to an acetylcholine challenge, i.e., an indicator of endothelial dysfunction [ 88 , 255 , 257 - 262 ]. Even brief episodes of mental stress, such as those encountered in daily life, may cause transient endothelial dysfunction even in young, healthy individuals ([ 263 , 264 ]. In turn, subsequent cytokine release may increase anxiety and have negative effects on emotional and memory functions [ 265 ]. Psychological stress has also been demonstrated to acutely reduce clearance of triglycerides [ 266 ], which could contribute to CNS leptin resistance [ 167 ]. There are many other ways in which psychological stress might increase the likelihood of developing metabolic syndrome and type 2 diabetes, for example, chronic psychological stress may also be related to central activation of the HPA (hypothalamo-pituitary-adrenal) axis and the sympathetic nervous system (SNS) [ 267 ]. Psychological stress also induces IL-6, TNFα, and other cytokine secretions from macrophages [ 267 - 271 ]. Repeated stress with the repeated induction of corticosteroids can damage the hippocampus, which is involved in the downregulation of corticosteroid production by corticosteroid feedback. Impairment of this feedback mechanism can lead to persisting elevated circulating cortisol levels [ 267 ], which might play a role in inducing VAT accumulation. Stress decreases splanchnic blood flow, impairs the integrity of the GI tract, increases intestinal permeability, and results in increased absorption of lipopolysaccharide endotoxin (LPS) from the gut (the greatest source of LPS). Elevated portal bloodstream LPS levels stimulate Kupffer cell receptors and cytokine release and possibly other immune-challenging activators, e.g., AGEs in food [ 271 ]. Stress and dietary carbohydrates Dietary carbohydrate has been known to stimulate SNS activity though a number of studies have emphasized the role of insulin. Recent studies in rats have demonstrated that adding glucose to the basic diet increased SNS activity in peripheral tissues and increased GLUT 4 activity in interscapular brown adipose tissue and retroperitoneal fat (but not in epididymal fat) [ 272 ]. Overfeeding results in high insulin levels. In the presence of glucose, insulin acts on the brain to increase the SNS tone, which, in turn enhances thermogenesis and dissipation of excess calories [ 163 ]. There is a close relationship between postprandial insulinemia, SNS activation, and adipose tissue blood flow (ATBF). ATBF increases in response to stress states such as exercise or mental stress, and also in response to nutrient intake [ 273 ]. High insulin levels and increased SNS tone are useful for the maintenance of caloric balance, but in the long term they are conducive to CHD, hypertension, sudden death, and obesity as the SNS receptors become down regulated [ 163 ]. Chronic stress leads to elevated cortisol levels, which may lead to accumulation of VAT and metabolic syndrome [ 274 ]. Stress-induced increased levels of glucocorticoids can also have a major effect on food intake [ 275 ]. A subset of stressed or depressed humans may overeat, especially comfort food (e.g., sugar and fat), in an attempt to reduce anxiety and activity in the chronic stress-response network. This is supported by the finding that these people have decreased cerebrospinal corticosteroid releasing factor, catecholamine concentrations, and HPA activity. While comfort foods may calm them down in the short term, they may lead to abdominal obesity if this becomes a long term "solution." The chronic elevation of systemic glucocorticoids may contribute to VAT deposition. By itself, being obese may be a stressful stimulus to overeating. A weight loss program can be stressful, which can sabotage its success by eliciting the release of stress hormones, which, in turn can make a person crave high energy foods [ 275 ]. Feeding rats a long-term high-sucrose diet along with supplemental dexamethasone has been shown to increase fat depots and induce liver steatosis [ 276 ]. In addition to dietary intervention, stress management may improve one's cognitive, behavioral, and physiologic responses to stress, including glycemia [ 277 ]. Summary The role of visceral adipose tissue (VAT) obesity in metabolic syndrome is critical and complex. (See figure 3 ). The paradigm of an individual critical VAT threshold (CVATT) has been presented along with a review of potential mechanisms and contributing factors. This includes the potential role of dietary carbohydrates in VAT obesity. As this area continues to evolve, perhaps the reviewed material and proposed concepts may have relevance to clinical assessment, prevention, and treatment of metabolic syndrome. Figure 3 This diagram provides an overview of how behavioral and environmental factors can lead to VAT obesity and, ultimately, metabolic syndrome and related disorders. Declaration of Competing Interests The author(s) declare that they have no competing interests.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC535537.xml

544870

Redesigned and chemically-modified hammerhead ribozymes with improved activity and serum stability

Background Hammerhead ribozymes are RNA-based molecules which bind and cleave other RNAs specifically. As such they have potential as laboratory reagents, diagnostics and therapeutics. Despite having been extensively studied for 15 years or so, their wide application is hampered by their instability in biological media, and by the poor translation of cleavage studies on short substrates to long RNA molecules. This work describes a systematic study aimed at addressing these two issues. Results A series of hammerhead ribozyme derivatives, varying in their hybridising arm length and size of helix II, were tested in vitro for cleavage of RNA derived from the carbamoyl phosphate synthetase II gene of Plasmodium falciparum . Against a 550-nt transcript the most efficient (t 1/2 = 26 seconds) was a miniribozyme with helix II reduced to a single G-C base pair and with twelve nucleotides in each hybridising arm. Miniribozymes of this general design were targeted to three further sites, and they demonstrated exceptional cleavage activity. A series of chemically modified derivatives was prepared and examined for cleavage activity and stability in human serum. One derivative showed a 10 3 -fold increase in serum stability and a doubling in cleavage efficiency compared to the unmodified miniribozyme. A second was almost 10 4 -fold more stable and only 7-fold less active than the unmodified parent. Conclusion Hammerhead ribozyme derivatives in which helix II is reduced to a single G-C base pair cleave long RNA substrates very efficiently in vitro . Using commonly available phosphoramidites and reagents, two patterns of nucleotide substitution in this derivative were identified which conferred both good cleavage activity against long RNA targets and good stability in human serum.

Background Hammerhead ribozymes were discovered as self-cleaving motifs in a number of small, circular, pathogenic RNAs in plants [ 1 - 3 ]. Uhlenbeck [ 4 ] showed that the ribozyme was able to act in a bimolecular fashion as a true enzyme, ie each ribozyme was able to cleave multiple substrates. Haseloff and Gerlach [ 5 ] divided the hammerhead into a form in which the majority of the conserved nucleotides were located on the enzyme strand, with the only sequence requirements for the substrate being UH (H = A, U or C) [ 6 - 8 ]. Since 1988 this configuration, as shown in Figure 1 , has been the paradigm for hammerhead ribozyme design. Hammerhead ribozymes are sequence-specific RNA cleaving agents with the potential to control the expression of genes by eliminating specific RNAs. This can be achieved by expressing the ribozyme within the target cell or by delivering it to the cell as a preformed entity. One of the difficulties associated with delivering preformed ribozymes is their instability in vivo , since RNA is degraded very rapidly by ribonucleases present in cells and extracellular fluids. Significant segments of ribonucleotides in the hammerhead ribozyme can be replaced with more nuclease-resistant analogues like DNA, phosphorothioate linkages, or 2' O-methyl analogues; however, within the conserved core of the hammerhead, the majority of ribonucleotides are sensitive to modification. Yang et al [ 9 ] demonstrated that predominantly DNA ribozyme analogues with at least 4 ribonucleotides (G 5 , G 8 , A 9 and A 15.1 or G 15.2 numbered according to [ 10 ]) displayed measurable cleavage activity (albeit reduced 5000-fold). Phosphorothioate modification of DNA hybridising arms and three of the conserved pyrimidines (C 3 , U 4 & U 7 ) resulted in significant increase in stability in human serum with a 6-fold loss in cleavage activity [ 11 ]. In the context of 2'-O-methyl substituted ribozyme analogues, at least 5 unmodified ribonucleotides (G 5 , G 8 , A 9 , A 15.1 and G 15.2 ) were required for activity [ 9 ]. Paolella et al [ 12 ] identified a minimum set of 6 ribonucleotides, U 4 , G 5 , A 6 , G 8 , G 12 and A 15.1 , in the conserved domain in which substitution with 2' O-allyl ribonucleotides inhibited activity. Eckstein's group [ 13 , 14 ] showed that good activity and stability in foetal calf serum could be achieved with 2'-amino-2'-deoxyuridines at U 4 and U 7 , and 2'-fluoro-2'-deoxycytidine at C 3 . Hammerheads in which helix II was shortened to only two base pairs, and all the pyrimidines were 2'-fluoro or 2'-methoxyethoxy derivatives except for U 4 and U 7 which were 2'-amino-2'-deoxy, were only 2–3 fold less active than the unmodified parent hammerhead [ 15 ], but showed a 10 4 -fold increase in nuclease stability. A study of the effects of various modified nucleotides on stability and activity of the 2'-O-methyl-pyrimidine modified hammerhead found a number of modifications, including 2'-amino-2'-deoxyuridine, 2'-O-methyl-uridine, and 2'-C-allyl-uridine at positions U 4 and U 7 , supported good rates of cleavage [ 16 ]. These modifications result in a greater than 10 3 -fold increase in stability in human serum, while the addition of an inverted thymidine at the 3' end of the oligonucleotide (a 3'-3' linkage) further improved the stability by two orders of magnitude. Figure 1 Schematic representation of Mrz-12/12-A bound to substrate S-30. Helices I and III are formed between the ribozyme and substrate. The standard hammerhead (eg Rz-12/12) has a 4 base pair helix II in place of the single g-c pair in the miniribozyme. The minizyme (Mz-12/12) has no helix II, but has a loop sequence gtttt connecting bases A 9 and G 12 . Upper-case letters represent ribonucleotides, and lower-case letters represent deoxyribonucleotides. Nucleotides which have been further modified in this study are shown in blue. Our laboratory is interested in the relationship between hammerhead design and reactivity. We have described a number of ribozyme derivatives that appear to have promise as RNA cleavage agents. Minizymes (Mz) possess a non-base-pairing tetranucleotide linker in place of helix II [ 17 ]. In general, such minizymes are less active than standard hammerheads, although in some instances they show comparable cleavage rate constants [ 18 , 19 ]. Miniribozymes (Mrz) have a single G 10.1 –C 11.1 base pair joined by a flexible linker in place of helix-loop II [ 19 ]. Asymmetric hammerheads are those in which the 5' hybridising arm is restricted to around 5 or 6 nucleotides; this modification eliminates the decrease in cleavage rate that occurs with standard hammerheads when the length of helix I, formed upon binding the substrate, increases to greater than about 6 base pairs [ 20 , 21 ]. The purpose of this study was to investigate the ability of these various derivatives to cleave an RNA molecule (sequence derived from the mRNA of the cpsII gene of the Malaria-causing organism Plasmodium falciparum [ 22 ]) in the context of two substrates, a 30-mer and a 550-mer. Having optimised the cleavage activity of the ribozyme, we planned to chemically modify the nuclease sensitive RNA nucleotides, using readily available protected nucleoside phosphoramidites, to extend the life of the ribozyme in the presence of human serum. Results Ribozyme design The primary target site for cleavage in this study was a previously identified site [ 23 ] in the cps II mRNA [ 22 ] of Plasmodium falciparum . The chosen target site, centred at position 3733 in the nucleotide sequence (Genbank reference L32150), has the local sequence 5' UAA CUU AUC AAG GUC* AAG AAC AUG AUG UUC 3', where the site of cleavage is denoted by the asterisk. A number of ribozyme designs were tested for their ability to cleave this RNA sequence either as a short (30-mer) oligonucleotide or in a transcribed RNA segment (550 nt). These designs included standard hammerhead ribozymes (Rz) which are defined as those with a helix II consisting of four base pairs closed at the end with a four-nucleotide loop, minizymes (Mz) [ 17 ] which lack helix II and instead the two segments of conserved nucleotides are linked between A9 and G12 with a non-base-paired linker (in this case consisting of 5 nucleotides), and finally miniribozymes (Mrz) [ 19 ] which have a single G-C base pair replacing helix II and in this instance a linker sequence consisting of four deoxythymidines. In this communication the core of each ribozyme is flanked by hybridising arms composed of DNA of various lengths, where the length is given in the ribozyme's name (e.g. Rz-6/12 has 6 nt in its 5' hybridising arm and 12 nt in its 3' arm). Hammerhead-ribozyme derivatives of these designs were tested for their ability to cleave the 30-mer substrate at pH 7.6, 37°C and 10 mM MgCl 2 under pseudo first-order conditions with an excess of ribozyme. Ribozyme concentrations ranged from 50 nM to 10 μM. Cleavage data fitted well to single exponential curves to yield observed rate constants which were plotted against the ribozyme concentration in each experiment to determine the apparent dissociation constant ("K d ") and the maximum cleavage rate constant (k max ) for each ribozyme-substrate pair (Table 1 ). In terms of catalytic efficiency, the most efficient ribozyme was the standard hammerhead Rz-12/12. Its efficiency is due to a very low "K d " of 7 nM, despite displaying a k max some 5-fold less than Mrz-12/12. The highest k max was displayed by Rz-6/12; however it had a "K d " about 60-fold greater than Rz-12/12. Table 1 Kinetic parameters for RNA/DNA unmodified hammerhead ribozyme derivatives. Ribozyme S-30 S-550 k max (min -1 ) "K d " nM k max /"K d " (min -1 μM -1 ) k max (min -1 ) "K d " nM k max /"K d " (min -1 μM -1 ) Rz-12/12 0.8 ± 0.1 7 ± 12 114 ± 200 0.7 ± 0.1 2600 ± 900 0.3 ± .2 Mz-12/12 0.22 ± 0.08 31 ± 6 7 ± 4 0.12 ± 0.01 240 ± 50 0.5 ± .2 Mrz-8/8 0.56 ± 0.06 210 ± 80 2.7 ± 1.3 0.005 ± 0.001 2200 ± 500 0.002 ± .001 Mrz-12/12 4.2 ± 0.3 75 ± 30 56 ± 26 1.6 ± 0.1 120 ± 50 13 ± 6 Rz-6/12 9 ± 1 400 ± 200 22 ± 13 0.6 ± 0.1 2800 ± 900 0.2 ± .1 Cleavage conditions; 37°C, pH 7.6, 10 mM MgCl 2 , ribozyme is in large excess over RNA substrates S-30 (30 nt) and S-550 (550 nt). The abilities of all the ribozymes to cleave the same target in the context of transcribed RNA (550 nt) was also examined. Mrz-12/12, by virtue of only a modest decrease in k max , and marginal increase in "K d ", was by far the most efficient of all the designs tested. In contrast Mrz-8/8, Rz-12/12 and Rz-6/12 displayed "K d "'s between 2 and 3 mM, which, in the case of Rz-12/12, is an increase of about 400-fold. Interestingly the k max values displayed by Rz-12/12 and Rz-6/12 were within experimental error, and were the same as displayed by Rz-12/12 for cleavage of the short substrate. Cleavage of other targets by the miniribozyme We examined whether the effectiveness of the miniribozyme design was limited to this target site. Miniribozymes, with long (>10 nucleotides) hybridising arms were designed to cleave RNAs of interest to other projects in the laboratory. Tet Mrz was a 53-mer oligonucleotide with conserved bases of RNA and hybridising arms and stem loop II composed of DNA. This Mrz, with hybridising arms of 18 and 19 nucleotides, was targeted to cleave the GUC triplet at position 60 of the Tetrahymena IVS ribozyme [ 24 ]. It cleaved the 388-nt transcript, L-21 ScaI, with a rate constant of 4.0 ± 0.2 min -1 (t 1/2 = 10 seconds), to about 70%, at pH 7.6, 37°C and 10 mM MgCl 2 . Another miniribozyme with 14-mer arms (HC Mrz), targeting a segment of Hepatitis C polyprotein mRNA, was tested against a synthetic 29-mer substrate. Under our standard conditions about 70% of the substrate was cleaved, and the cleavage rate (>5 min -1 ) was too fast to be measured reliably. In contrast, the standard ribozyme HC Rz cleaved 78% of the same substrate with a rate constant of only 0.2 ± 0.02 min -1 . Finally, PDGF MRz, with hybridising arms each of 10 ribonucleotides, was tested against both 25-mer synthetic and 707-nt transcribed substrates. Under standard conditions, the 25-mer was ~ 80% cleaved with a rate constant > 5 min -1 . Reducing the magnesium ion concentration to 1 mM yielded a rate constant of 2.8 min -1 and 63% cleavage for the 25-mer substrate, and 1.6 min -1 and 45% cleavage for the 707-nt transcript. Identification of nuclease susceptible sites in human serum Mrz-12/12 was the most efficient cleaver of the 550-nt cpsII RNA transcript and was used as the platform to test the effect of chemical modification on cleavage activity and nuclease stability. Firstly, the stability of unprotected Mrz-12/12 in RPMI + 10% human serum was determined using 32 P 5'-end labelled ribozyme at 37°C (Figure 2 ). Approximately 90% of the miniribozyme is degraded in 10 seconds, and no full-length material is observed after 10 minutes. Initially there are three main sites of cleavage, at U 4 , U 7 and C 15.2 . After 10 minutes, nearly all the end-labelled material co-migrates with the band generated by alkaline digestion at U 4 . That initial, major product has a half-life of about two hours under these conditions, as it is slowly converted to a product one nucleotide shorter, ie its 3' end corresponds to C 3 in the original miniribozyme. Figure 2 Degradation of 5' end-labelled Mrz-12/12 A in RPMI + 10% human serum at 37°C. Time of incubation in indicated above each lane. OH - indicates an alkaline digest of the same material. FL indicates the position of the full-length miniribozyme. The position of the fragments terminating at each of the ribonucleotides is indicated by the letters adjacent to the alkaline digest. Nuclease resistant modifications Commercially available modified phosphoramidites were used to generate hammerhead derivatives which were expected to be protected from nuclease degradation. The effect of the various modifications on the cleavage ability are given in Table 2 . In these experiments the concentration of the miniribozyme was fixed at 1 μM and the substrate S-30 at 5 nM. The nucleotide most sensitive to chemical modification was U 4 . All the 2' modifications tested (amino, deoxythymidine, deoxyuridine, and O-methyl, shown as D, F, G and H, respectively, in Table 2 ) diminished activity by more than 10-fold. Only the presence of a phosphorothioate linkage between U 4 and G 5 preserved the activity of the unmodified ribozyme. The ready availability of 2'-O-methyl phosphoramidites, coupled with the previous demonstrations [ 9 , 16 ] of tolerance to that modification at C 3 , U 7 and C 15.2 , lead us to synthesise Mrz-J, which we expected to have reasonable activity and nuclease stability. Its cleavage rate constant was actually twice that observed for the unmodified ribozyme, and its stability in serum was increased about 10 3 -fold. The kinetics of degradation in serum (Figure 3 ) were not straight-forward, displaying a rapid initial decay of about 25% of the starting material, followed by an approximately first-order decay with a half-life around 30 minutes. This second phase accounted for approximately 50% in the total starting material, ie after about 4 hours approximately 25% of the full-length material remained intact and thereafter decayed only very slowly. There was a single major product, corresponding to cleavage at U 4 , observed over the six hours of the experiment. Table 2 Cleavage Rate constants for cleavage of S-30 by Chemically Modified Mrz-12/12. Miniribozyme (Mrz-12/12) C 3 U 4 U 7 C 15.2 3' end Cleavage rate constant (min -1 ) Relative Stability A - - - - - 4.2 ± 0.3 1 B F NH 2 NH 2 F - 0.30 ± 0.05 - C - NH 2 NH 2 - - 0.18 ± 0.01 - D - NH 2 - - - 0.30 ± 0.03 - E - - NH 2 - - 7.7 ± 0.8 - F - dT - - - 0.014 ± 0.005 - G - dU - - - 0.024 ± 0.007 - H - OMe - - - 0.01 ± 0.007 - I - ps - - - 3.9 ± 0.06 - J OMe ps OMe OMe psps 7.3 ± 0.7 1400 K OMe NH 2 OMe OMe - 0.6 ± 0.1 8600 Rz-12/12-L - NH 2 - - - 0.05 ± 0.009 - Cleavage conditions; 37°C, pH 7.6, 10 mM MgCl 2. [Rz] = 1 μM, [S30] = 5 nM. (- = unmodified, ie 2'OH). F = 2'-fluoro, NH 2 = 2'-amino, dT = 2'-deoxythymidine, dU = 2'-deoxyuridine, OMe = 2'-O-methyl, ps = 3' phosphorothioate linkage. Stability is defined as the time required in 10% human serum to degrade 75% of full-length ribozyme, relative to unmodified miniribozyme A. Figure 3 Degradation of Mrz-12/12 J in RPMI + 10% Human Serum. Experimental conditions are as described in Figure 2. The single degradation site in Mrz-J suggested that a more robust modification at this position would have a significant effect on its lifetime in serum. Mrz-12/12 K was synthesised with a 2'-amino modification at the U 4 position and with a 2'-O-methyl modification at each of the three other conserved pyrimidines. Protection of the 3' end was omitted from Mrz-12/12 K because it did not appear to contribute significantly to the stability observed in Mrz-12/12 J. As expected, the cleavage activity of Mrz-12/12 K was significantly reduced compared to Mrz-12/12 J (Table 2 ), but stability in serum was greatly improved with only minor losses apparent after 5 hours incubation at 37°C (Figure 4 ). Even after 24 hours in 10% serum, approximately 25% of the radioactivity co-migrated with the full-length material, representing an almost 10 4 -fold increase in stability. Even in the absence of 3' terminal protection there was no 3' exonuclease activity apparent. The amount of 32 P label appearing in the gel lanes remained relatively constant throughout the experiment implying a lack of significant phosphatase activity in the serum. The main sites for degradation were the remaining unprotected (purine) ribonucleotides. Figure 4 Degradation of Mrz-12/12 K in RPMI + 10% Human Serum, and by alkaline digest. Experimental conditions are as described in Figure 2. The modifications to the conserved nucleotides in this study were all made in the context of a miniribozyme. Compared to a standard hammerhead, the catalytic domain is expected to be more conformationally flexible, and therefore it should not be assumed that all the changes described here can be directly applied to standard hammerheads with a helix II of four base-pairs. However, as for the miniribozyme, a standard ribozyme containing a 2'-amino modification at U 4 , was about 15-fold less efficient at cleaving S-30, (k obs = 0.05 min -1 , Rz-12/12-L, Table 2 ), compared to the unmodified ribozyme (Rz-12/12). Discussion Work in this laboratory [ 20 , 21 ] and elsewhere [ 25 ] has demonstrated that the cleavage kinetics of any conventional hammerhead ribozyme are significantly inhibited when the length of helix I exceeds about 6 bp. This has been ascribed to an interaction between helices I and II which stabilises an inactive conformation [ 20 ]. The angle between helices I and II changes with metal ion concentration [ 26 ], and we postulated that a similar change was required for the transformation between more and less active conformations of the ribozyme [ 20 ]. This general model is supported by more recent observations using a variety of techniques [ 27 - 31 ] which conclude that the dominant ground-state conformation of the hammerhead is inactive and is in equilibrium with the active form. The results of the present study are in accord with these conclusions. It is commonly observed, for conventional hammerhead ribozymes, that cleavage efficiencies for long transcripts are about 2 orders of magnitude lower than for short substrates [ 32 ]. This has been observed here also for ribozyme derivatives of both the conventional design and of the short-armed (8-nt) miniribozyme (Table 1 ). In these cases the cause appears to reside largely in the apparent dissociation constant "K d ". In contrast, the longer-armed miniribozyme (Mrz-12/12) shows excellent cleavage kinetics against both short and long substrates. It appears that the more flexible miniribozyme is better suited to binding to the target in the context of a long RNA. These data can be interpreted according to a simple model in which ribozymes possessing a stable helix II form a more rigid three-dimensional structure [ 33 , 34 ] which does not bind strongly to the long, folded substrate. When binding is achieved at very high ribozyme concentrations, the ribozyme-substrate complex is sterically hindered to such an extent that the complex is locked into a poorly active conformation. Hence Rz-6/12 and Rz-12/12 display similar maximal rate constants. In contrast, the Mrz lacking helix II is more flexible and readily adapts to binding the folded substrate with relatively minor effects on k obs and "K d ". This observation is not specific to target or cleavage triplet, since four unrelated targets, including three long transcripts, one with an AUC cleavage triplet, were cleaved with rate constants much higher than typically reported for cleavage even of short substrates. It is worth noting the magnitude of the observed cleavage rate constants; under our standard conditions these miniribozymes cleave their targets with half-lives in the range of < 5 to 25 seconds. The unmodified Mrz-12/12 was very unstable in 10% human serum. Degradation occurred by endonucleolytic cleavage at the 3' side of pyrimidine nucleotides. Since the hybridising arms and helix-loop II are composed of DNA, the four remaining ribo-pyrimidine nucleotides in the conserved domain were the critical residues for stabilisation. Modification of all four ribopyrimidines with 2'-aminouridine and 2'-fluorocytidine, (Mrz-12/12 B), resulted in a more than 10-fold decrease in maximum cleavage rate constant. This contrasts with some previous results where a U 4 , U 7 -amino derivatised ribozyme with a standard helix II was only marginally diminished in activity [ 14 ], and a U 4 , U 7 -amino derivatised ribozyme [ 16 ], in which all the remaining pyrimidines were modified with 2'-O-methyl groups, was only two-fold less active than the parent ribozyme. The study by Heidenreich et al [ 14 ] was confounded by the fact that the cleavage kinetics were measured against a long transcript and are relatively slow compared with rates typically observed for short substrates, and therefore it seems likely that access to the transcript, rather than chemistry of cleavage, may have been rate determining in that case. In a later report [ 15 ], a modification pattern identical to miniribozyme B ( ie C 3 , C 15.2 -fluoro and U 4 , U 7 -amino) but in a ribozyme with a standard helix II resulted in a more than 10-fold loss in turnover number (k cat ). The series of singly and doubly substituted miniribozymes (C-E) showed clearly that a 2'-aminouridine at position 4 was solely responsible for the loss in activity. This is consistent with the result of Beigelman et al [ 16 ], but, in contrast to that result, the cleavage activity of the miniribozyme was not rescued by the addition of an amino group at U 7 . A phosphorothioate linkage between U 4 and G 5 (Mrz-12/12 I) was the only modification tested here that did not suppress cleavage activity. This is in accord with interference studies [ 35 , 36 ] identifying U 4 as insensitive to phosphorothioate modification. The results for dU and 2'O-methyl substitutions contrast with Beigelman et al [ 16 ] where, in a background of 2'O-methyl-pyrimidines, the effect of those substitutions at U 4 were relatively minor. In serum, Mrz-12/12 J was degraded by cleavage at the U 4 phosphorothioate, where an initial, quite rapid degradation was followed by a slower step which is not complete at 6 hours. Chemically synthesised phosphorothioate linkages are a mixture of two enantiomers and it is likely that the multiphase kinetics observed were due in part to the different susceptibility of the two isomers (Rp and Sp) to the nucleases present in the serum [ 37 , 38 ]. Mrz12/12 K has 2'-amino modification at U 4 , is very stable in serum, and is only 7-fold less active than the unmodified parent. The products of nuclease cleavage of Mrz-12/12 K indicate that degradation occurred by cleavage at the unprotected ribopurine sites. This derivative was not protected at the 3' terminus, but unlike earlier work [ 11 , 16 , 38 ], there was no evidence in this experiment for 3'-DNA exonuclease activity. In recent times RNAi has surplanted catalytic RNA as the method of choice for suppression of gene expression in many eukaryotic cells [ 39 ]. Results published to date suggest that it may be effective in many settings, including possibly as a therapeutic. However, the mechanistic details of RNAi activity are quite complex and are still being unravelled, and it seems likely that there will be cell types, perhaps whole classes of organisms, that lack the necessary cellular machinery. For this reason, it may be premature to abandon the autonomous catalytic RNAs. During the 1990's, many studies aiming to demonstrate hammerhead ribozymes as gene silencing agents in vivo were not successful. The results obtained here, and in other recent studies on the effect of non-conserved sequences on the kinetics of cleavage [ 25 , 40 , 41 ], are beginning to reveal shortcomings of many studies of that era. That is the in vivo catalytic efficiency of the hammerheads under study was compromised by their simplistic design, which was based on in vitro studies performed with very short substrates. We were too quick to ascribe poor cleavage of transcripts to "accessibility" issues, and too slow to undertake comprehensive studies to understand the phenomenon. The miniribozyme design, and the effective nuclease protection afforded by the simple chemical modifications described here, provide an opportunity to revisit cellular and in vivo experiments with hammerheads that are around two orders of magnitude more efficient. Conclusions Long RNA substrates are much more effectively cleaved by miniribozymes than similarly targeted standard hammerheads. The presence of a stable helix II in the standard hammerhead appears to inhibit binding of the ribozyme to the substrate, and prevent the conformational mobility required for activity. Miniribozymes are affected by serum nucleases in a similar way to standard hammerheads, with cleavage occurring at the 3' side of unprotected ribopyrimidines. This work shows that it is readily possible to modify RNA/DNA hammerhead ribozymes, using commercially available reagents, to yield greatly improved nuclease resistance and retention of significant cleavage activity. These modified miniribozymes represent excellent candidates for cellular and in vivo studies on suppression of gene expression. Methods Oligonucleotide synthesis Oligoribonucleotides were synthesised using DNA phosphoramidites and ancillary reagents supplied by Perkin Elmer (Applied Biosystems Division, Foster City, CA), RNA and modified phosphoramidites from Glen Research (Sterling, VA) and using an Applied Biosystems Model 394 DNA synthesiser. Phosphorothioates were introduced by oxidation with Beaucage reagent (Glen Research, Sterling, VA). Deprotection and purification of oligonucleotides were as described previously [ 20 ]. The purity of each oligonucleotide was checked by labelling its 5'-end with 32 P phosphate using T4 polynucleotide kinase (New England Biolabs, Beverly, MA, USA) and [γ- 32 P]-ATP (Du Pont, Wilmington, DE), followed by electrophoresis in a 15% polyacrylamide gel containing 7 M urea and visualisation using a Molecular Dynamics PhosphorImaging system (Sunnyvale, CA). The concentrations of the purified oligonucleotides were determined by UV spectroscopy using the following molar extinction coefficients for the various nucleotides at 260 nm: A, 15.4 × 10 3 ; G, 11.7 × 10 3 ; C, 7.3 × 10 3 ; T/U, 8.8 × 10 3 l mol -1 cm -1 . All oligonucleotides were stored in distilled, deionised and autoclaved water at -20°C. The 550-nt CPSII substrate was transcribed by T7 RNA polymerase from the vector pCPS3b.1 [ 22 ] after linearisation with Eco RI restriction enzyme using an Ampliscribe T7 kit (Epicentre Technologies, Madison, WI, USA). The transcription reaction contained CTP, GTP and ATP (unlabelled) at 5 mM, UTP at 0.5 mM and 32 P UTP at about 1 nM. The transcription was for 90 minutes at 37°C, followed by the addition of 0.5 μL of DNAaseI and incubation for a further 20 minutes. The mixture was extracted three times with a 1:1 phenol/chloroform mix and once with chloroform, before precipitation of the transcript by the addition of 1/3 volume of 7.5 M ammonium acetate and incubation on ice for at least 2 hours. After recovery of the transcript by centrifugation and washing, the degree of incorporation of 32 P-labelled uridine was measured by Cerenkov counting of the transcript and the unincorporated UTP. The Tetrahymena intervening sequence transcript was generated from the ScaI digested pT7L-21 vector [ 24 ] (ATCC 40291) using the same method. The PDGF transcript was generated by T3 RNA polymerase transcription from the Eco RI digested vector pBSPDGF-LA using an Ampliscribe T3 kit (Epicentre Technologies, Madison, WI, USA). The vector pBSPDGF-LA was generated by insertion of the Xba I/Eco RI fragment from pmetPDGF-LA [ 42 ] into a similarly digested pBluescribe vector. Oligonucleotide sequences Deoxyribonucleotides are denoted by lower case letters, ribonucleotides by uppercase letters, modified nucleotides are shown in bold, C f = 2'-fluoro-2'-deoxycytidine, C me = 2' O-methylcytidine, U am = 2'-deoxy-2'-aminouridine, U me = 2' O-methyluridine, ps = phosphorothioate linkage. S-30, UAA CUU AUC AAG GUC* AAG AAC AUG AUG UUc, * denotes site of cleavage; Mrz-8/8, atgttctt CUGAUGA gttttc GAAAC cttgat; Mrz-12/12, catcatgttctt CUGAUGA gttttc GAAAC cttgataagt; Mz-12/12 catcatgttctt CUGAUGA gtttt GAAAC cttgataagt; Rz-12/12, catcatgttctt CUGAUGA GUCC UUUU GGAC GAAAC cttgataagt; Rz-6/12, gttctt CUGAUGA GUCC UUUU GGAC GAAAC cttgataagt; Mrz-12/12-A (= Mrz-12/12), catcatgttctt CUGAUGA gttttc GAAAC cttgataagt; Mrz-12/12-B, catcatgttctt C f U am GA U am GA gttttc GAAA C f cttgataagt; Mrz-12/12-C, catcatgttctt C U am GA U am GA gttttc GAAAC cttgataagt; Mrz-12/12-D, catcatgttctt C U am GAUGA gttttc GAAAC cttgataagt; Mrz-12/12-E, catcatgttctt CUGA U am GA gttttc GAAAC cttgataagt; Mrz-12/12-F, catcatgttctt CtGAUGA gttttc GAAAC cttgataagt; Mrz-12/12-G, catcatgttctt CuGAUGA gttttc GAAAC cttgataagt; Mrz-12/12-H, catcatgttctt C U me GAUGA gttttc GAAAC cttgataagt; Mrz-12/12-I, catcatgttctt CU ps GAUGA gttttc GAAAC cttgataagt; Mrz-12/12-J, catcatgttctt C me U ps GA U me GA gttttc GAAA C me cttgataagt ps t ps t; Mrz-12/12-K, catcatgttctt C me U am GA U me GA gttttc GAAA C me cttgataagt; Rz-12/12-L, catcatgttctt C U am GAUGA GUCC UUUU GGAC GAAAC cttgataagt; Tet Mrz, gcaatctattggtttaaa CUGAUGA gttttc GAAAC tagctaccaggtgcatg 3'; HC Mrz, 5' gtcgccacgacgac CUGAUGA gttttc GAAAC gttcccgctggt 3'; HC Rz, 5' gtcgccacgacgac CUGAUGA GGCC GAAA GGCC GAAAC gttcccgctggt 3'; HC S29, 5' ACCAGCGGGAACGUCGUCGUCGUGGCGAc 3'; PDGF Mrz, 5' CAGCUUCCUC CUGAUGA ggtaac GAAAU GCUUCUCt 3' ; PDGF S25, 5' GAAGAGAAGCAUCGAGGAAGCUGUc 3'. Cleavage kinetics Cleavage kinetics were studied at 37°C, pH 7.6 and 10 mM MgCl 2 under conditions of ribozyme excess; the substrate concentrations varied between 4 and 20 nM, and the ribozyme concentrations were between 20 nM and 10 μM. The short, synthetic substrates were labelled at their 5' ends using polynucleotide kinase (Roche Molecular Biochemicals) and γ- 32 P ATP. The long, transcribed substrates were uniformly labelled by including α- 32 P UTP in the transcription reaction. The ribozymes and 32 P-labelled substrates were mixed together in 20 μL of 50 mM Tris buffer and heated to 85°C for two minutes, then incubated at 37°C for 2 minutes. The tube containing the mix was centrifuged, and 2 μL was removed and put into 4 μL quenching solution which contained 90% formamide, 20 mM EDTA and 0.01 % xylene cyanol and bromophenol blue. The reaction was initiated by the addition of 2 μL of 100 mM MgCl 2 , and 2 μL samples were removed at various times and quenched as described above. The reaction products were separated using 15% denaturing polyacrylamide gels (containing 7 M urea), and then imaged using a Molecular Dynamics PhosphorImager. At each time-point, the amount of the substrate cleaved was calculated and plotted versus time. The data were fitted by a least-squares method using the program MacCurveFit [ 43 ], to an equation of the form: P t = P ∞ -(exp(-k obs t)P Δ ) where P t is the amount of product at time t, P ∞ is amount of product generated in the exponential phase of the reaction, k obs is the first-order rate constant for the reaction, and P Δ is the difference in the amount of product at t = 0 and P ∞ . Some reactions resulted in biphasic kinetics in which the first-order phase described above was followed by a slower step which was accommodated adequately in the curve-fitting process by the addition of a linear term, P t = {P ∞ -(exp(-k obs t)P Δ )} + d*t. Apparent dissociation constants "Kd" and the maximal first-order rate constants k max were determined from the plot of k obs versus ribozyme concentration according to the simple binding isotherm: k max = (k obs * [Rz])/("K d " + [Rz]) by least-squares fitting using the program MacCurveFit. Serum stability Ribozymes labelled at their 5' end with 32 P phosphate were dissolved in RPMI medium (Gibco) at 2 μM and the reactions were initiated by adding human serum (pooled human serum, Red Cross Blood Bank, Sydney, NSW), to a final concentration of 10%. A 2 μL sample was removed immediately and the remainder incubated at 37°C with samples removed at the times indicated in Figures 2 , 3 , 4 . The samples were quenched by addition to 4 μL of 90% formamide containing 20 mM EDTA and 0.01% bromophenol blue and xylene cyanol. The products of nuclease digestion in each sample were separated on 15% polyacrylamide gels containing 7 M urea and imaged using a Molecular Dynamics PhosphorImager. Authors' contributions The authors jointly conceived and designed this study. PH synthesised the ribozymes and performed the kinetic and stability analyses. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC544870.xml

529444

Morphological correlates of injury-induced reorganization in primate somatosensory cortex

Background Topographic reorganization of central maps following peripheral nerve injury has been well characterized. Despite extensive documentation of these physiological changes, the underlying anatomical correlates have yet to be fully explored. In this study, we used Golgi impregnation and light microscopy to assess dendritic morphology following denervation of the glabrous hand surface in adult primates. Results After survival durations that permit complete physiologically-defined reorganization, we find a systematic change in the dendritic arborization pattern of both layer II/III pyramidal and layer IV spiny stellate cells in the contralateral hand region of area 3b, compared to unaffected cortical areas. In general, our analyses indicate a progressive expansion of distal regions of the dendritic arbor with no appreciable changes proximally. This pattern of distal dendritic elaboration occurs for both basilar and apical dendrites. Conclusions These observations are consistent with the notion that latent inputs gain expression in reorganized cortex after nerve injury via their influence through contacts with more distally located termination sites.

Background The ability of the nervous system to modify its output in accordance with experiential demands is a central tenet of neuronal plasticity. For many years, the view of critical periods permeated our beliefs; almost dictating that plasticity beyond such epochs was, at best, minimal. The seminal experiments of Merzenich, Kaas and colleagues [ 1 , 2 ] have proved instrumental in moving the field beyond this restrictive mindset by showing that the central representation of the skin surface is subject to dramatic modification following peripheral nerve injury in adult primates. On the foundation of these observations, great strides have been made in understanding the mechanisms [ 3 - 9 ] and extent [ 10 - 13 ] of this phenomenon. These findings have generalized beyond sensory systems and collectively have been interpreted as reflective of fundamental properties of the nervous system. While physiological techniques are frequently used to characterize topographic (re)organization of central maps, the underlying anatomical correlates have not been thoroughly investigated. Using intracellular injection techniques, thalamic axons have been reported to innervate a much broader sector of cortex than necessary to represent typical receptive field size, suggesting the existence of "latent" inputs [ 14 , 15 ]. Disinhibition is a strong candidate as the primary mechanism during the immediate phase of somatotopic reorganization following nerve injury [ 16 - 18 ]. While unmasking of latent inputs may account for a portion of the overall reorganization [ 19 ], modification of central maps is neither complete immediately following nerve injury [ 1 , 20 - 22 ] nor dependent on a single mechanism [ 20 , 22 - 25 ]. Moreover, topographic reorganization appears to be permanent in nature [ 26 ], while at least some neurochemical changes have shown to be relatively transient [ 6 , 18 ]. Together, these observations suggest that alterations in the underlying anatomical connectivity might provide a stable platform for the maintenance of modified somatotopy. In this study, we report our examination of neurons in two cortical layers; spiny stellate cells in layer IV, as this is the primary input target of thalamocortical axons; and pyramidal neurons in layer II/III, as supragranular changes have been shown to precede somatotopic modification in the granular cell layer [ 27 ]. We predicted that dendritic arborization in the affected areas would be altered following peripheral nerve injury, providing an anatomical correlate of the functional changes. If the anatomical correlates of physiologically-defined changes can be readily observed, our understanding of the mechanisms underlying such changes would be greatly enhanced. Results Figure 1 presents typical Golgi-filled layer II/III pyramidal (Fig. 1A ) and layer IV spiny stellate cells (Fig. 1B ). Figures 1C (pyramidal) and 1D (stellate) are corresponding reconstructions of the same two cell types. Our initial inspection of the data revealed considerable heterogeneity as one moved from proximal to distal regions of the dendritic arbor. With regards to the proximal halves of the arbors, we observed no statistically significant differences between deprived and control groups for either basilar or apical dendrites. Layer II/III pyramidal neurons We found a greater number of intersections in the distal halves of dendritic arbors of pyramidal cells in deprived relative to control cortex. For basilar dendrites, there is a 92.0% increase in the number of intersections in the distal arbors of deprived cells relative to control cells (9/9, p < .01; see Fig. 2a ). Likewise, we find an 89.5% increase in total basilar dendritic length in the distal sectors of deprived cortical pyramidal cells, compared to control neurons (9/9, p < .01; see Fig. 2b ). For apical dendrites, there are 63.1% more intersections in distal portions of the arbors of deprived cells relative to controls (15/16, p < .01; see Fig. 2c ). Finally, the average length of the distal apical dendrites of deprived pyramidal cells is 37.4% greater than in controls (14/16, p < .01; see Fig. 2d ). Layer IV spiny stellate cells A largely comparable set of outcomes were found for spiny stellate cells in layer IV. For the distal sectors of basilar dendrites, there are 66.7% more intersections in arbors of deprived relative to control neurons (9/9, p < .01; see Fig. 3a ). Similarly, the overall average length of distal basilar dendrites is 92.4% longer in deprived stellate cells compared to controls (9/9, p < .01; see Fig. 3b ). For the distal apical dendritic arbors, deprived stellate cells have, on average, 25.5% more intersections than are found in control neurons (11/12, p < .01; see Fig. 3c ). Conversely, while the distal apical dendrites of deprived neurons are 20.5% longer than controls, on average, this difference is not statistically significant (8/12, p > .10; see Fig. 3d ). Differential effects of deprivation on basilar versus apical dendrites Deprivation of a specific region of somatosensory cortex by nerve transection clearly had a detectable effect on the distal portions of both the basilar and apical dendrites of both layer II/III pyramidal cells and layer IV spiny stellate cells. The data also support the contention that the basilar dendrites of both cell types were more profoundly affected by deprivation than were their apical dendrites. The magnitudes of the deprivation effects are more pronounced in basilar than in apical dendrites (see Fig. 4 ; Mann-Whitney = 0; p < .01). Discussion General observations In the present experiments, we investigated whether changes in dendritic morphology of neurons in deprived somatosensory cortex are correlated with the well-documented topographic reorganization that follows peripheral nerve injury in adult primates. Our general findings were that in deprived cortical areas, dendritic arbors were expanded distally, while being unaffected proximally. This pattern was found for both the basilar and apical dendrites of layer IV spiny stellate and layer II/III pyramidal neurons, though the effects were more pronounced for the basilar dendritic arbors, a difference that is consistent with previous reports [ 28 ]. Measures of dendritic length and the frequency of intersections, both well accepted metrics of dendritic arborization, yielded generally similar patterns of observations. These anatomical changes could well provide the means by which the functional changes in map topography proceed. Alternatively, they could reflect generic neural responses to deprivation per se , and have little or nothing to do with the functional reorganization. Does nerve injury-induced reorganization reflect the strengthening of normally latent inputs? Previous research has shown that the spread of thalamocortical arbors is much broader than necessary for the expression of typical receptive field size in primary somatosensory cortex [ 14 , 15 ]. Because of this disparity between the grain of the cortical topographic map [ 29 ] and the far more extensive thalamocortical anatomy, we have suggested that all parts of thalamocortical arbors cannot be equally effective in conveying suprathreshold receptive field information to the cortex, and that changes in synaptic efficacy could sustain the topographic plasticity that follows peripheral nerve injury [ 14 , 30 ]. Our observations of subthreshold, latent inputs to the cortex [ 3 ], and the emergence of their expression when dominant inputs are attenuated [ 31 ] lend support for this idea. Moreover, these presumptive latent inputs are largely prevented from gaining expression in cortex when NMDA glutamatergic receptors are blocked [ 20 , 23 ]. Such a blockade could prevent reorganization by preventing changes in the strength of existing synapses [e.g., [ 32 ]], by interfering with neurite outgrowth [ 33 ], or both. In any event, such latent inputs become evident only when the normally expressed, dominant inputs are somehow weakened – via pharmacological disinhibition [ 16 , 18 , 34 ], nerve injury [ 19 , 35 ], or use-dependency [ 36 - 38 ]. The data reported here are consistent with this notion, and suggest that distal sectors of the dendritic arbor may be selectively innervated by these latent inputs. Our observations that distal regions of apical dendrites, which clearly reside in upper cortical layers, are modified come as no surprise. Previous work has shown that layer IV spiny stellate cells act primarily as intracolumnar signal processors; while pyramidal cells integrate both horizontal and top-down information [ 39 ]. Supragranular layers appear to be particularly fertile to altered stimulation patterns as cortical reorganization occurs initially in the outermost layers of cortex, followed later by changes in the granular cell layer [ 27 ]. Measures of astrocytic recruitment mirror this outside-to-inside temporal progression of experience-dependent reorganization as well [ 40 ]. The selective elaboration of distal regions of the dendritic arbor is also consistent with data that implicate intracortical pathways as playing a major role in cortical reorganization [ 41 , 42 ], though, clearly, the contribution of bottom-up processes cannot be discounted [ 12 ]. Is reorganization a secondary consequence of other mechanisms/processes? While morphological changes may be less likely to account for acute changes in somatotopy after nerve injury, restructuring of the underlying anatomy could well correlate with the longer-term, persistent changes in cortical topographic maps. The modifications of distal dendritic regions reported here may be interpreted from at least three possible, non-exclusive, perspectives. First, expansion of the distal arbor may be a homeostatic response to a reduction in stimulation frequency/pattern following nerve injury. Progressive elaboration of the distal arbor might be an attempt to maintain optimal stimulation levels, and, thus, normal interneuronal trophic relationships. The altered somatotopy could be construed as simply the epiphenomenonal consequence of the activation of a homeostatic response. Second, the elaboration of the distal arbor may be a general property of the nervous system, a mechanism that permits the brain to respond in a dynamic and adaptable manner [ 43 ]. This supposes that the functional changes in cortex that follow nerve injury are adaptive, and that has not been convincingly demonstrated. Third, the observation that changes occur distally may simply reflect the fact that areas relatively distant to the soma are more vulnerable/susceptible to changes, regardless of the adaptability of such changes [ 44 ]. The reliability of progressively distal changes independent of dendritic location (apical or basilar) is certainly consistent with this idea. While these possibilities are not mutually exclusive, and certainly not all-inclusive, we believe that expansion of the distal arbor reported here is reflective of the altered activation pattern following nerve injury and serves as a long-term trace of this modified stimulation pattern. Conclusions Considering the range of survival durations following nerve injury in the current study, the observation of modifications to both layer IV spiny stellate and layer II/III pyramidal neurons was not unexpected. While this broad survival range may have "smeared" our snapshot with respect to the temporal integration of anatomical changes, our intention was simply to determine whether morphological changes were occurring at any point during the reorganization process. Our data clearly indicate that the anatomy in affected cortical areas is subject to modification and that the morphological changes observed may be related to the functional reorganization revealed electrophysiologically. We have begun experiments to better refine the temporal window in which these changes become evident. In sum, we have shown that just as the functional responsiveness of the mature primate nervous system is susceptible to change, so is the underlying anatomy. Our observations that the anatomical changes appear to be either potentiated in, or possibly restricted to, distal regions of the dendritic arbor provide additional insight into the mechanisms involved in the physiological changes. Further research will be instrumental in determining the exact role that the underlying anatomy plays in this complex reorganization process. Methods Adult squirrel monkeys ( Saimiri scireus or Saimiri bolivensius ) were socially housed with food and water available ad libitum . In six animals, the median and ulnar nerves to one hand were transected following the principles of animal care detailed in NIH publication no. 86–23. The local institutional animal care and use committee approved all procedures prior to initiation of any experiments. Briefly, monkeys were anesthetized with an intramuscular injection of a mixture of ketamine hydrochloride (25–30 mg/kg) and xylazine (0.5–1.0 mg/kg). Their forearms were shaved and prepared for surgery with alternate scrubbings of povidone-iodine and alcohol. Under sterile conditions an incision was made along the midline of the ventral forearm, the median and ulnar nerves were located by blunt dissection and cut about midway between the elbow and wrist. The epineural sheath of the proximal stump was retracted 0.5–1.0 cm, and the exposed nerve avulsed. The empty epineural sheath was re-extended, folded back upon itself and ligated. The nerve stumps were repositioned and the incision closed with sutures. Post-surgically, all subjects received penicillin, dopram hydrochloride, and dexamethasone injections. Subjects were permitted to recover for a period of time previously shown sufficient to permit complete reorganization of the hand representation in cortical area 3b (3–52 months, mean = 15.3). Two additional subjects served as naïve controls. Following electrophysiological mapping of the affected cortical areas, animals were overdosed and perfused transcardially with 0.9% saline. Brains were extracted and immersed in a modified Golgi–Cox solution for 11 days, thereafter dehydrated and embedded in celloidin. Tissue blocks were sectioned coronally at 150 μm in thickness for morphological assessment and every fourth section in the series was cut at a thickness of 90 μm for Nissl staining to facilitate cytoarchitectonic identification of area 3b borders. Free-floating sections were processed and mounted on glass, according to previously reported procedures [ 45 ]. Analysis of dendritic morphology was conducted blind to experimental condition on thoroughly impregnated cells using methods described by Sholl [ 46 ]. For each animal, ten layer IV spiny stellate and ten layer II/III pyramidal cells from the hand area of somatosensory area 3b contralateral to the nerve injury were drawn using Neurolucida (MicroBrightfield) at 600 × magnification. In addition, ten area 3b cells of the same two types located outside of the hand representation were drawn to serve as controls. We have shown previously that dominant and latent inputs from the three nerves innervating the hand have overlapping territories in area 3b [ 13 , 32 ], with the latent inputs gaining expression when the dominant inputs are weakened with peripheral nerve transection. These observations prompted us to treat the proximal and distal portions of the dendritic arbors separately in our statistical comparisons. To accomplish this goal, we divided the arbors into proximal and distal halves using the Sholl ring halfway between the soma and the most distal dendritic process as the dividing point. For dendritic length and intersection comparisons, the deprived and control averages for each Sholl ring were compared. A simple binomial test [ 47 ] was then applied to determine whether a systematic, statistically significant difference exists between those sets of means. Authors' contributions JDC participated in design of study, conducted the histological processing, drafted the manuscript: JAT conducted some of the histological processing: CLW participated in design of study: DRS participated in design of study: PEG participated in design of study, conducted statistical analyses, conceived and coordinated the study. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC529444.xml

545213

Completing the Public Health HIV/AIDS Alphabet

Dr. Gerberding outlines critical steps for arresting the HIV/AIDS epidemic [1] . She suggests moving ahead with “ABCs” and with “D” for diagnosis and “R” for responsibility. These are good suggestions—with increased HIV testing and individuals taking responsibility for their role in HIV spread, the epidemic might be slowed. We could continue to add incrementally to the alphabet soup of public health. But instead, we could choose to immediately implement the mainstays of public health—universal testing and contact tracing [ 2 , 3 , 4 ]. Every sexually active individual and every individual at risk for HIV deserves to know their HIV status. Thus, every HIV-infected individual must be called upon to be accountable for preventing HIV transmission. Contact tracing should be instituted for HIV just as it is for other infectious diseases. Those who have been exposed to HIV have a right to know how to protect themselves and if they too are infected, to be offered treatment [5] . HIV testing has too often focused on testing of women in a perinatal setting rather than universal testing in routine clinical care. Without universal voluntary HIV testing and contact tracing, we will see the continued tilt of the epidemic toward women, now at 55% of all HIV infections and in all likelihood at 75%–80% in another 8 to 10 years [ 6 , 7 ]. For too long the debate has been that contact tracing will result in physical abuse of women. Confining our definition of abuse of women to physical abuse alone is to have too narrow an ethical focus—HIV infection itself is an abuse of women or of anyone else. Universal HIV testing and contact tracing adds an essential comprehensive public health approach to the epidemic that will be successful in reducing the ever-escalating numbers of new infections.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC545213.xml

555941

A comparative sequence analysis reveals a common GBD/FH3-FH1-FH2-DAD architecture in formins from Dictyostelium, fungi and metazoa

Background Formins are multidomain proteins defined by a conserved FH2 (formin homology 2) domain with actin nucleation activity preceded by a proline-rich FH1 (formin homology 1) domain. Formins act as profilin-modulated processive actin nucleators conserved throughout a wide range of eukaryotes. Results We present a detailed sequence analysis of the 10 formins (ForA to J) identified in the genome of the social amoeba Dictyostelium discoideum . With the exception of ForI and ForC all other formins conform to the domain structure GBD/FH3-FH1-FH2-DAD, where DAD is the Diaphanous autoinhibition domain and GBD/FH3 is the Rho GTPase-binding domain/formin homology 3 domain that we propose to represent a single domain. ForC lacks a FH1 domain, ForI lacks recognizable GBD/FH3 and DAD domains and ForA, E and J have additional unique domains. To establish the relationship between formins of Dictyostelium and other organisms we constructed a phylogenetic tree based on the alignment of FH2 domains. Real-time PCR was used to study the expression pattern of formin genes. Expression of forC, D, I and J increased during transition to multi-cellular stages, while the rest of genes displayed less marked developmental variations. During sexual development, expression of forH and forI displayed a significant increase in fusion competent cells. Conclusion Our analysis allows some preliminary insight into the functionality of Dictyostelium formins: all isoforms might display actin nucleation activity and, with the exception of ForI, might also be susceptible to autoinhibition and to regulation by Rho GTPases. The architecture GBD/FH3-FH1-FH2-DAD appears common to almost all Dictyostelium , fungal and metazoan formins, for which we propose the denomination of conventional formins, and implies a common regulatory mechanism.

Background Eukaryotic cells rely on de novo nucleation mechanisms to generate actin filaments in order to elicit spatial and temporal remodeling of their actin cytoskeleton. Besides the Arp2/3 complex, nucleation activity has been recently demonstrated also for formins (reviewed in [ 1 ]). Formins are multidomain proteins conserved from plants to fungi and vertebrates. Their name originates from the mouse limb deformity gene. Mice with mutant alleles fail to form proper limbs and kidneys [ 2 ]. Subsequently, homologues were identified in Drosophila ( Diaphanous ) [ 3 ] and yeast (Bni1p and Cdc12p) [ 4 , 5 ]. Due to their pivotal role in the organization of the actin cytoskeleton formins are involved in processes as diverse as formation of filopodia, microspikes and lamellipodia, establishment and maintenance of cell polarity, vesicular trafficking, formation of adherens junctions, cytokinesis, embryonic development and signaling to the nucleus (reviewed in [ 6 ]). The FH2 (formin homology 2) domain is the defining feature of all formins. It is very well conserved and is almost invariably preceded by a proline-rich region, the FH1 (formin homology 1) domain [ 6 , 7 ]. In vitro, the FH2 domain competes with barbed-end capping proteins and is necessary and sufficient to nucleate actin polymerization, but the FH1 domain, which interacts with profilin-actin, funnels actin to the nucleation vicinity and confers full activity to the molecule [ 1 ]. Contrary to the Arp2/3 complex, which nucleates a new filament on the side of a preexisting filament, remains attached to the pointed end of the new filament and generates branched networks [ 8 ], the FH2 domain binds and stays associated to the barbed end, giving rise to unbranched filaments [ 9 - 11 ]. The crystal structure of the FH2 domain of two formins, Bni1p and mDia1, has been recently solved. Its fold is almost entirely α-helical and forms a ring-shaped flexible but stable dimer that caps the barbed end and allows processive elongation of the actin filament [ 12 , 13 ]. The FH1 domain is also a binding site for diverse SH3-domain containing proteins like Src-like non-receptor tyrosine kinases, WISH (WASP-interacting SH3 protein) and IRSp53 (insulin receptor substrate) in mammals, and Hof1p in yeast [ 6 ]. In most fungal and metazoan formins the FH1-FH2 core is accompanied by a less well conserved N-terminal FH3 (formin homology 3) domain involved in targeting [ 14 ]. In plants targeting might be mediated by membrane insertion signals or PTEN (phosphatase and tensin)-related domains [ 15 , 16 ]. Some formins, the so called Diaphanous-related formins, are able to interact with activated Rho GTPases through a poorly defined N-terminal Rho GTPase binding domain (GBD) that overlaps with the FH3 domain [ 6 , 7 ]. This binding releases the intramolecular inhibitory interaction between the GBD and a C-terminal Diaphanous autoregulatory domain (DAD) and renders the protein active [ 10 , 17 ]. The social amoeba Dictyostelium discoideum is an attractive model organism to investigate the components of the actin cytoskeleton and the signaling pathways involved in its regulation [ 18 , 19 ]. Dictyostelium amoebae are equipped with a complex actin cytoskeleton that endows the cells with motile behavior comparable to that of human leukocytes. In fact, a genome-wide survey revealed that the repertoire of cytoskeletal components of Dictyostelium is more similar to metazoa followed by fungi than to plants (Eichinger, et al., submitted). In Dictyostelium , nine formins have been previously identified but only three of them have been characterized to some extent [ 20 ]. Mutants lacking ForA, ForB or both showed no detectable phenotype, whereas disruption of the gene encoding ForC, which is expressed predominantly at late developmental stages, led to a cell autonomous developmental defect with the formation of aberrant fruiting bodies, suggesting this formin mediates actin remodeling during multicellular stages. In vivo experiments with GFP fusions showed that the N-terminal region of ForC targets the protein to places of active actin reorganization, like macropinosomes, phagocytic cups and cell-to-cell contacts [ 20 ]. We have made use of the information released by the Dictyostelium sequencing projects in order to achieve a complete inventory of formin genes. A detailed sequence analysis of the 10 formins identified revealed that, with the exception of ForI and ForC, all other formins conform to the domain structure GBD/FH3-FH1-FH2-DAD present in almost all fungal and metazoan formins, for which we propose the denomination of conventional formins. Our sequence analysis also indicates that the GBD and FH3 domains constitute a single domain also found in two Dictyostelium RasGEFs (guanine nucleotide exchange factors). The expression pattern of formin genes during asexual and sexual development was studied using real-time PCR. Our analysis allows some preliminary insight into the functionality of Dictyostelium formins: all isoforms might display actin nucleation activity and, with the exception of ForI, might also be susceptible to autoinhibition and regulation by Rho GTPases. Results Sequence analysis of Dictyostelium formin genes In a previous publication 9 genes that potentially encode proteins of the formin family were identified in Dictyostelium [ 20 ]. For some of the formins (ForA through D and ForF) full length sequences were available, whereas for the rest N- and C-terminal sequences were missing. For a complete analysis of this family in Dictyostelium we sought to exploit the available databases in order to achieve a complete inventory of formin genes in its entire length. The sequences already reported by Kitayama et al. [ 20 ] were used as queries for Blast searches of the Dictyostelium genomic DNA database. This allowed assembly of complete genomic sequence for forA through forI . In order to verify the predicted amino acid sequence for each formin, Blast searches were performed against the Dictyostelium EST database. In cases where no EST sequences were available, like forG and forI , introns were verified after RT-PCR. Inspection of the EST sequences led to the identification of one more formin gene, forJ , whose genomic sequence was also retrieved and inspected. Recent completion of the assembly of the Dictyostelium genome allowed us to confirm our gene predictions and map each formin gene to its corresponding chromosome locus (Eichinger et al., submitted). Formin genes are dispersed all over the six chromosomes (each chromosome harbors at least one formin gene), and in no case two or more genes are placed adjacent to each other (Table 1 ). Table 1 Features of Dictyostelium discoideum formins Sequences can be accessed through the Dictybase identifier at Gene Dictybase ID Chromosome Number of introns Numer of residues forA DDB0214996 3 5 1218 forB DDB0215000 3 1 1126 forC DDB0191362 5 2* 1158 forD DDB0205290 3 3 1214 forE DDB0190413 1 0 1561 forF DDB0188569 5 1 1220 forG DDB0169087 2 1 1074 forH DDB0186588 4 3 1087 forI DDB0186053 4 2 935 forJ DDB0183855 6 1 2546 * One intron upstream of the start codon. With the exception of forE , all other formin genes are interrupted by one or more introns, which are generally placed in the 5' half of the sequence, upstream of the region encoding the FH1 domain (Fig. 1 , arrowheads). Only in forC is an intron placed in the region encoding the FH2 domain. ForC is also the only case where an intron was identified upstream of the start codon. Dictyostelium formin genes do not appear to undergo alternative splicing, at least within the coding region. This is in contrast to metazoan and plant formins, where alternative splicing gives rise to a large number of variants that frequently differ in their pattern of tissue distribution and interaction with binding partners. Figure 1 Domain organization of Dictyostelium formins. With few exceptions, Dictyostelium formins conform to the domain structure GBD/FH3-FH1-FH2-DAD. Diagrams have been aligned with the FH2 domain. Regions with high probability of coiled coil structure are depicted as thin gray rectangles. C1 and C2 correspond to protein kinase C conserved regions 1 and 2, respectively. FHA is a forkhead-associated domain. Numbers inside the FH1 boxes indicate the number of XPPPPP motifs. Triangles denote the position of introns. Only introns placed in coding regions are shown. Intron positions shared by two or more genes have been labeled with letters. Only two intron positions are conserved among Dictyostelium formin genes (Figs. 1 and 4 ). Intron a is conserved in forA , forB , forD and forH , whereas intron b is conserved in forB and forH . The conserved FH3-FH1-FH2 core domain composition (see below) along with these two intron positions underscore the view that all Dictyostelium formin genes might have arisen from a common ancestor gene. After duplications and divergence from this ancestral formin gene introns were acquired or lost and additional domains and extensions were appended to some genes. Figure 4 Multiple alignment of the GBD/FH3 domains of Dictyostelium formins and two RasGEFs Amino acid sequences were aligned with ClustalX and the output file was subsequently edited manually. In addition to nine Dictyostelium formins, a GBD/FH3 domain was identified also at the N-terminus of RasGEF-L and RasGEF-V. The sequence of the human DRF3 has been included for reference. Dashes indicate gaps introduced for optimal alignment. In some places extensive repetitive stretches have been removed and replaced by a figure indicating the number of residues omitted. Residues identical or similar in at least 40% of the sequences are boxed in black or gray, respectively. Continuous and discontinuous lines indicate, respectively, the extension of the GBD and FH3 domains as defined in the Pfam database. Short arrows indicate boundaries of the FH3 domain as proposed by Petersen et al. [14]. Conserved intron positions are labeled a and b (see Fig. 1). Domain structure of Dictyostelium formins: the FH2 domain The domain structure and topology of all ten Dictyostelium formins was determined by means of bioinformatics tools and visual inspection. Although formins vary considerably in length (935 residues of ForI versus 2546 of ForJ), with few exceptions they have in common a core of about 1100 residues that harbors a GBD/FH3-FH1-FH2-DAD structure characteristic of most fungal and metazoan formins (Fig. 1 ). To better appreciate the relationships among the members of the Dictyostelium formin family and to analyze the requirements for their function, we have generated multiple alignments of the FH2-DAD domains as well as the GBD/FH3 domain. The FH2 domain is the best conserved domain of formins (Fig. 2 ). In general, the FH2 domain is about 400 residues long. Some Dictyostelium formins (ForC, D and I) have one or more stretches of intervening repetitive sequences of variable length rich in Arg, Gln or Ser. Such repetitive sequences are characteristic of many Dictyostelium genes. The crystal structure of the FH2 domain of two formins, Bni1p and mDia1, has been recently solved [ 12 , 13 ]. We will consider the FH2 domains of Dictyostelium formins in the context of these two structures. The FH2 domain fold is almost entirely α-helical. It is a stable dimer that forms a closed parallelogram-shaped ring. The structure of this domain can be subdivided into subdomains. At the N-terminus a so-called lasso is connected to a globular knob (helices α1 to α5 in red in Fig. 2 ) by a linker of variable length. The knob is followed by a three helix bundle with a coiled-coil structure (α6, α11 and α12 in blue). The C-terminal subdomain (helices α7 to α10 and α13 in green) forms a so-called post. The lasso subdomain of one subunit encircles the post subdomain of the other subunit in a dimer. The post also harbors the GNY/FMN sequence motif that originally defined the FH2 domain (box at the end of helix α7) [ 21 ]. Residues of the lasso/post interface are highly conserved, particularly Trp1 and 2 (substituted by Phe in ForB, E and F) that insert into hydrophobic pockets in the post flanked by Gly residues 6 and 8 (Fig. 2 ). Figure 2 Multiple alignment of FH2 and DAD domains of Dictyostelium formins. Amino acid sequences were aligned with ClustalX and the output file was subsequently edited manually. The sequence of the human Diaphanous-related formin 3 has been included for reference. Dashes indicate gaps introduced for optimal alignment. In some places extensive repetitive stretches have been removed and replaced by a figure indicating the number of residues omitted. Residues identical or similar in at least 40% of the sequences are boxed in black or gray, respectively. Secondary structure elements as determined for mouse Dia1 core FH2 domain [12] are indicated on top of the aligned sequences. Color coding denotes the N-terminal knob subdomain (red), three-helix-bundle (blue) and FH2 motif post-containg region (green). Regions involved in the formation of the lasso/post dimer interface, as determined for Bni1p [13] are also indicated, as well as the highly conserved GNY/FMN motif (boxed). Conserved residues discussed in the text are indicated by circles and are numbered consecutively. Below the DAD region triangles indicate conserved residues discussed in the text. All residues of the sequence motif GNY/FMN participate in dimerization. This motif is also highly conserved in almost all Dictyostelium formins (NY is substituted by SI in ForI) but the important methionine residue (Met7) [ 12 ] is present only in ForG and ForJ and is substituted by other hydrophobic residues in the rest of the Dictyostelium formins as well as in members of the FHOD (formin homology domain containing protein) and plant class1 subfamilies. Also very conserved are some residues probably involved in binding to actin, like Ile3 (absolutely conserved) in the N-terminal subdomain and Lys9 in the post region (substituted by Arg in ForB and ForH). Mutation of these residues in Bni1p to Ala and Asp, respectively, abolished actin nucleation and barbed end capping activity of the FH2 domain [ 13 ], and replacement of Lys9 and two adjacent Lys residues by Ala abolished alignment of microtubules and bundling of F-actin induced by activated mDia1 [ 22 ]. Other conserved residues are Asp4 (or the conservative substitution by Glu in most of the Dictyostelium formins) and Arg5 (substituted by Lys in ForB and ForE). These residues were found mutated in temperature-sensitive yeast mutants [ 23 , 24 ], and they probably participate in stabilization of the knob region [ 13 ]. In summary, all essential residues in the FH2 domain revealed by structural and functional studies in metazoan and fungal formins are conserved in Dictyostelium formins, indicating that all ten formins might be functional actin nucleators. Dictyostelium formins in the context of other organisms In order to establish the relationship between formins of Dictyostelium and other organisms and to investigate whether different species share subfamilies of formins, we constructed a phylogenetic tree based on the alignment of complete sets of sequences of FH2 domains from selected organisms, including representatives of fungi, plants, invertebrates and vertebrates. We retrieved sequences of already characterized formins and additionally we made a search of further available sequences through the SMART server with the FH2 domain as query. Appart from the ten sequences of Dictyostelium , we collected a total of 62 sequences, 21 from plants, 5 from yeasts, 6 from D. melanogaster , 6 from C. elegans and 14 from human. Taking into account that some genes might not have been predicted accurately and that predicted proteins not supported by EST sequences were not considered for our analysis, further metazoan formins, especially from human, most probably went unidentified in our search. The phylogenetic tree (Fig. 3 ) supports the high degree of conservation of the FH2 domain, as becomes evident from the homogeneous branch length for most of the sequences. Yeast formins and some C. elegans members are more divergent. The phylogenetic analysis reveals clustering of most formins into well defined classes. Yeast formins form a separate class whereas plant formins significantly group into any of two classes. Metazoan formins do not constitute a single cluster, rather they distribute into a number of subfamilies. The FHOD, Diaphanous and FMNL (formin in leukocytes) subfamilies have representatives in human, D. melanogaster and C. elegans . The Cappuccino/Formin and DAAM (Dishevelled-asssociated activator of morphogenesis) subfamilies, as well as a novel subfamily, is present in human and D. melanogaster , but seems to be absent in C. elegans . Delphilin constitutes a subfamily with a unique member present only in human. Finally, C. elegans has some additional divergent formins apparently unique to this organism. Figure 3 Phylogenetic tree of FH2 domains of formins from Dictyostelium and other organisms. Amino acid sequences of the FH2 domains (core and lasso region) were aligned with ClustalX and the output file was subsequently edited manually. A bootstrapped unrooted phylogenetic tree was constructed as described in the Methods section. Dictyostelium members are indicated in red. The other organisms considered are Arabidopsis thaliana (At), Saccharomyces cerevisiae (Sc), Schizosaccharomyces pombe (Sp), Drosophila melanogaster (Dm), Caenorhabditis elegans (Ce) and Homo sapiens (Hs). Nodes supported by either >75% or >50% bootstraps have been marked with red or green circles, respectively. For simplicity, nodes outside of a cluster supported by >50% bootstraps have not been indicated. Asterisks denote novel formin subfamilies. The scale bar indicates percent substitutions. On average, Dictyostelium formins are 45.5% similar (23.8% identical) to each other, with ForC being only slightly more divergent (40.0%/20.4% similarity/identity to the rest of Dictyostelium formins). A comparable degree of similarity (identity) was found to members of several subfamilies of metazoan formins, and ranged between 40% (20%) and 48% (24%). Similarity (identity) to plant and yeast formins was lower: 38% (19%) and 36% (17%) respectively. Dictyostelium ForC and ForG cluster together with the Cappuccino/Formin group (75% bootstraps), whereas ForI very weakly clusters with the FHOD subfamily (53% bootstraps). However, taking into account that the FH2 domain is highly conserved, the position of these three Dictyostelium formins in the tree does not necessarily mean functional relationship with the mammalian counterpart, because other domains are probably responsible for diversity of localization and function. Bootstrapping does not support a significant clustering of the rest of the Dictyostelium formins, and only few members cluster together with a reasonably high number of bootstraps (ForE, D, A and F, 51% bootstraps). Domain structure of Dictyostelium formins: FH1, FH3 and other domains The FH1 domain is a proline-rich region situated immediately upstream of the FH2 domain. It is present in almost all known formins, including that of Dictyostelium , with the notable exception of ForC. The length of the FH1 domains is very variable among formins (10 to >500 amino acids). It constitutes a binding site for the actin monomer binding protein profilin, as well as for SH3 and WW domain containing signaling proteins [ 25 , 26 ]. Binding to profilin is well established for a large number of formin proteins and might take place through type 1 proline-rich motifs with the sequence XPPPPP, where X is usually Gly, Leu, Ile or Ser. Dictyostelium formis have a variable number of these motifs, between 1 in ForD and 8 in ForA (Fig. 1 ). In most cases Gly occupies the X position. In general the motifs are separated by a short stretch of up to five residues, two or more of them usually glycines. In some formins, like ForA and ForF, the proline-rich motifs might be the product of internal duplications. ForE has one additional short proline-rich region located at the N-terminus of the protein. The FH3 domain was initially identified and characterized in the yeast formin Fus1p as a region consisting of three blocks of similarity in the same relative order in several formins [ 14 ]. It is less well conserved than the FH2 domain and is thought to be important for determining the intracellular localization of formins. Two domains of the Pfam database are recognized in this region that overlap with the FH3 domain of Petersen and co-workers [ 14 ], the Diaphanous GTPase-binding domain (PF06371) and the Diaphanous FH3 domain (PF06367). Automatic domain analysis identified a GBD and a FH3 domain in ForA, B, D, E, F and H. In ForC and ForJ a GBD was identified with confidence values slightly below the default threshold of the SMART tool. This was also the case for a FH3 domain in ForG and ForJ. A multiple alignment of the N-terminus of Dictyostelium formins with metazoan and fungal homologues revealed a homology region of approximately 380 residues in all Dictyostelium formins with the exception of ForI (Figs. 1 and 4 ). We will consider this region as a single GBD/FH3 domain (see discussion). In ForJ this domain is considerably longer due to stretches of intervening repetitive sequences rich in Arg and Ser residues. On average the GBD/FH3 domain of Dictyostelium formins displays 39% similarity to that of human DRF3 taken as reference for figure 4 . Interestingly, inspection of the Dictyostelium genome for proteins with a GBD as defined by Pfam PF06371 yielded two genes encoding RasGEF proteins of identical domain composition, RasGEF-L and RasGEF-V. Both proteins harbor a complete GBD/FH3 domain that is 35% similar to that of human DRF3 and constitute the first case where this domain is observed outside of a formin. We constructed a phylogenetic tree based on a multiple alignment of the GBD/FH3 domain of Dictyostelium formins (except ForI), RasGEFs, fungal formins and members of the Diaphanous, DAAM, FMNL and FHOD subfamilies (Fig. 5 ). With few exceptions automatic domain analysis identified GBD and FH3 domains in the metazoan and fungal formins. For example, a weak GBD was identified in D. melanogaster and C. elegans FHOD, but not in the human homologs, and conversely, a weak FH3 domain was identified in HsFHOD3 but not in other members of the subfamily. In those cases the missing domain could be reliably identified in multiple alignments. We could not identify a GBD/FH3 domain in members of the cappucino/formin subfamily. DmAE003560 has a FH3 domain and a short piece of a GBD but, interestingly, the human homolog KIAA1727 completely lacks an N-terminal region and starts at the FH1 domain. Inspection of the sequence databases did not allow clearing whether the available sequences correspond to spliced variants of longer proteins. The multiple alignment of the GBD/FH3 domain showed several blocks where similarity is higher among sequences, generally in the central part of the domain (Fig. 4 ). In many cases these blocks are separated by intervening stretches of variable length in the different subfamilies. We removed these insertions from our alignment prior to calculating the tree. The phylogenetic tree showed significant clustering of members of the respective metazoan subfamilies, and additionally the FMNL and DAAM subfamilies clustered together (73% bootstraps). Bootstrap analysis did not support clustering of fungal or Dictyostelium sequences into distinct classes, but interestingly, ForC and ForG significantly clustered with the FHOD family (92% bootstraps). Figure 5 Phylogenetic tree of the GBD/FH3 domains of formins and two RasGEFs from Dictyostelium and formins from other organisms. Amino acid sequences of the GBD/FH3 domains were aligned with ClustalX and the output file was subsequently edited manually and intervening sequences between blocks of high similarity were removed. The sequence available for DmAE003560 only contains a FH3 domain and a short part of the GBD. A bootstrapped unrooted phylogenetic tree was constructed as described in the Methods section. Dictyostelium members are indicated in red. The other organisms considered are as in the legend to figure 3. Labeling is also as in the legend to figure 3. The DAD immediately follows the FH2 domain and is required for autoinhibition by intramolecular interaction with the N-terminus of formins [ 21 ]. Inspection of the multiple alignment of the C-terminus of Dictyostelium formins revealed a DAD in all members with the exception of ForI (Figs. 1 and 2 ). This formin ends abruptly at the last α-helix of the FH2 domain. In all cases the DAD was placed in the vicinity of and no more than approximately 60 residues beyond the FH2 domain. The DAD is composed of two sections, a core leucine-rich sequence and a short stretch of basic residues. Both elements are present in the DAD of most Dictyostelium formins, in particular three hydrophobic residues shown to be required for activity in mouse Dia2 (indicated by triangles in Fig. 2 ) [ 17 ]. In ForJ, where these residues are substituted by polar or charged aminoacids, the DAD might not be functional. Like metazoan and fungal formins, most Dictyostelium formins have predicted coiled-coil regions adjacent to the FH3 domain that could act as protein-protein interfaces for yet unidentified ligands (Fig. 1 ). For example, in mammalian formin1 this region constitutes the binding site of α-catenin and is involved in recruitment of formin1 to nascent adherens junctions [ 27 ] and in Bni1p the coiled coil region harbors the binding site for Spa2, a protein involved in recruitment of Bni1p to the bud cortex [ 28 ]. A few Dictyostelium formins have additional predicted coiled coil regions upstream of the FH3 domain (ForE and ForI) or downstream of the FH2 domain (ForD and ForJ) that might constitute potential protein interaction sites with regulatory or targeting functions. Three Dictyostelium formins have additional recognizable domains at their N-terminus (Fig. 1 ). ForA has a C2 (protein kinase C conserved region 2) domain. This domain, present in phospholipases, protein kinases C, synaptotagmins and diverse other proteins, is thought to be involved in calcium-dependent phospholipid binding [ 29 ]. ForE has a C1 (protein kinase C conserved region 1) domain, a cysteine-rich region involved in zinc-dependent binding to diacylglycerol [ 30 ]. Finally, ForJ has a FHA (forkhead-associated) domain, a phospho-specific protein-protein interaction motif found in nuclear proteins [ 31 ]. None of these domains are found in formins from other organisms. ForJ is the only Dictyostelium formin with a long C-terminal extension. Similar extensions, although of unrelated sequence, can be observed in formins from other organisms, like yeast Cdc12 and For3, C. elegans Cyk-1 and AF106580, D. melanogaster AE003560 and human KIAA1727. Expression analysis Dictyostelium cells can propagate following either an asexual or a sexual life cycle. Characteristic of the asexual life cycle is the transition from single cell amoebae to a multicellular fruiting body consisting of at least two differentiated cell types. In the sexual life cycle some amoebae become sexually mature under dark and submerged conditions, fuse and form macrocysts. Either life cycle involves coordinated transcription of certain sets of genes. We have used quantitative real-time PCR to study the expression of the formin genes during sexual and asexual development (Fig. 6 ). Figure 6 Expression analysis of Dictyostelium formin genes. Expression analysis was performed using quantitative real time PCR on two independently isolated mRNA samples both in sexual and asexual developmental stages. Average and standard deviation of two independent expression ratios obtained from independent cDNA samples are shown. IC, fusion incompetent cells; FC, fusion competent cells; LS, light submerged cells. The expression patterns observed during asexual development can be classified into two major groups. Expression of forC, D, I and J displayed an increase during transition to multi-cellular stages, and except for forI , levels remained constantly high throughout the rest of development. The rest of genes displayed less marked developmental variations, and expression was either kept at constant levels ( forG and H ) or gradually increased ( forF ) or decreased ( forA, B and E ) after the onset of development. When expression was analyzed during sexual development only forH and forI displayed a significant increase of about 3-fold in fusion competent cells compared to fusion incompetent cells. Cells cultured in light submerged conditions have a reduced sexual fusion competency [ 32 , 33 ]. In parallel with this, forH and forI were enriched in fusion competent compared to light submerged cells, indicating that this enrichment is related to the acquisition of the fusion competence rather than to the submerged condition that was included to induce the fusion competence. Discussion We have performed a detailed sequence and expression analysis of the formin family of Dictyostelium , which in this organism comprises 10 genes. A comparison of the domain composition of formins from diverse phyla allows their grouping into four major classes (Fig. 7 ). In general, Dictyostelium formins can be grouped within the class of what we designate conventional formins (see below), which includes all fungal and almost all metazoan formins. This is in agreement with a genome wide analysis that places Dictyostelium closer to fungi and metazoa than to plants (Eichinger et al., submitted). Figure 7 Classification of formins according to structural and functional elements. Most formins of metazoans as well as formins of Dictyostelium and fungi can be classified as conventional formins, with a GBD/FH3-FH1-FH2-DAD structure, although in particular members or alternatively spliced variants a domain (but never the FH2) might be absent. Plant formins can be grouped into one of two classes. Delphilin is an unconventional formin lacking GBD/FH3 and DAD only found in metazoa. Formins are not drawn to scale. SP, signal peptide. TM, transmembrane region. With very few exceptions all formins have in common a FH2 domain immediately preceded by a FH1 domain. The FH1-FH2 combination constitutes the minimal core that is fully functional in terms of actin nucleation and elongation activity (reviewed in [ 1 ]). This FH1-FH2 core is very ancient, and its remarkable degree of conservation points at an essential role within the cell. The diverse formin classes differ in their N-terminal regions, which have regulatory and targeting roles. Plant formins characteristically lack GBD/FH3 and DAD domains and there is no evidence for an interaction with Rop GTPases. In plants the N-terminus is unrelated to that of other organisms. Class 1 plant formins are integral membrane proteins by virtue of a signal peptide or membrane anchor followed by a transmembrane domain, whereas some class 2 plant formins have a PTEN-related domain [ 15 , 16 ]. Conventional formins have characteristically a GBD/FH3 domain at the N-terminus. Together with the DAD region at the very C-terminus this domain confers in most cases regulatable autoinhibition through binding of activated Rho GTPases. Finally, Delphilin is a variation only present in vertebrates. Instead of a GBD/FH3 domain it has a PDZ domain that interacts with a glutamate receptor, and it has been proposed that receptor binding causes activation of this formin [ 6 , 34 ]. The GBD/FH3 domain: a targeting and regulation domain Our sequence analysis defines a putative GBD/FH3 domain in most Dictyostelium as well as fungal and metazoan formins. The two domains identified in the Pfam database at the N-terminus of several formins, the Diaphanous GTPase-binding domain and the Diaphanous FH3 domain, overlap with the FH3 domain proposed initially by Petersen et al. [ 14 ]. This distinction is apparently based on reports on binding of activated Rho GTPases, however the boundaries of each domain have not been defined experimentally. We propose that these two regions constitute a single domain for two reasons. First, when present, these two domains as defined in the Pfam database invariably appear adjacent to each other and are separated by only very few residues. Some cases of sequences where only a FH3 domain is present correspond to alternatively spliced variants of proteins that in their full length possess GBD and FH3 domains. This is the case for example of HsDRF3 [ 35 ]. Second, the GBD/FH3 domain appears as a single block in two formin-unrelated proteins of Dictyostelium , indicating that the domain was shuffled as a unit during remodeling of the genome (Fig. 4 ). The role of the GBD/FH3 domain appears to be twofold. On one hand the N-terminal region of formins is involved in subcellular localization through interaction with diverse targets. For instance, the N-terminus of yeast Fus-1 is responsible for recruitment to the projection tip during conjugation [ 14 ]. In mouse Dia3 the analogous region is required for localization at mitotic spindles [ 36 ]. In Dictyostelium an N-terminal fragment that encompasses most of the GBD/FH3 domain of ForC is sufficient for targeting to crowns and macropinosomes [ 20 ]. The N-terminus appears thus as a major determinant of localization and therefore function of formins. The low degree of sequence conservation of this region might correlate with the diversity of binding partners, not only Rho GTPases, and subcellular localization patterns described. On the other hand the GBD/FH3 domain is involved in regulation of activation by releasing of an intramolecular interaction between the DAD and the N-terminus, as initially proposed by Watanabe et al. [ 21 ]. As already mentioned, the boundaries of the GBD region remain poorly defined and while a CRIB-like (Cdc and Rac interactive binding) region has been described in mammalian DRF [ 37 ], such motif cannot be identified in any other formin, whether regulated by Rho GTPases or not. Although initially not appreciated [ 6 ], with very few exceptions a GBD/FH3, a DAD or both can be identified in almost all conventional formins, including all fungal formins, FMNL, FHOD and DAAM ([ 11 , 14 , 38 - 40 ] and Figs. 4 and 5 ). Although a FH3 domain was reported also in cappuccino and in one alternatively splice variant of mouse formin 1 (formin1 IV) [ 14 ], we were not able to identify a GBD/FH3 domain in these proteins. We cannot exclude that a strongly divergent GBD/FH3 be present in members of this subfamily. In fact, cappuccino interacts with activated RhoA [ 41 ], and the N and C-terminal segments of formin1 (IV) interact with each other [ 27 ], two features characteristic of conventinal formins. The designation Diaphanous-related formin has been applied to those formins that interact with activated Rho GTPases [ 7 ]. However, the number of formins shown to posses this property is increasing, and includes to date at least one member of each family of conventional formins as well as Dictyostelium formins (our unpublished data). We therefore propose the use of the name conventional formins for those subfamilies with the general structure GBD/FH3-FH1-FH2, although in particular members or in alternatively spliced variants a domain (but never the FH2) might be absent, indicating whether the protein is Rho-regulated where documented experimentally. The name Diaphanous-related formin should be restricted to the metazoan members of the Diaphanous subfamily, like human DRF1 to 3. Functionality and roles of Dictyostelium formins Functional data on Dictyostelium formins is scarce. Only three isoforms, formins A, B and C have been characterized to some extent. Mutants lacking ForA, ForB or both showed no detectable phenotype, whereas deletion of forC led to formation of aberrant fruiting bodies with short stalks and unlifted sori, suggesting this formin mediates actin remodeling during multicellular stages [ 20 ]. Dictyostelium formins are expected to be functional according to their highly conserved FH1-FH2 structure; therefore a certain degree of functional redundancy is expected. However, diversity might arise through specific targeting and activation by Rho GTPases conferred by the GBD/FH3 domain, through interaction of specific SH3-domain containing proteins with the FH1 domain and by virtue of unique additional domains. These issues need to be addressed experimentally in the future. ForC and ForI might be exceptions in terms of regulation. ForC lacks a FH1 domain and consequently does not bind to profilins [ 20 ]. Although the FH2 domain is necessary and sufficient for nucleation, FH2-induced nucleation is very slow and requires binding of profilin to the FH1 domain for full functionality [ 9 - 11 ]. While other scenarios are possible, in the case of ForC fueling of the actin polymerization process by profilin-actin might be furnished by heterodimerization with another formin possesing an FH1 domain. Regarding ForI, that lacks GBD/FH3 and DAD domains, it is not clear how this isoform could be regulated. Three Dictyostelium formins have domains at their N-termini that are not found in other formins and might confer unique additional functions or ways of regulation or targeting. The C2 and C1 domains of ForA and ForE, respectively, might regulate activation or targeting of the molecule through interaction with specific lipids [ 29 , 30 ], while the FHA domain of ForJ might be involved in interactions with components of the cell nucleus [ 31 ]. In general, well defined domains others than the ones characteristic of formins are very rare. Most plant class 1 formins carry transmembrane domains and proline-rich regions in their N-termini that together might mediate anchorage of actin nucleation sites to the cell wall across the plasma membrane [ 15 , 16 ] and the PTEN-related domain of some class 2 plant formins might also be involved in membrane anchoring [ 16 ]. Apart from Delphilin (see above) we have identified only one more case of additional domains in metazoan formins, CeZ22171. This protein, that also lacks a FH1 domain, has a zinc finger domain and might be involved in nucleic acid interactions. The C-terminal extensions found in ForJ and several other fungal and metazoan formins also probably harbor recognition sites for additional binding partners that remain to be identified. Functional diversity might also be related to different patterns of local and temporal gene expression. Our gene expression analyses also suggest specific roles during asexual and sexual development. Four genes in particular, forC, D, I and J , displayed an increase in expression during transition to multi-cellular stages. During this phase cells acquire aggregation competence in parallel with maturation of signaling pathways involved in remodeling of the cytoskeleton. At least for forC gene expression data correlate with a developmental role, as mentioned above [ 20 ]. ForH and ForI might play specific roles during sexual development, based alone on their patterns of gene expression. Interestingly, expression of rac1b and racF2 was found increased during the analysis of a gamete-enriched cDNA library [ 33 ]. It is therefore conceivable that one or more formins, irrespective of their expression pattern, play roles during sexual development upon activation by those GTPases. Conclusion The social amoeba Dictyostelium discoideum expresses 10 formins that with few exceptions conform to the domain structure GBD/FH3-FH1-FH2-DAD. This arhitecture and the high degree of conservation of the FH2 domain allow some preliminary conclusions about the functionality of Dictyostelium formins: all isoforms may display actin nucleation activity and, with the exception of ForI, may also be susceptible to autoinhibition and to regulation by Rho GTPases. Although functional redundancy may be expected to occur to some extent among Dictyostelium formins, specific roles may be conferred by the GBD/FH3 domain, which is less well conserved than the FH2 domain, and by specific patterns of gene expression during asexual and sexual development. We propose four major classes of formins based on a comparison of the domain composition of proteins from diverse phyla. Dictyostelium , fungal and most metazoan formins can be grouped within the class of what we designate conventional formins, characterized by the structure GBD/FH3-FH1-FH2-DAD. The GBD and FH3 domains, whose boundaries had not been defined previously, probably constitute a single domain. The architecture shared by conventional formins implies a common regulatory mechanism based on autoinhibition through intramoleculr interaction of the GBD/FH3 and the DAD domains and activation through release of this interaction upon binding of Rho GTPases. Formins of the other classes (plant formins and Delphilin) lack GBD/FH3 and DAD domains and must therefore have other mechanisms of activation. Note. While our manuscript was under review a phylogenetic analysis of the FH2 domain by H. N Higgs and K. J. Peterson has been published. These authors used a larger set of FH2 domains that includes only three formins from Dictyostelium . The topology of the phylogenetic tree described in that article and that of our tree are essentially coincident, and all seven metazoan groups identified by their authors can be found in our tree, with the novel subfamily INV comprising our HsKIAA1727, DmAE003560 and CeAF106580 sequences. Higgs and Peterson, however, do not recognize the GBD/FH3 region as a domain present in a larger number of formin subfamilies. Methods Sequence analysis The amino acid or DNA sequences of Dictyostelium formins were used as query for BLAST searches [ 42 ] of the Dictyostelium genome project databases at The Welcome Trust Sanger Institute, Baylor College of Medicine, The University of Cologne and the Department of Genome Analysis of the Institute of Molecular Biotechnology in Jena. Nearly all of this data was generated at the aforementioned institutes with a small part of it produced at the Institute Pasteur. After assembly of the genome further analyses were performed through the Dictybase server [ 43 ]. BLAST searches against EST sequences were performed at NCBI [ 44 ]. Accession numbers for Dictyostelium formins can be found in Table 1 . Accession numbers of the sequences retrieved for phylogenetic analyses are as follows. S. cerevisiae Bni1p, P41832; Bnr1p, P40450; S. pombe Fus1, L37838; Cdc12, 786133; For3, AL035247. D. melanogaster Cappuccino, U34258; Diaphanous, U11288; FHOD, AE003554; FMNL, BT003654; DAAM, AAF45601; a novel formin, AE003560. C. elegans FHOD, U88314; Cyk-1, U40187; FMNL, AC024798; novel formins, Z78013, AF106580 and Z22174. H. sapiens Formin 1, AK127078; Formin 2, XM_351329; FHOD1, AF113615; FHOD3/FHOS2, KIAA1695; DRF1, AF05187; DRF2, Y15909; DRF3, BC034952; Delphilin, XM_353725; FMNL1, AF432213; FMNL2/FHOD2, KIAA1902; WBP3/FMNL3, NM_175736; DAAM1, NM_014992; DAAM2, AL833083; a novel formin KIAA1727. Sequences of plant formins were obtained from Cvrčková et al. [ 16 ]. Dictyostelium RasGEF-L and RasGEF-V can be accessed at Dictybase [ 43 ] under DDB0217789 and DDB0216586, respectively. Protein sequences were aligned using the ClustalX [ 45 ] program with a BLOSUM62 matrix and default settings, followed by manual edition with the Bioedit program [ 46 ]. Phylogenetic trees were constructed using the neighbor-joining algorithms of the ClustalX program with correction for multiple substitutions; positions with gaps were not excluded. Construction of trees was done with TreeView [ 47 ]. Bootstrap analysis (1000 bootstraps) was applied to provide confidence levels for the tree topology. The domain analysis was done using the SMART tool [ 48 ] and InterProScan [ 49 ]. FH1 domains were identified by visual inspection. GBD/FH3 and DAD domains were identified in part by inspection of multiple alignments. Cell culture D. discoideum AX2 strain was grown at 21°C in shaking suspension in axenic HL5 medium [ 50 ]. AX2 cells were also cultured for sexual gametes in Bonner's salt solution (BSS) as described [ 33 ]. In brief, cells were cultured for 15 hours in a dense suspension of K. aerogenes in BSS either in the darkness or in the light. Cells cultured for 15 hours in the dark become fusion-competent cells. However, cells cultured for 15 hours in the light condition exhibit reduced fusion competency, and are designated light submerged cells. Cells on SM agar plates [ 50 ] are fusion incompetent cells. Isolation of total RNA and quantitative real-time PCR Total RNA was purified from both asexually and sexually developing cells with the TRIZOL reagent (GIBCO BRL, USA). Asexually developing cells on phosphate agar plates [ 50 ] were collected every 4 hours. Total RNA was treated with RNase-free DNase to remove contaminating genomic DNA, and then used to synthesize the first strand cDNA using SuperscriptII (Invitrogen, USA). For each time point cDNA was synthesized using two independently isolated mRNA samples. Specific primer sets for each formin gene were designed. To equalize the concentrations of template cDNAs, amplification was conducted using the control primer set for the Ig7 gene, which is expressed constitutively. Quantitative real-time PCR was performed with an ABI 7900HT Sequence Detection System according to the manufacturer's instructions. The amplifications were carried out using Ex Taq R-PCR Version (Takara Bio, JAPAN) and SYBR Green I. Each sample had 2 replicates containing 1-, 4-, or 16-fold diluted cDNA. Miscellaneous methods For RT-PCR, first strand cDNA synthesis was performed with M-MLV reverse transcriptase (Promega Corporation, Madison, WI) on poly A+ mRNA purified with the Oligotex system (Qiagen GmbH, Hilden, Germany) from total RNA. PCR fragments were cloned into the pGEM-T Easy vector system (Promega Corporation, Madison, WI) and sequenced. DNA sequencing was done at the service laboratory of the Center for Molecular Medicine, Cologne, using an automated sequencer (ABI 377 PRISM, Perkin Elmer, Norwalk, CO). Authors' contributions FR conceived the study, performed the assembly of genomic sequences and the sequence alignments and drafted the manuscript. TM and HU carried out the gene expression studies. AKM performed RT-PCR and cloning. CK and TQPU participated in the design of the study and the assembly of genomic sequences. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC555941.xml

555955

Emergence of unusual species of enterococci causing infections, South India

Background Enterococci tend to be one of the leading causes of nosocomial infections, with E. faecalis and E. faecium accounting up to 90% of the clinical isolates. Nevertheless, the incidence of other species of enterococci from clinical sources shows an alarming increase with the properties of intrinsic resistance to several antibiotics including beta-lactams and glycopeptides. Thus proper identification of enterococci to species level is quintessential for management and prevention of these bacteria in any healthcare facility. Hence this work was undertaken to study the prevalence of unusual species of enterococci causing human infections, in a tertiary care hospital in South India. Methods The study was conducted in a tertiary care hospital in South India from July 2001 to June 2003. Isolates of enterococci were collected from various clinical specimens and speciated using extensive phenotypic and physiological tests. Antimicrobial susceptibility testing were performed and interpreted as per NCCLS guidelines. Whole cell protein (WCP) fingerprinting of enterococci were done for species validation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed computationally. Results Our study showed the prevalence of unusual (non-faecalis and non-faecium enterococci) and atypical (biochemical variant) species of enterococci as 19% (46 isolates) and 5% (12 isolates) respectively. The 7 unusual species (46 isolates) isolated and confirmed by phenotypic characterization includes: 15 E. gallinarum (6.2%), 10 E. avium (4.1%), 6 E. raffinosus (2.5%), 6 E. hirae (2.5%), 4 E. mundtii (1.7%), 3 E. casseliflavus -including the two atypical isolates (1.2%) and 2 E. durans (0.8%). The 12 atypical enterococcal species (5%) that showed aberrant sugar reactions in conventional phenotyping were confirmed as E. faecalis, E. faecium and E. casseliflavus respectively by WCP fingerprinting. The antimicrobial susceptibility testing depicted the emergence of high-level aminoglycoside and beta-lactam resistance among different species apart from intrinsic vancomycin resistance by some species, while all the species tested were susceptible for linezolid and teicoplanin. Conclusion Our study reveals the emergence of multi-drug resistance among unusual species of enterococci posing a serious therapeutic challenge. Precise identification of enterococci to species level enables us to access the species-specific antimicrobial resistance characteristics, apart from knowing the epidemiological pattern and their clinical significance in human infections.

Background Enterococci, generally regarded as normal flora of gastrointestinal and genitourinary tract of humans, have emerged as the etiogen of several nosocomial as well community-acquired infections since last two decades. Globally, many studies have revealed that enterococci tend to be one of the leading causes of several nosocomial infections, with the emergence and spread of multi drug resistance among isolates [ 1 - 3 ]. Since the inception of separate genus Enterococcus, there are 23 species of enterococci with clinical significance to date [ 4 ], of which Enterococcus faecalis and Enterococcus faecium accounts up to 90% of clinical isolates belonging to this genus [ 1 ]. Nevertheless, the incidence of other species of enterococci from clinical sources shows an alarming increase with the properties of intrinsic resistance to several antibiotics including beta-lactams and glycopeptides [ 5 , 6 ]. But the incidence of non-faecalis and non-faecium enterococci is underestimated because of frequent misidentification. On several instances only one phenotypic character differentiates one species from another, and to further complicate some strains of enterococci do not posses the exact phenotypic character of the type strains, and there comes confusion over their exact taxonomic status [ 7 ]. Thus proper identification of enterococci to species level is quintessential for management and prevention of these bacteria in any health care facility. Many studies focus on the two most common species E. faecalis and E. faecium , and only few reports or studies of non-faecalis and non-faecium enterococci are prevalent [ 5 , 6 ]. Hence the aim of our study was to check the prevalence of unusual and atypical species of enterococci causing human infections, in a tertiary care hospital in South India over a time period. Methods i. Bacterial isolates and conventional phenotypic characterization of enterococci The study was conducted in a 900-bedded tertiary care hospital at Pondicherry, South India from July 2001 to June 2003. Isolates of enterococci were collected over the time period from various clinical specimens such as blood, urine, wound swabs and pus (surgical and non-surgical), catheters, ascitic fluid, synovial fluid, by plating them on 5% Sheep Blood agar and Mac-conkey agar, as well on Bile esculin azide agar (Hi-media, Mumbai, India) as per nature of the specimen. Extensive phenotypic and physiological characterization was carried out by the conventional tests devised by Facklam et al [ 3 , 8 ]. Carbohydrate fermentation tests were performed using 1% sugar discs in Brain heart infusion (BHI) broth with Andrade's indicator (Hi-media, Mumbai, India) as per manufacturer's instructions. The following sugars were tested for fermentation by isolates using commercial discs: mannitol, sorbitol, inulin, arabinose, melibiose, sucrose, raffinose, trehalose, lactose, glycerol, salicin, maltose, adonitol, and xylose, while sorbose and ribose were added to a final concentration of 1% to the broth base directly after sterilization (due to non-availability of discs). Group D antigen was detected using a commercial latex agglutination kit (The Binding site limited, Birmingham, B29 6AT) as per manufacturer's™ instructions. ii. Antimicrobial susceptibility testing Antibiotic susceptibility testing of the clinical isolates along with the quality control strains were performed using BHI agar instead of Muller Hinton agar by disk diffusion method (for the antibiotics: penicillin [10 units], ampicillin [10 μg], gentamicin-high content [120 μg], streptomycin-high content [300 μg], ciprofloxacin [5 μg], nitrofurantoin-for urinary isolates only [300 μg], vancomycin [30 μg], teicoplanin [30 μg] and linezolid [30 μg]), standard agar dilution (for the antibiotics mentioned in Table- 1 ) and agar screening methods (for vancomycin and high-level aminoglycoside resistance) and interpreted as per NCCLS guidelines [ 9 ]. Production of β-lactamase was determined by using nitrocefin discs (BBL Microsystems) as per manufacturer's™ instructions. Table 1 Analysis of MIC ranges of unusual species of enterococci. Species tested (no.of.isolates) Antibiotic Tested No. of isolates at specified MIC, in μg/mL Susc a ,% 2 4 8 16 32 ≥ 64 E. avium (10) Pen. 9 9 9 9 7 6 10 Amp. 9 6 6 6 6 0 40 Van. 4 2 0 0 0 0 100 Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 100 HLGm. NA NA NA NA NA NA 10 HLStr. NA NA NA NA NA NA 50 E. casseliflavus (3) Pen. 0 0 0 0 0 0 100 Amp. 0 0 0 0 0 0 100 Van. 3 3 0 0 0 0 NA Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 66.6 HLGm. NA NA NA NA NA NA 100 HLStr. NA NA NA NA NA NA 100 E. durans (2) Pen. 2 1 1 1 0 0 50 Amp. 1 1 1 1 0 0 50 Van. 2 1 0 0 0 0 100 Te. 1 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 50 HLGm. NA NA NA NA NA NA 50 HLStr. NA NA NA NA NA NA 0 E. gallinarum (15) Pen. 9 9 8 8 8 8 46.6 Amp. 8 8 8 8 7 6 46.6 Van. 13 9 2 0 0 0 NA Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 53.3 HLGm. NA NA NA NA NA NA 46.6 HLStr. NA NA NA NA NA NA 66.6 E. hirae (6) Pen. 3 3 2 2 0 0 66.6 Amp. 6 2 2 2 0 0 66.6 Van. 0 0 0 0 0 0 100 Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 100 HLGm. NA NA NA NA NA NA 100 HLStr. NA NA NA NA NA NA 100 E. mundtii (4) Pen. 1 1 1 0 0 0 100 Amp. 0 0 0 0 0 0 100 Van. 2 2 0 0 0 0 100 Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 50 HLGm. NA NA NA NA NA NA 100 HLStr. NA NA NA NA NA NA 50 E. raffinosus (6) Pen. 4 4 4 4 4 4 33.3 Amp. 4 4 4 4 4 0 33.3 Van. 0 0 0 0 0 0 100 Te. 0 0 0 0 0 0 100 Va.Scr. NA NA NA NA NA NA 100 HLGm. NA NA NA NA NA NA 66.6 HLStr. NA NA NA NA NA NA 66.6 NOTE: a Interpretations based on NCCLS guidelines, NA-not applicable; Susc- Susceptiblity, Pen-Penicillin; Amp-Ampicillin; Van-Vancomycin; Te-Teicoplanin, Va. Scr- Vancomycin resistance (6 μg/mL) agar screening, HLGm- High-level gentamicin resistance (500 μg/mL) agar screening, HLStr- High-level streptomycin resistance (2000 μg/mL) agar screening, iii. Molecular phenotyping of enterococci Whole cell protein (WCP) analysis of the enterococcal isolates, including atypical biochemical variants of enterococci and the reference/type strains of enterococci (a kind gift from Dr. Richard.R.Facklam, CDC, Atlanta, GA. USA) were done using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) as described previously with minor modifications [ 10 , 11 ] for species identification, as well confirmation of species identities of atypical strains. Briefly, enterococcal test strains were grown for 18 hours at 37°C on Trypticase soy agar with 5% sheep blood. The samples were prepared by removing the bacterial growth from the surface of agar plate carefully with a sterile disposable loop, and suspended in 5 ml of sterile saline solution in order to obtain a turbidity equal to that of No.8 MacFarland density standard, centrifuged, and resuspended in 0.5 ml of an aqueous lysozyme solution (10 mg/ml). The suspensions were incubated in a water bath preset at 37°C for 2 hours. The WCP extracts were obtained by mixing one part of whole cell extract to one part of sample loading buffer, and boiled for 5 minutes and separated by SDS-PAGE along with a broad range molecular weight marker (New England Biolabs Inc.,) as per standard procedure [ 11 ]. The SDS-PAGE was performed using 5% stacking gel and 10% separating gel at a constant current of 20 mA using a mini-gel electrophoresis system (Bangalore Genei, India) and stained with Coomassie brilliant blue. Visual comparisons of the gels were made, and documentation done using a Gel doc system (Vilber loubert, France) for further analysis. The gel images were analyzed and dendrogram constructed using appropriate software (Bionumerics, version 2.5, Applied Maths, Kortrijik, Belgium) for validating the taxonomy of the enterococcal species studied. Results Conventional and Molecular phenotyping of enterococci We isolated a total of 242 enterococci during our 2-year study period from different clinical samples. The biochemical phenotyping results revealed 46 isolates (19%) belonging to 7 different unusual species of enterococci (excluding E. faecalis and E. faecium -data not shown) which included 15 E. gallinarum (6.2%), 10 E. avium (4.1%), 6 E. raffinosus (2.5%), 6 E. hirae (2.5%), 4 E. mundtii (1.7%), 3 E. casseliflavus (1.2%) and 2 E. durans (0.8%). The distribution by site of isolation for the 46 unusual enterococcal species included 30 isolates- 12 E. gallinarum , 6 E. avium , 3 each of E. hirae , E. casseliflavus and E. raffinosus , 2 E. mundtii and 1 E. durans (65.2%) from bloodstream, 6 isolates- 3 E. raffinosus , 2 E. avium and 1 E. mundtii (13 %) from surgical and non-surgical wound swabs, 10 isolates- 3 each of E. hirae and E. gallinarum , 2 E. avium and 1 each of E. durans and E. mundtii (21.8%) from miscellaneous sites, including muscle tissues sent for anaerobic culture, catheter tips, peritoneal fluid, ear swab and urine. Of the 46 persons from whom unusual enterococci were obtained, 56.5% were males and 43.5% were females including newborn/neonates. The infections were polymicrobial in 6 (13%) of the 46 cases from which unusual enterococci were isolated, including 2 (6.7%) of 30 bloodstream infections. The 12 atypical enterococcal strains (5%) showing aberrant sugar reactions included 6 mannitol negative variant E. faecalis like species, 1 arginine negative variant E. faecalis like species, 3 mannitol negative variant E. faecium like species and, 2 arginine negative variant E. casseliflavus like species. The WCP analysis by SDS-PAGE confirmed the species identities of seven different species. Atypical strains showed a similar banding pattern like the reference strains from CDC ( E. faecalis SS-1273, E. faecium SS-1274, E. casseliflavus SS-1229) except for minor quantitative differences, with no qualitative difference. The computational analysis of the WCP gel images of atypical strains were performed by Dice coefficient, and the dendrogram constructed using unweighted pair group method using arithmetic averages (UPGMA) as shown in Figure- 1 , and validated their exact taxonomic status as E. faecalis, E. faecium and E. casseliflavus respectively. The 2 (atypical) isolates of arginine negative variant E. casseliflavus like species after taxonomic validation were included as an unusual species of enterococci accounting to 3 E. casseliflavus isolated overall. Figure 1 Cluster analysis of atypical strains of Enterococci using Dice coefficient and UPGMA method (Bionumerics, Applied Maths, Belgium). Note : SS- Designation of CDC standard strains, E. porcinosus is currently designated as E. villorum , E. pseudoav.- E. pseudoavium ; E. malodorat.- E. malodoratus , E. casselifla.- E. casseliflavus , MNV- Mannitol negative variant; ANV- Arginine negative variant. Antimicrobial susceptibility testing The antimicrobial susceptibilities of the isolates given in Table- 1 depict the ranges of MICs for various antimicrobial agents tested by standard agar dilution, and agar screening methods. The E. gallinarum and E. casseliflavus isolates showed reduced susceptibility to lesser concentrations of vancomycin ranging 2–8 μg/ml. Other species were highly susceptible for vancomycin and teicoplanin except one isolate of E. durans . High-level aminoglycoside resistance for gentamicin and streptomycin was found absent only in E. casseliflavus and E. hirae , while other species exhibited variable susceptibilities ranging 0 – 66.7% for either aminoglycoside tested. The disk diffusion testing showed 100% susceptibility for linezolid and teicoplanin by all isolates tested, while E. casseliflavus and E. mundtii showed 100% susceptibility for penicillin and ampicillin. Only 37% of unusual enterococcal isolates were susceptible to ciprofloxacin, with resistance exhibited by 9 E. avium (n = 10), 2 E. durans (n = 2), 11 E. gallinarum (n = 15), 2 E. hirae (n = 6), 2 E. mundtii (n = 4) and, 3 E. raffinosus (n = 6). Only E. casseliflavus (n = 3) exhibited 100% susceptibility to ciprofloxacin. None of the 46 isolates was positive for β-lactamase, but resistance for β-lactam agents were prevalent variably among different species. The results of MICs for penicillin, ampicillin, high-level gentamicin and high-level streptomycin resistance were in accordance with the disk diffusion testing results except for vancomycin. Disk diffusion testing showed vancomycin resistance for 6 isolates (1 E. durans , 2 E. mundtii , 3 E. gallinarum ), but the agar screening method exhibited vancomycin resistance for 11 isolates (2 E. mundtii , 1 E. casseliflavus , 1 E. durans , 7 E. gallinarum )(including 5 isolates-4 E. gallinarum and 1 E. casseliflavus , which showed susceptibility to vancomycin by the disk diffusion method). Discussion Our study reveals that the prevalence of unusual species of enterococci as 19% in our clinical setup in South India. Many studies and reviews show the prevalence of non-faecalis and non-faecium enterococci as 2–10% [ 3 , 6 , 12 ]. Previous studies from India have reported E. faecalis and E. faecium as the only prevalent species [ 13 - 16 ], which may not reflect the true incidence rate. From our perspective the real incidence tends to be higher which in part can be explained as, misidentification of species due to exhibition of aberrant sugar reactions by some enterococci or, due to lack of application of the complete range of tests to identify non-faecalis and non-faecium enterococci [ 7 , 17 ]. The prevalence rate (19%) of our study was partly in accordance with another Indian study [ 18 ] showing 14.8% (excluding E. faecalis and E. faecium ) prevalence of unusual species of enterococci from catheterized patients with urinary tract infections. E. mundtii and E. durans were not reported in their study, whose prevalence was 1.7% and 0.8% respectively in our study. E. gallinarum (6.2%) and E. avium (4.1%) were the most commonly identified species, which markedly differs in isolation rate (0.3–1.2%) from other studies [ 6 , 19 , 20 ]. The incidence of infections caused by unusual enterococcal species is of serious concern, since 43.5% of the isolates were from cases of septicemia without endocarditis. Apart from septicemia, the unusual species of enterococci were isolated frequently from cases of urinary tract infections, surgical and non-surgical wound infections and peritonitis. Most of the patients with the bloodstream infections had a peripheral or central catheter. Further, only 13% of enterococcal infections were polymicrobial, with majority from non-bloodstream isolates that underscores the clinical significance of these unusual enterococcal species. Although the unusual species of enterococci were isolated at regular intervals throughout our study period, we could find clustering of specific species during a specific time period from specific units/wards. Interestingly, 10 among the 15 isolates of E. gallinarum isolated during our study period were from pediatrics unit, while 7 of the 10 isolates exhibiting a similar antibiotype were isolated from the same ward within a span of 2 months. The remaining 3 of the 10 E. gallinarum were isolated from the same ward in the preceding 3 months, one of which showed an antibiotype similar to the cluster of 7 isolates. The same was the case of 3 E. casseliflavus isolated from the same pediatrics unit within a span of 2 months in the preceding year. Most of these (8 of 10 E. gallinarum , and all 3 E. casseliflavus ) isolates were from cases of septicemia. Although molecular epidemiological studies have not been done to compare the genetic similarities of these isolates, the data depicts the nosocomial spread of these species. WCP analysis by SDS-PAGE had been proven to assist in validating the species identities as well, to identify strains that do not exhibit phenotypic characteristics identical to the type strains of each species [ 4 , 10 , 21 ]. We were able to validate the authenticity of the unusual species, and the exact taxonomic status of the atypical phenotypic variant strains identified by conventional biochemical testing as shown in Figure- 1 , using WCP fingerprinting by SDS-PAGE. Ciprofloxacin resistance was 63% among isolates (excepting E. casselifalvus ) which proves that it may be successful only in treating enterococcal urinary tract infections [ 1 , 9 ], since most of our isolates were from bloodstream and other related specimens. None of the isolates were β-lactamase producer, but penicillin and ampicillin resistance were exhibited by 54.3% and 45.7% isolates. We suggest penicillin binding protein modification based resistance for our isolates as a basis for β-lactam resistance, as depicted previously [ 22 , 23 ], but markedly differs from other Indian studies [ 15 , 24 ] showing up to 50% β-lactamase associated resistance. The prevalence of high-level gentamicin resistance (43.4%) and high-level streptomycin (37%) among unusual enterococcal isolates from our study partially correlates with studies from Japan [ 25 ] and United States [ 12 , 26 ]. In our study, most strains with high-level gentamicin resistance lacked high-level streptomycin resistance, and vice versa, thus facilitating the combination therapy (cell wall inhibitor plus aminoglycoside) treatment options for serious enterococcal bloodstream infections [ 1 , 9 ]. The prevalence of vancomycin resistance was 24% by agar screening /agar dilution method and 13% by disk diffusion. The difference may be attributed to the intrinsic low level vancomycin resistance (van C genotype), exhibited by 4 E. gallinarum and, 1 E. casseliflavus isolates, which may go undetected by disk diffusion testing [ 27 ]. Of serious concern was the low-level vancomycin resistance exhibited by one E. durans and two E. mundtii (MIC ≤ 6 μg/ml). The genotypic basis of vancomycin resistance for these 3 isolates yet to be studied, will give us a definitive picture regarding its clinical significance, since studies have reported the prevalence of vancomycin resistance in these two species, and its transferable nature from E. durans to E. faecium [ 28 - 30 ]. Conclusion Precise identification of enterococci to species level enables us, to access the species-specific antimicrobial resistance characteristics, apart from knowing the epidemiological pattern and their clinical significance in human infections. The difficulty in detecting (intrinsic) low-level vancomycin resistance by disk diffusion testing [ 28 ] emphasizes the necessity for including agar screening methods as per NCCLS guidelines in routine susceptibility testing of all enterococci isolated from clinical specimens [ 9 ]. Further as shown in our study, the increase in the rate of prevalence of the unusual and atypical species and the emergence of multidrug resistance among them, highlights the significance of rapid and accurate identification of enterococci to the species level for initiating appropriate therapeutic regimen, and reemphasizes the importance of the implementation of appropriate infection control measures to limit the nosocomial spread of these unusual species in any nosocomial setting. Competing interests The author(s) declare that they have no competing interests. Authors' contributions PVP designed the study and carried out the experimental works and analysis, and drafted the manuscript. RSR supervised and participated in the design of the study and coordination, and helped to draft the manuscript. SCP participated in the coordination of the study and helped to draft the manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC555955.xml

529450

Age-related changes in Serum Growth Hormone, Insulin-like Growth Factor-1 and Somatostatin in System Lupus Erythematosus

Background Systemic lupus erythematosus is an age- and gender-associated autoimmune disorder. Previous studies suggested that defects in the hypothalamic/pituitary axis contributed to systemic lupus erythematosus disease progression which could also involve growth hormone, insulin-like growth factor-1 and somatostatin function. This study was designed to compare basal serum growth hormone, insulin-like growth factor-1 and somatostatin levels in female systemic lupus erythematosus patients to a group of normal female subjects. Methods Basal serum growth hormone, insulin-like growth factor-1 and somatostatin levels were measured by standard radioimmunoassay. Results Serum growth hormone levels failed to correlate with age (r 2 = 3.03) in the entire group of normal subjects (i.e. 20 – 80 years). In contrast, serum insulin-like growth factor-1 levels were inversely correlated with age (adjusted r 2 = 0.092). Of note, serum growth hormone was positively correlated with age (adjusted r 2 = 0.269) in the 20 – 46 year range which overlapped with the age range of patients in the systemic lupus erythematosus group. In that regard, serum growth hormone levels were not significantly higher compared to either the entire group of normal subjects (20 – 80 yrs) or to normal subjects age-matched to the systemic lupus erythematosus patients. Serum insulin-like growth factor-1 levels were significantly elevated (p < 0.001) in systemic lupus erythematosus patients, but only when compared to the entire group of normal subjects. Serum somatostatin levels differed from normal subjects only in older (i.e. >55 yrs) systemic lupus erythematosus patients. Conclusions These results indicated that systemic lupus erythematosus was not characterized by a modulation of the growth hormone/insulin-like growth factor-1 paracrine axis when serum samples from systemic lupus erythematosus patients were compared to age- matched normal female subjects. These results in systemic lupus erythematosus differ from those previously reported in other musculoskeletal disorders such as rheumatoid arthritis, osteoarthritis, fibromyalgia, diffuse idiopathic skeletal hyperostosis and hypermobility syndrome where significantly higher serum growth hormone levels were found. Somatostatin levels in elderly systemic lupus erythematosus patients may provide a clinical marker of disease activity in these patients.

Background Systemic lupus erythematosus (SLE) is the protean autoimmune disorder with strong familial penetrance. Immunologically, SLE is characterized by aberrations in T cell and B cell function [ 1 , 2 ], over-production of autoantibodies directed principally against nuclear antigens [ 3 ] as well as other tissue antigens, and deficiences in the complement system [ 4 ]. SLE is predominantly a disease of young females with peak incidence occuring between 20 and 40 yrs with a female to male ratio of 6–10:1 [ 5 ]. Although many of the principal pathophysiological changes associated with SLE indicate organ involvement consistent with vascular inflammation and immune complex deposition [ 6 ], several prominent SLE-related pathologic findings suggest systemic disturbances consistent with metabolic abnormalities [ 7 ]. However, surrogate blood or serum markers of systemic dysfunction such as erythyrocyte sedimentation rate and C-reactive protein levels, although frequently elevated in SLE compared to normal subjects are often uninformative and unreliable as surrogate markers of SLE disease activity [ 7 ]. We have shown that diverse rheumatic and musculoskeletal disorders, including osteoarthritis (OA) [ 8 - 12 ], diffuse idiopathic skeletal hyperostosis (DISH) [ 12 , 13 ] and hypermobility syndrome [ 14 ] as well as fibromyalgia [ 15 ] were characterized, in part, by elevated serum growth hormone levels. Growth hormone was also found sequestered in erythrocytes in OA and DISH patients at levels that significantly exceeded serum growth hormone levels [ 16 ] suggesting a putative mechanism by which "toxic" levels of growth hormone could be confined, or in cases of vascular inflammation, transported to joint synovial fluid or peripheral end-organs [ 11 , 16 ]. Further, medical therapy of OA and DISH principally with non-steroidal anti-inflammatory drugs (NSAIDs) which resulted in pain suppression and reduced stiffness as well as improved range of motion correlated with lower serum growth hormone levels consistent with levels found in normal subjects [ 10 , 13 ]. More recently, we showed that symptomatic rheumatoid arthritis (RA) patients were also characterized by elevated serum growth hormone levels [ 17 ], but treatment of RA with prednisone failed to significantly lower serum growth hormone levels. Insulin-like growth factor-1 (IGF-1) synthesis is coupled to growth hormone via its capacity to stimulate hepatocyte IGF-1 production [ 11 ]. In several rheumatic and musculoskeletal disorders, elevated serum growth hormone was correlated with elevated IGF-1 levels [ 9 , 11 , 13 , 14 ] with OA [ 8 - 10 , 12 ] and RA [ 17 ] being notable exceptions. In the case of OA, IGF-1 levels are significantly lower compared to normal control subjects [ 8 - 10 , 12 ]. However, medical therapy of OA principally with NSAIDs resulted in growth hormone and IGF-1 levels approaching normal [ 10 ] whereas in DISH patients treated with NSAIDs, reduced serum growth hormone levels failed to result in concomitant changes in IGF-1 [ 13 ]. Somatostatin is a 14 amino acid polypeptide whose principal function is to regulate growth hormone release from the pituitary [ 18 ]. Elevated serum and synovial fluid somatostatin levels have been associated with inflammatory responses [ 19 ] most notably in RA [ 20 ]. A recent study showed that patients with symptomatic RA were, in part, characterized by a skewed upward serum growth hormone to somatostatin ratio [ 17 ]. The present study was performed to determine the extent to which serum growth hormone, IGF-1 and somatostatin levels were modulated in patients with SLE. A linear regression analysis was performed to determine the relationship between age and serum growth hormone and IGF-1 levels in a group of normal female subjects so that these values could be employed for comparison to a group of predominantly young, female SLE patients. Methods All studies were performed at University Hospitals of Cleveland (UHC) and the Wade Park Veterans Administration Medical Center (VAMC), Cleveland, Ohio. The UHC and VAMC Institutional Review Boards approved the study design with the research protocol, which included informed consent, being in keeping with the Declaration of Helsinki. Normal subjects and SLE patients were all volunteers. SLE Patients met the clinical and laboratory criteria for the diagnosis of SLE according to previously published classifications [ 21 ]. Patients with co-morbid conditions such as diabetes mellitus or hyperglycemia were excluded from the normal subject group as was any normal individual with evidence for rheumatic disorders in family members. This information was obtained by questioning potential normal subjects. Blood drawn by venipuncture was clotted at room temperature, centrifuged, serum aliquots separated and stored at -70°C until assayed. Blood samples were generally collected during an identical 3–4 hr morning period to normalize the potential contribution of growth hormone pulses and serum glucose levels to serum growth hormone determinations [ 8 - 10 ]. Serum samples were included for serum growth hormone, IGF-1 or somatostatin determinations only if glucose levels measured by the highly sensitive hexokinase assay [ 8 - 10 ] were between 65 and 135 mg/dl attained either by overnight fasting or a fast of at least 4 hours or more. Insulin levels were measured as previously described [ 8 - 10 ]. An insulin level in the range of 5–27 μU/ml was considered normal. Basal serum growth hormone and IGF-1 levels were determined by standard radioimmunoassay (RIA) (INCSTAR, Stillwater, MN) as previously described [ 8 - 10 ]. The lower limit of detection for serum growth hormone by the RIA was 0.4 ng/ml [ 8 - 10 ]. Serum growth hormone levels in samples falling at or below the lower limit of detection were excluded from the statistical analysis. Basal somatostatin levels in serum of 112 normal subjects and 55 SLE patients stratified by age (i.e. <45, between 45 and 55 yrs and >55 yrs of age) were separately measured by RIA [ 17 ]. The 2-tailed T-test was employed to analyze the differences in means of serum growth hormone, IGF-1 and somatostatin concentrations in groups of unequal size where p < 0.05 was significant. The population sample size was sufficient to detect a 20% difference in serum growth hormone and IGF-1 levels between control subjects and SLE patients and a 15% difference in serum somatostatin levels. The relationship between serum growth hormone and IGF-1 levels as a function of age was analyzed from scatter plots by linear regression analysis employing SPSS 11.1 (SPSS, Inc., Chicago, IL) and SigmaPlot 8.0 (SPSS, Inc.) to calculate the adjusted r 2 -value and regression line, respectively. Results and discussion Glucose and insulin concentration was determined in serum from normal female subjects and from patients with SLE. No normal female subjects were excluded from the study as a result of detecting hyperglycemia or hyperinsulinemia However, 2 SLE patients were excluded from the statistical analysis on this basis (data not shown). Over the course of this study, SLE patients received medical therapy with NSAIDs, prednisone (10–60 mg/day), hydroxychloroquine sulfate or methotrexate as well as combinations of these drugs. No SLE patients were treated with azathioprine or cyclophosphamide during this study. Previously it was shown that basal serum growth hormone levels among normal male and female subjects did not significantly differ on the basis of age [ 16 ]. As noted, SLE has a high female to male prevalence ratio and is predominant in young females between 20 and 40 yrs of age [ 5 , 21 ]. Thus, it was critical to determine the extent to which serum growth hormone and IGF-1 differed among female normal subjects on the basis of age. Serum growth hormone levels did not correlate with age in normal female subjects between the ages of 20 and 80 (Figure 1A ). However, a strong inverse correlation between age and IGF-1 levels (adjusted r 2 = 0.269) in this group of normal female subjects was found (Figure 1B ). In contrast to the results obtained from basal serum growth hormone measurements in the entire normal female subject population (Figure 1A ), a strong direct correlation (adjusted r 2 = 0.092) between age (age, 30.8 ± 7.0, mean ± SD; 95% confidence, 3.74) and basal serum growth hormone levels in the young female normal subjects was found (Figure 2A ). However, the correlation between age and basal serum growth hormone levels was weak in the older (age, 60.6 ± 9.4; mean ± SD; 95% confidence, 3.29) normal female subjects (Figure 2B ). Based on the above considerations, basal serum growth hormone and IGF-1 concentration was determined in study groups subdivided by age in normal female subjects and these values compared with basal serum growth hormone and IGF-1 levels in SLE patients. The results showed that SLE was not characterized by elevated serum growth hormone whether or not all normal female subjects or age-matched normal female subjects were employed as the comparison group (Table 1 ). Serum IGF-1 levels were significantly lower in the normal female subject group compared to SLE patients (Table 1 ), but there was no significant difference if serum IGF-1 levels in the SLE group were compared to serum IGF-1 levels in the age-matched normal female group (Table 1 ). A trend towards elevated somatostatin levels in normal subjects as a function age was previously found [ 17 ]. In the present study, there was also a trend towards elevated serum somatostatin levels in the <45 yr old SLE patient group or 45 – 55 yr old group compared to their age-matched normal counterparts (Table 1 ). However, a significant difference was found only in the older (>55 yrs) SLE patients compared to their age-matched control counterparts (Table 1 ). The results of the present study emphasized the critical requirement to control for age and gender when basal serum growth hormone and IGF-1 levels in normal subjects are compared to patients with autoimmune musculoskeletal diseases which, like SLE, are characterized by a strong age and gender association. Several studies from our laboratory have consistently shown basal serum growth hormone to be higher in females than in males [ 8 , 9 , 14 ]. One study, in particular, examined the correlation between age, gender and race with basal serum growth hormone and concluded that, in general, older Causcasian women had slightly higher growth hormone levels compared to older African-American women [ 8 ]. However, in that study (8) no statistical differences were shown when serum growth hormone levels in young Caucasian women (age, 28 ± 6; mean ± SD) were compared to serum growth hormone levels in African-American women (age, 34 ± 10). This finding is particularly noteworthy to studies of SLE because, in most cases, SLE onset is prominent among young females during their reproductive years, and African-American women are over-represented in the SLE patient population [ 21 ]. The present analysis also extends the results of previous studies [ 8 , 9 , 14 ] and partially supports the conclusions of Ghigo et al . [ 22 ] who showed that basal growth hormone levels were similar in young and older individuals. Ghigo et al . [ 22 ] further suggested that the somatotroph response in young versus older individuals to the combined administration of arginine and growth hormone-releasing substance also did not vary with age. In contrast, the present results do not support the conclusions that growth hormone decreases as a function of age as reported by Kelijman [ 23 ]. In fact, the results of the present study showed a strong correlation between age and serum growth hormone only in the circumscribed young normal female (age 20 – 46 yrs) group (Figure 1A ). The decrease in basal serum IGF-1 levels with age (Figure 1B ) confirmed previous studies by Hochberg et al . [ 24 ] who studied patients with osteoarthritis of the knee as well as earlier studies by Ghigo et al . [ 22 ] who reported a significant difference in IGF-1 levels between young and older individuals. Thus, it was not unexpected that basal serum IGF-1 levels in SLE was significantly elevated when compared to basal serum IGF-1 levels in the general population of normal subjects, but not so, when basal serum IGF-1 levels from SLE patients were compared to their age-matched counterparts (Table 1 ). In this regard, Bennett et al . [ 25 ] also failed to find differences in serum IGF-1 levels when normal subjects (age, 45.1 ± 8.6) were compared to 15 age-matched SLE patients (age, 42.5 ± 7.0). The relationship between putative abnormalities in the hypothalamic-pituitary axis, systemic disturbances and SLE pathogenesis and progression remains conjectural. In this regard, Rovensky et al . [ 26 ] found no correlation between plasma prolactin, growth hormone, interleukin-6, cortisol or C-reactive protein in adult SLE patients. However, studies by Chikanza et al . [ 27 ] reached a different conclusion. They suggested that a "pro-inflammatory hormonal bias" existed in juvenile SLE which was identical to adult SLE. They also concluded that the role of the neuroendocrine-immune system in adult SLE was, at the present time, limited to deficiencies in prolactin. Of note, two recent case reports suggested a link between growth hormone and exacerbation of lupus nephritis in a male teenager with SLE [ 28 ] as well as in juvenile SLE [ 29 ] where when growth retardation treated with growth hormone was terminated, clinical improvement in lupus symptoms was observed. These findings suggested that exogenously-administered growth hormone may result in "toxic" levels of growth hormone accompanied by lupus "flares" with progressive autoimmune dysfunction. A recent study from this laboratory showed that the growth hormone to somatostatin ratio was skewed upward in patients with RA [ 17 ]. In the present study, somatostatin levels in the age groups encompassing the average age of the SLE patients were not different from than of normal subjects (Table 1 ). Although previous studies have suggested that somatostatinergic activity increased with age [ 20 ], the present analysis (Table 1 ) does not support that view (at least from measurements of basal somatostatin levels) as lower somatostatin levels in the older SLE patients reached statistical significance when compared to age-matched controls with the caveat that the present study did not relate changes in somatostatin to SLE disease activity. Although somatostatin may alter growth hormone effects and immune responses in chronic autoimmune diseases, the relationship between somatostatin and "specific" somatostatin receptor (sSR) in SLE remains to be elucidated. In this regard, van Hagen [ 30 ] showed that 97% of patients with sarcoidosis, 100% of patients with tuberculosis or Wegener's granulomatosis, 75% of patients with Sjogren's syndrome but only 50% of SLE patients exhibited sSRs on mitogen-activated human peripheral lymphocytes compared to 97% in normal individuals. Of note, somatostatin receptor levels appeared to be unrelated to disease progression or remission. In the present study, a trend towards reduced serum somatostatin levels was seen only in the older SLE patients (Table 1 ). As functional somatostatin may change in autoimmunity and result in altered growth hormone release, reduced somatostatin levels could also influence basal levels of growth hormone in elderly SLE patients. Thus, changes in somatostatin could be one of several environmental stress factors resulting in the progression of clinically active disease in older SLE patients [ 31 ]. The therapeutic implications and diagnostic utility of serum growth hormone, IGF-1 and somatostatin measurements in SLE as well as in other musculoskeletal disorders appears central to assigning a role for these factors in disease progression. Serum growth hormone remained elevated in some DISH and OA patients where clinical symptoms were significant [ 9 , 11 , 12 ]. Thus, single serum growth hormone determinations appear to accurately reflect a pattern of serum growth hormone levels associated with these clinical disorders. Further, improvement in the clinical symptoms in OA and DISH patients with medical therapy [ 10 , 13 ] resulted in a sustained reduction in serum growth hormone levels reaching levels comparable to those found in normal subjects. Although longitudinal measurements of serum somatostatin in SLE and other rheumatic diseases have not yet been performed, reduced somatostatin levels appear to be most strongly associated with joint inflammation (as was seen in RA) [ 17 ] as well as in older patients (>55 yrs) with the inflammatory complications of knee OA (Denko and Malemud, submitted). Thus, it could be informative if elevated somatostatin levels correlated with clinical improvement in SLE patients. Competing interests The authors declare that they have no competing interests. Authors' contributions CWD participated in the design of the study and performed the clinical analysis. CJM participated in the design of the study and performed the statistical analysis. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC529450.xml

546040

What Does an Airline Traveler Have in Common with a Glowing Fish?

In William Gibson's novel Pattern Recognition , the protagonist posits a theory of jet lag: “Souls can't move that quickly, and are left behind, and must be awaited, upon arrival, like lost luggage.” Science has yet to address the issue of a spiritual speed limit, but it is generally accepted that jet lag actually results from the upset of the body's circadian clock, a biochemical pacemaker that dictates daily rhythms in sleep and wakefulness as well as body temperature and metabolic activity. In humans, the circadian rhythm responds to many factors, but daytime–nighttime (or, more precisely, light–dark) cycles are one major regulator. It is possible to gradually reset an upset circadian clock; if travelers remain in the same place for long enough, their circadian rhythm will eventually adjust to the new time zone and ambient light patterns, and the symptoms of jet lag will disappear. The more scientists know about the workings of the circadian clock, the closer they can come to manipulating it. Much is known about the molecular machinery of the circadian clock in the fruitfly, Drosophila melanogaster . Two circadian proteins, Clock and Cycle, cooperate to induce expression of two other proteins, Per and Tim, and when levels of Per and Tim are high enough, they cooperate to shut off their own expression. This negative feedback loop leads to periodic fluctuations in the level of Per and underlies the circadian rhythm in flies. However, until recently, not much was known about the mechanics of the circadian clock in vertebrates. A fusion gene (period3-luciferase) was used to track circadian rhythms Maki Kaneko and Gregory Cahill have created a new tool for investigating the components of the circadian clock in vertebrates: a zebrafish that luminesces (glows) in sync with the periodicity of its circadian clock. To do this, the researchers created a transgene that places expression of the firefly luciferase gene under the control of the promoter of the zebrafish circadian gene period3 (per3) . Each cell of the transgenic fish has one normal copy of the per3 gene and one copy of the period3-luciferase fusion gene (per3-luc) . Therefore, whenever per3 expression is normally turned on in a cell, the cell produces Per3 protein and also produces the luciferase protein. While characterizing their transgenic zebrafish, the authors made some interesting findings. First, contrary to earlier studies, the authors found that per3 periodicity is not hardwired into zebrafish embryos; instead, per3 periodicity is entrained by alternating light–dark cycles, which must occur at specific stages in early development. Also, other external factors such as ambient temperature can influence both the level of per3 mRNA expressed in the animal and the magnitude of its protein-level oscillations. Because the establishment of circadian rhythms in the adult animal can be so strongly influenced by conditions experienced by the embryos, the authors suggest using a standardized set of conditions for the culture of transgenic embryos in future experiments involving adult fish. Under these controlled conditions, Kaneko and Cahill anticipate that these transgenic zebrafish will be quite useful in examining the molecular machinery of the vertebrate circadian clock. For example, researchers can use the per3-luc transgenic zebrafish in forward genetic screens (where researchers mutagenize the animal to induce a desired phenotype and then identify the mutated gene responsible for the phenotype). In this case, mutagenized zebrafish could be examined for disruptions of per3-luc periodicity or expression. What is more, luminescence can be measured quickly and noninvasively, making this animal an ideal candidate for high-throughput screening aimed at identifying components of the circadian clock in the zebrafish. Thanks to luminescent fish, scientists may someday gain enough insight to make jet lag a thing of the past.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC546040.xml

526763

Traumatic deep vein thrombosis in a soccer player: A case study

A 42 year-old male former semi-professional soccer player sustained a right lower extremity popliteal contusion during a soccer game. He was clinically diagnosed with a possible traumatic deep vein thrombosis (DVT), and sent for confirmatory tests. A duplex doppler ultrasound was positive for DVT, and the patient was admitted to hospital for anticoagulation (unfractionated heparin, warfarin). Upon discharge from hospital the patient continued oral warfarin anticoagulation (six months), and the use of compression stockings (nine months). He followed up with his family doctor at regular intervals for serial coagulation measurements, and ultrasound examinations. The patient's only identified major thrombotic risk factor was the traumatic injury. One year after the initial deep vein thrombosis (DVT) the patient returned to contact sport, however he continued to have intermittent symptoms of right lower leg pain and right knee effusion. Athletes can develop vascular injuries in a variety of contact and non-contact sports. Trauma is one of the most common causes of lower extremity deep vein thrombosis (DVT), however athletic injuries involving lower extremity traumatic DVT are seldom reported. This diagnosis and the associated risk factors must be considered during the initial physical examination. The primary method of radiological diagnosis of lower extremity DVT is a complete bilateral duplex sonography, which can be augmented by other methods such as evidence-based risk factor analysis. Antithrombotic medication is the current standard of treatment for DVT. Acute thrombolytic treatment has demonstrated an improved therapeutic efficacy, and a decrease in post-DVT symptoms. There is a lack of scientific literature concerning the return to sport protocol following a DVT event. Athletic individuals who desire to return to sport after a DVT need to be fully informed about their treatment and risk of reoccurrence, so that appropriate decisions can be made.

Introduction Athletes are susceptible to a variety of vascular injuries, secondary to either repetitive motion, or high-speed collisions [ 1 ]. The differential diagnosis for lower extremity trauma in sport seldom invites a diagnosis of vascular injury, such as a deep vein thrombosis (DVT). Failure of the physician to recognize a vascular injury can have catastrophic limb or life threatening (pulmonary embolism) implications. The epidemiology, diagnosis, treatment, and recurrence of DVT, as well as the prevention of post-thrombotic symptoms are the most current areas of clinical research. Research-based guidelines concerning an athlete's return to sport after a DVT is an important area for future investigation. Case Report A 42 year old Polish born male former semi-professional soccer player was seen on May 16 th , 2003 in the emergency department, with the chief complaint of right leg pain. The patient had been playing soccer 10 days prior to this visit, and recalled a traumatic "tackle" injury to the posterior area of his right lower extremity. He denied experiencing any sensation of tearing or popping in the right knee during the index trauma, and was able to complete the game with only minor discomfort. On day 3 post-injury the patient noted severe pain in his knee and calf with ambulation. The patient visited his primary doctor on post-injury day 8 and was diagnosed with a right lower extremity soft tissue injury. A right lower extremity echo-doppler ultrasound (US), and a semi-quantitative D-dimer automated latex procedure were ordered to rule out a vascular disorder. The US investigation demonstrated a DVT in the distal femoral, popliteal, and distal calf veins, with a heterogenous mass (5 cm × 3 cm × 4 cm, resembling a hematoma) without a doppler signal in the right popliteal fossa. The D-dimer result was also positive for a suspected thrombosis (1.0–2.0 ug/ml; range = <0.25 ug/ml). The patient was instructed by his physician to proceed immediately to the emergency department for further evaluation and treatment. The past medical and family history of the patient was non-contributory for a history of thrombophilia or other thrombotic major risk factors. The patient had a remote (11 years old) surgical history of a right-sided inguinal hernia that could have created scar tissue contributing to vascular obstruction and stasis. The initial emergency department examination demonstrated an exquisitely tender right calf with a 3 cm difference in mid-calf girth (10 cm. distal from each inferior patellar pole); a 1+ right knee supra patellar effusion; and a palpable popliteal mass with visible ecchymosis. Laboratory tests (CBC, Lytes, PT, PTT, ESR, CPK, Anti-throbomin, Factor V Leiden, Lupus Screen, ANA, Anti-Cardiolipin, Protein C, and Protein S) were negative for metabolic, hematological or familial abnormalities. A repeat US investigation confirmed the results of the previous outpatient results. The patient was anticoagulated simultaneously with unfractionated heparin and Warfarin sulfate. A multiview plain film x-ray examination of the right lower extremity demonstrated no fracture, dislocation, or bony mass. A magnetic resonance image (MRI) of the right knee was done several days after admission, to verify a torn right knee meniscal cartilage that had been previously diagnosed. The official MRI radiological report included a small free-edge tear of the posterior horn root junction of the lateral meniscus, chondromalasia (lateral patella and lateral femoral articular cartilage), and a moderate joint effusion with a bursal cyst or dilated semimembranous-gastronemius bursa. Anticoagulation was achieved on day 6 of the patient's hospitalization. He was discharged on 5 mg of warfarin per day, with instructions to continue the use of compression stockings. The patient was also advised to follow up with his primary physician for regular monitoring, and to avoid contact or collision activities during anticoagulation. The patient was maintained on warfarin for six months, with weekly physician monitoring (symptoms, PT, INR) for the first three months post-injury. The monitoring interval was changed to once per month for the remainder of the treatment period. Hematologic investigations (APTT, PT, INR, Cardiolipin antibody, C-reactive protein, Lupant anticoagulant, Factor V Leiden, Antithrombin, ANA, Protein C, Protein S, and RPR) were obtained three months post injury. There were no contributory thrombophilic factors found in these investigations. Laboratory levels of Protein C activity 22% (range = 70–140%), Protein S activity 48% (range = 75–140%), INR 2.57 (range = 0.88–1.12), and PT 27.5 sec (range = 9.6–12.0 sec); APTT 38.5 sec (range = 23.4–35.4 sec) were found to be appropriately reactive to the anticoagulant therapy. The patient underwent two arthrocentesis procedures to remove small amounts of serous fluid from the joint, and each time was injected with a lidocaine/corticosteroid combination. US examinations after the hospitalization period failed to demonstrate a recurrence or new onset of DVT, however residual echogenic material characteristic of a chronic thrombus was demonstrated in the popliteal vein. Compression stocking use was maintained after hospital discharge, and was discontinued after nine months. The patient returned to soccer after anticoagulation, with a full understanding of his increased risk of DVT recurrence. One-year post injury the patient continued to suffer from intermittent right lower extremity discomfort and swelling often unrelated to activity. An elective arthroscopy was recently performed on the patient's right knee to investigate his long-standing meniscal disruption and effusion. The arthroscopy demonstrated several areas of arthrosis (patellar lateral and medial facets, lateral and medial femoral condyles), and a torn lateral meniscus. Appropriate partial lateral menisectomy and debridement, and chondroplasty of the areas of arthrosis were preformed. An arthroscopic examination of the posterior compartment demonstrated a small cleft-like area just medial to the semimembranosis where the Baker's cyst likely originated. The patient returned to the orthopedist one week post-op with a large (150 cc's) hemarthrosis that was aspirated from the knee. He was requested to follow-up in one month for re-evaluation. Discussion This case study illustrates the importance of considering deep vein thrombosis in the diagnosis of sport-related extremity trauma. DVT is classically related to venous stasis, intimal injury, and coagulation diathesis (Virchow's triad). The estimated incidence of DVT from all causes is 0.5 to 1.6 per 1000 persons per year, and may be an underestimation due to the number of DVT that are asymptomatic [ 2 ]. Standard risk factors for DVT are immobilization, pregnancy, recent surgery (particularly orthopedic), malignancy, older age, smoking, coagulation deficits or hypercoagulable states, connective tissue disorders, sex steroid administration, severe dehydration, and major trauma. Bates et al. [ 3 ] presented a table of the estimated relative risk (RR) for individual DVT risk factors. These factors include inherited conditions (e.g. Factor V Leiden, RR = 50, Antithrombin deficiency, RR = 25, Protein C and S deficiency, RR = 10); acquired conditions (e.g. major surgery or trauma, RR = 5–200; history of venous thromboembolism, RR = 50); and hereditary, environmental, or idiopathic conditions (e.g. hyperhomocysteinemia, and elevated levels of Factor VIII, RR = 3: elevated levels of Factor IX, RR = 2.3). Coagulation diathesis through congenital or acquired thrombophilia may promote coagulation [ 3 ]. Coagulation deficits in previously healthy athletes are becoming increasingly identified through laboratory tests, and must be considered as contributing factors for DVTs [ 4 - 7 ]. Hilberg et al. [ 6 ] found that the risk of hereditary exists in elite athletes, corresponds to the general population. These authors proposed that countermeasures (e.g. early anticoagulation during periods of immobilization/injury; single dose of low molecular weight heparin and/or leg exercises on long-distance flights; and avoiding hemoconcentration with adequate hydration) for athletes who are carriers of a congenital coagulation deficit [ 6 ]. The testing for hypercoagulable states in an individual after a single episode of thrombosis is a costly, yet routine procedure in many centers. The common assumption that an identified presence of a thrombophilic abnormality increases the risk of recurrence, and justifies prolonged therapy is without clear supportive evidence. A review of the current literature concerning the treatment of individuals with coagulation deficits concludes that there is no clear evidence that modifying treatment because of an identified hypercoaguable state alters the outcome, or that more intensive therapy is required in patients with laboratory evidence of thrombophillia [ 3 ]. Exercise is thought to act as a protective mechanism against thrombosis, due to the controlled balance between the exercise activated coagulation and fibrinolytic pathways [ 8 ]. Upper extremity thrombosis that is not related to primary diseases or well known risk factors are rare (2–4% of DVTs).This type of thrombosis has been documented in a variety of sports as effort thrombosis or "Paget-Schroetter's syndrome" [ 9 - 14 ]. This syndrome is been described as a primary thrombosis of the subclavicular and axillary veins, usually proceeded by a strenuous effort or repetitive action involving retroversion and hyperabuction of the extremity [ 10 ]. Vascular compression by adjoining bone, ligament and muscle or resulting intimal traumas have been documented as contributing factors toward the development of upper and lower extremity thrombosis [ 15 - 27 ]. Lower extremity DVT with a traumatic sporting injury in otherwise healthy active adults is seldom mentioned in the medical literature [ 16 - 29 ]. This lack of reported cases of this type of thrombosis may be due to either underreporting or incorrect diagnosis. Very few cases of sport-related lower extremity DVT involved direct externally trauma [ 29 , 30 ]. There is one case report (Finnish language) that specifically related DVT development to soccer-related trauma [ 30 ], and one case report of lower extremity DVT in a soccer player with coagulation deficiencies [ 31 ]. There is also one case report in the literature of a traumatic popliteal thrombosis in a hockey player, which resulted in a fatal pulmonary embolism (PE) [ 29 ]. The popliteal, posterior tibial and peroneal veins are susceptible to intimal trauma by the sudden hyperextension and torsion that the lower extremity experiences in a soccer "kick" or "tackle" motion. The popliteal arteries and veins are susceptible to direct, sheering, and muscular compressive forces due to their anatomical position, especially with rapid knee hyperextension or anterior dislocation [ 13 , 22 ]. The literature demonstrates the importance and efficacy of a complete bilateral duplex sonography as the primary method of DVT diagnostic investigation [ 32 ]. US findings can be augmented by other methods (e.g. evidence-based risk factor analysis) [ 33 , 34 ]. A review of the current literature also suggests the need for comprehensive evidence-based guidelines concerning the use of radiological diagnostic investigations of suspected DVT [ 35 ]. Anticoagulation is effective in preventing DVT propagation and PE, but has no chemical fibrinolytic activity. This type of therapy allows for intrinsic fibrinolysis to occur. Radiographically demonstrable clot lysis occurs in only 50% of anticoagulated patients, and the incidence of complete resolution is less than 5%. Intrinsic fibrinolysis that occurs slowly does not preserve the function of the venous valves, which become fibrotic and fixed after a few weeks of being trapped in clot [ 36 ]. The symptoms experienced by individuals without complete clot resolution include heavy or achy legs, edema, throbbing paresthesia, purities, numbness, stiffness, and difficulty standing or ambulating. Postthrombotic syndrome (PTS) is characterized by brawny edema of the leg, stasis dermatitis, hyperpigmentation, induration, ulceration and chronic leg pain. This syndrome is associated with an extraordinary level of chronic pain and disability, and approximately 40% of the total cost of treating DVT is spent on PTS [ 36 ]. Zeigler et al. [ 37 ] investigated the long-term clinical outcome of individuals with a first DVT. These authors found that 82% of the patients suffered from recurrent symptoms, with a mean follow-up period of 6.6 years. Four level DVT, calf vein thrombosis, recurrence of ipsilateral DVT, and a non-sufficient oral anticoagulation are of prognostic significance for developing clinically relevant symptoms within 10 to 20 years after the first DVT [ 37 ]. There is growing evidence that the early lysis provided by thrombolytic therapy is more likely to preserve valve function, decreasing the likelihood of DVT recurrence, and the occurrence of PTS [ 38 , 39 ]. Recent trials of new antithrombotic agent used an endpoint of 'symptomatic recurrent DVT', which was defined as the combination of persistent or recurrent symptoms along with the radiographic evidence of primary clot progression or new thrombus formation. The rate of symptomatic recurrent DVT was reported to be between 4–7%, and does not reflect those individuals who simply continue to be symptomatic after the primary event [ 35 ]. The general knowledge concerning quality of life and burden of illness in patients with persistent post-DVT symptoms is limited. This issue is especially important to the athletic patient, as participation in sport is usually an extremely important component of quality of life. For routine monitoring of outcomes in chronic venous disorders there are questionnaires that are available [ 40 , 41 ]. Hedner et al. [ 42 ] have recently developed an instrument that measures health and treatment-related quality of life factors in DVD patients. The athlete's primary concern upon the initial DVT diagnosis is return to play. The issue of return to sport after a lower extremity DVTs has only been addressed only once in the literature concerning return to non-contact sport [ 43 ]. General guidelines for sedentary individuals allow for a gradual return to return to daily activities over a six week period [ 43 ], with no contact activities allowed during the period of anticoagulation. Roberts and Christie [ 43 ] provided a theoretical framework, based on the natural history of animal models for the safe and expeditious return of the athlete. These authors suggested a protocol that combines a graduated return to activity and anticoagulation therapy with regular physician based reevaluation [ 43 ]. An athlete who wants to return to a contact or collision sport should be informed of the possible increased risk of recurrent DVT that he or she may face, above the current estimates derived from the general population. There is no current evidence in the literature that investigates the specific risk factor of a traumatic collision, and the recurrence of a DVT. This lack of evidence suggests that the patient and physician should work together to make an informed return to play decision involving the patient's current individual risk profile, the likelihood of DVT recurrence, athletic goals, and the perceived importance of the particular sport to quality of life. The potential limitations of this case study include the lack of testing for prothrobin mutation, and fibrynolitic parameters (level of tPA, PAI-1 or PAI-1 polymorphism 4G/5G). Competing Interests The author(s) declare that they have no competing interests. Author Contributions PE developed, researched, wrote and revised the case study; RU assisted in study development and manuscript revision; DM assisted in manuscript development and revision; HJ assisted in manuscript development and revision.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC526763.xml

523828

Palliative Care in Africa and the Caribbean

In many of the world's poorest countries, dying is often accompanied by avoidable pain and other distressing symptoms. How can we improve care at the end of life?

“If someone is condemned to a premature death because of the injustice of global health inequality, it is doubly unjust for that person to be condemned to an agonising death racked by preventable pain.” [ 1 ] In many resource-poor countries, death is accompanied by avoidable pain and other distressing symptoms. Unfortunately, governments in these countries usually give care at the end of life a low priority compared with preventive and curative services [ 2 ]. This prioritization makes little sense, especially when applied to treating patients with cancer and HIV/AIDS, since prevention efforts are often failing to reduce the disease burden, while treatments aimed at cure or prolonging life are still too expensive to be made widely available. As three physicians in Jamaica, Uganda, and Rwanda, we believe that providing quality care at the end of life should be seen as a global public health priority. By using relatively low-cost palliative care approaches and community-based strategies, thousands of terminally ill patients in Africa and the Caribbean could be relieved of their pain and suffering. The Burden of Cancer and HIV/AIDS In the countries where we work, the burden of cancer and HIV/AIDS is overwhelming. In Africa about 2.5 million people die annually from HIV/AIDS, and more than 0.5 million die from cancer [ 3 , 4 ]. Sepulveda and colleagues have estimated that each year, at least one in 200 people in the five African countries that they studied (Botswana, Ethiopia, Tanzania, Uganda, and Zimbabwe) need palliative care at the terminal stages of HIV/AIDS or cancer [ 2 ]. This figure does not include those needing palliative care for other diseases or those suffering from a serious illness in the pre-terminal stages. Thus, perhaps one in 100 people in these countries needs some level of palliative care each year [ 2 ]. In Rwanda, as in most other African countries, infectious diseases are still rife. Health professionals are often faced with the terrible dilemma of having to choose between saving lives and easing the suffering of the dying. Indeed the authorities usually believe that any investment in palliative care would be at the expense of providing life-saving treatments for those suffering from curable, often infectious illness. In many Caribbean countries, while the scourge of water- and insect-borne infectious diseases is largely under control, the prevalence rates of HIV in the adult population are some of the highest in the world [ 5 ]. In Jamaica, the largest English-speaking country in the Caribbean (population 2.5 million), in 2001, there were an estimated 20,000 people living with HIV and 980 deaths from AIDS [ 6 ]. Further, Jamaica's proximity to the United States means that many people aspire to a lifestyle more representative of a wealthy, industrialized nation, and are thus susceptible to diseases such as cancer, coronary artery disease, and diabetes. Unfortunately, the island's struggling public health system is often unable to provide adequately for patients with these diseases. Better distribution of analgesics would improve palliative care provision The Arguments for Palliative Care Prevention efforts—including health promotion, education, and screening—and treatments aimed at cure or prolonging life are key strategies needed to reduce the burden of HIV/AIDS and cancer in resource-poor countries [ 7 ]. However, when it comes to prevention, in many countries the effects of health education, health promotion, and screening programs have yet to make an impact on rates of HIV infection or cancer. When it comes to treatment, the provision of high-quality, affordable treatments for patients with HIV/AIDS and cancer requires the development of appropriate and accessible infrastructure and technology with sustainable funding. At present, access to treatment where we are working is essentially controlled by the ability of the patient to pay. Thus, only about one in 200 people with HIV in Uganda are able to obtain antiretroviral medicines [ 8 ]. Furthermore, patients in developing countries often present with far advanced malignant disease, and as many as 80% of people with cancer may be incurable at diagnosis [ 9 ]. Given that prevention isn't taking effect in many places, and curative services are poorly available or inappropriate, we believe that the provision of palliative care ( Box 1 ) in the Caribbean and Africa should be viewed as an urgent public health problem. About 80% of cancer patients will have pain in the terminal phase of their disease [ 1 ], and we estimate that at least 25% of HIV/AIDS patients have substantial pain during their illness. Box 1. The WHO Definition of Palliative Care The WHO has defined palliative care as an approach that improves the quality of life for patients and their families facing the problems associated with life-threatening illness, through the prevention and relief of suffering. This is done through early identification, careful assessment, and treatment of pain and other problems—physical, psychological, and spiritual. Dying is regarded as a normal process, and death is neither hastened nor postponed [ 2 ]. The philosophy of hospice and palliative care acknowledges death, dying, and bereavement as a reality of life. Effective and relatively cheap methods exist for controlling pain and other symptoms. For example, the World Health Organization (WHO) has outlined a relatively cheap way of relieving cancer pain in about 90% of patients, which could be extended to patients with HIV/AIDS [ 2 ]. Sadly, most people in Africa and the Caribbean who need pain relief aren't receiving it [ 10 ]. Assessing Patients' Needs Several studies in East Africa have looked at the experience of dying, the quality of care at the end of life, and patients' unmet needs [ 2 , 11 , 12 ]. Recurring themes are (1) unmet physical needs, including the need for relief of pain and other symptoms, (2) the need for food, (3) the high cost or unavailability of appropriate analgesic drugs, (4) the severe financial constraints on the family and caregivers, (5) the need for training of family caregivers, (6) lack of psychosocial support, and (7) social isolation due to the stigma attached to a diagnosis of HIV/AIDS. In the Caribbean, patients' needs at the end of life appear to be similar to those of patients in many East African countries. A qualitative study in Grenada, in the Eastern Caribbean, showed that people preferred to die at home rather than in hospital and—in the absence of pain relief and much-needed counseling, information, and financial support—they took solace in spiritual comfort [ 13 ]. In Jamaica ( Box 2 ), although data are scarce, it seems that patients' needs are very similar to those in Grenada. Christianity is the principal religion of Jamaica, and faith in God and family support are critical factors in patient care at the end of life. Outside of the hospital setting, appropriate analgesics are difficult to access and are often unaffordable. Patients and caregivers are not provided with enough information to help them understand disease processes, and what to expect as the ill person nears death. There is little or no palliative care provision for patients with HIV/AIDS. Box 2. Dying in Jamaica—A Typical Case Scenario This fictional case scenario gives an impression of the sorts of problems that patients face at the Hope Institute, Kingston—Jamaica's first public hospice. A 50-year-old woman is diagnosed with inoperable lung cancer. Because of brachial plexus involvement, she experiences severe pain and weakness of her arm. She is treated at Kingston Public Hospital with palliative radiotherapy, which helps the pain for a few months. But then the pain returns, and she requires a high dose of slow-release morphine for pain control. She lives in the mountains, and her house is a two-and-a-half-hour bus ride from Kingston, the capital city. Unfortunately, the public pharmacy in Kingston is unwilling to dispense more than a week's supply of morphine at any one time, because they have limited supplies (there is a shortage of the drug in Jamaica) and because they think the patient's dose is unacceptably high. So she has to make the exhausting five-hour round trip every week. Her husband's health has also recently declined, and the woman's sister now has to care for the patient and her husband. The family now has the financial means to afford only one small meal a day, and they rely on donations from their church community in order to survive. Because the family's savings dwindle, and the public pharmacy faces further shortages of morphine, the woman with cancer requires multiple admissions to the hospice in Kingston over the last six months of her life in order to get suitable analgesia. Uganda's Public Health Approach Uganda has made palliative care for patients with AIDS and cancer a priority in its National Health Plan [ 10 ]. In 1993, after conducting a feasibility study, Hospice Uganda was established in Kampala, making palliative care available to a population of about 2 million people (Uganda's population is 22 million people). There are now two other hospices, one in Mbarara serving 1 million people, and one in Hoima serving 350,000 [ 8 ]. The hospice care provided by these units is all home-based care. This type of care provision is designed to meet the cultural and practical needs of the people in Uganda, where most people prefer to die in their own homes, and where people are often buried in their household gardens. Hospice Uganda provides community-based care principally to patients suffering from HIV/AIDS and cancer. Almost all patients coming to the hospice have pain, and a great deal of attention is focused on good pain management. Uganda is only the third African country to have made morphine available and affordable to its patient population. Because of the dearth of legal prescribers (doctors, dentists, and vets only), in May 2004, Uganda changed the statute. This allowed midwives to prescribe pethidine, and allowed clinical palliative care nurses and clinical officers who are specially trained and registered to prescribe morphine. How was Uganda—an African country with a relatively under-funded health service—able to provide a palliative care service? A national program using a public health approach to reach those in need was established following principles outlined in the WHO's National Cancer Control Guidelines [ 4 ]. These guidelines outline the importance of assessing the magnitude of the problem, setting measurable objectives, evaluating possible strategies, and choosing priorities for initial activities. A series of workshops were held in Uganda between 1998 and 2000, where the WHO's “little cost, big effect” measures began to be addressed. The three key measures involve education, increased drug availability, and changes in government policy ( Figure 1 ). Figure 1 The WHO's Triangle of Foundation Measures (Adapted with permission from “A Clinical Guide on Supportive and Palliative Care for People with HIV/AIDS” [ http://hab.hrsa.gov/tools/palliative/ ]) Other African Initiatives Four other African countries—Botswana, Ethiopia, Tanzania, and Zimbabwe—have made the development of home-based care a priority in dealing with the HIV/AIDS epidemic [ 2 ]. Botswana has an operational home-based care program integrated into its national health system, while in the other three countries, care is largely provided through private organizations. But few of the home-based care services in these countries include the capacity for providing effective pain relief [ 2 ]. The Next Steps By using strategies such as providing access to an essential short list of relatively cheap generic medications, and other methods recommended by WHO, it has now been proven that palliative care in the African context is affordable and achievable [ 2 , 7 , 14 ]. We believe that, following the Ugandan and Botswanan models, palliative care should be integrated into national government strategies. In order to begin to show governments the importance and economic justification for developing a palliative care health policy, it is clear that needs assessments are an essential first step. It is likely to be much less expensive to provide community-based care with family and community support at the end of life than to burden already overcrowded hospital wards with patients suffering end-stage disease. There is a long tradition, both in Africa and in the Caribbean, of caring for the disabled, the mentally ill, and the young and elderly sick at home. Both start-up and sustainable funding are enormous issues that will need to be addressed by local governments, international funding agencies, and charitable bodies. Advocating palliative care to decision makers, providing training programs for health professionals, and making medications available and affordable are important challenges. Research in individual countries is needed to assess whether the above recommendations are suitable locally. Hospice Africa Uganda is advocating to other African governments and assessing other African countries where local laws and customs may dictate the most suitable way to provide palliative care together with government support. Partnerships and a public health approach to palliative care must be the way forward. Palliative Care Resources for the Developing World African Palliative Care Association Representing Kenya, South Africa, Tanzania, Uganda and Zimbabwe E-mail: apca@hospiceafrica.or.ug Hospice Africa (Uganda) Resource and Training Centre PO Box 7757, Kampala, Uganda Tel: +256 41 266 867 / 510089; Fax: +256 41 510087—residence E-mail: info@hospiceafrica.or.ug ; E-mail: anne@hospiceafrica.or.ug Centre for Palliative Learning Hospice Association of the Witwatersrand PO Box 87600, Houghton, Johannesburg 2041, South Africa Hospice Information At http://www.hospiceinformation.info . Click on “Training” to search for courses and conferences in palliative care and bereavement. Requires member's password to access this part of the website but membership is free to people in developing countries—contact hospice information at + 44 (0)870 903 3 903 (telephone), + 44 (0)20 8776 9345 (fax), or info@hospiceinformation. Information is also circulated quarterly by E-mail to members under the title of e-Choices. Palliative Care in Resource-Poor Settings A freely available overview of HIV/AIDS palliative care, written by Kathleen Foley, Felicity Aulino, and Jan Stjernswärd. At http://hab.hrsa.gov/tools/palliative/chap19.html . Living Well with HIV/AIDS A freely available manual on nutritional care and support for people with HIV/AIDS, by the Food and Agriculture Organization of the United Nations. At http://www.fao.org/DOCREP/005/Y4168E/Y4168E00.HTM . Cancer Pain Relief: A Guide to Opioid Availability A section of this guide, by the WHO, is freely available at http://www.medsch.wisc.edu/painpolicy/publicat/cprguid.htm .

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC523828.xml

340939

Bridging the Science–Policy Divide

The Millennium Ecosystem Assessment focuses on the benefits people obtain from ecosystems and aims to improve ecosystem management and contribute to human well-being and poverty alleviation

Nobody questions the importance of good scientific information for sound environmental decision-making. But designing mechanisms to link scientific research to the decision-making process is no easy matter. Research and decision-making often seem to operate in different worlds. Policy-makers' needs for applied findings and best judgment typically clash with scientists' pursuit of basic research and statistical significance. Despite this challenge, as needs for scientific input into decision-making arise, a number of institutions have been established to help bridge the science–policy divide. Regulatory agencies like the United States' Food and Drug Administration have met this need in areas of public health, for example, and Environmental Impact Assessment procedures have helped to introduce better science into project-level decisions. At the scale of global environmental challenges, highly regarded mechanisms have been established such as the Scientific Assessment of Ozone Depletion, which has guided decisions by governments, the private sector, and nongovernmental organizations (NGOs), and the Intergovernmental Panel on Climate Change, which has become the authoritative source of policy-relevant information on climate science. But a significant gap remains in the landscape of institutions designed to link science with policy-making: no mechanism has existed to provide decision-makers with authoritative information on the causes and consequences of changes in the planet's ecosystems and on the options for response. Human well-being and progress toward sustainable development are vitally dependent upon improving the management of Earth's ecosystems to ensure their conservation and sustainable use. The benefits that human beings extract from nature are the foundation of all economies and the basis of major industries, are sources of knowledge, and are central to many cultures. While demands for ecosystem services such as food and clean water are growing, human actions are at the same time diminishing the capability of many ecosystems to meet these demands. And while many of the changes to ecosystems, such as increased agricultural production, have greatly enhanced human well-being, many others have not. World fisheries are now declining due to overfishing, for instance, and some 40% of agricultural land has been degraded in the past half-century. Other human-induced impacts on ecosystems include alteration of the nitrogen, phosphorous, sulfur, and carbon cycles, causing acid rain, algal blooms, and fish kills in rivers and coastal waters, along with contributions to climate change. The benefits that human beings extract from nature are the foundation of all economies and the basis of major industries, are sources of knowledge, and are central to many cultures. Recognizing the growing scale of these problems, United Nations Secretary General Kofi Annan, in his 2000 Millennium Report to the General Assembly, called for a Millennium Assessment of Global Ecosystems to provide definitive information on the consequences of ecosystem change for human well-being. With further authorization received through three international conventions (on Biological Diversity, Desertification, and Wetlands) and with financial support from the Global Environment Facility, the United Nations Foundation, the David and Lucile Packard Foundation, and the World Bank, the Millennium Ecosystem Assessment (MA) ( www.millenniumassessment.org ) was launched one year later, in 2001. More than 700 authors from 80 countries are now involved in the expert working groups preparing the global assessment; 100 experts will serve on the Editorial Review Board; more than 1,000 experts will be asked to review the materials, and hundreds more are undertaking subglobal assessments as part of the MA. The MA focuses on ecosystem services (the benefits people obtain from ecosystems), how changes in ecosystem services have affected human well-being, how ecosystem changes may affect people in future decades, and response options that might be adopted at local, national, or global scales to improve ecosystem management and thereby contribute to human well-being and poverty alleviation ( Figure 1 ). Figure 1 MA Conceptual Framework The MA examines both indirect and direct drivers (both human-caused and natural) of change in ecosystems, how those changes affect ecosystem services, how those changes, in turn, influence human well-being and poverty reduction, and opportunities for interventions that can ensure ecosystem conservation and enhance human well-being. The assessment must take into consideration the multiple time and spatial scales over which these interactions take place. (Schematic is used by permission from the Millennium Ecosystem Assessment [2003] and published under the terms of the Creative Commons Attribution License.) The first report— Ecosystems and Human Well-Being: A Framework for Assessment —was published in 2003 and describes the approach and methods used in the MA. The four main assessment volumes—Conditions and Trends, Scenarios, Response Options, and Subglobal Assessments—began the first of two rounds of peer review in January 2004, and the final assessment reports will be published in early 2005. Unlike previous global scientific assessments, the MA is a “multiscale” assessment. Assessments at subglobal scales are needed because ecosystems are highly differentiated in space and time and because sound management requires careful local planning and action. Local assessments alone are insufficient, however, because some processes are global and because local goods, services, matter, and energy are often transferred across regions. The MA subglobal assessments will directly meet needs of decision-makers at the scale at which they are undertaken, strengthen the global findings with on-the-ground reality, and reinforce the local findings with global perspectives, data, and models. In Southern Africa, for example, a series of community-level assessments are being conducted using the MA conceptual framework. The findings from these assessments inform, and are informed by, assessments underway in the Gariep and Zambezi river basins. These local and river basin assessments, in turn, are linked to a regional assessment encompassing the countries in the Southern African Development Community. Other subglobal assessments are now underway in such regions as São Paulo, Brazil; coastal British Columbia, Canada; the Caribbean Sea; western China; Colombia; the Sinai Peninsula, Egypt; several regions within India; Indonesia; the Laguna Lake Basin, the Philippines; Portugal; and Sweden. The ultimate impact of the MA will depend on its credibility, legitimacy, and utility. To ensure its scientific credibility, the assessment involves leading scientists from around the world and has established an independent peer-review process. To ensure the political legitimacy, all of the stakeholders—governments, the private sector, and NGOs—have a role in governing the process, and governments themselves have approved the process through decisions in international conventions. And, to ensure its utility, ongoing interactions with stakeholders are designed to ensure a focus on their questions and issues. The scientific information now available concerning ecosystems and human development holds the promise of significantly improving the choices that the public and decision-makers take concerning the environment. But for that promise to be fulfilled, a bridge needs to be built between the research community holding this information and the decision-makers seeking it. The MA is an attempt to establish that bridge.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC340939.xml

544858

Changing treatment patterns for coronary artery revascularization in Canada: the projected impact of drug eluting stents

Background To evaluate current treatment patterns for coronary artery revascularization in Canada and explore the potential impact of drug eluting stents (DES) on these treatment patterns. Methods Eleven cardiologists at multiple Canadian academic centers completed a questionnaire on coronary artery revascularization rates and treatment patterns. Results Participating physicians indicated slightly higher rates of PTCA, CABG, and stent implantation than reported in CCN publications. Participants estimated that 24% of all patients currently receiving bare metal stents (BMS) would receive DES in the first year following DES approval in Canada, although there was a large range of estimates around this value (5% to 65%). By the fifth year following DES approval, it was estimated that 85% of BMS patients would instead receive DES. Among diabetic patients, estimates ranged from 43% in the first year following approval to 91% in the fifth year. For all patients currently receiving CABG, mean use of DES instead was estimated at 12% in the first year to 42% at five years; rates among diabetic patients currently undergoing CABG were 17% in the first year to 49% in the fifth year. Conclusions These results suggest a continued increase in revascularization procedures in Canada. Based on the panel's responses, it is likely that a trend away from CABG towards PTCA will continue in Canada, and will be augmented by the availability of DES as a treatment option. The availability of DES as a treatment option in Canada may change the threshold at which revascularization procedures are considered.

Background Coronary artery disease (CAD) is a common condition in western society [ 1 ]. Treatment of CAD often involves surgical revascularization, that is, removal of coronary artery stenoses to restore sufficient myocardial blood flow. Currently, there are three major treatment options for coronary artery revascularization: coronary artery bypass grafting (CABG), percutaneous transluminal coronary angioplasty (PTCA), and coronary artery stenting [ 2 ]. Treatment choice is based on a variety of factors, including patient age, comorbidities, extent of disease (i.e., number and location of affected coronary arteries), and disease severity. Treatment choices and treatment patterns for CAD have changed over the past several years, and are likely to evolve further in the next few years. Drug eluting stents (DES) are a newly available treatment modality. DES are stents that incorporate bioactive coatings (polymer or non-polymer) permitting the release of associated molecules to attenuate the processes of restenosis. Preliminary clinical data suggests that use of DES can substantially reduce the rates of restenosis seen following implantation of bare metal stents but at much lower incidences of severe procedure-related complications as compared to CABG surgery [ 3 , 4 ]. DES may also be an important treatment option for populations such as individuals with diabetes and multi-vessel disease, who appear to have better outcomes with CABG [ 2 ]. It is likely that the availability of DES will change treatment patterns for patients with CAD [ 5 ]. To better understand current treatment patterns for CAD and the potential role of DES in Canada, we developed and administered a questionnaire to a panel of Canadian cardiologists. This manuscript describes the estimates regarding current and future CAD treatment patterns in Canada provided by this panel. Methods The objective of this study was to obtain opinions and estimates from Canadian cardiologists regarding the annual rates of PTCA, CABG, and stenting in Canada, treatment patterns associated with these procedures, and the potential future impact of DES on coronary revascularization treatment patterns. The results from this questionnaire were used in the development of health economic models used to evaluate these stents for use among Canadian patients. The questionnaire asked participating physicians to assess the projected annual procedure rates for CABG, PTCA and stent implantations published by the Cardiac Care Network of Ontario (CCN) Target-Setting Working Group [ 6 ]. In addition to the estimated rates of the revascularization procedures, respondents were also requested to provide a lower and upper bound for each rate. Additional sections of the questionnaire provided recommendation rates for repeat revascularization procedures (following restenosis) depending on the type of first procedure. For example, physicians were asked for the percentage of patients they would recommend CABG after restenosis following PTCA. In the final sections, the questionnaire requested information on the projected use of DES among patients currently receiving bare metal stents or CABG. The questionnaire instructed respondents that if they agreed with provided reference values (based on published estimates) for a particular item, they could leave that item blank. However, respondents free to provide estimates that were greater, lesser, or unchanged compared to the reference values. Cardiologists at multiple academic centers distributed across Canada were identified for this study. These physicians were contacted, provided a brief introduction to the study, and offered a study honorarium of $400 dollars to participate. The questionnaire was distributed to physicians willing to participate. Completed questionnaires were collected and data was entered into Excel. Summary statistics including mean, median, maximum, and minimum values were computed for the various estimates provided by the participating physicians. The summary statistics were returned to those doctors who completed questionnaires. Study participants were then asked to comment on the summary results and indicate any questions or comments on the results. At the time of the study, only one drug-eluting stent (CYPHER™, Cordis Corporation) had been approved for use by Health Canada (Nov. 2002). Results Study participants A total of 18 Canadian physicians were contacted and invited to participate. Eleven physicians agreed to participate and complete the questionnaire, a response rate of 61%. All eleven physicians were male, specialized in cardiology, and had experience in this specialty ranging from 6 to 30 years. The average length of time specializing in cardiology was 12.8 years. The average age of participating physicians was 44.0 years. In each of the result tables, the number of panel members responding to each item is reported. Procedure annual rates The CCN Target-Setting Working Group (TSWG) projected annual rates of CABG surgery and PTCA of 110 and 160 per 100,000 Canadians, respectively [ 6 ]. Based on responses from the questionnaire, participating physicians reported slightly higher mean estimates of 112.7 (range of 100 to 150) for CABG and 172.3 (range of 125 to 200) for PTCA (Table 1 ). The TSWG also estimated that the rate of stent implantations among PTCA cases is over 90%. Questionnaire results indicated a similar proportion of patients receiving stents (92.1%, range 90% to 99%). Table 1 Estimated rates of CABG, PTCA, and Stenting Mean Median Min Max # of Panel Members Responding Annual rate of CABG per 100,000 (CCN estimate: 110) estimate 112.7 110 100 150 11 not lower than 91.1 90 60 110 9 not higher than 131.1 120 110 190 9 Annual rate of PTCA per 100,000 (CCN estimate: 160) estimate 172.3 180 125 200 11 not lower than 147.8 150 110 180 9 not higher than 214.4 200 160 300 9 Percent of PTCA patients receiving stents (CCN estimate: >90%) estimate 92.1% 90.0% 90.0% 99.0% 11 not lower than 81.8% 84.5% 70.0% 90.0% 8 not higher than 96.8% 96.0% 94.0% 100.0% 8 Recommendations for subsequent revascularization procedures Table 2 presents results for recommendations from participating physicians for patients who need a subsequent revascularization procedure after having received either stenting or CABG. For patients requiring subsequent revascularization after stenting, PTCA would be recommended for approximately 40% of cases. Brachytherapy was the second most frequently recommended procedure (23%), followed by similar rates for CABG and a second stenting. For revascularization following CABG surgery, stenting was recommended for almost 80% of patients while PTCA and CABG were recommended for 6.8% and 13.6% of patients, respectively. Table 2 Subsequent revascularization procedures following stenting or CABG Mean Median Min Max # of Panel Members Responding Percent of patients receiving each procedure following initial stenting PTCA 38.5% 40.0% 9.0% 75.0% 11 Stenting 18.4% 15.0% 0.0% 65.0% 11 CABG 19.8% 15.0% 1.0% 50.0% 11 Brachytherapy 23.3% 20.0% 0.0% 90.0% 11 Percent of patients receiving each procedure following initial CABG PTCA 6.8% 5.0% 0.0% 15.0% 11 Stenting 79.5% 80.0% 55.0% 90.0% 11 CABG 13.6% 10.0% 5.0% 40.0% 11 Projected use of DES among Canadian CAD patients Tables 3 through 6 present results from the cardiologist panel regarding the projected use of DES once after they are approved in Canada. For these projections, we separately asked the panel members to provide estimates on the proportion of patients currently receiving bare metal stents who would likely receive DES instead versus the proportion currently undergoing CABG who would likely receive DES instead. We also requested rates DES adoption separately for the entire Canadian CAD population versus the subpopulation of Canadian CAD patients with diabetes, as the diabetic population is at higher risk for adverse clinical outcomes [ 7 ] and therefore may have different treatment patterns. In all cases, responses for the proportion of patients likely to receive DES were requested for all patients in the specified population as well as separately for patients with single- vs. multi-vessel CAD. Projections were requested annually for the first five years following approval of DES in Canada. Table 3 Estimated Percentage of BMS Patients Likely to Receive DES by Year Following DES Approval Mean Median Min Max # of Panel Members Responding % of Bare Metal Stent Patients Receiving DES, 1st Year Following Approval all patients 24.0% 22.5% 5.1% 65.0% 10 all patients with single-vessel 18.8% 12.6% 5.0% 50.0% 8 all patients with multi-vessel 32.9% 25.0% 5.0% 80.0% 6 % of Bare Metal Stent Patients Receiving DES, 2nd Year Following Approval all patients 36.6% 40.0% 10.2% 65.0% 10 all patients with single-vessel 28.4% 30.4% 10.0% 50.0% 9 all patients with multi-vessel 43.9% 40.0% 5.1% 100.0% 8 % of Bare Metal Stent Patients Receiving DES, 3rd Year Following Approval all patients 57.1% 60.0% 20.5% 80.0% 10 all patients with single-vessel 53.5% 60.0% 20.0% 80.9% 9 all patients with multi-vessel 61.2% 65.0% 10.3% 100.0% 8 % of Bare Metal Stent Patients Receiving DES, 4th Year Following Approval all patients 76.7% 80.0% 50.8% 91.0% 10 all patients with single-vessel 69.6% 80.0% 30.0% 91.0% 9 all patients with multi-vessel 85.0% 80.0% 80.0% 100.0% 7 % of Bare Metal Stent Patients Receiving DES, 5th Year Following Approval all patients 85.0% 90.0% 49.0% 100.0% 10 all patients with single-vessel 81.7% 90.0% 49.0% 100.0% 9 all patients with multi-vessel 88.4% 90.0% 60.0% 100.0% 8 Table 4 Estimated Percentage of Diabetic BMS Patients Likely to Receive DES by Year Following DES Approval Mean Median Min Max # of Panel Members Responding % of Diabetic Bare Metal Stent Patients Receiving DES, 1st Year Following Approval all patients 43.2% 40.0% 10.2% 90.0% 11 all patients with single-vessel 39.5% 30.0% 10.2% 80.0% 9 all patients with multi-vessel 61.4% 50.0% 30.0% 100.0% 7 % of Diabetic Bare Metal Stent Patients Receiving DES, 2nd Year Following Approval all patients 60.6% 60.0% 25.5% 100.0% 11 all patients with single-vessel 57.9% 50.0% 25.5% 100.0% 9 all patients with multi-vessel 75.0% 75.0% 40.0% 100.0% 7 % of Diabetic Bare Metal Stent Patients Receiving DES, 3rd Year Following Approval all patients 77.9% 80.0% 50.0% 100.0% 11 all patients with single-vessel 75.7% 80.0% 50.0% 100.0% 9 all patients with multi-vessel 81.4% 80.0% 50.0% 100.0% 7 % of Diabetic Bare Metal Stent Patients Receiving DES, 4th Year Following Approval all patients 86.1% 90.0% 60.0% 100.0% 11 all patients with single-vessel 85.2% 90.0% 60.0% 100.0% 9 all patients with multi-vessel 87.1% 90.0% 60.0% 100.0% 8 % of Diabetic Bare Metal Stent Patients Receiving DES, 5th Year Following Approval all patients 90.9% 90.0% 70.0% 100.0% 11 all patients with single-vessel 88.0% 90.0% 70.0% 100.0% 10 all patients with multi-vessel 89.0% 90.0% 70.0% 100.0% 10 Table 5 Estimated Percentage of CABG Patients Likely to Receive DES by Year Following DES Approval Mean Median Min Max # of Panel Members Responding % of CABG Patients Receiving DES, 1st Year Following Approval all patients 12.3% 5.4% 0.0% 50.0% 11 all patients with single-vessel 7.8% 5.1% 0.0% 20.0% 9 all patients with multi-vessel 15.7% 6.0% 0.0% 80.0% 8 % of CABG Patients Receiving DES, 2nd Year Following Approval all patients 17.5% 12.6% 5.0% 50.0% 10 all patients with single-vessel 9.7% 10.0% 2.0% 20.0% 8 all patients with multi-vessel 21.4% 15.1% 5.0% 80.0% 8 % of CABG Patients Receiving DES, 3rd Year Following Approval all patients 31.7% 30.0% 5.0% 90.0% 10 all patients with single-vessel 29.1% 27.7% 2.0% 90.0% 8 all patients with multi-vessel 28.9% 25.0% 5.0% 90.0% 8 % of CABG Patients Receiving DES, 4th Year Following Approval all patients 37.3% 33.0% 5.0% 90.0% 10 all patients with single-vessel 32.0% 30.2% 5.0% 90.0% 8 all patients with multi-vessel 33.2% 30.0% 5.0% 90.0% 8 % of CABG Patients Receiving DES, 5th Year Following Approval all patients 42.1% 40.2% 5.0% 90.0% 10 all patients with single-vessel 35.1% 30.2% 5.0% 90.0% 7 all patients with multi-vessel 37.6% 30.0% 5.0% 90.0% 7 Table 6 Estimated Percentage of Diabetic CABG Patients Likely to Receive DES by Year Following DES Approval Mean Median Min Max # of Panel Members Responding % of Diabetic CABG Patients Receiving DES, 1st Year Following Approval all patients 16.9% 10.0% 1.1% 65.0% 10 all patients with single-vessel 17.4% 5.1% 1.1% 50.0% 7 all patients with multi-vessel 19.5% 10.0% 1.1% 80.0% 9 % of Diabetic CABG Patients Receiving DES, 2nd Year Following Approval all patients 26.0% 17.5% 5.1% 80.0% 10 all patients with single-vessel 21.3% 15.2% 5.0% 50.0% 8 all patients with multi-vessel 21.1% 15.0% 5.1% 80.0% 9 % of Diabetic CABG Patients Receiving DES, 3rd Year Following Approval all patients 33.5% 25.0% 10.0% 90.0% 10 all patients with single-vessel 30.1% 25.2% 5.0% 90.0% 8 all patients with multi-vessel 26.3% 17.5% 5.1% 90.0% 9 % of Diabetic CABG Patients Receiving DES, 4th Year Following Approval all patients 42.6% 40.0% 10.0% 90.0% 10 all patients with single-vessel 37.0% 32.8% 5.0% 90.0% 8 all patients with multi-vessel 34.0% 30.0% 10.2% 90.0% 9 % of Diabetic CABG Patients Receiving DES, 5th Year Following Approval all patients 48.6% 50.0% 10.0% 90.0% 10 all patients with single-vessel 41.4% 45.0% 5.0% 90.0% 8 all patients with multi-vessel 42.6% 37.5% 15.0% 90.0% 10 Tables 3 and 4 present estimated percentage for all CAD patients and diabetic patients (respectively) who are currently receiving bare metal stents but are likely to receive DES once approved. As presented in Table 3 , the mean estimated percentage for all CAD patients in the first year of approval is 24%. A large range was present around this mean, from a minimum of 5.1% to a maximum of 65%. However, the median (22.5%) was similar to the mean, suggesting that outliers did not substantially skew the mean value. Patients with single-vessel disease were less likely to receive DES (18.8%), while those with multi-vessel disease were more likely (32.9%). The proportion of patients projected to receive DES rather than bare metal stents increased with each subsequent year after DES approval. In each year, a greater proportion of multi-vessel disease patients are projected to receive DES than are single vessel disease patients. During the fifth year following approval, the panel estimated that 85% of all bare metal stents patients are likely to receive DES instead. Ranges around the annual mean values continued to be large, with the minimum estimate being 49% and the maximum estimate of 100%. Projected use of DES among diabetic patients who currently receive bare metal stents is presented in Table 4 . Among patients with diabetes, the estimated percentage likely to receive DES is higher than the corresponding values of the overall population. In the first year following DES approval, 43.2% of patients with diabetes who would have received bare metal stents are projected to receive DES instead. While the median proportion of diabetic patients receiving DES in this first year (40%) is similar to the mean, suggesting that outliers do not skew the projections, a very large range of responses was present (10.2% to 90%). The estimated proportion of patients receiving DES rather than bare metal stents in the first year was 39.5% for single vessel disease patients with diabetes, and 61.4% for multi-vessel disease patients. As with the overall population of CAD patients currently receiving bare metal stents (Table 3 ), the proportion of patients with diabetes receiving DES instead of bare metal stents increases in each subsequent year, and the percentage is greater for multi-vessel disease patients than for single vessel disease patients. In the fifth year following DES approval, it is estimated that 90.9% of patients with diabetes who would have received bare metal stents will instead receive DES (range 70% to 100%). Table 5 presents estimates from the cardiologist panel for CABG patients who are likely to receive DES after approval. For each year, the proportion of CABG patients who would instead receive DES is approximately half the proportion of bare metal stent patients who would receive DES instead (Table 3 ). Of all CABG patients, 12.3% are estimated to likely receive DES during the first year following approval in Canada. The range of estimates for receipt of DES rather than CABG was substantial, from a minimum of 0% to a maximum of 50%. The median estimate, 5.4%, is lower than the mean, suggesting that higher estimates may be skewing the mean. In the first year following approval, 7.8% of single vessel disease patients and 15.7% of multi-vessel disease patients would receive DES rather than CABG. In years three through five after DES approval, the estimated proportions of single vessel disease and multi-vessel disease patients likely to receive DES rather than CABG are both less than the proportion among the overall CABG population. This is due to missing data, in that some panel members provided projections only for the overall population and/or one of the population subgroups. In these cases, the relative projections for single vessel and multi-vessel disease patients cannot be directly compared to the estimates for the overall population. Similar to DES adoption among bare metal stent patients, the likelihood of DES use among CABG patients increases with each year after approval. At year five, 42% of all CABG patients are likely to receive DES compared to 85% of bare metal stent patients. In all years except year three, the proportion of single vessel disease CABG patients instead receiving DES is less than the proportion for multi-vessel disease CABG patients. Table 6 provides the estimated rates of DES use for patients with diabetes currently receiving CABG. In the first year following approval, 16.9% of patients with diabetes who would have received CABG are projected to likely to receive DES instead. The range of estimates around this value is large (1.1% to 65.0%). The mean estimate of DES adoption among diabetic CABG patients increased each year, and is larger each year than the corresponding mean estimate for the overall population of patients who would receive CABG. However, the estimated proportion of patients with diabetes receiving DES instead of CABG is less than the estimated proportion for bare metal stent patients. Rates for single and multi-vessel disease patients with diabetes are similar; however, as noted above, missing data makes comparison of these subpopulations to the overall diabetic population difficult. During the fifth year following approval, an estimated 48.6% of diabetic patients who would have received CABG surgery are instead projected receive DES. Recommended use of DES among Canadian CAD patients Tables 3 through 6 present the proportion of bare metal stent and CABG patients who are likely to receive DES rather than these other revascularization procedures. In a final question to the cardiologist panel, we asked for estimates of the proportion of bare metal stent and CABG patients in the overall CAD population who should receive DES rather than these other procedures. In requesting this additional information, the questionnaire specified that respondents could indicate that the proportion of patients who should receive DES is the same as or different from the proportion that are likely to receive DES (as presented in Tables 3 and 5 ). The questionnaire also specified that in estimating the proportion of patients who should receive DES instead of bare metal stents of CABG, panel members should assume that funding is available for this intervention. Thus, this question addressed the projected use of DES in a best-case scenario, without economic restrictions. The estimated percentages of patients who should receive DES are summarized in Table 7 . During the first year of approval, the mean estimated percentage of bare metal stent patients who should receive DES is 42.8%; this is close to double the estimate of the proportion of bare metal stent patients who are likely to receive DES during the first year following approval (24.0%, Table 3 ). The estimated percentage of CABG patients who should receive DES during the first year following approval is 16.8%, an increase of 37% over the proportion of CABG patients likely to receive DES that year (12.3%, Table 5 ). These estimated percentages of patients who should receive DES increase with each year after DES approval. At year five, the panel indicated that 86.8% of bare metal stent patients and 43.7% of CABG patients should be receiving DES. The median responses are very similar to these values, suggesting that outliers are not distorting the presented means. However, while the range around the mean proportion of bare metal stent patients who should receive DES has decreased (minimum 60.8%, maximum 100%), the range around the proportion of CABG patients who should receive DES remains very large (5% to 90%). Thus, there are considerable differences in opinion regarding the appropriate patients to convert from CABG to DES. Table 7 Estimated Percentage of BMS and CABG Patients who Should Receive DES by Year Following DES Approval* Mean Median Min Max % of Patients Who Should Receive DES, 1st Year Following Approval Bare metal stent patients 42.8% 30.0% 5.1% 100.0% CABG patients 16.8% 10.0% 5.0% 50.0% % of Patients Who Should Receive DES, 2nd Year Following Approval Bare metal stent patients 56.0% 50.0% 10.2% 100.0% CABG patients 25.5% 20.0% 5.0% 80.0% % of Patients Who Should Receive DES, 3rd Year Following Approval Bare metal stent patients 70.1% 90.0% 30.0% 100.0% CABG patients 31.4% 30.0% 5.0% 90.0% % of Patients Who Should Receive DES, 4th Year Following Approval Bare metal stent patients 79.7% 90.0% 40.6% 100.0% CABG patients 37.8% 35.0% 5.0% 90.0% % of Patients Who Should Receive DES, 5th Year Following Approval Bare metal stent patients 86.8% 90.0% 60.8% 100.0% CABG patients 43.7% 40.0% 5.0% 90.0% *In answering this question, respondents were asked to assume that funding for DES was available. All questions were responded to by all 11 members of the study panel. Discussion This study presents results from a panel of Canadian cardiologists on treatment patterns for coronary artery revascularization and the potential future adoption of DES in these treatment patterns. Previous reports have debated whether the rate of coronary revascularization in Canada is likely to decrease [ 8 ] or increase [ 9 ] during the present decade. The estimated procedure rates provided by the panel were slightly higher than those from the CCN, suggesting a continued increase in revascularization procedures. In addition, multiple reports have indicated that PTCA is replacing CABG among broad populations of patients requiring coronary revascularization, and CABG is being performed more frequently among higher risk patients [ 10 ]. Based on the panel's responses, it is likely that a trend away from CABG towards PTCA will continue in Canada, and will be augmented by the availability of DES as a treatment option. There are a number of limitations associated with this study. The panel members were recruited from academic medical centers and thus may be more familiar with and more likely to use newer technologies. This may limit the generalizability of the rates provided by the panel to the overall population of Canadian cardiologists and may explain the differences between the panel's estimates and those of the CCN. The study panel was also relatively small; this small sample size may result in estimates that are subject to change if a larger population of cardiologists is surveyed. Despite these limitations, the results of the panel questionnaire indicate that DES will be an important treatment option for Canadian CAD patients, both among patients currently receiving bare metal stents and for patients currently undergoing CABG surgery. It is difficult to assess the validity of these results, as they relate to future events. A recent report by Poses et al. indicated that physicians were likely to underestimate survival for medically managed CAD patients and overestimate the benefits for such procedures [ 11 ]. If this finding applies to the present study, then the rate of DES adoption may be lower than that reported. However, other reports have suggested that coronary revascularization procedures are currently underused, with resulting adverse clinical outcomes [ 12 ]. Even if the adoption rates are lower than the projected values presented in Tables 3 through 6 , DES is likely to be a commonly used treatment modality. Further, the availability of this less invasive yet more efficacious treatment option may address the potential underuse issues, resulting in greater adoption rates than reported by the panel. The recommended adoption rates presented in Table 7 may then be more realistic estimates for the future use of DES. Little information is available regarding the impacts of "converting" patients from CABG to stent implantation. Lee et al. evaluated in the impact of bare metal stent use among patients who were at high operative risk or refused CABG [ 13 ]. In the Lee et al. study, stent implantation was reported to be safe and clinically beneficial [ 13 ]. Use of DES as a treatment option is likely to improve clinical outcomes while maintaining the safety of this less invasive revascularization approach. We requested information separately for the projected use of DES among individuals with diabetes. Previous studies have reported that CAD patients with diabetes have better outcomes following CABG than with PTCA [ 14 , 15 ]. Available data also suggest that use of bare metal stents improves outcomes among patients with diabetes compared to angioplasty alone [ 16 ], although it is unclear whether or not diabetics have worse outcomes following stenting than do non-diabetics [ 7 , 17 ]. While few published data are yet available regarding outcomes among individuals with diabetes following DES implantation, reductions in subsequent restenoses and revascularizations in the general population receiving DES may also occur in the diabetic population. The cardiologist panel felt that DES would become a frequently used treatment option in this population, with adoption rates surpassing those of the overall CAD population. Comparing the estimated proportion of patients who the panel indicated were likely to receive DES versus the proportion the panel reported "should" receive DES provides interesting findings. The proportion of bare metal stent patients that the panel indicated should receive DES (Table 7 ) is greater than the proportion who are likely to receive DES (Table 3 ) for each of the first five years following approval. These estimated proportions become approximately equal at five years following approval (85.0% likely to receive DES, 86.8% should receive DES). A number of factors may influence the difference in proportions between patients who are "likely to" versus "should" receive DES, such as available funding and attitudes towards adoption of new technologies. The estimated proportion of CABG patients who should receive DES (Table 7 ) is greater than the proportion that are likely to receive DES (Table 5 ) during the first two years following approval. For years three through five, the "likely to" and "should" proportions are approximately equal for the CABG population. This more rapid convergence of projected rates may reflect the perceived benefits of the less invasive stenting with DES compared to CABG as well as the potential cost savings from DES versus CABG. A number of reports have indicated that the rate of coronary revascularization procedures in Canada is less than that in the U.S. Bourassa et al. reported that more anginal symptoms were present in Canadian patients prior to revascularization compared to U.S. patients, although Canadian patients apparently experienced greater improvements in quality of life following revascularization procedures [ 18 ]. The availability of DES as a treatment option in Canada may change the threshold at which revascularization procedures are considered. The projected uptake rates presented in Tables 3 through 6 certainly indicate that DES is likely to be used for a substantial proportion of revascularization procedures. It will therefore be important to evaluate the impact of this new technology on patient-reported outcomes, such as satisfaction with treatment, satisfaction with the medical care system (e.g., time until treatment), and change in health-related quality of life. These metrics will help to assess further the potential benefits of DES in Canada. Conclusions Cardiologists at tertiary care hospitals in Canada expect the use of drug-eluting stents (DES) for coronary artery revascularization to increase over the next five years. DES will both be used instead of bare metal stents and an alternative to CABG surgery. This increase in DES use will increase initial procedure-related costs compared to bare metal stents, but is likely to decrease subsequent costs due to the decreased need for repeat revascularizations. Medical care decision makers and planners need to prepare for this increased use, in terms of both facility and budget allocation as well as staffing availability and training. Competing interests This study was performed under a research contract from Boston Scientific Corporation. MH has received research funding from Boston Scientific. ML, MAC, and MV are employees of Boston Scientific Corporation. Boston Scientific is providing the article-process charge for this manuscript. Boston Scientific holds a number of patents on the TAXUS Express2 Paclitaxel-Eluting Coronary Stent System. Publication of this article may result in increased consulting work for Exponent, Inc. Authors' contributions MH directed the overall study and participated in all aspects. ML and MV participated in study design and questionnaire development. MAC participated in data analysis. All authors participated in preparation of the manuscript and read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC544858.xml

545207

KRAS Mutations and Primary Resistance of Lung Adenocarcinomas to Gefitinib or Erlotinib

Background Somatic mutations in the gene for the epidermal growth factor receptor (EGFR) are found in adenocarcinomas of the lung and are associated with sensitivity to the kinase inhibitors gefitinib (Iressa) and erlotinib (Tarceva). Lung adenocarcinomas also harbor activating mutations in the downstream GTPase, KRAS, and mutations in EGFR and KRAS appear to be mutually exclusive. Methods and Findings We sought to determine whether mutations in KRAS could be used to further enhance prediction of response to gefitinib or erlotinib. We screened 60 lung adenocarcinomas defined as sensitive or refractory to gefitinib or erlotinib for mutations in EGFR and KRAS . We show that mutations in KRAS are associated with a lack of sensitivity to either drug. Conclusion Our results suggest that treatment decisions regarding use of these kinase inhibitors might be improved by determining the mutational status of both EGFR and KRAS .

Introduction Genes of the ERBB family encode receptor tyrosine kinases that mediate cellular responses to growth signals. Somatic mutations in the tyrosine kinase domains of two ERBB genes, epidermal growth factor receptor (EGFR) and HER2, have been found in a proportion of lung adenocarcinomas [ 1 , 2 , 3 , 4 ]. For EGFR, mutations are associated with sensitivity to the small-molecule kinase inhibitors gefitinib (Iressa) [ 1 , 2 , 3 ] and erlotinib (Tarceva) [ 3 ]. ERBB signaling pathways include downstream GTPases encoded by RAS genes. Some 15%–30% of lung adenocarcinomas contain activating mutations in the RAS family member KRAS . These mutations are most frequently found in codons 12 and 13 in exon 2 [ 5 , 6 ], and may be associated with unfavorable outcomes [ 7 ]. Interestingly, EGFR and KRAS mutations are rarely found in the same tumors, suggesting that they have functionally equivalent roles in lung tumorigenesis ([ 8 ]; M. Meyerson, personal communication). Furthermore, EGFR mutations are common in tumors from patients who have smoked less than 100 cigarettes in their lifetimes (“never smokers”) [ 3 ], while KRAS mutations more commonly occur in individuals with a history of substantial cigarette use [ 9 ]. We sought to determine whether KRAS mutations could also be used to predict primary sensitivity or resistance to gefitinib or erlotinib. We systematically evaluated 60 lung adenocarcinomas from patients with known responses to either of these drugs for the presence of mutations in EGFR (exons 18 through 21) and KRAS2 (exon 2). Here, we show that mutations in KRAS are associated with primary resistance to single-agent gefitinib or erlotinib. Our results suggest that a determination of mutational status for both EGFR and KRAS may help define which patients are likely to benefit from receiving gefitinib or erlotinib. Methods Tissue Procurement Tumor specimens were obtained through protocols approved by the institutional review board of Memorial Sloan-Kettering Cancer Center, as previously described [ 3 ] (see Protocols S1–S3 ). Tumor material, obtained from patients prior to kinase inhibitor treatment for lung cancer, was collected retrospectively for patients on gefitinib, who received 250 mg or 500 mg orally once daily ( n = 24), and prospectively for patients on erlotinib, who received 150 mg orally once daily ( n = 36). The latter cohort of patients was part of a clinical trial of erlotinib for patients with bronchioloalveolar carcinoma. The analysis presented here includes specimens we previously reported on ( n = 17 for gefitinib and n = 17 for erlotinib) [ 3 ]. All specimens were reviewed by a single reference pathologist (M. F. Z.). Imaging studies were assessed by a single reference radiologist (R. T. H.), who graded responses according to Response Evaluation Criteria in Solid Tumors (RECIST) [ 10 ]. Both observers were blinded to patient outcomes. Eight of nine patients with tumors sensitive to gefitinib had objective partial responses as defined by RECIST, i.e., at least a 30% decrease in the sum of the longest diameters of target lesions, taking as reference the sum measured at baseline. The ninth patient had marked clinical improvement, as ascertained by two independent reviewing physicians and manifested by lessened dyspnea and cancer-related pain. However, this individual had radiographic lesions (pleural and bone metastases) that were deemed nonmeasurable by RECIST criteria. As erlotinib-treated patients were all in a clinical trial, all had disease measurable using RECIST guidelines. For both drugs in this study, tumors were considered refractory if they did not undergo sufficient shrinkage to qualify for partial response. This definition includes patients whose “best overall response” was either progression of disease ( n = 26) or stable disease ( n = 12) as defined by RECIST. No patients had a complete response. Mutational Analyses of EGFR and KRAS in Lung Tumors Genomic DNA was extracted from tumors embedded in paraffin blocks, except for tumor 109T, which was a fresh-frozen tumor specimen. Primers for EGFR analyses (exons 18–21) were as published [ 3 ]. For KRAS analyses, the following nested primer sets for exon 2 were used: huKRAS2 ex2F, 5′- GAATGGTCCTGCACCAGTAA-3′; huKRAS2 ex2R, 5′- GTGTGACATGTTCTAATATAGTCA-3′; huKRAS2 ex2Fint, 5′- GTCCTGCACCAGTAATATGC-3′; and huKRAS2 ex2Rint, 5′- ATGTTCTAATATAGTCACATTTTC-3′. For both EGFR and KRAS, PCR was performed using the HotStarTaq Master Mix Kit (Qiagen, Valencia, California, United States), as per manufacturer's instructions. Use of this method often obviated the need for nested PCR sets. All sequencing reactions were performed in both forward and reverse directions, and all mutations were confirmed by PCR amplification of an independent DNA isolate. In 12 cases, exon 19 deletions were also studied by length analysis of fluorescently labeled PCR products on a capillary electrophoresis device, using the following primers: EGFR -Ex19-FWD1, 5′- GCACCATCTCACAATTGCCAGTTA-3′, and EGFR -Ex19-REV1, 5′-Fam- AAAAGGTGGGCCTGAGGTTCA-3′. Using serial dilutions of DNA from the H1650 non-small-cell lung cancer cell line (exon 19 deletion-positive [ 11 ]), this assay detects the mutant allele when H1650 DNA comprises 6% or more of the total DNA tested, compared to a sensitivity of 12% for direct sequencing. These same cases were also screened for the exon 21 L858R mutation by a PCR–restriction fragment length polymorphism assay, based on a new Sau96I restriction site created by the L858R mutation (2,573T→G). The Sau96I-digested fluorescently labeled PCR products were analyzed by capillary electrophoresis, and the following primers were used: EGFR -Ex21-FWD1, 5′- CCTCACAGCAGGGTCTTCTCTGT-3′, and EGFR -Ex21-REV1, 5′-Fam- TCAGGAAAATGCTGGCTGACCTA-3′. Using serial dilutions of DNA from the H1975 cell line (L858R-positive [ 11 ]), this assay detects the mutant allele when H1975 DNA comprises 3% or more of the total DNA tested, compared to a sensitivity of 6% for direct sequencing (Q. Pan, W. Pao, and M. Ladanyi, unpublished data). Statistics Fisher's Exact Test was used to calculate p- values, and confidence intervals were calculated using Statistics with Confidence software [ 12 ]. Results We identified 60 lung adenocarcinomas from individual patients with tumors shown to be sensitive or refractory to single-agent gefitinib or erlotinib and evaluated these tumors for mutations in EGFR and KRAS . Collectively, nine of 38 (24%) tumors refractory to either kinase inhibitor had KRAS mutations, while zero of 21 (0%) drug-sensitive tumors had such mutations ( p = 0.02) ( Table 1 ). The 95% confidence intervals (CIs) for these observations are 13%–39% and 0%–16%, respectively. Conversely, 17 of 22 (77%) tumors sensitive to either kinase inhibitor had EGFR mutations, in contrast to zero of 38 (0%) drug-resistant tumors ( p = 6.8 × 10 −11 ). The 95% CIs for these observed response rates are 57%–90% and 0%–9%, respectively. All 17 tumors with EGFR mutations responded to gefitinib or erlotinib, while all nine tumors with KRAS mutations did not ( p = 3.2 × 10 −7 ). Table 1 EGFR and KRAS Mutation Status in Lung Adenocarcinomas Sensitive or Refractory to Gefitinib or Erlotinib Lung tumors were examined for mutations in EGFR (exons 18–21) and KRAS (exon 2). In gefitinib-treated patients, six EGFR mutations involved exon 19 deletions that lacked the amino acids Leu-Arg-Glu-Ala, and two were exon 21 amino acid substitutions (L858R). A seventh case with an exon 19 deletion (involving 12 nucleotides) was detected by fluorescent capillary electrophoresis only, so exact sequence deletion information was unavailable (see Methods). In erlotinib-treated patients, three EGFR mutations were exon 19 deletions that lacked amino acids Leu-Arg-Glu-Ala, and five were exon 21 amino acid substitutions (L858R) a One erlotinib-sensitive tumor was unavailable for KRAS examination b The incidence of KRAS mutations in this cohort was low, probably because only cases involving bronchioloalveolar carcinoma were tested (see text) Correlation of EGFR and KRAS mutational status with drug and treatment response is detailed in Table 1 . The spectrum of KRAS mutations is shown in Figure 1 and Table 2 . Results with gefitinib and erlotinib were similar overall. However, the incidence of KRAS mutations in the patients treated with erlotinib was low, probably because of the fact that all patients treated with this drug had bronchioloalveolar carcinoma, which rarely has RAS mutations [ 13 ]. Alternatively, our analyses involving only exon 2 of KRAS2 may have missed some RAS mutations. However, in our analysis of the exonic regions encoding the first 100 amino acids of KRAS in 110 surgically resected early-stage non-small-cell lung cancers, we have found 18 mutations, and all were in either codon 12 or codon 13, encoded by exon 2 (W. Pao, R. Wilson, H. Varmus, unpublished data). Another possibility is that the erlotinib-treated tumors have mutations in other RAS genes, since a minority of RAS mutations in lung cancer have been reported to occur in N- or HRAS [ 5 , 6 ]. Figure 1 Sequence Chromatograms Displaying the Types of KRAS Mutations Found in This Study Table 2 KRAS Exon 2 Mutations Found in Non-Small-Cell Lung Cancers Refractory to Treatment with Gefitinib or Erlotinib a Bold indicates mutations Discussion These results have important clinical implications. First, they extend previous data from our group and others showing that lung adenocarcinomas containing EGFR mutations are associated with sensitivity to gefitinib or erlotinib (17 of 17 in this series; 100% observed response rate; 95% CI, 82%–100%). Second, these data show that tumors with KRAS exon 2 mutations ( n = 9) are associated with a lack of response to these kinase inhibitors (0% observed response rate; 95% CI, 0%–30%). Third, no drug-sensitive tumors had KRAS exon 2 mutations ( n = 21). Whether KRAS mutational status can be used to predict responses to gefitinib or erlotinib in patients whose tumors have wild-type EGFR sequence is still under investigation: our analysis comparing response rates for tumors with neither EGFR nor KRAS mutations versus tumors with wild-type EGFR but mutated KRAS does not reach statistical significance (five of 22 versus zero of nine; p = 0.29). Nevertheless, these findings suggest that patients whose lung adenocarcinomas have KRAS mutations will not experience significant tumor regression with either drug. The incidence of EGFR mutations in tumors responsive to EGFR kinase inhibitors has varied from 71% to 100% ([ 1 , 2 , 3 ] and this paper). Thus, at this point, patients whose tumors test negative for EGFR mutations should not necessarily be precluded from treatment with either gefitinib or erlotinib. Data presented here suggest that clinical decisions regarding the use of these agents in patients with lung adenocarcinomas might be improved in the future by pre-treatment mutational profiling of both EGFR and KRAS . These findings warrant validation in large prospective trials using standardized mutation detection techniques. Supporting Information Protocol S1 Preclinical Studies of Blood, Urine, Bone Marrow, and Tissues Collected from Patients with Thoracic Malignancies (32 KB PDF). Click here for additional data file. Protocol S2 Multicenter Phase II Trial of OSI-774 (Erlotinib, Tarceva) in Patients with Advanced Bronchioloalveolar Cell Lung Cancer (1.9 MB PDF). Click here for additional data file. Protocol S3 Protocol Approval Letters (60 KB PDF). Click here for additional data file. Accession Numbers The LocusLink ( http://www.ncbi.nlm.nih.gov/LocusLink/ ) accession number for the KRAS2 sequence discussed in this paper is 3845; the GenBank ( http://www.ncbi.nlm.nih.gov/Genbank/ ) accession number for the KRAS2 sequence discussed in this paper is NT_009714.16. Patient Summary Background Two drugs, gefitinib (Iressa) and erlotinib (Tarceva), have been developed that can make lung cancers smaller in some patients. The drugs work by blocking the effect of a molecule called the epidermal growth factor receptor (EGFR), which relays instructions to cells to grow and divide. Recently, researchers found that these drugs most effectively shrink tumors that have acquired abnormal variations (mutations) in the EGFR gene. These mutations somehow allow tumor cells to escape normal safety mechanisms that keep cells from growing out of control. Some lung cancers also have mutations in another gene called KRAS . Interestingly, KRAS mutations and EGFR mutations are rarely ever found in the same tumor. Why Was This Study Done? Unfortunately, EGFR mutations are only found in a minority of patients with lung cancer. This means that gefitinib or erlotinib might be given to a lot of patients who may not benefit from this treatment. Ideally, the drugs would be given only to patients who we know will benefit from them. This study examined whether studying the KRAS gene (to see if it had a mutation) could help predict which patients had tumors that would respond well to the drugs. What Did the Researchers Do? They took 60 lung cancer samples from patients who had been treated with one of the drugs and either responded (that is, their tumors shrunk in size) or not, and tested whether the tumors had normal or abnormal KRAS . What Did They Find? Tumors that got significantly smaller while treated with gefitinib or erlotinib (a total of 22) had a normal KRAS gene . Most of these tumors had EGFR mutations. Conversely, tumors that had abnormal KRAS (a total of nine) did not shrink while treated with gefitinib or erlotinib. What Does This Mean? Both gefitinib and erlotinib are expensive and have side effects. Testing for EGFR and KRAS mutations is relatively straightforward, and one could test for abnormalities in both genes first and then decide which patients should be treated with either of the two drugs. What Next? Before doing EGFR and KRAS tests on a routine basis and taking the results into account when making a decision about who should be treated with gefitinib or erlotinib, larger studies need to be done to see whether the results reported here hold up. More Information Online US Food and Drug Administration information page on Iressa: http://www.fda.gov/cder/drug/infopage/iressa/iressaQ&A.htm Cancer Research UK information page about erlotinib: http://www.cancerhelp.org.uk/help/default.asp?page=10296

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC545207.xml

535904

Survival of patients treated with intra-aortic balloon counterpulsation at a tertiary care center in Pakistan – patient characteristics and predictors of in-hospital mortality

Background Intra-aortic balloon counterpulsation (IABC) has an established role in the treatment of patients presenting with critical cardiac illnesses, including cardiogenic shock, refractory ischemia and for prophylaxis and treatment of complications of percutaneous coronary interventions (PCI). Patients requiring IABC represent a high-risk subset with an expected high mortality. There are virtually no data on usage patterns as well as outcomes of patients in the Indo-Pakistan subcontinent who require IABC. This is the first report on a sizeable experience with IABC from Pakistan. Methods Hospital charts of 95 patients (mean age 58.8 (± 10.4) years; 78.9% male) undergoing IABC between 2000–2002 were reviewed. Logistic regression was used to determine univariate and multivariate predictors of in-hospital mortality. Results The most frequent indications for IABC were cardiogenic shock (48.4%) and refractory ischemia (24.2%). Revascularization (surgical or PCI) was performed in 74 patients (77.9%). The overall in-hospital mortality rate was 34.7%. Univariate predictors of in-hospital mortality included (odds ratio [95% CI]) age (OR 1.06 [1.01–1.11] for every year increase in age); diabetes (OR 3.68 [1.51–8.92]) and cardiogenic shock at presentation (OR 4.85 [1.92–12.2]). Furthermore, prior CABG (OR 0.12 [0.04–0.34]), and in-hospital revascularization (OR 0.05 [0.01–0.189]) was protective against mortality. In the multivariate analysis, independent predictors of in-hospital mortality were age (OR 1.13 [1.05–1.22] for every year increase in age); diabetes (OR 6.35 [1.61–24.97]) and cardiogenic shock at presentation (OR 10.0 [2.33–42.95]). Again, revascularization during hospitalization (OR 0.02 [0.003–0.12]) conferred a protective effect. The overall complication rate was low (8.5%). Conclusions Patients requiring IABC represent a high-risk group with substantial in-hospital mortality. Despite this high mortality, over two-thirds of patients do leave the hospital alive, suggesting that IABC is a feasible therapeutic device, even in a developing country.

Background Intra-aortic balloon counterpulsation (IABC) has an established role in the treatment of patients presenting with cardiogenic shock [ 1 - 3 ], refractory heart failure [ 4 , 5 ], ischemia [ 6 ] and arrhythmias [ 7 ] as well as for prophylaxis [ 8 , 9 ] and treatment of complications of percutaneous coronary intervention (PCI). Patients requiring IABC represent a high-risk subset with an expected high mortality [ 10 ]. In an international registry of over 16,000 cases selected from primarily developed nations [ 11 ], the overall adjusted in-hospital mortality was 21.2%. However, there were geographic differences with lower mortality rates in U.S. patients compared to their non-US counterparts (20.1% vs. 28.7%; p < 0.001) [ 12 ]. Major predictors of mortality in these patients include age, gender, and presentation with cardiogenic shock. There is paucity of data on the usage patterns as well as outcomes of patients undergoing IABC in the Indo-Pakistan region. This is partly due to the limited availability and capacity to implant the device as only a few centers in Pakistan have the required logistical as well as technical expertise. Our institution has previously reported on our initial experience of 15 patients undergoing IABC prior to coronary artery bypass graft (CABG) surgery [ 13 ]. We now report on an extended experience with intra-aortic balloon counterpulsation and describe the patterns of usage as well as the independent predictors of in-hospital mortality in patients undergoing IABC. Methods Patient population We reviewed the charts of 95 patients undergoing IABC at the Aga Khan University Hospital (AKUH), Karachi, Pakistan between January 2000 and December 2002. Patients requiring IABC in the operating room immediately following CABG to assist weaning off cardiopulmonary bypass were excluded from this study. However, those patients who underwent IABP implantation prior to surgery were included. The AKUH is a tertiary care hospital located in the metropolitan city of Karachi that receives a mixture of affluent as well as low and middle income patients and serves the entire city as a referral center for patients requiring high-intensity tertiary care. Variables collected included age, gender, indication for IABC (shock or non-shock), history of diabetes, hypertension, smoking, prior PCI or CABG, left ventricular function, refractory ischemia and treatment (revascularization vs. no revascularization). Cardiogenic shock was defined as a systolic blood pressure (SBP) of < 90 mm Hg for at least 30 minutes (or requirement of inotropes to maintain a SBP > 90 mm Hg) associated with hypoperfusion (decreased urine output or cool extremities) and a heart rate of ≥ 60 beats per minute. Left ventricular (LV) ejection fraction (EF) was assessed by visual estimation. LV function was recorded as normal for an EF of ≥ 55%, mildly impaired for an EF 40–54%, moderately impaired for an EF 26–39% and severely impaired if the EF was ≤ 25%. Heart failure was diagnosed using clinical signs as defined by the Framingham criteria [ 14 ]. Refractory heart failure was defined as heart failure failing to respond to therapy including inotropic support. Refractory ischemia was defined as on-going ischemic chest pain and/or dynamic ECG changes (ST depression or ST elevation ≥ 1 mm in two or more contiguous leads) despite adequate medical therapy including antiplatelet drugs, beta-blockers and heparin. The outcome of interest was in-hospital mortality. Statistical methods All variables were entered into Statistical Package for Social Sciences (SPSS) version 10. Means and standard deviations were calculated for continuous variables and frequencies for categorical variables. Variables were analyzed by simple logistic regression to calculate the unadjusted odds ratios for factors associated with in-hospital mortality. Those variables with a p value of ≤ 0.25 on univariate analysis were entered into the multivariable model and adjusted odds ratios for factors associated with in-hospital mortality were calculated. Finally, the model fit was assessed using the Hosmer-Lameshow test. A p value of < 0.05 was considered significant. Results Table 1 summarizes the patient characteristics. The mean age of the study group was 58.8 (± 10.4) years. The majority of subjects were male (78.9%) and a high proportion had hypertension (55.8%), diabetes (43.2%), a smoking history (37.9%), previous PCI (30.5%) or CABG (48.4%). About half (48.4%) of the patients presented with cardiogenic shock and a similar number (52.6%) had moderate or severe depression of left ventricular function at presentation. All except two patients underwent coronary angiography and over two-thirds had three-vessel coronary artery disease. A revascularization procedure (either surgical or PCI) was performed in 74 patients (77.9%). In the remaining 21 patients, the main reasons for not performing revascularization were as follows: diffuse disease not amenable to PCI or CABG (5 patients), CABG felt to be too high-risk on account of comorbid conditions (6 patients), death in the catheterization laboratory prior to revascularization (6 patients), failed PCI (1 patient) and no need for revascularization (3 patients). The overall in-hospital mortality rate in this study group was 34.7% with six patients (6.3%) dying in the laboratory while the remaining 27 (28.4%) died during the hospital stay. Sixty-five patients (65.3%) left the hospital alive. Table 1 Patient Characteristics. Characteristic N (%)* Age (mean/SD) 58.8 (10.4) Males 75 (78.9) Female 20 (21.1) Diabetes 41 (43.2) Hypertension 53 (55.8) Smoking 36 (37.9) Previous PCI 29 (30.5) Previous CABG 46 (48.4) Coronary Anatomy Single vessel disease 6 (6.3) 2-vessel disease 15 (15.8) 3-vessel disease 72 (75.8) LV function – normal or mildly impaired 45 (47.4) LV function – moderate or severely impaired 50 (52.6) Cardiogenic shock 46 (48.4) Underwent revascularization 74 (77.9) Percutaneous 26 (27.4) Surgical 48 (50.5) * mean/Standard Deviation for age; (%) for others LV = left ventricular; PCI = percutaneous coronary intervention; CABG = coronary artery bypass grafting Table 2 shows the indications for the implantation of an intra-aortic balloon pump (IABP). Almost half were inserted for cardiogenic shock. In the univariate analysis (Table 3 ), variables associated with in-hospital mortality included (odds ratio [95% CI]) age (OR 1.06 [1.01–1.11] for every year increase in age; diabetes (OR 3.68 [1.51–8.92]) and cardiogenic shock at presentation (OR 4.85 [1.92–12.2]); left ventricular dysfunction, hypertension and 3-vessel (versus no 3-vessel) coronary artery disease were not significantly associated with in-hospital mortality in these patients. A significant protective effect of a prior history of CABG surgery (OR 0.12 [0.04–0.34]) and in-hospital revascularization, either surgical or percutaneous, (OR 0.05 [0.01–0.189]) was noted in this study. Table 2 Indications for Intra-aortic balloon counterpulsation Indication N (%) Cardiogenic shock 22 (23.2) Cardiogenic shock with mechanical complication 24 (25.3) Left Main disease, no chest pain 9 (9.5) Left Main disease, chest pain in laboratory 6 (6.3) Refractory heart failure 8 (8.4) Refractory Ischemia 23 (24.2) Complication during PCI 2 (2.1) PCI = percutaneous coronary intervention (includes abrupt closure, severe "no-reflow") Table 3 Univariate Predictors of In-Hospital Mortality Survived (%) (n = 62) Died (%) (n = 33) Unadjusted OR (95% CI) p value Age (SD) 56.9 (10.1) 62.5 (10.3) 1.06 (1.01–1.11) * 0.016 Male Gender 51 (82.3) 24 (72.7) 0.58 (0.21–1.57) 0.281 Diabetes 20 (32.3) 21 (63.6) 3.68 (1.51–8.92) 0.004 Hypertension 32 (51.6) 21 (63.6) 1.64 (0.69–3.93) 0.263 Smoking 25 (40.3) 11 (33.3) 1.35 (0.56–3.27) 0.504 Previous PCI 20 (32.3) 9 (27.3) 0.79 (0.31–2.0) 0.616 Previous CABG 40 (64.5) 6 (18.2) 0.12 (0.04–0.34) <0.001 Cardiogenic Shock 22 (35.5) 24 (72.7) 4.85 (1.92–12.2) 0.001 3-vessel disease** 44 (72.1) 28 (87.5) 2.70 (0.83–8.89) 0.101 LV dysfunction *** 30 (48.4) 20 (60.6) 1.64 (0.70–3.87) 0.258 Revascularized 58 (95.1) 16 (48.5) 0.05 (0.01–0.19) < 0.001 SD = standard deviation. PCI = percutaneous coronary intervention. CABG = coronary artery bypass graft. LV = left ventricular * for every 1 year increase in age ** vs. no 3-vessel disease *** moderate/severely impaired LV function vs. normal/mildly impaired In the multivariate analysis (Table 4 ), the significant independent predictors of in-hospital mortality were age (OR 1.13 [1.05–1.22] for every year increase in age); diabetes (OR 6.35 [1.61–24.97]) and cardiogenic shock at presentation (OR 10.0 [2.33–42.95]). Revascularization during hospitalization remained a significant protective factor against mortality (OR 0.02 [0.003–0.12]) The Hosmer-Lemeshow test indicated a good fit for the model (χ 2 6.09; p = 0.637). In the adjusted analysis, a prior history of CABG did not remain a significant predictor of survival primarily because forty-five out of 46 patients underwent revascularization. Table 4 Multivariate Predictors of In-hospital Mortality* Survived (%) (n = 62) Died (%) (n = 33) Adjusted OR (95% CI) p value Age (SD) 56.9 (10.1) 62.5 (10.3) 1.13 (1.05–1.22) * 0.001 Diabetes 20 (32.3) 21 (63.6) 6.35 (1.61–24.97) 0.008 Cardiogenic Shock 22 (35.5) 24 (72.7) 10.0 (2.33–42.95) 0.002 Revascularized 58 (95.1) 16 (48.5) 0.02 (0.003–0.12) < 0.001 * adjusted for gender, previous CABG, hypertension and LV dysfunction (none/mild vs. moderate/severe). Hosmer-Lemeshow χ 2 6.09; p = 0.637 When age as a risk factor was further analyzed by plotting an ROC curve, an age cut-off of 66.5 years had a high specificity for the outcome of in-hospital mortality (specificity 83.9%; area under ROC-curve 0.66; p = 0.01). Thus older patients requiring IABC suffer worse outcomes than younger subjects. The overall complication rate related to the device implantation was low. Eight patients (8.5%) developed limb ischemia necessitating removal of the IABP; however, only one of these eight required surgery. There were no significant bleeding complications although one patient developed a hematoma following removal of the device; however this patient did not require blood transfusion or surgical repair of the arteriotomy site. Discussion Patients requiring IABC are at high risk for death on account of their critical underlying conditions. Despite this, several data have suggested that IABC can improve morbidity and mortality in specific subsets of patients including those presenting with cardiogenic shock. The use of IABC in developing countries is limited on account of lack of equipment as well as skilled personnel who can insert and manage the device. Ours is the first report on a sizeable experience with IABC from the Indo-Pakistan subcontinent. Our experience is similar to that of other centers in the West. We report a high in-hospital mortality rate in patients undergoing IABC (almost 35%). However, given that nearly half of the subjects had cardiogenic shock at presentation, this mortality rate is reasonably acceptable. Advanced age (over 66.5 years), diabetes and cardiogenic shock at presentation were strong independent predictors of in-hospital mortality, while revascularization (either surgical or PCI) was associated with high odds of survival. The latter finding is consistent with recently reported data from the IABP Benchmark Registry [ 15 ]. Of particular interest is the finding that patients with a prior history of CABG were more likely to survive, a finding driven by the fact that the majority underwent revascularization. This suggests that repeat revascularization of patients with a prior history of bypass surgery (a clearly high-risk subset) is not only feasible but also effective in a developing country setting. Our complication rates were acceptably low, supporting the feasibility of using IABC in our setting. Several limitations of this study should be acknowledged. First, the sample size is fairly small and this is reflected in the relatively wide confidence intervals for the odds ratios. Due to a small sample size, it is difficult to make a comparison of correlates of mortality between subgroups, for example those presenting with cardiogenic shock versus those who did not and those undergoing surgical versus percutaneous revascularization. However, as expected, patients presenting with shock had a significantly higher mortality (72.7% vs. 27.3%; p = 0.001). Second, the patient group selected may not be representative of other centers in Pakistan given that our institution is a unique tertiary care hospital in the country. Third, our cohort did not contain patients undergoing prophylactic IABC prior to high-risk PCI for indications other than cardiogenic shock. This probably represents practice patterns at our institution whereby, largely due to cost constrains, very high-risk patients (for example those with multivessel disease and/or severe impairment of LV function) are preferentially send for surgery. The cumulative cost of IABC with multivessel stenting far exceeds that of a bypass operation. Only two patients required emergent IABC during PCI in the study period. This may reflect a selection of lower risk patients for PCI at our institution. Fourth, while the survival rate following IABC is nearly 65%, no analysis has been made of the cost effectiveness of this therapy. Conclusions In conclusion, cardiogenic shock and refractory ischemia are common indications for IABC in a Pakistani setting. Patients requiring an IABP represent a high-risk group with substantial in-hospital mortality. This is consistent with the nature of the presenting illnesses in these patients and is similar to western data. Despite this high mortality, over two-thirds of patients do leave the hospital alive, suggesting that IABC is a feasible therapeutic device, even in a developing country. Age (particularly over 66.5 years), diabetes and cardiogenic shock at presentation are significant predictors of mortality in this group of patients. Revascularization is a significant predictor of survival and complication rates are acceptably low. Larger studies are needed to evaluate which subsets of patients benefit the most from this device and further cost effectiveness analyses are warranted. Competing interests The author(s) declare that they have no competing interests. Authors' contributions FHJ, SAK conceptualized this study and participated in the study design. FHJ performed the statistical analysis. HK, NFM collected the data. KAK, SD, AS were involved in manuscript review. AH, JT and NN participated in manuscript drafting and review. All authors have read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC535904.xml

514699

Alternative mapping of probes to genes for Affymetrix chips

Background Short oligonucleotide arrays have several probes measuring the expression level of each target transcript. Therefore the selection of probes is a key component for the quality of measurements. However, once probes have been selected and synthesized on an array, it is still possible to re-evaluate the results using an updated mapping of probes to genes, taking into account the latest biological knowledge available. Methods We investigated how probes found on recent commercial microarrays for human genes (Affymetrix HG-U133A) were matching a recent curated collection of human transcripts: the NCBI RefSeq database. We also built mappings and used them in place of the original probe to genes associations provided by the manufacturer of the arrays. Results In a large number of cases, 36%, the probes matching a reference sequence were consistent with the grouping of probes by the manufacturer of the chips. For the remaining cases there were discrepancies and we show how that can affect the analysis of data. Conclusions While the probes on Affymetrix arrays remain the same for several years, the biological knowledge concerning the genomic sequences evolves rapidly. Using up-to-date knowledge can apparently change the outcome of an analysis.

Background In a relatively short time microarrays have become a well established technique, widely used by researchers. Microarrays offer nothing less than to be able to monitor simultaneously the expression levels for thousands of genes. The RNA molecules from the biological sample are called targets , and the polymers of nucleic acids fixed on the surface are called probes . The very large number of genes represented on each microarray requires the use of computer based approaches. Although such approaches currently constitute a very rich and active area of research, for many data analyses this step can be summarized simply: under the prior assumption that for the large majority of the genes represented on a microarray the expression will not vary significantly across experiments, the main focus is be to isolate the few genes of interest from the rest. Short oligonucleotide arrays are a particular type of microarrays. Short oligonucleotide arrays are constituted of short probes (oligonucleotides) with several probes designed to match different part of the target sequences. The use of techniques originating from the micro-electronics industry proved very successful in the making of short oligonucleotides arrays [ 1 ], helping the Affymetrix company establish itself as one of the primary manufacturers for microarrays. The pre-processing of oligonucleotide array data differs from other microarray data, specifically because probe intensities associated with each gene are generally summarized by an expression value, or expression index. As it contributes to the computation of the expression values, this step alone is of importance. Different algorithms have been suggested to replace and improve Affymetrix's original algorithms [ 2 ], including E. Lazaridis et al .'s playerout [ 3 ], Li and Wong's model [ 4 ], medianpolish [ 5 ], and Affymetrix's own improvements to its algorithms [ 6 ]. However, information about the individual probes was not disclosed until a few years ago. Only with the release of the probe sequences for a significant number of Affymetrix chips, data analysis approaches considering the nature of individual probes have been made possible. The use of the chemical nature of the probes, on which depends the binding energy with complementary sequences, has already been suggested to improve pre-processing of Affymetrix data at the probe level [ 7 , 8 ]. The annotation for sequenced genomes have progressed considerably since the design of the Affymetrix chips, even the most recent ones, and matching the latest transcriptomic (or genomic) data available with the chip designs is an obvious thing to do. We have performed such a remapping for a few Affymetrix chips, and we show that the resulting probe-to-gene mapping can differ substantially from the original Affymetrix mapping. This can affect the interpretation of experimental data. As annotations of genomes continue to evolve, it is also desirable to have a framework to perform and handle up-to-date probe-to-gene mapping. We provide an open source and documented implementation to do so. Results The results obtained are subdivided in two main categories: the matches between probes and reference sequence obtained, and the difference in the outcome of an analysis when using an alternative mapping. Probes matching multiple RefSeq entries A fair number of probes were found to match several reference sequences, as shown in Figure 1 . For example, the RefSeq NM_001544.2 is found to have 21 matching probes. Eleven of these matching probes also match another reference sequence: NM_022377.1 . A quick look at the annotation reveals that both reference sequences are two different transcripts variants of the same gene 'Homo sapiens intercellular adhesion molecule 4, Landsteiner-Wiener blood group (ICAM4)' and that the same probes are found matching these two sequences. However the ten remaining probes matching NM_001544.2 are also found matching a fairly large number of other reference sequences (from a little less than 300 to almost 600 reference sequences, depending on the probe). We found that these probes are designed to match sub-sequences frequently found in mRNA: ALU repeats. All the probes matching ALU repeats are in the official mapping grouped in a common probe set, called 'human ALU'. Besides 'human ALU' probes, other probes matching multiple reference sequences were found. In that case, the reference sequences matching a given probe have closely related annotations, or even identical annotations. Complex mixtures of partial overlaps for the probe sets can then be observed. As an example, the probe 88322 is found matching the reference sequences NM_017445.1 , NM_003519.3 , NM_003520.3 , NM_003521.2 , NM_003525.2 , NM_003528.2 , NM_080593.1 and XM_301109.1 , annotated 'H2B histone family, member S (H2BFS), mRNA', 'histone 1, H2bl (HIST1H2BL), mRNA', 'histone 1, H2bn (HIST1H2BN), mRNA', 'histone 1, H2bm (HIST1H2BM), mRNA', 'histone 1, H2bi (HIST1H2BI), mRNA', 'histone 2, H2be (HIST2H2BE), mRNA', 'histone 1, H2bk (HIST1H2BK), mRNA' and 'similar to Histone H2B 291B (LOC350694), mRNA' respectively. Such matches in each case require expert annotators to curate the alternative mappings , so we chose to simply ignore the probes matching several reference sequences in the rest of this study. Other multiple matches are more easy to handle, and potentially more harmful when included in an analysis. Some probes are found to hybridize to several unrelated reference sequences, as shown in Figure 2 . Only 290 probes that were labeled mismatches in the official mapping, were found to be legitimate perfect match probes in the alternative mapping. Reference sequences matching all the probes from probe sets We also found a significant number of reference sequences for which all the matching probes belong to one probe set in the official mapping . When a 'one to one' association can be established between a probe set of the official mapping and a probe set in an alternative mapping , which means that a given reference sequence matches all the probes associated with one Affymetrix ID , we conclude a complete agreement between the alternative mapping and the official one (See Figure 3 , top). That is the case for 6274 out of 17426 reference sequences (17426 is the number of reference sequences for which at least one matching probe was found). When the association is 'one-to-many', in the sense that several complete probe sets in the original mapping are matching one reference sequence, one could conclude that, alternative splicing events left aside, some probe sets are redundant (See Figure 3 , bottom). We obtain 1168, 212, 38, 4 and 2 reference sequences for which the matching probes are coming from 2, 3, 4, 5 and 6 original probe sets respectively. The Figure 4 shows that it represents a significant part of the cases. This represents 8% of the original probe sets of a HG-U133A that are potentially redundant. Effect on the outcome of an analysis Naturally the expression values, or expression indexes, computed from the probes intensities are sensitive to differences in the mapping: different probes will give different summary expression values, which can have an effect on the outcome of an analysis. To verify it, we performed a standard exploratory analysis of Affymetrix data (two samples, looking for the genes that are significantly differentially expressed). The probe level intensities were pre-processed and expression values computed. The original mapping was used to obtain a first set of expression values (set Affy), while the alternative mappings made from matching NCBI's reference sequences was used to obtain two more sets of expression values (sets Alt1 and Set Alt2), using all the matching probes or all the probes matching uniquely respectively. In other words, the set Alt2 differs from the set Alt1 in the sense that probes in Alt1 matching several reference sequences were removed from Alt2. The set Affy describes 22283 probe sets, the set Alt1 describes 18076 probe sets (for a total of 184735 probes) and the set Alt2 11640 probe sets (for a total of 153257 probes). The number of mismatch probes in the official mapping that are found in the alternative mappings is low: 290 in Alt1 and 87 in Alt2. For each one of the three sets, 'significantly differentially expressed genes' (SDEGs) are searched for: a Welch's two-sample t-test is performed on all the expression values in each set, and the selection for significant p-values done as described by Ventura and collaborators [ 9 , 10 ] (qvalue set to 1%). The number of SDEGs obtained in the sets Affy, Alt1 and Alt2 are 163, 163, and 103 respectively. Table 1 shows how many of these represent identical probe sets. However, there is also a significant fraction of probe sets for which no simple equivalence could be made. As shown when discussing the matches, the situation is rather complex and a detailed examination for each case would be needed before a comparison between the mappings is possible. The presence of mismatch probes in the alternative mappings does not appear to have much influence. A legitimate concern can be that some of the probe sets in the alternative mappings only contain one or two probes, therefore the results obtained with theses probe sets may be dubious. In fact, only very few of these probe sets contain a small number of probes, as shown in Figure 5 . Moreover, the minimal acceptable number of probes in a probe set has been reported to be lower than the number of probes commonly used [ 11 ]. Validation of the results found with our alternative mappings will have to be done in silico through the curation of the mappings by expert annotators and experimentally with techniques like RT-PCR. Software for inter-exchange mappings We present a complex situation, with new probe sets built on matches between NCBI's RefSeq reference sequences and the sequences of the short oligonucleotide probes found on commercial arrays. This would be of little practical use for the research community without an easy access to the data or the tools used to obtain them. The framework developed in the package 'affy', an open-source and documented collection of data structures and functions for the analysis of GeneChip oligonucleotide arrays at the probe level, is currently used by a growing number of researchers. We take advantage of the features it offers by providing alternative mapping objects that can be 'plugged in', and used instead of the original ones, whenever wanted. The Bioconductor package 'altcdfenvs' contains helping functions, and documentation explaining how to achieve this. Discussion We performed a matching of the Affymetrix probes against the latest reference sequences from NCBI's RefSeq data bank. A number of probes appear to match a large number of reference sequences, hence to match a large number of transcripts. When analyzing a real data set with state-of-the-art processing methods, we observe that the outcome of an analysis can be influenced by inaccuracies in the probes mapping. This is a potential problem when more and more people use 'high-throughput' procedures, to select 'significantly important genes' in an automated or semi-automated fashion. We introduce an alternative mapping between probes and identifiers, our identifiers being NCBI's RefSeq IDs, and offer for download alternative mappings for the Affymetrix chips HG-U95Av2 and HG-U133A. The affy package is a free environment to work with Affymetrix data at the probe level. It has capabilities to include alternative mapping and use them in a simple way. Our work aims at showing potential problems. Biological expertise remains necessary to discuss the exact nature of each match. Affymetrix offers at their NetAffx web site [ 12 ] a tool that allows visualization of probes matching to multiple sequences. As the annotation of the human genome improves over time, our environment allows to update the probe-to-gene mapping accordingly and analyze microarray data using the best biological data and knowledge available. The environment for alternative mappings we present could also be of use when using GeneChips designed for a certain organism with mRNA known to differ (a mutant or a slightly different organism for example). Conclusions We built new 'probe to gene' mappings by matching the sequences for the probes found on human Affymetrix GeneChip arrays against NCBI's RefSeq database of sequences. Solely observing the distribution of the probes matching reference sequences, and comparing the results with the mapping provided by the manufacturer of the chips, we found potential problems such as probes matching several reference sequences and probes from different probe sets in the official mapping matching all the same reference sequence. Depending on the mapping used, the outcome of the data analysis will change as different genes may be selected as differentially expressed. We suggest that a good mapping of probe to genes changes in time, follows the most recent updates in sequence databases (the databases of sequences are constantly modified, with sequences corrected, and sometimes hypothetical gene sequences withdrawn). We like to picture the current situation, where the mappings are frozen, as a Dorian Gray-like syndrome: the apparent eternal youth of the mapping does not reflect that somewhere the 'picture of it' decays. We developed a set of open-source tools, perfectly integrated to an already existing working environment for Affymetrix arrays. The tools are documented and made available to the research community. They let one reproduce our results, and build other mappings. Methods Reference sequences and probe sequences The database of reference sequences used for remapping the Affymetrix chips is NCBI's RefSeq (first release of NCBI's RefSeq, dated June 30th, 2003), freely available for download on the RefSeq website [ 13 ]. Only sequences tagged as Homo sapiens mRNA are considered. The Affymetrix chip types HG-U95Av2 and HG-U133A, both designed for the study of the human transcriptome, were used. While similar features were observed for both, we focus on the HG-U133A to describe our results, as this is the chip with the most recent design. The probe sequences for the chips are freely available on the Affymetrix website [ 14 ], as well as on the Bioconductor website [ 15 ] as meta-data packages. In total, there are 495930 probe sequences on the chips of type HG-U133A (245965 perfect match probes and 245965 mismatch probes). Probe matching Affymetrix GeneChips have several probes per probe set, and a probe set usually represents a gene. Each probe is 25 bases long, and on most arrays probes are grouped in pairs. A probe pair is constituted of a perfect match (pm) probe, designed to match perfect a target gene sequence, and a mismatch (mm) probe, designed to measure non-specific hybridization. The mismatch probe differs from its associated perfect match probe only in the 13th base. In some cases, a probe set represents a fragment of a gene, e.g, 3-prime and 5-prime extremities of the same gene are used as internal control for the efficiency of the reverse transcription. We prefer the term 'probe set ID' to 'gene' since probe sets are not always genes. We call the association 'probes – probe set ID' a mapping . We refer to the grouping of probes in probe sets given by Affymetrix as the official mapping , while a grouping of the probes matching a reference sequence into a probe set is referred to as an alternative mapping . The short length of the probes, as well as several authors reporting a successful use of perfect matches only [ 3 - 5 , 16 ], suggest that the hybridization signals coming from exact matches alone are able to capture the expression signal reliably. We consider all the probes (including mm ) as potential pm probes to perform the matching. The matching of a probe to a sequence is done using the Bioconductor package matchprobes . It performs an exact string matching, as done by the standard C library string: only complete sequence identity between a probe and a fragment of a reference sequence is counted as a match. Experimental data We collected pancreatic tumor tissue from 8 patients and normal pancreatic tissue from 5. Tumor samples were collected from patients undergoing surgery for pancreatic tumors, quick frozen in liquid nitrogen and kept at -80°C until RNA extraction (the study was approved by the ethical committee for Copenhagen). The tissue was homogenized using a Polytron (kinematica. AG, Littau-Luzern, Schwitzerland). 5 μg RNA was extracted from each sample and labeled. The RNA was extracted according to the Trizol protocol (Invitrogen). The samples were applied to Affymetrix HG-U133A GeneChips according to manufacturer's instruction. Clinical results from this study will be published elsewhere (manuscript in preparation). Data processing The probe level data are pre-processed using the affy software package [ 17 ]. No background correction is performed, and probe level intensities are normalized using the vsn [ 18 ] normalization method. Summary expression indexes are computed using the method medianpolish , using only perfect match probes. The exclusive use of perfect match probes is needed to have an identical pre-processing step for the official mapping and alternative mappings , to allow comparison of the results obtained. Availability All the datasets and software used are freely available, and on a wide range of platforms (Linux, Microsoft Windows, MacOS X, other UNIX-like operating systems). The original material used to obtain our results, software and data, is available from the web page: Alternative mappings . The package altcdfenvs, new with the Bioconductor release 1.4 of May 2004 and available on the Bioconductor website, is designed to help researchers to build their own alternative mappings, using their own set of reference sequences and the probe sequences for the Affymetrix chip of their choice. Authors contributions LG conceived, developed and implemented the concept of alternative mapping for Affymetrix chips, and drafted the manuscript. MM and LFH provided original experimental data. SK provided guidance with the choice of a source for reference sequences, and with comments and input on the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC514699.xml

548680

Elevated glutamine/glutamate ratio in cerebrospinal fluid of first episode and drug naive schizophrenic patients

Background Recent magnetic resonance spectroscopy (MRS) studies report that glutamine is altered in the brains of schizophrenic patients. There were also conflicting findings on glutamate in cerebrospinal fluid (CSF) of schizophrenic patients, and absent for glutamine. This study aims to clarify the question of glutamine and glutamate in CSF of first episode and drug naive schizophrenic patients. Method Levels of glutamine and glutamate in CSF of 25 first episode and drug-naive male schizophrenic patients and 17 age-matched male healthy controls were measured by a high performance liquid chromatography. Results The ratio (126.1 (median), 117.7 ± 27.4 (mean ± S.D.)) of glutamine to glutamate in the CSF of patients was significantly (z = -3.29, p = 0.001) higher than that (81.01 (median), 89.1 ± 22.5 (mean ± S.D.)) of normal controls although each level of glutamine and glutamate in patients was not different from that of normal controls. Conclusion Our data suggests that a disfunction in glutamate-glutamine cycle in the brain may play a role in the pathophysiology of schizophrenia.

Background Multiple lines of evidence suggest that a dysfunction in glutamatergic neurotransmission might be involved in the pathophysiology of schizophrenia [ 1 - 6 ]. The amino acid glutamate plays a central role in nitrogen metabolism and participates in multiple biochemical pathways. Released glutamate is taken up by glia, where it is converted to glutamine, transported back to the presynaptic neuron, and reconverted to glutamate [ 6 , 7 ]. Thus, it seems that glutamate-glutamine cycle plays a role in the neuron-glia communication in the synapse, and that impairment of glutamate-glutamine cycle may be implicated in the pathophysiology of schizophrenia [ 1 - 6 ]. By means of in vivo proton magnetic resonance spectroscopy (MRS), a significant increase in glutamine level was detected in the medial prefrontal cortex of never-treated schizophrenic patients compared with controls [ 8 ]. In addition, a recent 4.0 T MRS study demonstrated that levels of glutamine in the left anterior cingulate cortex and thalamus of the never-treated patients with schizophrenia were significantly higher than those of healthy controls [ 9 ]. In contrast, significant lower levels of glutamine were found in the left anterior cingulate cortex of medicated patients with chronic schizophrenia than in the healthy controls, suggesting the role of chronic medication [ 10 ]. Thus, it is possible that the glutamate-glutamine cycle between neuron and glia may play a role in the glutamate hypothesis of schizophrenia. Although Kim et al. [ 11 ] first reported reduction of cerebrospinal fluid (CSF) levels of glutamate in patients with schizophrenia, the findings of subsequent studies are inconsistent, with several report of no alteration in CSF levels of glutamate [ 12 - 14 ]. Furthermore, it was reported that a gradient for glutamate and glutamine in CSF was lack, and that there were significant correlations between the CSF and serum levels of glutamate (r = 0.67, p < 0.05) and glutamine (r = 0.84, p < 0.01)[ 15 ]. Moreover, sodium-dependent neutral amino acids transporters, located in the abluminal membranes of the blood brain barrier, are capable of actively removing neutral amino acids from the brain [ 16 ]. These findings suggest that concentration of neutral amino acids in the extracellular fluid of brain are maintained at ~10% of those of the blood [ 15 , 16 ]. In this study, we investigated whether levels of glutamate and glutamine or ratio of glutamine to glutamate in CSF of first episode and drug naive schizophrenic patients are different from those of age-matched healthy normal controls. Methods Twenty-five male patients with schizophrenia (mean age 26.1 years, range 18–39) and 17 age-matched male healthy subjects (mean age 27.3 years, range 22–44) with no psychiatric disease were enrolled in Uppsala University and Linkoping University Hospital, Sweden. All patients diagnosed according to the DSM-III-R were first episode and drug naive, i.e. they had never been treated with antipsychotic drugs. In the morning (8:00–9:00) from May 1997 to January 1998, CSF of subjects was obtained by lumbar puncture (L4-L5), and twelve to eighteen μL of CSF was collected with a 0.9 mm needle and the samples were immediately frozen at -80°C, as reported previously [ 17 ]. The ethics committee of each institute approved the present study, and we received the informed consent from the participants of the study. Measurement of glutamate and glutamine levels were carried out according to established methods [ 18 ] with a slight modification using a high performance liquid chromatography (HPLC) system with fluorescence detection (Shimadzu Corporation, Kyoto, Japan). Ten μL of the human CSF was added with 10 μL of 0.1 M borate buffer (pH 8.0) and 30 μL of 50 mM 4-fluoro-7-nitro-2,1,3-benzoxadiazole (NBD-F; Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan) in CH 3 CN. The reaction mixture was then heated at 60°C for 2 min, and immediately supplemented with 100 μL of H 2 O/CH 3 CN (90/10) containing 0.1 % trifluoroacetic acid (TFA) to stop the reaction. Ten μL of the resultant solution was injected into the HPLC system. A reversed-phase ODS column (TSKgel ODS-80Ts, Tosoh Corporation, Tokyo, Japan) was used for the separation and quantification of glutamate and glutamine, and the gradient elution of the mobile phase was kept at a constant flow rate of 0.8 mL/min. Mobile phase 1a consisted of H 2 O/CH 3 CN (90/10) containing 0.1 % TFA, and phases 1b and 1c, of H 2 O/CH 3 CN (10/90) containing 0.1 % TFA and CH 3 CN, respectively. The time program for gradient elution was programmed as follows: 0–50.5 min 1a: 1b : 1c = 95 : 5 : 0, 50.5–55.5 min 1a : 1b : 1c = 0 : 100 : 0, and 55.5–57 min, 1a : 1b : 1c = 0 : 0 : 100. The column temperature of all columns was maintained at 35°C. Fluorescence detection was made at 530 nm with an excitation wavelength at 470 nm. Differences between two groups were analyzed using the Mann-Whitney U-test. The relationship between two variables was examined using Spearman correlation coefficients. A p < 0.05 was considered significant. Results The CSF levels (421.7 μM (median), 468.1 ± 146.1 μM (mean ± S.D.), 254.0–775.1 (range)) of glutamine in the patients were not different (z = -1.038, p = 0.299) from those (410.5 μM (median), 405.6 ± 108.6 μM (mean ± S.D.), 219.8–689.0 (range)) of normal controls. The CSF levels (4.17 μM (median), 4.25 ± 1.77 μM (mean ± S.D.), 2.22–8.88 (range)) of glutamate in the patients were not different (z = -1.307, p = 0.191) from those (5.26 μM (median), 4.73 ± 1.29 μM (mean ± S.D.), 2.54–6.51 (range)) of normal controls. However, the ratio (126.1 (median), 117.7 ± 27.4 (mean ± S.D.), 42.0–161.6 (range)) of glutamine to glutamate in the CSF of patients was significantly (z = -3.29, p = 0.001) higher than that (81.01 (median), 89.1 ± 22.5 (mean ± S.D.), 59.7–134.0 (range)) of controls (Table 1). Furthermore, we found significant correlations between glutamate and glutamine in normal controls (r = 0.549, p = 0.022) or patients (r = 0.780, p < 0.001). Discussion In this study, we found that the ratio of glutamine to glutamate in the CSF of first episode and drug naive schizophrenic patients was significantly higher than that of normal controls although each level of glutamine and glutamate in the CSF of patients was not significantly different from that of normal controls. To our knowledge, this is a first report demonstrating that the ratios of glutamine to glutamate in the first episode and drug naive patients are significantly higher than those of normal controls. In contrast, it was supposed earlier that alterations in CSF levels of glutamate are not so prominent compared with those in the brain [ 14 ]. Therefore, it is likely that a difference in glutamate (or glutamine) levels between our CSF study and MRS studies may be due to the difference between CSF samples and specific corticolimbic regions. However, it should be noted that alterations in the ratio of glutamine to glutamate are detected in the CSF samples of first episode and drug naive schizophrenic patients, suggesting an abnormality of the glia-neuronal glutamate-glutamine cycle in the brain of patients with schizophrenia. In general, glutamine is synthesized in astrocytes from glutamate by the enzyme glutamine synthetase, found exclusively in brain glia cells. Glutamine then crosses the astrocytes to be transported into nerve cell terminals, where it is converted again into the neurotransmitter glutamate by glutaminase. It is reported that activities of glutaminase and glutamic acid decarboxylase (GAD; the rate-limiting enzyme in the synthesis of GABA by decarboxylation of glutamate) are significantly greater in the dorsolateral prefrontal cortex (DLPFC) of schizophrenia than in the control group, whereas activities of glutamate dehydrogenase, glutamine synthetase, and GABA transaminase in the DLPFC of schizophrenia are not different from the control group [ 19 ]. These findings suggest that metabolism of glutamate and GABA might be altered in the DLPFC of schizophrenic patients. Furthermore, it has been reported that activity of glutamine synthetase and glutamate dehydrogenase, the key enzymes involved in glutamate-glutamine cycle between neuron and glia, were markedly altered in the prefrontal cortex of schizophrenic patients, suggesting abnormalities in the function of glutamate-glutamine cycle in schizophrenic brain [ 20 ]. It is also well known that the glutamate-glutamine cycle between neuron and glia is tightly related to glutamate neurotransmission, glutamatergic function, and their regulation in human brain [ 7 ]. Taken together, it is likely that a dysfunction in glutamate-glutamine cycle in the brain may play a role in the pathophysiology of schizophrenia, supporting the glutamate hypothesis of schizophrenia. As described in introduction, sodium-dependent amino acids transporters, located in the abluminal membranes of the blood brain barrier, are capable of actively removing amino acids from the brain [ 16 , 20 , 21 ]. Sodium-dependent amino acids transporter are capable of pumping both glutamine (system N) and glutamate (glutamate transporters EAAT-1, 2, and -3) from the extracellular fluid into endothelial cells [ 20 , 21 ]. The luminal facilitative carriers for both glutamate and glutamine can then transport them to the blood [ 16 , 20 , 21 ]. Therefore, the concentrations of naturally occurring amino acids in the CSF [presumably similar to the extracellular fluid of brain] are ~10% of those of the blood [ 15 , 16 ]. Taken together, it seems that alteration in the transport mechanisms regulating levels of glutamate and glutamine in CSF may be implicated in elevated glutamine/glutamate ratio in CSF of schizophrenic patients although further study is necessary. Conclusion Our findings suggest that a dysfunction in glutamate-glutamine cycle between neuron and glia may play a role in the pathophysiology of schizophrenia, supporting the glutamate hypothesis of schizophrenia. Competing interests The author(s) declare that they have no competing interests. Authors' contribution KH conceived of the study, its design and coordination, and edited the manuscript. GE participated in the design of the study. CN and LHL recruited subjects and collected CSF samples. ES and MI assisted HPLC analysis and data analyses. All authors read and approved the final manuscript. Figure 1 Levels of glutamine and glutamate, and ratio of glutamine to glutamate in CSF of normal controls, and first episode and drug naive schizophrenic patients. (A) CSF levels of glutamine in patients were not different from those of normal controls. (B) CSF levels of glutamate in patients were not different from those of normal controls. (C) Ratios of glutamine to glutamate in patients were significantly higher than those of normal controls. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548680.xml

514706

Tamoxifen and the Rafoxifene analog LY117018: their effects on arachidonic acid release from cells in culture and on prostaglandin I2 production by rat liver cells

Background Tamoxifen is being used successfully to treat breast cancer. However, tamoxifen also increases the risk of developing endometrial cancer in postmenopausal women. Raloxifene also decreases breast cancer in women at high risk and may have a lower risk at developing cancer of the uterus. Tamoxifen has been shown to stimulate arachidonic acid release from rat liver cells. I have postulated that arachidonic acid release from cells may be associated with cancer chemoprevention. Methods Rat liver, rat glial, human colon carcinoma and human breast carcinoma cells were labelled with [ 3 H] arachidonic acid. The release of the radiolabel from these cells during incubation with tamoxifen and the raloxifene analog LY117018 was measured. The prostaglandin I 2 produced during incubation of the rat liver cells with μM concentrations of tamoxifen and the raloxifene analog was quantitatively estimated. Results Tamoxifen is about 5 times more effective than LY117018 at releasing arachidonic acid from all the cells tested. In rat liver cells only tamoxifen stimulates basal prostaglandin I 2 production and that induced by lactacystin and 12-O-tetradecanoyl-phorbol-13-acetate. LY117018, however, blocks the tamoxifen stimulated prostaglandin production. The stimulated prostaglandin I 2 production is rapid and not affected either by preincubation of the cells with actinomycin or by incubation with the estrogen antagonist ICI-182,780. Conclusions Tamoxifen and the raloxifene analog, LY117018, may prevent estrogen-independent as well as estrogen-dependent breast cancer by stimulating phospholipase activity and initiating arachidonic acid release. The release of arachidonic acid and/or molecular reactions that accompany that release may initiate pathways that prevent tumor growth. Oxygenation of the intracellularly released arachidonic acid and its metabolic products may mediate some of the pharmacological actions of tamoxifen and raloxifene.

Background The successful treatment and prevention of estrogen-dependent breast cancer in women by tamoxifen is attributed to its estrogen receptor (ER) occupancy [reviewed in [ 1 , 2 ]]. In the N-nitrosomethylurea (NMU) induced breast cancer model in rats, tumor growth is estrogen dependent and tamoxifen is considerably more effective than raloxifene [ 3 ]. In the dimethylbenzanthracene (DMBA)-induced model in rats, in which tumor growth is predominantly dependent on prolactin for growth, tamoxifen and raloxifene show effective anti-tumor action. Tamoxifen and raloxifene have several properties in common; e.g. prevention of tumors in the DMBA induced rat mammary model, maintenance of bone density in the ovariectomized rat and reduction of low density lipoprotein cholesterol. The partial estrogen agonist activity of tamoxifen on uterine tissue, however, increases the risk of developing endometrial cancer. This does not appear to occur with raloxifene. Tamoxifen stimulates arachidonic acid release from rat liver cells [ 4 ]. In this report, I have compared tamoxifen and the raloxifene analog LY117018 for effectiveness at releasing arachidonic acid (AA) from rat liver, rat glial, human colon carcinoma and human breast carcinoma cells and their effects on prostaglandin (PG) I 2 production by the rat liver cells. Although both compounds release AA from these cells, LY117018 is less effective. Only tamoxifen stimulates both basal and PGI 2 production induced by incubation of rat liver cells with lactacystin in the presence of 12-O-tetradecanoyl-phorbol-13-acetate (TPA). LY117018, however, inhibits the PGI 2 production stimulated by tamoxifen. The intracellular release of AA and/or the cellular reactions that accompany that release may initiate pathways that prevent tumor growth. The tissue specific effects of tamoxifen and LY117018 may be associated with the AA or with cyclooxygenase (COX) activity and/or one of the many bioactivities resulting from oxygenation and metabolism of the released AA. Methods The C-9 rat liver and BT-20 human breast carcinoma cells were purchased from the American Type Culture Collection (Manassas, VA, USA) and maintained in MEM supplemented with 10% fetal bovine serum. The C-6 rat glial cell line was obtained from Dr. Elaine Y. Lai in the Department of Biology, Brandeis University and maintained in medium 199. The human colon carcinoma cells (HT-29) were obtained from Dr. Basil Rigas, American Health Foundation, Valhalla, NY and maintained in McCoy's medium. [ 3 H]AA (91.8 Ci/mmol) was purchased from NEN Life Science Products, Inc. (Boston, MA, USA); ICI-182,780 from Tocris Cookson, Inc. (Ballwin, MO, USA); tamoxifen and 4-OH-tamoxifen were from Sigma Chemical Co. (St. Louis, MO, USA). LY117018 was obtained from Dr. David A. Cox, Eli Lilly and Co. (Indianapolis, IN, USA). Raloxifene was extracted from EVISTA ® tablets with dimethylsulfoxide. Two days prior to experiments, the cells were treated with 0.25% trypsin-EDTA and, after addition of minimum essential medium (MEM), medium 199 or McCoy's medium containing 10% fetal bovine serum, the floating cells were seeded onto 35 mm culture dishes. The plating densities varied from 0.1 to 0.5 × 10 5 cells/35 mm dish. The freshly seeded cultures were incubated for 24-h to allow for cell attachment. After decantation of incubating media, 1.0 ml fresh media (MEM for the rat liver and BT-20 cells, medium 199 for the rat glial or McCoy's for the HT-29 cells respectively) containing 10% fetal bovine serum and [ 3 H] AA (0.2 μCi/ml) was added and the cells incubated for 24-h. The cells were washed 4 times with MEM and incubated for various periods of time with 1.0 ml of MEM, medium 199 or McCoy's containing 1.0 mg bovine serum albumin (BSA)/ml and different concentrations of each test compound. The culture fluids were then decanted, centrifuged at 2000 × g for 10 min, and 200 μl of the supernate counted for radioactivity. Radioactivity recovered in the washes before the 6-h incubation was compared to input radioactivity to calculate the % radioactivity incorporated into the cells [ 5 ]. As measured by thin layer chromatography (TLC), of the [ 3 H] released from radiolabelled methylcholanthrene transformed fibroblasts (about 20% after a 3-h stimulation by serum), 92% was AA, 4% was PGE 2 , 0.6% and 1% were PGF 2α and phospholipids respectively [ 6 ]. When labelled to equilibrium, the [ 3 H] AA had been incorporated into phosphatidylcholine (50%), phosphatidylethanolamine (36%), phodphatidylserine (9%) and triglycerides (10%) [ 7 ]. Such distribution varied among 12 cell lines [ 8 ]. The [ 3 H] AA label released from human colorectal cancer cell lines (HCT116 and SW180) was AA as measured by TLC [ 9 ]. For PGI 2 production by the rat liver cells, 1.0 ml of MEM supplemented with 10% fetal bovine serum, lacking [ 3 H] AA, was added after the first 24-h incubation. The cells were incubated for another 24-h, washed three times with MEM, then incubated with lactacystin and TPA in the presence and absence of the test compounds in MEM/BSA for various periods of time. The culture fluids were decanted and analyzed for 6-keto-PGF 1α , the stable hydrolytic product of PGI 2 , by radioimmunoassay [ 10 ]. At the sub-confluent cell densities used in these experiments (about 5 × 10 4 or less per dish) only the major COX product of rat liver cells (97% 6-keto-PGF 1α ) could be measured. The other two COX products, PGE 2 (2%) and PGF 2α (1%) could not be measured. The release of [ 3 H] AA is presented as a percentage of the radioactivity incorporated by the cells. Except for the time-course experiments, which used duplicate dishes (Fig. 5 ), three to six culture dishes were used for each experimental point. The data are expressed as mean values ± SEM. The data were evaluated statistically by the unpaired Student's t-test . A P value < 0.05 was considered significant. Results AA Release The release of AA after incubation of the rat liver cells with tamoxifen, 4-OH-tamoxifen, LY117018 and raloxifene for 6 h is shown in Fig. 1-A . Tamoxifen is the most effective: it releases AA at 10 μM. 4-OH-tamoxifen, an active metabolite of tamoxifen [ 2 ] is also effective. Raloxifene and LY117018 release AA from the rat liver cells, but are about 5 times less potent (at the doses tested) than tamoxifen. At μM concentrations, all induce apoptosis [ 11 , 12 ]. Tamoxifen, but not 4-OH-tamoxifen, has also been reported to induce apoptosis in p53(-) normal human mammary epithelial cells [ 13 ]. Yet, the metabolite binds to the ER with higher affinity than tamoxifen [ 14 , 15 ]. The release of AA by tamoxifen and LY117018 is not unique to the rat liver cells. Both release AA from rat glial cells (Fig. 1-B ). Tamoxifen is more effective at releasing AA than is LY117018. The highest level of [ 3 H] AA released into the medium which contains BSA (1 mg/ml) would be 0.25 nM (the specific activity of the [ 3 H] AA is 92 Ci/mmole). Since TLC or HPLC analyses were not done, the concentration of the total released AA could not be quantitatively estimated. However, even treatment of a colorectal cancer cell line with 200 μM AA for 48-h leads to apoptosis [ 9 ]. The release of AA is neither species nor tissue specific. Both drugs also release AA from human breast carcinoma (BT-20) and human colon carcinoma (HT-29) cells (Fig. 2-A and 2-B ). Again, LY117018 is less effective than tamoxifen. PGI 2 production Of the cultured cells studied, the COX mediated metabolic profile of AA has been described only in the rat liver cells. Thus, the effects of tamoxifen and LY117018 on PGI 2 production by these cells were determined. Tamoxifen stimulates basal PGI 2 production as well as that induced by lactacystin in the presence of TPA. LY117018 and raloxifene (Fig. 3-A ) are ineffective in stimulating basal PGI 2 production. LY117018, at a concentration that does not affect PGI 2 production (50 μM), inhibits the PGI 2 production stimulated by lactacystin in the presence of TPA (Fig. 3-B ). The increased PGI 2 production probably reflects some stimulation of both induced and basal COX activity. In mouse neuronal and embryonic rat mesencephalic cells, COX-2 is induced by lactacystin as measured by western and northern blot analyses [ 16 ]. Identification of the induced and basal isoforms stimulated by tamoxifen in these rat liver cells was deduced from the relative effectiveness of inhibition of PGI 2 production by celecoxib and piroxicam. Celecoxib is 7.5 times more effective than piroxicam as an inhibitor of COX-2, but 30,000 times less effective as an inhibitor of COX-1. Piroxicam is a selective inhibitor of COX-1 [ 17 ]. Celecoxib, the nonsteroidal anti-inflammatory drug (NSAID) that is a selective inhibitor of COX-2, inhibits both induced and basal COX activity 20–80 times more effectively than piroxicam (Fig. 4-A and 4B ). Thus, based on inhibition by these two NSAIDs, COX-2 is the most likely isoform expressed both constitutively and after induction by lactacystin in the presence of TPA in rat liver cells. This conclusion was confirmed by western blot analyses (Levine L, Tashjian A, unpublished data). Is COX-2 production non-genomic and ER-independent? The time-course of stimulation by tamoxifen (15 μM) of basal activity is shown in Fig. 5 . Even after a 5-minute incubation (not shown in Fig. 5 ), stimulation by tamoxifen is significant statistically [0.08 ± 0.011 (n = 5) vs 0.13 ± 0.011 (n = 5) ng 6-keto-PGF 1α per ml in the absence and presence of tamoxifen, respectively]. Similar to AA release [ 4 ], stimulation of basal PGI 2 production by tamoxifen is not affected by preincubation of the cells for 2-h with 1 μM actinomycin D. The induction obtained by incubation with lactacystin in the presence of TPA is inhibited (Fig. 6-A ). Stimulation of basal PGI 2 production by tamoxifen is unaffected by ICI-182,780 (Fig. 6-B ). This estrogen antagonist, which binds to the ER with high affinity [ 18 ] did not affect the release of AA by tamoxifen [ 4 ] but does partially inhibit release of AA stimulated by 17β-estradiol, 22 (R)-cholesterol, indomethacin, trans -retinoic acid and the tyrosine analog of thiazolidinedione, GW7845 [ 19 ]. The rapidity of stimulation (5 min) and the lack of effect of actinomycin D or ICI-182,780 suggest that the stimulation of PGI 2 production by tamoxifen is non-genomic and does not require ER occupancy. LY117018 blocks PGI 2 production stimulated by tamoxifen (Fig. 7 ). This blockade is unaffected by preincubation of the cells for 2-h with either actinomycin D or ICI-182,780 (data not shown). LY117018, however, does not inhibit the release of AA stimulated by tamoxifen; their combined effect is at least additive (Fig. 8 ). Discussion These studies demonstrate that, at μM concentrations, tamoxifen and the raloxifene analog, LY117,018 stimulate the release of AA from cells in culture. At nM levels, the release is not observed. The effectiveness of these two compounds at prevention of estrogen-dependent breast cancer reflects competition for the ER [ 1 , 2 ]. In addition to occupancy of the ER, I am postulating that these drugs, at μM concentrations, may also prevent breast cancer of estrogen independent tumors. Consistent with this hypothesis are the findings that even at 100 μM concentrations, ethanol extracts of Femora ® (letrozole) or Aramidex ® (anastrozole) do not release AA from cells in culture (unpublished data). These two drugs inhibit estrogen synthesis by blocking aromatase enzymes and also prevent estrogen-dependent breast cancer [ 20 ]. In view of the stimulations of AA release by tamoxifen and LY117018 from human colon carcinoma cells (Fig. 2-A ), the occurrence of colon cancer in women undergoing long term treatment with high levels of tamoxifen would be of interest. A mechanism that most simply explains the release of AA by tamoxifen and LY117018 is the ability of such compounds to intercalate into cell membranes and affect phospholipase activities. The release of AA from endothelial cells by the Ca 2+ ionophore A-23187 reflects phospholipase activities. It is regulated by phosphorylation of the enzyme [ 21 ]. The enzymic properties of the altered membrane may impact signaling mechanisms e.g., pathways leading to apoptosis and COX induction. The intracellularly released AA also can serve as substrate for oxygenases. AA has intrinsic biologic activities that may also affect signaling pathways. Such changes would depend on the lipophilic properties of the compound and the composition of a particular membrane [ 22 ] and would vary from cell type to cell type, organelle to organelle and with the growth phase of the cell. The AA release by tamoxifen and other reagents studied in my laboratory occurs with μM concentrations [ 4 , 5 , 19 , 23 , 24 ]. These experiments were carried out in the presence of BSA (1.0 mg/ml), and therefore do not differentiate between the protein-bound and free reagent. Thus, they are likely to be overestimated values. Nevertheless, the possibility that general necrotic cell death may cause AA release, must be considered. Tamoxifen, was found not to be toxic at concentrations of 10 to 20 μM for A549 human lung adenocarcinoma (ER-negative) cells [ 25 ]. Nor was 10 μM tamoxifen toxic when tested on rat glial cells and breast cancer MCF-7 cells [ 26 ]. Even when cell viability of three different breast cell lines (ER-positive MCF-7; ER-negative MDA-MB-239 and ER-negative BT-20 cells) was measured after incubation with 25 μM tamoxifen for 24-h, the loss in viability was due to apoptosis [ 12 ] and was not the result of necrotic cell death. Concentrations of tamoxifen used in this report are comparable to those found to induce apoptosis, not necrotic cell death. The median concentration of tamoxifen and its metabolites for clinical effectiveness in the treatment of breast cancer varies from 0.8 μM to 2.4 μM, depending on the age of the woman [ 27 ]. When measured after a 6-h incubation, 10 μM tamoxifen stimulates deesterification of membrane phospholipids as measured by extracellular AA release, (Fig. 1 and 2 ). The 6-h incubation may not be optimum for apoptosis to be observed, e.g. after a 6-h incubation of 5 μM tamoxifen with breast cancer cells, apoptosis was induced in about 10% of the cells compared to 8% in control cells [ 12 ]. After 24- and 48-h incubation, apoptosis in the tamoxifen treated cells increased to 40 and 70%, respectively. Apoptosis, induced by this membrane perturbation after 6-h incubation, increased 4 to 7 fold after longer incubation. It is apoptosis that may mediate the cancer preventative action of tamoxifen [ 11 , 12 ]. Some tissue specific effects of tamoxifen and raloxifene, e.g. on endometrial cells of the uterus, may be related to COX activity. At the low cell densities used in the present studies, the PGI 2 produced by the rat liver cells can be quantitatively determined. The effects of tamoxifen and LY117018 on the COX activity of the rat liver cells may be similar to their effects on other cells that express COX. Tamoxifen and LY117018 affect COX activity differently; only tamoxifen stimulates PGI 2 production. AA, which regulates the production of lipoxygenase, COX and cytochrome P-450 epoxygenase products could impact many of the pharmacological actions of these two selective estrogen receptor modulators. AA can be oxygenated by COX isoforms, lipoxygenases, and cytochrome P-450 epoxygenases and their products converted to the prostaglandins, leukotrienes, epoxyeicostetranoic acids [ 28 ] and AA to the oxidative stress-related isoprostanes [ 29 ]. In the COX pathway, enzymes convert PGH 2 to the major physiologically active products, PGD 2 , PGE 2 , PGF 2α , PGI 2 and thromboxane A 2 [ 28 ]. They, in turn, can be converted enzymatically and nonenzymatically to other biologically active compounds that possess different pharmacological properties. Often, only one major product is synthesized by the cell [ 8 ]. Thus, tissues can be affected differently by changes in COX activity. AA or its metabolites, would have different biological effects depending upon the genetic capabilities of the individual cell being affected. In addition to the spectrum of activities of the prostanoids, oxygenation of AA by 5-, 12- and 15-lipoxygenase and epoxygenases yield other products located in different cells and tissues [ 30 , 31 ]. Competing interests None declared. Author's contributions L.L. is the sole author and all of the experiments, with the exception of the western blots referred to as Levine L and Tashjian A (unpublished data), were conducted by L.L. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC514706.xml

387282

Evaluating Disease Trends in Marine Ecosystems

After the recent mad cow scare in the United States, 61% of Americans said they would start eating more fish, according to a Wall Street Journal Online poll. The respondents may not know that populations of large predatory fish, such as tuna, swordfish, and marlin, have declined 90% over the past 50 years or that less-prized species are increasingly overfished. Or that ever more fish and seafood species show rising levels of mercury contamination, rendering them unfit for human consumption—and contaminating other organisms in the ocean food chain. Humans are also affecting marine life in unexpected ways, as when large numbers of seals in Antarctica in 1955 and in Siberia in 1987 succumbed to canine distemper virus, presumably contracted from domestic dogs. In 2000, more than 10,000 Caspian seals—which also had contact with domestic dogs—died of the same virus. Such human incursions cause even more damage by exacerbating the effects of naturally occurring parasitic and pathogenic diseases that already wreak havoc as they ripple through the food chain. A dead gorgonian sea fan on a wall in Palau (Photograph, with permission, by Drew Harvell) With recent studies suggesting that disease rates have increased over the past 30 years—and are expected to increase even more, thanks to global climate change—prospects for protecting marine ecosystems depend on understanding the causes and nature of these disease outbreaks. While all indicators point to a real increase in disease rates, scientists have no baseline data to measure these increases against and so cannot directly test the hypothesis that marine diseases are increasing. Now Jessica Ward and Kevin Lafferty report a method that uses the recorded incidence of disease as a proxy for baseline data to identify disease trends in major groups of marine organisms. Ward and Lafferty conducted an online search of 5,900 journals published from 1970 to 2001 for reports of disease in nine taxonomic groups: turtles, corals, mammals, urchins, mollusks, seagrasses, decapods (crustaceans), sharks/rays, and fishes. Their approach takes into account three potentially confounding factors in determining trends in this type of search. Fluctuations in publication numbers could skew results, since an increase in the number of scientific reports published in a particular taxonomy might not reflect a true increase in the incidence of disease; a particularly prolific author could bias the search results by turning up more cases of disease in a population than actually occurred; or a single disease event reported multiple times in different papers could create the impression that disease had suddenly increased. To normalize publication rates over time, Ward and Lafferty used a proportion of disease reports from a given population relative to the total number of reports in that group. To determine whether there was an “author effect,” they removed the most prolific author in each taxonomic group and found that an author's abundant contributions did not skew the results. Finally, they confirmed that a single disease didn't bias their results by removing multiple reports of the same disease from the literature before analyzing the trends. When they analyzed the searches without adjusting for the total number of reports published, Ward and Lafferty found that reports of disease increased for all groups. But when they analyzed the normalized results, they found that trends varied. While there was a clear increase in disease among turtles, corals, mammals, urchins, and mollusks, they found no significant trends for seagrasses, decapods, and sharks/rays. And they found that disease reports actually decreased for fishes. (One explanation for this decrease could be that drastic reductions in population density present fewer opportunities for transmitting infection.) Ward and Lafferty tested the soundness of this approach by using a disease (raccoon rabies) for which baseline data exist and showing that normalized reports of raccoon rabies increased since 1970, just as the disease increased from one case reported in Virginia in 1977 to an “epizootic” outbreak, affecting eight mid-Atlantic states and Washington, D.C., by 1992. The pattern of increased reports, the authors propose, confirms scientists' perceptions about the rising distress of threatened populations and thus reflects a real underlying pattern in nature. The fact that disease did not increase in all taxonomic groups suggests that increases in disease are not simply the result of increased study and that certain stressors, such as global climate change, likely impact disease in complex ways. By demonstrating that an actual change in disease over time is accompanied by a corresponding change in published reports by scientists, Ward and Lafferty have created a powerful tool to help evaluate trends in disease in the absence of baseline data. It is only by understanding the dynamics of disease outbreaks that scientists can help develop sound methods to contain them.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC387282.xml

524174

Regulation of membrane fatty acid composition by temperature in mutants of Arabidopsis with alterations in membrane lipid composition

Background A wide range of cellular responses occur when plants are exposed to elevated temperature, including adjustments in the unsaturation level of membrane fatty acids. Although membrane bound desaturase enzymes mediate these adjustments, it is unknown how they are regulated to achieve these specific membrane compositions. Furthermore, the precise roles that different membrane fatty acid compositions play in photosynthesis are only beginning to be understood. To explore the regulation of the membrane composition and photosynthetic function in response to temperature, we examined the effect of temperature in a collection of mutants with altered membrane lipid fatty acid composition. Results In agreement with previous studies in other species, the level of unsaturation of membrane fatty acids in Arabidopsis was inversely correlated with growth temperature. The time required for the membrane fatty acids to attain the composition observed at elevated temperature was consistent with the timing required for the synthesis of new fatty acids. Comparisons of temperature-induced fatty acid alterations in membranes were made among several Arabidopsis lines including wild-type Columbia, and the compositional mutants, fad5 , fad6 , act1 and double mutants, fad7 fad8 and act1 fad6 . The results revealed key changes that occur in response to elevated temperature regardless of the specific mutations in the glycerolipid pathway, including marked decreases in trienoic fatty acids and consistent increases in unsaturated 16:0 and in dienoic 18:2 levels. Fluorescence measurements of various mutants indicated that photosynthetic stability as well as whole plant growth at elevated temperature is influenced by certain membrane fatty acid compositions. Conclusions The results of this study support the premise that defined proportions of saturated and unsaturated fatty acids in membrane lipids are required for photosynthetic thermostability and acclimation to elevated temperature. The results also suggest that changes in the membrane fatty acid composition brought about in response to temperature are regulated in such a way so as to achieve highly similar unsaturation levels despite mutations that alter the membrane composition prior to a high-temperature exposure. The results from examination of the mutant lines also suggest that interorganellar transfer of fatty acids are involved in mediating temperature-induced membrane alterations, and reveal steps in the fatty acid unsaturation pathway that appear to have key roles in the acclimatization of membranes to high temperature.

Background One of the most prevalent environmental challenges encountered by plants is the exposure to a broad range of temperatures. Plants use a variety of anatomical, metabolic and cellular strategies to deal with changing environmental temperatures. Acclimation to elevated temperatures is mediated at the cellular level in part by the induction of general stress responses, which include the increased expression and activity of heat shock proteins [ 1 - 6 ]. These proteins enable organisms to withstand elevated temperatures by functioning as molecular chaperones to offset damage from misfolded proteins that would otherwise accrue during exposure to high temperature. Changes in the properties of cellular membranes, on the other hand, occur to ensure the proper function of processes that take place within them. The chloroplast thylakoids house the photosynthetic electron transport machinery, and are the most abundant membranes of leaf tissue. They are responsible for all light harvesting and photosynthetic energy conversion in the cell. Some alterations within thylakoid membranes occur rapidly to diminish stress triggered by immediate changes in the environment. For example, short-lived alterations in membranes can take place to counter excess absorbed energy in response to sudden exposures to high light [ 7 , 8 ]. Thermal dissipation of such excess light energy is mediated by the function of the carotenoid zeaxanthin, which is converted from violaxanthin by a de-epoxidase that is rapidly activated in response to high light [ 9 ]. In general, however, temperature-induced compositional changes of membranes follow time scales that reflect the acclimatization of plants at different temperatures [ 10 ]. Early studies of the ability of plants to acclimate to higher temperatures were conducted on plants adapted to high temperature growth [ 10 - 12 ]. For example, the desert shrub Atriplex lentiformis , changes its membrane fatty acid composition by decreasing the level of unsaturated fatty acids such as hexadecatrienoic acid (16:3) and increasing the level of saturated lipids at higher growth temperatures [ 13 ]. Membrane fatty acids of plants from temperate environments show similar trends in response to temperature, an observation that suggests that alterations in membrane lipids generally contribute to the ability of plants to acclimate to different temperatures [ 14 , 15 ]. Of the numerous cellular processes likely to be affected by the membrane composition, photosynthesis is probably the most critical. Adjustments in the level of unsaturation of thylakoid membranes may therefore affect the capacity of plants to adapt to elevated temperature conditions in order to avoid a reduction in photosynthetic efficiency. In this regard, the reaction center of PSII is considered to be the most sensitive component of the thylakoid membrane to thermal breakdown, and the function of the water-splitting D1 protein within PSII has been implicated as the most readily damaged by high temperature [ 10 ] as well as the being the primary target for photoinhibition [ 16 , 17 ]. High-temperature induced changes in the membrane composition may therefore play a role in the stability of such proteins leading to positive effects on whole plant growth at elevated temperature. Studies on cyanobacteria have indicated that the mechanism by which temperature mediates compositional alterations in membranes is at the level of transcription. The level of desaturase transcripts in cyanobacterial cells increases in response to growth at low temperature [ 18 ]. It is likely that the transcription of cyanobacterial desaturase genes is dependent on the physical properties of the membranes themselves [ 19 ]. The idea that changes in the degree of fluidity of cyanobacterial thylakoid membranes might be a means of sensing thermal stress is supported by the finding that heat shock proteins were induced when the membrane fluidity was chemically altered [ 20 ]. However, most investigations conducted with plants suggest that transcription is not the primary means of controlling membrane fatty acid unsaturation. Studies examining the level of expression of desaturase genes in Arabidopsis have shown that the majority of them do not respond to either decreases or increases in temperature via changes in transcription [ 21 - 23 ], or to alterations in the membrane fatty acid composition due to mutations in the desaturase pathway [ 24 , 25 ]. Only the relatively unusual case of the desaturation step catalyzed by the Arabidopsis fad8 gene [ 25 ], and other highly specific examples for particular plant species [ 26 ], have been shown to be regulated at the level of transcription by temperature. In this study we examined the leaf membrane fatty acid composition of Arabidopsis in response to temperature. After establishing the changes that occur at different elevated temperatures, we examined a number of mutant lines of Arabidopsis that possess well-characterized alterations in their membrane fatty acid content. The lines studied differ primarily in the degree of unsaturation of various lipids in chloroplast membranes. Examination of the fatty acid profiles of these lines grown at high temperature reveals a regulatory system that mediates membrane compositional changes even in the absence of specific desaturase steps. The results also suggest some general control points that might be important in the mechanism by which the membrane fatty acid unsaturation level is adjusted in response to temperature. In the course of carrying out these studies, we also examined the relationship between membrane fatty acid composition and growth at elevated temperatures. Results Elevated temperatures alter the membrane fatty acids of Arabidopsis Arabidopsis lines were germinated and grown at 17°C for seven days before being shifted to designated temperatures. Total leaf fatty acids were extracted from lines grown at various temperatures for 30 to 35 d, converted to methylesters and analyzed by gas chromatography. Changes in the relative abundance of the major fatty acid species from the Columbia wild-type line grown at 17°C, 20°C, 29°C and 36°C were plotted and are shown in Fig. 1 . The tendency of polyunsaturated species to decrease in abundance upon increasing growth temperature is evident within both 16 and 18 carbon-length fatty acids. The level of 18:3 in leaves decreased from 54 to 21 mol% of the total fatty acids from plants grown at 17°C and 36°C, respectively, and 16:3 species decreased from approximately 16 to 2.3 mol% for the same growth temperatures. In contrast, the proportion of diunsaturated linoleic acid (18:2) and monounsaturated oleic acid (18:1) showed an opposite response to elevated growth temperatures, with incremental increases of these species detected in plants grown at progressively higher temperatures. Specifically, the level of 18:2 increased from 10.1 mol% to 31.5 mol% of total fatty acids at 17°C and 36°C growth temperatures, respectively. These changes did not occur linearly over the temperatures measured but showed larger changes between 17°C and 20°C. At temperatures above 20°C, the relative abundance of the fatty acids generally changed in a more linear fashion. However, 16:0, unlike most other fatty acids, changed more abruptly between the two uppermost growth temperatures of 29°C and 36°C. To provide an additional estimation of the total membrane unsaturation level, the double bond index was calculated from the mol% fatty acid data (Table 2 ). The double bond index showed a corresponding decrease as the growth temperature was increased. Likewise, the levels of saturated palmitic acid, 16:0, accumulated to higher proportions in plants grown at elevated temperature, from 12.9 to 31 mol%. The relative abundance of other fatty acids (16:1, 16:2, and 18:0) showed alterations that were less pronounced in response to growth at elevated temperatures. Temporal evaluation of membrane alterations induced by high temperature Based on the substantial changes detected after the shift to elevated temperature, we wished to determine whether these changes occurred on time scales similar to the timing known for new synthesis of fatty acids. To estimate the time required for a temperature increase to induce measurable changes in the membrane fatty acid composition, Arabidopsis seedlings were germinated and grown at 22°C for seven days and then shifted to 29°C. The fatty acid composition was then evaluated in rosette leaves sampled at various times after the shift to elevated temperature by gas chromatographic analysis (Fig. 2 ). The leaf fatty acid profile of plants grown at 22°C (0 h) was comparable to previous determinations made on plants grown at similar temperatures (Fig. 1 ). No significant changes were detected in the fatty acid profile 60 h after the shift to 29°C. The earliest time point reflecting an alteration in response to elevated temperature occurred at 108 h, with 16:0 and 16:3 exhibiting the largest and most abrupt shifts, and 18:2 and 18:3 showing detectable, but more gradual changes at this time. At the 204-h time point, 18:2 and 18:3 species exhibited alterations in their accumulation to a degree similar to those observed to change markedly in plants grown at different temperatures. Specifically, 18:2 increased in a relatively linear manner from approximately 16 to 19.5 mol% of the total while 18:3 decreased from about 39 to 34 mol% and these proportions remained relatively constant at the final time point measured, 240 h after the shift to high temperature. The time course observed for the fatty acid composition to change in response to elevated temperatures correlates with the time determined for the incorporation of precursors to attain steady state levels in radiolabel feeding experiments [ 27 ]. The same general pattern of fatty acid changes was observed after the shift to high temperature. For example, 16-C fatty acids, 16:0 and 16:3, exhibited an abrupt shift from 18.9 to 24.8 mol% and 13.6 to 8.2 mol%, respectively, at the 108-hr time point, and remained relatively constant, until the final time point, except that 16:0 tended to decrease somewhat, after the increase observed at 108 h. On the other hand, major changes in the relative proportions of 18-C fatty acids, a large percentage of which are derived through the eukaryotic pathway, were not evident until the 204-h time point. This distinction between the onset of changes of 18:2 and 18:3 and of 16:0 and 16:3 is also reflected in the more linear response of 18:2 and 18:3 compared to the more abrupt transition and then fairly linear response of 16:0 and 16:3. Control experiments in which wild-type Arabidopsis lines were maintained at 22°C and sampled for membrane fatty acids over the same time intervals as shown in Figure 2 , did not show equivalent alterations at 108 h but maintained similar profiles throughout (data not shown). Thus, detection of the first alterations in the fatty acid composition in response to increased growth temperature is consistent with the time required for new synthesis of fatty acids. These results also suggest that changes in membrane fatty acids in response to temperature require turnover and resynthesis of new lipid to achieve the temperature-adapted composition. Temperature-induced changes in membrane fatty acid composition of mutant lines to probe membrane regulation A number of Arabidopsi s lines possessing well-characterized alterations in their membrane fatty acid profiles (Table 1 ) were used to examine how a specific block in the glycerolipid pathway affects the membrane composition in response to high temperature. For this analysis, we examined the membrane fatty acid content of plants grown at 17°C and 36°C. As in the previous determinations (Fig. 1 ), all plants were germinated and grown in an environmental chamber at 17°C for seven days before shifting to one set to 36°C. The fatty acid profile was then determined after 25–30 d of growth. The lower growth temperature of 17°C was chosen based on the response of the membrane fatty acid composition of leaves to this temperature determined previously, as well as the overall growth performance of the plants. The upper temperature of 36°C was selected based on the pronounced alterations observed in the fatty acid composition of wild-type plants grown at this temperature and the fact that this temperature represents the physiological upper limit for vegetative growth in soil. Indeed, growth at this temperature led to sterile plants (data not shown). This loss of seed production was apparently due to insufficient pollen formation and was consistent with notable decreases in seed yield (approximately 50%) from lines grown at 29°C (data not shown). Incubation at 36°C was also used to provide conditions by which the leaf fatty acid profiles would be most clearly delineated among the various Arabidopsis mutant lines as well as to establish the magnitude of the alterations in the fatty acid composition in response to elevated temperature. Each line used in this analysis possessed a mutation that primarily impacts the composition of the photosynthetic membranes of the chloroplasts (Table 1 ). Several of these mutants have been shown previously to display variable degrees of thermal tolerance [ 28 - 31 ]. Examination of the fatty acid profiles from these lines reflect the same general trends seen for wild-type plants grown at temperatures above 17°C (Fig. 1 ). As observed previously, the level of polyunsaturated fatty acids generally decreased with a concomitant increase in diunsaturated and saturated fatty acids when compared to plants grown at 17°C (Fig. 3 ). Specifically, trienoic fatty acids (16:3 and 18:3) decreased sharply and the level of dienoic 18:2 (but not 16:2) increased at the elevated temperature. Saturated 16:0 and, in some lines, monounsaturated 18:1, also accumulated to higher levels in plants grown at 36°C. A general estimation of the membrane unsaturation level was also provided by the double bond indices derived from this data (Table 2 ). Mutant lines fad5 , fad6 , fad7 fad8 , and act1 fad6 showed a lower double bond index from plants grown at 17°C compared to wild type, with fad6 displaying the most pronounced decrease in overall unsaturation. In contrast, the double bond indices obtained from membranes of the mutant lines grown at 36°C were similar to each other and the wild type with an average value of 1.5. Thus the total membrane unsaturation level is relatively equivalent among the different lines grown at the higher temperature. Two principal changes were apparent among the individual lines in response to high temperature: an increase in the abundance of 16:0 and 18:2 fatty acids and a pronounced decrease of both trienoic polyunsaturated species, 16:3 and 18:3 (Fig. 3 ). In the case of the fad5 mutant line, which is deficient in the desaturation of 16:0 to 16:1 at the sn -2 position only on chloroplastic monogalactosyldiacylglycerol [ 32 ], the 16-C fatty acids from plants grown at 17°C resembles those of the wild type grown at high temperature (Fig. 3B ). This includes the proportion of 16:3, which is essentially absent due to the block in the prokaryotic pathway at the level of 16:0. At 36°C, the composition of fad5 membranes is most similar among all of the mutants examined to that of the wild type grown at high temperature. The high degree of similarity of the fad5 and wild-type compositions may implicate the step catalyzed by the FAD5 desaturase to be an important control point in the acclimatization of the membrane composition in response to temperature. Examination of the membrane fatty acid composition of the fad6 mutant line in response to high temperature similarly reveals the general trends seen in the wild-type line grown at elevated temperature (Fig. 3C ). Because fad6 is deficient in the desaturation of 16:1 and 18:1 to 16:2 and 18:2, respectively, on all chloroplastic lipids [ 33 ], the accumulation of both 16:1 and 18:1 in this line are evident. The proportions of these two monounsaturated fatty acids remain virtually unchanged in fad6 plants grown at low and high temperature, showing that the overaccumulated levels of these species, due to the mutation in the FAD6 desaturase, is not subject to alterations in response to temperature. The mutation in the act1 line blocks entry of fatty acids into the prokaryotic pathway by a deficiency in chloroplast GPAT [ 34 ]. This is reflected in the fatty acid composition of act1 plants grown at both low and high temperature, with a decrease in all 16-C fatty acid species (Fig. 3D ). Temperature-induced changes in the fatty acid composition of the act1 line shows a profile highly similar overall in 18-C fatty acids compared to the wild type, suggesting that the desaturases controlling the conversion of 18:2 to 18:3 is a major control point capable of adjusting unsaturation levels of leaf membranes. The slightly elevated level of 18:2 at both temperatures in act1 compared to the wild type likely reflects the previously demonstrated increase of fatty acid flux into the chloroplast membranes via transfer of primarily this fatty acid [ 27 , 34 ]. The deficiency in the fad7 fad8 line is due to mutations at two loci, each encoding desaturase isozymes that convert 16:2 and 18:2 to 16:3 and 18:3, respectively, in the chloroplast [ 35 ]. In this line, the dienoic fatty acids, 16:2 and 18:2, are elevated and the effect of high temperature resulted in a reduction of this elevated 16:2 from 9.8 to 5.4 mol% (Fig. 3E ). The proportion of 18:2, on the other hand, increased in response to high temperature from 36.8 to 55.7 mol%. The trienoic fatty acid, 18:3, which due to the fad7 fad8 mutations was lower than the level in the wild type at 17°C (28 vs 54.8 mol% in the wild type), decreased to 6 mol% of total fatty acids at 36°C, the lowest leaf 18:3 levels of all the lines tested. Thus, the general trend of increasing 18:2 and decreasing 18:3 in response to high temperature growth was retained in the absence of most FAD7 and FAD8 desaturase activities, which catalyze the formation of the majority of trienoic fatty acids in the leaf membranes. A line derived from a cross between act1 and fad6 was produced during this study and used to evaluate the performance and temperature response of plants containing a membrane fatty acid composition which is distinguished by an elevated level of 18:1 (due to the deficiency in the FAD6 desaturase) but no increase in the relative amounts of 16:1 (because of the block into the prokaryotic pathway, due to diminished GPAT activity). The resulting composition in the act1 fad6 double mutant line grown at 17°C indicates the profile expected for mutations at each of these steps. The fatty acid profile altered in response to elevated temperature is similar to each parental mutant line grown at high temperature (Fig. 3F ). The act1 fad6 line thus provides a membrane fatty acid profile dissimilar to other lines in this study in that it contains elevated 18:1 but no increases in the proportion of 16:1 or 16:3. Such a profile can be used to address the functional significance of having only increased 18:1 but relatively similar proportions of all other fatty acids (see below). Fluorescence yield enhancement Six Arabidopsis lines possessing distinct membrane fatty acid compositions were evaluated by measuring chlorophyll fluorescence parameters to determine the thermal stability of PSII. The use of chlorophyll fluorescence has been used extensively to investigate photosynthesis and previous studies have examined photosynthetic stability using similar methods on isolated chloroplasts or detached leaves from the fad5 and act1 lines [ 29 , 36 ]. However, the analysis conducted here is the first to use in planta measurements in side-by-side comparisons to examine differences in thermal tolerance. In addition, prior to the fluorescence stability measurements, the fatty acid compositions of all lines used in this analysis were determined and all were found to be within the standard error of those values determined for the fatty acid profiles of the lines grown at 17°C (Fig. 3 , data not shown). Table 3 shows the results of the fluorescence analysis. Plants were grown at 17°C and evaluated at approximately four weeks of age. In all plants grown at this temperature, it is apparent that mutant lines fad6 , fad5 and ac t1 show a statistically significant increase in the fluorescence transition point (TP) temperature while fad7 fad8 shows a slight but insignificant increase. These results are likely to be due to differences stemming predominantly from the distinct membrane fatty acid compositions in leaves from the lines grown at this temperature. The TP values obtained for these intact leaf measurements here, are very similar to various determinations previously conducted on some of the lines using isolated chloroplast preparations or detached whole leaves [ 29 , 36 , 37 ]. Thus, only some lines display enhanced stability of photosynthetic quantum yield as the leaf temperature is increased and this seems to be positively correlated with a decrease in trienoic fatty acids, particularly 16:3, derived from the prokaryotic pathway. However, the relative proportion of 16:3 does not appear to be the only determinant for increased thermal stability. The double mutant act1 fad6 provided a line with a fatty acid profile that is distinct from that of other lines analyzed in this study (i.e., an increase in the percentage of 18:1 but no increased level of 16:1 at 17°C) (Fig. 3F ). To determine if this composition resulted in differences in the photosynthetic thermostability, we subjected the act1 fad6 line to fluorescence measurements. The act1 fad6 double mutant did not show a significant difference in the TP temperature from that of the wild type, even though each of the parental lines that possess only the single respective mutation exhibited a measurable enhancement of thermal stability (Table 3 ). The act1 ++ line is a transgenic line that over expresses GPAT (i.e., the act1 gene product). It was included as an additional line here to test whether its fatty acid composition would impact the thermal stability measurements as determined by chlorophyll fluorescence. However, the leaf fatty acid composition of act1 ++ plants grown at 17°C is similar to the wild-type composition except that the percentage of 16:0 and 16:3 fatty acids is increased to 2.0 and 1.5 mol%, respectively, over the proportions determined in the wild type. No differences in thermostability as indicated by chlorophyll fluorescence measurements were apparent for this line compared to the wild type. Although the most equivalent comparisons for stability measurements among the different lines would be obtained from those grown at temperatures considered moderate for Arabidopsis, such as 17°C used here, we also determined the TP temperatures for those lines that exhibited enhanced thermal stability at 17°C after growth at high temperature. Since this study revealed clear differences in the membrane fatty acid composition at various elevated temperatures, an indication of high-temperature acclimation based on fluorescence measurements may be evident. Four plants including wild type were grown at 29°C for three weeks before conducting the fluorescence stability tests. The results indicated an increase in the TP temperature from 42.7°C for wild type grown at 17°C to about 44.8°C in plants grown at 29°C (Table 3 ). However, the fad5 , fad6 and act1 lines, which showed reproducible increases in thermal stability compared to wild type when grown at 17°C, showed essentially the same TP temperatures as the wild type when grown at the elevated temperature. The lack of a detectable increase in thermal stability in mutant plants above the temperature found for the wild-type line grown at high temperature may be a consequence of the altered composition of the membrane fatty acids after growth at high temperature. The similar TP temperatures are also consistent with the overall similar unsaturation levels as indicated by the double bond index calculated for these lines. These results may also reflect the upper tolerance limit attainable that arises from fatty acid adjustments in response to elevated growth temperature, at least with the different compositions in the lines studied here. Growth at elevated temperature Most of the mutant lines used in this study were chosen on the basis of their altered fatty acid profiles in chloroplast membranes. As shown here, reductions in the total level of triunsaturated fatty acids in many lines grown at 17°C reflects the composition of the wild-type line grown at elevated temperature. To establish whether the observed changes translates into improved performance at the whole plant level, growth rates were determined for the lines at various temperatures. Growth rates were determined only during the first six days after the shift to elevated temperatures in order to maintain equivalent membrane fatty acid compositions of the lines before changes occurred in membranes during acclimation to high temperature (Fig. 2 ). As shown in Table 4 , the relative growth rates of fad5 , fad6 , act1 and fad7 fad8 seedlings are at least the same or slightly higher than wild type within several days after the shift to high temperature. In agreement with findings by Murakami et al. [ 31 ], the fad7 fad8 mutant line exhibited the most pronounced tolerance to high temperature based on growth rate. These results indicate that a less polyunsaturated membrane fatty acid composition than normally found in wild type favors seedling growth at elevated temperatures. Discussion Several clear alterations are evident in the membrane fatty acid composition in response to high temperature growth. A decrease in trienoic fatty acids, including strongly diminished 16:3, produced exclusively within the prokaryotic pathway in the chloroplast, is consistent with the general decrease of polyunsaturated fatty acids in response to high temperature. Concurrent with the reduced accumulation of trienoic fatty acids at high temperature is an increase in linoleic acid, 18:2, and an increase in 16:0 (Figs. 1 and 3 ). The observation that 16-C fatty acids show a pattern distinct from that of 18-C fatty acids suggests that individual fatty acid classes may have specific roles in maintaining optimal membrane function as well as different mechanisms governing their synthesis. This idea of distinct roles for particular fatty acids in the membrane is also supported by the relatively similar degree of membrane unsaturation in most of the lines examined (Table 2 ), despite the differences found in growth or photosynthetic stability. It is remarkable that these general alterations are observed even in mutant lines that are deficient in steps early or late in the desaturation pathways. For example, the profile of fad5 , which is deficient in 16:0 desaturation in the chloroplast compared to fad7 fad8 , which is deficient in the final desaturase step forming trienoic fatty acids, show similar overall trends in response to high temperature. These temperature responses must take into account the two glycerolipid pathways that operate in parallel in Arabidopsis, one in the chloroplastic envelope membranes and one in the endoplasmic reticulum [ 38 , 39 ]. Previous characterizations of all of the individual mutant background lines used in this study have clearly indicated the primary lipid species acted upon by the specific desaturase (or acyltransferase) activity missing in each mutant (Table 1 ). The temperature-induced alterations in the membrane composition can be considered in the context of the particular mutant background with its corresponding deficiency in a given step of the glycerolipid pathway. In this case, the mutations examined primarily affect chloroplastic lipids. In addition, the polyunsaturated 16-C fatty acids can be used as reliable markers for the major chloroplastic lipids monogalactosyldiacylglycerol and digalactosyldiacylglycerol, since they are the only lipids that contain these fatty acids. In regard to lipid compositions following growth at elevated temperature, examination of the proportions of individual lipids from the related species, Brassica napus , has shown that significant changes do not occur in leaf membranes from plants grown at 20°C and 30°C [ 15 ]. This finding suggests that it is the degree of fatty acid unsaturation that varies most appreciably at these temperatures and not the levels of the major leaf lipids themselves [ 15 ]. Temporal basis for temperature-induced membrane alterations The time required for the fatty acid composition to adjust to high-temperature growth conditions demonstrated that major alterations in leaf membranes do not occur rapidly in response to elevated temperature. The ~60-h period before changes become evident in the fatty acid profile corresponds with the occurrence of new lipid synthesis and turnover and therefore does not suggest a mechanism by which temperature induces modifications of existing membrane fatty acids. These results are also consistent with labeling studies that show fatty acids produced by the prokaryotic pathway accumulate prior to those synthesized via the eukaryotic pathway. The abrupt changes in 16-C fatty acids beginning 60 h after the shift to high temperature compared to the gradual and more continuous alterations observed for 18-C fatty acids (Fig. 2 ) also fits with these labeling patterns. It is well established by a number of time-course radiotracer labeling studies, including several conducted on most of the mutant lines used here, that earlier stages in the glycerolipid pathway show alterations prior to those formed through the eukaryotic pathway [ 27 , 33 - 35 , 40 ]. These include the formation of saturated fatty acids prior to the accumulation of unsaturated species and the earlier production of palmitate-containing species in the prokaryotic pathway. Overall, these time-course observations suggest that the modulation of membrane unsaturation levels plays a role in longer-term acclimation of the plant. Transient fluxes in environmental temperature are therefore not likely to result in pronounced alterations in the composition of leaf membranes. Membrane composition in response to temperature in mutant lines The fatty acid composition resulting from high-temperature growth in the different genetic backgrounds reveals a complex regulatory system. Thus, despite deficiencies in several enzymatic steps in the different mutants, the membrane fatty acid composition undergoes adjustments similar to those observed in the wild type in response to elevated temperature. The similar increase in 18:2 levels after high temperature growth in the fad5 , fad6 , act1 , act1 fad6 and wild-type lines suggests that the percentage of 18:2 may be important in membrane acclimatization. These results also suggest that desaturase activities as well as flux from extrachloroplastic membranes might also be controlled to mediate the response to temperature (see below). The accumulation of 18:2 (as opposed to 18:1) in the lines tested implicates it as a preferred species in membranes adapted to high-temperature growth. Although unknown, one speculation for 18:2 for this may be due to the "intermediate" level of disorder represented by diunsaturated acyl chains in the membrane. Fatty acids with one double bond, such as 18:1, impart a greater relative degree of disorder to the membrane than acyl groups containing two double bonds which, in turn, confer only slightly less disorder to the membrane relative to triunsaturated 18:3. The composition of the fad7 fad8 mutant line supports this contention, in which the mol% of 18:2 is almost 2-fold higher than that of wild-type membranes and this line exhibited the best growth at elevated temperature. However, biological membranes are complex, dynamic structures and it is likely that other, unknown factors will be important to maintain optimally functioning membranes. The fatty acid profiles of several mutants grown at high temperature suggests that the control of flux of fatty acids from the eukaryotic pathway is partly responsible for the changes observed in 18:2 and 18:3 due to temperature. In the fad6 mutant grown at high temperature, the proportions of 18:2 and 18:3 are highly similar to the proportions observed in the wild type, despite that essentially all of the 18:2 must be derived from the eukaryotic pathway. Labeling studies have shown previously that the fad6 line exhibits a decrease in lipid synthesis via the prokaryotic pathway [ 33 ], and this same mechanism, which presumably operates to maintain specific physical properties of the chloroplast membranes, may also be involved in mediating compositional adjustments of the membranes in response to increased temperature. In this case, a possible mechanism might be a reduction in the amount of 18:2 fatty acids transferred to the chloroplast for desaturation by the FAD7 and FAD8 desaturases. The fad6 profile also reveals that activity of the FAD2 enzyme, operating in the endoplasmic reticulum, might also be subject to temperature regulation. Similarly, in the act1 mutant line, in which entry of fatty acids into the prokaryotic pathway is blocked by the step catalyzed by GPAT, the primary difference is a decrease in the level of 16:0 when grown at elevated temperatures. This presumably reflects an increase into the eukaryotic pathway but with a similar temperature-responsive regulation. Flux of 18:2 from the eukaryotic pathway back into the chloroplast also appears to be modulated, as the proportions of 18:1 and 18:2 are slightly elevated in act1 at both temperatures, while the 18:3 level is highly similar to wild type. The fatty acid composition of the fad7 fad8 mutant reveals alterations that occur in response to elevated temperature in the absence of all trienoic fatty acid-forming desaturase activity in the chloroplast (Fig. 3E ). This profile also suggests that transfer of fatty acids from the eukaryotic pathway may be an important component in temperature regulation of the membrane composition. Such a mechanism is possible considering that the major proportion of 18:3 produced in this line must be synthesized through the eukaryotic pathway via the FAD3 desaturase [ 27 ]. The resulting low level of 18:3 detected in plants grown at high temperature is evidence that the FAD3 enzyme in the endoplasmic reticulum also is subject to temperature regulation. Thus it appears that desaturase enzyme activity is inversely regulated by increased temperature, in agreement with previous proposals as a likely mechanism [ 41 ]. Analysis of the data presented here suggests that the desaturases that catalyze trienoic fatty acid formation (FAD7, FAD8 and FAD3) and the FAD5 desaturase are the enzymes likely to have the greatest impact if regulated in this way. Membrane fatty acid composition has a role in enhancing photosynthesis to tolerate high temperature Studies conducted on the ability of plants to acclimate to elevated temperature have mainly focused on components most likely to affect the stability of photosynthetic electron transport, particularly PSII. In this respect, the composition of the chloroplast thylakoids is expected to be important in the thermal tolerance of photosynthetic electron transport [ 10 ]. Moon et al., [ 42 ] have implicated the unsaturation level of PG as being important in the removal and replacement of damaged D1 proteins in plants. However, the direct relationship pointing to protein-lipid associations being involved in stabilizing the D1 protein at high temperature has only recently been suggested [ 43 ]. The results presented here imply a regulatory mechanism that confers a similar overall composition in response to temperature regardless of the initial fatty acid alteration in the membranes due to mutation. A possible reason for such apparently stringent control might be that compositions that are deleterious for membrane function are curtailed, utilizing desaturase pathways that are present in both chloroplastic and extrachloroplastic locations. Measurement of PS II activity during leaf heating was used here as a sensitive indicator of the thermostability that might be conferred by the different membrane compositions. In this case, the temperature at which the quantum yield of PS II electron transport collapses (TP) was determined in intact leaves. The fluorescence measurements were conducted on plants grown at 17°C, to minimize variation and other potential responses, such as increases in the synthesis of heat shock proteins that might be induced at higher temperatures. In addition, only at the 17°C growth temperature were differences observed in the total membrane unsaturation level among the different lines tested as estimated by the double bond index, which might suggest that the largest influence of the fatty acid composition occurs at lower and more moderate temperatures. The fad6 mutant exhibited the maximum fluorescence yield enhancement of the lines grown at 17°C. This maximum TP is essentially the same as that obtained from several plants grown at 29°C, including the wild-type line (Table 3B ). The 2°C difference apparent in the mutant lines grown at 17°C may be an accurate indication of the magnitude of PSII thermal stability that can be conferred by adjustments in the membrane fatty acid composition and therefore may reflect the extent to which these adjustments can contribute to high-temperature acclimation in Arabidopsis. While it is unknown how specific fatty acids influence thermal stability, analysis of the act1 fad6 mutant described in this study demonstrates that alterations in the relative abundance of 16:1 and 18:1 have an effect. Both act1 and fad6 mutant plants display enhanced stability compared to wild type as determined by fluorescence measurements whereas the act1 fad6 line, which does not exhibit elevated 16:1 but slightly higher levels of 18:1, shows no statistical difference in the TP temperature compared to wild type (Table 3 ). This difference in fluorescence TP temperatures among these three mutants, as well as a lack of a correlation between the TP values and the double bond index, illustrates that the relative level of a specific lipid class can influence thermal stability. Measurements of diffusion rates of light-harvesting complexes in desaturase mutants of cyanobacteria also point to distinct roles that lipids may have in photosystem stability. In a recent study, the interaction of phycobilisomes with reaction centers was proposed to be stabilized by lipids as opposed to being affected by the level of membrane unsaturation directly [ 44 ]. Studies using spectroscopic methods have also suggested that overall lipid acyl chain disorder in cyanobacterial membranes is similar despite differences in growth temperature or unsaturation levels [ 45 ] and that protein-to-lipid interactions in membranes seem to be a key parameter in membrane dynamics [ 45 , 46 ]. Measurements of a single physical parameter attributable to alterations in the membrane composition have often not correlated with results apparent in whole plant performance tests. Similar to the results presented here, Murakami, et al., [ 31 ] have shown that chlorophyll fluorescence measurements might not be the ideal indicator to assess whole-plant performance. For example, although the fad7 fad8 line showed the fastest growth rate at high temperature, it did not show a significant difference in thermal stability based on fluorescence yield. The use of other functional measurements, such as CO 2 uptake rates or O 2 evolution, may provide indicators that more reliably assess potential high-temperature tolerance, although these measurements can also lead to unpredictable results regarding whole-plant physiology. In an Arabidopsis mutant devoid of virtually all digalactosyldiacylglycerol in chloroplast membranes, O 2 evolution was found to be unaffected despite large changes in thylakoid membrane organization and in fluorescence energy transfer characteristics [ 47 ]. Thus it is likely that additional differences stemming from the relative levels of distinct fatty acids in the membrane affect thermal tolerance and that membrane unsaturation levels will not be the exclusive factor. Other induced changes that impact membrane structure and function can be important. For example, alterations in the length and positions of double bonds within the acyl chain of a lipid molecule can confer widely differing properties and suggest that specific proportions of distinct fatty acid classes may be necessary for optimum membrane function [ 48 ]. A number of processes are called into play in response to stresses such as high temperature. Induction of heat shock proteins serves as one countermeasure to respond to elevated temperatures [ 5 ] and their specific roles in thermal tolerance are becoming clearer [ 2 - 4 , 49 , 50 ]. Correlations between the antioxidant status of plants and thermotolerance have also been noted recently [ 51 - 53 ]. Additional, perhaps more species-specific, responses are likely to be important in protecting against harmful effects of high temperature growth, including the accumulation of small molecules such as glycinebetaine and through the regulation of carbon fixation via rubisco activase [ 6 , 54 ]. Although decreases in the amounts of trienoic fatty acids may be an important determinant for plant thermal tolerance, lines possessing elevations in other, distinct fatty acid species, exhibit different characteristics. A triple desaturase mutant of Arabidopsis demonstrated that all trienoic fatty acids in leaf membranes are not essential for growth at low temperature [ 55 ]. This fad7 fad8 fad3 mutant also displayed thermostability but actually died after prolonged exposure to 33°C, suggesting that some trienoic fatty acids are indeed essential for high temperature growth [ 56 ]. The upper limit of chronic high temperature exposure for all lines tested here was about 34°C, where plants continue to grow and flower but exhibit reduced seed yield (data not shown), consistent with the idea that some trienoic fatty acids are essential. Although not addressed in the present study, it would be of interest to determine if this decreased seed yield from high-temperature-grown plants was related to insufficient levels of linolenic acid, a precursor required for the synthesis of jasmonic acid, a signaling molecule necessary for pollen development [ 57 ]. The results relating to growth performance at high temperature in this study are noteworthy in view of the fact that Arabidopsis is not considered a high temperature tolerant plant [ 58 ]. In more heat tolerant species, adjustment of the membrane fatty acid composition may well have greater significance in providing a rational means to control plant high-temperature tolerance. For example, suppression of a FAD7 homolog to concomitantly raise 18:2 and decrease 18:3 in tobacco enabled enhanced growth at elevated temperature [ 31 ]. Thus elimination of trienoic fatty acids might be the most critical aspect of altering the membrane composition to favor such enhanced growth. It is unclear, however, if the corresponding increases in monounsaturated and diunsaturated fatty acids that occur in such lines also contribute to the improved tolerance. Several of the mutant lines examined in this study also possessed very low levels of trienoic fatty acids but were accompanied by distinct alterations in other fatty acids. Most of the mutant lines exhibited elevations in the fatty acid species that serves as a precursor to the mutated step. The fad5 line grown at 17°C displayed 16-C fatty acids that were most similar to that of the wild-type line grown at 36°C, and had a membrane fatty acid profile almost identical to the wild type when each was grown at 36°C (Fig. 3A,3B ). While such a composition has no deleterious effects at moderate or elevated temperatures, this line, as well as the fad6 line, shows impaired growth and chloroplast development at low temperature [ 28 ]. Based on the similar composition to high temperature-grown wild-type plants, the reaction catalyzed by the FAD5 desaturase, which has been identified as an expressed sequence tag [ 59 ], may be an additional target to further manipulate tolerance to high temperature. Conclusions This study has shown that temperature precisely modulates the membrane fatty acid composition and these changes occur via mechanisms that are not profoundly affected by a number of mutations in the fatty acid desaturation pathway. These observations support the idea that the unsaturation level of plant membranes plays some role in enabling the plant to tolerate elevated temperatures but other characteristics of membrane lipids will likely also be important. Examination of a number of fatty acid desaturase mutants also suggests that alterations in the unsaturation level of membrane fatty acids in response to growth temperature apparently occurs by controlling the level of desaturase activity at several steps within the lipid biosynthetic pathway as well as by modulating inter-compartmental flux between the chloroplastic and cytosolic pathways. Methods Plant material and growth conditions Arabidopsis lines were grown in environmentally controlled chambers (Conviron, Inc.) adjusted to a given temperature under continuous light at 120–150 μmol m -2 s -1 . Seeds were sown on pots containing vermiculite-perlite mix (1:1:1) potting soil irrigated with a commercial fertilizer at every third watering. For growth in pots at elevated temperatures, seeds were germinated and grown at 17°C for seven days and then shifted to growth chambers adjusted to the respective temperatures and set to maintain humidity levels to at least 80% or higher, which was found to be essential to maintain growth at temperatures above 32°C. These chamber-grown plants were utilized for determining leaf membrane fatty acid compositions and for analyzing fluorescence yield characteristics. Growth rates of seedlings were determined on lines germinated and grown in Murashige and Skoog salts media (Sigma) containing 1% sucrose and 0.7% agar. Surface-sterilized seeds were plated, treated at 4°C for 2 to 3 days to synchronize germination, and grown for 16 d at 16°C in a growth chamber before shifting to elevated temperatures. Light intensity remained constant at 60 μmol m -2 s -1 . Four seedlings were sampled at days 0, 2, 4 and 6 d after the shift to elevated temperatures and were dried for 2 d at 70°C before weighing. The relative growth rates (ω -1 ) were calculated as the slope of the natural log of the seedling dry weight plotted over time in days. Growth rate determinations for seedlings grown at 22.5°C and 30°C were conducted 3 separate times to obtain SE and growth rates determined for 28 and 34°C were conducted once. The Arabidopsis mutant lines used in this study are listed in Table 1 . All of these are descendents of the Columbia ecotype. The act1 fad6 double mutant line was derived by first crossing act1 to the fad6 line and then crossing progeny from the resulting heterozygotes back to fad6 before selecting for lines possessing a fatty acid composition indicative of both homozygous mutations. The act1 ++ line, is a transgenic line containing the gene encoding glycerol-3-phosphate acyltransferase (GPAT) (Schneider and Somerville, unpublished, [ 60 ]) under the control of a dual 35S cauliflower mosaic virus promoter. It was included as an additional control for the chlorophyll fluorescence studies and possesses a membrane fatty acid composition virtually identical to the wild-type profile, except for a very slight elevation in the amount of 16:3 (~1.5 mol%, over the percentage in wild type at 17°C). Fatty acid analysis For each determination, at least two plants were used as a source of leaf material for extraction and analysis. Three mid-to-fully expanded (2 to 5-cm long) rosette leaves were harvested from each plant by removing approximately one-third of the leaf and immediately storing at -80°C in Teflon-capped tubes until needed. Preparation of methyl esters from these leaves and gas chromatographic analysis of the resulting extracts was performed using established procedures [ 61 ] with a Hewlett-Packard 5800 series gas chromatograph equipped with Supelco SP2330 glass capillary column (0.75-mm × 20-m). An estimation of the membrane total unsaturation level (double-bond index) was calculated from the mol% values derived from the gas chromatographic data, according to the equation: [Σ(mol% fatty acid content × no. of double bonds)]/100 as described by Skoczowski, et al. [ 62 ]. Fluorescence measurements Chlorophyll fluorescence was measured using an Opti-Sciences OS500-FL pulse modulated fluorometer (Opti-Sciences, Inc. Tyingsboro, MA). A custom designed aluminum heating block was constructed to hold an Arabidopsis rosette leaf, the associated optical cables and the temperature probes to conduct the leaf chlorophyll excitation and fluorescence measurements. An Arabidopsis leaf (~2.4 cm long) was fitted into a recessed (~0.5-mm) portion of the heating block to prevent damage to the leaf section. An ethylene glycol solution was circulated within the block to control temperature. The block and clamp module also provided an adjustable fixture to stably hold the leaf of a potted plant during the measurements. The photosynthetic thermostability of the different lines was evaluated by determining the transition point temperature of fluorescence quantum yield (TP). An attached rosette leaf was fitted into the thermally controlled heating block and fluorescence was monitored with a saturating light pulse (0.8 s duration) every 90 s. After stabilization at 29°C (4–5 pulses), the temperature of the heating block was increased at a rate of 0.75°C min -1 to a final temperature of 48.5°C using a digital temperature control unit (Omega Technologies Co. Stamford, CT). Leaf temperatures indicated were measured with a thermocouple in contact with the leaf surface. Quantum yield was calculated by the equation F' ms -F s /F' ms according to Genty [ 63 ], where F s is the steady state fluorescence under environmental conditions, and F' ms is the maximal fluorescence yield obtained after a saturating light pulse. F m -F o /F m , which is a measure of the photochemical efficiency of photosystem II, is based on the variable fluorescence (F m -F o ), which is the minimum (F o ) and maximum (F m ) fluorescence of a dark adapted leaf before and after, respectively, of a saturating light pulse. All determinations were made on leaf samples that gave an initial F m -F o /F m ratio of at least 0.80, prior to initiating the temperature increase. This measure was taken to ensure that the integrity of the leaf section introduced into the clamp module was sound and to verify that the leaf was properly fitted into the measurement module. The mean value ratio of 0.83 is indicative of a well functioning photosynthetic apparatus in an unstressed leaf as has been shown in a variety of other plant species [ 64 ]. Authors' contributions DF executed the fatty acid analysis, photosynthetic measurements, data analysis and drafted the manuscript. JO undertook the production of the act1 overexpressing lines and assisted in the construction of the double mutant line. CS conceived of the study. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC524174.xml

555766

GC-compositional strand bias around transcription start sites in plants and fungi

Background A GC-compositional strand bias or GC-skew (=(C-G)/(C+G)), where C and G denote the numbers of cytosine and guanine residues, was recently reported near the transcription start sites (TSS) of Arabidopsis genes. However, it is unclear whether other eukaryotic species have equally prominent GC-skews, and the biological meaning of this trait remains unknown. Results Our study confirmed a significant GC-skew (C > G) in the TSS of Oryza sativa (rice) genes. The full-length cDNAs and genomic sequences from Arabidopsis and rice were compared using statistical analyses. Despite marked differences in the G+C content around the TSS in the two plants, the degrees of bias were almost identical. Although slight GC-skew peaks, including opposite skews (C < G), were detected around the TSS of genes in human and Drosophila , they were qualitatively and quantitatively different from those identified in plants. However, plant-like GC-skew in regions upstream of the translation initiation sites (TIS) in some fungi was identified following analyses of the expressed sequence tags and/or genomic sequences from other species. On the basis of our dataset, we estimated that >70 and 68% of Arabidopsis and rice genes, respectively, had a strong GC-skew (>0.33) in a 100-bp window (that is, the number of C residues was more than double the number of G residues in a +/-100-bp window around the TSS). The mean GC-skew value in the TSS of highly-expressed genes in Arabidopsis was significantly greater than that of genes with low expression levels. Many of the GC-skew peaks were preferentially located near the TSS, so we examined the potential value of GC-skew as an index for TSS identification. Our results confirm that the GC-skew can be used to assist the TSS prediction in plant genomes. Conclusion The GC-skew (C > G) around the TSS is strictly conserved between monocot and eudicot plants (ie. angiosperms in general), and a similar skew has been observed in some fungi. Highly-expressed Arabidopsis genes had overall a more marked GC-skew in the TSS compared to genes with low expression levels. We therefore propose that the GC-skew around the TSS in some plants and fungi is related to transcription. It might be caused by mutations during transcription initiation or the frequent use of transcription factor-biding sites having a strand preference. In addition, GC-skew is a good candidate index for TSS prediction in plant genomes, where there is a lack of correlation among CpG islands and genes.

Background A prominent GC-compositional strand bias or GC-skew (=(C-G)/(C+G)), where C and G denote the numbers of cytosine and guanine residues, was reported recently around the transcription start sites (TSS) of Arabidopsis genes [ 1 ]. It is well known that GC-skews occur bi-directionally in circular bacterial genomes along the direction of replication, and that GC-skew is an effective index for predicting the replication origin in some bacteria [ 2 , 3 ]. In this case the numbers of G and T (thymine) residues in the leading strand of these genomes exceed those of C and A (adenine). Several models have been proposed to explain this bias [ 4 ]. A similar strand bias related to the direction of replication has also been observed in human mitochondrial genomes [ 5 ]. Also, mammalian and enterobacterial genomes have been reported to show a strand bias associated with transcribed regions [ 6 - 8 ]. An excess of G+T over A+C was observed in mammals within the sense strand of genes. A transcription-coupled DNA-repair system might be involved in this bias [ 9 ]. However, existing models cannot explain the excess of C over G in the sense strand around the TSS in Arabidopsis . Although slight GC-skews (regardless of the direction) were reported recently around the TSS in metazoans [ 10 , 11 ], it remains unclear whether the strong GC-skew of Arabidopsis is similar to that observed in metazoans. Although large amounts of genomic and full-length cDNA sequence data from plants are now publicly available, knowledge of the promoters and TSS in plants is still limited compared to mammals, such as human and mouse. It has been reported that the CpG island [ 12 ] is the most effective index for predicting the promoter regions or TSS in mammals[ 13 , 14 ]. However, the CpG islands are not specifically located in the promoter regions in Arabidopsis , so they cannot be used for the prediction of TSS or promoters [ 15 ]. Identifying another, more suitable, index for the prediction of plant-specific TSS has therefore become a priority. Answers are required for two key issues. First, which eukaryotic species, phyla or kingdoms have Arabidopsis -like GC-skews around the TSS? Second, what is the biological significance of these regions? In the present study, we used sequences from various animal, fungus, protist and plant species to conduct comparative analyses. We explored the potential value of GC-skew as an index for TSS prediction in plants. Finally, we considered the biological meaning of the GC-skew around the TSS of plant genes. Results GC-skew around the TSS of plant genes The shift in GC-skew values around TSS was assessed by calculating the GC-skew for regions between 1.0-kb upstream and 0.5-kb downstream of the TSS in Arabidopsis (a dicotyledonous plant), rice (a monocotyledonous plant), human and Drosophila . Full-length cDNAs and genomic sequences were exploited from the species with available TSS data. The sliding-window method was used, and a value for the GC-skew was computed at the central position for each 100 bp. The average of the GC-skew values at each position was calculated for all of the genes. The results confirmed that the mean GC-skew spectra of both Arabidopsis and rice peaked at the same position as the TSS (Fig. 1 ). The mean GC-skew values were approximately zero in regions upstream (<-0.2 kb) from the TSS, as the numbers of C and G residues were equal in both plant species. The values increased from the proximal region (-0.2 to -0.1 kb) and peaked at the TSS. Downstream from the TSS, the mean GC-skew values were inversely low. Although a small GC-skew was detected around the TSS in Drosophila and an opposite skew (C < G) was observed in human, these were less prominent when compared to those in plants. Our findings for Drosophila and human were consistent with previous data [ 10 , 11 ], indicating that this is a plant- or angiosperm-specific phenomenon. Intriguingly, the mean GC-skew values at the TSS were identical in Arabidopsis and rice genes, which were 0.20 (standard deviation = 0.30) and 0.20 (standard deviation = 0.29), respectively. However, it should also be noted that the mean values for the G+C content in the TSS region (-50 to +50 bp) in Arabidopsis and rice were significantly different: 37 and 53%, respectively. In contrast, the G:C ratios at the TSS in Arabidopsis and rice were identical, despite the considerable difference in G+C content. Figure 1 GC-skew in up- and downstream regions of the TSS. The up- and downstream regions of the TSS, which were 1.0 and 0.5-kb long, were analyzed using data from four species: 7,708 loci for Arabidopsis , 14,868 loci for rice, 14,053 loci for human and 8,344 loci for Drosophila . The graph shows the mean GC-skew values calculated for sequences of the four species using the sliding-window technique (window size = 100 bp; shift size = 1 bp). We also determined the frequencies of the four nucleotides in the regions up- and downstream of the TSS in plants (see Fig. S1 in Additional file 1 ). High C-residue frequencies were observed in both Arabidopsis and rice (approximately 50 to 100 bp from the TSS). In contrast, G-residue frequencies decreased slightly around the TSS (+/-10 bp). These findings suggest that the peaks of GC-skew values observed near the plant TSS were caused by an increased frequency of C residues and a slight reduction in the G-residue frequency. No significant biases were observed in the frequencies of A and T residues, indicating a lack of AT-skew at the TSS (See Fig. S2 in Additional file 1 ). The number of plant genes with a strong GC-skew in proximal regions of the TSS, was calculated by assessing the distributions of GC-skew values in regions +/-100 bp of the TSS. In order to take into account the wobble of the TSS, and to avoid experimental or mapping artifacts, only the maximum GC-skew values from these regions were used. The distributions of the maximum GC-skew values in the proximal regions of the TSS were similar in both plant species (Fig. 2 ). Over 70% of Arabidopsis genes and over 68% of rice genes showed a strong GC-skew (>0.33) near the TSS (that is, the number of C residues being more than double that of the G residues). Figure 2 Distribution of maximum GC-skew values around the TSS in Arabidopsis and rice genes. The graph illustrates the distribution of maximum GC-skew values within 100-bp up- and downstream of the TSS in Arabidopsis and rice. The GC-skew values were computed using the sliding-window technique (window size = 100 bp; shift size = 1 bp). The numbers of sequences analyzed were 7,708 for Arabidopsis and 14,868 for rice. Our results suggest that many plant ( Arabidopsis and rice) genes have strong GC-skews around the TSS. Furthermore, this characteristic is common among both monocot and eudicot plants. In contrast, the GC-skews (including the C < G skew) that were observed in human and Drosophila were qualitatively and quantitatively different from those identified in plants. GC-skew in eukaryotes The existence of GC-skew peaks around the TSS in various eukaryotes was examined by calculating GC-skew values 100-bp downstream of the 5' -ends of virtually assembled transcripts [ 16 , 17 ]. Although these did not represent the actual TSS, we were able to approximate the GC-skew values in the downstream regions. Table 1 shows the mean GC-skew values in the downstream regions of the TSS for several species. A prominent GC-skew (an excess of C residues) was confirmed in the 5' -ends of the transcripts in seven out of the nine plant species examined, and in five out of seven species of fungi. Although opposite skews (C < G) were observed in several protist and animal species, no significant excess of C residues was detected in any of the 10 animal species or the 11 protist species analyzed. Table 1 GC-skew in various eukaryotes Group Species GC-skew Mean Std. Dev. No. of sequences Plant Sorghum bicolor ++ 0.126 0.278 4,482 Oryza sativa ++ 0.118 0.304 18,676 Triticum aestivum + 0.095 0.255 13,133 Arabidopsis thaliana + 0.094 0.303 18,714 Gossypium + 0.092 0.304 1,561 Zea mays + 0.075 0.247 7,904 Glycine max + 0.050 0.311 6,162 Chlamydomonas reinhardtii 0.032 0.184 3,343 Pinus luchuensis -0.045 0.227 3,896 Fungus Filobasidiella neoformans ++ 0.222 0.341 243 Neurospora crassa ++ 0.184 0.276 2,763 Coccidioides immitis ++ 0.174 0.317 52 Aspergillus nidulans ++ 0.139 0.238 254 Magnaporthe grisea ++ 0.126 0.224 2,799 Saccharomyes cerevisiae -0.012 0.188 2,642 Schizosaccharomyces pombe -0.032 0.189 1,489 Protist Eimeria tenella 0.046 0.195 300 Tetrahymena thermophila 0.040 0.249 171 Trichomonas vaginalis 0.037 0.167 47 Dictyostelium discoideum 0.015 0.328 2,032 Neospora caninum -0.004 0.169 636 Toxoplasma gondii -0.019 0.184 1,328 Sarcocystis neurona -0.035 0.183 91 Trypanosoma brucei - -0.080 0.230 231 Plasmodium berghei -- -0.102 0.308 86 Cryptosporidium parvum -- -0.124 0.259 70 Plasmodium falciparum -- -0.191 0.354 1,905 Animal Caenorhabditis elegans 0.008 0.194 8,848 Drosophila melanogaster -0.011 0.170 14,310 Amblyomma variegatum -0.015 0.152 77 Ictalurus punctatus -0.017 0.216 382 Rattus norvegicus -0.023 0.193 12,594 Danio rerio -0.029 0.198 9,350 Homo sapiens -0.045 0.213 53,459 Mus musculus -0.045 0.217 50,029 Xenopus laevis - -0.064 0.212 13,444 Schistosoma mansoni - -0.088 0.230 195 The mean values and standard deviations (Std. Dev.) of the GC-skew values 100-bp downstream of the 5' -end were calculated in virtually assembled transcripts of nine plant species, seven species of fungus, 11 protist species and 10 animal species, which were downloaded from [16, 17]. The symbols + and ++ denote the predominance of C: ++ (≥0.10) and + (≥0.05). The symbols - and -- denote the predominance of G: -- (≤-0.10) and - (≤-0.05). We determined whether GC-skew peaks were actually present in the TSS of fungal genes, by investigating the regions around the translation initiation sites (TIS) of fungal genomic sequences. Genomic sequence data and information on open reading frames (ORFs; including predicted ones) are publicly available for some fungi, although there is insufficient information about the TSS. Nevertheless, it was possible to estimate the tendency towards GC-skew around the TSS by analyzing sequences both up- and downstream of the TIS in the available genomic sequences. Figure 3 shows the mean GC-skew values in regions between 1.0-kb upstream and 0.5-kb downstream of the TIS in fungal species: Aspergillus nidulans , Fusarium graminearum, Magnaporthe grisea , Neurospora crassa , Saccharomyces cerevisiae and Schizosaccharomyces pombe . GC-skew peaks similar to those of plants were observed in 50- to 100-bp upstream regions of the TIS in all of these species, with the exception of S. cerevisiae . Figure 3 GC-skew around TIS in fungal genes. The mean GC-skew values in regions 1.0-kb upstream and 0.5-kb downstream of TIS in the genomes of six species of fungi were calculated using the sliding-window technique (window size = 100 bp; shift size = 1 bp) for 9,432, 11,407, 10,054, 9,872, 5,825 and 4,305 loci in A. nidulans , F. graminearum , M. grisea , N. crassa , S. cerevisiae and S. pombe , respectively. Correlation between the two nucleotide frequencies in plants A possible mechanism causing the GC-skew around the TSS is nucleotide substitution, raising the question of what kind of substitution could be responsible. Correlations between the two nucleotide frequencies in regions both up- and downstream of the TSS were expected to indicate the answer (see Fig. 4 for Arabidopsis and rice, and Fig. S3 in Additional file 1 for human and Drosophila ). If the substitution rate between two specific nucleotides in one region is higher than that in other regions, a larger negative correlation would be expected between the two nucleotide frequencies, compared to other regions. In both plant species, the correlation coefficient ( r ) of A-T and G-C decreased dramatically around the TSS: the values were -0.4 to -0.7, and 0.0 to -0.5, in Arabidopsis (Fig. 4(a) ), respectively; and 0.18 to -0.2, and 0.2 to -0.4, in rice (Fig. 4(b) ), respectively. In contrast, the r values of A-G and T-C increased significantly around the TSS: values were -0.4 to 0.1, and -0.4 to 0.2, in Arabidopsis (Fig. 4(a) ), and -0.5 to 0.0, and -0.6 to -0.3, in rice (Fig. 4(b) ). Figure 4 Correlation between the two nucleotide frequencies in plants. The figure illustrates the correlation coefficients ( r ) between the two nucleotide frequencies (A-T, A-G, A-C, G-C, T-C and T-G) at each position around the TSS in ( a ) Arabidopsis and ( b ) rice genes. Each nucleotide frequency at a particular position was defined as the frequency of the nucleotide in a 100-bp window at that position. Correlation coefficients were calculated using these frequencies (see Methods section). The numbers of sequences analyzed were 7,708 for Arabidopsis and 14,868 for rice. GC-skew and gene-expression level In order to examine the relationship between GC-skew at the TSS and gene expression in plants, we conducted a statistical test using serial analysis of gene expression (SAGE) data for Arabidopsis (10-day-old seedlings) [ 18 ]. The mean GC-skew value in the TSS of highly-expressed genes (see Methods section for details) was significantly higher ( P = 0.0003, paired t-test) than in genes with low expression in Arabidopsis : the mean values were 0.25 (standard deviation = 0.31) and 0.19 (standard deviation = 0.31). Potential value of GC-skew as an index for TSS prediction Many of the GC-skew peaks were preferentially located near the TSS, therefore we assessed the potential value of GC-skew as a predictive index for TSS in plants. Sequences 1-kb upstream of the predicted ORF start positions in plant genomic sequences were used in the following analysis. First, the GC-skew values were computed using the sliding-window technique, in which one window and the shift size were set to 100 and 1 bp, respectively. GC-skew peaks satisfying a particular cut-off value were identified as primitive TSS candidates from the noisy GC-skew spectrum using the Savitzky-Golay (S-G) filter [ 19 ], which simultaneously smoothes and differentiates. Next, TSS candidates that were located within 50 bp of another candidate were considered to be identical and were merged into the position with the highest GC-skew. TSS prediction was validated by counting as true positives ( TP ), the candidates that were located within 100-bp up- or downstream of the actual TSS. If more than two candidates coexisted in the appropriate region, they were regarded as one TP . Using this method and the criteria described above, we validated the predictive performance of GC-skew under cut-off values ranging from -0.9 to 0.9 (Table 2 ). The specificity ( SP = TP/(TP+FP) ) ranged from 14 to 87%, and the sensitivity ( SN = TP/(TP+FN) ) ranged from 1 to 95% in Arabidopsis . The SP in rice ranged from 12 to 56%, and the SN ranged from 1 to 99%. The false-positive rate ( FPR = FP/ ( TN+FP )) varied with the cut-off values in a similar manner to SN . The difference between our results and random cases was clarified using a receiver-operating characteristic (ROC) curve (Fig. 5 ). This is a plot of FPR versus SN , with each cut-off value corresponding to a point on the curve. Good ROC curves lie closer to the top left-hand corner, whereas the random cases are represented as a diagonal line (defined by FPR = SN ). Predictions made using the GC-skew appeared to differ from the random cases, and lay closer to the top left-hand corner in both Arabidopsis and rice. In addition, the correlation coefficient ( φ ; see Methods section for details) corresponding to each GC-skew cut-off value was calculated in order to compare their predictive performances. The GC-skew value that maximized φ was 0.4 in both Arabidopsis and rice ( SP = 45%, SN = 41% and FPR = 8%, and SP = 47%, SN = 38% and FPR = 10%, respectively). Table 2 TSS prediction results for a stepwise increase of cut-off values Arabidopsis Rice GC-skew cutoff SN (%) SP (%) FPR (%) φ SN (%) SP (%) FPR (%) φ -0.9 95.3 13.6 24.4 0.149 98.7 11.6 6.4 0.068 -0.8 95.3 13.6 24.5 0.149 98.7 11.6 6.4 0.068 -0.7 95.3 13.6 24.5 0.150 98.6 11.7 6.5 0.068 -0.6 95.2 13.7 24.7 0.150 98.6 11.7 6.7 0.069 -0.5 95.2 13.7 25.2 0.153 98.4 11.7 7.2 0.072 -0.4 94.9 13.9 26.5 0.157 98.2 11.8 8.4 0.078 -0.3 94.2 14.3 29.4 0.168 97.8 12.1 11.1 0.092 -0.2 93.0 15.1 34.6 0.187 96.6 12.6 16.4 0.115 -0.1 90.7 16.4 42.3 0.213 94.4 13.6 25.3 0.147 0.0 86.7 18.6 52.6 0.247 90.4 15.4 37.9 0.186 0.1 80.0 21.8 64.2 0.282 83.5 18.2 53.2 0.230 0.2 70.2 26.4 75.6 0.315 74.1 22.7 68.5 0.279 0.3 58.7 32.9 85.0 0.345 62.0 29.4 81.4 0.322 0.4 45.0 40.9 91.9 0.354* 47.3 37.7 90.2 0.340* 0.5 31.6 50.5 96.1 0.343 32.3 46.7 95.4 0.327 0.6 20.6 60.6 98.3 0.312 19.1 55.3 98.1 0.281 0.7 11.1 70.2 99.4 0.251 9.1 60.2 99.2 0.204 0.8 4.6 77.4 99.8 0.171 3.2 59.4 99.7 0.119 0.9 1.1 86.9 100.0 0.089 0.8 56.3 99.9 0.057 Only the TSS of genes in which a TIS was defined within the 1.0-kb downstream region are included. Sequence data from 1.0-kb upstream of the TIS (6,850 sequences for Arabidopsis and 13,111 sequences for rice) were used for the TSS prediction. The asterisks denote maximum φ . Figure 5 ROC curves for predictions by GC-skew peak. ROC curves for ( a ) Arabidopsis and ( b ) rice. The areas under the curves (AUCs) for ROC curves were 0.80 for Arabidopsis and 0.78 for rice. The diagonal lines in the two graphs correspond to the ROC curve produced by random prediction. Discussion This study confirmed the presence of significant GC-skews (C > G) around the TSS (or upstream regions of the TIS) of genes in some species of plants and fungi. In contrast, our analysis revealed no significant excess of C residues in either animals or protists. However, an opposite GC-skew (C < G) has been reported previously in several animal species ( Mus musculus, Rattus norvegicus, Fugu rubripes, Danio rerio and human) [ 10 ]. Although small skews were detected for these species in our present analysis (Table 1 ; Fig. 1 ), they were not significant compared to those observed in plants. Aerts et al . [ 10 ] reported a GC-skew (C > G) close to the TSS in two nematode species ( Caenorhabditis elegans and Caenorhabditis briggsae ) however, no significant GC-skew was detected in C. elegans in our analysis. This inconsistency was probably due to the fact that our analysis targeted only regions downstream of the TSS in C. elegans (thus, regions upstream were not examined); alternatively, the skew might have been too small to be detected using our method. In either case, it is difficult to compare GC-skews between nematodes and plants, since trans-splicing at the 5' -ends of genes has been reported in nematodes [ 20 , 21 ], as pointed out by Aerts et al [ 10 ]. As noted above, the GC-skews near the TSS in plants and fungi differed from those of other species (or kingdoms) in both quality and quantity. To our knowledge, this is the first report to describe the prominent GC-skew (C > G) around the TSS specific to plants and fungi. We propose two possible explanations for the GC-skew peaks found close to the TSS. First, regulatory elements, such as transcription factor-biding sites (TFBS), which are present in regions both up- and downstream of the TSS, might contribute to (or even cause) this phenomenon. Moreover, some TFBS have a strand preference (see for example [ 22 ]). Therefore, if these types of TFBS are preferentially located around the TSS of plant and fungal genes, they might influence the local GC-compositional strand bias. Second, the GC-skew might be involved in transcription-coupled events, such as transcription-associated mutational asymmetry [ 6 - 9 ]. Tatarinova and colleagues [ 1 ] mentioned such transcription-associated mutational asymmetry and suggested that the GC-skew around the TSS of genes might be caused by the substitution of C residues with T residues, due to C deamination in the template strand. However, this hypothesis cannot fully explain all of the findings of our present study. If the C-to-T transition occurred preferentially around the TSS in the template strand, an AT-skew would also be expected in these regions. However, no significant AT-skew (see Fig. S2 in Additional file 1 ) was observed at the TSS of either Arabidopsis or rice genes, indicating that the C-to-T transition was not the main cause of the GC-skew. Furthermore, in our analyses, significant changes in the correlation coefficient (decreased A-T/G-C and increased A-G/T-C) were observed in both Arabidopsis and rice (Fig. 4 ). Assuming that the GC skew is caused by mutations during transcription initiation, changes in the correlation coefficient might be interpreted as an increase in the transversion ratio around the TSS. The increased negative correlation between C and G, coupled with the high GC-skew value in the same region, led us to speculate that G-to-C transversion occurred at relatively high rates in this region in plants and fungi, but not in animals and protists. If mutations that yield a GC-skew occur mainly around the TSS of single-stranded DNA, highly-expressed genes would be expected to have a high GC-skew around the TSS. In fact, the mean GC-skew value in the TSS of highly-expressed genes was significantly higher than in genes with low expression in Arabidopsis . This indicates that the GC-skew is associated with the level of gene expression, at least in Arabidopsis . Strand-specific mutational rates are believed to be a by-product of transcription-coupled DNA repair in mammals [ 8 ], therefore the GC-skew observed around the TSS of plant genes might result from the plant-specific DNA lesion and repair system. Alternatively, if the GC-skew was caused by the higher frequency of strand-specific TFBS, the higher mean GC-skew value in highly-expressed genes might be interpreted as a greater effect of strand-specific TFBS for the transcription efficiency. In either strand-specific TFBS or mutation, highly-expressed genes appear to have a high GC-skew around the TSS. As a GC-skew was also detected in the upstream regions of the TIS in fungal genes, this feature might be generated by a common mechanism in both groups. However, additional sequence data, and investigations into the patterns of nucleotide substitutions and their rates around the TSS, both between and within groups, will be necessary to elucidate the origin and mechanism of GC-skew around the TSS of plant and fungal genes. TSS prediction by the GC-skew peak was validated through a stepwise increase of cut-off values which demonstrated that the GC-skew could contribute to TSS prediction. Although the optimal GC-skew cut-off value depends on the specific situation, our results will be helpful in determining the optimal cut-off. Figure 6 shows representative cases in which the GC-skew peaks are located near the TSS in Arabidopsis and rice genes. The single index presented in this paper might not be sufficient to achieve accurate TSS prediction. However, our results indicate that GC-skew is a good candidate index for the TSS, promoter or first exon in plants. Thus, the combined use of GC-skew and other indices, or the incorporation of this index into pre-existing programs appears to be a realistic and effective approach for TSS prediction. Figure 6 Representative cases of GC-skew peaks located in the TSS of plant genes. These figures show typical examples of GC-skew peaks located near the TSS in Arabidopsis ( a ) and rice ( b ). Red solid triangles represent the positions of the GC-skew peak. The corresponding GenBank entry IDs are: AF361813, AF380632, AK105340 and AK102437. Conclusion Significant GC-skew (C > G) around the TSS is strictly conserved among monocot and eudicot plants (that is, angiosperms), and a similar skew is also seen in some fungi. The mean GC-skew at the TSS in the highly expressed genes was greater than that in the group with low expression. We therefore propose that the GC-skew around the TSS in some species of plants and fungi is associated with transcriptional activity. This is probably a result of DNA mutations during transcription initiation or the frequent use of strand-specific TFBS. Our findings also confirm that GC-skew has the potential to assist TSS prediction in plant genomes, where there is a lack of correlation among CpG islands and genes. Methods Data sources Data for 13,095 full-length cDNAs from Arabidopsis thaliana were downloaded from The Institute for Genomic Research (TIGR) [ 23 ] (as of March 2, 2001) and the RIKEN Arabidopsis Genome Encyclopedia [ 24 ] (as of May 9, 2002). Genomic sequences for Arabidopsis were downloaded from The National Center for Biotechnology Information (NCBI) [ 25 ] (as of January 31, 2003). ORF information for Arabidopsis genes was obtained from the NCBI Entrez database [ 26 ], based on each of the original sequence IDs. For rice ( O. sativa ssp. japonica c.v. Nipponbare), 28,469 full-length cDNA sequences [ 27 ] were retrieved from the Knowledge-Based Oryza Molecular Biological Encyclopedia (KOME) [ 28 ] with ORF information, and the genomic sequences were obtained from the Rice Genome Research Program [ 29 ] (as of October 16, 2002) and Syngenta Biotechnology Inc. (SBI) [ 30 ]. For analysis of human ( Homo sapiens ) DNA, 21,245 full-length cDNAs and genomic sequences were downloaded from the DNA Data Bank of Japan (DDBJ) [ 31 ] and Ensembl (Ver. 9.30) [ 32 ], respectively. Data for 9,872 full-length cDNA sequences (as of July 17, 2002) and genomic sequences of Drosophila melanogaster were obtained from the Berkeley Drosophila Genome Project [ 33 , 34 ]. Fungal genomic sequence data, including ORF information, were downloaded from the NCBI [ 35 , 36 ] for S. cerevisiae and S. pombe , and from the Fungal Genome Initiative (FGI) [ 37 ] for A. nidulans , F. graminearum , M. grisea , and N. crassa . Virtually assembled transcripts from 10 animal species, nine plant species, six species of fungi and 11 protist species were retrieved from the TIGR Gene Indices (TGI) [ 16 , 17 ]. Mapping of cDNA to the genomic sequences Redundant cDNA sequences showing at least 95% similarity in at least 95% of the regions, compared with other sequences were excluded from the full-length cDNA dataset for four species – Arabidopsis , rice, human and Drosophila . They were identified using BLASTN homology searching and CAP3 [ 38 ]. Also, any poly(A) tracts at the 3' -end of the cDNA sequences were eliminated. Finally, by mapping the cDNA sequences for corresponding genomic sequences using the BLASTN program and SIM4 [ 39 ], sequences both up- and downstream of the TSS were determined. Gene-expression data SAGE data for Arabidopsis (10-day-old seedlings) [ 18 ] were used to examine the relationship between GC-skew in the TSS and gene-expression levels. In assigning the SAGE tags to genes, only data that showed one-to-one correspondence between the genes and tags were included. Genes with <100 counts per million were defined as the low-expression group (504 genes) and those with >100 per million the high-expression group (689 genes). The mean GC-skew values in a 100-bp window at the TSS were calculated for both groups. Genes with a G+C content at the TSS of <0.3 were eliminated from the dataset in advance. GC-skew peak detection As an initial step towards detecting the GC-skew peaks, GC-skew values at each position in the sequences were calculated using the sliding-window technique (window size = 100 bp; shift size = 1 bp). Next, we attempted to smooth the spectrum, to reduce the noise and to simultaneously determine the peaks, using the S-G filter [ 19 ]. The first-order derivative of the smoothed GC-skew value g i at position i is given by the following equation: Here, f i + n is the original GC-skew value in the position i + n . n l and n r represent the lengths of the filter window to the left and right of position i , and were set to 20 in this analysis. C n denotes the set of weight coefficients and corresponds to the first-order derivatives of the quartic polynomial. The position of the GC-skew peak i was defined as the zero-crossing point, which is the point satisfying the following condition: g i · g i + 1 ≤ 0 ∩ g i + 1 < 0 The peak was only counted as a TSS candidate if a GC-skew value at that peak satisfied the particular cut-off value, and the G+C content was ≥ 0.3 in the window. When more than one peak occurred within 50 bp, they were regarded as identical and the peak with the highest GC-skew value was accepted. Using this procedure, TSS prediction was conducted with stepwise changes in the cut-off values of the GC-skew (-0.9 to 0.9) at the peak. The predictive accuracy was verified by counting TSS candidates located within 100-bp up- and downstream of the actual TSS as true positives ( TP ). Where more than two TSS candidates coexisted within 100-bp either up- or downstream of the TSS, they were counted as one TSS candidate. The rest of the candidates, which were located in inappropriate regions, were counted as false positives ( FP ). To compare results for different cut-offs, the correlation coefficient ( φ ), which is one of the measures used to compare predictive performance [ 40 ], was calculated as follows: Here, TN and FN denote the number of true and false negatives, respectively. In this analysis, the total number of negatives ( N ) was defined as the maximum number of possible TSS candidates in the non-TSS regions. Thus, TN was calculated as N - FP . Correlation coefficient between the two nucleotide frequencies The correlation coefficient r ij ( p ) between nucleotide i and j at a position p up- and downstream of the TSS was defined as follows: Here, i k ( p ) and j k ( p ) are the frequencies of i and j at a position p in the sequence k , respectively. and are the mean frequencies of i and j at p . The nucleotide frequency in each position was the number of each nucleotide in a 100-bp window. Authors' contributions SF carried out the computational and statistical analysis, and drafted the manuscript. TW and MT participated in the design and coordination of the study. All of the authors have read and approved the final manuscript. Supplementary Material Additional File 1 This file contains three figures: S1, showing the nucleotide frequency around the TSS; S2, showing the AT/GC-skew in both up- and downstream regions of the TSS in plants; and S3, showing the correlation between the two nucleotide frequencies around the TSS in human and Drosophila . Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC555766.xml

548694

Autonomic nervous system dysfunction predicts poor prognosis in patients with mild to moderate tetanus

Background Autonomic nervous system (ANS) dysfunction is present in up to one third of patients with tetanus. The prognostic value of ANS dysfunction is known in severe tetanus but its value is not well established in mild to moderate tetanus. Methods Medical records of all patients admitted with tetanus at two academic tertiary care centers in Karachi, Pakistan were reviewed. The demographic, clinical and laboratory data was recorded and analyzed. ANS dysfunction was defined as presence of labile or persistent hypertension or hypotension and sinus tachycardia, tachyarrythmia or bradycardia on EKG. Patients were divided into two groups based on presence of ANS dysfunction (ANS group and non ANS group). Tetanus severity was classified on the basis of Ablett criteria. Results Ninety six (64 males; 32 females) patients were admitted with the diagnosis over a period of 10 years. ANS group had 31 (32%) patients while non ANS group comprised of 65 (68%) patients. Both groups matched for age, gender, symptom severity, use of tetanus immunoglobulin and antibiotics. Twelve patients in ANS group had mild to moderate tetanus (Ablett I and II) and 19 patients had severe/very severe tetanus (Ablett III and IV). Fifteen (50%) patients in ANS group required ventilation as compared to 28 (45%) in non-ANS group (p = 0.09). Fourteen (47%) patients died in ANS group as compared to 10 (15%) in non ANS group (p= 0.002). Out of those 14 patients died in ANS group, six patients had mild to moderate tetanus and eight patients had severe/ very severe tetanus. Major cause of death was cardiac arrhythmias (13/14; 93%) in ANS group and respiratory arrest (7/10; 70%) in non ANS group. Ten (33%) patients had complete recovery in ANS group while in non ANS group 35(48%) patients had complete recovery (p= 0.05). Conclusions ANS dysfunction was present in one third of our tetanus population. 40% patients with ANS dysfunction had only mild to moderate tetanus. ANS dysfunction, irrespective of the need of mechanical ventilation or severity of tetanus, predicted poor outcome.

Background Tetanus is still a major health problem in developing world. Care of this potentially fatal, though preventable, disease has been revolutionized with advent of mechanical ventilation. However, still a significant minority, at least 10%, would die despite every effort [ 1 ]. Respiratory arrest used to be a common problem during preventilation era, however, ANS dysfunction has been emerged as a major problem in these patients in post ventilation era [ 1 ]. ANS dysfunction can cause a variety of arrhythmias which can potentially be lethal. There are reports that certain arrhythmias e.g. tachycardia carry poor prognosis [ 2 ], however, ANS dysfunction has been attributed to severe tetanus, which carry poor prognosis anyways. Violent autonomic disturbances, severe hypertension and tachycardia alternating with hypotension and bradycardia is reported in relation to severe tetanus. The association of so called milder forms i.e. stages I and II on Ablett's criteria [ 3 ] to ANS dysfunction and its affect on outcome is not well established. Our study highlights the importance of this previously not well-known aspect of tetanus. Methods All patients admitted to two tertiary care centers in Karachi, Pakistan over a period of 10 years, were identified through ICD-9 coding system of the hospital medical records. The charts were reviewed retrospectively. The demographic, clinical, laboratory data was recorded and analyzed. The autonomic dysfunction was defined as presence of labile or persistent hypertension (>140/90 mmHg) or hypotension (<90/60 mmHg) and persistent sinus tachycardia (heart rate >100 bpm), tachyarrhythmia or bradycardia (heart rate <50 bpm) alternating with tachycardia on EKG. These values were recorded at resting state, not during tetanic spasm. Brief episodes of tachycardia or hypertension are common in ICU and could be related to pain or anxiety. The values in ICU were mostly recorded in sedated and relaxed position. Continuous automated blood pressure and heart rate monitoring was done in all ICU patients. Blood pressure and heart rate were checked every four hour for the ward patients. The data regarding arterial lines is not available. The values recorded during tetanic spasms were not included in the study. The decision to admit these patients was based mostly on availability of ICU beds. Patients with only locked jaw without any autonomic abnormalities on admission were electively admitted to ward. Severity of tetanus was classified based on Ablett criteria. The poor outcome was defined as either death or partial recovery at the time of discharge from the hospital. Partial recovery was defined as neck pain, back pain, walking difficulty and abnormal gait at discharge, in various combinations. Complete recovery was defined as absence of all of the symptoms and normal gait on examination at discharge. The data was analyzed on SPSS version 10.0. Statistical analysis employed descriptive and univariate (chi-square and t test) methods. Results Ninety six patients were identified. Sixty four (67%) were men and 32 (33%) were women. Their mean age was 44 (11–75) years. Identifiable risk factors were present in only 49 (51%) patients, recent trauma being most common which was noted in 27(28%) of the patients. Other risk factors were skin wounds, either infected or uninfected; 10 (11%), needle injury; 8 (9%) and recent surgery; 8 (9%). Most common clinical features were trismus; 89 (93%), neck stiffness; 73 (76%), muscle spasms; (67) 70%, followed by dysphagia; 57 (59%), back stiffness; 42 (44%), dysarthria; 19 (20%), abdominal pain; 10 (11%) and subjective breathing difficulty; 7 (8%). All patients received TIG (500–8000 international units) and almost all (99%) received benzodiazepine infusion. About one third (34%) of the patients received paralytics and 4 (5%) were also given magnesium sulphate. Majority of patients (71; 72%) were given antibiotics. The EKG abnormalities include sinus tachycardia in 44 (46%) patients, cardiac ischemia in 5 (5%), sinus bradycardia in 2 (2%) and heart block in 1 (1%) patient. Blood pressure abnormalities include persistently elevated blood pressure in 8 (8%) patients, persistent hypotension in 2 (2%) and labile hypertension in 14 (15%). Forty five (47%) patients recovered fully, 27 (28%) had partial recovery and 24 (25%) patients died. Major cause of death was cardiac arrhythmias (14/24; 58%) followed by respiratory arrest (7/24; 29%). All of the patients who had respiratory arrest were in general wards and never intubated. Fifty eight (60%) patients admitted to ICU. 43 (44%) required mechanical ventilation. Mean hospital stay was 17 (1–63) days. The hospital stay was prolonged in patients who required ventilation (25 vs 9 days) ANS group had 31 (32%) patients while non ANS group comprised of 65 (68%) patients. Severity of tetanus in ANS group was mild to moderate (Ablett I and II); 12 patients and severe/very severe (Ablett III and IV); 19 patients. 15 patients out of these 19 with severe tetanus required mechanical ventilation. Fourteen of 31 (47%) patients died in ANS group as compared to 10/62 (15%) in non ANS group (p= 0.002) (Table 1 ). Out of 14 patients who died, six had mild to moderate tetanus and eight patients had severe tetanus. Cause of death was cardiac arrhythmias (13 patients) and sepsis (1 patient) in ANS group while in non ANS group 7 patients died of respiratory arrest, one died of sepsis and one died of myocardial infarction. Cause of death was not known in one patient. None of these patients were witnessed to die during tetanic spasm. All patients who died were receiving benzodiazepines but not at a dose that could cause respiratory failure. All patients in ICU and majority of patients in the ward were receiving DVT prophylaxis in form of subcutaneous heparin and compression stockings. Ambulatory, non-bedridden patients (considered low risk for DVT) were not given DVT prophylaxis. Pulmonary embolism was not considered as a cause of death in any of the patients because these patients had no clinical evidence of DVT. The possibility that these patients died of acute autonomic dysfunction or arrhythmia could not be ruled out. Autopsy was not performed in any of the patients. Ten of 31 (33%) patients had complete recovery in ANS group as compared to 35/62 (48%) in non ANS group (p= 0.05). Both groups were similar in other aspects (Table 2 ) Table 2 Tetanus severity, autonomic dysfunction and mortality. Tetanus severity ANS dysfunction (n = 31) Non-ANS dysfunction (n = 65) >Mild/Moderate 12 (6; died) 40 (1; died) severe 19 (8; died) 25 (9; died) Table 1 Comparison of Autonomic neuropathy versus no autonomic dysfunction in patients with tetanus. Variable ANS group (n = 31) Non ANS group (n = 65) P value Mean age 39 years 46 years 0.2 Male: Female 23:8 41:24 0.1 Locked jaw 89 (93%) 89 (93%) 0.5 Dysphagia 18 (58%) 39 (60%) 0.12 Ventilation 15 (48%) 28 (43%) 0.09 Death 15 (48%) 10 (15%) 0.002 Complete recovery 10 (32%) 35 (54%) 0.05 Partial recovery 7 (23%) 20 (31%) 0.25 Discussion Tetanus is disease of antiquity. It was first described in Egypt over 3000 years ago [ 1 ]. It is caused by tetani clostridium which is an obligate non aerobe and ubiquitous organism. Tetanus kills approximately one million people every year world wide and is a major health problem in developing world [ 1 ]. The clinical presentation of our patients is similar to that described in literature [ 2 , 4 , 5 ], most common presenting features being, trismus, neck stiffness and spasms (vide supra). Prognosis of patients with tetanus has been variable. Overall mortality is approximately 10–50% [ 1 , 6 ], however, it is as high as 90–95% in certain select groups i.e. neonates [ 7 ] and tetanus associated with intramuscular quinine [ 8 ]. Various factors have been known to affect the prognosis. The poor prognostic factors include shorter incubation and onset periods [ 2 , 9 ], fever [ 9 ], tachycardia [ 2 ], fluctuating blood pressure [ 2 ], tetanus associated with intramuscular injections specially quinine [ 8 ], extreme of ages [ 1 , 7 ], hypoxia and acidosis at admission [ 10 ]. The major complication used to be respiratory arrest before advent of artificial ventilation, however, nosocomial infections and autonomic disturbances are major complications in post ventilator era [ 1 ]. Though the autonomic instability has been recognized as a major complication in the post ventilator era, this has been known since 1960s [ 11 ] and reported to occur in about a third of the tetanus patients [ 4 ]. An electrocardiographic study from India showed the incidence of sinus tachycardia as high as 85% [ 12 ]. The autonomic disturbance often develops a few days later in course. Pathogenesis of autonomic disturbances is unclear, however, several theories have been put forward, including damage to brain stem and hypothalamic nuclei [ 5 ] and direct disturbances in autonomic nerves [ 5 , 13 ]. The role of tetanus induced damage to brain stem nuclei has been discarded by many authorities. [ 14 ] Initially these were considered to be sympathetic overactivity [ 11 ], however, later studies with hemodynamic monitoring showed that both sympathetic as well as parasympathetic systems are involved [ 13 ]. The autonomic disturbances can be fatal. Though sudden cardiac arrest is most devastating complication, various tachy and bradyarrhthmias can be fatal. We also noted that the autonomic system dysfunction was associated with higher mortality (table I ). In our cohort the major cause of death in patients with ANS dysfunction was cardiac arrhythmias. The plausible mechanisms for higher mortality in patients with ANS dysfunction include fatal tachyarrhthmias e.g. VF, cardiac arrest, myocardial infarction. Complications of intervention for ANS dysfunction is another potential mechanism of catastrophic end result as these patients are exquisitely sensitive to pharmacotherapy. To avoid these complication careful hemodynamic monitoring and cautious and judicious use of fluids and drugs like beta blockers, atropine may improve outcome in these patients. Morphine has been reported to control autonomic dysfunction. [ 15 ] It was used in only a few patients in our series in ICU patients, however, not as first line therapy. Data prior to use of morphine was used for presence or absence of autonomic dysfunction. The ANS disturbances are considered to be part of severe tetanus [ 1 ], however, we did note the disturbances even in milder cases. As mentioned earlier the ANS dysfunction carries poor prognosis but this has been considered to be part of severe tetanus [ 1 ], which carries poor prognosis anyways. The association of so called milder forms i.e. stages I and II on Ablett's criteria [ 3 ] to ANS dysfunction and its affect on outcome is not well established. In our cohort 40% of the patients with ANS dysfunction were in grade I or II. Our study is limited by its retrospective nature and paucity of hemodynamic and autonomic testing. However, based on the results we conclude that the presence of autonomic nervous dysfunction irrespective of the respiratory/ventilatory status, predicts poor outcome in tetanus. Studies have shown that reduced tetanus mortality in the recent era is most probably due to advances in ICU management. [ 16 ] We also strongly think that the ANS dysfunction is inherent to tetanus even in milder cases so all patients with tetanus should be carefully monitored in ICU settings to avoid mortality from this deadly but potentially reversible disease. Further prospective studies are required to confirm our findings Conclusions Autonomic dysfunction is not uncommon in patients with Tetanus. It was present in one third of our tetanus population. 40% patients with ANS dysfunction had only mild to moderate tetanus. ANS dysfunction, irrespective of the need of mechanical ventilation or severity of tetanus, predicted poor outcome. Competing interests The author(s) declare that they have no competing interests. Authors' contributions MW: conceptualization, data analysis, manuscript preparation. BAK: data entry and analysis, statistical analysis, manuscript preparation. NT: data acquisition, manuscript preparation. RS: data acquisition, data analysis, manuscript preparation. NAS: conceptualization, manuscript preparation. NS: conceptualization, manuscript preparation. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548694.xml

514712

The solution structure of ChaB, a putative membrane ion antiporter regulator from Escherichia coli

Background ChaB is a putative regulator of ChaA, a Na + /H + antiporter that also has Ca + /H + activity in E. coli . ChaB contains a conserved 60-residue region of unknown function found in other bacteria, archaeabacteria and a series of baculoviral proteins. As part of a structural genomics project, the structure of ChaB was elucidated by NMR spectroscopy. Results The structure of ChaB is composed of 3 α-helices and a small sheet that pack tightly to form a fold that is found in the cyclin-box family of proteins. Conclusion ChaB is distinguished from its putative DNA binding sequence homologues by a highly charged flexible loop region that has weak affinity to Mg 2+ and Ca 2+ divalent metal ions.

Background The regulation of cellular ion concentrations is an essential process in all organisms, necessary to sustain a multitude of physiological processes including pH balance and ion homeostasis. This process is accomplished mainly through membrane ion transporters. In Escherichia coli , among the membrane proteins that catalyze the exchange of ions across the cell membrane [ 1 ] are the Na + /H + antiporters NhaA, NhaB and ChaA, which are involved in sodium ion extrusion. Within E. coli and other enteric bacteria, antiporters encompass the primary systems responsible for adaptation to growth in conditions of high Na + concentrations and varying pH [ 2 - 8 ]. It is common for bacteria to have multiple systems for a similar function. The use of one system is preferred depending on the stress as a means to adapt to varying environmental conditions [ 9 , 10 ]. Of the Na + /H + antiporters, ChaA is unique in that it also shows pH-independent Ca + /H + antiporter activity. ChaA is also regulated by Mg 2+ , which inhibits both its Na + /H + and Ca + /H + antiporter activity [ 11 ]. The Cha operon consists of 3 genes, chaA , chaB and chaC found at ~27 minutes on the E. coli chromosome [ 12 ]. Both ChaB and ChaC are proposed to be regulators of ChaA however, the biological function for either remains to be established. ChaB is a 76-residue protein that contains a conserved 60-residue region found in several other bacteria and baculoviruses. We report here the three-dimensional structure of ChaB determined by NMR spectroscopy and examine key differences between the ChaB families of proteins. Results and Discussion Assignment of resonances The 1 H- 15 N HSQC spectrum of ChaB (Fig. 1 ) is well dispersed suggesting ChaB is a globular, folded protein. Complete 1 H, 15 N and 13 C backbone assignments were made for ChaB, except residues S39 and H40, which yielded no apparent amide cross peaks. Virtually complete assignments (> 98%) were made for the 1 H, 13 C, and 15 N side chain resonances. Resonance assignments have been deposited at BMRB (code 6117). Six signals in the 1 H- 15 N HSQC spectrum originated from the 21 residue N-terminal His-tag (Fig. 1 ). The low heteronuclear NOE values (Fig. 2A ) and the relatively low number of long range NOE's (Fig. 2B ) for these residues indicate the His-tag to be flexible in solution. Figure 1 1 H- 15 N HSQC spectrum of ChaB at 600 MHz, pH 6.3 in 50 mM CaCl 2 . Folded resonances are indicated by asterisks. Numbering shown includes the N-terminal His-tag (residues 1–21). The native ChaB sequence starts at P22. The unassigned peak is denoted as a question mark. Folded arginine/lysine side chain resonances are indicated by SC. Figure 2 Plots of 1 H- 15 N heteronuclear NOE, NOE constraints and RMSD statistics for ChaB. (A) 1 H- 15 N heteronuclear NOE acquired at 600 MHz. (B) Summary of all unassigned unambiguous NOE constraints: intra-residue, sequential, medium and long range NOEs are shown as blue, green, red and black bars respectively. (C) Backbone RMSD's calculated for the 17 lowest energy ChaB structures based on superposition of residues P22-S96. Solution structure of ChaB The 3D structure of ChaB (Fig. 3 ) is well defined by the structural constraints (Table 1 ) and is dominated by two, relatively long central helices comprising residues H40-Q55 (helix α2) and D65-E83 (helix α3) and a small N-terminal helix (helix α1, E31-K34), which is terminated by a proline (P30). At the C-terminus, a short two-strand β-sheet is observed involving residues Y84-K86 and W92-K94. A tightly packed hydrophobic core stabilizes the overall fold of ChaB. The following hydrophobic residues have < 10% of their surface area exposed to the solvent: Y23, L29, V33, L37, A41, I44, Y45, A48, F49, A52, A72, A76, V80, Y84, A85 and W92. Many of the hydrophobic contacts are between the two long helices (α2 and α3). The C-terminal β-sheet acts as a "cap" for hydrophobic residues from loop 1 (V36, L37), which connects helices α1 and α2, and residues at the N-terminus of helix α2 (A41) and the C-terminus of α3 (V80). Both central helices are largely amphipathic, with residues D43, K46, E47, D54, of helix α2 and E70, K74 and K81 of helix α3 exposed to the bulk solvent and contributing to a highly charged ChaB surface. Most notable, an area of negative charge is observed at the highly mobile loop 2 and the helices immediately surrounding it. In addition, K74, K81, K86, and K95 contribute to a positively charged area while Y56, V75 and A79 present a small hydrophobic patch. Table 1 Constraints and structural statistics for ChaB Constraints used for structure calculation (all residues) Total NOE constraints 2140 Intraresidue NOEs (n = 0) 515 Sequential NOEs (n = 1) 432 Medium Range NOEs (n = 2,3,4) 379 Long Range NOEs (n > 4) 486 Total Unambigous NOEs 1794 Ambiguous NOE restraints 346 Dihedral angle constraints 49 15 N- 1 H residual dipolar couplings 58 Average RMSD to mean structure (Å) (residues 22–96) Backbone atoms 0.397 All heavy (non-hydrogen atoms) 0.807 Average energy values (kcal mole -1 ) quoted for residues 1–96 E total -332.42 ± 9.46 E bond 11.93 ± 0.89 E angle 84.06 ± 1.56 E improper 16.35 ± 0.58 E VdW -515.11 ± 10.31 E NOE 41.39 ± 3.75 E dihedral 0.86 ± 0.21 E sani 28.09 ± 3.11 Deviation from idealised covalent geometry Bonds (Å) 0.0028 ± 0.0001 Angles (°) 0.4470 ± 0.0042 Improper (°) 0.353 ± 0.006 RMSD from experimental data Distance restraints (Å) 0.016 ± 0.0007 Dihedral angle restraints (°) 0.302 ± 0.094 Average Ramachandran statistics for 17 lowest energy structures (residues 22–96) Residues in most favored regions 78.2% Residues in additional allowed regions 8.6% Residues in generously allowed regions 3.3% Residues in disallowed regions 0.0% Analysis of residual dipolar coupling RMSD (Hz) 1.495 ± 0.097 Q-factor 0.138 ± 0.0059 Correlation coefficient 0.98 Figure 3 Solution structure of ChaB . Ensemble of the 17 lowest energy structures showing (A) backbone and (B) heavy-atom traces. Superposition was made over residues comprising the native ChaB sequence (P22-S96). (C) Ribbon representation of the lowest energy ChaB conformer. In general, the secondary structure elements of ChaB are well defined exhibiting RMSD's of 0.14 Å and 0.39 Å for backbone and all non-hydrogen atoms, respectively. This is confirmed by the heteronuclear NOE data, which show a 10% trimmed weighted mean of 0.77 ± 0.03 in the structured regions and indicate lack of motions on the nanosecond timescale (Fig. 2A ). Regions connecting the secondary structural elements exhibited lower heteronuclear NOE values. In particular, the loops connecting helices α2 and α3 (loop 2, Y56-D64) and the two strands of the β-sheet (loop3, G87-K91) exhibited NOE values below 0.65, indicating large amplitude nanosecond motions in these regions. These motions manifest as regions with large RMSD values in the structural ensemble (Fig. 2C ). The small sheet region at the C-terminus was also seen to exhibit some motional freedom particularly for the second strand. However, it is notable that its NOE values are substantially higher than the surrounding loop. The ChaB family and structurally similar proteins Figure 4 shows the alignment of protein sequences related to E. coli ChaB. The most conserved residues make up the hydrophobic core of ChaB, particularly the two long helices and the small sheet. With the exception of P38 and A79 all these residues exhibit < 5% solvent accessibility. These residues are critical for defining the overall fold of ChaB and suggest that all proteins within this family adopt a similar fold. Figure 4 Sequence alignment of the family of ChaB protein s. (A) Alignment of ChaB from E. coli aligned with a series of related proteins identified by Pfam [34]. In bold above alignments, are residues most conserved among ChaB proteins. Cartoon diagram above represents the secondary structure of ChaB. ChaB from E. coli (in bold) Salmonella typhi (Q8XGJ2)and Methanosarcina mazei (Q8PYS9) are classified as group I ChaB proteins. Group II ChaB proteins are all found in Baculoviridae. The figure was created using BOXSHADE (EMBnet). Identical amino acids are highlighted in black and homologous residues in gray. (B) Sequence alignments of ChaB and related proteins σ 70 and σ RN based structural composition (see text for details). Structural homologues of ChaB in the PDB were identified using the DALI server [ 13 ] (Fig. 5 ). This yielded several matches with fragments of other structures, the best match being sigma factor σ 70 (PDB code 1SIG, [ 14 ]) with a DALI Z-score of 4.7. A DALI-Z score greater than 2.0 is considered structurally similar. Another sigma factor, σ RN [ 15 ], with little sequence homology to σ 70 was identified with a DALI-Z score of 3.5. ChaB, however, exhibits no significant sequence similarity with these proteins. Sigma factors are proteins, which bind to DNA dependent RNA polymerases to form the holeoenzyme [ 16 , 17 ]. Although the observed structural similarities do not define a functional role for ChaB it is worth noting that the σ RN domain is classified in the cyclin-box fold of proteins [ 15 ], a class of proteins that bind a diverse set of proteins and nucleic acids [ 18 ]. Figure 5 Comparison of ChaB with structurally similar proteins. Structural similarities between (A) ChaB, (B) σ 70 (PDB code 1SIG [14]) and (C) σ RN (PDB code 1H3L, [15]) proteins identified from the DALI server. The three helices in ChaB are colour coded, with the equivalent helices in σ 70 and σ RN similarly coloured. An interesting observation can be made when aligning the sequences of ChaB, σ 70 and σ RN based on their structural similarity (Fig. 4B ). Residues in the three structures with < 10% of the surface exposed to the solvent are highlighted in red. Clearly, the hydrophobic core, critical for the ChaB fold (marked above the sequence alignments in Fig. 4 ) is also important for the fold observed in the sigma factors. Key hydrophobic residues appear in similar locations in their "structural space" between the three proteins, forging contacts important for stabilizing the fold. These residues are among the most conserved in the two sigma factor families and within ChaB proteins. Loop 2 has weak affinity for divalent ions Given the proposed function of ChaB as a regulator and the effect of magnesium as an inhibitor of the Ca + /H + antiporter ChaA, we examined the influence of calcium and magnesium ions on ChaB 15 N- 1 H chemical shifts. The pattern of perturbed shifts (summarised in Fig. 6A for Ca 2+ ) indicates that the highly charged (Fig. 6B ) flexible loop 2 and surrounding regions are most important for binding. Chemical shift perturbations of similar magnitude and direction were witnessed upon addition of MgCl 2 indicating that Mg 2+ has a similar binding site and affinity for ChaB as Ca 2+ . The observed association constant for CaCl 2 is weak (the K D was estimated to be > 10 mM by NMR) and not likely to be physiologically significant. Given ChaB's proposed role as a regulatory protein, it is possible that the affinity for Ca 2+ or Mg 2+ is increased in the presence of ChaA or ChaC. Figure 6 CaCl 2 titration and surface potential map of ChaB. (A) Summary of the perturbations of CaCl 2 on the backbone 1 H- 15 N chemical shifts of ChaB. The change in chemical shifts (determined from a weighted vector sum of 1 H and 15 N ppm deviations) are mapped onto the structure of ChaB using a colour gradient from blue, to red to yellow, where yellow is the largest perturbation and blue the smallest. Residues that could not be analyzed such as overlapping residues or residues that do not exhibit NH resonances are coloured grey. (B) Potential map of the surface of ChaB calculated using MOLMOL [35] shown in the same orientation as (A). Residues most perturbed by Ca 2+ cluster around a highly negatively charged patch on the ChaB surface comprising a flexible loop. A functional role for loop 2? ChaB proteins are classified into two major groups based on their sequence alignments (Fig. 4 ). Group I consists of ChaB proteins found in bacteria ( E. coli and Salmonella typhi ) and archeabacteria ( Methanosarcina mazei ), while Group II contain ChaB related proteins that are found in Baculoviridae. Thus far, no ChaB domains have been identified in vertebrate and plant species. One major difference between the two classes of ChaB proteins is the presence of the charged loop (loop 2, Fig. 4 ) that we have shown to bind weakly to Ca 2+ and Mg 2+ ions. The EMBL European bioinformatic database annotates ChaB proteins found within group II (Baculoviridae, Fig. 4 ) as putative DNA binding proteins. Interestingly, the σ-factor domains that are structurally similar to ChaB are known to bind DNA at the position equivalent to helix α3 in ChaB. However, the composition of the corresponding loop in sigma factors is more hydrophobic and/or shorter than in ChaB. Members of the ChaB sequences belonging to group I (Fig. 4 ) are annotated as cation transport regulators based on being part of the ChaA operon. The alignment results suggest that the loop 2 region, which is only observed in the group I family of ChaB, is correlated to its function as a cation transport regulator protein. Clearly, further experiments are required to test this hypothesis. Conclusion As part of the E. coli structural genomics project, we report the first 3D structure of a member from the ChaB protein family. E. coli ChaB is a putative cation transport regulator protein whose structure resembles the cyclin-box fold. ChaB was shown to have weak affinity for calcium and magnesium ions at a highly charged and mobile loop that is only present in ChaB family members associated with a cation transporter. We hypothesise that this loop may play a role in the function of ChaB as a regulator of cation transport. Methods Cloning, expression and purification of ChaB The gene encoding full length ChaB (residues 1–76) was amplified from genomic E. coli DNA strain O157:H7 using oligos OPI403 AAAAAAG GATCCC CGTATAAAACGAAAAGCGACCTG and OPI499 AAAAAA GAATTC TTACGATTTTTTATGCCATTTATCATCA. Underlined are restriction sites BamHI and EcoRI for both oligos, respectively. The product was cloned into the BamHI/EcoRI site of pFO-1. The plasmid pFO-1 is a tailored pET-15b vector (Novagen Inc., Madison, WI), which contains an extended poly-linker region and an 8× N-terminal histidine tag with a modified thrombin cleavage site. The ChaB construct (plasmid ID: pPI489) was expressed in E. coli BL21-Gold (DE3) cells (Stratagene) as an 8× His-tagged fusion protein. At an OD 600 of 0.8, the cells were induced with 1 mM IPTG and grown for another 3 hours at 30°C. The protein was purified to homogeneity by absorption onto a Ni 2+ charged chelating sepharose column (Amersham Biosciences) under native conditions. The recombinant protein used for NMR studies consists of the ChaB sequence with an extra 21 residues (MGSSHHHHHHHHSSGFNPRGS) at the N-terminus containing the 8× His-tag and a thrombin cleavage site. This tag replaced the first residue, a methionine, of the genomic sequence of ChaB. In the analysis vide ante ChaB refers to residues P22-S96 of our construct (i.e. the wild type ChaB sequence excluding the first Met residue). Thus, ChaB begins at P22 in our numbering scheme, corresponding to P2 in the native ChaB sequence. The mass of ChaB was confirmed by SDS-PAGE and electrospray mass spectroscopy. NMR Spectroscopy Uniform enrichment of ChaB with 15 N and/or 13 C was achieved by growing the bacteria in M9 minimal medium supplemented with BME vitamins (SIGMA) and ( 15 NH 4 ) 2 SO 4 and/or 13 C 6 -glucose as the sole nitrogen and carbon sources at 37°C. ChaB was purified as described above. NMR samples were obtained by exchanging ChaB into an NMR buffer comprising 50 mM CaCl 2 at pH 6.3 using a PD-10 column and subsequent concentration to ~200–300 μL using an Amicon Ultra-4 (5 KDa cutoff, Millipore). Typical protein concentrations ranged from 1.5–2.0 mM. NMR spectra for resonance assignments were recorded at 303 K on a Bruker Avance DRX 600 MHz spectrometer equipped with a triple-resonance CryoProbe and processed with NMRPipe [ 21 ]. Backbone 1 H, 13 C and 15 N assignments were completed from CBCA(CO)NH, CBCANH and HBHA(CBCACO)NH spectra using a combination of NMRView [ 22 ] and SMARTNOTEBOOK (a module designed for semi-automated assignment in NMRView) [ 23 ] packages. 1 H, 13 C and 15 N sidechain assignments were obtained by manual analysis of the H(CC)(CO)NH, C(C)(CO)NH and HCCH-TOCSY experiments using NMRView and in-house written scripts. 1 H, 13 C and 15 N chemical shifts were referenced to DSS according to the IUPAC recommendation [ 24 ]. Distance constraints were obtained from a simultaneous 3D 13 C/ 15 N-edited NOESY experiment (τ m = 120 ms) in 90% H 2 O/10% D 2 O, and 13 C-edited NOESY (τ m = 100 ms) and 13 C-edited NOESY (aromatic region) (τ m = 100 ms) experiments acquired in 99.9% D 2 O. The experiments in D 2 O were acquired at 800 MHz on a Varian INOVA spectrometer at NANUC). A 4D 13 C- 13 C edited NOESY experiment (τ m = 100 ms) was acquired at 600 MHz to resolve ambiguities involving methyl groups. For all experiments at 600 MHz, the minimal number of scans dictated by the phase cycle was used in combination extensive folding in 15 N and 13 C to reduce experimentation time. Additional restraints used in structure calculations were: dihedral restraints, derived from 3J HN-Cα coupling constants obtained from the HNHA experiment [ 25 ] and 1 H- 15 N residual dipolar couplings extracted from comparison of IPAP-HSQC experiments recorded on ChaB with and without 11 mg/mL Pf1 phage [ 26 ]. For the measurement of dipolar couplings, the NMR buffer was altered to 50 mM phosphate and 100 mM NaCl, pH 6.3 since the presence of CaCl 2 precipitated the Pf1 phage. Steady state { 1 H}- 15 N NOE spectra were acquired in an interleaved manner in which each individual FID was collected with and without presaturation and a recycle delay of 4 s [ 27 ]. Saturation was achieved using a train of 120° pulses separated by 5 ms for a total irradiation time of 3 s. Structure calculations A set of unambiguous NOE constraints were extracted from the 3D-NOESY spectra and used in conjunction with dihedral angle restraints to generate a preliminary fold of ChaB using CNS1.1 [ 28 ]. The resulting structures were used as model templates for automated assignment of NOE peaks using the ARIA 1.1 package [ 29 ]. In many cases, the 4D 13 C- 13 C NOESY experiment was important for manually assigning a number of ambiguous assignments. A total of 1794 unambiguous and 346 ambiguous NOE restraints were obtained from this method and used in combination with dihedral restraints to calculate an ensemble of ChaB structures using CNS [ 28 ]. These structures were further refined using residual dipolar coupling restraints. The axial and rhombic components of the alignment tensor were obtained from the histogram method [ 30 ] and optimized by a grid search [ 31 ] and determined to be D a = 13.7 and R = 0.325. Only residues exhibiting a heteronuclear NOE > 0.65 were included as residual dipolar couplings. Seventeen lowest energy structures with the fewest violations were selected to represent the ChaB structure. No NOE violations over 0.2Å were observed. Structural statistics for this ensemble as calculated by CNS [ 28 ], PROCHECK [ 32 ] and SSIA [ 33 ] are summarised in Table 1 . The coordinates have been deposited in the RCSB under PDB code 1SG7. Titration with calcium and magnesium The effect of calcium on ChaB was determined from addition of aliquots of 5 M CaCl 2 or 2 M MgCl 2 to 15 N labeled ChaB. Prior to titration, metal impurities were removed by addition of EDTA to the ChaB sample followed by exchange into 20 mM Bis-Tris buffer, pH 6.3 using a PD-10 column. Aliquots of CaCl 2 or MgCl 2 were added up to a final concentration of 50 mM. Minimal changes in pH and volume were ensured throughout. Chemical shift perturbations were measured as a weighted vector sum of the 1 H and 15 N chemical deviations: {[(Δ 1 H ppm) 2 + (Δ 15 N ppm × 0.2) 2 ] 0.5 }. List of abbreviations NMR: nuclear magnetic resonance NOE: nuclear Overhauser enhancement HSQC: heteronuclear single quantum coherence PPM: parts per million RMSD: root mean squared deviation PDB: Protein Data Bank Authors' contributions MJO expressed and purified isoptically enriched ChaB, collected all NMR spectra at 600 MHz, processed and analyzed NMR data, performed structural calculations and structural refinement. NS identified ChaB among a series of E. coli proteins cloned as part of the structural genomics initiative. PI completed the initial cloning of chaB. NS expressed, purified and characterized ChaB by mass spectrometry. MJO drafted and NS contributed to the written manuscript. KG coordinated and provided financial support for this study.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC514712.xml

387269

Avoiding URL Reference Degradation in Scientific Publications

Arguments are presented concerning the deposit of Internet-based information into the Internet Archive, a digital library of Internet sites and other digital data

While we applaud PloS' use of Digital Object Identifiers (“The What and Whys of DOIs,” PLoS Biol 1: e57 doi: 10.1371/journal.pbio.0000057 ), we also note the lack of provisions in your instructions for authors for preserving access to electronic information residing at a cited Internet addresses via Uniform Resource Locators (URLs). Medical and scientific literature increasingly cites information only found on the Internet. However, URLs may become inaccessible shortly after article publication. Please consider requiring PLoS' authors to (1) submit all cited URLs to the Internet Archive ( www.archive.org ), a nonprofit organization that has been preserving electronic content since 1996, and (2) maintain a printed copy of the electronic information for future communication until the URL becomes available at the Internet Archive (about a six-month lag time). The Internet Archive, the largest digital library of Internet sites and other digital data, stores cited Internet information at no cost to the author, reader, or publisher. By requiring PLoS' authors to submit all cited Internet-based information to the Internet Archive, PLoS will better preserve the integrity of its content for the future. PLoS' Response Ms. Kelly and colleagues raise an important issue about the ephemeral nature of many information sources on the Internet. In the case of online scholarly literature, information is more likely to be archived and able to be found—indeed, an open-access article is one in which, according to the Bethesda Definition, “A complete version of the work and all supplemental materials, including a copy of the permission…, in a suitable standard electronic format is deposited immediately upon initial publication in at least one online repository that is supported by an academic institution, scholarly society, government agency, or other well-established organization that seeks to enable open access, unrestricted distribution, interoperability, and long-term archiving (for the biomedical sciences, PubMed Central is such a repository).” Other types of Internet-based information are more likely to change, move, or be removed. We agree that wherever possible we must find a way to preserve the relevant information from the sources cited in our articles. PLoS has always encouraged authors to submit supporting information for their research articles, including raw datasets, spreadsheets, multimedia, and snapshots of Web-based interactive tools. PLoS makes this supporting information available to everyone for download and use. PLoS also requires authors to deposit all appropriate datasets, images, and information in public databases and to list the relevant accession and version numbers in the article. The question, then, is how PLoS and its authors can preserve access to other Internet-based information, including organizational Web sites, articles in the popular media, or interactive databases. Although submitting cited URLs to the Internet Archive is worthwhile, it is still (unfortunately) far from ideal. The Internet Archive is best at archiving simple HTML and may be the most appropriate place to archive a Web site an author has cited for its static information content. The Internet Archive does not, however, archive content with password restrictions or “crawling” restrictions, and it allows the removal of already archived content at the request of Web administrators; it would therefore not be an effective archive, for example, for popular press articles that have restricted access. In addition, the Internet Archive cannot preserve functions that interact with the originating server, so it is not an appropriate way to archive a Web site an author has cited, for example, for its useful interactive tools. Finally, there is currently no automated way for publishers to redirect links from the original address to the address on the Internet Archive. For the time being, PLoS plans to review all electronic citations on a case-by-case basis and, when appropriate, request that authors submit the cited Web site URL to the Internet Archive and additionally submit a digital copy of the information to PLoS for internal archiving. We would also like to encourage further input on this issue from the scientific and medical community and urge them to support the Internet Archive and other organizations working to preserve the digital record for future generations.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC387269.xml

539045

Is It Ethical to Use Enhancement Technologies to Make Us Better than Well?

Background to the debate: A variety of biomedical technologies are being developed that can be used for purposes other than treating disease. Such “enhancement technologies” can be used to improve our appearance and regulate our emotions, with the goal of feeling “better than well.” While these technologies can help people adapt to their rapidly changing lifestyles, their use raises important ethical issues.

Arthur Caplan's Viewpoint: Nobody Is Perfect—But Why Not Try to Be Better? Perfection has come in for a lot of bad press recently. A torrent of books and articles has recently appeared [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 ], all raising serious ethical questions about the wisdom and morality of trying to use biomedical knowledge to perfect ourselves or our offspring. Biomedical scientists and physicians might be inclined to ignore this literature as just so much abstract philosophical handwringing. After all, it is almost impossible to find mainstream scientists arrogant enough to proclaim their interest in perfecting anything, much less themselves or their fellow human beings. Beating up on the pursuit of perfection is silly. As Salvadore Dali famously pointed out, “Have no fear of perfection—you'll never reach it.” Critics of those who allegedly seek to perfect human beings know this. While often couching their critiques in language that assails the pursuit of perfection, what they really are attacking is the far more oft-expressed—albeit far less lofty—desire to improve or enhance a particular behavior or trait by the application of emerging biomedical knowledge in genetics, neuroscience, pharmacology, and physiology. Those who might accurately be termed “anti-meliorists” wonder how we will ever resist the obvious temptation to put this knowledge to use to alter ourselves. They are quick to note that we have already given in to such temptation—we augment our breasts, smooth our wrinkles, and pump ourselves full of antidepressants. It is in our nature as humans to strive for self-improvement (Illustration: Margaret Shear) Putting the brakes on biologically driven human betterment would have real consequences for science. Some lines of research would be slowed or restricted [ 3 , 5 , 8 ]. Their application would be declared off-limits or at least tightly regulated [ 1 , 2 , 3 , 4 , 5 , 7 , 8 , 9 ]. Why is the drive to improve ourselves so disturbing to the anti-meliorists? Their arguments cluster around three key worries: that the pursuit of perfection by biomedical means is vain, selfish, and unrewarding [ 1 , 2 , 3 , 6 , 7 ], that improving ourselves is unfair [ 1 , 3 , 4 ], and that enhancement or improvement violates human nature [ 2 , 4 , 5 , 7 , 8 , 9 ] and may actually destroy it [ 2 , 5 , 7 , 9 ]. It is the last of these arguments that is at the core of anti-meliorist concerns. It cannot simply be the pursuit of improvement that is making anti-meliorists nervous. Many religious traditions and spiritual movements seek perfection [ 10 , 11 , 12 , 13 ], but these evoke no negative commentary from the anti-meliorists. Nor do efforts to improve animals and plants set this crowd aflutter. Rather, it is biomedical knowledge being applied to you and me that is the crux of their concern. They fear that in applying new biomedical knowledge to improve human beings, something essential about humanity will be lost. If biomedical tinkering is allowed, we will destroy the very thing that makes us human—our nature. Anti-meliorism rests, however, on a very shaky foundation. To support their position, the anti-meliorists must state what human nature is. They do not. They must also be very clear about why they see human nature as static. They are not. And they must advance an argument about why human nature, which has presumably evolved in response to an enormous array of random forces, tells us anything about what is good or desirable in terms of the traits humans should possess. They cannot. The fight over whether there is any such thing as human nature is a long-standing one [14] . But one can concede that we are shaped by a causally powerful set of genetic influences and still remain skeptical as to whether these produce a single “nature” that all members of humanity possess. Is there a single trait or fixed set of traits that defines the nature of who we are and have been throughout our entire existence on this planet? Unless they can articulate this Platonic essence, anti-meliorists do not have a foundation for their argument that change, improvement, and betterment are grave threats to humanity. Worse still for anti-meliorists, we are clearly creatures who have long tinkered with ourselves, using all manner of technologies from clothing to telescopes to computers to airplanes. Our view of our “nature” is closely linked to the technologies that we have invented and to which we have adapted [15] . We are already technological creatures. Nor is there any normative guidance offered by our evolutionary history that shows why we should not try to improve upon the biological design with which we are endowed. Augmenting breasts or prolonging erections may be vain and even a waste of scarce resources, but seeking to use our knowledge to enhance our vision, memory, learning skills, immunity, or metabolism is not obviously either. Ultimately, anti-meliorism posits a static vision of human nature to which the anti-meliorists mandate we reconcile ourselves. If anything is clear about human nature, it is that this is not an accurate view of who we have been or what we are now, or a view that should determine what we become. Carl Elliott's Viewpoint: Pharma's Gain May Be Our Loss Those of us who worry about medical enhancement are usually less worried about the technologies themselves than about the larger social effects of embracing them too enthusiastically. Just as you do not need to object to cars to worry about urban sprawl, you do not need to object to enhancement technologies to question where these technologies may be taking us. It is not just technophobes who wonder whether a society that consumes 90% of the world's supply of methylphenidate (Ritalin), where the most profitable class of drugs is antidepressants, and where cosmetic surgeons perform liposuction on prime-time television is a society that has somehow lost its way. Let's look at three of the most commercially successful medical enhancements of recent years: selective serotonin reuptake inhibitors, hormone replacement therapy, and the diet drug fenfluramine-phentermine (Fen-Phen). What can we learn from these interventions? First, the manufacturers of enhancement technologies will usually exploit the blurry line between enhancement and treatment in order to sell drugs. Because enhancement technologies must be prescribed by physicians, drug manufacturers typically market the technologies not as enhancements, but as treatments for newly discovered or under-recognized disorders. Selective serotonin reuptake inhibitors were marketed not as personality enhancers, or even only as treatments for clinical depression, but as treatments for questionable illnesses like “premenstrual dysphoric disorder” [16] . Fen-Phen was sold not as a mere diet drug but as a treatment for obesity, which Wyeth, the manufacturer, portrayed as a dangerous public health problem [17] . Estrogen replacement therapy was initially marketed as a risk-free way for women to extend their youthfulness. But when a 1974 study found that estrogen replacement therapy was associated with an increased risk of endometrial cancer, the manufacturers added progesterone, renamed the combination “hormone” replacement therapy, and recast it as a treatment for medical problems associated with menopause such as osteoporosis [6] . Where is the pursuit of the perfect face, body, and mind taking us? (Illustration: Margaret Shear) Second, an alarming number of supposedly risk-free enhancements have later been associated with unanticipated side effects, some of them deadly. Wyeth has set aside over $16 billion to compensate the thousands of patients who have developed valvular heart disease and pulmonary hypertension after taking Fen-Phen [18] . A 2002 National Institutes of Health study found that hormone replacement therapy was associated with such an elevated risk of heart disease, stroke, pulmonary emboli, and breast cancer that the study was stopped prematurely [19] . Selective serotonin reuptake inhibitors are currently embroiled in controversy over whether they are associated with an elevated risk of suicide [20] . Third, the most successful enhancement technologies have been backed by tremendously influential public relations campaigns. These campaigns have included ghostwritten journal articles, industry-funded front groups, and lucrative payments to academics, professional societies, and university centers [21] . For example, GlaxoSmithKline marketed paroxetine (Paxil) by promoting the previously obscure diagnosis of “social anxiety disorder” through phony support groups, celebrity spokespeople, a direct-to-consumer illness awareness campaign, and generous payments to key opinion leaders [22] . The manufacturers of estrogen replacement therapy marketed the hormone in the 1960s by funding a “research foundation” for Robert Wilson, the gynecologist and author of the best-selling book Feminine Forever [6] . Wyeth marketed Fen-Phen by funding obesity research centers, launching public fitness campaigns, contracting with a medical education company to produce a series of ghostwritten journal articles, and making generous payments to academic physicians who then published extensively and testified for the drug's safety to the Food and Drug Administration [17] . The traditional worry about enhancement technologies is that users of the technologies are buying individual well-being at the expense of some larger social good. I may improve my own athletic ability by taking steroids, but I set off a steroid arms race that destroys my sport. I may get cosmetic surgery for my “Asian eyes” or use skin lighteners for my dark skin, but I reinforce the implicitly racist social norms that say that Asian eyes or dark skin are traits to be ashamed of. The worry is that some aspect of the way we live together, collectively, is going to be damaged by actions that we take individually [4] . A market-driven health-care system brings this worry much closer to home. The pharmaceutical industry is now the most profitable and politically powerful industry in the United States [23] . It also has a huge financial interest in creating a demand for enhancement technologies. The pharmaceutical industry can buy politicians to pass industry-friendly legislation; it can buy academic scientists to publish favorable journals articles; it can buy professional societies and patient support groups to spread the word on the newly medicalized disorders that its interventions are developed to treat [24] . It can even buy bioethicists to dispense with any moral concerns [25] . In this kind of political and economic climate, how likely is it that dissenting voices will have any effect before it is too late? Caplan's Response to Elliott's Viewpoint Elliott professes to be unhappy about enhancement. What arguments does he present to support his unhappiness? Not many, and the arguments that he does offer miss the point completely. If people want to feel better, sleep less, have fewer hot flashes, better vision, or fewer wrinkles, then they may want to use enhancement technologies to achieve these things. Technology in itself isn't driving us in any particular direction—I believe that we decide where it should go. Elliott, however, gravely warns us that you and I do not really decide a direction when it comes to matters of enhancement. It is—listen carefully for the Darth Vader–esque hissing—drug companies! The rest of Elliott's viewpoint amounts to what is his increasingly familiar harangue against the pharmaceutical industry. The drug companies sucker us into buying enhancement by getting us hooked on pseudotherapies. The drug companies rob us of our will to fend off their siren-like messages of better living through their chemistry. And the drug companies get us feeling so bad about ourselves that we empty our wallets on their latest overpriced geegaws. Pharmaceutical companies may be evil incarnate. And we may be putty in their pecuniary little hands. But that has nothing at all to do with the question of whether there is anything wrong with pursuing enhancement. When Elliott eagerly dons his hair shirt to bemoan Big Pharma, he finds so much sin to revel in that he forgets to give a reason, any reason, why enhancement is, in itself, immoral. At most he presents an argument for keeping the pharmaceutical industry out of enhancement. Okay, so let's take Big Pharma out of the picture. If we left the encouragement of enhancement to the government, the military, schools, foundations, doctors, or parents, would this now be morally acceptable? I think sometimes it would be. And nothing that Elliott says provides any reason to think otherwise. Elliott's Response to Caplan's Viewpoint Caplan does not defend medical enhancement so much as attack its critics. Or rather, he attacks a small group of conservative critics who want to preserve “human nature.” He dispatches those critics with admirable precision, but I am not sure why he believes that group of critics includes me. My worry about enhancement technologies has little to do with human nature. My worry is that we will ignore important human needs at the expense of frivolous human desires; that dominant social norms will crowd out those of the minority; that the self-improvement agenda will be set not by individuals, but by powerful corporate interests; and that in the pursuit of betterment, we will actually make ourselves worse off. It's no secret that many Americans are deeply ashamed of their personal shortcomings and inadequacies. Nor is it any secret that these shortcomings and inadequacies can be exploited for commercial profit. But do we really want to submit our health-care system to the same forces that have made millionaires out of motivational speakers and diet book authors? Skepticism about enhancement technologies is not equivalent to a wish to set back medical research and declare some applications off-limits. This is a debate about enhancing human traits, not curing human illness. To say that our medical research agenda will be set back if we restrict enhancement technologies makes no more sense than saying that cancer surgery will be set back if the American Broadcasting Corporation cancels its cosmetic surgery reality TV show Extreme Makeover . We live in a country where 46 million uninsured people cannot get basic medical care, while the rest of us spend a billion dollars a year on baldness remedies. It is not just the inequity here that is so impressive. It is the fact that we have gotten so accustomed to the inequity that we do not see it as obscene.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539045.xml

524028

Amino acids as regulators of gene expression

The role of amino acids as substrates for protein synthesis is well documented. However, a function for amino acids in modulating the signal transduction pathways that regulate mRNA translation has only recently been described. Interesting, some of the signaling pathways regulated by amino acids overlap with those classically associated with the cellular response to hormones such as insulin and insulin-like growth factors. The focus of this review is on the signaling pathways regulated by amino acids, with a particular emphasis on the branched-chain amino acid leucine, and the steps in mRNA translation controlled by the signaling pathways.

Introduction Recent advances in biomedical research reveal a key role for amino acids as nutritional signals in the regulation of a number of cellular processes. Studies employing a variety of cell types and different tissues demonstrate that one such process affected is the regulation of gene expression through modulation of the translation of messenger RNA (mRNA). The studies show that cells recognize changes in amino acid availability and generate alterations in signal transduction pathways that are also regulated by hormones and growth factors. The cells then respond to the integrated signaling input by either upregulating or downregulating translation initiation, i.e., the process during which initiator methionyl-tRNA (met-tRNA i ) and mRNA bind to a 40S ribosomal subunit followed by the joining of a 60S ribosomal subunit to form a translationally competent 80S ribosome. The response of translation initiation to a change in amino acid and/or hormone availability can be general, i.e., affecting the translation of most if not all mRNAs, and/or specific, i.e., affecting the translation of a single class or subset of mRNAs. Both the general and specific responses can be mediated through regulation of either the met-tRNA i and/or mRNA binding steps. The specific response may also involve an additional regulatory site, i.e., the phosphorylation status of ribosomal protein rpS6, one of the proteins composing the 40S ribosomal subunit. Learning how the cell recognizes a sufficiency of amino acids is presently the objective of intense research. Present evidence, however, suggests multiple recognition sites and multiple signaling pathways. Below, we summarize our current knowledge of the signaling pathways known to respond to changes in amino acid availability. In addition, the translation initiation factors and mRNA structural elements that are involved in changes in both global and specific modulation of mRNA translation are discussed. mRNA translation initiation The first step in translation initiation involves the binding of met-tRNA i to the 40S ribosomal subunit, a reaction mediated by the eIF2•GTP complex [reviewed in [ 1 ]]. In a subsequent step, the GTP bound to eIF2 is hydrolyzed to GDP and eIF2 is released from the 40S subunit complexed with GDP, leaving met-tRNA i behind. Exchange of GDP bound to eIF2 for GTP is mediated by the guanine nucleotide exchange factor eIF2B, and as described below, there are at least three known mechanisms for modulating eIF2B activity in vivo. The second step in translation initiation involves the binding of mRNA to the 40S ribosomal subunit containing the eIF2•GTP•met-tRNA i complex and eIF3 [ 1 ]. The protein that mediates this step is a heterotrimeric complex referred to as eIF4F which consists of the initiation factors eIF4A, eIF4E, and eIF4G. eIF4A is an RNA helicase that serves to unwind secondary structure in the 5'-untranslated region (5'-UTR) of the mRNA, allowing the 40S ribosomal subunit to migrate from the 5'-m 7 GTP cap to the AUG start codon. The helicase activity of eIF4A is stimulated by eIF4B and eIF4H. eIF4E binds to the m 7 GTP cap at the 5'-end of the mRNA and thus plays a crucial role in the binding of the mRNA to the ribosome. eIF4G is a scaffolding protein that binds to eIF4A, eIF4E, and eIF3. Thus, eIF4G is a molecular bridge that links the mRNA, which is bound by eIF4E, to the 40S ribosomal subunit, which is bound by eIF3. Assembly of the eIF4F complex is regulated in part through the reversible association of eIF4E with the translational repressors, eIF4E-binding proteins 4E-BP1, 4E-BP2, and 4E-BP3. The domain on eIF4E to which eIF4G binds overlaps with the binding domain for the 4E-BPs, such that either eIF4G or 4E-BP can bind to eIF4E, but both cannot bind at the same time. Thus, association of eIF4E with a 4E-BP precludes the binding of mRNA to the 40S ribosomal subunit by preventing the binding of the eIF4E•mRNA complex with eIF4G. Association of eIF4E with the 4E-BPs is regulated by phosphorylation of 4E-BP, whereby hypophosphorylated 4E-BPs bind to eIF4E but the hyperphosphorylated proteins do not. Regulation of mRNA translation through Phosphorylation of eIF2 or eIF2B Of the three known mechanisms for regulating eIF2B activity, the best characterized involves phosphorylation of eIF2 on Ser51 of its α-subunit. Phosphorylation of eIF2α converts eIF2 from a substrate into a competitive inhibitor of eIF2B and represses the translation of most mRNAs, but paradoxically enhances the translation of mRNAs containing multiple upstream open reading frames (uORF) and internal ribosome entry sites (IRESs). Phosphorylation of eIF2α is mediated by any of four known eIF2α kinases in mammalian cells: the mammalian ortholog of the yeast general control non-derepressing kinase-2 (mGCN2), the heme-regulated inhibitor (HRI), the protein kinase dsRNA-activated (PKR), and the PKR-like endoplasmic reticulum kinase [PERK, reviewed in [ 2 ]] (Fig. 1 ). In both cells in culture [e.g. [ 3 ]] and livers perfused in situ [ 4 ], deprivation of single essential amino acids promotes phosphorylation of eIF2α with a concomitant inhibition of eIF2B. The phosphorylation of eIF2α that occurs in vivo [ 5 ], in perfused rat liver [ 6 ], and in cells in culture [ 7 ] in response to altered amino acid availability is mediated by the eIF2α protein kinase referred to as mGCN2. In yeast deprived of amino acids, uncharged tRNA accumulates and binds to a domain on Gcn2p that exhibits sequence homology to histidyl-tRNA synthetase resulting in its activation [reviewed in [ 5 ]]. In fasted rats, feeding a meal containing a complete mixture of essential amino acids stimulates protein synthesis in the liver and skeletal muscle, but has no effect on eIF2α phosphorylation or eIF2B activity [ 8 ]. In contrast, feeding a diet lacking a single essential amino acid results in both an increase in eIF2α phosphorylation and a reduction in eIF2B activity in liver [ 9 ], suggesting that an imbalance in plasma concentrations of essential amino acids results in activation of signaling pathways within the liver that result in increased phosphorylation of eIF2α. The enhanced phosphorylation of eIF2α that occurs in response to an imbalanced amino acid mixture is mediated by the eIF2α kinase mGCN2, because in mice lacking the kinase, feeding a diet lacking leucine does not promote eIF2α phosphorylation or inhibition of eIF2B [ 5 ]. However, the mechanism through which severe amino acid deprivation activates mGCN2 in cells in culture, i.e. accumulation of uncharged tRNA, probably isn't relevant in vivo because plasma amino acids are typically maintained at concentrations well above the K m of the aminoacyl-tRNA synthetases, even during fasting, and therefore significant amounts of uncharged tRNA are unlikely to accumulate. Modulation of eIF2α phosphorylation also occurs in response to changes in the availability of nutrients other than amino acids. For example, either hypoglycemia or hyperglycemia promotes eIF2α phosphorylation. Hypoglycemia is thought to activate the endoplasmic reticulum-associated eIF2α kinase termed PERK through induction of the ER stress response [ 10 ]. However, the kinase that phosphorylates eIF2α in response to hyperglycemia is unknown. In vivo, the transient hypoglycemia that occurs shortly after birth results in altered translation of mRNAs encoding several transcription factors such as C/EBPβ that induce the transcription of a number of genes involved in gluconeogenesis and glucose storage such as PEPCK, glucose-6-phosphatase, pyruvate carboxylase, and glycogen synthetase [ 11 ]. A recent study using mice containing a homozygous mutation in the gene encoding eIF2α that replaces Ser51 with an unphosphorylatable Ala residue (eIF2S51A) demonstrated that phosphorylation of eIF2α is a critical component in the response of the newborn to hypoglycemia [ 12 ]. Thus, in neonatal eIF2S51A mice, the activity of PEPCK in the liver is significantly reduced compared to wildtype mice and its induction immediately after birth is severely attenuated. The mechanism through which eIF2α phosphorylation might promote induction of PEPCK gene transcription is as yet unexplored, but has been postulated to be a consequence of altered translation of mRNAs encoding specific transcription factors, e.g. C/EBPα and C/EBPβ. Figure 1 Regulation of eIF2α phosphorylation. Phosphorylation of eIF2α is mediated by four known protein kinases that are regulated by diverse cellular stresses. Phosphorylation of eIF2α inhibits eIF2B which can have both general and specific effects on mRNA translation as described in detail in the text. HRI was first identified in rabbit reticulocytes and shown to be activated in response to hyperoxia and iron and heme deficiency [ 13 ]. Subsequent studies have shown that HRI is also expressed in multiple tissues and is activated by heavy metals and nitric oxide (NO). In fact, NO binds directly to HRI [ 14 ]. Because non-erythroid cells seldom experience large fluctuations in heme content, it has been suggested that NO may be a principle regulator of HRI in such cells [ 14 ]. PKR is a ubiquitously expressed serine-threonine protein kinase that is activated by double-stranded RNA and is induced by interferon [reviewed in [ 15 ]]. PKR is also activated by lipopolysaccharide and cytokines such as IL-1 and TNF-α, and is a key component of the proinflammatory response to bacterial infection. It is a potent inhibitor of cell growth when over-expressed in yeast, mammalian, or insect cells, an effect that is mediated by eIF2α phosphorylation because co-expression of a non-phosphorylatable eIF2α prevents the growth repressive effect [ 16 ]. Unlike the other three eIF2α kinases, eIF2α is not the only substrate for PKR; for example, PKR is reported to phosphorylate the regulatory subunit of protein phosphatase 2A [ 17 ]. PKR also binds to the IκB kinase complex and is involved in NF-κB signaling. In addition to changes in eIF2α phosphorylation, eIF2B activity can be altered through changes in expression of the catalytic ε-subunit. Knockdown of the catalytic ε-subunit using RNA i essentially halts cell growth and triggers apoptosis [ 18 ]. In contrast, overexpression of eIF2Bε, as occurs in many transformed cells, results in increased growth [ 19 ]. Because the ε-subunit alone is not inhibited by phosphorylated eIF2, overexpression of eIF2Bε provides a means of enhancing mRNA translation under stress conditions that promote eIF2α phosphorylation. The mechanism(s) through which eIF2Bε expression is regulated are unknown, but our laboratory has found that a preferential increase in eIF2Bε expression occurs in response to acute resistance exercise and is blocked by pre-treatment with rapamycin, a specific inhibitor of the mammalian target of rapamycin (mTOR) (unpublished observation). Because both nutrients and growth-promoting hormones stimulate the mTOR signal transduction pathway (see the next section for further discussion of mTOR signaling), it is tempting to speculate that expression of eIF2Bε might be enhanced by such stimuli. The guanine nucleotide exchange activity of eIF2B may also be subject to regulation through phosphorylation of its ε-subunit. In vitro, at least four kinases phosphorylate eIF2Bε including casein kinases (CK)-I and -II, glycogen synthase kinase (GSK)-3, and DYRK. Phosphorylation of eIF2Bε by either CK-I [ 20 ] or CK-II [ 20 , 21 ] reportedly stimulates the activity of eIF2B, although this conclusion has been questioned by another group [ 22 ]. Whether or not phosphorylation by GSK-3 alters the activity of eIF2B is likewise controversial. One study [ 20 ] reports that phosphorylation by GSK-3 has no direct effect on eIF2B activity, even though phosphorylation by GSK-3 prevents the subsequent phosphorylation, and thus activation, by CK-I. In contrast, other studies suggest that phosphorylation of Ser535 in rat eIF2Bε (Ser540 in the human sequence) by GSK-3 is required, but not sufficient, for inhibition of eIF2B activity by insulin [ 23 ]. Regulation of mRNA translation through downstream targets of the mTOR signaling pathway The protein kinase mTOR is a common intermediate in both nutrient and hormone signal transduction pathways (Fig. 2 ). Signaling through mTOR is enhanced by nutrients and anabolic hormones, such as insulin or IGF-I [ 24 , 25 ], and repressed by elevation of cAMP [ 25 - 27 ] or activation of AMPK [ 28 - 30 ], suggesting that one function of mTOR is to integrate the anabolic response to nutrients and insulin and the catabolic response to counter-regulatory hormones, such as glucagon. However, mTOR may not be a direct target of nutrient and hormone signaling. Instead, a number of recent studies have identified TSC1•TSC2 as a potential branch point in the nutrient, insulin, and AMPK signaling pathways to mTOR [ 31 , 32 ]. The results of these studies support a model wherein insulin and leucine would repress the inhibitory action of TSC1•TSC2 on mTOR signaling whereas glucagon would stimulate it. In this model, insulin stimulates signaling to mTOR through Akt-mediated phosphorylation of TSC2. Leucine would also modulate signaling through mTOR through the TSC1•TSC2 complex. However, the mechanism through which leucine signals to TSC1•TSC2 is unknown, but is distinct from Akt. Leucine may also modulate signaling through mTOR by altering the association of the kinase with one or more regulatory proteins, such as the regulatory associated protein of mTOR (raptor) and G protein β-subunit-like protein (GβL). In the paragraphs that follow, the evidence supporting these various mechanisms for regulating mTOR is discussed. Figure 2 Regulation of the mTOR signaling pathway. The mTOR signaling pathway is controlled through various upstream kinases (e.g. AMPK, AKT, and MK2) that converge on the tuberous sclerosis complex, TSC1•TSC2. TSC2 is a GTPase-activator protein for Rheb which is a positive effector of signaling through mTOR. mTOR signals to downstream targets such as 4E-BP1 and S6K1 as a complex with the regulatory proteins raptor and GβL as described in detail in the text. The activity of mTOR toward downstream targets such as 4E-BP1 and S6K1 is controlled in part through the interaction of mTOR with the regulatory proteins raptor and GβL. Evidence linking raptor with nutrient signaling through mTOR is provided by studies wherein raptor expression was downregulated using siRNA [ 33 , 34 ]. In such studies, leucine-induced phosphorylation of S6K1 is greatly repressed to an extent similar to that observed in cells in which mTOR expression is reduced. In part, leucine may modulate signaling through mTOR by altering the stability of the mTOR•raptor complex. In this regard, in one study the stability of the mTOR•raptor complex was found to be enhanced in cells subjected to amino acid deprivation [ 33 ]. However, a study by another group [ 35 ] failed to observe a change in binding of raptor to mTOR in cells starved for amino acids. In part, this discrepancy may be explained by the identification of GβL as a second mTOR-interacting protein [ 34 ]. Like raptor, GβL has been shown to co-immunoprecipitate with mTOR [ 34 ]. GβL is a positive regulator of mTOR because co-expression of GβL with mTOR results in greatly increased kinase activity of mTOR toward 4E-BP1 and S6K1 compared to expression of mTOR alone [ 34 ]. Moreover, reducing GβL expression using siRNA represses leucine- and serum-induced phosphorylation of S6K1 [ 34 ], suggesting that GβL is involved in hormone and amino acid signaling though mTOR. Importantly, GβL is necessary for leucine-mediated changes in mTOR•raptor association. In cells deprived of leucine, the binding of both raptor and GβL is high and readdition of leucine to leucine-deprived cells decreases the amount of raptor, but not GβL, associated with mTOR [ 34 ]. However, leucine-induced changes in mTOR•raptor association requires GβL, suggesting that the binding of GβL to mTOR renders the binding of raptor to mTOR sensitive to changes in amino acid availability. The most proximal upstream protein that has been identified in the mTOR signaling pathway is the Ras homolog enriched in brain (Rheb). Rheb is a small G protein that enhances phosphorylation of S6K1, rpS6, and 4E-BP1 in an mTOR-dependent fashion when overexpressed [reviewed in [ 31 , 36 ]]. Moreover, in cells overexpressing Rheb, S6K1 phosphorylation is maintained during starvation for amino acids, suggesting that Rheb is involved in transducing signals from amino acids through mTOR [ 37 , 38 ]. Rheb activity is controlled in part by a GTPase activating protein (GAP) referred to as TSC2 or tuberin. TSC2, and its binding partner TSC1 (a.k.a. harmartin) were originally identified as the product of two genes that are causative in the autosomal dominant syndrome tuberous sclerosis [reviewed in [ 39 - 43 ]]. Mutations in either gene are associated with the widespread development of benign growths in multiple organs and tissues, suggesting that the normal role of these proteins is to restrict cell size and proliferation. This idea has been confirmed in studies in which the Drosophila orthologs of TSC1 and TSC2, dTsc1 and dTsc2, respectively, were shown to function in a complex that acts downstream of AKT but upstream of Drosophila TOR (dTOR) to restrict cell growth and proliferation [ 44 - 46 ]. Studies in both Drosophila [ 47 ] and mammalian cells [ 48 ] have implicated TSC1 and TSC2 in amino acid signaling through TOR. In Drosophila, downregulated expression of either protein causes cells to become resistant to amino acid deprivation [ 47 ]. Thus, S6K phosphorylation is largely maintained during amino acid starvation in cells with reduced expression of either dTsc1 or dTsc2 [ 47 ]. Similarly, in mammalian cells lacking either TSC1 or TSC2, S6K1 phosphorylation is resistant to amino acid deprivation [ 49 ]. Moreover, in mammalian cells in culture, co-overexpression of TSC1 and TSC2 prevents amino acid-dependent activation of S6K1 [ 48 ]. Together, these studies strongly suggest that TSC1 and TSC2 are required for amino acid induced signaling through mTOR. The mechanism(s) involved in the regulation of TSC2 GAP activity are poorly understood, but likely involve phosphorylation of the protein by multiple upstream protein kinases. For example, TSC2 has been shown to be directly phosphorylated by AKT on multiple serine and threonine residues and phosphorylation by AKT represses the inhibitory action of the TSC1/TSC2 complex on signaling through mTOR to 4E-BP1 and S6K1 [ 50 - 53 ]. Likewise, phosphorylation of TSC2 by the MAP kinase regulated protein, MK2, reportedly inhibits TSC2 and leads to activation of mTOR [ 54 ]. In contrast, phosphorylation by the AMP-activated protein kinase (AMPK) on distinct residues activates TSC2 and results in repressed signaling through mTOR, suggesting that the GAP activity of TSC2 is enhanced by AMPK [ 55 ]. Until recently, the kinase that regulates AMPK was unknown. However, a recent study reports that LKB1 phosphorylates AMPK on the activating residue, Thr172, and likely represents an authentic AMPK kinase [ 56 ]. LKB1 was originally identified as a tumor suppressor that functions to limit cell growth, and is ubiquitously expressed in mammalian tissues [ 57 , 58 ]. Alone, LKB1 does not phosphorylate AMPK, but when complexed with two adapter proteins, STRAD and MO25, it exhibits potent AMPK activity [ 56 ]. Two isoforms (α and β) of each protein exist in human cells, and the complex of LKB1 with the α-isoform of each protein, i.e. LKB1•STRADα•MO25α, exhibits greater AMPK kinase activity compared to other permutations of the complex [ 56 ]. In addition to enhancing its AMPK kinase activity, STRAD and MO25 also target LKB1 to the cytoplasm; LKB1 normally is found primarily in the nucleus [ 59 ]. LKB has multiple phosphorylation sites and mutation of either Thr336 or Ser431 prevents LKB1 from inhibiting cell growth [ 60 ]; Thr336 is an autophosphorylation site, whereas Ser431 is phosphorylated by both PKA and p90 rsk [ 61 ]. These studies provide a possible mechanism by which glucagon might downregulate mTOR activity, i.e. phosphorylation of LKB1 by PKA might repress the AMPK kinase activity of LKB1. However, such an idea is still speculative at this point as the effect of LKB1 phosphorylation by PKA on its ability to phosphorylate AMPK has yet to be investigated. mRNA cis-acting elements mediating translational control Changes in translation initiation can manifest as either altered translation of most or all mRNAs (i.e. global changes) or as altered translation of mRNAs encoding specific proteins. The mRNAs that encode proteins whose expression are specifically regulated through changes in mRNA translation (as opposed to changes in global mRNA translation) typically have one or more structural elements within the 5'-untranslated region (5'-UTR) that mediate translational control. Examples of such elements include multiple upstream open reading frames (uORF), internal ribosome entry sites (IRES), highly structured 5'-UTRs, terminal oligopyrimidine (TOP) tracts immediately downstream of the 5'-m 7 GTP cap, and binding domains for specific regulatory proteins (e.g. the iron-responsive element in the ferritin mRNA). Each of these elements serves to modulate the translation of a subset of mRNAs in response to various stimuli. For example, uORF elements repress the translation of most mRNAs under normal growth conditions. The translation of mRNAs bearing multiple uORFs is paradoxically enhanced in response to phosphorylation of the α-subunit of eIF2, an event that is associated with repressed translation of most mRNAs. The mechanism through which eIF2α phosphorylation enhances the translation of mRNAs containing multiple uORFs is complex and involves inhibition of the guanine nucleotide exchange activity of a second translation initiation factor, eIF2B, by phosphorylated eIF2. Examples of enhanced translation of mRNAs containing uORFs concomitant with eIF2α phosphorylation include the induction of the transcription factors ATF4 in mouse embryo fibroblasts deprived of amino acids (5) and CD36 in response to hyperglycemia (6) and the induction of the cationic amino acid transporter (CAT-1) in response to deprivation of amino acids [ 62 ] or glucose [ 63 ]. However, although enhanced translation of mRNAs with uORF sequences has been demonstrated in yeast and in cell lines, a similar phenomenon has not been demonstrated in an intact tissue. A second 5'-UTR structure that allows preferential translation when eIF2α is phosphorylated is an IRES. An IRES allows the ribosome to bind to an internal site in the 5'-UTR and bypass the normal route of association with the mRNA, i.e. binding to the 5'-cap structure [reviewed in [ 64 , 65 ]]. The best characterized IRES-containing mRNA that is regulated by eIF2 phosphorylation is that encoding CAT-1 [ 62 ]. Like many IRES-containing mRNAs, the 5'-UTR of the CAT-1 mRNA has both an IRES and uORFs, and both elements are required for optimal regulation of CAT-1 mRNA translation. Thus, translation of an uORF adjacent to the IRES promotes a rearrangement of the IRES structure, resulting in its activation. However, this mechanism alone is unlikely to account completely for the enhanced translation of the CAT-1 mRNA because enhanced CAT-1 synthesis is delayed several hours after induction of eIF2α phosphorylation, suggesting that synthesis of another protein might be required for translation of the CAT-1 mRNA. Proteins that bind to IRES elements and modulate their function are referred to as IRES-transacting factors (ITAFs). Although poorly characterized, it has been suggested that ITAFs function as RNA chaperones that, upon binding to the IRES, promote refolding of the domain into the correct structure for 40S ribosome binding. Examples of ITAFs include the polypyrimidine tract binding protein (PTB) and upstream of N-ras (unr) that activate the Apaf-1 IRES [ 66 ]. Although eIF2α phosphorylation is one mechanism for enhancing the translation of mRNAs containing an IRES element(s), it is not unique. For example, during apoptosis or infection by certain types of viruses, eIF4G is cleaved. The normal function of eIF4G is to assemble the translation initiation factors eIF4A and eIF4E and the poly(A) binding protein into a complex that mediates the binding of mRNA to the 40S ribosomal subunit. Cleavage of eIF4G during apoptosis or viral infection separates the binding domain for PABP and the mRNA cap binding protein, eIF4E, from the domains that bind eIF4A and allow ribosome attachment (referred to as the middle fragment of eIF4G or M-FAG). A recent study reported that M-FAG generated in etoposide-treated cells M-FAG promotes the preferential translation of certain, but not all, IRES-containing mRNAs including Apaf-1 and death-associated protein (DAP)-5 [ 67 ]. Moreover, a number of IRES-containing mRNAs are preferentially translated under conditions that promote dephosphorylation or decreased function of eIF4E, for example when eIF4E is associated with one of the eIF4E binding proteins such as 4E-BP1. Thus, IRES function can be regulated through multiple mechanisms. Another structural element within the 5'-UTR of some mRNAs that is involved in selective mRNA translation is an oligopyrimidine tract, referred to as a TOP sequence, immediately downstream of the 5'-cap structure [ 68 , 69 ]. Messages containing a TOP sequence include those encoding the ribosomal proteins, eukaryotic elongation factors-1A and 2, PABP, and eIF4G; in other words, proteins involved in protein synthesis. Thus, enhanced translation of TOP mRNAs is one mechanism for increasing ribosome biogenesis and the long-term capacity to synthesize protein. In liver of fasted rats, inhibition of mTOR by rapamycin prevents completely the leucine-induced phosphorylation of S6K1 and rpS6 as well as the increased association of TOP mRNAs with polysomes, suggesting an important role for S6K1 activation in the regulation of TOP mRNA translation [ 70 ]. Similarly, rapamycin prevents the feeding-induced increase in S6K1 phosphorylation in liver and skeletal muscle of neonatal pigs [ 71 ]. However, recent studies [ 72 - 74 ] suggest that activation of S6K1 may not be the only mechanism for enhancing translation of TOP mRNAs, although possible alternatives have not been identified. Most mRNAs that are efficiently translated, e.g. GAPDH and β-actin, have 5'-UTRs that are short (<200 nt), have a low content of G and C residues, and are relatively unstructured [ 75 ]. In contrast, other mRNAs contain long, highly-structured 5'-UTRs. It isn't surprising that in order for the 40S ribosome to reach the AUG start codon of mRNAs with highly-structured 5'UTRs, the RNA helicase activity of eIF4A is essential. However, the results of a recent study suggest that both eIF4A and the eIF4A enhancer eIF4B are required for optimal translation of most mammalian mRNAs, including those such as the β-actin mRNA [ 76 ]. Although little is known about the mechanism(s) through which eIF4A and eIF4B might be regulated, eIF4B is phosphorylated on Ser422 in vitro by S6K1 and leucine-deprivation of cells in culture promotes dephosphorylation of eIF4B [ 77 ]. Thus, eIF4B phosphorylation by S6K1 provides a possible link between hormone and nutrient signaling through mTOR and eIF4A/eIF4B function. Secondary structure may also function as a cis-acting regulatory element through the binding of specific trans-acting factors. A well-characterized example of such regulation is the modulation of ferritin and δ-aminolevulinate (ALA) synthase mRNA translation in response to changes in iron availability [ 78 ]. Both the ferritin and ALA synthase mRNAs contain hairpin structures near the 5'-end of their mRNAs, termed an iron-responsive element (IRE), that specifically binds to the IRE-binding proteins IRP1 and IRP2. Low intracellular iron enhances the IRE binding activity of IRP1 and the stability of IRP2 allowing them to bind to the IRE structure and stabilize it. Because of its proximity to the 5'-cap structure, the IREIRP1/2 complex blocks the binding of the 40S ribosome to the mRNA, thereby preventing translation of the ferritin and ALA synthase mRNAs. Conclusions Nutrients, and in particular certain amino acids, play important roles in the control of gene expression through their ability to modulate the initiation phase of mRNA translation. All essential amino acids have the potential to globally regulate mRNA translation through the eIF2α kinase mGCN2. In addition, changes in eIF2α phosphorylation can selectively modulate the translation of mRNAs encoding particular proteins if the 5'UTR of the mRNA contains uORFs and/or IRES elements. Selective control of mRNA translation can also occur through changes in signaling through mTOR. Activation of S6K1 by mTOR leads to phosphorylation of rpS6 and eIF4B which is thought to promote preferential translation of TOP mRNAs and mRNAs with highly structured 5'-UTRs, respectively. In addition, mTOR phosphorylates the eIF4E binding proteins leading to enhanced assembly of the eIF4F complex. In combination with eIF4B phosphorylation, enhanced eIF4F assembly leads to preferential translation of mRNAs with highly structured 5'-UTRs. Although other amino acids have been shown to increase signaling through mTOR, leucine is arguably the most potent of the amino acids in activating the pathway. Authors' contributions Both authors contributed equally to the writing of this manuscript. Competing interests None declared.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC524028.xml

514896

Primary care follow up of patients discharged from the emergency department: a retrospective study

Background The visit to the emergency department (ED) constitutes a brief, yet an important point in the continuum of medical care. The aim of our study was to evaluate the continuity of care of adult ED visitors. Methods We retrospectively reviewed all ED discharge summaries for over a month 's period. The ED chart, referral letter and the patient's primary care file were reviewed. Data collected included: age, gender, date and hour of ED visit, documentation of ED referral and ED discharge letter in the primary care file. Results 359 visits were eligible for the study. 192 (53.5%) of the patients were women, average age 54.1 ± 18.7 years (mean ± SD). 214 (59.6%) of the visits were during working hours of primary care clinics ("working hours"), while the rest were "out of hours" visits. Only 196 (54.6%) of patients had a referral letter, usually from their family physician. A third (71/214) of "working hours" visits were self referrals, the rate rose to 63.5% (92/145) of "out of hours" visits (p < 0.0001). The ED discharge letter was found in 50% (179/359) of the primary care files. A follow-up visit was documented in only 31% (111/359). Neither follow up visit nor discharge letter were found in 43% of the files (153/359). Conclusions We have found a high rate of ED self referrals throughout the day together with low documentation rates of ED visits in the primary care charts. Our findings point to a poor continuity of care of ED attendees.

Background The emergency department (ED) is intended to treat medical urgencies or emergencies, but a large proportion of visits are due to problems that could be treated in the primary care setting [ 1 , 2 ]. ED services are available 24 hours a day while primary care facilities have limited service hours. In the Israeli health system patients can be referred to the ED by their family practitioner, or by other community health providers, or be self referred. Recently many out of hours community based services have been established but without a significant reduction in visiting rates to in-hospital ED. The visit to the ED constitutes a brief, yet an important point in the continuum of medical care. In today's era of cost effectiveness and increasingly competent primary care physicians, ambulatory investigation, treatment and follow-up have largely replaced prolonged and costly hospitalizations [ 3 , 4 ]. The ED visit however, remains a cross-road which may mark a sudden change in the patient's medical condition. In many cases it may result in introducing new medications, withdrawing others and recommendation of further investigations. The family practitioner is the one expected to coordinate and carry out the treatment and follow-up. The new information given from the ED should be effectively delivered to the family practitioner, the modality usually used is the discharge letter. The continuation of treatment between hospital departments and the primary care physician had been issued in several studies using discharge letters audit [ 5 - 7 ]. Raval et. al. assessed the adequacy of the discharge summary in reporting important investigative results and future management plans in patients hospitalized and discharged with a diagnosis of heart failure [ 5 ]. They found substantial inadequacies in communicating to the community physician that may have implications for continuity of care and subsequent clinical outcome. Wilson et. al. examined the reliability, effectiveness, accuracy and timeliness of hospital to general practitioner information transfer by discharge summaries. In a retrospective audit of 569 patient discharge summaries and related medical records they found that summaries written for patients discharged from hospital were estimated to be received by the patient-nominated general practitioner in 27.1% of cases [ 6 ]. Bolton et al assessed the quality of communications between hospitals and general practitioners. The general practitioner's(GP) name was recorded in 88% of audited records. Few inaccuracies were detected in the medications recorded in the discharge summaries, and on contrary to Wilson et al 77% of discharge summaries were received by the GP [ 7 ]. The continuity of care between the ED and the primary care physician had been assessed for children with asthma [ 8 ] but we did not find data about the continuity of care for adults. To evaluate the continuity of care after ED visits, we evaluated the ED referral and discharge letters, their content, and the documentation of the ED visit in the patients' primary care files. We have focused on discharges from the internal medicine ED. We expected that in these cases the patients would be followed up by their family practitioner. Methods The study was conducted in the district medical center (Kaplan), serving more than 500,000 inhabitants, and in 12 primary care clinics (32 family practitioners), of The Clalit Health Services in the Rehovot region, Israel. In Israel the entire population have a national health insurance by law and each citizen can choose to be a member of one of four HMOs. Every member of the Clalit Health Services, the largest HMO in Israel, is registered to a single family physician, and have a medical record in his physician's clinic. Visits to the emergency department are regulated in the national health insurance law. A referral by a physician or by ambulance is free of charge, but this referral should be with a referral letter and not by a phone call to the ED or to the patient. A self referral may cost the patient a co-payment of up to 100 USD. We reviewed retrospectively all the charts of the ED visits for a period of one month, excluding the visits to the pediatric and the gynecologic-obstetrics EDs (see flow-chart 1). 5,898 visits documented that month, resulted in 4,256 discharges and 1,642 hospitalizations. There were 1,564 discharges from the general ED. Trauma, surgery and orthopedics accounted 2,209 discharges and the rest 483 were from other specialties (ophthalmology, ENT, dermatology etc.). Inclusion criteria were: visit to the general ED, age above 18 years, discharge to the community (not hospitalized) at that visit, living and getting medical care in a family medicine group practice in the Rehovot region. Visits due to accidents, trauma, surgery, orthopedics, ENT, ophthalmology and other specialities were excluded from the study. The 1,564 discharges from the general ED were reviewed and 359 were found to be eligible to this study. Two physicians reviewed independently each ED medical chart. Data extracted included: age and gender of the patient, attendance date and hour, self referral, or a referral by a physician and the final diagnosis in the discharge letter. In cases of referrals the content and format of the referral letter were assessed, including: hand writing quality and whether the referring physician referred the patient with a specific question (for example: rule out new onset angina pectoris, suspected pneumonia, please make a chest X ray etc.). Continuity of communication and care: The primary care files of ED visitors were retrieved and checked for the existence of the ED discharge letter and comments about the visit in the follow-up chart. If the discharge letter and / or any comment on the ED visit in the follow-up chart had been found the case was defined as "a case with good continuity of care". The cases in which the family physician was the referring physician we looked for documentation of the encounter prior to ED attendance. Visits to the ED were divided into "working hours" visits – when the visit took place during working hours of primary care clinics in the community (Sunday to Thursday from 08:00 to 20:00, and Friday 08:00 to 14:00), and "out of hours" visits when primary care clinics in the community are closed (from 20:00 to 08:00 weekdays, and weekends from 14:00 on Friday to 08:00 on the following Sunday). A referral letter was defined as "any document written by a medical authority in the community prior to the index ED visit", including referrals from family practitioners, other practitioners in the community and arrival by an ambulance. A recurrent visit to the ED was defined as a patient's visit to ED within less than two weeks from a previous visit with the same complaint, when in both cases the patient was not hospitalized. Diagnoses at discharge were coded for a specific diagnosis and for the system involved according to the ICPC coding system. Statistical analysis: Data was analyzed using distribution analysis and χ 2 tests to investigate the association between categorized variables. Student's t tests were used to analyze continuous variables. The analysis was performed using the SPSS package. Results During the study period there were 359 ED visits that were eligible to be included in the study (table 1 , flow chart 1). . 214 (59.6%) visited the ED during the "working hours" of primary care clinics, 28 (7.8% of all visits) were recurrent visits to the ED. Table 1 Data on 359 visitis to the Emergency Department All visits (359) Referral letter (196) Self referral (163) P Value* Gender Women - 192 (53.5%) 106 (54%) 86 (52.7%) NS Men – 167 (46.5%) 96 (46%) 77 (47.3%) Age (years, mean ± SD) 54.1 ± 18.7 55.1 ± 19.0 52.9 ± 18.4 NS Age distribution <45 127 (35.4%) 68 (34.7%) 59 (36.2%) NS 46–65 105 (29.3%) 53 (27%) 52 (31.9%) 66–75 63 (17.5%) 38 (19.4%) 25 (15.3%) >75 64 (17.8%) 37 (18.9%) 27 (16.6%) Visit time Working hours – 214 (59.6%) 143 (73%) 71 (43.6%) <0.0001 Out of hours – 145 (40.4%) 53 (27%) 92 (56.4%) * – p Value for comparison between patients with referral letter and self referrals Out of all ED visits only 196 (54.6%) patients had a referral letter, the rest were self-referrals. Referral letters were mainly from the family practitioner (147/196, 75%), 14 (7%) from other practitioners in the community, and 35 (18%) of referrals were by ambulance. The referral letters from the community were legible in 43.4% (70/161), 46.5% (75/161) were barely legible and 10% (16/161) illegible. In only 25/161 letters (15.7% of the referrals) a specific question was asked by the referring physician and in another 32 (20.2%) there was only a general question. The main diagnostic groups according to the ICPC were: respiratory (15.7%), digestive system (18.1%), musculo-skeletal (15.2%) and cardio-vascular (11%). In 9.6% of the cases the discharge letter did not contain a specific diagnosis and the diagnosis fell in the "general" category. The most common specific diagnoses were: chest pain (5.9%), abdominal pain (3.9%), other respiratory tract infections (3.7%), asthma (3.1%), back pain (2.8%), COPD exacerbation (2.8%) headache (2.5%), nephrolithiasis (2.5%), vertigo or dizziness (2.5%) and gastroenteritis (2.3%). The "out of hours" visitors tended to be younger (52.2 ± 17.5 vs. 55.4 ± 19.5, p = NS) (table 1 ). A third of the "working hours" visits (71/214) were self referrals as opposed to 63.5% (92/145) of "out of hours" visits (p < 0.0001). Table 2 compares the referring practitioners according to ED visiting hours. Table 2 Comparison between the source of referral, in 196 visits according to ED visit hours* The referring practitioner "Working hours"** "Out of hours" Total The Family practitioner 125 (87.5%) 22 (41.5%) 147 By ambulance 16 (11%) 19 (36%) 35 Other practitioner*** 2 (1.5%) 12 (22.5%) 14 Total 143 (100%) 53 (100%) 196 * – p < 0.0001 ** – "working hours" visits – when the visit was during the working hours of primary care clinics in the community *** – private practitioners, and out of hours commmunity emergency clinics In 147 cases the reffering physician was the family physician, documentation of the ED referral was found in 32% (47/147) of primary care files. The ED discharge letter was found in 50% (179/359) of the primary care files. A follow-up visit was documented in only 31% (111/359). Neither follow up visits nor discharge letters were found in 43% of the files (153/359). No associations between clinic characteristics (size, place) or family practitioner qualification and ED visit documentation was found. Discussion The Emergency Department (ED) acts as a link between community and hospital based medicine. In Israel a patient who needs non elective admission to a hospital unit must pass through the ED, either with a referral note from a medical practitioner, or as a self referral. Most ED visits, however, do not result in hospitalization, and many could be regarded as primary health care problems [ 1 , 2 , 9 ]. These patients are discharged directly from the ED to the community and further care of the family practitioner. A visit to the ED is generally not prompted by a benign complaint; The most common reasons for referral include, chest pain, asthma exacerbations and nephrolithiasis, subsequent follow up by the family practitioner can be vital. It was found that most children do not have outpatient follow-up after an ED asthma visit [ 8 ]. However, those patients that present for outpatient follow-up have an increased likelihood for repeat ED asthma visits, and this visit should be a key opportunity to prevent future ED asthma visits. The increasing role played by the ED in treating primary care problems has been discussed in a number of recent articles [ 9 - 11 ]. One aspect, which is important to the ED team, is the logistics and manpower needed to optimize the treatment of these non-urgent patients in ways that will not interfere with emergencies yet providing them adequate care. It is unclear whether the capability and quality of primary care services in the ED should be improved and compete with the community family physicians. This is true especially in Israel where there is a universal national health insurance and every patient can have a personal family practitioner. The continuity of comprehensive management is expected from the family practitioner, and is gaining importance nowadays [ 12 ]. To achieve this goal the communication between health care providers who treat the patient is mandatory. In the case of the ED visit, where we found many self referrals and referrals from other physicians, it becomes even more important. The modes of communication are the referral letter and the discharge letter. We have found that the referral letter can be improved both in style (printed instead of illegible hand writing) and content (the referring physician should define and clarify the reasons for referral and his expectations). These problems exist in discharge letters as well [ 13 ]. Documentation in the primary care file was poor, only one third of referrals were documented and less than 60% of discharges. This figure is between the 27%–77% that was found by others [ 6 , 7 ]. A possible bias is that some follow-up visits were to specialists. But in the case of discharge from the general medicine ED we presume that most patients were advised to return to their family physicians. It is well known that medical notes are poor in other areas, Miller et al [ 14 ] found documentation of only 15% of prescriptions given by family practitioners. They explained one of the causes for this discrepancy as the need of double writing (both the prescription itself and in the medical notes). By introducing carbon copy prescriptions, they achieved an 82% documentation rates in patients' files. Opila [ 15 ] found documentation in out patient medical records greatly improved after employing quality control and a feedback system. With the introduction of computerized medical files in primary care clinics in our region, the need of "double writing" will disappear, and this in turn should dramatically improve documentation rate of referrals and discharges to the ED; particularly if a computerized reminder system is used to encourage follow up of referrals by the family practitioner. Limitations Israeli health care system works in regard to ED use quite different from the US and other countries. Likewhise these results may not automatically be generalized to other health care systems. This study described the written communication between the emergency department and the primary care physician, which is the first and mandatory step in establishing continuation of care. This is only one of the four dimensions of continuity of care in family practice: chronological, geographical, interdisciplinary, and interpersonal [ 16 ]. Each of these dimensions may influence the quality of care and be evaluated and studied. Further study is needed to prove the link between documantion of ED visit and good contuniuity of care. Large scale prospective intervention studies are needed to prove that continuity of care between ED and the primary care physician improves outcome and saves money. Conclusion ED visits may have important implications for the patient and his family practitioner. The high rate of ED self referrals together with low documentation rates of ED visits in the primary care charts result in poor continuity of care of ED visitors. Competing interests None declared. Authors' contributions All authors read and approved the final manuscript. VS Conceived and designed the study, participated in the collection, analysis and interpretation of data and drafted the manuscript. KE Participated in the statistical analysis, interpretation of data and draft of the manuscript.OY participated in the design of the study, data collection and interpretetio hand draft of the manuscript. SN participated in the design of the study, interpretation of data and draft of the manuscript. All authors have read and approved the final manuscript. Figure 1 Emergency department (ED) visits that included in the study Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC514896.xml

535938

GOtcha: a new method for prediction of protein function assessed by the annotation of seven genomes

Background The function of a novel gene product is typically predicted by transitive assignment of annotation from similar sequences. We describe a novel method, GOtcha, for predicting gene product function by annotation with Gene Ontology (GO) terms. GOtcha predicts GO term associations with term-specific probability (P-score) measures of confidence. Term-specific probabilities are a novel feature of GOtcha and allow the identification of conflicts or uncertainty in annotation. Results The GOtcha method was applied to the recently sequenced genome for Plasmodium falciparum and six other genomes. GOtcha was compared quantitatively for retrieval of assigned GO terms against direct transitive assignment from the highest scoring annotated BLAST search hit (TOPBLAST). GOtcha exploits information deep into the 'twilight zone' of similarity search matches, making use of much information that is otherwise discarded by more simplistic approaches. At a P-score cutoff of 50%, GOtcha provided 60% better recovery of annotation terms and 20% higher selectivity than annotation with TOPBLAST at an E -value cutoff of 10 -4 . Conclusions The GOtcha method is a useful tool for genome annotators. It has identified both errors and omissions in the original Plasmodium falciparum annotation and is being adopted by many other genome sequencing projects.

Background It is now often possible to obtain the complete genome sequence of an organism in a few months, but without a directed approach, determining the function of potential gene products through biological experimentation is inefficient. Accordingly, methods for function prediction are required to direct experiments in function verification. In the context of this paper the term function is used to refer to all aspects of a gene product's behaviour. This includes the concepts described by the Gene Ontology classifications for Molecular Function, Biological Process and Cellular Component. It is explicitly stated in the text when a more specific interpretation of function is intended. A powerful tool in the annotation of novel genomes is the prediction of function by similarity to a sequence of known function. Such 'transitive function assignment' can work very well where there is a clear match to a homologue with a well established function. However, accurate functional assignment is difficult in cases where the match is less well defined, either due to lower sequence similarity or the presence of many candidates with differing functions. Gerlt and Babbitt [ 1 ] reviewed a number of examples where sequence similarity alone cannot provide full function specificity. The examples they discussed included classes of proteins where the function is similar but sequences are diverse, and classes where sequences are similar but function is diverse, indicating potential pitfalls for automated analyses. These examples are however quite extreme; sequence similarity can be used to infer function for a large proportion of genes with good results. Function annotation of sequences by tools such as PEDANT [ 2 ] and GeneQuiz [ 3 ] was dependent on free text annotations in the sequence databases and was complicated by the difficulty of mining and interpreting natural language. For example, a function may be described in one way in one sequence annotation, only to have the same function described in a different way in another sequence annotation. Such inconsistencies make computational determination of function equivalence difficult if not impossible. The use of restricted vocabularies and keywords has gone some way towards addressing this problem since it allows direct comparison of sequences with identical annotation schemes, at least to a match/no match level. Ouzounis and Karp [ 4 ] proposed the Transitive Annotation Based Score (TABS) to assess qualitatively the differences between annotations provided by different schemes. This scale relies on a human curator to determine manually the relationship between potentially conflicting terms, so is not readily applicable to the automated analysis of annotations. Keywords and restricted vocabularies do not solve the problem of conflicting assignments. Unless some computable form of relationship is present between terms, it is not possible to provide any automated form of conflict resolution between terms or to identify computationally where one term is a more specific descriptor than another. An ontology represented as a graph can provide a solution to this problem. Ontologies are restricted vocabularies, or sets of terms where each term is explicitly related to parent terms and child terms (and hence to sibling terms). The Gene Ontology (GO) [ 5 ] is a description of biology represented as a directed acyclic graph (DAG) where each node represents a clearly defined biological concept. Gene Ontology is continually being developed but contained approximately 14,000 nodes as at March 2003. The availability of the Gene Ontology has provided for the first time, a broadly accepted classification system for function assignment that can be analysed computationally. Previous work using other classification schemes, such as restricted vocabularies based on SwissProt keywords, suffered because of the lack of a distinct relationship between terms and/or due to typographical differences [ 6 , 7 ]. Since the establishment of GO, several authors have prepared tools that provide function assignment to Gene Ontology or a subset thereof. Jensen and co-workers [ 8 ] used neural networks to provide predictors for a small subset of 190 relatively non-specific GO terms. Schug et al. [ 9 ] used similarity to protein families defined as ProDom [ 10 ] or CDD [ 11 ] domains, by assigning the most specific common function represented in the set of proteins belonging to the family. This was a relatively conservative approach, taking similarity to clearly defined families annotated with relatively non-specific functions as a basis for transitive annotation. Xie and coworkers [ 12 ] have combined sequence similarity data with protein domain matches, cellular location prediction and literature mining data to improve transitive assignments. Their tools provided mappings to individual GO terms using a complex collection of probabalistic models and single linkage clustering. The method appears to be extremely powerful, taking input from a wide variety of sources, but it is difficult to assess the overall accuracy. Two tools based on BLAST searches have recently been presented in the literature. OntoBlast [ 13 ] provided a list of GO-terms prepared from gene-association links to similarity matches from BLAST searches. GO terms associated with the matching sequences are scored according to the E -value of the pairwise match. GOblet [ 14 ] also applies BLAST searches as the basis for assigning GO terms but does not give any estimates of validity beyond restricting matches to those below a user defined E-value threshold and counting the number of matching sequences. In this paper we present a novel method, GOtcha, that can be applied to any database search technique that returns scored matches. We have initially implemented this with BLAST searches and extend the analysis from the similarity match scores for a search in order to provide an empirical estimate of the confidence in each predicted function. We have applied this method to Malaria ( Plasmodium falciparum ) [ 15 ] and six other well annotated genomes and compared the results obtained by the GOtcha method to the results of annotation with the top informative BLAST match. The two methods have been assessed quantitatively with seven-fold cross validation by comparing the predictions obtained by GOtcha with those provided by the curators of the respective genome sequence consortia. The assessment of the global accuracy of a particular annotation method is extremely problematic in the absence of a computable annotation scheme. Gene Ontology provides such a computable scheme and we present here a quantitative measure for comparison of function annotations based on assignment to GO terms. This provides a metric for direct objective comparison of annotation methods that is independent of arbitrary cut off values. The new accuracy measure encompasses true positives, false positives and false negatives, so combining sensitivity and selectivity in one value. Results Two sets of annotation predictions were determined for each data set in the study. One was based on all available GO annotations and the other on a reduced set of GO annotations that excluded gene-associations with the evidence code IEA (Inferred by Electronic Annotation). IEA annotations are usually considered to be less reliable as they have not been assessed by a human curator. In contrast, ISS annotations (Inferred from Sequence Similarity) are annotations which, whilst being derived electronically, have been assessed by a human curator and can be considered sufficiently reliable. IEA annotations may however give a broader coverage than non-IEA annotations. On average, each dataset contained slightly more than 50% IEA annotations, though the vast majority of the sequences had some non-IEA annotation. The number of sequences for each dataset is listed in Table 1 along with a summary of the number of sequences annotated both with and without IEA annotations. Function assignment using all gene-associations Figure 1 illustrates the recovery of annotations by the two function assignment methods. In this and the following analyses the predicted term associations for all three ontologies are combined. The derivation of the P-score accuracy estimate (see Methods) normalises the data allowing combination of the three separate sets of results in one graph. The y -axis indicates the proportion of annotations provided by the genome project (given annotations) that were annotated to some degree by either GOtcha (Figure 1a ) or TOPBLAST (Figure 1b ). At a P-score of 50% GOtcha recovered 47% of the given annotations (35–59% s.d. 7.7%) whereas TOPBLAST with a cutoff of E = 10 -4 recovered 28% of annotations (20–38% s.d. 5.1%). This E -value cutoff is at the top end of the E -values between 10 -4 and 10 -20 typically used as a threshold for confident function assignment [ 16 - 20 ]. The proportion of annotations recovered by TOPBLAST was on average 60% (s.d. 4.1) of the proportion of annotations recovered by GOtcha, clearly indicating the presence of much useful function information throughout the BLAST search results, even at relatively high E -values. Figure 2 illustrates for each genome the total number of predicted GO term associations (GOtcha in Figure 2a , TOPBLAST in Figure 2b ) and the number of sequences annotated (GOtcha in Figure 2c , TOPBLAST in Figure 2d ) with respect to a scoring cut off for the annotation by each method. Figure 2e (GOtcha) and Figure 2f (TOPBLAST) illustrate the number of annotations per annotated sequence. Figures 2a,2c and 2e show the results for GOtcha with the x -axis representing the minimum P-score. A low P-score represents low confidence in the annotation. A high P-score represents high confidence in the annotation. Figures 2b,2d and 2f show the results for the top informative BLAST hit with the x -axis representing the maximum E -value. A low E -value represents high confidence in the annotation. A high E -value represents low confidence in the annotation. In Figure 2c the total number of sequences annotated by GOtcha with a P-score for the annotation above the value on the x -axis approaches the maximum relatively quickly when moving from high P-score to low P-score, typically coming very close to the total number of sequences annotated well before the P-score has dropped to 50%. This represents a broad coverage of sequence space, assigning annotation at a relatively nonspecific level to most sequences. In terms of the total number of annotations, these rise steadily as the P-score cut off drops. At very low P-scores (below 10%) the total number of annotations increases rapidly, indicating an increase in the spectrum of functions matched with only weak similarity. The number of annotations per sequence increases gradually as the P-score drops until a rapid rise at low P-scores (Figure 2e ). The rapid increase in number of sequences annotated is a reflection of high confidence in GO term associations at a general level of specificity. At lower P-score values more specific terms can be associated with sequences but the total number of sequences annotated has already approached the maximum. In comparison, the average number of associated GO terms per sequence by the genome projects varies from 14.5 to 19.8 (mean 16.6 s.d. 1.9). Figure 2d shows the number of sequences annotated using the top BLAST hit with a score below the E -value indicated by the x -axis. In this case the number of sequences annotated increases more slowly with E -value (Figure 2d ) but the number of annotations per sequence remains relatively constant, rising only modestly as E -value rises (Figure 2f ). This arises from the key difference between GOtcha and TOPBLAST. In GOtcha a term-specific probability is calculated which allows some functions for a given sequence to be assigned more confidently than others. For a given sequence only the more general terms will appear in the prediction list above the P-value threshold. With TOPBLAST the whole set of annotations from the top matching hit is assigned with a common score, irrespective of the term's specificity. Thus either all or no terms for that sequence will appear below the E -value threshold. The specificities of function prediction for both GOtcha and TOPBLAST are illustrated in Figure 3 . Figure 3a shows the proportion of predictions by GOtcha that are correct with a P-score above the cutoff on the x -axis. At a P-score cutoff of 50%, the selectivity of GOtcha is 61.4% (54–68% s.d. 4.9). Figure 3b shows the proportion of predictions by TOPBLAST that are correct with an E -value below the cutoff on the x -axis. At an E -value of 10 -4 TOPBLAST shows a selectivity of 53.4% (43–60% s.d. 5.7). Accordingly, GOtcha outperforms TOPBLAST with improved coverage and better selectivity for each genome examined. Both the GOtcha and the TOPBLAST analyses include gene associations that are children of obsolete (GO:0008369) and the three 'unknowns' (cellular_component_unknown, GO:0008372; molecular_function_unknown, GO:0005554; biological_process_unknown, GO:0000004). The obsolete terms comprise a very small proportion (1.5% mean 0 – 3.1% s.d. 1.1) of the total number of annotations (shown in Table 1 ) and would not be expected to have any significant effect on the results. The three 'unknowns' however are considered to be valid function descriptions. They indicate a clearly observed similarity to a sequence with a function that has not been determined more specifically. Function annotation excluding IEA annotations Function assignment was repeated using the same BLAST search results but excluding the IEA coded gene-associations. Figure 4 illustrates the recall rate for function assignments. Recovery was lower in all but one genome compared to when IEA terms were included. GOtcha retrieved 39% (30–54 s.d. 7.3) of annotations with a P-score above 50%. This is 83% (54–100 s.d. 14) of the proportion of annotations retrieved by GOtcha when IEA based term associations are included. TOPBLAST retrieved 18% (9–25 s.d. 5.3) of annotations with an E -value below 10 -4 . This is 31–81% of the proportion of annotations retrieved when IEA based term associations are included. TOPBLAST only recovers 47% (23–63 s.d. 11) of the number of annotations recovered by GOtcha. The number of annotations per sequence was reduced by comparison to the data shown in Figure 2 though the trends were very similar (Data not shown). The difference between the analysis with and without IEA terms is consistent with the relative numbers of IEA and non-IEA annotations provided by the genome projects as there are only 9.7% (3.8–14.7 s.d. 4.0) GO term associations per sequence, 62% (23–100% s.d. 30) of the number of GO term associations per sequence when IEA terms are included. Figure 5 illustratess the selectivity for the analyses with IEA terms excluded. Figure 5a shows the proportion of assosciations correctly predicted by GOtcha with a P-score above the cutoff on the x -axis. Figure 5b shows the proportion of assosciations correctly predicted by TOPBLAST with an E -value below the cutoff on the x -axis. GOtcha with a P-value cutoff of 50% shows a selectivity of 60% (35–79% s.d. 14). TOPBLAST with an E -value of 10 -4 shows a selectivity of 49% (25–59% s.d. 11). In all cases except that of Arabidopsis GOtcha shows a clear improvement over TOPBLAST with a mean improvement in selectivity of 1.2 fold (0.85 – 1.4 s.d. 0.17). One issue with excluding IEA annotations is that the coverage of functions in the genome is lowered. This inevitably will lead to a higher number of positives that have incorrectly been assigned as false as a result of the incomplete sequence annotations. Despite excluding terms for which there is no annotation to the ontology under examination, the results are skewed by assigning a proportion of true positives as false positives. This indicates that the method is performing more poorly than is in fact the case. We have examined the nature of the false positives in more detail below. A metric for quantitative assessment of function annotation Comparing function assignment methods is difficult. Typically the standard against which they are assessed is an incompletely annotated dataset. Both a lack of experminental data confirming potential functions and a lack of knowledge about potential functions can lead to the standard data being less perfectly annotated that would be desired. It is not realistically possible in an automated analysis to cope with unrecorded true positives that are registered in the analysis as false positives. It is therefore the case that any analysis of accuracy can only give an estimate of minimum accuracy. Accuracy can also be difficult to compare between two methods that annotate to different subsets of GO. One method may only annotate to relatively general terms, allowing for a better claimed specificity than a method that attempts to annotate to a more specific level. GOtcha predicts at all levels of the GO hierarchy. It assigns a probability to every combination of GO term – sequence association and should be compared to other function assignment algorithms using a global metric, one which can account for over-specificity and under-specificity in a set of predictions as well as incorrect assignment. Ouzounis and Karp [ 4 ] described the TABS system for qualitative assignment of function annotation to eight categories. The TABS categories are reproduced in Table 2 . When applied to annotation using a DAG such as GO the number of potential categories is reduced from the eight described in TABS to three. TABS was developed to compare annotations where the terms used are not implicitly related through a computable structure such as a DAG. As we are using a DAG where ancestor nodes are implicitly associated with the gene through direct association of a child node, the prediction for a particular sequence becomes a set of GO terms (the nodeset) comprising all nodes that match the prediction rather than just the most specific terms. The accuracy of a prediction can then be assessed by observing the presence of nodes in both the node sets for annotations and for the predictions rather than assigning qualitative values. The more distant a given prediction node set is from the annotation node set, the smaller a proportion of nodes (GO terms) they will have in common. The effect of a quantitative approach on the TABS categories is as follows: TABS category 0 is unchanged. This is an exact match and is represented by the presence of the term in the node sets for both original annotation and current prediction. TABS category 1 is no longer relevant. A controlled vocabulary is being used so there is no scope for typographical errors of the type described by Ouzounis and Karp or by Tsoka [ 21 ] or Iliopoulos [ 22 ]. TABS category 2 is also irrelevant. GO has no undefined terms (though a small proportion of terms lack complete descriptions) and all annotation sources are attributed using evidence codes and references. TABS category 4 is an extreme case of category 3. Both these categories are represented by the existence of a function annotation in the original annotation node set but not in the predicted node set. Likewise TABS category 7 is an extreme case of TABS category 6. In many cases a false positive is represented as an underprediction in the true branch of the GO DAG and an overprediction in a false branch. TABS category 5 describes the mechanism of occurrence of an error rather than the error itself and is not relevant to this analysis. In this analysis we reduce the eight TABS categories describing the accuracy of a function prediction for an individual sequence to three categorise that describe each node in the nodeset comprising a function prediction for an individual sequence. These categories correspond to false positive, false negative and true positive nodes. A particular sequence annotation node set could potentially contain nodes from all three categories. Quantitation of the analysis Given two sets A and B corresponding to a given annotation set and a predicted set (each node in the set comprising a sequence – GO term association) we are interested in the true matches (intersection of A and B, n ∈ A ∩ B ), false positives (term associations in B but not in A, n ∈ B , n ∉ A ) and the false negatives (term associations in A but not in B, n ∈ A , n ∉ B ). The aim of any prediction method is to maximise the number of matches (true positives) whilst minimising the errors (false positives and false negatives). The number of true negatives does not need to be considered as this number is very large and essentially constant over the analysis. We can use the following relation as an error quotient to assess prediction methods. where REQ is the Relative Error Quotient, n is the total false negatives, p is the total false positives, w is a weighting factor and t is the total true positives. A low REQ represents a low proportion of errors. A higher REQ indicates a higher proportion of errors. Such a measure is dependent upon the population of the node set which in turn is dependent upon the cut off used for selecting predictions in the node set. Figure 6a shows the change in REQ with respect to P-score cutoff for the GOtcha analysis and Figure 6b the REQ with respect to E -value cutoff for the TOPBLAST assignments. A weighting factor of 1 was used in both cases, thus giving equal weight to both false positives and false negatives. In this figure the minima indicate optimum cutoffs for maximising the similarity between annotation and prediction nodesets. The GOtcha results (Figure 6a ) indicate broad minima, suggesting that small differences in cut off selection may have only a slight effect on the accuracy of the results. The minima for BLAST are difficult to see as they are skewed to very high E -values as a result of a large proportion of false negatives. This indicates that the TOPBLAST search is rejecting important information present in matches with E -values approaching 1, much higher than those normally used for genome annotation. The REQ metric therefore appears to perform quite robustly. This metric assigns identical weight to each GO term association. More complex weighted measures of semantic similarity have been proposed by Lord and coworkers for searching databases based on annotation [ 23 ] but these are difficult to apply to the present problem in a manner that uses a non-arbitrary weighting. In the absence of IEA annotations the spread of the REQ curves changes dramatically as shown in Figure 7 . Figure 7a illustrates the REQ for GOtcha with the differences between the genomes far less marked than for Figure 6a . In contrast, the REQ for TOPBLAST is shown in Figure 7b and shows much higher and more diverse REQ than when IEA terms are included (Figure 6b ). Minimum REQ (i.e. maximum accuracy) has been determined for both GOtcha and top BLAST hit annotation sets, both with and without the use of automated annotations (IEA evidence code) for transitive function assignment (Table 3 ). When automated annotations (IEA codes) are included in the analysis, there is no significant difference between the minimum REQ obtained using GOtcha or that from TOPBLAST. The minimum REQ for TOPBLAST is obtained at very high E -values, 0.011–0.71 when IEA terms are included (Figure 6 ) and 0.12–0.71 when IEA terms are excluded (Figure 7 ). When IEA annotations are excluded from the analysis GOtcha performs significantly better than TOPBLAST (p ≤ 0.016 using the Wilcoxon signed rank test). GOtcha excluding IEA terms performs better (though this small number of genomes does not give a statistically significant result) than when IEA annotations are included (mean change: 15% reduction in REQ s.d. 11%, p = 0.2 using the Wilcoxon signed rank test). It may be that the annotation set used as the reference in comparing these results was incomplete. This would result in some true positives being incorrectly assigned as false positives with a corresponding increase in REQ . However, this would apply similarly to GOtcha and to the top BLAST hit analysis. Assessment of incorrectly assigned false positives Samples of the false positive function predictions by GOtcha with the highest P-scores from three P. falciparum chromosomes (representing the three genome centres in the Malaria Genome Sequencing Consortium) were assessed by hand to give an indication of the completeness of the curated annotations. Results for selected sequences in this set are shown in additional file 1 . Twenty sequences were examined: ten from chromosome 12, and five taken from each of chromosomes 2 and 3. Chromosome 3 was the first to be sequenced and is the most carefully annotated of the chromosomes. In each case the sequences selected were those with the highest scoring false positive function assignments. Representative results from the analysis of GOtcha annotation with and without IEA terms are available as supplementary material. The proportion of correct annotations generally performed better than the P-score would suggest. Taking a P-score of > 50% as a cutoff, most GOtcha predictions agreed with the function assigned by the curator. The false positives fell into several categories: Differences in curator judgement In some examples, genes that were annotated as encoding hypothetical proteins could be re-annotated based on GOtcha predictions. GO terms had not been assigned during the manual curation phase of the P. falciparum genome project if no function had been identified during the first-pass automatic annotation. However, the addition of GO terms to sequences by GOtcha prompted the original annotation to be re-evaluated. For example, PFL1875w shows a hit to the Pfam K+ tetramerisation domain (Pfam:0224, E = 10 -9 ) supports the GOtcha annotation although it is at a level that genome annotators may feel is marginal. In PFL1780w, stronger supporting evidence (a hit with E = 10 -12 to Pfam:04140, isoprenylcysteine carboxyl methyl transferase domain) indicates again that GOtcha can suggest GO annotations that have been previously overlooked. In several examples, GOtcha predicted either additional functions or more specific GO terms to describe previously annotated functions. PFC0495w encodes a putative aspartyl protease. When all evidence codes were included, a molecular function of pepsin A activity is predicted. This protein matches pepsin A domains defined by the InterPro entry IPR001461 ('Peptidase_A1 pepsin A'), thus the term from GOtcha is likely to be correct. Human error PFL2465 encodes a thymidylate kinase, which was correctly annotated by GOtcha as being involved in dTTP biosynthesis. GOtcha also indicates 'dTDP biosynthesis' as a suitable GO process term. Thymidylate kinase catalyses the synthesis of dTDP, a necessary step in dTTP biosynthesis. However, the human annotator missed the fact that dTDP biosynthesis is not a 'part of' dTTP biosynthesis within the ontology structure and in such cases, terms describing both processes must be employed. Sometimes, GOtcha highlighted erroneous omissions in the GO annotation of the P. falciparum genome, many of which have arisen from retrospective corrections and amendments to gene models. For instance, GOtcha provides detailed annotation for a putative ATPase synthase F1 alpha subunit (PFB0795w) almost completely lacking useful GO terms. GOtcha also suggested GO terms relating to translation elongation for PFL1710c. A highly significant hit to Pfam:00009 (Elongation factor Tu GTP binding domain, E = 10 -46 ) indicates that this GOtcha prediction may well be more accurate then the original genome annotation. IEA vs non IEA Annotations performed with IEA terms appeared to be more specific than those where IEA terms were excluded. In many cases, such as PFC0495w, the difference was quite pronounced. Here the protein was implicated in 'proteolysis and peptidolysis' when all annotations were included but filtering out IEA annotations resulted in the more general, and less useful, description of 'metabolism'. Real false positives Out of the 20 genes inspected, PFL1825w was the only example where GO terms were incorrectly suggested for the biological process, molecular function and cellular component aspects of GO. In other cases, mis-annotations often had low I scores (predictions made with P-scores > 50% but very low associated I-scores ≪ 0.1) or were due to terms taken from slightly too far down a branch in the ontology structure. For example 'ATP-binding and phosphorylation-dependent chloride channel' was predicted for PFB0795w, an ATP synthase. The cellular component of gene products are hard to annotate – often BLAST is insufficient to recognise the targeting information encoded in signal and transit peptides and specific signal sequence detection methods such as PSORT II [ 24 ] must be used instead. GOtcha consequently made incorrect predictions of subcellular localisation in some cases. For instance PFL1710c is annotated as having mitochondrial and apicoplast localisation based on separate lines of evidence [ 15 ] but GOtcha predicted cytoplasmic localisation with a P-score of 52%. It is hard to measure what proportion of the calculated false positives does in fact represent serious mis-annotation. Although the hand analysis may provide representative examples, it is too small to be of statistical significance. Genuine false positives (with high P- and I-scores) were fewer than would be expected from the P-score. Despite the small sample size, these results show that GOtcha performs well as a guide to the manual assignment of GO terms. Not only can it provide suggestions for more granular annotation but it can highlight terms that would otherwise be missed by a human annotator. Discussion Data interdependency and annotation accuracy One of the major problems facing assessment of function assignment is the separation of annotation and test datasets. In this analysis we have tackled this issue by taking individual genome datasets as the test sets and using other genome datasets for the annotation source from which to transitively assign function. The scoring mechanism used for estimating accuracy values is independent of both test and annotation datasets, since it makes use of sequences that are found in neither. Whilst the sequences are independent, the annotations associated with these sequences may not be. Many of the computationally assigned annotations are derived from analyses involving the 'independent' datasets and can therefore not be regarded as entirely independent. IEA annotations are primarily obtained from sequence similarity searches. As a consequence it is not surprising that the results obtained for both GOtcha and TOPBLAST when IEA annotations are included are so similar. Interestingly, when IEA based annotations are excluded from the TOPBLAST analysis, the REQ goes up. This may well indicate a degree of inaccuracy in the IEA based annotations, or incomplete coverage by the human curated annotations. GOtcha, however, makes a significantly better use of the BLAST search result in the quality and coverage of the annotation. False positive/false negative balance in the relative error quotient The REQ analyses performed weighted under prediction errors (false negatives) equally to over prediction errors (false positives). In order to examine the effect of the weighting on REQ , the GOtcha predictions for the human genome were compared to the genome consortium annotations with weights ranging from 0.5 to 15 (Figure 8 ). As expected, an increased emphasis on false positives shifts the minimum REQ towards a higher P-score cutoff. Weighting can be adjusted depending on the aims of the study in question. The minimum REQ should give the best tradeoff between accuracy and coverage and can be used to estimate an optimum P-score cutoff for transitive assignment of function. Investigations that emphasise accuracy over coverage may increase the weight to reduce false positives. Investigations with less concern for accuracy but a greater emphasis on coverage will use a lower weight for minimal REQ determination to increase coverage. The metric presented here is an objective measure of method performance but has some drawbacks. Using the REQ as described in this paper, each term in the nodeset is weighted equally. This may not be the most appropriate measure. The granularity of terms in Gene Ontology is not constant across the ontologies, nor is it readily quantifiable. This may lead to bias in the metric, where differences in the presence or absence of closely related terms is weighted equally to presence or absence of more distantly related terms even though they have the same graph path distance between them. There is also the issue of prevalence. Some terms occur in almost every nodeset, others are less prevalent. The most appropriate form for a quantitative metric will need to be examined in future work. Transitive function assignment is limited by the sensitivity of the underlying search method and the scope of the dataset being searched. The GOtcha method of preparing a weighted composite view of the functions from a complete set of search results provides a significant improvement in the annotation of sequences when compared to a method that selects the most significant annotated hit. GOtcha also provides a confidence measure for the putative function assignments, allowing for the determination of an appropriate level of specificity for the annotation set. Hennig and co-workers examined the ability of BLAST analysis to transitively assign function from distant taxa, concluding that for the majority of cases, GO-based annotation would give a good result [ 14 ]. In this study we have performed seven-fold cross validation with seven distinct genomes across the taxonomic range. It is intended to improve the performance of this method by including further genomes and updating the annotations on those already used. Conclusions The GOtcha method has several significant advantages over the transitive assignment of function by TOPBLAST. Firstly each function assignment has a directly understandable accuracy estimate that can be interpreted without any knowledge of the prediction methodology. This accuracy estimate is function-specific, unlike general rules of thumb that are applied to interpretation of BLAST search results. Secondly, the GOtcha method provides much greater coverage than a top annotated match approach, annotating more sequences with reasonable confidence. In many cases it provides annotations for sequences that otherwise would have no annotations. Finally, it provides term specific annotation accuracy estimates. This is a significant advantage over TOPBLAST where every term in the set predicted for an individual sequence has the same value and a biologist interpreting the results is given little indication of which terms can reasonably be accepted. In contrast, GOtcha provides individual P-scores for each term. This allows a rapid visual examination of the prediction as a graph or a list, indicating appropriate points at which experimental verification may best be directed. In order to assess the accuracy of annotations to tree-like ontologies we have developed an objective flexible scoring metric that provides a global analysis, including assessment of both false positives and false negatives. This metric also provides a means for comparison of methods that is not dependent on the selection of any particular parameter threshold or cutoff in the scoring method used. The underlying mapping methodology applied in GOtcha can readily incorporate other search methods that provide a more sensitive similarity search. Combining search methods should also provide a better coverage of sequence space occupied by distant homologues [ 25 ], and such potential improvements are the subject of further work. Methods Data sources All data were obtained in the same week (week 9, 2003) to provide a consistent time point at which to perform the analysis. Sequence data Malaria ( Plasmodium falciparum ) sequence data for the recently determined genomic sequence [ 15 ] were obtained from the malaria consortium. The whole genome annotated peptide set was dated 3 October 2002 and comprised 5334 peptides varying in length from 17 to 10589 amino acid residues. Fruit fly ( Drosophila melanogaster ) data were obtained from Flybase [ 26 ] as release 3.1 of the annotated full genome transcript set. This set contained 18484 transcript sequences corresponding to 13656 genes. A non-redundant set was created for subsequent analysis by selecting the longest transcript to represent each gene. Transcript lengths varied from 15 to 69162 nucleotides (5 to 23054 amino acid residues). Yeast ( Sacchyromyces cerevisiae ) data were obtained from the Sacchyromyces Genome Database [ 27 ]. The set of translated open reading frames for the whole genome was used and comprised 6356 peptides varying in length from 25 to 4911 amino acid residues. Cholera Vibrio cholerae data were obtained from The Institute for Genomic Research [ 28 ]. The dataset contained 3836 sequences varying in length from 26 to 4588 amino acid residues. Human ( Homo sapiens ) data were obtained from Swiss-Prot using the conceptual complete human proteome from the Swiss-Prot/EnsEMBL collaboration dated 6 March 2003 [ 29 ]. The dataset contained 39080 proteins with lengths varying from 3 to 34350 amino acid residues. Worm ( Caenorhabditis elegans ) data were obtained from Wormbase release 97 [ 30 ]. The dataset contained 30753 peptides varying in length from 4 to 13100 amino acid residues. Thale cress ( Arabidopsis thaliana ) data were obtained from The Arabidopsis Information Resource [ 31 ]. The complete genome peptide set dated 31 July 2002 was used. The dataset contained 27288 sequences varying in length from 20 to 4707 amino acid residues. Each data set was formatted for BLAST searching with the formatdb program from the BLAST2 suite [ 32 , 33 ]. The URI for each genome dataset are listed in Table 4 Gene association and Gene Ontology data Data for the Gene Ontology and gene associations for all proteome sets except Arabidopsis were downloaded from the Gene Ontology CVS repository in week 9, March 2003, parsed and loaded into a relational database. Arabidopsis data were obtained from The Arabidopsis Information Resource using gene association data dated 13 February 2003. A flat file database containing the Gene Ontology and gene association data was developed and indexed to allow rapid retrieval of individual entries by custom written Perl modules. The number of annotated sequences in each data set is shown in Table 1 . Software The BLAST2 programs were obtained from NCBI. Analyses were performed on a cluster of 50 HP Netserver L1000 dual processor machines configured with two 1.4 GHz Pentium III processors, 70 Gb hard disk, 2 Gb RAM and running a customised Linux operating system. Job scheduling was performed with Grid Engine (Sun Microsystems). Results were stored in a relational database (PostgreSQL version 7.3) or as flat files where appropriate. BLAST result parsing was performed with the BioPerl toolkit (release 0.7) [ 34 ]. Sequence manipulation was performed with EMBOSS [ 35 ]. All processing scripts were written in Perl. A set of Perl modules were developed for accessing and manipulation of data entries. Methods GOtcha method overview The GOtcha method is illustrated by a cartoon in Figure 9 . We have implemented this method by searching against a cohort of seven well defined and annotated genomes. To predict the association of GO terms with a specific individual gene product a BLAST search is run against each genome data set using the appropriate program (blastx/tblastn when D. melanogaster was the query/subject set, blastp otherwise). Default parameters were used (Maximum expectancy score 10; maximum list sizes 250 and 500 hits). Each sequence database search produces a ranked set of sequences similar to the query sequence. The search result for each genome database search is parsed and a list of pairwise matches between the query sequence and the subject database sequences obtained. For each similarity match between the query sequence and a database sequence, a set of GO terms corresponding to the gene-associations for the database sequence is retrieved from the appropriate gene-association dataset. The set of GO terms and all ancestral terms (the nodeset) are assigned a score R = max { -log 10 ( E ), 0 } where E is the expectancy score for that pairwise match. In this way the whole subtree to the root node is assigned the R-score. The GOtcha method allows mappings obtained from many sequence matches to be combined. For each node (which corresponds to an individual GO term, either directly associated or the ancestor of an associated GO term), R-scores for all pairwise matches which contain annotation to that node are summed and normalised to the total R-score for the root node of that ontology (Cellular Component, GO:0005575; Molecular Function, GO:0003674; or Biological Process, GO:0008150). This normalisation gives an internal relative score (the I-score), producing a weighted composite subgraph of the GO. This normalisation effectively removes bias in the E-value due to database size or search program used. A confidence measure is calculated as loge of the root node score (the C-score). Accordingly, this provides two measures for an individual predicted gene-association; A score relative to the other predicted gene-associations in the node set (the I-score) and a score for the function prediction as a whole (the C-score). Each genome was searched individually and I-score and C-score for each GO term association were averaged across all genome searches that provide at least one annotated pairwise match. Averaging across genomes in this way provides some correction for individual genes with exceptionally high copy numbers in certain genomes. In this paper the term 'function prediction' relating to an individual sequence refers to a prediction of a set of GO term – sequence associations (also referred to as a node set). Averaging of the individual search results avoids the over-representation of large genomes in the final annotation set and allows the final result to be weighted towards a particular taxonomic grouping should that be desired. Each gene association represents a function assignment of a gene product with a GO term and is annotated with an evidence code providing an indication of the reliability of a particular annotation. The GOtcha method allows specific classes of annotation, such as those derived exclusively from computational analyses, to be excluded from the analysis if required. Background accuracy estimates for individual GO terms – P-score table construction Although higher C-score and I-score values correspond to greater confidence in the transitive assignment of function than lower C-score or I-score values, it is not immediately apparent how these values should be interpreted. Examination of preliminary results indicated that there was considerable variation between GO terms in the confidence that can be placed in a prediction with a given I-score and C-score (data not shown). Accordingly we have created an empirically based estimate of accuracy (the P-score, expressed as a percentage) that can be used to indicate confidence in the prediction of association between a GO term and a gene product. A background set of 518226 annotated sequences from the SwissProt gene associations were included in the accuracy estimate after excluding taxa corresponding to the search databases and their subspecies. All background sequences were subject to a search against all 7 species specific datasets and a set of function predictions obtained as described above. A scoring table for each GO term was prepared by segregating all predictions for that GO term on I-score and C-score. I-scores were divided into ten rows by dividing the range (0 – 1) evenly. C-scores were divided into columns by unit ranges (0–1, 1–2, 2–3 and so on). This gave rise to approximately one hundred cells for each GO term table. Each prediction was assigned to a cell based upon its I-score and C-score. The overall accuracy of each cell was determined by comparison of the predicted associations in that cell to the annotations provided by the GO Annotation project (GOA) and calculated as the proportion of true positives to the sum of true and false positives. The table for a specific GO term was then used to deliver the P-score based on any given I-score and C-score pair for a predicted association between that GO term and the query sequence. A similar set of tables was constructed from background analyses from which terms with IEA associations were excluded. For GO terms where there are few datapoints with which to estimate accuracy reliably, accuracy estimation falls back to a scoring table that combines results over all GO terms from that ontology with the same number of ancestors. Function assignment by top informative BLAST hit The same BLAST searches used for function assignment with the GOtcha method were analysed. Function assignments for the nodeset corresponding to the top annotated BLAST match (TOPBLAST) for each genomic dataset were transferred to the query sequence with a score corresponding to the E -value for that hit. List of abbreviations DAG, Directed Acyclic Graph. URI, Uniform Resource Identifier. BLAST, Basic Local Alignment Search Tool. TOPBLAST, Top annotated BLAST match. Perl, Practical Extraction and Report Language. GO, Gene Ontology. TABS, Transitive Annotation Based Score. NCBI, National Centre for Biological Information. s.d., Standard deviation. Authors' contributions The GOtcha method was devised and implemented by DMAM who also prepared the manuscript. MB performed the manual assessment of false positives and provided feedback on the presentation of results. GJB provided essential guidance for the performance assessment and revision of the manuscript. Supplementary Material Additional File 1 The supplementary data contains representative examples from the manual assessment of false positives. It is portrayed in tabular format and indicates the benchmark annotation, the highest scoring predicted incorrect annotation by GOtcha and the lowest scoring predicted annotation by GOtcha. Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC535938.xml

521193

Cheyne-Stokes respiration in patients hospitalised for heart failure

Background Previous studies showing a strong relationship between Cheyne-Stokes respiration and the severity of left ventricular systolic dysfunction have usually been done in selected patient populations with lower age and a higher proportion of males than the "typical" in-hospital patient with heart failure. The purpose of the present study was test the strength of this relationship in unselected patients admitted to hospital due to decompensated chronic heart failure. Methods We evaluated 191 patients (32% women), mean age 73 years, ready for discharge from the heart failure unit in the University Hospital of Malmö, Sweden. The patients underwent echocardiography for determination of left ventricular ejection fraction and left ventricular inner diastolic diameter. A respiratory investigation during sleep was performed the last night before discharge. Results We found that 66% of the patients had Cheyne-Stokes respiration more than 10% of the total recording time. Only 7 (3.6%) of the patients had predominantly obstructive apnoeas. There was a significant but very weak relationship between left ventricular ejection fraction and left ventricular inner diastolic diameter on one hand and Cheyne-Stokes respiration on the other. Age was a stronger determinant of Cheyne-Stokes respiration than any of the cardiac or other clinical variables. Conclusion Although presence of Cheyne-Stokes respiration indicates left ventricular dysfunction, its severity seems only weakly related to the severity of heart failure. Age was found to be a stronger determinant, which may reflect the underlying age-dependency found also in healthy subjects. Due to age restrictions or other selection criteria, the importance of age may have been underestimated in many previous studies on factors associated with Cheyne-Stokes respiration.

Background Cheyne-Stokes respiration (CSR) during sleep, is common in patients with heart failure [ 1 , 2 ]. Cheyne-Stokes respiration during sleep has been claimed to be an independent risk factor for death [ 3 , 4 ], speculatively through increased neurohumoral stress on the heart [ 5 ]. Results from other studies have, however, been contradictory [ 6 ]. It has also been claimed that sleep disturbance from Cheyne-Stokes respiration may cause daytime sleepiness [ 7 ]. Most previous studies have included selected patients with impaired left ventricular systolic function. A large proportion of heart failure patients are elderly and have relatively preserved left ventricular systolic function. However, elderly heart failure patients have often been excluded from sleep studies. Such studies are not representative of the everyday clinical spectrum of heart failure patients, including the increasing number of elderly patients usually seen in general medical wards. The proportion of women, with predominantly diastolic heart failure, may also be greater in this group. The aims of the present study were to test the strength of the relationship between left ventricular dysfunction and nocturnal Cheyne-Stokes respiration in unselected patients admitted to hospital due to decompensated chronic heart failure and to determine the correlation between Cheyne-Stokes respiration and other clinical variables to provide source material for subsequent analyses of quality of life and survival. Methods Patients All patients admitted to the Heart Failure Unit at the Department of Cardiology, Malmö University Hospital from January 1996 to november 1999 primarily due to decompensated chronic heart failure were eligible for inclusion. Malmö University Hospital serves as the main hospital for the whole population of Malmö (population 250,000) and is the only hospital in the city. Patients were excluded only if unable to comply to the study protocol due to some other condition, to complete the study questionnaire or to provide informed consent. Heart failure was diagnosed according to European Society of Cardiology guidelines for the diagnosis of heart failure [ 8 ]. All patients had been stabilised following treatment for heart failure and were studied on the day prior to planned discharge. For practical reasons not all patients discharged could be included into the study. Patients were included if they were to be discharged on a weekday and only when a nurse trained in the use of the registration apparatus was available. Furthermore, we could not register more than one patient per night. Unfortunately, the precise number of patients discharged alive from this unit during this particular period was not possible to obtain, but a crude estimate is that we have investigated between 25 and 50% of the available patients. All patients provided informed consent to participate in the study, which was performed in accordance with the principles of the Declaration of Helsinki and approved by the Medical Ethics Committee at Lund University. Clinical evaluation Ischaemic heart disease was diagnosed based on findings from previous coronary angiography, documented myocardial infarction or typical signs of ischaemia at exercise testing. Hypertension (according to the local guidelines at the time of the study) was diagnosed if blood pressure was >150/95 mmHg or the patient was receiving drug therapy for hypertension. Diabetes was diagnosed if fasting blood glucose levels were >7 mmol/l or the patient was treated with oral anti-glycaemic medication or insulin. The diagnosis of other concomitant diseases was based on patient history and/or patient records. Echocardiography Echocardiographic examinations were performed using a Hewlett-Packard Sonos 2000 (Andover, Mass, USA). Parasternal and apical views were obtained with the patient in a left lateral recumbent position. Measurements were acquired during silent respiration or end-expiratory apnoea. Left ventricular systolic function was assessed by determination of the mean left atrioventricular plane displacement (AVPD), global qualitative assessment and/or single plane ellipse (modified Simpson's rule) [ 9 - 11 ]. Respiration during sleep We recorded oronasal airflow by thermocouples, electrocardiogram, chest wall movement by electrical impedance, and finger pulse oximetry using the EdenTrace II Plus Multirecording System (EdenTech Corp, Eden Prairie, MN, USA) [ 12 , 13 ]. The recordings started when the patients went to bed and were discontinued the following morning when the patients woke up. No attempt was made to define the amount of sleep. The recordings were printed out and scored manually. Hypnotics were allowed, and were taken by 48% of the patients at the time of the study. Patients were considered to be habitual snorers if they answered "often" or "always" to the question "Do you snore loudly and disturbingly?" Scoring and analysis of breathing patterns The records were scored manually for Cheyne-Stokes respiration (gradual waxing and waning of respiration followed by a central apnoea or hypopnoea) [ 14 , 15 ] and for obstructive sleep apnoea. Patients with purely or predominantly obstructive sleep apnoea (n = 7) were excluded from the analyses of CSR. Patients with occasional obstructive apnoeas occuring during extended periods of CSR (n = 28) were included in the analyses of CSR. The total time spent in Cheyne-Stokes respiration was divided by the total recording time to compute the percentage of time in bed spent in Cheyne-Stokes respiration (CSR%). Since CSR% was not normally distributed, we used Spearmans rank correlation test to relate CSR% to the continuous variables age, BMI, LVEF and LVIDD. For analysis of CSR% with respect to the categorical variables gender, cerebrovascular disease, ischaemic heart disease, NYHA class, atrial fibrillation and habitual snoring, we used the Mann-Whitney U-test. There are no data in the literature to allow categorisation of patients as normal and abnormal according to any specific level of CSR%. In table 2 , the data concerning CSR% are, however, categorised; this is for demonstrational purpose only. Table 2 Cheyne-Stokes respiration related to physiologic variables All CSR incl mixed apnoeas CSR% CSR% CSR% Significance <10 10–50 >50 n = 58 n = 66 n = 60 Age (years) 68.8 ± 11.7 72.1 ± 8.5 75,9 ± 8.6 p < 0.01 R=0.24 Body Mass Index (kg/m 2 ) 26.5 ± 5.5 26.1 ± 4.3 24.6 ± 4.1 NS LVEF (%) 36.1 ± 12.1 38.8 ± 12.6 32.4 ± 11.5 P < 0.05 R=-0.17 LVIDD (mm) 55.1 ± 10.0 57.1 ± 9.4 58.0 ± 8.2 P < 0.01 R = 0.20 Ischaemic heart disease (percentage of patients) 51 59 72 p < 0.05 NYHA class 3–4 (percentage of patients) 60 39 55 NS Atrial fibrillation (percentage of patients) 29 33 40 NS Cerebrovascular disease (percentage of patients) 16 11 27 NS Habitual snorers (percentage of patients) 12 12 12 NS Males (percentage of patients) 57 77 70 NS Values are given as mean ± SD. Patients are divided according to the severity of Cheyne-Stokes respiration (quantified as percentage of total recording time); the limits are arbitrarily chosen. Significance testing is made with Spearman's rank correlation test for continuous variables and with Mann-Whitney U-test for categorical data. Results Clinical data Two hundred and three patients were included, however final analysis only included 191 patients (32% women). Three patients were excluded because of total recording time less than two hours. Eight patients were excluded due to the poor quality of their recordings and one patient was excluded because of an abnormal irregular breathing pattern that could not be categorized as Cheyne-Stokes or obstructive sleep apnoea. The aetiology of the heart failure was ischaemic heart disease in 60%, dilated cardiomyopathy in 3%, hypertension in 14%, valvular disease in 8% and other or unknown reason in 15%. Half of the patients had had heart failure diagnosis for more than a year. All but eight patients were prescribed diuretics, 43% digitalis, 70% ACE inhibitors and 28% beta-blockers. Almost half (48%) of the patients used hypnotics (usually bensodiazepines). The clinical characteristics of the study patients are presented in Table 1 . Five patients were in NYHA class 1, 83 in NYHA 2, 89 in NYHA 3 and five in NYHA 4. Only 10 patients were free from concomitant disease. Twenty five percent had diabetes, 13% chronic obstructive pulmonary disease and 18% cerebrovascular disease (reversible or permanent cerebral ischaemia or haemorrhage) at any time prior to the investigation. Five percent had cancer diagnosed and treated within the last year. Table 1 Clinical characteristics of the study patients All patients Females (n = 61) Males (n = 130) Significance Age (years) 72. 6 ± 10.0 75.1 ± 8.9 71.4 ± 10.3 P < 0.05 BMI (kg/m 2 ) 25.7 ± 4.6 25.4 ± 5.4 25.9 ± 4.3 NS LVIDD (mm) 56.5 ± 9.4 51.2 ± 9.5 59.0 ± 8.3 P < 0.001 LVEF (%) 36.2 ± 12.1 39.0 ± 13.0 34.9 ± 11.6 P < 0.05 LVEF ≥45% (percentage of patients) 26 36 22 P < 0.05 Ischaemic heart disease (percentage of patients) 60 57 62 NS NYHA 3–4 (percentage of patients) 52 57 49 NS Atrial fibrillation (percentage of patients) 35 33 35 NS Cerebrovascular disease (percentage of patients) 18 15 19 NS Habitual snorers (percentage of patients) 12 7 15 P < 0.05 Values for continuous data (age, BMI, LVIDD and LVEF) are given as mean ± SD. Significance testing is made with T-test for continuous data and chi-square for the other data. Overall 26% had heart failure due to left ventricular diastolic dysfunction (ejection fraction >45%). Female patients were older (Table 1 ) and more likely to have heart failure with preserved left ventricular systolic function; 36% had ejection fraction ≥45% as compared to 22% of the males (p < 0.05, chi square). There was no significant gender difference in the prevalence of ischaemic heart disease as the cause of heart failure. Respiration during sleep The average recording time was 424 (SD 75, median 444, interquartile 400–472) minutes and average CSR% 35 (SD 30, median 28, interquartile 6–59). Predominantly obstructive apnoeas were found in seven patients (3.7%) and CSR (arbitrarily defined as CSR% >10%) in 126 patients (66%). Sixty (31%) of the patients had CSR more than 50% of the recording time (Table 2 ). In figure 1 , four examples of different breathing patterns are demonstrated. Figure 1 Examples of nocturnal respiratory recordings. The upper part of each panel is the flow signal from the thermocouples and the lower is the impedance signal from the ECG electrodes. The duration of each example is 6 minutes. Panel A depicts an unequivocal period of Cheyne-Stokes respiration. Panel B was interpreted as normal by the software of the recording device, but was interpreted by us as Cheyne-Stokes respiration. Panel C is an example of obstructive sleep apnoeas and panel D is a period of Cheyne-Stokes respiration with a small obstructive component, classified by us as Cheyne-Stokes respiration rather than obstructive sleep apnoeas. Univariate rank correlation analysis showed that CSR% was most strongly correlated to age (R = 0.24, p < 0.01, figure 2 ), but also to left ventricular ejection fraction (LVEF) and to left ventricular diastolic diameter (table 2 ). Stepwise multiple regression analysis with gender, age, BMI, LVEF and LVIDD as independent variabels confirmed the relative importance of age vs that of LVEF and LVIDD (r 2 0.06 vs 0.03 and 0.03). The severity of CSR was also greater in patients with ischaemic heart disease. There were no relationships between medication (e.g. betablockers or benzodiazepines) and CSR%. Figure 2 Cheyne-Stokes respiration as a function of age. Cheyne-Stokes respiration (% of total recording time) as a function of age, linear regression line and confidence bands are drawn. Filled circles denote patients with LVEF 45 and above, i.e. patients with diastolic heart failure. Seven patients (3.6%) were found to have predominantly obstructive sleep apnoea (average AHI 16.0, SD 7.0, median 16.2, interquartile range 8.7 – 21.8). Excessive snoring, obesity or male gender were not overrepresented in this group. Discussion We have shown that CSR is common in elderly patients hospitalised due to decompensated chronic heart failure and that age was a stronger determinant of CSR than any of the cardiac or other clinical variables Selection of patients and timing of the study Malmö University Hospital serves a population of 250,000 inhabitants. In general all patients primarily admitted to hospital due to decompensated chronic heart failure are treated at the Heart Failure Unit until discharge from hospital. Due to this and to the liberal inclusion criteria we believe that the study patients are representative of the general population of patients with heart failure. Many other studies suffer from the disadvantages of a selection bias by excluding older patients, including only men or including only patients specifically referred for sleep studies [ 16 ] or for evaluation for heart transplantation [ 17 ]. One major difference between our study design and that of the majority of other studies is that we included patients immediately after an episode of decompensation, whereas most other authors have studied patients 1–3 months after discharge. We do not, however, know to what extent this approach affects nocturnal breathing patterns. The results reported by Tremel et al [ 2 ] suggest that sleep respiratory disturbances are stable during the second month after an episode of worsening heart failure, whereas there are no data, as far as we know, examining patients prior to that phase. Nonetheless, many observations of apnoeas (by nurses and relatives, posing questions to the clinician) are made during the hospitalisation. Our results are therefore relevant with respect to the factors associated with apnoeas in this situation. Respiratory disturbances during sleep At the time of the study, the most convenient and easy manageable system available at our department was the EdenTrace system. The semiquantitative nature of chest wall impedance measurement may have reduced the sensitivity of our measurements. However, by combining impedance measurement of respiratory movements, recording of oronasal airflow with thermocouples and finger pulse oxymetry with a careful visual analysis af the traces by an experienced physician (SB), we postulate that the distinction between central and obstructive events is sufficiently accurate. Furthermore that the bias introduced, would tend to underestimate, rather than overestimate the prevalence of sleep disordered breathing in the studied population. Whereas there are widely accepted standards for analysis of sleep apnoea the definition of CSR is less precise, and there is no accepted standard method for its quantification. We quantified the severity of CSR as percentage of total recording time. This approach is suggested by Ancoli-Israel et al [ 14 ] and is accepted also by the American Academy of Sleep Medicine Task Force [ 15 ] since it is simple and well suited for routine clinical use, irrespective of the technical methods available for respiratory recordings. The overall prevalence of CSR in this material was 66%, using an arbitrary limit of CSR% >10%. This figure is in the same magnitude as the 50–60% previously reported in stable heart failure outpatients [ 18 , 19 ] and close to that found by Ancoli-Israel (70%) in a small group of elderly, hospitalised patients [ 14 ]. We suggest that the main cause of the prevalence differences between studies is the age of the population studied, rather than the timing with regard to worsening heart failure [ 2 ]. We found 28 patients with mixed apnoeas (see figure 1 panel D for example). We believe that these patients should be considered to have a variant of CSR rather than to have an obstructive sleep apnoea syndrome (OSAS). There was no excess of snoring, obesity or male gender in this group, factors that are otherwise considered to be associated with OSAS. Furthermore, exclusion of these patients from the analyses did not change our results. It has been suggested that obstructive sleep apnoeas and CSR in heart failure patients both are part of a spectrum of periodic breathing [ 20 , 21 ], our data are compatible with this hypothesis. Age The most consistent result of our study is that age was more strongly related to Cheyne Stokes respiration than any other variable recorded. This corroborates data from other large studies without an upper age limit [ 14 , 16 , 22 ]. Many other studies that failed to demonstrate a similar relationship are constrained to patients below an arbitrary age limit or to patients referred for cardiac transplantation [ 17 ], which strongly affects validity for the general in-hospital patients. The strength of the association with age remained also when we excluded the youngest outliers (see figure 2 ) from the statistical analysis. We therefore consider our finding of age-dependency to be valid for unselected in-hospital patients with heart failure. It should be emphasized, however, that the predictive value of age was very weak, only explaining 6% of the total variability. Increasing prevalence of central apnoeas with age in normal subjects has been demonstrated by e.g. Bixler and coworkers [ 23 ], although the prevalence is much lower than that found in our patients with heart failure. Bixler et al suggest that a conservative approach should be applied when interpreting sleep studies in elderly. Our data suggest that this may also be a valid strategy for patients with heart failure. Although the presence of CSR seems to be associated with the presence of heart failure, its severity gives little information about the severity of the heart disease. Gender We found no gender effect (table 2 ), in contrast to the findings of e.g . Sin and coworkers [ 16 ]. One important reason may be that their 450 patients (only 15% women vs 32% in our study) were not a random sample of heart failure patients, but represent a much younger population (mean age 60 years vs 72 in our study) specifically referred for a sleep study. Indices of heart failure Heart dilatation (in our study measured as LVIDD) has been claimed to be an important factor for the development of CSR [ 24 ]. Impaired systolic function, as demonstrated by a low LVEF, is another factor that is usually considered to be associated with CSR. In our data, we found that these factors explained only 3% each of the variability of CSR. This contrasts to the results of many other studies, but is well in accordance with the findings of Sin and coworkers in their large study [ 16 ]. The idea of CSR mainly being a function of low cardiac output (as estimated by e.g. LVEF) may therefore be an oversimplification. Atrial fibrillation is a third factor that has been claimed to be associated with CSR [ 16 , 18 , 22 ], but this could not be confirmed in our study. One reason for the discrepancy may be that the patients in the studies of Sin [ 16 ] and Javaheri [ 18 ] were considerably younger than our patients, with a lower prevalence of atrial fibrillation. Thus their findings may have been confounded by an age effect. However the study of Blackshear [ 22 ] with a non-selected sample of elderly heart failure patients with a high prevalence of atrial fibrillation demonstrates a strong relationship between atrial fibrillation and CSR. The reason for the discrepancy between their results and our present study is not obvious. Concomitant diseases Patients with a history of stroke are usually excluded from studies on CSR and heart failure. This may be a limitation of the external validity of such studies, since many patients with heart failure (18% in our material) also have a history of minor or major cerebrovascular disease. We found however, quite unexpectedly, that this was not associated with a higher occurrence of CSR, thereby corroborating the findings of Blackshear et al [ 22 ]. Neither was any other concomitant disease associated with CSR. There was no association between intake of hypnotics and CSR. Conclusions Although presence of Cheyne-Stokes respiration indicates presence of left ventricular dysfunction, its severity seems only weakly related to severity of heart failure. Age was a stronger determinant, which may reflect the underlying age-dependency found also in healthy subjects. Follow-up of the current patient cohort will be performed, but from the present data, we cannot conclude if Cheyne-Stokes respiration is of clinical importance or not. Authors' contributions LM coordinated and performed the study which was designed by CC and BM. LE and CC were responsible for the cardiac investigations and SB and BM for the interpretation of the nocturnal respiratory recordings.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC521193.xml

529256

Potentially inappropriate prescriptions for older patients in long-term care

Background Inappropriate medication use is a major healthcare issue for the elderly population. This study explored the prevalence of potentially inappropriate prescriptions (PIPs) in long-term care in metropolitan Quebec. Methods A cross sectional chart review of 2,633 long-term care older patients of the Quebec City area was performed. An explicit criteria list for PIPs was developed based on the literature and validated by a modified Delphi method. Medication orders were reviewed to describe prescribing patterns and to determine the prevalence of PIPs. A multivariate analysis was performed to identify predictors of PIPs. Results Almost all residents (94.0%) were receiving one or more prescribed medication; on average patients had 4.8 prescribed medications. A majority (54.7%) of treated patients had a potentially inappropriate prescription (PIP). Most common PIPs were drug interactions (33.9% of treated patients), followed by potentially inappropriate duration (23.6%), potentially inappropriate medication (14.7%) and potentially inappropriate dosage (9.6%). PIPs were most frequent for medications of the central nervous system (10.8% of prescribed medication). The likelihood of PIP increased significantly as the number of drugs prescribed increased (odds ratio [OR]: 1.38, 95% confidence interval [CI]: 1.33 – 1.43) and with the length of stay (OR: 1.78, CI: 1.43 – 2.20). On the other hand, the risk of receiving a PIP decreased with age. Conclusion Potentially inappropriate prescribing is a serious problem in the highly medicated long-term care population in metropolitan Quebec. Use of explicit criteria lists may help identify the most critical issues and prioritize interventions to improve quality of care and patient safety.

Background Inappropriate medication use is a major health care issue for the elderly population [ 1 - 3 ]. Older patients are more at risk for adverse medication outcomes because they often have complex drug regimens and because of the age-related changes in drug pharmacokinetics and pharmacodynamics [ 1 ]. Potentially inappropriate prescriptions (PIPs), defined as prescriptions in which risks outweigh benefits, have been assessed in various settings using lists of explicit criteria most often based on that developed by Beers [ 4 ]. PIPs have been estimated to affect 4.8% to 45.6% of the elderly population [ 5 - 12 ]. Prevalence estimates of PIPs are likely to vary with the criteria that are applied. Some authors have based their assessment on the Beers criteria [ 5 - 7 , 9 - 12 ]. However, in all these studies but one [ 7 ], criteria applied were a subset only of Beers criteria as dosage and duration was not evaluated. Despite controversy about which explicit criteria should be used, there is a strong body of evidence showing that suboptimal prescribing is disturbingly common in older patients. In Canada, a list of explicit criteria was developed by a panel of experts in 1997 [ 13 ]. The Canadian criteria required diagnostic information which is not easily accessible in the long-term care setting [ 6 , 14 ]. Using various methodologies, several studies have investigated the extent of the problem in Canada. A 1995 study of community-dwelling and institutionalized older patients reported large variations in PIPs among provinces, ranging from 4.8% in the prairies to 12.8% in Quebec [ 9 ]. More recently, the prevalence of PIPs in long-term care patients in Ontario was reported to range between 14.9% and 20.8% [ 15 - 17 ]. In Quebec, a 1990 retrospective database survey of 63,268 older Medicare patients reported that 45.6% of non-institutionalized patients received high-risk prescriptions of questionable appropriateness [ 8 ], while a recent survey of 3,400 elderly patients in the Quebec general population reported that 6.5% had a potentially inappropriate prescription (PIP) [ 18 ]. A 1995 physician survey reported that 77.1% of nursing home patients in Quebec had been taking benzodiazepine for over one year [ 19 ]. The long-term care elderly population is particularly vulnerable to inappropriate medication use; it is composed of frail older patients who typically have functional disabilities and acute and chronic medical histories that require complex medication regimens [ 20 , 21 ]. Assessing PIPs using the data available in long-term care, in particular data on dosage and duration of use, may help designing efficient interventions to improve prescribing practices in one of the frailest populations. The objectives of this study were (1) to describe prescribing patterns in elderly patients residing in long-term care facilities in the Quebec metropolitan area, (2) to assess the prevalence of PIPs in this long-term care setting using published explicit criteria [ 4 , 13 , 22 ] adapted for this study, and (3) to identify patient-related predictors of PIPs. Methods Design and data sources A cross-sectional chart review of long-term care patients aged 65 years and over living in the Quebec City area was performed in the period between April 1995 and December 1996. All long-term care facilities located in the Quebec City area were contacted and the majority (29 out of 33) agreed to participate in the study. Within the 29 participating facilities, there were a total of 71 long-term care units. Numbers of beds in these units averaged 41 (10 to 190). Units were visited once during the study period. Data on drugs currently being prescribed the day of the visit was collected using medication charts. Demographic data included age, gender and length of stay. This study was approved by the ethics committees at Université Laval, Hôpital Saint-François d'Assise and Hôpital de l'Enfant-Jésus. For each medication order, the name, dosage, frequency of dosing and nature of prescription (scheduled or given on an as-needed basis) were collected. To capture the fullest possible extent of potentially inappropriate prescribing, it was assumed that all medications prescribed on an as-needed basis were taken. The total daily dose of an as-needed prescription was calculated by multiplying the prescribed unit dose with the indicated daily frequency of administration. Prescriptions for creams, ointments and drops were not included. Each medication was classified using the Anatomical Therapeutic Chemical (ATC) classification system [ 23 ]. The maximal prescribed daily dose was calculated for each medication order. Classification of potentially inappropriate prescribing A list of explicit criteria for PIP in older patients was developed based on a review of the literature [ 4 , 6 , 10 , 11 , 13 , 14 , 22 ]. Criteria referring to medications unavailable in Canada were excluded. Because diagnostic information is difficult to obtain in the long-term care setting [ 6 , 14 ], criteria involving clinical information were also excluded. The list of criteria was elaborated using a modified Delphi method [ 24 ]. A consensus panel of four local experts was convened including a general practitioner with a geriatric practice (RV), a family physician (LB), a clinical pharmacist and a pharmacoepidemiologist (JPG), all involved in practice or research on medication issues in the elderly population. In the first step, experts were asked to review and comment independently on the preliminary list of published criteria. Responses from the experts were used to revise this list. In the second step, the panel discussed each criterion until a consensus was reached. A total of 111 explicit criteria were included in the list to assess the quality of prescribing (Appendix). Medication charts were reviewed and compared with the list of explicit criteria. PIPs were categorized as: • Potentially inappropriate medication; • Potentially inappropriate duration; • Potentially inappropriate dosage; and • Potentially inappropriate drug-drug interaction. Data analyses Drug prescribed and PIP data were stratified by age and gender. Chi-square and Student t tests were used to compare proportions and means, respectively. Association between age and drug utilization was evaluated by analysis of variance. Factors predicting PIP were identified by logistic regression analyses. Independent variables were age, sex, number of prescribed drugs and length of stay. An initial bivariate analysis allowed calculation of crude odds ratios, identification of variables individually associated with the risk of PIP, and determination of the appropriate scale for each variable. A multivariate analysis with a significance threshold of 0.10 for the inclusion of variables subsequently yielded adjusted odds ratios for the number of prescribed medications, age and length of stay. Data were analyzed for collinearity and overdispersion. Data analyses were performed using SAS version 6.12 (SAS Institute Inc. Cary, NC). Results Study population The study population included 2,633 individuals, aged 65 years and older, residing in long-term care facilities for a mean duration of 8.5 years. Mean age was 82 ± 8 years and the majority of individuals were women (74.2%). Women were older than men (84 ± 8 years versus 79 ± 8 years, p = .0001). Drug utilization Most residents (94%, n = 2,481) had one or more prescribed medications and 48% (n = 1,266) of the total population had five or more medications. Residents had on average 4.8 prescribed medications. Proportions of patients by number of prescribed medications were similar for men and women, but varied according to age. The oldest patients, aged 85 years and more, received significantly less medications than their youngest counterparts aged between 65 and 74 years; 43.8% of patients aged over 85 years received five medications or more, compared to 59.4% of those aged 65 to 74 years. Of the 12,707 medications prescribed, 86% were scheduled administrations and 82% were prescribed for more than three months. A majority of patients (85.5%, n = 2,251 patients) had a prescription for medications of the central nervous system (CNS). Cardiovascular medications (46.4%, n = 1,221 patients) and medications of the alimentary tract and metabolism (29.3%, n = 772 patients) were the following most frequently prescribed anatomical groups of medications. Most commonly prescribed therapeutic classes included analgesics (48.0%), anxiolytics (41.4%), antipsychotics (35.0%) and loop (high-ceiling) diuretics (18.6%) (Table 1 ). There were differences in therapeutic classes prescribed to men and women. Acetaminophen (36.7% of patients), haloperidol (20.5% of patients) and lorazepam (20.2% of patients) were the three most frequently prescribed drugs (Table 2 ). Table 1 Proportion (in %) of elderly patients on medication by therapeutic class and sex* Proportion of patients (%) Therapeutic class All (n = 2,633) Men (n = 680) Women (n = 1,953) p value Analgesics & antipyretic 48.0 44.6 49.2 0.037 Anxyolitics 41.4 41.4 41.4 0.983 Antipsychotics 35.0 39.5 33.5 0.004 Loop (high ceiling) diuretics 18.6 16.2 19.4 0.062 Antiepileptics 14.9 21.6 12.6 <0.001 Thyroid preparations 14.6 8.4 16.8 <0.001 Vasodilators 14.6 11.0 15.8 0.002 Antidepressants 13.7 11.0 14.7 0.017 Cardiac glycosides 12.4 11.6 12.7 0.462 Drugs for peptic ulcer 11.0 11.3 10.9 0.765 Hypnotics & sedatives 10.9 9.8 11.3 0.293 Anticholinergics 10.9 11.2 10.8 0.788 Selective calcium channel blockers 10.6 7.2 11.7 0.001 Angiotensin converting enzyme inhibitors 10.0 10.0 10.0 0.991 * Only therapeutic classes prescribed to 10% or more of the elderly are displayed Table 2 Most frequently prescribed medications among the elderly in long-term care Proportion of patients (%) ATC code Medication Men (n = 680) Women (n = 1,953) All (n = 2,633) N02BE01 Acetaminophen 30.5 38.9 36.7 N05AD01 Haloperidol 21.9 20.1 20.5 N05BA06 Lorazepam 20.2 20.2 20.2 C03CA01 Furosemide 16.2 19.2 18.6 N02BA01 Acetyl salicylic acid 19.1 16.7 17.3 N05BA04 Oxazepam 16.3 17.1 16.9 H03AA01 Levothyroxin sodium 8.4 16.8 14.6 C01DA02 Nitroglycerin 10.0 14.6 13.4 C01AA05 Digoxin 11.6 12.7 12.4 ATC: Anatomical Therapeutic Classification Potentially inappropriate prescribing Overall, 51.5% of the population under study had one or more PIPs. Of the 2,481 patients with at least one prescribed drug, more than half (54.7%) had one or more PIPs; 29.5% had one PIP, 12.5% had two PIPs, 7.5% had three PIPs and 5.3% had four or more PIPs. A total of 12,707 drugs were prescribed of which 1807 were given on an as-needed basis. The proportion of PIPs among scheduled and as-needed prescriptions were 9.2% and 11.5%, respectively. If we exclude as-needed prescriptions, 46.4% of all residents had one or more PIPs. The most common type of PIP was drug-drug interaction, affecting 33.9% of patients treated with drugs, followed by potentially inappropriate duration (23.6%), potentially inappropriate medication (14.7%), and potentially inappropriate dosage (9.6%) (Figure 1 ). The proportion of patients receiving any type of PIP decreased with age, from 66.7% for patients aged 65 to 74 years to 56.4% for those aged 75 to 84 years and 47.7% for patients aged 85 years and more. PIPs were the most frequent for CNS medications, representing 9.3% of prescribed medications. Figure 1 Potentially inappropriate prescribing including inappropriate medication, dosage, duration and potential drug-drug interaction among three age-groupsof long-term care elderly (N = 2,481) ♂ = male ♀ = female The most common PIP was a potentially inappropriate duration for intermediate and short-acting benzodiazepines for more than one month (22.9%); more than half of those PIPs were for the anxiolytic oxazepam (Table 3 ). A substantial number of patients treated with pharmacotherapy were receiving repeat (dual) prescriptions of antipsychotics (16.5%) or benzodiazepines (14.9%). Almost 6% of patients treated with pharmacotherapy were prescribed potentially inappropriate long-acting benzodiazepines and 5.2% were receiving haloperidol at a potentially inappropriate dosage. The most common PIP among cardiovascular drugs was repeat prescription of calcium channel blockers, affecting 3.1% of treated patients. Table 3 Most common potentially inappropriate prescriptions (PIPs) among older patients receiving medication in long-term care Criteria Number of patients Proportion of all patients prescribed a medication (%) (N = 2,481) Potentially inappropriate medication 365 14.7 Long-acting benzodiazepines 138 5.6 Preparations including an antihistaminic 112 4.5 Flurazepam 54 2.2 Doxepin 31 1.3 Amitryptiline 27 1.1 Propanolol 27 1.1 Chloral hydrate 22 0.9 Potentially inappropriate duration 585 23.6 Intermediate and short-acting benzodiazepines at bedtime for more than one month 567 22.9 Oxazepam at bedtime for more than one month 313 12.6 Potentially inappropriate dosage 239 9.6 Haloperidol > 3 mg daily 129 5.2 Thioridazine > 30 mg daily 53 2.1 Lorazepam > 3 mg daily 34 1.4 Potential drug-drug interaction 842 33.9 Repeat* prescription of antipsychotics 409 16.5 Repeat* prescription of benzodiazepine 369 14.9 Clonazepam and other benzodiazepine 46 1.9 Benzodiazepine and hypnotic or sedative 93 3.8 Repeat* prescription of calcium channel blockers 77 3.1 Repeat* prescription of tricyclic antidepressants 37 1.5 Repeat* prescription of angiotensin converting enzyme inhibitors 19 0.8 Repeat* prescription of β-blockers 11 0.4 Repeat* prescription of non-steroidal anti-inflammatory drugs (except acetylsalicylic acid) 10 0.4 Repeat* prescription of barbiturate 10 0.4 Total potential inappropriate prescriptions ** 1,358 54.7 *Repeat prescription indicates that two agents of the same drug class are being prescribed **Numbers do not add up since one prescription may be linked to more than one PIP (e.g., duration and dosage) Predictors Multivariate analysis indicated that patients with a length of stay 10 years or over were 1.78 times at greater risk of being prescribed a PIP than those with less than 10 years of stay (adjusted odds ratio [OR]: 1.78, 95% confidence interval [CI]: 1.43–2.20) (Table 4 ). The risk of PIP also increased significantly as the number of drugs prescribed increased (OR: 1.36, CI: 1.32–1.41) whereas it decreased with age. Gender was not a significant predictor of PIP. No problems of collinearity or overdispersion were observed in the multivariate model. Table 4 Predictors of potentially inappropriate prescription among elderly patients in long-term care (N = 2,481) Predictor Proportion of patients with PIP (%) Crude odds ratio (95% CI) Adjusted odds ratio (95% CI)* Number of prescribed drugs (increments of one drug) 54.7 1.38 (1.33–1.43) 1.36 (1.32–1.41) Gender Women 54.5 1.00 - Men 55.5 1.04 (0.87–1.25) - Age 65 to 74 years 66.7 1.00 1.00 75 to 84 years 56.4 0.65 (0.51–0.81) 0.74 (0.58–0.96) 85 years or more 47.7 0.46 (0.36–0.57) 0.60 (0.47–0.77) Length of stay <10 years 51.1 1.00 1.00 ≥10 years 67.4 1.98 (1.62–2.41) 1.78 (1.43–2.20) CI: confidence interval; *Adjusted for number of prescriptions, age, and length of stay Discussion The long-term care elderly population evaluated in this study was highly medicated and a majority of patients receiving medication had a PIP. These results indicate that potentially inappropriate prescribing was significant at the time of the study in institutionalized older patients in the Quebec metropolitan area. A total of 94% of residents in this long-term care population were prescribed at least one drug, compared to 60% in community-dwelling elderly patients in Quebec [ 25 ]. The mean number of medications was also higher (4.8) than in community-dwelling individuals in Quebec (2.9) [ 25 ], but lower than in American long-term care (7.2) [ 7 ]. The total prevalence of PIPs among the population under study was high (51.5%). Estimates of PIP prevalence in the literature vary between 4.8% [ 6 ] and 45.6% [ 8 ] for both institutionalized and community-dwelling older patients. Caution must be used when comparing these results, as the delivery of care may vary from one setting and one region to another [ 9 ]. The current lack of consensus when defining lists of criteria and variations with respect to methodologies also contribute to the observed differences [ 26 ]. For example, Zhan and colleagues [ 5 ] estimated the proportion of potentially inappropriate medication use in the community-dwelling elderly in the United States. Applying criteria on the indication for the use of 33 drugs, they observed a prevalence of 21.3% for 1996. In our study, PIPs were identified using an explicit criteria list that was primarily based on Beers and McLeod criteria [ 4 , 13 , 22 ] and that was updated and validated by local experts to apply to the long-term care context in Quebec. As we had access to dosage and duration information, we were able to apply a broader set of criteria which can explain the higher prevalence of PIPs we have observed. Explicit criteria lists, such as those developed by Beers and McLeod, define inappropriate prescription according to the drug overall risk-benefit profile for elderly patients. These lists were previously used in studies examining inappropriate prescribing in elderly populations [ 3 , 5 , 6 , 11 , 15 , 27 - 30 ] and undergo a continuous process of revision and updating to reflect the most current clinical information on the risks and benefits of medications [ 31 ]. A large number of patients were receiving CNS medication (85%) and the most common PIPs were related to that category of drugs. Thirty-five percent of patients were prescribed antipsychotics and 22.9% had benzodiazepine for potentially inappropriate duration, defined as more than a month [ 32 ]. A number of studies have reported the inappropriate use of CNS drugs [ 5 , 8 , 33 , 34 ], particularly benzodiazepines [ 18 , 19 ]. Many factors may contribute to the continued use of inappropriate CNS medications, including prescriber attitudes, patient demands and the design of the health care system [ 34 ]. A survey of physicians in Quebec reported that the psychological distress of aging patients and the quasi-absence of reported side-effects justified the long-term use of psychotropic medication, which was seen as the most effective way of helping the patient [ 35 ]. Moreover, side effects of psychoactive medication are often believed to be a consequence of the aging process [ 34 ]. Almost three quarters of potentially inappropriate psychoactive medications can produce a physical dependence [ 34 ]. Psychoactive pharmacotherapy increases risk of hip fractures and is advocated for use with caution to prevent falls in elderly populations [ 36 , 37 ]. Anticonvulsants, antidepressants and short- and long-acting benzodiazepines were reported to increase risk of falls in older women [ 38 ]. The length of stay was positively associated with PIPs, while the prevalence of PIPs decreased with age. Although the association between length of stay and the likelihood of receiving a PIP in nursing homes was studied in the past [ 6 ], to our knowledge, this is the first time it is being shown to be a predictor of PIPs. On the other hand, the risk of receiving a PIP was previously reported to decrease with age in nursing home patients over 65 years [ 7 , 12 ]. Data on clinical status was not considered in these studies and it can be hypothesized that either the oldest residents were less ill or that physicians were more cautious when prescribing to very old patients. As reported in previous studies [ 12 , 26 ], the number of medications was also a predictor of PIP in older patients. Patients in long-term care frequently have multiple diseases resulting in complex medication regimens, which makes assessment of the risks versus benefits of treatments often difficult. Female gender was previously reported as a predictor of PIP [ 7 , 12 ]. Although we observed gender differences in the prescribed therapeutic classes, female gender was not a predictor of PIPs in our study. The results presented here should be viewed in light of potential limitations. As in previous studies [ 15 ], we did not abstract information on diagnoses from the patients charts and drug prescriptions were considered as surrogates for disease conditions. Thus, the explicit criteria used in this study apply to general circumstances, but may not be applicable to specific cases, since they do not consider clinical information. For example, lipid-lowering drugs may be potentially inappropriate in patients aged 75 and over, but evidence from clinical trials suggests that statins may be of benefit if the patient's life expectancy exceeds two years [ 39 ]. Thus, misidentification of potential cases of appropriate or inappropriate prescribing may have occurred, since complex medical conditions can alter the risk-benefit profile of medications. However, due to the frail condition of most patients, it is unlikely that such misidentifications have occurred frequently. Since access to clinical data is often difficult in the nursing home setting, a list of explicit criteria that does not require that type of information may be easier to apply on a larger scale. This study evaluated prescription patterns rather than the actual consumption of medication. The low prevalence of as-needed medication (14%) and the long-term care setting, in which medication is administered to patients by a health caregiver, suggest that this limitation did not have a significant impact on the results. As-needed prescriptions may have accounted for repeat prescriptions, which may in turn have led to overestimation of the number of drug-drug interactions. However, even after excluding as-needed prescriptions from the analysis, the proportion of residents with a PIP remains high. Predictors of PIPs were assessed using a multivariate analysis. It allowed us to adjust for potential confounding variables. However, we were not able to adjust for facility variables as those were not available. This study is the first to describe and qualify prescribing practices in long-term care facilities in urban Quebec. In particular, it highlights the extent of potentially inappropriate prescribing in elderly long-term care patients, which are among the frailest of society [ 4 , 21 ]. Inappropriate prescribing is one component of the major health care problem of suboptimal prescribing that also includes underuse of effective agents, drug-disease interactions and prescription errors. Substantial morbidity, mortality and cost are attributed to suboptimal prescribing [ 1 , 2 ]. Although a decline in the prevalence of PIPs was reported in community-dwelling older patients in the United States between 1987 and 1996 [ 40 ], the continued use of inappropriate medications is a major concern. A growing body of evidence suggests that clinical pharmacy and multidisciplinary team interventions can modify suboptimal prescribing in older patients. Modern data management [ 15 , 41 ] and use of the best clinical evidence could help practitioners improve the management of complex cases [ 40 , 42 ]. Recent studies in long-term care settings showed that physician or pharmacist interventions reduce PIPs [ 1 , 12 , 16 , 43 ], while a clinical review program of prescriptions for community-dwelling patients conducted by a team of physicians, pharmacists and nurses did not seem to improve prescribing practices [ 44 ]. Conclusions Inappropriate prescribing is highly prevalent in the elderly long-term care population in metropolitan Quebec. The use of a explicit criteria list to identify PIPs is a first step towards identifying most critical issues and implementing strategies to improve quality of care and patient safety. Identifying predictors of PIPs may help to target problems and prioritize interventions that are most needed in the rapidly expanding older population. Competing interests Carol Rancourt and Jean-Pierre Grégoire were employed by Merck Frosst Canada at the time of the preparation of this article. Authors' contributions CR, in partial fulfillment for the grade of M.Sc., lead the protocol development, expert panel consultation, data analyses, discussion of results, and manuscript preparation. JM contributed to all steps of this research project and manuscript preparation. LB contributed to protocol development, presentation and discussion of results and manuscript preparation and participated in the expert panel to define the explicit criteria. RV is the principal investigator for the initial research project which generated the drug prescription data used for this study. He contributed to protocol development, presentation and discussion of results, manuscript preparation and participated in the expert panel to define the explicit criteria. DL was a co-investigator for the initial research project, which generated the drug prescription data used for this study, and contributed to protocol development, data analyses and manuscript preparation. All authors read and approved the final manuscript. JPG contributed to protocol development presentation and discussion of results, manuscript preparation and participated in the expert panel to define the explicit criteria. Pre-publication history The pre-publication history for this paper can be accessed here: Supplementary Material Additional File 1 Appendix: List of explicit criteria used to assess the quality of prescribing in long-term care for elderly patients provide as additional file Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC529256.xml

549549

The responses of Ht22 cells to oxidative stress induced by buthionine sulfoximine (BSO)

Background glutathione (GSH) is the most abundant thiol antioxidant in mammalian cells. It directly reacts with reactive oxygen species (ROS), functions as a cofactor of antioxidant enzymes, and maintains thiol redox potential in cells. GSH depletion has been implicated in the pathogenesis of neurological diseases, particularly to Parkinson's disease (PD). The purpose of this study was to investigate the change of cellular antioxidant status and basic cell functions in the relatively early stages of GSH depletion. Results in this study, GSH was depleted by inhibition of glutamylcysteine synthetase using buthionine sulfoximine (BSO) treatment in Ht22, a neuronal cell line derived from mouse hippocampus. Treatment with BSO produced dose-dependent decreases in total GSH level, Fe3+-reducing ability (FRAP assay), Cu2+-reducing ability (Antioxidant Potential, AOP assay), and ABTS free radical scavenging ability (ABTS assay) of the cells, but the sensitivity of these indicators to dosage varied considerably. Most of the changes were completed during the first 8 hours of treatment. Cell viability was tested by MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromid) assay, and cells at lower density in culture were found to be more sensitive to GSH depletion. The activity of antioxidant enzymes, such as glutathione peroxidase (GPx), glutathione reductase (GR), and copper/zinc superoxide dismutase (Cu/Zn-SOD) were affected by GSH depletion. A cDNA expression array assay of the effects of BSO treatment showed significantly decreased mRNA level for 3 genes, and significantly increased mRNA level for 10 genes, including the antioxidant enzymes Cu/Zn-SOD and thioredoxin peroxidase 2 (TPxII). Conclusions the study suggests that there are BSO-sensitive and BSO-resistant pools of GSH in Ht22 cells, and that different categories of antioxidant react differently to GSH depletion. Further, the effect of GSH status on cell viability is cell density dependent. Finally, the alterations in expression or activity of several antioxidant enzymes provide insight into the various cellular responses to GSH depletion.

Background Glutathione (GSH, tripeptide γ-L-glutamyl-L-cysteinyl-glycine) is the most abundant thiol antioxidant in mammalian cells. It reacts directly with reactive oxygen species (ROS), or functions as a cofactor of antioxidant enzymes such as the glutathione peroxidases (GPxs). In addition, GSH keeps sulfhydryl groups of cytosolic proteins in reduced form by maintaining thiol redox potential in cells [ 1 ], and regulates cell signaling pathway in apoptosis [ 2 , 3 ]. The requirement for GSH and total antioxidant capacity is particularly high in brain. Brain consumes 20% of total oxygen in the body, and thus undergoes high levels of oxidative challenge. Cumulative oxidative damage has been strongly implicated in neurodegeneration and neurological diseases, such as Parkinson's disease (PD), Alzheimer's disease (AD), and amyotrophic lateral sclerosis (ALS). The importance of GSH in particular has been indicated in the case of PD. Among the progressive oxidative stresses that occur in the pathogenesis of PD, an characteristic is the decrease in total GSH concentrations in the substantia nigra in preclinical stages, when other biochemical changes are still undetectable [ 4 ]. Investigation of the consequences of intracellular GSH depletion in neuronal cell lines has relied predominantly on one of three methods. These are: (a) treatment with homocysteic acid (HCA) or glutamate to block the uptake of cystine, a substrate for GSH synthesis [ 5 - 7 ]; (b) treatment with BSO [ 5 ] to inactivate γ-glutamylcysteine synthetase, the rate limiting enzyme in GSH synthesis, and (c) treatment with ethacrynic acid [ 8 ] or diethyl maleate [ 9 ] to react with the thiol group of GSH. The oxidative stress caused by GSH depletion further affects the status of other antioxidants. The concept of total antioxidant capacity (TAC) has been developed to assess the general antioxidant activity in biological samples without distinguishing the contribution from each individual component. Measurements of TAC have been intensively developed in the past 20 years. The principle of these methods is to evaluate the total effect of all contributing compounds in the system by one criterion, such as free radical scavenging, electron donation or protection against the oxidative damage to lipids, proteins or DNA. But considering the fact that different categories of antioxidant work through different mechanisms and need specified conditions for maximal function, it is impossible to cover the activities of all antioxidants in one assay. Thus multi-dimensional measurements on TAC have been suggested [ 10 ]. In this study, GSH was depleted by BSO in Ht22, a neuronal cell line derived from mouse hippocampus. GSH level, TAC, antioxidant enzyme activity, cell viability and gene expression were assessed. Results and discussion Intracellular total GSH level vs . cell density in culture Before depleting GSH from cells, total GSH levels in Ht22 were monitored from pre-log phase to the end of log phase of cell growth (Figures 1 & 2 ). Intracellular GSH levels decreased as cell density increased, with this effect being more dramatic as the cells entered log phase growth. Whether the decrease of GSH with cell growth is due to limited nutrient supply or to programmed regulation is not known. It has been reported that the GSH content of brain cells depends strongly on the availability of precursors for GSH [ 11 ]. It was also noticed that as Ht22 cells grow denser in culture, the intracellular ROS level decreased (Chen et al., unpublished data), as apparently balances with the decreased GSH concentration. Dose responses Intracellular GSH The dose responses of Ht22 to BSO were analyzed by varying BSO concentration from 0.03 to 10 mM in a 15-hr treatment, and the changes of intracellular GSH level and TAC were measured (Figure 3 ). Treatment with 0.03 mM BSO resulted in a dramatic decrease of total GSH level to 35% of the control level. Increasing the BSO concentration to 10 mM further decreased the GSH level to 22% of control, representing an additional drop of only 13%. In comparison to 0.03 mM of BSO, increasing the concentration to 1 mM or higher caused significant decreases in GSH levels (P values ≤ 0.0399). This result suggests there are two pools of GSH in the cell, one easily depleted by BSO, and the other more resistant to depletion. A previous study by Seyfried et al. [ 12 ] showed that BSO treatment of PC12 cells was more efficient at depleting cytosolic GSH than mitochondrial GSH, indicating that the BSO-sensitive and BSO-resistant GSH pools in Ht22 might localize to cytosol and mitochondria, respectively. After 10 mM BSO treatment for 15 hrs, the predominant form of glutathione was the reduced form (GSH); only about 5% of the total glutathione was found in the oxidized form (GSSG). After 15 hrs BSO treatment, malondialdehyde (MDA) assay showed the increases of Abs at 586 nm caused by MDA formation were: control = 0.011 ± 0.006; 1 mM BSO = 0.012 ± 0.006; 3 mM BSO = 0.009 ± 0.004; and 10 mM BSO = 0.011 ± 0.004 (average ± SD, 3–5 independent experiments), no increase in lipid peroxidation after the treatments was observed, indicating GSH depletion at these levels was not yet destructive to the cells. Total antioxidant capacity Three methods were employed in this study to investigate antioxidant status following BSO treatment of Ht22 cells. The Cu2+-reducing ability assay (Antioxidant Potential, AOP) and Fe3+-reducing ability assay (FRAP) both measure the activity of metal ion-reducing antioxidants, but the FRAP assay is characterized by its low pH (3.7), thus excluding the antioxidant function of thiols. Figure 3 shows that from 0.03 to 10 mM BSO treatment, the FRAP value had a sharp decrease from 76 to 32% of the control, indicating that some non-thiol antioxidants were expended to preserve the BSO-resistant GSH or to cope with oxidative stress caused by the depletion of BSO-sensitive GSH. In comparison to FRAP, the AOP value showed a gradual decrease from 76 to 50 % of control through the range of BSO concentrations, and this lesser decrease may be partially maintained by the BSO-resistant GSH pool in the cells. In contrast to FRAP and AOP, the ABTS free radical scavenging ability (ABTS assay) of the cells maintained at about 80% of the control level at all BSO concentrations. Cao et al. [ 13 ] previously showed that GSH is highly reactive to ABTS radical. Unlike the FRAP and AOP assays, the ABTS assay appears to be less sensitive to the metal ion-reducing antioxidants that were consumed by GSH depletion. Cell viabilities and bioreduction activity Cell viabilities Effects of GSH depletion on cell viability were assayed using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromid)-based cell viability assay, an indicator of mitochondrial activity (Figure 4 ). Ht22 cells at higher density (1e5 cells per well, equal to 1.3e5 cells per square cm) were more resistant to GSH depletion, and in spite of the change in BSO concentration, the decrease in cell viability remained at about 10% of control. Lower cell density (5e4 cells per well, equal to 6.5e4 cells per square cm) rendered higher sensitivity; as BSO concentration increased from 0.03 to 10 mM, the viability dropped from 100 to 50% of control. At 3 and 10 mM BSO concentrations, the differences caused by cell density are significant (P values ≤ 0.0027). These results show that the mitochondrial activity of cells is influenced by GSH depletion, and that this effect is sensitive to the density of cells in culture. The reason(s) behind the cell density factor is not clear. Either proliferation status or communication between cells may contribute. In neurodegenerative diseases, the effect of cell death, thus the decrease of cell density, on the susceptibility of neurons to GSH depletion will be worth investigating. Wither sever cell death can accelerate GSH loss is not known. No changes in cell viability in response to BSO treatment were detected by trypan blue staining, which distinguishes dead cells from live ones. Cell viabilities were 98.9% (N = 25) for untreated cells, and 99.6% (N = 5), 99.1% (N = 5), and 98.1% (N = 12), after 15 hrs treatment with 1, 5, and 10 mM BSO, respectively. The oxidative stress produced in these treatments is not sufficient to be lethal to the cells, in accordance with the MDA assay, which shows no increase in lipid peroxidation. Bioreduction activity Figure 5 shows a slight tendency of increase in the bioreduction activity in cells after BSO treatment, as is more evident when cells were at lower density. The effective resazurin-reducing compound(s) in Ht22 is not clear, but the possible background from BSO was excluded. The mechanism behind the change is yet to be understood. Time course The responses of Ht22 cells to 0.1 mM BSO treatment were tested at 4, 8, 12 and 15 hrs, and results are expressed as percentage of 0 hr (Figure 6 ). For total GSH, FRAP and ABTS, 4-hr treatment induced significant decreases (P values ≤ 0.0001), while AOP value did not change significantly. From 4 hrs to 8 hrs, all of the assays showed significant decreases (P values ≤ 0.0326). Whereas from 8 through 15 hrs, no significant change was seen. The data suggests that for 0.1 mM BSO treatment, the majority of the GSH depletion occurred in the first 8 hrs, and that the depletion of other antioxidants were dynamically parallel to GSH depletion. Cell death was visible under microscopy after 20 hrs of 0.1 or 10 mM BSO treatments. Seyfried et al. [ 12 ] found that PC12 cells treated with 0.5 mM BSO maintained 100 % of mitochondrial GSH in the first 4 hrs, while cytosolic GSH was completely depleted. Extending the treatment from 4 hrs to 6 hrs resulted in 50 % depletion of mitochondrial GSH. These data are in agreement with the significant decreases of total GSH level from 0 to 4 hrs, and from 4 hrs to 8 hrs in this study, although the differences in cell line and dosage of BSO in the two studies should be kept in mind while comparing the results. Antioxidant enzyme activities Glutathione peroxidase Glutathione peroxidases (GPxs) are selenium-containing antioxidant enzymes that reduce hydrogen peroxide to water, or lipid peroxides to ethanols, with GSH as reducing cofactor. Five isoforms of the GPx family are commonly known. In this study, the major activity detected by the assay is cellular GPx (GPx1). Figure 7 shows the change of GPx activity in response to 1, 3 and 10 mM BSO treatments for 15 hrs. Treatment with 1 mM BSO significantly increased the GPx activity to 127% of control (P = 0.0401). Treatment with 3 or 10 mM BSO decreased the activity by about 20% of control. This treatment was not significant compared to control, but is significant when compared to 1 mM BSO treatment (P values ≤ 0.0421). The results show that the regulation of GPx activity is dependent on the level of GSH depletion, and that the reserve pool of GSH may have critical function, since slight depletion caused significant changes. Glutathione reductase Glutathione reductase (GR) reduces GSSG to GSH using NADPH as cofactor. When Ht22 cells were subjected to 1, 3 and 10 mM BSO treatment for 15 hrs, the GR activity in the cells dropped to 97%, 95% and 94% of control, respectively (Figure 8 ). At 10 mM BSO concentration, the decrease is significant (P = 0.045) in comparison to the control level. However, the toxicology cDNA expression array study in Section 5 did not show significant changes in the mRNA level of GR. The activity decrease of this enzyme may be due to enzyme inactivation caused by oxidative stress. Superoxide dismutase Superoxide dismutase (SOD) converts superoxide anion to hydrogen peroxide, which can be further detoxified by GPx1 or catalase. Two isoforms of SOD are found in mammalian cells, known as Cu/Zn-SOD and Mn-SOD, and localized in cytosol and mitochondria, respectively. Incubating Ht22 cells with 1, 3, and 10 mM BSO for 15 hrs increased the Cu/Zn-SOD activity to 104%, 112%, and 152% (P= 0.0619) of the control level (Figure 9 ). The toxicology cDNA expression array study found a 2-fold increase of Cu/Zn-SOD mRNA level in response to treatment with 10 mM BSO treatment. In contrast to GPx, SOD responds more to severe GSH depletion or higher level of oxidative stress. Toxicology cDNA expression array cDNA expression arrays are powerful tools for studying gene expression under various circumstances. In this study, we employed the Atlas rat toxicology array II (BD) to monitor the changes in gene expression after 15 hrs treatment with 10 mM BSO. Out of a total of 465 genes in the array, mRNA level of 3 genes were significantly decreased, and 10 significantly increased (Table 1 ). The increased mRNA level of heat shock protein (HSP, the induction of which correlates with the abundance of unfolded polypeptide chains) and eukaryotic peptide chain release factor 1 (ERF1, which functions in termination of translation) indicate the stress caused by BSO treatment affected proteins at translational and structural levels. The increased mRNA level of antioxidant enzymes Cu/Zn-SOD and thioredoxin peroxidase 2 (TPx II, a peroxidase that requires thioredoxin or thiol-containing intermediates to carry out its peroxidase function) suggests these enzymes have important functions in coping with the oxidative stress caused by GSH depletion. Furthermore, both of these enzymes have been shown to protect cells from different inducers of apoptosis [ 14 , 15 ], indicating that they may have contributed to maintaining high cell viability after BSO treatment in this study. Conclusions Inhibition of glutamylcysteine synthetase in Ht22 cells by BSO revealed two pools of GSH in the cells, one susceptible to depletion by low concentration of BSO, and the other more resistant to depletion. TAC values measured by FRAP, AOP and ABTS methods showed parallel time courses to GSH depletion, but different dose-responses. The GSH depletion studied did not result in increases in GSH/GSSG ratio, lipid peroxidation, or cell death, but affected MTT-based cell viability. The antioxidant enzyme activities of GPx, GR and Cu/Zn-SOD were affected by the GSH depletion. The mRNA levels of HSP90-beta, ERF1, Cu/Zn-SOD and TPx II were significantly increased after 10 mM BSO-15 hr treatment. Methods Cell maintenance and treatment Ht22 cells were fed with Dulbecco's Modified Eagle Medium (DMEM) with 10% fetal bovine serum (FBS) supplement, and cultured at 50–55% relative humidity (RH), in 5% CO 2 at 37°C. For MTT-based cell viability assay and Resazurin-based bioreduction activity assay, cells were seeded at a density of 1 or 0.5 million cells per well (corresponding to 1.3e5 and 6.5e4 cells per square cm, respectively) in 48-well cell culture plates (Costar) 12 hrs before BSO treatment. For other assays, the cells were allowed to reach 80–90% of optical confluency (about 2e5 cells per square cm) in 100 mm cell culture dishes (Costar) before the BSO treatment. TAC, GSH/GSSG, MDA assays In this study, three methods were employed for total antioxidant capacity assay. The AOP assay (Antioxidant Potential assay kit, Oxis) tests the ability of samples to reduce Cu2+ to Cu+ at physiological pH, (assayed according to manufacturer's instructions with minor adjustments). The FRAP assay [ 16 ] tests the ability of samples to reduce Fe3+ to Fe2+ at pH 3.6, a low pH that inactivates thiol antioxidants. The ABTS assay [ 17 ] tests the ability of samples to scavenge ABTS radical at physiological pH. Modified versions [ 18 ] of FRAP and ABTS assays were used in this study. GSH was assayed for total level as well as GSH/GSSG ratio using GSH/GSSG ratio assay kit (Oxis). Manufacturer's instructions were followed with minor adjustments. Lipid peroxidation in cells was assayed using malondialdehyde (MDA), a product of lipid peroxide decomposition, as an indicator. The assay was carried out using a Malondialdehyde assay kit (Oxis). Cell viabilities and bioreduction activity assays A common method of testing cell viability is trypan blue staining. As a vital dye, trypan blue enters dead cells, distinguishing them from live ones. In this study, 0.4% (w/v) trypan blue-PBS solution was mixed with properly diluted cells at 5:1 ratio, and cell numbers were counted using a hemocytometer. Cell viability can also be reflected by dehydrogenase activity, which indicates the activity of mitochondria (Cell growth determination kit/MTT based, Sigma). Dehydrogenase converts MTT into purple MTT formazan, causing a colorimetric change that can be monitored photometrically. The bioreduction activity of cells was monitored by an in vitro toxicology assay kit (Sigma), based on a blue to red color change when the oxidoreduction dye, resazurin, is reduced by the bioreduction activity of the cells. Both MTT and Resazurin assays were carried out following the kit instructions. GPx, GR, and SOD activity assays GPx activity assay was based on the classical principle [ 19 ] with optimization to the Ht22 cell lysis. The peroxide used in this study was t-butyl hydroperoxide (0.323 mM), the concentration of GSH was reduced to 1.875 mM, and the pH of the assay was increased to 7.6. The GR and SOD activities were assayed by corresponding kits from Oxis. Toxicology cDNA array assay Atlas rat toxicology array II was purchased from BD, and the assay was carried out following the instruction manual. Statistical work F-test in SAS procedure "Proc GLM" was used for statistical work. Authors' contributions JC designed experiments, treated and harvested samples, optimized and participated in the antioxidant assays, participated in the toxicology cDNA array assay. ASH participated in the toxicology cDNA array assay. AY participated in harvesting samples, and the antioxidant assays. MJB contributed to conception and design, critical revision and final approval of the article.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549549.xml

544938

Homologs of wingless and decapentaplegic display a complex and dynamic expression profile during appendage development in the millipede Glomeris marginata (Myriapoda: Diplopoda)

Background The Drosophila genes wingless ( wg ) and decapentaplegic ( dpp ) comprise the top level of a hierarchical gene cascade involved in proximal-distal (PD) patterning of the legs. It remains unclear, whether this cascade is common to the appendages of all arthropods. Here, wg and dpp are studied in the millipede Glomeris marginata , a representative of the Myriapoda. Results Glomeris wg ( Gm-wg ) is expressed along the ventral side of the appendages compatible with functioning during the patterning of both the PD and dorsal-ventral (DV) axes. Gm-wg may also be involved in sensory organ formation in the gnathal appendages by inducing the expression of Distal-less ( Dll ) and H15 in the organ primordia. Expression of Glomeris dpp ( Gm-dpp ) is found at the tip of the trunk legs as well as weakly along the dorsal side of the legs in early stages. Taking data from other arthropods into account, these results may be interpreted in favor of a conserved mode of WG/DPP signaling. Apart from the main PD axis, many arthropod appendages have additional branches (e.g. endites). It is debated whether these extra branches develop their PD axis via the same mechanism as the main PD axis, or whether branch-specific mechanisms exist. Gene expression in possible endite homologs in Glomeris argues for the latter alternative. Conclusion All available data argue in favor of a conserved role of WG/DPP morphogen gradients in guiding the development of the main PD axis. Additional branches in multibranched (multiramous) appendage types apparently do not utilize the WG/DPP signaling system for their PD development. This further supports recent work on crustaceans and insects, that lead to similar conclusions.

Background The genes wingless ( wg ) and decapentaplegic ( dpp ) are important factors for the normal development of the Drosophila legs. Both genes encode secreted morphogens that generate combinatorial gradients across the developing imaginal leg discs (e.g. [ 1 ]). These gradients form the top level in a PD axis patterning cascade and they control expression of the genes at the next level of the cascade, the leg-gap genes (e.g. Distal-less ( Dll ), dachshund ( dac )) (e.g. [ 2 , 3 ]). Thus, wg and dpp are key factors involved in the early events of PD axis formation. In recent years several comparative studies in other arthropod species have suggested that the action of the leg-gap genes in PD patterning is evolutionarily conserved [ 4 - 9 ]. Tus, the question arose as to whether the regulation of the leg-gap genes by the WG/DPP morphogen gradient is also conserved. The currently available data provide no clear answer. Initially, the expression patterns did not support the conservation of this top level of the PD axis patterning cascade [ 9 , 10 ]. Other authors, however, have argued in favor of a conservation of WG/DPP morphogen signaling in PD axis formation [ 6 ]. Furthermore, many arthropods have appendages with more than one PD axis. It is currently debated whether these multiple axes are all patterned by a cascade involving wg and dpp at the top level, or whether the different branches are patterned through branch-specific mechanisms. The comparative analyses of wg and dpp expression during appendage formation to date mainly focus on insects (e.g. Tribolium , Gryllus , Schistocerca , Athalia [ 9 - 12 ]). Only a few representatives of the crustaceans and chelicerates have been studied from other arthropod classes [ 6 , 13 ]. Here, I report on results concerning wg and dpp expression in the appendages of a representative of the fourth extant arthropod class, the myriapod Glomeris marginata . The wg gene of Glomeris is expressed on the ventral side of the appendages compatible with a conserved role in PD axis development. Additionally, Glomeris wg may induce expression of the genes Dll and H15 in the sensory organs of the mouthparts. The results with the Glomeris dpp gene are more ambiguous. Although the data can be interpreted in favor of a conservation of the WG/DPP morphogen gradients, clearly more work on the subject is necessary to clarify the evolution of PD axis patterning in arthropod appendages. In particular, it will be necessary to elucidate the mechanisms through which the additional PD axes in multibranched appendages are patterned. Results Cloning of Gm-dpp cDNA fragments A fragment of a Glomeris gene that shows sequence similarity to members of the TGF-beta gene family was isolated. In order to establish the orthology of this fragment, I performed a phylogenetic analysis incorporating a selection of TGF-beta proteins from Drosophila , other arthropods and a variety of deuterostome taxa. The resulting phylogenetic tree distinguishes two groups of proteins. One group comprises the arthropod dpp genes and their deuterostome homologs, the BMP2/4 genes. The second group consists of the Drosophila TGF-beta genes screw ( scw ) and glass bottom boat ( gbb ), and the remaining deuterostome BMP genes with TGF-beta homology, including zebrafish anti dorsalizing morphogenetic protein (ADMP). The Glomeris fragment resides in the dpp /BMP2/4 group and is therefore designated as Gm-dpp (Fig. 1 ). The resolution within the dpp /BMP2/4 group is low, with many nodes lacking statistical support. The Glomeris fragment forms a group together with dpp from the two-spotted cricket ( Gryllus bimaculatus ) and BMP2/4 from the yellow acorn worm ( Ptychodera flava ). However, support for this grouping is not statistically significant (reliability value = 27). A higher resolution of the dpp /BMP2/4 group may be achieved in the future by the aquisition of additional sequence information. Figure 1 Phylogenetic analysis of the Glomeris dpp fragment. The analysis included TGF-beta genes from mouse (Mm), zebrafish (Dr), lancet ( Branchiostoma floridae ; Bf), acorn worm ( Ptychodera flava ; Pf), the sea urchins Strongylocentrotus purpuratus (Sp) and Lytechinus variegatus (Lv), fruit fly (Dm), flour beetle ( Tribolium castaneum ; Tc), sawfly ( Athalia rosae ; Ar), buckeye butterfly ( Junonia coenia ; Jc), the grasshoppers Schistocerca americana (Sa) and S. gregaria (Sg), cricket ( Gryllus bimaculatus ; Gb), pill millipede ( Glomeris marginata ; Gm), and the spiders Achaearanea tepidariorum (At) and Cupiennius salei (Cs). Shown is the unrooted Puzzle tree computed from 1000 intermediate trees produced with the Quartet Puzzling method [41]. The numbers at the edges denote the reliability values. Expression of Gm-dpp during embryogenesis The expression of a number of developmental genes has been studied in the pill millipede Glomeris marginata [ 7 , 14 - 17 ]. Of all genes studied so far the expression of Gm-dpp is the weakest. A specific in situ hybridization signal is observed after approximately six hours of staining, whereas the normal staining interval of other genes ranges between 15 and 30 minutes. This extended staining time is responsible for the intense artificial background that is visible in the preparations displayed in this paper (see Figs. 2 , 3 ). Figure 2 Expression of Gm-dpp in G. marginata embryos. (A) stage 2. The arrow points to expression in the dorsal portion of the neuroectoderm. (B) stage 3. The arrows point to the dorsal and ventral (middle) portion of the neuroectoderm, respectively. (C) stage 3. Aspect of the head. The arrow points to expression in the brain. (D) stage 4. Arrow: expression in the optic lobe. Asterisk: expression in the antennal neuromere. Arrowheads: expression in the heart. (E) stage 5. Arrow: expression in the optic lobe. Arrowheads: expression in the heart. (F) stage 6.1. The arrowheads denote expression in the dorsal portion of the germ band that is probably correlated with heart formation. A-E are in ventral aspect. F is in lateral aspect. Abbreviations: md, mandible; mx, maxilla; an, antenna; t1, t2, t3, first three trunk legs. Figure 3 Expression of Gm-dpp during appendage development. (A, E, I) trunk legs. (B, F, J) maxilla. (C, G, K) mandible. (D, H, L) antenna). The arrows in A, E, I, D, H, L point to the distal expression domain in the trunk legs and the antenna, and denote the border of this domain against the ventral side of the appendage, where no Gm-dpp expression is detected. The arrows in B, F, G, K point to a ventral expression domain in the gnathal appendages. In all panels arrowheads denote expression along the dorsal side. The asterisk in K denotes expression in the external lobe. The asterisk in D denotes expression in the antennal neuromere at the base of the antenna. The asterisk in L denotes expression within the base of the antenna. The square in L is located next to a weak ring of expression in the antenna. Stages are as indicated in the top right corner the panels. Abbreviations: max, maxilla; mdb, mandible; ant, antenna. In younger stages, a specific staining is seen in the forming appendage buds and along the external, i.e. dorsal, rim of the neuroectoderm (Fig. 2A ; arrow). It is known that the neuroectoderm of each hemisegment is divided into a dorsal, medial and ventral portion [ 16 ]. Judging from its expression, it is possible that Gm-dpp has a role in the development of the dorsal portion of the neuroectoderm. A role in the developing ventral portion is also possible, since Gm-dpp is transiently expressed along the ventral midline (Fig. 2B ; arrows). A further expression domain in the central nervous system is seen in the area of the developing optic centers of the brain (Fig. 2C,2D,2E ; arrows). Starting with stage 4, Gm-dpp is expressed along the external rim of the germband in tissue that will later form the heart (Fig. 2D,2E ; arrowheads). Later on, segmentally repeated patches of weak Gm-dpp expression appear on the dorsal side of the embryos (Fig. 2F ; arrowheads). These patches are presumably also correlated with the developing heart of the embryos. Finally, expression of Gm-dpp is found in the stomodaeum, and very weakly in the proctodaeum. Expression profile of Gm-dpp during appendage development The appendages buds show weak expression of Gm-dpp at the very beginning of their formation (Fig. 2A,2B ). Later on, different appendages display appendage-specific expression patterns. In the trunk legs, the strongest expression is seen at the leg tips. In early developmental stages the expression fills almost the entire tip, and the border against the ventral portion of the legs (which is devoid of expression) is rather distinct (Fig. 3A ). There is also expression of Gm-dpp along the dorsal side of the trunk legs, but this is visibly weaker than the expression in the leg tips. The expression at the leg tips is clearly confined to the dorsal side of the tip in legs of stage 5 embryos (Fig. 3E ), while the expression along the dorsal side persists, but becomes weaker and diffuse. Finally, expression of Gm-dpp in the legs vanishes almost completely at stage 6 (Fig. 3I ). The dorsal expression is virtually undetectable, and only a few cells express Gm-dpp at the tip. In the maxilla there are two expression domains of Gm-dpp , a dorsal and a ventral one (Fig. 3B ). The ventral domain is located on the internal side at the base of the maxilla. This domain slowly vanishes during development (Fig. 3F ) and finally disappears around stage 6 (Fig. 3J ). The dorsal expression domain runs along the dorsal edge of the base of the maxilla (Fig. 3B,3F ). This domain also gradually disappears during development, and at stage 6.1 only a faint dorsal expression is detectable (Fig. 3J ). In the mandible, a dorsal expression domain that runs along the entire dorsal rim of the appendage is visible (Fig. 3C ). Later on, however, this expression is restricted to the basal portion of the mandible and has a distinct border against the external lobe (Fig. 3G ). Additional expression domains are detectable at later stages within the external lobe (Fig. 3K ; asterisk) in addition to the internal side of the internal lobe (Fig. 3G,3K ; arrow). In the antenna, Gm-dpp is expressed in the dorsal half of the appendage with a distinct border against the non-expressing ventral half (Fig. 3D ). In addition, a patch of weaker Gm-dpp expression is located on the ventral side of the antenna (Fig. 3D ; square). Another patch of Gm-dpp expression is visible at the transition between the antennal base and the neuroectoderm of the antennal neuromere (Fig. 3D ; asterisk). The latter two patches of expression disappear during the further course of development. By stage 5 the ventral spot has disappeared completely, and the patch at the antennal base is virtually gone as well (Fig. 3H ). Similar to the other appendages at stage 6.1, the level of Gm-dpp expression has also significantly decreased, though one can discern three specific expression domains at this stage. There are two groups of cells (at the tip and at the base of the antenna) weakly expressing Gm-dpp (Fig. 3L ; arrow and asterisk, respectively), and a ring of cells at the distal third of the antenna, where Gm-dpp expression is even weaker (Fig. 3L ; square). Expression profile of Gm-wg during appendage development The expression of Gm-wg during germ band segmentation, neurogenesis, and the development of the digestive system has already been described [ 14 ]. Here, I focus on Gm-wg expression during the development of the appendages. Before the onset of limb development, Gm-wg is expressed in a stripe in each hemisegment. This stripe is located approximately in the middle of the segment and runs across the neuroectoderm and the presumptive appendage tissue. Comparison to the expression of engrailed ( Gm-en ) has shown that the stripe of Gm-wg expression abuts the parasegment border and, thus, is located in cells of the anterior segmental compartment [ 14 ]. The buds of the appendages form from the tissue at the external ends of the Gm-wg stripes. Preparations of complete hemisegments of stage 3 embryos show, that expression of Gm-wg extends more or less contiguously across the neuroectoderm into the forming limb buds in all four different appendage types (Fig. 4A,4B,4C,4D ). The extent to which the expression reaches into the limb buds varies depending on the appendage type. In the antennal bud the expression of Gm-wg is restricted to the ventral side (Fig. 4D ). In the maxillary and mandibulary buds expression includes larger areas, approximately two thirds of the buds (Fig. 4B,4C ). Finally, expression is most extensive in the buds of the trunk legs: almost the entire buds express Gm-wg (Fig. 4A ). Figure 4 Expression of Gm-wg during appendage development. (A, E, I, M, Q) trunk leg. (B, F, J, N, R) maxilla. (C, G, K, O, S) mandible. (D, H, L, P, T) antenna. The arrowheads in A-D, and the arrows in E, I, M, Q and H, L, P, T point to the transition of ventral to dorsal tissue in the appendages. The arrows in A-D point to expression in the neuroectoderm of the respective body segment. The arrows in F, J, N, R denote the expression surrounding the maxillary sensory organs. The squares and asterisks in G, K, O, S denote expression of Gm-wg in the internal and external lobe, respectively. Stages are as indicated in the top right corner of the panels. Abbreviations see Fig. 3. In the trunk legs expression of Gm-wg is restricted to the ventral side during the further course of development (Fig. 4E,4I,4M,4Q ). The expression is contiguous from the base of the legs to the tips, but the level of expression is somewhat heterogeneous. The strongest expression is seen near the base and in the distal part of the legs, while expression is visibly weaker between these parts. A similar phenomenon is present in the antenna (Fig. 4H,4L,4P,4T ), where expression is restricted to the ventral side of the antenna and the level of expression at the distal end is much stronger than in more proximal parts. However, unlike the pattern in the legs, the intensity of expression at the base of the antenna is not increased. The maxilla displays a rather dynamic expression profile of Gm-wg . Beginning at stage 4 the gene is expressed along the ventral edge of the maxilla (Fig. 4F ). Three domains can be distinguished that are not completely separated. The innermost domain is more diffuse than the other two domains and at stage 5 separates into two separate patches of expression (Fig. 4J,4N,4R ; two-headed arrow). The two other domains remain separate during the development of the maxillary appendage and are reminiscent of the expression pattern of Gm-Dll (see below). In the mandible, a similar fragmentation of the initial mostly homogeneous expression pattern takes place. In the external lobe the expression is strong throughout and is separated from the expression domain in the internal lobe by an area of very weak expression (Fig. 4G,4K,4O,4S ). The expression domain in the internal lobe splits (Fig. 4K ), then retracts from the tip of the lobe (Fig. 4O ) and decreases in expression strength (Fig. 4S ). Expression of Gm-wg and Gm-Dll in the gnathal sensory organs As mentioned above, the expression pattern of Gm-wg in the maxilla is reminiscent of the pattern described for Gm-Dll [ 7 ], and at first glance both patterns appear virtually identical. The Gm-Dll gene is expressed in the primordia of the maxillary sensory organs. Expression of the Gm-wg gene, however, is at least partially complementary to the pattern of Gm-Dll . In older stages, strong expression of Gm-wg is not detected within the primordia of the sensory organs, but rather it surrounds the primordia (Fig. 4R ). In preparations simultaneously labeled with probes against Gm-wg and Gm-Dll the composite expression pattern stains the entire internal side of the maxilla (Fig. 5A,5C ), indicating that both patterns complement each other. However, the maxillary expression of Gm-wg in three (later four) domains is more extensive than the more restricted expression pattern of Gm-Dll in two (later three) well-defined stripes (see [ 7 ]). Therefore, the expression patterns are certainly not mutually exclusive. The presumed overlap of the two expression patterns, however, cannot be detected with the double labeling technique used here. Figure 5 The relation between the expression of Gm-wg and Gm-Dll . Preparations of maxillae (A, C) and mandibles (B, D) simultaneously labeled with a mixture of probes against Gm-wg and Gm-Dll . In the maxilla the patterns complement each other to stain the entire ventral edge of the appendage, whereas in the mandible no significant difference is observed compared to Gm-wg expression detected alone. Compare to Fig. 4. Stages are as indicated in the top left corner of the panels. Abbreviations see Fig. 3. In addition, a complex and dynamic pattern of Gm-Dll has been described in the mandible [ 7 ]. In contrast to the maxilla, the patterns of Gm-wg and Gm-Dll appear to overlap completely in the mandible. In preparations of mandibles labeled with a cocktail of probes against both genes no significant difference to the pattern of Gm-wg alone is observed (Fig. 5B,5D ), indicating that the Gm-Dll pattern is entirely included in the Gm-wg pattern. Discussion Establishment of the primary PD axis In Drosophila dpp is expressed in a narrow dorsal sector in the leg imaginal discs, whereas wg is expressed in a similar sector on the ventral side (e.g. [ 1 ]). Together these two genes generate morphogen gradients in the developing leg imaginal discs. These gradients are utilized by several genes to guide the development of the PD axis of the leg imaginal discs. Evolutionary developmental studies have shown that the expression of wg homologs along the ventral side of the appendages is highly conserved in the arthropods (e.g. [ 6 , 9 , 10 , 13 ]). In contrast, dpp expression differs from the expression pattern found in Drosophila in all arthropods studied thus far (e.g. [ 6 , 9 - 12 , 18 , 19 ]). At early stages expression of arthropod dpp homologs is restricted to the leg tip, while at later stages expression rings of unclear significance appear in some species. Despite these differences in expression, it has been argued that the combined action of the WG and DPP morphogen gradients is conserved, and that the differences in expression of dpp are correlated with the differences in the mode of leg development between Drosophila (via imaginal discs) and most other arthropods (normal leg outgrowth) [ 6 ]. The data from Glomeris presented here may be interpreted in favor of this hypothesis. The Gm-wg gene is expressed along the ventral side in the legs and Gm-dpp is expressed most strongly in the leg tips. Taking these expression loci as the sources of Gm -WG and Gm -DPP protein, the resulting hypothetical protein gradients would facilitate PD patterning events similar to the ones in the Drosophila leg discs (see also Fig. 11 in [ 6 ]). However, Gm-dpp is weakly expressed along the dorsal leg side. This is similar to the Drosophila situation, but is contrary to the predictions of the above hypothesis since Glomeris does not develop the legs via flat imaginal discs and therefore should show a dpp expression pattern typical of directly developing legs rather than a pattern similar to Drosophila . The fact that Gm-dpp is also weakly expressed along the dorsal side of the legs may be explained by several possibilities. It may be argued that the dorsal expression is so weak that it has no significant influence on the shape of the Gm -DPP protein gradient, which would therefore mainly be dependent on the morphogen source at the tip. It is also possible that the dorsal expression is unrelated to PD axis formation and instead functions during DV axis formation (see below). In any case, the picture emerging from the available data on dpp expression in arthropods is that the dorsal sector in Drosophila seems to be an exception rather than the rule. The hypothesis proposed by Prpic et al. [ 6 ] attempts to explain this by the differences in leg architecture between Drosophila and most other arthropod species. However, according to their hypothesis, the presence of combinatorial protein gradients is conserved. It should be pointed out in this context that the existence of a DPP gradient (or a WG gradient for that matter) has yet to be demonstrated in an arthropod other than Drosophila . Thus, although the expression data may be interpreted as the PD axis patterning using WG/DPP signaling being conserved among arthropods, it is obvious that comparative expression analyses alone cannot answer the question satisfactorily. It must now be considered whether experiments capable of demonstrating WG/DPP signaling during leg development in arthropods other than Drosophila may be conceived. Establishment of secondary PD axes Aside from the primary PD axis, many arthropods have limbs with additional branches (rami). It has been proposed that these additional rami are patterned in the same way as the main branch, simply by duplications of the WG/DPP signaling system [ 20 ]. Recent results from the study of insect mouthparts argue against this notion [ 11 ]. The insect labium and maxilla have ventral branches (endites) that apparently do not utilize a combinatorial WG/DPP gradient system to guide their outgrowth. A similar conclusion has been reached by a study of the development of crustacean multibranched appendages [ 13 ]. The presence of endites in the mouthparts of myriapods is unclear, mainly because of the modified morphology of the adult gnathalia. Certain elements of the centipede mandible and first maxilla are probably derived from endites (e.g. [ 21 , 22 ]) and there are attempts to assign parts of the diplopod mandible as homologous to crustacean or insect endites (e.g. [ 21 , 23 ]). Indeed, the embryonic mandible and maxilla in Glomeris develop ventral lobes that are very reminiscent of the endite lobes of the embryonic mouthparts in insects. The exclusive ventral origin of these lobes is further corroborated by the lack of expression of the dorsal marker optomotor-blind [ 17 ]. Furthermore, the Glomeris lobes possess Dll -positive sensory organs, which is typical of arthropod endites [ 7 , 24 - 26 ]. Thus, although the interpretation of the millipede mouthparts is disputed (see e.g. [ 27 ]), these ventral lobes are likely homologous to the endites present in insect mouthparts. The embryonic Glomeris mandible develops two lobes, the internal and external mandibular lobe. Both lobes express Gm-dpp , but not at a position suggestive of a role in PD axis formation (Fig. 6C ). In addition, the expression domain in the external mandibular lobe appears after the lobe has already grown and is therefore unlikely to be involved in PD outgrowth. The development of the maxillae is more complex. They start out as separate appendages, but around stage 6 the left and the right maxilla fuse along their internal sides (Fig. 6A,6B ). Each maxilla has a single lobe, containing the primordia of three sensory organs that can also be visualized by Gm-Dll expression [ 7 ]. The two external sensory organ primordia form the lobus medius and the lobus exterior in the adult (see [ 27 ] for a detailed description of Glomeris maxillary morphology). These two sensory structures grow from the internal side of the stipes (Fig. 6A ). The internal sensory organ primordium is different from the other two in the sense that it will not end up on the stipes, but will form the lobus interior that grows from the lamella lingualis (Fig. 6A ). The lamellae linguales of the right and left maxilla fuse around stage 6 to form the intermaxillary plate (Fig. 6B ). The single lobe in the Glomeris embryonic maxilla therefore has a rather complex fate in the adult mouthpart (the gnathochilarium [ 27 ]). The portion of the maxillary lobe that will give rise to the stipes expresses Gm-dpp at only later stages when the expression along the dorsal side of the maxillary base is extending weakly into the stipes. The portion forming the lamella lingualis also expresses Gm-dpp , but at its internal edge, a location hardly suggesting a role in PD outgrowth of the maxillary lobe. Figure 6 Possible endite homologs in the mouthparts of Glomeris . Schematic representations of the mouthparts of Glomeris (A-C). An insect mouthpart (maxilla of Schistocerca ) is shown for comparison (D). (A) Glomeris maxilla. (B) Left and right maxilla of Glomeris already fused forming the gnathochilarium. (C) Glomeris mandible. The palp is proposed to be lost in Glomeris gnathalia ([7]; grey hatched line). The black hatched line indicates where the appendage inserts on the segment. Expression of dpp is shown in grey (hatched area in the maxilla: very weak expression). Please note that the figure shows all observed expression domains in a single drawing although really some domains appear at different time points (see text for a description of the temporal expression profile). None of the possible endite homologs in Glomeris mouthparts (lli, st, iml, and eml) expresses Gm-dpp in a fashion suggestive of a role in PD axis patterning. See text for details. Expression of dpp in Schistocerca is after [11]. Abbreviations: bs, base; ca, cardo; eml, external mandibular lobe; ga, galea; iml, internal mandibular lobe; lc, lacinia; le, lobus exterior; li, lobus interior; lli, lamella lingualis; lm, lobus medius; plp, palp; st, stipes. In summary, none of the maxillary and mandibulary lobes in Glomeris appear to utilize conventional WG/DPP signaling to organize PD growth. Similar results have been obtained recently for the endites in the grasshopper Schistocerca and the beetle Tribolium [ 11 ]. In Schistocerca at least one endite (the galea) grows without dpp expression (Fig. 6D ), and in Tribolium both maxillary endites lack detectable dpp expression [ 11 ]. This indicates that the development of the PD axis of the endites does not generally require the WG/DPP morphogen system. Relation of wingless and dpp expression to DV axis formation A second role of wg and dpp in Drosophila is the activation of some factors involved in DV axis formation in the legs [ 28 , 29 ]. wg , being expressed along the ventral side, is an instructor of ventral fate, whereas dpp is expressed on the dorsal side and establishes dorsal fates. The primary factors controlled by wg and dpp are H15 on the ventral side and omb on the dorsal side. These factors have been recently studied in Glomeris and in a spider ( Cupiennius salei ) [ 6 , 17 ]. The expression patterns suggest that the role of omb as dorsal instructor is evolutionarily conserved, but H15 does not seem to be a general ventralizing factor in all arthropods. Thus, the dorsal, but not the ventral developmental mechanisms seem to be conserved. It is interesting that the expression data of wg and dpp suggest that the opposite is true. The wg expression on the ventral side is highly conserved among the arthropods, but the dpp patterns differ between species and in most part expression is not localized to the entire dorsal side. This paradox clearly demonstrates the limited understanding of the evolution of DV axis formation in arthropod appendages. Patterning of appendicular sensory organs The maxilla of Glomeris has several sensory organs. Recent studies have identified the genes Dll , dac and H15 , which show a restricted expression pattern in the primordia of the maxillary sensory organs [ 7 , 17 ]. Two of these genes, Dll and H15 , are known from Drosophila to be activated upon signaling through the wingless pathway [ 29 , 30 ]. It is interesting to note that expression of Gm-wg surrounds the sensory primordia in the Glomeris maxilla. It may therefore be the case that cells expressing Gm-wg in the surrounding of the primordia signal to their neighbors within the primordia and stimulate them to activate Gm-Dll and Gm-H15 . Minimally the activation of Dll appears to be a general feature of appendicular sensory organs in arthropods since Dll expression has been observed in appendicular sensory organs in chelicerates, crustaceans, myriapods and insects (e.g. [ 7 , 24 , 25 , 31 ]). Moreover, data from Drosophila suggest that Dll expression is critically required for sensory organ formation, as mutants lacking Dll fail to develop Keilin's organs (the sensory structures of the embryonic leg anlagen) [ 32 , 33 ]. Conclusions The expression of Gm-wg and Gm-dpp during appendage development indicates a role for both genes in guiding this process. Involvement of wg and dpp in appendage development appears to be conserved among all extant arthropod classes including myriapods. The data from Glomeris and other arthropods suggest that the WG/DPP morphogen signaling system as it is known from Drosophila leg discs is present in all arthropods. However, this morphogen system apparently functions in only the main branch of the appendages, the so-called telopodite [ 34 ]. Limb types with additional branches (e.g. endites) obviously use additional, yet unidentified mechanisms to organize proximal-distal growth of the extra branches. Gene expression in potential endite homologs present in Glomeris mouthparts supports this notion. Aside from the role in PD axis formation, the expression profile of Gm-wg suggests an additional role for this gene in patterning appendicular sensory organs. Methods Animal stocks Animals were collected during Spring 2003 in beech forests in the vicinity of Cologne, Germany and near Kranenburg, Germany. They have been treated as described before [ 6 , 14 ]. The animals were released after the end of the breeding season (Summer '03). Molecular cloning The cloning assays were based on cDNA transcribed from polyA-RNA extracted from selected Glomeris embryos of all developmental stages up to stage 6.1 (see [ 14 , 35 ] for a description of embryonic stages) and were performed in duplicate. For the amplification of dpp -like gene fragments, the primers dpp-fw-1 (GAY GTN GGN TGG GAY GAY TGG) and dpp-bw-1 (CKR CAN CCR CAN CCN CAN AC) were used in the initial PCR, and the primers dpp-fw-2 (GGN TAY GAY GCN TAY TAY TG) and dpp-bw-1 were used in the nested PCR. Additional sequence information was gained by RACE PCR. No full-length fragment could be obtained and several artificial clones were encountered, probably representing chimeric products resulting from jumping PCR between different TGF-beta-like cDNAs. Using species specific primers, artificial and genuine fragments were identified. A confirmed genuine fragment of almost 360 bp was isolated and cloned. This fragment was used for sequence analysis and probe synthesis. The isolation of Gm-wg has been previously reported [ 14 ]. The GenBank accession numbers are as follows: Gm-wg (AJ616907); Gm-dpp (AJ843875). Alignments and sequence analysis Pairwise alignments of aminoacid sequences were performed by searching GenBank [ 36 ] using the Gapped BLAST program [ 37 ]. The alignments were calculated based on the BLOSUM 62 matrix [ 38 ] (gap costs: 11 for opening, 1 for extension). Multiple sequence alignments were calculated based on the GONNET matrix [ 39 ] (gap costs: 10 for opening, 0.2 for extension) implemented in CLUSTAL_X [ 40 ]. The resulting alignments were subjected to maximum likelihood analysis using the Quartet Puzzling method [ 41 ] as implemented in PAUP* 4.0b10 [ 42 ]. In situ hybridizations, specimen preparation and microscopy In situ hybridization with digoxigenin-labeled RNA probes has been performed as previously described [ 7 ]. Whole-mount embryos were photographed in PBST under a Leica dissection microscope. Appendages were dissected with fine insect needles and photographed in 50% glycerol under a Zeiss Axioplan microscope. All images were corrected for color values, brightness and contrast using Adobe Photoshop 5.5 for Apple Macintosh. The image processing software has also been used to enhance image backgrounds by retouching dirt or yolk remains, and to group together single pictures into multipanel figures.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC544938.xml

499542

Significant receptor affinities of metabolites and a degradation product of mometasone furoate

Mometasone furoate (MF) is a highly potent glucocorticoid used topically to treat inflammation in the lung, nose and on the skin. However, so far no information has been published on the human glucocorticoid receptor activity of the metabolites or degradation products of MF. We have now determined the relative receptor binding affinities of the known metabolite 6β-OH MF and the degradation product 9,11-epoxy MF to understand their possible contribution to undesirable systemic side effects. In competition experiments with human lung glucocorticoid receptors we have determined the relative receptor affinities (RRA) of these substances with reference to dexamethasone (RRA = 100). We have discovered that 6β-OH MF and 9,11-epoxy MF display RRAs of 206 ± 15 and 220 ± 22, respectively. This level of activity is similar to that of the clinically used inhaled corticosteroid flunisolide (RRA 180 ± 11). Furthermore we observed that 9,11-epoxy MF is a chemically reactive metabolite. In recovery experiments with human plasma and lung tissue we found a time dependent decrease in extractability of the compound. Hence, we provide data that might contribute to the understanding of the pharmacokinetics as well as the clinical effects of MF.

Introduction Mometasone furoate (MF) is a highly potent topical glucocorticoid for the treatment of asthma [ 1 ], allergic rhinitis [ 2 ] and various skin diseases [ 3 ]. The clinical efficacy of MF is comparable to that of fluticasone propionate [ 4 ]. Both compounds have a very high affinity to the human glucocorticoid receptor. With reference to dexamethasone, fluticasone propionate has an eighteen-fold higher relative receptor affinity (RRA) of 1800 [ 5 , 6 ], while MF displays a RRA of about 2200 [ 7 ]. These high receptor affinities as well as the administered doses, the absolute lung deposition and a prolonged retention time in the lung tissue contribute to the clinical success of both compounds. Besides the efficacy of a corticosteroid, safety issues have to be taken into consideration. For topically applied glucocorticoids, the high local anti-inflammatory activity should be paralleled by a low systemic exposure. Therefore, a prolonged redistribution from lung tissue into systemic circulation and a rapid and complete hepatic metabolism of the compounds to inactive derivatives are favorable. For MF, a very low systemic bioavailability of less than 1 % has been reported [ 8 ]. However, there have been discussions about the appropriate methodology and the validity of the conclusion has been questioned [ 9 , 10 ]. Indeed, the claimed low systemic bioavailability of MF would appear to be inconsistent with the considerable suppression of the hypothalamic-pituitary-adrenal (HPA) axis recorded in a clinical study [ 11 , 12 ]. Frequently, various researchers called attention to the formation of active MF metabolites that would account for undesirable systemic side effects [ 9 , 13 ]. In an early study by Isogai et al. more than ten different metabolites and related compounds of MF displayed varying binding affinities to the rat glucocorticoid receptor [ 14 ]. There had been, however, not much information on the extent and site of metabolite formation in humans. Recent studies now provided some of the required information [ 7 , 13 , 15 , 16 ]. In rat liver microsomes, 6β-hydroxy MF (6β-OH MF) was identified as the major metabolite [ 16 ]. This metabolite was also found after incubation of MF with human liver and intestine microsomes [ 13 ]. Additionally, the degradation product 9,11-epoxy MF was detected in plasma and urine. 9,11-epoxy MF is formed in aqueous solutions [ 15 ] indicating a general time- and pH-dependent instability of MF [ 7 ]. Recently, we discovered 9,11-epoxy MF in incubation mixtures of human lung tissue as well as in fresh human plasma [ 7 ]. We pointed out that this degradation product might form covalent adducts with proteins in follow-up reactions. Despite the recent discovery of the major metabolite 6β-OH MF and the abundant degradation product 9,11-epoxy MF it is still not clear whether these compounds retain any significant binding affinity to the human glucocorticoid receptor. In the present study we addressed this open question and we present some evidence that the degradation product might bind tightly, most possibly covalently, to protein structures in human lung tissue and plasma. Materials and Methods Chemicals and reagents Mometasone furoate (MF), 6-hydroxy mometasone furoate (6-OH MF), mometasone and 9,11-epoxy mometasone furoate (9,11-epoxy MF) were generous gifts from GlaxoSmithKline (Greenford, England). [ 3 H]-Dexamethasone was obtained from Amersham (Freiburg, Germany). All other chemicals were obtained from Sigma-Aldrich-Chemie (Taufkirchen, Germany) or E. Merck (Darmstadt, Germany). Source and handling of human specimen Human lung tissue resection material was obtained from patients with bronchial carcinomas who gave informed consent. Cancer-free tissue was used for the experiments. None of the patients was treated with glucocorticoids for the last 4 weeks prior to surgery. Tissue samples were shock frozen in liquid nitrogen after resection and stored at -70°C until usage. To collect sufficient material for the experiments, tissue samples of three or more patients were pooled. Lung cytosol for receptor competition experiments was prepared as detailed in [ 6 ]. Plasma samples were obtained from healthy volunteers who gave informed consent. Samples were either used immediately or were shock frozen in liquid nitrogen and stored at -70°C until usage. Determination of relative receptor affinity by competition tests The competition experiments were performed according to the procedure described earlier [ 6 ]. The displacement of a constant concentration of [ 3 H] labelled dexamethasone by various concentrations of 6-OH MF, mometasone and 9,11-epoxy MF was determined. Recovery of MF and 9,11-epoxy MF from human plasma, lung tissue and buffer MF or 9,11-epoxy MF, respectively, were added to human plasma, lung tissue suspension (0.5 g / 20 ml) or buffer (0.2 M phosphate buffer, pH 7.4) yielding an initial concentration of 0.3 μg/ml. Only glass lab ware was used for these experiments to exclude any non-specific binding effects of the highly lipophilic compounds to plastic material. Samples were incubated at 37°C in a shaking water bath. At designated time intervals samples of 1.0 ml were removed, subjected to a fluid extraction with diethylether and analyzed by HPLC. Sample preparation and HPLC conditions Samples were prepared and analyzed as described previously [ 7 ]. The HPLC system consisted of a Waters HPLC (Milford, MA) with a 1525 binary pump, a 717plus autosampler and 2487 dual wavelength absorbance detector set at the detection wavelength of 254 nm. Data collection and integration were accomplished using Breeze™ software version 3.2. Analysis was performed on a Symmetry C 18 column (150 × 4.6 mm I.D., 5 μm particle size, Waters, MA). Results We determined the relative receptor affinities (RRAs) of 6β-OH MF, 9,11-epoxy MF and mometasone base by competition assays with reference to dexamethasone (RRA = 100). Both, the metabolite 6β-OH MF and the degradation product 9,11-epoxy MF displayed residual receptor binding affinities about twice as high as dexamethasone (Table 1 ). This level of activity is between that of the clinically used inhaled corticosteroids flunisolide (RRA 180 ± 11) and triamcinolone acetonide (RRA 361 ± 26) [ 5 ]. Mometasone which is formed by hydrolysis of the furoate ester, revealed an even higher RRA of almost 800. For comparison, the RRA of the parent compound MF is about 2200 [ 7 ]. Table 1 Relative receptor affinities of mometasone furoate (MF, data from [7]), its metabolites 6β-hydroxy mometasone furoate (6β-OH MF), mometasone and the major degradation product 9,11-epoxy mometasone furoate (9,11-epoxy MF) in relation to dexamethasone (Dexa). Values represent mean and mean deviation of the mean of n = 3 independent experiments. Compound Relative receptor affinity (RRA) Mean deviation of the mean MF 2244 ± 142 Dexa 100 ± 10 6β-OH MF 206 ± 15 9,11-epoxy MF 220 ± 22 Mometasone 781 ± 27 To investigate the putative reactivity of the degradation product 9,11-epoxy MF we monitored the recovery of MF and 9,11-epoxy MF from human plasma by organic solvent extraction (Fig. 1 ). The determination of recovery was limited to a period of three hours since MF is successively degraded to 9,11-epoxy MF [ 7 ]. The retrieval of 9,11-epoxy MF from human plasma decreased steadily and was clearly more pronounced than for MF. After three hours 9.14 ± 2.3 % of 9,11-epoxy MF was not recovered from plasma while 4.8 ± 1.4 % of MF was not extractable any more. Figure 1 Recovery of mometasone furoate (MF) and its degradation product 9,11-epoxy MF from incubation mixtures with human plasma over three hours. Each data point represents the mean and mean deviation of the mean of three experiments. The decrease in recovery of 9,11-epoxy MF from human lung tissue was even more evident (Fig. 2 ). While there was no change in the control incubation mixture comprising of buffer (pH 7.4) a pronounced and steady decrease in recovery rates of 9,11-epoxy MF was revealed. After three hours 16.61 ± 0.58 % of the degradation product was not extractable any more. No new peaks were observed in the HPLC to indicate a further degradation of 9,11-epoxy MF. Figure 2 Recovery of 9,11-epoxy MF from incubation mixtures with human lung tissue and buffer (control experiment) over three hours. Each data point represents the mean and mean deviation of the mean of three experiments. Discussion In the present study we have determined the relative receptor binding affinities of the mometasone furoate (MF) metabolite 6β-OH MF and its degradation product 9,11-epoxy MF to understand their possible contribution to undesirable systemic side effects. For the first time we provide data that both compounds are significantly active at the human glucocorticoid receptor with binding affinities twice as high as dexamethasone and similar to that of the clinically used inhaled corticosteroids flunisolide and triamcinolone acetonide [ 5 ]. Furthermore, our data demonstrate that the ubiquitous degradation product 9,11-epoxy MF undergoes follow-up reactions. Glucocorticoids currently used for topical application in asthma therapy all share the safety relevant property of extensive metabolism and formation of inactive metabolites. For MF, however, data was sparse so far. Though putative metabolites and degradation products with binding affinity to the rat glucocorticoid receptor have been previously suggested [ 14 ], it was not clear whether this might have any implications to humans. Potential human metabolites such as 6β-OH MF, mometasone or MF-epoxide have been proposed [ 8 ], but experimental evidence of in vivo formation of these compounds was still lacking. Studies of Teng et al. identified 6β-OH MF and 9,11-epoxy MF as candidate compounds that can indeed be formed in vivo either by hepatic metabolism or by simple degradation of MF [ 13 , 16 ]. We discovered that 9,11-epoxy MF is also formed in human lung tissue suspensions and plasma [ 7 ]. Usually hydroxylation at the 6β position results in inactivation of the corticosteroid. The 6-OH metabolite of various glucocorticoids displays little or no residual binding affinity to the receptor (e.g.) [ 17 , 18 ]. This, however, is different for MF with its 6β-OH metabolite exhibiting a relative receptor affinity of more than 200 (dexamethasone: 100). Obviously, the substitution pattern of the D-ring of MF confers such potent binding affinity that hydroxylation in 6β position does not result in complete inactivation of this corticosteroid. Notably, neither the RRA we determined for 6β-OH MF nor for mometasone are coherent with the binding results of the early studies with the rat glucocorticoid receptors [ 14 ]. This emphasizes the need for data derived from human receptor studies. The MF degradation product 9,11-epoxy MF also displays a significant receptor binding affinity with an RRA of about 200. This RRA is within the range that could be expected from the studies of Isogai et al . [ 14 ]. Since 9,11-epoxy MF is also formed in the lung tissue suspensions [ 7 ], it can be assumed that it contributes to the effects after inhalation of MF. It can, however, be predicted that this compound might be also responsible for undesired effects such as HPA axis suppression. Besides the significant residual receptor binding affinity of 9,11-epoxy MF we discovered that this compound undergoes follow-up reactions. After incubation with plasma clearly less of 9,11-epoxy MF compared to the parent compound MF was recovered by extraction with an organic solvent. This extraction procedure usually reliably retrieves all non-covalently bound substance from the incubation mixture. In human lung tissue, it was even more obvious that 9,11-epoxy MF was recovered completely from buffer, but not from the tissue suspension. About 17 % of 9,11-epoxy MF was "lost" after three hours of incubation. This observation cannot be explained by simple non-specific tissue binding since the tissue adsorption reaches equilibrium very quickly after about 20 min [ 7 ]. Also, the non-specifically bound compound would be still extractable by organic solvents. Generally, epoxides are chemically reactive molecules that tend to bind irreversibly to cellular macromolecules. If this were the case for 9,11-epoxy MF it would have two implications. Firstly, irreversibly bound 9,11-epoxy MF escapes detection and feigns a low bioavailability after inhalation. The fact that after inhalation of a single dose of tritium labelled MF only 88% (63–99 %) of total radioactivity was recovered over seven days in humans [ 8 ] seems to support this conclusion. Secondly, if 9,11-epoxy MF is indeed covalently bound to cellular macromolecules the adduct might lead to allergic reactions. Such reactions to corticosteroids for asthma therapy do occur occasionally [ 19 ]. However, it cannot be excluded that 9,11-epoxy MF is further degraded although we did not observe any new peaks that emerged in the HPLC chromatograms. The chromatographic conditions were chosen for rather lipophilic compounds, thus, if a further degradation product of 9,11-epoxy MF with pronounced hydrophilic character was formed, it might have escaped our attention. However, the possibility of covalent adduct formation of 9,11-epoxy MF should be further investigated. Conclusions In contrast to other inhaled corticosteroids MF generates an active metabolite, 6β-OH MF, in the liver. The degradation product 9,11-epoxy MF, which is formed in human lung tissue and plasma, exhibits significant receptor affinity as well. Additionally, we found that 9,11-epoxy MF undergoes follow-up reactions. Our data contribute to the understanding of how the claimed low bioavailability of MF parent compound after inhalation might still be accompanied by HPA axis suppression. Thus, our findings are consistent with both pharmacokinetic and clinical data. We strongly suggest a clinical trial that determines both efficacy and safety in parallel as well as all known metabolites and degradation products after application of MF. Authors' contributions AV carried out all experiments and the data analysis and participated in the design of the study. PH conceived of and designed the study and wrote the manuscript. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC499542.xml

212702

What Controls Variation in Human Skin Color?

There is a large range of human skin color, yet we know very little about the underlying genetic architecture. Is the number of skin color genes close to five, 50, or 500?

Diversity of human appearance and form has intrigued biologists for centuries, but nearly 100 years after the term “genetics” was coined by William Bateson in 1906, the genes that underlie this diversity are an unsolved mystery. One of the most obvious phenotypes that distinguish members of our species, differences in skin pigmentation, is also one of the most enigmatic. There is a tremendous range of human skin color in which variation can be correlated with climates, continents, and/or cultures, yet we know very little about the underlying genetic architecture. Is the number of common skin color genes closer to five, 50, or 500? Do gain- and loss-of-function alleles for a small set of genes give rise to phenotypes at opposite ends of the pigmentary spectrum? Has the effect of natural selection on similar pigmentation phenotypes proceeded independently via similar pathways? And, finally, should we care about the genetics of human pigmentation if it is only skin-deep? Why Should We Care? From a clinical perspective, inadequate protection from sunlight has a major impact on human health ( Armstrong et al. 1997 ; Diepgen and Mahler 2002 ). In Australia, the lifetime cumulative incidence of skin cancer approaches 50%, yet the oxymoronic “smart tanning” industry continues to grow, and there is controversy over the extent to which different types of melanin can influence susceptibility to ultraviolet (UV) radiation ( Schmitz et al. 1995 ; Wenczl et al. 1998 ). At the other end of the spectrum, inadequate exposure to sunlight, leading to vitamin D deficiency and rickets, has been mostly cured by nutritional advances made in the early 1900s. In both cases, understanding the genetic architecture of human skin color is likely to provide a greater appreciation of underlying biological mechanisms, much in the same way that mutational hotspots in the gene TP53 have helped to educate society about the risks of tobacco ( Takahashi et al. 1989 ; Toyooka et al. 2003 ). From a basic science perspective, variation in human skin color represents an unparalleled opportunity for cell biologists, geneticists, and anthropologists to learn more about the biogenesis and movement of subcellular organelles, to better characterize the relationship between genotypic and phenotypic diversity, to further investigate human origins, and to understand how recent human evolution may have been shaped by natural selection. The Color Variation Toolbox Historically, measurement of human skin color is often based on subjective categories, e.g., “moderate brown, rarely burns, tans very easily.” More recently, quantitative methods based on reflectance spectrophotometry have been applied, which allow reddening caused by inflammation and increased hemoglobin to be distinguished from darkening caused by increased melanin ( Alaluf et al. 2002b ; Shriver and Parra 2000 ; Wagner et al. 2002 ). Melanin itself is an organic polymer built from oxidative tyrosine derivatives and comes in two types, a cysteine-rich red–yellow form known as pheomelanin and a less-soluble black--brown form known as eumelanin ( Figure 1A ). Discriminating among pigment types in biological samples requires chemical extraction, but is worth the effort, since the little we do know about common variation in human pigmentation involves pigment type-switching. The characteristic phenotype of fair skin, freckling, and carrot-red hair is associated with large amounts of pheomelanin and small amounts of eumelanin and is caused by loss-of-function alleles in a single gene, the melanocortin 1 receptor (MC1R) ( Sturm et al. 1998 ; Rees 2000 ) However, MC1R variation has a significant effect on pigmentation only in populations where red hair and fair skin are common ( Rana et al. 1999 ; Harding et al. 2000 ), and its primary effects—to promote eumelanin synthesis at the expense of pheomelanin synthesis, or vice versa— contribute little to variation of skin reflectance among or between major ethnic groups ( Alaluf et al. 2002a ). Figure 1 Biochemistry and Histology of Different Skin Types (A) Activation of the melanocortin 1 receptor (MC1R) promotes the synthesis of eumelanin at the expense of pheomelanin, although oxidation of tyrosine by tyrosinase (TYR) is required for synthesis of both pigment types. The membrane-associated transport protein (MATP) and the pink-eyed dilution protein (P) are melanosomal membrane components that contribute to the extent of pigment synthesis within melanosomes. (B) There is a gradient of melanosome size and number in dark, intermediate, and light skin; in addition, melanosomes of dark skin are more widely dispersed. This diagram is based on one published by Sturm et al. (1998) and summarizes data from Szabo et al. (1969) , Toda et al. (1972) , and Konrad and Wolff (1973) based on individuals whose recent ancestors were from Africa, Asia, or Europe. More important than the ratio of melanin types is the total amount of melanin produced. In addition, histological characteristics of different-colored skin provide some clues as to cellular mechanisms that are likely to drive pigmentary variation ( Figure 1B ). For the same body region, light- and dark-skinned individuals have similar numbers of melanocytes (there is considerable variation between different body regions), but pigment-containing organelles, called melanosomes, are larger, more numerous, and more pigmented in dark compared to intermediate compared to light skin, corresponding to individuals whose recent ancestors were from Africa, Asia, or Europe, respectively ( Szabo et al. 1969 ; Toda et al. 1972 ; Konrad and Wolff 1973 ). From these perspectives, oxidative enzymes like tyrosinase (TYR), which catalyzes the formation of dopaquinone from tyrosine, or melanosomal membrane components like the pink-eyed dilution protein (P) or the membrane-associated transporter protein (MATP), which affect substrate availability and activity of TYR ( Orlow and Brilliant 1999 ; Brilliant and Gardner 2001 ; Newton et al. 2001 ; Costin et al. 2003 ), are logical candidates upon which genetic variation could contribute to the diversity of human skin color. Of equal importance to what happens inside melanocytes is what happens outside. Each pigment cell actively transfers its melanosomes to about 40 basal keratinocytes; ultimately, skin reflectance is determined by the amount and distribution of pigment granules within keratinocytes rather than melanocytes. In general, melanosomes of African skin are larger and dispersed more widely than in Asian or European skin ( Figure 1 ). Remarkably, keratinocytes from dark skin cocultured with melanocytes from light skin give rise to a melanosome distribution pattern characteristic of dark skin, and vice versa ( Minwalla et al. 2001 ). Thus, at least one component of skin color variation represents a gene or genes whose expression and action affect the pigment cell environment rather than the pigment cell itself. Genetics of Skin Color For any quantitative trait with multiple contributing factors, the most important questions are the overall heritability, the number of genes likely to be involved, and the best strategies for identifying those genes. For skin color, the broad sense heritability (defined as the overall effect of genetic vs. nongenetic factors) is very high ( Clark et al. 1981 ), provided one is able to control for the most important nongenetic factor, exposure to sunlight. Statements regarding the number of human skin color genes are attributed to several studies; one of the most complete is by Harrison and Owen (1964) . In that study, skin reflectance measurements were obtained from 70 residents of Liverpool whose parents, grandparents, or both were of European (“with a large Irish component”) or West African (“mostly from coastal regions of Ghana and Nigeria”) descent and who were roughly classified into “hybrid” and “backcross” groups on this basis. An attempt to partition and analyze the variance of the backcross groups led to minimal estimates of three to four “effective factors,” in this case, independently segregating genes. Aside from the key word minimal (Harrison and Owen's data could also be explained by 30–40 genes), one of the more interesting findings was that skin reflectance appeared to be mainly additive. In other words, mean skin reflectance of “F1 hybrid” or “backcross hybrid” groups is intermediate between their respective parental groups. An alternative approach for considering the number of potential human pigmentation genes is based on mouse coat color genetics, one of the original models to define and study gene action and interaction, for which nearly 100 different genes have been recognized ( Bennett and Lamoreux 2003 ; Jackson 1994 ). Setting aside mouse mutations that cause white spotting or predominant effects outside the pigmentary system, no more than 15 or 20 mutations remain, many of which have been identified and characterized, and most of which have human homologs in which null mutations cause albinism. This brings us to the question of candidate genes for skin color, since, like any quantitative trait, a reasonable place to start is with rare mutations known to cause an extreme phenotype, in this case Mendelian forms of albinism. The underlying assumption is that if a rare null allele causes a complete loss of pigment, then a set of polymorphic, i.e., more frequent, alleles with subtle effects on gene expression will contribute to a spectrum of skin colors. The TYR, P , and MATP genes discussed earlier are well-known causes of albinism whose primary effects are limited to pigment cells ( Oetting and King 1999 ); among these, the P gene is highly polymorphic but the phenotypic consequences of P gene polymorphisms are not yet known. Independent of phenotype, a gene responsible for selection of different skin colors should exhibit a population signature with a large number of alleles and rates of sequence substitution that are greater for nonsynonymous (which change an amino acid in the protein) than synonymous (which do not change any amino acid) alterations. Data have been collected only for MC1R , in which the most notable finding is a dearth of allelic diversity in African samples, which is remarkable given that polymorphism for most genes is greater in Africa than in other geographic regions ( Rana et al. 1999 ; Harding et al. 2000 ). Thus, while MC1R sequence variation does not contribute significantly to variation in human skin color around the world, a functional MC1R is probably important for dark skin. Selection for Skin Color? Credit for describing the relationship between latitude and skin color in modern humans is usually ascribed to an Italian geographer, Renato Basutti, whose widely reproduced “skin color maps” illustrate the correlation of darker skin with equatorial proximity ( Figure 2 ). More recent studies by physical anthropologists have substantiated and extended these observations; a recent review and analysis of data from more than 100 populations ( Relethford 1997 ) found that skin reflectance is lowest at the equator, then gradually increases, about 8% per 10° of latitude in the Northern Hemisphere and about 4% per 10° of latitude in the Southern Hemisphere. This pattern is inversely correlated with levels of UV irradiation, which are greater in the Southern than in the Northern Hemisphere. An important caveat is that we do not know how patterns of UV irradiation have changed over time; more importantly, we do not know when skin color is likely to have evolved, with multiple migrations out of Africa and extensive genetic interchange over the last 500,000 years ( Templeton 2002 ). Figure 2 Relationship of Skin Color to Latitude (A) A traditional skin color map based on the data of Biasutti. Reproduced from http://anthro.palomar.edu/vary/ with permission from Dennis O'Neil. Erratum note: The source of this image was incorrectly acknowledged. Corrected 12/19/03. (B) Summary of 102 skin reflectance samples for males as a function of latitude, redrawn from Relethford (1997) . Regardless, most anthropologists accept the notion that differences in UV irradiation have driven selection for dark human skin at the equator and for light human skin at greater latitudes. What remains controversial are the exact mechanisms of selection. The most popular theory posits that protection offered by dark skin from UV irradiation becomes a liability in more polar latitudes due to vitamin D deficiency ( Murray 1934 ). UVB (short-wavelength UV) converts 7-dehydrocholesterol into an essential precursor of cholecaliferol (vitamin D 3 ); when not otherwise provided by dietary supplements, deficiency for vitamin D causes rickets, a characteristic pattern of growth abnormalities and bony deformities. An oft-cited anecdote in support of the vitamin D hypothesis is that Arctic populations whose skin is relatively dark given their latitude, such as the Inuit and the Lapp, have had a diet that is historically rich in vitamin D. Sensitivity of modern humans to vitamin D deficiency is evident from the widespread occurrence of rickets in 19th-century industrial Europe, but whether dark-skinned humans migrating to polar latitudes tens or hundreds of thousands of years ago experienced similar problems is open to question. In any case, a risk for vitamin D deficiency can only explain selection for light skin. Among several mechanisms suggested to provide a selective advantage for dark skin in conditions of high UV irradiation ( Loomis 1967 ; Robins 1991 ; Jablonski and Chaplin 2000 ), the most tenable are protection from sunburn and skin cancer due to the physical barrier imposed by epidermal melanin. Solving the Mystery Recent developments in several areas provide a tremendous opportunity to better understand the diversity of human pigmentation. Improved spectrophotometric tools, advances in epidemiology and statistics, a wealth of genome sequences, and efficient techniques for assaying sequence variation offer the chance to replace misunderstanding and myths about skin color with education and scientific insight. The same approaches used to investigate traits such as hypertension and obesity—genetic linkage and association studies—can be applied in a more powerful way to study human pigmentation, since the sources of environmental variation can be controlled and we have a deeper knowledge of the underlying biochemistry and cell biology. This approach is especially appealing given the dismal success rate in molecular identification of complex genetic diseases. In fact, understanding more about the genetic architecture of skin color may prove helpful in designing studies to investigate other quantitative traits. Current debates in the human genetics community involve strategies for selecting populations and candidate genes to study, the characteristics of sequence polymorphisms worth pursuing as potential disease mutations, and the extent to which common diseases are caused by common (and presumably ancient) alleles. While specific answers will be different for every phenotype, there may be common themes, and some answers are better than none. Harrison and Owen concluded their 1964 study of human skin color by stating, “The deficiencies in the data in this study are keenly appreciated by the writers, but since there appear at present to be no opportunities for improving the data, it seems justifiable to take the analysis as far as possible.” Nearly 40 years later, opportunities abound, and the mystery of human skin color is ready to be solved.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC212702.xml

539051

Cancer as a Complex Phenotype: Pattern of Cancer Distribution within and beyond the Nuclear Family

Background The contribution of low-penetrant susceptibility variants to cancer is not clear. With the aim of searching for genetic factors that contribute to cancer at one or more sites in the body, we have analyzed familial aggregation of cancer in extended families based on all cancer cases diagnosed in Iceland over almost half a century. Methods and Findings We have estimated risk ratios (RRs) of cancer for first- and up to fifth-degree relatives both within and between all types of cancers diagnosed in Iceland from 1955 to 2002 by linking patient information from the Icelandic Cancer Registry to an extensive genealogical database, containing all living Icelanders and most of their ancestors since the settlement of Iceland. We evaluated the significance of the familial clustering for each relationship separately, all relationships combined (first- to fifth-degree relatives) and for close (first- and second-degree) and distant (third- to fifth-degree) relatives. Most cancer sites demonstrate a significantly increased RR for the same cancer, beyond the nuclear family. Significantly increased familial clustering between different cancer sites is also documented in both close and distant relatives. Some of these associations have been suggested previously but others not. Conclusion We conclude that genetic factors are involved in the etiology of many cancers and that these factors are in some cases shared by different cancer sites. However, a significantly increased RR conferred upon mates of patients with cancer at some sites indicates that shared environment or nonrandom mating for certain risk factors also play a role in the familial clustering of cancer. Our results indicate that cancer is a complex, often non-site-specific disease for which increased risk extends beyond the nuclear family.

Introduction Highly penetrant susceptibility variants explain only a small fraction of the genetics of all cancer cases. As an example, mutations in the BRCA1 and BRCA2 genes account for around 2%–3% of all breast cancer cases [ 1 , 2 ], although more prevalent founder mutations in these genes can explain up to about 10% of the disease in some populations [ 3 , 4 , 5 , 6 , 7 ]. However, the role of genetics in the remaining breast cancer cases and the majority of other cancers is not clear. Family studies have given insight into the contribution of genetic and environmental factors to the etiology of cancer. Case-control, registry- and population-based studies have evaluated familial clustering using either risk ratio (RR) estimations for relatives of cancer patients, or kinship coefficient (KC) estimations for cancer patients. The largest of these studies, utilizing either the Utah Population and Cancer Registry Database or the Swedish Family-Cancer Database, have demonstrated excess familial clustering at practically all cancer sites in the body [ 8 , 9 , 10 , 11 , 12 ]. Most of these studies have been able to evaluate familial clustering only within the nuclear family, thus making it more difficult to separate the roles of shared environmental and genetic factors in the familial aggregation of cancers. However, in one of these studies [ 12 ], in which familial clustering was evaluated for more distant relatives, significant clustering outside the nuclear family was demonstrated for a number of cancer sites. Extended familial clustering has also been reported in studies of individual cancers [ 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ]. Twin studies have also evaluated the role of genes versus environment in cancer susceptibility. The largest study involved close to 45,000 twins from Denmark, Sweden, and Finland where the RR of same type of cancer was calculated for individuals with affected twins and compared to those without an affected twin [ 23 ]. The authors concluded that for the majority of cancer sites only a limited part of the risk could be explained by heritable factors. Exceptions to this were cancers of the prostate, colon and breast. In addition to well documented familial clustering for the majority of individual cancers, aggregation of different types of cancers in families has also been observed. Reports have been published on the results of systematic analysis of the aggregation of different cancers using the Utah Population and Cancer Registry Database [ 24 , 25 ]. In addition to demonstrating excess familial clustering for most cancer sites, these studies also indicate that an excess is also shared by different cancer sites. In these studies, cancer clustering was evaluated either by calculating the RR for first-degree relatives or KC between different cancer sites. While distant relationships contributed to the overall calculation of KC, their contributions were not evaluated separately in the studies between cancer sites, hence making it more difficult to separate the effects of genetic and environmental factors in these studies. We have studied a registry of all cancer cases diagnosed in Iceland from 1 January 1955 to 31 December 2002, with the aim of searching for evidence of genetic factors both at individual cancer sites and those shared by different sites. By cross-referencing cancer prevalence in relatives of cases with the aid of a comprehensive nationwide genealogy database, we have estimated RR separately for first- to fifth-degree relatives of all cancer patients diagnosed in Iceland over 48 y. We demonstrate here an increased cancer risk in relatives outside the nuclear family (third- to fifth-degree relatives) for many cancer sites. These relatives share significant genetic makeup but are less likely to share environmental factors beyond those shared by the general population, indicating that genetic factors may be involved. By applying the analysis across different cancer sites we also demonstrate shared familiality between certain cancer sites both in close and distant relatives. These results suggest that cancer can be considered a broad phenotype with shared genetic factors crossing different cancer sites. That is, the difference between cancers at various sites may in part be the consequence of variable expressivity of the same cancer-predisposing genes. Methods This study was approved by the National Bioethics Committee of Iceland, the Data Protection Authority of Iceland, and the Icelandic Cancer Society. All names of patients listed in the Icelandic Cancer Registry (ICR) and the genealogic database were encrypted through a process approved by the National Bioethics Committee and the Data Protection Authority before being analyzed [ 26 ]. Cancer Registry The ICR of the Icelandic Cancer Society is a carefully constructed database containing practically complete records of all cancer cases diagnosed in Iceland after 1 January 1955 [ 27 ]. Records are received at the ICR from all hospitals in the country that treat cancer patients, and the very few not listed are individuals who are diagnosed while living abroad. Furthermore, the records are verified by a continuous interaction between the ICR and Icelandic hospitals and clinicians. Approximately 95% of cases are histologically verified [ 28 ]. In the present study we used International Classification of Disease version 10 codes as the basis for defining phenotypes. A total of 81 unique phenotypes (sites) were analyzed. In this paper we present data from 27 sites with more than 200 cases each ( Table 1 ). For the 48 years (1 January 1955 to 31 December 2002) a total of 32,534 individuals were found in our genealogy database. Cancer incidence in Iceland is comparable to the Nordic countries of Europe and is detailed in [ 27 ]. Table 1 RR Estimates of Cancer at the Same Site for Relatives and Mates for Cancer Sites with 200 or More Cases Shown are the estimated RRs with 90% confidence interval for first- to fifth-degree (1°–5°) relatives and mates of the 27 cancer sites with ≥200 cases, in bold when the 90% CI does not include 1.00, which corresponds to one-sided p < 0.05. Also shown are combined p values to evaluate the significance of the increased RR for all relatives (first- to fifth-degree) and for close (first- and second-degree) and distant relatives (third- to fifth-degree). p values nominally significant at the 0.05 level are shown in bold a Nominal p values that remained significant after Bonferroni correction for the 27 individual tests ( p < 0.00185) b na, not applicable (sex-specific cancers) c NHL, non-Hodgkin's lymphoma d Number of mates with cancer/total number of mates for each cancer site ICD10, International Classification of Disease version 10 Genealogic Database deCODE Genetics has built a computerized genealogy database of more than 687,500 individuals [ 29 , 30 ]. The names of all 288,000 Icelanders currently alive and a large proportion of all Icelanders who have ever lived in the country are in the database. The genealogy of the entered individuals is recorded from multiple sources including church records and censuses from previous centuries and, more recently, from published genealogy books. The genealogy database is quite complete from the 18th century on, thus allowing quite distant relationships to be traced accurately. Mates are defined as individuals of the opposite sex who have one or more children in common, regardless of marital status. Calculations of RRs The RR for relatives is a measure of the risk of disease for a relative of an affected person compared to the risk in the population as a whole. More precisely, for a given relationship the RR for disease B in the relatives of probands with disease A is defined as where P A denotes the event that the proband is affected with disease A, and R B denotes the event that the relative is affected with disease B. Note that disease A and disease B can be the same in this definition which applies when estimating RR at individual cancer sites. Using Bayes' rule it can be shown that for symmetric relationships, RR is the same if the roles of A and B are switched, i.e., the RR for disease A in the relatives of probands with disease B is the same as the described above. In this study we always chose the less common phenotype as the proband when estimating RR. A basic underlying assumption in our estimation of RR is that of conditional independence of ascertainment, or censoring, ( O RB and O PA are the events that the relative and proband are observed with diseases A and B, respectively): P ( O RB , O PA | P A , R B ) = P ( O RB | R B ) P ( O PA | P A ). Some form of this assumption is used by most methods estimating RR [ 31 ]. Obtaining valid estimates of the RR is not always straightforward, since the method of ascertainment of affected cases critically affects the estimation, and inappropriate estimators can lead to bias or inflated estimates [ 32 ]. The use of a nationwide registry of patients covering close to five decades decreases much of the potential sampling bias. However, the ascertainment of the ICR depends on the year of birth of individuals. This dependence needs to be addressed when estimating the RR. The approach chosen here is to estimate the RR for a number of subpopulations, where prevalence is reasonably constant, and combine them into a single estimate of RR for the full population. Let r be the number of relatives of probands, counting multiple times individuals who are relatives of multiple probands [ 33 ], let a be the number of relatives of probands that are affected (again possibly counting the same individual more than once), let n be the size of the population, and finally let x be the number of affected individuals in the population. If P ( R B ) and P ( R B | P A ) can reasonably be assumed to be constant in the population, then x/n and a/r, respectively, are estimates of these probabilities. Given these estimates, RR is consistently estimated by Assuming the population can be split into N subpopulations, such that within each subpopulation P ( R B ) and P ( R B | P A ) can be assumed to be constant, although they may vary between subpopulations, and assuming furthermore that RR is the same in all the subpopulations, then the RR is consistently estimated by a convex combination of the estimates for the subpopulations. We selected weights for the combination such that the efficiency of the estimator was at maximum for RR equal to one. Making the simplifying assumption that the relatives are independent (while this assumption is not entirely correct, it affects only efficiency, not validity), the optimal weight for group j is (this is the inverse of the variance of the estimate for RR in subpopulation j ), where a, r, x, and n are defined as above, restricted to the subpopulation j . Note that probands are not restricted to the subpopulation. Given these weights, our estimate of RR is In this study, the most relevant variations in P ( R B ) and P ( R B | P A ) stem from time-dependent censoring of affected status and sex-specific differences. Hence, we have stratified the population so that j runs over groups of people of the same sex and born in the same 5-y periods. For a fixed year-of-birth stratum, there is censoring of affected status (missing data) based on year of onset because of the fact that records cover only the period 1955–2002. Our approach is designed to address this type of missing data. As an example of the stratification, the breast cancer patients in our analysis were born in the years 1865 to 1970 (5-y strata), yielding 35 subpopulations, 22 for female patients, but only 13 for male, as this cancer is rare for males. To assess the significance of the RR obtained for a given group of patients, we compared their observed values with the RR computed for up to 100,000 independently drawn and matched groups of control individuals. Each patient was matched to a single control individual in each control group. The control individuals were drawn at random from the genealogic database with the conditions that they had the same year of birth, the same sex, and the same number of ancestors recorded in the database at five generations back as the matched patients. Empirical p values can be calculated using the control groups; thus, a p value of 0.05 for the RR would indicate that 5% of the matched control groups had values as large as or larger than that for the patient's relatives or mates. The number of control groups required to obtain a fixed accuracy of the empirical p values is inversely proportional to the p value. We therefore selected the number of control groups generated adaptively up to a maximum of 100,000. When none of the values computed for the maximum number of control groups were larger than the observed value for the patient's relatives and mates, we report the p value as being less than 0.00001. Using a variance-stabilizing square-root transform, an approximate confidence interval may be constructed based on the distribution of RR for control groups [ 33 ]. As another test for significance of RR between cancer sites, we used combined estimators for risk in relatives of degree 1 and 2 together, degrees 3, 4, and 5 together, and degrees 1 through 5 together. If RR d is the RR for relatives of degree d, then RR d – 1 is known to decrease proportional to 2 -d as d increases for a monogenetic single variant or additive disease models, and faster for more complex disease models [ 34 ]. With the estimate of RR d denoted by , we then chose a test statistic of the form with d summed over the relevant degrees. For RR d close to one, the variance of the estimate is inversely proportional to the number of relatives of degree d for the proband. Based on the Icelandic genealogy for the cancers being studied here, the number of relatives is proportional to γ d , where the value of γ quantifies how the number of relatives grows with each degree of relatedness to the proband. This factor γ varies only slightly between cancers and is on average 2.46. Minimizing the variance of the test statistic in equation 6 with respect to the weights yields the statistic As above, the choice of weights and the form of the statistic affects only power, not validity. To assess significance, the observed value of the statistic was compared to its value for multiple matched control groups as described above. Although our evaluations of familial clustering, for both close and distant relatives, are based on RR, an alternative approach based on comparing KCs among patients and among controls exists [ 12 , 24 , 25 ]. The two approaches are closely related, and our choice was made in part because relative risk is a less technical concept and its application to genetic counseling more direct. Also, the relationship between relative risk and the power to map disease genes by linkage analysis has been thoroughly investigated [ 34 , 35 ]. Results We have studied the familial clustering of cancer by estimating RR for first- and up to fifth-degree relatives both within and between all cancer sites. Here we present results for 27 sites that contain 200 or more cancer cases each, based on International Classification of Disease version 10 codes. These 27 sites represent 89% of all cancer cases in the ICR. Risk Estimations for Cancer at Same Site A significantly increased RR to first-degree relatives of patients with cancer was seen for 22 of the 27 cancer sites ( Table 1 ). Among the statistically significant RRs, the highest estimates were for lymphoid leukemia, Hodgkin's disease, and cancer of the thyroid, meninges, lip, testis, and larynx (RR above three). These cancers, except for thyroid cancer, were among the least prevalent sites (200–400 cases), as reflected in the large standard deviation of the RR estimates ( Table 1 ). First-degree relatives of individuals with breast, lung, kidney, pancreatic, ovarian, and esophageal cancer and multiple myeloma, had between 2- and 3-fold increased risk of developing the same cancer. The medians of the estimated RR values for the 27 sites in first- to fifth-degree relatives were 2.00, 1.32, 1.21, 1.10, and 1.04, respectively. Combined p -values incorporating the increased risk for first- to fifth-degree relatives identified 21 sites being significant at a nominal level of 0.05. Sixteen of those sites remained significant after Bonferroni adjustment for the 27 individual tests ( p value < 0.00185) ( Table 1 ). To discriminate between familial clustering in close and distant relatives, combined p values were also calculated for first- and second-degree relatives on one hand and for third- to fifth-degree relatives on the other hand ( Table 1 ). Fourteen sites were nominally significant for the distant relationships (third- to fifth-degree relatives) of which eight were significant after Bonferroni adjustment. These eight sites were all within the group of 16 sites demonstrating significant familial clustering in all relationships. The RR for developing cancer at the same site was also estimated for mates of cancer patients at 22 out of the 27 individual sites. The remaining five sites are sex-specific and calculations thus not applicable. For seven rare cancer sites, affected mates were not observed, corresponding to a RR of zero. Only lung, stomach, and colon cancer were characterized by significantly increased RR values in mates ( Table 1 ). Risk Estimations between Cancer Sites We calculated RR between all cancer sites for first- and up to fifth-degree relatives and mates (results for the 27 largest sites are shown in Table S1 ). As done for the individual cancer sites, p values were calculated for all (first- to fifth-degree), close (first- and second-degree), and distant (third- to fifth-degree) relationships. Figure 1 shows a diagram representing 20 pairs of cancer sites that associate with a combined p value, significant at a level of 1 × 10 -4 , for first- to fifth-degree relationships. This level was significant at the 0.05 level after Bonferroni adjustment for the 351 tests (number of unique pairs of cancers). The strength of the distant familiality (i.e., the p value for third- to fifth-degree relatives) between these pairs of cancers is represented by the thickness of the lines joining sites in Figure 1 . Figure 1 A Schematic Representation of Cancer Pairs Demonstrating Significant Familial Aggregation Cancer pairs that demonstrate significant familial co-clustering (first- to fifth-degree relatives) at the 0.05 level after adjustment for multiple testing (nominal p value < 1 × 10 -4 ) are joined by lines. The thickness of the lines joining the pairs are based on nominal p values corresponding to the significance of the familiality in distant relatives (third to fifth degree): bold, p ≤ 0.001; solid, p ≤ 0.01; and dashed, p ≤ 0.05. The number on the lines joining each pair indicates the cross-cancer RR in first-degree relatives. Shaded ovals correspond to individual cancer sites that were significant for the combined group of first- to fifth-degree relatives at the 0.05 level after Bonferroni adjustment (see Table 1 ). In total, 17 cancer sites were involved in 20 significant pairs of sites ( Figure 1 ). Stomach and prostate cancer were involved in most pairs, seven and six pairs, respectively, followed by colon, ovarian, and cervical cancer, each involved in three pairs. The estimated RRs for the 20 pairs are between 1.1 and 1.7 for first-degree relatives and between 1.1 and 1.5 for second-degree relatives ( Figure 1 ; Table S1 ). The highest RRs in first-degree relatives between cancer sites were seen for esophagus–cervix, with a RR of 1.74, pancreas–ovary, with a RR of 1.66, and colon–rectum, with a RR of 1.64. All of the 20 pairs shown in Figure 1 were nominally significant ( p value < 0.05) for distant relationships, of which nine were significant at the 0.001 level. In the latter group, prostate, rectum, stomach, and cervical cancers each appeared in two pairs, and colon cancer in three. Discussion In this study we have comprehensively analyzed familial aggregation of cancer cases in a whole nation, both within and between pairs of cancer sites. The completeness of our genealogy database allows us to accurately trace distant relationships, which we believe is unique to this study. Linking the ICR to our nationwide genealogy database thus has made it possible to uncover distant familial connections between cancer cases, and reach beyond shared environmental factors to identify individual and combined cancer sites with the strongest genetic influences. Furthermore, even though the genetic effect decreases with more distant relationships, the sample sizes used to estimate familiality are dramatically larger for the distant relationships than for the closer ones. This compensates to some extent for the lower effect and adds considerable statistical power to the study. In this paper we restrict the presentation and discussion to the most significant findings. However, we provide results for all pairs of 27 cancer sites in Table S1 , as a resource for other researchers interested in the familiality of specific cancers. The largest population-based studies reported to date, evaluating familial clustering within the same cancer site, are from Utah and Sweden [ 8 , 9 , 10 ]. These studies report RR values for first-degree relatives [ 36 ] that are comparable to those presented here for first-degree relatives. For example, the median RRs for the occurrence of the same cancer in first-degree relatives were 2.15, 1.86, and 2.00 for the Utah, the Sweden, and our study, respectively. Also, RR values in first-degree relatives ranged between 1.5 and 3.0 for the majority of sites, i.e., 69%, 82%, and 60%, in Utah, Sweden, and this study, respectively. As seen in Utah and Sweden, high RR values were found in this study for multiple myeloma, lymphoid leukemia, and thyroid, testicular, and laryngeal cancer. The RR for thyroid cancer in first-degree relatives was much higher in Utah and Sweden (8.48 and 9.51) than in Iceland (3.02). One possible explanation of the lower RR may be the high incidence of thyroid cancer in Iceland, due to an excess of the papillary subtype [ 18 , 37 ], which is not a part of the multiple endocrine neoplasia syndromes. The cancer sites showing the highest RR for first-degree relatives tend to be among the rarer sites. There are two potential reasons why rare tumors tend to show higher RRs than common cancers. Being common, the baseline frequency is not low and that creates a bound on how large the RR can be. Also, common cancers are expected to be genetically complex, whereas it is more likely for a rare tumor to be closer to a Mendelian trait, caused by rare alleles with high penetrances. Most individual cancer sites, or 16 out of the 27 studied here, showed familiality as evidenced by significant p values (after adjustment for multiple testing) for the combined group of first- to fifth-degree relatives. Furthermore, eight of these 16 sites remained significant even after exclusion of the first- and second-degree relatives (after adjustment for multiple testing). The majority of the 16 significant cancer sites are among the sites of the most prevalent cancers, indicating that we may lack power to detect extended familiality for the less prevalent cancer sites. Indeed the median number of cases per cancer site was 943 for the 16 significant sites compared to 342 for the non-significant sites. Nevertheless, significant familial clustering (first- to fifth-degree relatives) is seen for some of the less prevalent sites, i.e., lymphoid leukemia and esophagus and meningeal cancer. The largest cancer twin study reported to date [ 23 ] documented significant heritability of prostate (42%), colorectal (35%), and breast cancer (27%) and provided suggestive evidence for limited heritability of leukemia and stomach, lung, pancreas, ovarian, and bladder cancer. All of these cancer sites showed significant familial clustering in our study. However, when the analysis was restricted to distant relatives, lymphoid leukemia, pancreatic, and ovarian cancer were no longer significant. Although close to 45,000 pairs of twins were included in the study (of which 10,803 had been diagnosed with cancer), the study clearly lacked statistical power to detect the effects of heritable factors for the less prevalent cancer sites. A significantly increased risk of the same cancer was seen in mates only for individuals diagnosed with stomach, lung, or colon cancer. These results are in accordance with previous reports, including Swedish population-based studies, except for colon cancer [ 38 , 39 , 40 , 41 ]. Environmental factors in adult life (including lifestyle and infections) or nonrandom mating could explain the higher risk of these cancer types in mates. The RR was not significant or not observed in mates for other sites. We also assessed the significance of familial clustering between cancer sites by calculating combined p values corresponding to the increased risk for first- to fifth-degree relationships. With this method, we detected 17 cancers that linked into 20 pairs of sites that were significant after adjustment for multiple testing. Stomach and prostate cancer appeared more frequently in the pairs than other cancer types, followed by colon, ovarian, and cervical cancer. We emphasize again, as with the same-cancer calculations, that we might lack power to connect rare cancers to other cancer sites. This possibility is highlighted by the fact that the 17 cancers in the significant pairs are the most prevalent cancer sites in Iceland. Some connections seen here between cancer sites may be partly explained by known high-risk genes involved in heritable syndromes. Thus, mutations in genes associated with hereditary nonpolyposis colorectal cancers could explain a part of the risk shared between stomach, colon, rectal, and endometrial cancer, and possibly brain and ovarian cancer [ 42 , 43 ]. In a similar manner, mutations in BRCA1 and BRCA2 may explain in part the cluster seen between prostate, breast, ovarian, and possibly pancreatic cancer [ 20 , 44 , 45 , 46 ]. Other known but even rarer cancer syndromes are likely to explain only a handful of cases. Undiscovered genetic factors could contribute to some connections seen here to a much greater extent than the known susceptibility factors. Although these could include unknown high-risk susceptibility genes, they are more likely multiple genetic variants, each conferring small to moderate risk. Familial clusters were identified between cancer sites, both in close and distant relatives, that do not correspond to known cancer syndromes. These include lung, esophageal, cervical, and stomach cancer, which, interestingly, have been associated with environmental rather than genetic factors. One explanation for this excess familiality between these cancer sites is an interaction of genetic susceptibility factors with environmental carcinogens (e.g., tobacco and diet) or infectious agents. Thus, the same environmental factor could interact with the same genetic susceptibility factor or factors to induce different cancers (i.e., smoking in lung and cervical cancer). Alternatively, different environmental factors could interact with the same genetic susceptibility factor or factors to increase the risk for different cancers (i.e., smoking in lung cancer and human papilloma virus in cervical cancer). Hormone-related cancers form another risk cluster. Thus, shared genetic susceptibility factors could directly influence the hormonal metabolism to induce breast, prostate, thyroid, or ovarian cancer in carriers. Alternatively, shared genetic factors could interact with dietary factors to induce aggregation of cancers at these sites in related individuals. A significantly increased risk of breast, prostate, cervical, and non-melanoma skin cancer was recently reported in first-degree relatives of early-onset breast cancer patients from Sweden that tested negative for BRCA1 and BRCA2 mutations [ 47 ]. Our data support the notion that unknown susceptibility variants that increase the risk of breast and prostate cancer and melanoma remain to be characterized. Two more groups of cancers with shared risk were identified that each include sites that share the same developmental progenitors: the prostate, kidney, and bladder are sites derived from the nephrogenic ridge while colon, rectum, and stomach are derived from the primitive gut tube. Therefore, the sites in each group may share risk alleles that regulate embryonic development, which can later play a role in oncogenesis. Interestingly, three cancer sites/types, non-melanoma skin, brain, and melanoma, that do not have significant same-cancer familial clustering demonstrate significant cross-cancer familial clustering with more prevalent cancer sites, i.e., rectum, stomach, and kidney cancers, respectively. Previous reports systematically evaluating the significance of co-clustering of cancer pairs in families have utilized the Utah Population Database. In these studies lip and prostate cancers appear to associate most frequently with other cancer sites. The same is true for prostate cancer in our study, whereas lip cancer does not significantly associate with any other cancer sites. This can at least in part be explained by the difference in age-standardized incidence rates for lip cancer in Iceland and Utah (Iceland 1.1 and Utah 2.4) [ 48 ]. In contrast, stomach cancer associates with seven other cancer sites out of the 20 significant pairs in our study, but only three other sites in the Utah study. Of the 20 cancer pairs that significantly associate in our study, eight concur with the Utah studies. Because the increased cross-site RR extends beyond the nuclear family, shared genetic factors may contribute to the risk of more than one cancer type. This suggests that cancer could be considered a broad phenotype with shared genetic factors across cancer sites. Therefore cancer should in certain cases be studied in a broader context than previously done. Combining multiple cancers that show increased cross-site RR may serve to increase the power of linkage and case-control studies. Our results also have implications for genetic counseling and imply that the focus of attention should broaden to the history of multiple cancer types in relatives within and outside the nuclear family. These results also suggest the utility of comparing expression profiles and in vitro biological processes across the cancers that we have identified as sharing genetic risk. The isolation of cancer predisposition genes with broad effects may define new rate-limiting pathways that can be used to search for drug targets for a more focused treatment with fewer side effects but with utility across multiple cancers. Supporting Information Table S1 Cross-Site RR Estimates for Relatives and Mates of Patients Diagnosed with Cancers at 27 Sites with 200 Cases or More (1.9 MB DOC). Click here for additional data file. Accession Numbers The LocusLink ( http://www.ncbi.nlm.nih.gov/projects/LocusLink/ ) accession numbers for the genes discussed in this paper are BRCA1 (LocusLink ID 672) and BRCA2 (LocusLink ID 675). Patient Summary Background Although a few cancers have a fairly simple genetic cause, most, especially the most common cancers, do not, and what makes one person rather than another develop cancer is not clear. One way of trying to work out how much genes rather than environment contribute to disease is to study large populations. One such population is the Icelandic nation: not only is detailed health information about individuals available, including information on cancer, but also very good genealogical information and a substantial amount of genetic data. What Did the Study Find? Researchers examined all cancer records dating back to 1955 and then analyzed the chances of relatives and mates of these patients having cancer. They found that some cancers, especially rare ones, had a higher than baseline chance of occurring in relatives, but so did many common cancers, and for some cancers, the higher chances extended to quite distant relatives. In addition, the risk sometimes involved different cancer types. What Does the Study Mean for Patients? Even for the highest risk cancers, the absolute increased risk for relatives remains very small. In addition, despite the large numbers of patients studied, the numbers of cancer cases are still not large enough to be completely certain of the results, apart from very common cancers, which had the lowest chance of occurring in relatives. So these results will not help doctors much at the present time in telling an individual patient what their risk is of getting cancer if a relative has it—but they will be useful for other researchers in knowing how to plan future studies to look at the underlying causes of cancer. Where Can I Get More Information? Icelandic Cancer Society: http://www.krabb.is/cancer/ The United States National Cancer Institute's Cancer Information Service: http://cis.nci.nih.gov/ CancerHelp UK, a free information service about cancer and cancer care: http://www.cancerhelp.org.uk/ deCODE Genetics: http://www.decode.com

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539051.xml

549207

Assessing stability and change of four performance measures: a longitudinal study evaluating outcome following total hip and knee arthroplasty

Background Physical performance measures play an important role in the measurement of outcome in patients undergoing hip and knee arthroplasty. However, many of the commonly used measures lack information on their psychometric properties in this population. The purposes of this study were to examine the reliability and sensitivity to change of the six minute walk test (6MWT), timed up and go test (TUG), stair measure (ST), and a fast self-paced walk test (SPWT) in patients with hip or knee osteoarthritis (OA) who subsequently underwent total joint arthroplasty. Methods A sample of convenience of 150 eligible patients, part of an ongoing, larger observational study, was selected. This included 69 subjects who had a diagnosis of hip OA and 81 diagnosed with knee OA with an overall mean age of 63.7 ± 10.7 years. Test-retest reliability, using Shrout and Fleiss Type 2,1 intraclass correlations (ICCs), was assessed preoperatively in a sub-sample of 21 patients at 3 time points during the waiting period prior to surgery. Error associated with the measures' scores and the minimal detectable change at the 90% confidence level was determined. A construct validation process was applied to evaluate the measures' abilities to detect deterioration and improvement at two different time points post-operatively. The standardized response mean (SRM) was used to quantify change for all measures for the two change intervals. Bootstrapping was used to estimate the 95% confidence intervals (CI) for the SRMs. Results The ICCs (95% CI) were as follows: 6MWT 0.94 (0.88,0.98), TUG 0.75 (0.51, 0.89), ST 0.90 (0.79, 0.96), and the SPWT 0.91 (0.81, 0.97). Standardized response means varied from .79 to 1.98, being greatest for the ST and 6MWT over the studied time intervals. Conclusions The test-retest estimates of the 6MWT, ST, and the SPWT met the requisite standards for making decisions at the individual patient level. All measures were responsive to detecting deterioration and improvement in the early postoperative period.

Background Osteoarthritis, the most common reason for total hip (THA) and knee arthroplasty (TKA), accounts for more difficulty with climbing stairs and walking than any other disease [ 1 , 2 ]. Physical performance measures, therefore, play an important role in the measurement of outcome in patients undergoing total joint arthroplasty. Although the past two decades have seen considerable development and evaluation of self-report functional status measures [ 3 - 7 ] these advances have not been paralleled to the same extent in performance measures. Information about customary or normal values often exists for performances measures, however, information concerning sensitivity to change and clinically important change are rarely available [ 8 ]. This gap is exemplified in the case of commonly used performance measures in the assessment of patients post TKA and THA. Measures such as self-paced walk tests (SPWTs) [ 9 - 11 ], the timed up and go test (TUG) [ 9 , 12 , 13 ], stair measures (STs) [ 9 - 11 , 14 ] and the six minute walk test (6MWT) [ 14 - 18 ] lack information on responsiveness in this population [ 8 ]. Although the literature contains varied definitions of responsiveness, in this case, it is used to indicate the ability of a measure to detect change [ 19 ]. A few studies have examined the responsiveness of the 6MWT and STs in patients following arthroplasty. Kreibich et al [ 15 ] investigated the responsiveness of six outcome measures using paired t tests and found that the 6MWT was more responsive than a thirty-second stair climb, yet not as responsive as the two disease specific measures studied. Parent et al [ 14 ] compared the responsiveness of 3 locomotor tests and 2 questionnaires using 4 different responsiveness statistics and recommended the 6MWT and the Physical Function subscale of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) for assessment in the early recovery period after TKA. No studies were found that examined the responsiveness of the SPWT and TUG. Several studies used performance test components in other tools, however, they were not reported in their original format [ 20 , 21 ]. Responsiveness statistics such as the standardized response mean (SRM) and effect size (ES) are important for making relative comparisons between measures. However, clinicians still require estimates to quantify the error in patients' scores and to determine if change has truly occurred. In the absence of population specific benchmarks, clinicians and researchers apply the results available from other populations. For example, Mahon et al [ 17 ] used the 6MWT as one outcome measure to examine the association between waiting time and postoperative health-related quality of life in patients undergoing THA. They considered a change of greater than 30 meters in the 6MWT to be clinically important, based on the work of Guyatt et al [ 22 ] in respiratory patients. Enhancing the interpretability of commonly used performance measures in the end stage OA-arthroplasty population would assist clinicians and researchers to better quantify decline and recovery. The importance of determining THA and TKA population specific benchmarks is further underlined when one considers the growing number of North Americans requiring total joint arthroplasty [ 23 , 24 ]. In Canada alone, the number of THR and TKR increased 31.7% from 1994/1995 to 1999/2000 [ 25 ]. The purposes of this study were therefore to examine the reliability and sensitivity to change of the SPWT, TUG, ST and the 6MWT in patients with end-stage hip or knee osteoarthritis (OA) who subsequently underwent a total joint arthroplasty. Methods Subjects The sample consisted of patients with a diagnosis of OA who were scheduled to undergo primary, unilateral THA or TKA and was part of a larger, observational, longitudinal study. A sample of convenience was chosen and included one hundred fifty consecutive, eligible patients (69 hips, 81 knees) investigated over the one-year period, November 2001 to 2002. Eligibility criteria included the following: diagnosis of OA, scheduled for primary total joint arthroplasty; sufficient language skills to communicate in written and spoken English; and absence of neurological, cardiac, psychiatric disorders or other medical conditions that would significantly compromise physical function. Patients were excluded if they were scheduled for revision, bilateral or staged arthroplasties. All of the surgeries took place at a specialized, orthopaedic tertiary care hospital in Toronto. The characteristics of the patients with respect to age, height, weight, and body mass index (BMI) are reported in Table 1 . All patients provided informed consent and the study received approval from the institution's research and ethics review board. Table 1 Sample Characteristics n = 150 Mean, SD Quartiles Age (yr) 63.7, 10.7 57, 64, 72 Height (m) 1.69, 0.09 1.62, 1.68, 1.76 Weight (kg) 85.5, 15.4 74.3, 83.3, 94.2 Body mass index (kg/m 2 ) 30.0, 4.9 26.3, 28.9, 33.4 n, Number of subjects Yr, year M, meter Kg, kilogram SD, standard deviation Outcome measures As noted earlier, patients completed four timed performance measures; the fast SPWT, TUG, ST, and 6MWT, at each assessment point. Time was measured on a stopwatch to the nearest 1/100 of a second. The order of testing was as follows: SPWT, TUG, ST, and 6MWT with a 10 minute rest between the ST and 6MWT. Standardized guidelines for performing the SPWT, TUG, and ST have been reported previously for a similar patient population [ 9 , 11 ]. In terms of the fast SPWT, patients were timed while they walked two lengths (turn excluded) of a 20-m indoor course in response to the instruction: "walk as quickly as you can without overexerting yourself." The ST required patients to ascend and descend 9 stairs (step height, 20 cm) in their usual manner, and at a safe and comfortable pace. To complete the TUG, patients were required to rise from a standard arm chair, walk at a safe and comfortable pace to a tape mark 3-m away, then return to a sitting position in the chair [ 26 ]. During the performance of the 6MWT, patients were instructed to cover as much distance as possible during the 6 minute time frame with opportunity to stop and rest if required. The test was conducted on a pre-measured, 46 meter unobstructed, uncarpeted, rectangular circuit. The course was marked off in meters and the distance traveled by each subject was measured to the nearest meter. As encouragement has been shown to improve performance [ 27 ], standardized encouragement, "You are doing well, keep up the good work" was provided at 60 second intervals. During the administration of each of the four performance measures, patients were permitted to use their regular walking aids. Study design As noted previously, the data for this study represent a subset from a larger ongoing study that examines recovery profiles using a number of self-report and physical performance measures. The study has two arms, in Phase 1 patients are recruited from the caseload of two orthopaedic surgeons with high volumes and long waiting lists to examine the impact of waiting time on recovery profiles. In phase 2, patients are recruited from all of the orthopaedic surgeons' lists at their preoperative visit to the hospital's standardized patient orientation program, which is scheduled one to two weeks prior to surgery. There are no differences in the postoperative follow-up for both of the Phases and all patients receive standardized treatment, following either a primary total hip or knee care pathway. To provide an accurate model of change over time, patients' follow-up measurements are scheduled at different intervals. The format is that of an observational repeated measures' design (Figure 1 ). Figure 1 Study Design Test-retest reliability was assessed preoperatively in a sub-sample of 21 patients from Phase 1. These 21 patients represented individuals who had progressed to surgery and follow-up by the time of this analysis. Data from patients' initial consultations with the surgeon, an intermediate assessment, and then again at patients' preoperative orientation visits contributed to the reliability analysis. Although the median interval between the first and second assessments was 91 days (1 st , 3 rd quartiles: 72, 133 days) and between the first and third assessments was 178 days (1 st , 3 rd quartiles: 140, 204 days), there is evidence to suggest that the amount of change in function while on the waiting list is minimal [ 28 ]. A second strategy was also employed to examine the stability of the twenty-one patients' measures over the aforementioned time period using data from the larger study on the Lower Extremity Functional Scale (LEFS). Previous research has determined the LEFS minimal detectable change at a 90% confidence level (MDC 90 ) to be 9 LEFS points [ 29 ]. Using this benchmark, data from only 17 of the 21 patients were retained for the reliability analysis. It is important when assessing responsiveness that a research design be employed in a period where change is expected. Based on the results of prior work [ 9 ], it was recognized that the early period following joint arthroplasty would provide such a framework in which the measures' abilities to detect deterioration and improvement could be determined. A construct validation process was therefore applied to evaluate the measures' abilities to detect change at two different time points post-operatively. The first postoperative assessment occurred within 15 days of surgery. The median interval between the preoperative and first postoperative assessment was 8 days (1 st , 3 rd quartiles: 7, 9 days). It was theorized that patients' lower extremity functional status, as represented by either the time to complete a task or the distance covered in the case of the 6-minute walk test, would demonstrate deterioration compared to their preoperative values [ 9 ]. Next it was theorized that patients' lower extremity functional status would improve over the interval between the first and second postoperative assessments with the minimum interval between these assessments set to 20 days. The median interval between these postoperative assessments was 38 days (1 st , 3 rd quartiles: 32, 46 days). Analysis Descriptive statistics including the mean, standard deviation, and quartiles were applied to summarize the data. Shrout and Fleiss Type 2,1 intraclass correlation coefficients (ICC) were used to describe the measures' test-retest reliabilities [ 30 ]. Standard errors of measurement (SEMs) were used to quantify the measurement error in the same units as the original measurement [ 31 ]. The 95% confidence intervals for all ICCs and SEMs [ 30 , 31 ] were calculated. In addition, the error associated with a measured value (i.e., 90% confidence interval) and the minimal detectable change at the 90% confidence level (MDC 90 ) was calculated [ 19 ]. The error calculation for a measured value was obtained by multiplying the point estimate for the SEM by the z-value associated with the 90% confidence interval (z = 1.65). To calculate MDC 90 , the value obtained from the error calculation was multiplied by the square root of two (i.e. MDC 90 = SEM × 1.65 × ). The interpretation of MDC 90 is that 90% of truly stable patients will demonstrate random variation of less than this magnitude when assessed on multiple occasions. A change greater than MDC 90 is often interpreted as a true change. The standardized response mean (SRM) was used to quantify change [ 3 ] and SRMs were calculated for all measures for the two change intervals. A minus sign was applied to all SRMs that represented deterioration in functional status. For example, a decrease in distance, and an increase in time were assigned negative values. Although sample values of the SRM for the measures represent estimates of the population parameters for these measures, it is impossible to directly ascertain their sampling distributions. We applied a bootstrap procedure to obtain approximate representations of the sampling distributions for the measures' SRMs and to estimate their 95% confidence intervals [ 32 ]. Bootstrapping involves sampling with replacement. Specifically, 1000 samples of size n – where n equaled the number of observations for the specific analysis of interest – were selected with replacement. Estimates of SRMs were ordered from lowest to highest; accordingly, the 25 th and 975 th observations from the bootstrap samples represented the 95% confidence limits. This method provides a distribution free estimate of the confidence limits. Results Figures 2 , 3 , 4 , 5 provide the distributions of preoperative scores for each of the performance measures. Table 2 provides a summary of the reliability analyses and estimates of SEM and MDC 90 . There was no systematic difference between the test and retest assessments for any of the measures (p > 0.05). All of the estimates were greater or equal to 0.90 with the exception of the TUG. Table 3 summarizes the measured performance values (means and quartiles) for the three assessment points and Table 4 presents a summary of the change scores and SRMs. The number of patients in Tables 3 and 4 differ as a result of the pattern of missing values. The results presented in Tables 3 and 4 provide consistent evidence that lower extremity functional status, as represented by the time/distance concept, deteriorates between the preoperative and first postoperative assessment. The measures demonstrated uniform improvement from the first to second postoperative assessments: time decreased, and distance for the 6-minute walk increased. As apparent in Table 4 , the SRMs were greatest for the ST and 6MWT over the two measured time intervals. Table 5 provides an accounting of the missing data. It is evident from this table that a substantial number of patients were unable to complete the ST and 6MWT when administered within 16 days of surgery. Independent t-tests were performed to test if the preoperative values differed for patients who were and were not able to complete the ST and 6MWT at the first postoperative visit. No significant differences (p > 0.05) in the preoperative ST or 6MWT were observed for patients in the two groups. Figure 2 Distribution of Times to Complete the Fast Self-Paced Walk Test Figure 3 Distribution of Preoperative Stair Test Times Figure 4 Distribution of Preoperative Timed Up and Go Test Times Figure 5 Distribution of Preoperative 6 Minute Walk Test Distances Table 2 Reliability Coefficients and Minimal Level of Detectable Change Measure R (95% CI) SEM (95% CI) Confidence in Score (90% CI) MDC 90 Fast Self-paced Walk Time (completed over 40 meters) 0.91 (0.81, 0.97) 1.73 (1.39, 2.29) ± 2.86 s 4.04 s Stair Time 0.90 (0.79, 0.96) 2.35 (1.89, 3.10) ± 3.88 s 5.49 s Timed Up and Go Time 0.75 (0.51, 0.89) 1.07 (0.86, 1.41) ± 1.76s 2.49 s Six Minute Walk Test Distance 0.94 (0.88, 0.98) 26.29 (21.14, 34.77) ± 43.37 m 61.34 m R, Reliability Coefficient SEM, Standard Error of Measurement MDC 90 , Minimal detectable change at the 90% confidence Level s, seconds m, meters Table 3 Mean and Quartile Scores of the Performance Measures across Time Measure Preop Mean, SD Quartiles n = 150 Postop 1 <16 Days Postop Mean, SD, n Quartiles Postop 2 >20 Days From Postop 1 Mean, SD, n Quartiles Self-paced Walk Time (seconds) 31.7, 9.2 25, 30, 36 85.7, 62.7, 115 53, 66, 93 33.7, 10.9, 92 26, 32, 38 Stair Time (seconds) 17.1, 8.2 11, 15, 22 40, 12, 87 29, 39, 48 20.0, 9.7, 91 12, 18, 27 Timed Up and Go Time (seconds) 9.8, 3.2 7, 9, 11 24.7, 14.2, 116 15, 21, 31 10.3, 4.2, 91 7, 9, 12 Six minute Walk Test Distance (meters) 412, 123 329, 412, 508 193, 87, 82 120, 194, 263 408, 116, 91 328, 393, 477 SD, Standard Deviation n, Number of subjects Table 4 Change Scores and Standardized Response Means Measure Preop to First Postop Interval Mean Change*, SD, n SRM* (95% CI) First to Second Postop Interval Mean Change*, SD, n SRM* (95% CI) Self-paced Walk Time (seconds) -54.8, 61.6, 115 -0.89 (-1.42, -0.68) 47.7, 60.7, 89 0.79 (0.66, 1.45) Stair Time (seconds) -23.8, 13.8, 87 -1.74 (-2.13, -1.45) 20.59, 10.40, 73 1.98 (1.68, 2.42) Timed Up and Go Time (seconds) -14.9, 13.8, 116 -1.08 (-1.38, -0.92) 13.57, 13.04, 89 1.04 (0.84, 1.61) Six minute Walk Test Distance (meters) -232, 133, 82 -1.74 (1.60, 1.97) 207, 109, 61 1.90 (1.46, 2.39) * Negative sign indicates a worsening in the measured value; positive sign indicates an improvement in the measured value SD, Standard Deviation SRM, Standardized Response Mean n, Number of subjects CI, Confidence Intervals Table 5 Missing Values Details Measure Time 2 Eligible n = 119 Time 3 Eligible n = 93 Self-paced Walk Completed Test 115 92 Unable to Complete Test 4 0 Missing 0 1 Stair Test Completed Test 87 91 Unable to Complete Test 29 1 Missing 3 1 Timed Up and Go Test Completed Test 116 91 Unable to Complete Test 3 0 Missing 0 2 Six minute Walk Test Completed Test 82 87 Unable to Complete Test 33 1 Missing 4 5 Discussion This study has provided information concerning the measurement properties of four performance measures used to complement information concerning lower extremity functional status in patients with advanced OA undergoing THA or TKA. The test-retest reliability component of this study was conducted over a median interval of 178 days, which is a longer period than would typically be chosen to assess stability. This extended reassessment interval was chosen to accommodate the fact that random measurement error is often time dependent, and in practice, the period between clinical visits is often greater than several months [ 33 ]. A potential concern when applying a reassessment of this duration is that true change in the sample will occur; however, in this study the LEFS MDC 90 was applied to further define a stable patient sample. The reliability coefficients (Table 2 ) for the time and distance components of the tests met or exceeded 0.90 with the exception of the TUG. They are believed to represent conservative estimates of the reliability likely to be associated with most clinical reassessment intervals. It is important to remember that the reliability of a measure intended for individual patient application must be greater than the reliability of a measure designed for group use [ 34 ]. Different authors have advocated different standards for individual patient use, Nunnally [ 34 ] recommended 0.95, Kelley [ 35 ] 0.94 and Weiner and Stewart suggested 0.85 [ 36 ]. Although the reliability of the TUG at 0.75 would meet the standards for group application, it would not meet the aforementioned standards for individual patient use. The SPWT, ST and 6MWT would meet one or all of these standards. In reviewing the mean and quartile scores of the performance measures preoperatively (Table 3 ), the scores indicate higher function than those reported in other studies [ 14 , 16 , 17 ], including the findings from our own prior work which examined a large dataset of over 1800 patients [ 11 ]. One potential explanation for these findings may have been the age of our sample, 25% of the patients were 57 or younger. As noted in the Canadian Joint Replacement Registry, the numbers of THA and TKA in the 45–54 year age group has increased between 1994/1995 to 1999/2000 [ 25 ]. A second factor potentially accounting for the preoperative scores is the nature of the study. Individuals who could not complete all the performance measures preoperatively would not be included, thereby filtering out the individuals with the highest disability. To be useful in clinical practice, the scores obtained on outcome measures must have meaning to clinicians. In this study, the SEM was used to identify the error associated with a patient's reported score and to estimate the value of MDC 90 . Because the SEM is reported in scale points, it enhances the interpretability of a patient's score and change score. To the authors' knowledge this is the first study to provide estimates for MDC 90 for each of the four physical performance measures in the hip/knee end stage OA-arthroplasty population. These benchmarks will assist clinicians to more effectively monitor change in these types of patients. Using a different methodology, Redelmeier et al [ 37 ] determined the smallest difference in the 6MWT associated with a noticeable difference in perceived walking ability for COPD patients to be a distance of 54 meters. Using this as a benchmark in arthroplasty patients would underestimate the distance required to be confident that a change had truly occurred. This illustrates the importance of population specificity when determining MDC 90 . Many studies assessing change have focused on improvement only; the current investigation assessed deterioration and improvement [ 14 , 21 , 38 , 39 ]. Based on prior work, it was hypothesized that surgical intervention would induce a reduction in lower extremity functional status when assessed within 16 days of surgery [ 9 ]. All time/distance performance measures demonstrated deterioration over this interval. Subsequently all of the measures demonstrated significant improvements between the first and second postoperative visits. These findings suggest that the four performance measures are adept at assessing both types of change. The greatest changes were associated with the ST and 6MWT. Examination of the SRMs for these two tests demonstrated similar responsiveness over the studied time intervals. This parallels the findings in the study by Parent et al [ 14 ] examining early recovery after TKA using locomotor tests, including gait speed, stair ascent cycle duration, and the 6MWT. Of these measures, the authors found the 6MWT to be most responsive over the study's three time points, ranging from preoperatively to 4 months postoperatively. Of interest, the stair ascent cycle duration, measured using a 2-dimensional biomechanical analysis system was least responsive and the authors recommended evaluating the responsiveness of a timed stair measure, which has been accomplished in this study. In addition to providing information concerning the psychometric properties of the performance measures, our results also offer insights into the clinical application of these measures. The TUG was originally developed to easily evaluate the risk of falls using balance and basic functional mobility [ 8 ]. Tested in the frail elderly population, scores under 10 seconds were associated with individuals who were functionally independent [ 26 ]. Considering this benchmark and normative values reported for community dwelling elders [ 40 ], the patients' mean TUG score, in this sample, did not demonstrate much disability. Consequently, there would not be as much opportunity for detecting change. However, the usefulness of the TUG in an elderly orthopaedic population, including patients post THA and TKA, has been reported. [ 13 ]. In considering the SPWT and the 6MWT, it is not surprising that the 6MWT demonstrated greater responsiveness in this study, as it was measured over a longer distance and duration. Unlike the SPWT, which in this study was used to determine fast walking speed, the 6MWT has both speed and endurance components. However, as apparent in Table 5 , the TUG and SPWT tests might be preferred if the goal was measurement in the early acute post-operative phase when patients deteriorate and may be unable to perform the ST or 6MWT. This was the case for over 25% of the current study's sample when assessed within 16 days of surgery. Therefore, the time period of administration and the patient's preoperative level of disability can serve as useful guides for clinicians faced with the decision of choosing the most informative measures. This study has several limitations. As apparent in the tables, different numbers of patients were assessed at postoperative assessment one and two. This is partially a reflection of the study design, as mentioned earlier, not all patients were assessed at the same time points due to the goals of the larger ongoing observational study. However, some patients were also missed at both time points due to unexpected changes in appointments without communication to the investigators. Referral bias might also be a potential concern due to the nature of the institution being a specialized tertiary care facility. This must be balanced against the fact that it is one of the largest joint arthroplasty centers in Canada and draws from a wide catchment area. Considering the higher preoperative function of the patients in this sample, it will be important to replicate the current study's findings in different settings with other samples of arthroplasty patients. In addition, as responsiveness is a highly contextualized attribute [ 19 ], it would be informative to study the results over additional time points in the postoperative continuum. Conclusions This study has examined selected psychometric properties in four commonly used performance measures to assess change in the end-stage OA-arthroplasty population. The test-retest reliability estimates of the SPWT, ST and 6MWT met the requisite standards for making decisions at the individual patient level. All of the measures were responsive to detecting deterioration and improvement in the early postoperative time period following arthroplasty. The time period of administration and the patient's preoperative level of disability can serve as useful guides for clinicians faced with the decision of choosing the most informative measures. Estimates of MDC 90 have been reported for each of the performance measures to assist clinicians in assessing change. Competing interests The author(s) declare that they have no competing interests. Authors' contributions DMK conceived and designed the study, assisted with the statistical analysis and prepared the manuscript. PWS assisted with the design, performed the statistical analysis and assisted with the manuscript preparation. JW consulted in the conception and design of the study and assisted with the manuscript preparation. JDG assisted with the design and execution of the study and manuscript preparation. DP assisted in the coordination of the study, data collection and assisted with the manuscript preparation. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549207.xml

548131

Immunoreactivity of the Mycobacterium avium subsp. paratuberculosis 19-kDa lipoprotein

Background The Mycobacterium tuberculosis 19-kDa lipoprotein has been reported to stimulate both T and B cell responses as well as induce a number of Th1 cytokines. In order to evaluate the Mycobacterium avium subsp. paratuberculosis ( M. avium subsp. paratuberculosis ) 19-kDa lipoprotein as an immunomodulator in cattle with Johne's disease, the gene encoding the 19-kDa protein (MAP0261c) was analyzed. Results MAP0261c is conserved in mycobacteria, showing a 95% amino acid identity in M. avium subspecies avium , 84% in M. intracellulare and 76% in M. bovis and M. tuberculosis . MAP0261c was cloned, expressed, and purified as a fusion protein with the maltose-binding protein (MBP-19 kDa) in Escherichia coli . IFN-γ production was measured from 21 naturally infected and 9 control cattle after peripheral blood mononuclear cells (PBMCs) were stimulated with a whole cell lysate (WCL) of M. avium subsp. paratuberculosis or the recombinant MBP-19 kDa. Overall, the mean response to MBP-19 kDa was not as strong as the mean response to the WCL. By comparison, cells from control, non-infected cattle did not produce IFN-γ after stimulation with either WCL or MBP-19 kDa. To assess the humoral immune response to the 19-kDa protein, sera from cattle with clinical Johne's disease were used in immunoblot analysis. Reactivity to MBP-19 kDa protein, but not MBP alone, was observed in 9 of 14 infected cattle. Antibodies to the 19-kDa protein were not observed in 8 of 9 control cows. Conclusions Collectively, these results demonstrate that while the 19-kDa protein from M. avium subsp. paratuberculosis stimulates a humoral immune response and weak IFN-γ production in infected cattle, the elicited responses are not strong enough to be used in a sensitive diagnostic assay.

Background Paratuberculosis (Johne's disease) is caused by Mycobacterium avium subsp. paratuberculosis (referred to hereafter as M. avium subsp. paratuberculosis ) and induces a chronic enteritis in ruminants. The disease signs include weight loss, diarrhea, and decreased milk production. In the United States alone the economic burden of Johne's disease is estimated at over $200 million in lost annual revenue to the dairy industry [ 1 ]. Prevalence studies in the United States have estimated that between 20 to 30% of dairy herds are infected with M. avium subsp. paratuberculosis [ 2 , 3 ]. Neonatal calves are most susceptible to infection and are likely to become infected after ingestion of contaminated milk or colostrum [ 4 , 5 ]. During the subclinical stage of infection, the host cell-mediated immune response is robust and appears to control the infection. As the disease progresses from the subclinical to the clinical stage, the cell-mediated response diminishes, and a humoral immune response predominates [ 6 ]. Vaccines are not completely protective, but have been reported to reduce fecal shedding and delay the onset of clinical disease [ 7 - 9 ]. Lipoproteins have long been considered immunomodulators and mycobacteria are especially rich in these post-translationally modified proteins. There are approximately 100 open reading frames identified in the M. tuberculosis genome that possess a characteristic amino-terminal acylation motif [ 10 ]. The 19-kDa lipoprotein from Mycobacterium tuberculosis is immunodominant in both mice [ 11 - 13 ] and humans [ 14 , 15 ] and has been shown to stimulate CD4 + T cell proliferation as well as the release of IL-2, IFN-γ, and IL-12 [ 16 , 17 ]. Acylation near the N-terminal portion of the 19-kDa protein is believed to occur at amino acids 19–24 and contributes to its immunogenicity [ 12 ]. Furthermore, glycosylation of the M. tuberculosis 19-kDa protein inhibits innate immune responses, such as the release of TNF-α, IL-6, and IL-10 from macrophages, but does not affect antibody binding [ 18 - 20 ]. The 19-kDa protein was also shown to induce CD8+ cells to secrete IFN-γ and specifically lyse M. tuberculosis -infected monocytes [ 21 ] as well as promote neutrophil priming and activation [ 22 ]. Finally, B cell epitopes have been described that localize to the linear sequences of amino acids 11–30, 29–47, 61–80, and 140–159, as well as a conformation-dependent epitope at the amino and carboxy-terminal ends because of intramolecular disulfide bonding of cysteine residues [ 21 , 23 ]. Experimental infection and staining of macrophages has shown that the 19-kDa protein is secreted by live M. tuberculosis residing within the phagolysosomal compartment [ 24 ]. Homologues of the 19-kDa lipoprotein exist in M. bovis , M. avium , and M. intracellulare but are absent from M. phlei , M. smegmatis , M. fortuitum , M. gordonae , and M. leprae [ 25 ]. Despite decades of research, little is known about the M. avium subsp. paratuberculosis proteins involved in metabolism, cell wall synthesis, macrophage entry and survival, disease pathogenesis, or host immune evasion. However, several antigens have recently been identified and their immunogenicity examined by serodiagnostic and/or lymphocyte stimulation assays [ 26 - 30 ]. With the genome sequence of M. avium subsp. paratuberculosis recently defined [ 31 ], all proteins produced by this pathogen are now identified and can be characterized. Novel opportunities arising from the genome sequence of M. avium subsp. paratuberculosis enable us to select and characterize genes of interest. A major goal of this laboratory is to define a complete catalog of immunodominant antigens in M. avium subsp. paratuberculosis . With the immunostimulatory capabilities of the M. tuberculosis 19-kDa antigen in mind, the objective of this study was to determine if the 19-kDa protein of M. avium subsp. paratuberculosis possessed a similar capacity. In this study, the 19-kDa lipoprotein from M. avium subspecies paratuberculosis was cloned, expressed, and characterized. In addition, the purified recombinant protein was used to assess cellular immune responses in subclinically infected cattle as well as humoral immune responses in cattle with clinical Johne's disease. Results Sequence analysis of the mycobacterial 19-kDa coding region The 19-kDa coding sequence was identified from the M. avium subsp. paratuberculosis genome project as MAP0261c. Comparison of amino acid sequences from other species of mycobacteria show that this gene product is conserved. Sequence alignment shows the N-terminal half is more variable and the region between amino acids 99 and 123 is highly conserved (Figure 1 ). MAP0261c displays a 95% amino acid identity in M. avium , 84% in M. intracellulare and 76% in M. bovis and M. tuberculosis . MAP0261c has a G+C content of 66.2% and encodes for 161 amino acids with a predicted molecular mass of 15.2 kDa. The first 22 amino acids of the M. tuberculosis 19-kDa protein are hydrophobic and were previously noted to represent a signal peptide that is post-translationally cleaved to expose an N-terminal cysteine [ 32 ]. Signal peptidase cleavage analysis of MAP0261c (SignalP3.0; ) detected a signal peptide generated from a putative cleavage site between amino acids 34 and 35, to expose an N-terminal serine (Figure 1 ). The SignalP-NN (neural networks) model assigned the highest cleavage probability values to amino acid 35 (C score = 0.324; Y score = 0.439) with the predicted length of the signal peptide being 34 amino acids. This is slightly longer than most signal sequences which range from 18 to about 30 amino acid residues in length [ 33 ]. The predicted peptidase cleavage sight for M. tuberculosis is between amino acids 21 and 22 and therefore falls within this range (Figure 1 ). Despite a predicted signal peptidase cleavage site, PSORTb analysis software could not predict if the protein was cytoplasmic or membrane located. Cloning and expression of the M. avium subsp. paratuberculosis 19-kDa protein In order to perform immunoassays with purified 19-kDa protein, MAP0261c was amplified from M. avium subsp. paratuberculosis genomic DNA and cloned into the pMal-c2 expression vector and transformed in E. coli . Induced expression resulted in production of a maltose binding protein (MBP)-19 kDa fusion protein that was affinity-purified from E. coli lysates. MBP-19 kDa was analyzed by SDS-PAGE (Figure 2 ) to assess yield, purity and size. The predicted mass of MBP alone is 42 kDa, while the predicted mass of the MBP-19 kDa fusion protein is 56 kDa. The purified MBP-19 kDa protein migrated to a position around 50 kDa in SDS-PAGE (Figure 2A ). Approximately 5 mg of purified protein was easily obtained from a 500-ml broth culture at O.D. 600 nm = 0.9. The purified protein was further characterized by immunoblot analysis using a monoclonal antibody (mAb) that detects the MBP affinity tag (Figure 2B ). Both the fusion protein and MBP alone are detected by the mAb. In addition, the fusion protein is expressed at higher levels under inducing conditions. Immunoblot analysis of the 19-kDa protein The M. tuberculosis 19-kDa protein is known to be immunodominant, therefore immunoblot analysis was performed to determine if cattle naturally infected with M. avium subsp. paratuberculosis produce antibodies against the 19-kDa protein. Immunoblots were probed with sera from 9 non-infected and 14 clinically infected cattle. Sera from 8 of 9 non-infected cattle did not react with either MBP or MBP-19 kDa, while 1 non-infected cattle weakly recognized both MBP and MBP-19 kDa protein (data not shown). By comparison, sera from 9 of 14 infected cattle reacted specifically with the 19-kDa protein, but not MBP alone. Antibody reactivity to the 19-kDa protein from 3 clinical cows is shown in Figure 3 . Sera from the remaining five infected cattle detected both MBP and MBP-19 kDa proteins. This result made it difficult to distinguish if sera from the animal recognized the MAP0261c gene product or if it simply recognized the MBP affinity tag. Collectively, these data suggest that the 19-kDa protein is detectable and immunogenic in cattle with Johne's disease. IFN-γ responses of infected cattle As an indicator of the cell-mediated responses of infected cattle to the 19-kDa protein, whole blood containing PBMCs from 9 control and 21 infected cows was stimulated with M. avium subsp. paratuberculosis sonicated whole cell lysate (WCL), MBP, or MBP-19 kDa and IFN-γ production was assayed by ELISA. These cattle were selected from a larger group because they showed no IFN-γ stimulation in response to MBP alone. IFN-γ production in response to WCL stimulation allowed for the segregation of infected cattle into three groups: suspect, positive, and high positive. After subtracting the IFN-γ responses of non-stimulated cells, suspect animals had less than 0.1 absorbance units of IFN-γ production, positive animals had 0.1 – 0.3 absorbance units of IFN-γ production, and high positive animals had more than 0.3 absorbance units of IFN-γ production. IFN-γ responses by blood mononuclear cells from infected cattle exceeded responses from control cattle for both the WCL and MBP-19 kDa (Table 1 . Significant differences (P < 0.05) were found between control and high positive groups for both WCL and MBP-19 kDa protein stimulation. However, direct comparisons of the two antigen preps using mononuclear cells from the same animal clearly showed the WCL was a stronger stimulator of IFN-γ production (Table 1 ). Discussion A majority of the research on individual mycobacterial proteins has been performed in M. tuberculosis , whereas little is known about the M. avium subsp. paratuberculosis proteome. Indeed, all currently available antigen-based diagnostic tests for Johne's disease use an undefined mixture of proteins, such as purified protein derivative (PPD) or WCL, which may not be specific for M. avium subsp. paratuberculosis . Recent completion of the M. avium subsp. paratuberculosis genome has already advanced efforts to identify novel antigens [ 26 , 34 ]. Furthermore, the genome will be a critical resource in proteomic studies directed at defining the proteins present in mixtures such as johnin PPD. The present study was performed in order to characterize the M. avium subsp. paratuberculosis 19-kDa protein, as well as assess its immunostimulatory capabilities in cattle. In this study, we show that the 19-kDa protein of M. avium subsp. paratuberculosis can be readily overexpressed as a fusion protein in E. coli . This is not true for many other proteins encoded by M. avium subsp. paratuberculosis [ 34 ] and suggests the putative lipoprotein is not toxic to E. coli . Previous studies have suggested the M. tuberculosis 19-kDa protein undergoes posttranslational modification by the addition of fatty acids to form a lipoprotein [ 35 ]. Although not demonstrated directly by our studies, it is possible that the 19-kDa was not posttranslationally modified by the heterologous E. coli host. It is unclear whether posttranslational modification of this protein would affect its immunological activity. The recombinant antigen was detected by sera from cattle with Johne's disease; however, it was not as strong a stimulator of proliferative T-cell responses as has been reported for its counterpart in M. tuberculosis [ 36 ]. Furthermore, the 19-kDa protein from M. tuberculosis was shown to induce both cellular and humoral immune responses from mice and humans [ 11 , 14 ]. These studies, combined with the present study, may suggest that acylation is more important in cell-mediated immune responses than in the humoral immune response. It is generally accepted that cellular and humoral immune responses of M. avium subsp. paratuberculosis -infected cattle are biphasic, with IFN-γ responses detected early and antibody responses detected late in infection. However, evidence suggests that an unknown M. avium subsp. paratuberculosis protein can be detected by antibodies from cattle just 3 weeks after infection [ 37 ]. As a measure of cellular immune responses, blood mononuclear cells from infected cattle were stimulated with both the recombinant 19-kDa protein and a whole-cell sonicated lysate of M. avium subsp. paratuberculosis (WCL), and IFN-γ production was measured. Results from this study suggest that while the 19-kDa protein is a stimulator of IFN-γ production, it is not as potent when compared to WCL. Additionally, we found that a majority (9 of 14) of infected cattle produced antibodies to the 19-kDa protein, as determined by immunoblot analysis. By comparison, sera from the majority of control, non-infected cattle (8 of 9) did not react to the 19-kDa protein. The single non-infected cow that did show reactivity to the MAP0261c gene product may be attributed to exposure of environmental mycobacteria such as M. avium subsp. avium , which has a similar protein (Figure 1 ). Sera from four clinical cows reacted to the MBP protein, but this reactivity was extremely weak. MBP is found in environmental E. coli and likely accounts for reactivity seen in some cattle. In order to avoid potential cross-reactivity with MBP, we attempted to cleave the MBP portion from the MBP-19 kDa fusion protein, but cleavage was not 100% efficient (data not shown). The M. tuberculosis 19-kDa protein was reported to contain a 21 amino acid N-terminal signal peptide [ 22 ], however our SignalP analysis for M. avium subsp. paratuberculosis identified a putative 34 amino acid N-terminal signal sequence. Furthermore, the computer algorithm PSORTb predicts a putative signal sequence, but it cannot determine if the protein is actually secreted. Antibodies will be produced against the recombinant MBP-19 kDa protein to determine if the protein is secreted by M. avium subsp. paratuberculosis . Conclusions The results from this study show that the recombinant 19-kDa protein stimulates a weak host immune response in infected cattle. The 19-kDa protein may be used in conjunction with other antigens from M. avium subsp. paratuberculosis to identify infected cattle but should not be used as a "stand alone antigen" in new diagnostic assays. Methods Bacterial strains and culture conditions M. avium subsp. paratuberculosis strain 19698-1974 (originally isolated in 1974 from a clinical cow housed at the National Animal Disease Center, Ames, IA) was grown in Middlebrook 7H9 liquid media (pH 6.0) supplemented with 10% oleic acid albumin dextrose complex (Becton Dickinson Microbiology, Sparks, MD), 0.05% Tween 80 (Becton Dickinson Microbiology), and 2 mg/ml mycobactin J (Allied Monitor Inc., Fayette, MO). M. avium subsp. paratuberculosis cultures were grown to log phase at an optical density 540 nm (OD 540 ) of 0.4, at 37°C without shaking. Escherichia coli DH5α cells were routinely grown in Luria-Bertani (LB) broth or LB agar plates at 37°C supplemented with ampicillin (100 μg/ml) for selection. Cattle The Johne's Disease Research Project at the National Animal Disease Center has a repository of sera from cattle that were euthanized with clinical signs of Johne's disease, which included shedding, weight loss and diarrhea. Fecal samples from each of these clinical cattle were found to contain more than 100 CFU per gram of feces as determined by colony counts on Herrold's egg yolk media (HEYM) agar slants by standard culture methods [ 38 ]. Sera from 14 clinical cattle were selected for immunoblot analysis. The National Animal Disease Center also maintains a small herd of non-infected cattle as well as a herd of cattle naturally infected with M. avium subsp. paratuberculosis . The animals used in IFN-γ experiments were placed in four groups consisting of 9 non-infected healthy cows, 9 suspect subclinical cows (IFN-γ responses < 0.1), 6 positive subclinical cows (IFN-γ responses 0.1 – 0.3), and 6 high-positive cows (IFN-γ responses > 0.3). The non-infected control cows were characterized by repeated negative fecal cultures performed quarterly over a 3- to 5-year period. In addition, these animals were negative on all immunological assays (i.e. ELISA and IFN-γ production) performed during that period. Subclinical cattle (suspect, positive, and high-positive) were characterized by shedding less than 10 CFU/g of feces and were intermittently positive by IFN-γ assays (response > 0.1) performed quarterly over a 3- to 5-year period. The institutional Animal Care and Use Committee approved all animal procedures described in this study. Comparison of the mycobacterial 19 kDa coding region The nucleotide sequences for the 19 kDa coding regions from M. tuberculosis , M. bovis , M. avium and M. intracellulare were obtained from the NCBI nucleotide sequence database. The sequences were assembled and compared using MegAlign software (DNASTAR, Inc., Madison, WI). Cloning and expression of the M. avium subsp. paratuberculosis 19-kDa gene A maltose binding protein (MBP) fusion of the M. avium subsp. paratuberculosis 19-kDa sequence (MBP-19 kDa) was constructed using the pMAL-c2 vector (New England Biolabs, Beverly, MA). To amplify the 19-kDa coding region from M. avium subsp. paratuberculosis , primers were designed directly from the MAP0261c sequence. MAP0261c was amplified with Expand High Fidelity PCR system using the primers 19-kDa-pMal5' (5'-GCGC CAGCTG ACGATCGCGGTCGCGGGCGCGGC-3') and 19-kDa-pMal3' (5'-GCGC AAGCTT CAGGTCACATCGATCTCGAAC-3'). The 19-kDa-pMal5' and 19-kDa-pMal3' primers contained Pvu II and Hin d III restriction sites (underlined) for cloning, respectively. This primer set amplified the 19-kDa coding sequence minus the N-terminal 4 amino acids and the C-terminal 3 amino acids. The pMal-c2 vector was digested with Xmn I and Hin d III and the PCR amplicon was digested with Pvu II and Hin d III. The two products were ligated overnight at 12°C with T4 DNA ligase (Life Technologies Inc., Rockville, MD), which resulted in an in-frame fusion between the vector-encoded malE gene and a majority of the 19-kDa gene. The resulting recombinant plasmid, designated pMal-19-kDa, was transformed into E. coli DH5α competent cells. Recombinant clones were selected by plating on LB-ampicillin plates overnight at 37°C. Individual clones were picked and inserts were confirmed by DNA sequencing. The resulting fusion protein was overexpressed and purified by maltose affinity chromatography using amylose resin (New England Biolabs). A detailed expression and purification protocol has been published previously [ 39 ]. Expression and purification of the MBP-19 kDa fusion protein was monitored by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels either stained with GelCode Blue (Pierce Biotechnology Inc., Rockford, IL) or checked by immunoblot analysis with a monoclonal antibody against MBP developed at the National Animal Disease Center. Expression and purification of the MBP alone (without 19-kDa) was previously described [ 26 ]. SDS-PAGE and immunoblotting E. coli lysates expressing MBP-19 kDa were prepared as previously described [ 40 ]. SDS-PAGE was performed using 12% (wt/vol) polyacrylamide gels. Proteins were electrophoretically transferred onto nitrocellulose membranes (Schleicher and Schuell, Keene, NH) using the Trans Blot Cell (Bio-Rad Laboratories, Hercules, CA) in sodium phosphate buffer (25 mM; pH7.8) at 0.9 amps for 90 minutes. After transfer, the blots were blocked overnight with PBS plus 2% bovine serum albumin (BSA) and 0.1% Tween 20 (PBS-BSA). For immunoblots, serum from cattle with Johne's disease was diluted 1:500 in PBS-BSA. Sera were incubated on the blots at room temperature for 2 hours. After 3 washes in PBS plus 0.1% Tween 20, blots were incubated for 1.5 hours in anti-goat peroxidase-conjugated secondary antibody diluted 1:20,000 in PBS-BSA (Pierce Biotechnology Inc.). After secondary antibody incubation, the blots were washed 3 times as described above and were developed for chemiluminescent detection using SuperSignal detection reagents (Pierce Biotechnology Inc.). IFN-γ assays Blood was collected from the jugular vein of subclinically-infected cattle into sodium heparin vacutainer blood collection tubes. One ml aliquots of whole blood from each animal were plated into 4 wells of 24-well culture plates and cultured alone (non-stimulated) or with 10 μg/ml of M. avium subsp. paratuberculosis sonicate (WCL), 10 μg/ml of MBP, or 10 μg/ml of MBP-19 kDa. The WCL was prepared by sonication of bacilli and centrifugation exactly as described previously [ 37 ]. Blood-antigen mixtures were incubated for 18 hours at 39°C in a 5% CO 2 humidified atmosphere. Plates containing blood-antigen samples were centrifuged at 500 × g for 15 minutes and the plasma was harvested from each well. Plasma samples were frozen at -20°C until being analyzed for IFN-γ concentration by enzyme-linked immunosorbent assay (ELISA) using a commercial kit (Bovigam, BioCor, Omaha, NE) as recommended by the manufacturer. Samples were analyzed in duplicate and were determined to be positive for IFN-γ production if the absorbance of the stimulated sample (WCL, MBP, MBP-19 kDa) was 0.1 units greater than the absorbance of the nonstimulated well for that animal. This classification of IFN-γ positive samples has been previously reported by our laboratory as well as others [ 41 , 42 ]. Statistical analysis ANOVA and unpaired t tests were performed to analyze the IFN-γ stimulation data. Analyses were performed to compare average stimulation of control, non-infected cattle to the infected cattle groups (suspect, positive, high positive). Differences were considered significant when P < 0.05. Authors' contributions JFH carried out all the experiments and drafted the manuscript as part of his PhD dissertation. JRS provided advice, participated in its design and coordination and helped edit the manuscript. JPB conceived of the study, participated in its design and helped to draft the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548131.xml

535864

Family physicians' perspectives on practice guidelines related to cancer control

Background Family physicians (FPs) play an important role in cancer control. While FPs' attitudes towards, and use of guidelines in general have been explored, no study has looked at the needs of FPs with respect to guidelines for the continuum of cancer control. The objective of this study was to understand which guideline topics FPs consider important. Methods Five group interviews were conducted by telephone with FPs from across Ontario, Canada. Transcripts were analyzed inductively. Content analysis identified emergent themes. Themes are illustrated by representative quotes taken from the transcripts. Results The main areas where FPs felt guidelines were needed most included screening – a traditional area of responsibility for FPs – and treatment and follow-up – areas where they felt they lacked the knowledge to best support patients. Confusion over best practice when faced with conflicting guidelines varied according to disease site. FPs defined good guideline formats; the most often cited forms of presentation were tear-off sheets to use interactively with patients, or a binder. Computer-based dissemination was acknowledged as the best way of widely distributing material that needs frequent updates. However, until computer use is a common aspect of practice, mail was considered the most viable method of dissemination. Guidelines designed for use by patients were supported by FPs. Conclusions Preferred guideline topics, format, dissemination methods and role of patient guidelines identified by FPs in this study reflect the nature of their practice situations. Guideline developers and those supporting use of evidence-based guidelines (e.g., Canadian Strategy for Cancer Control) have a responsibility to ensure that FPs are provided with the resources they identify as important, and to provide them in a format that will best support their use.

Background Family physicians (FPs) play an important role in cancer control. Their traditional involvement has been primarily focused on opposite ends of the cancer control continuum: prevention, screening and diagnosis at the beginning of the continuum, and provision of palliative care at the other end. Treatment and follow-up have typically been the responsibility of secondary or tertiary care physicians. There are indications that FPs would like their traditional roles to include involvement in treatment and follow-up [ 1 , 2 ], and that they are, in fact, becoming more involved in these areas [ 3 , 4 ]. A survey of Canadian FPs showed that they felt a significant proportion of long term follow-up care could be transferred to the FP decreasing the burden on consultants [ 5 ]. For example, in a randomized trial of routine follow-up of breast cancer patients, the care provided by FPs was found to be equivalent to that of an oncology specialist [ 6 ]. A concern, however, with increased involvement is that FPs receive little oncological training in medical school [ 4 ], and thus, are not adequately prepared for involvement in certain aspects of cancer care particularly as treatment practices change as new evidence emerges. Clinical practice guidelines (CPGs) can provide FPs with information and guidance on evidence-based best practices. While FP attitudes on the use of CPGs have been shown on the whole to be positive [ 7 - 11 ], FPs have suggested that involvement of FPs in the development of guidelines that take into account the nature of primary care would improve their uptake among FPs [ 11 - 14 ]. Dowswell et al[ 11 ] suggested that rather than ask how to get physicians to follow guidelines, it would be more productive to ask physicians about their information needs and how they would like them met. A broad partnership of key stakeholders known as the Canadian Strategy for Cancer Control (CSCC) is currently working towards a national cancer control strategy. The CSCC initiative builds upon previous work done in Canada, and has the goal of developing, adopting and implementing a national strategy. A focus on "Guidelines and Standards" is one of the CSCC's top priorities with an aim to establish mechanisms and improve capacity for collaborative guideline and standards development [ 15 ]. In order to ensure wise use of resources, it will be important to ascertain the needs of FPs in Canada with respect to CPGs relevant to cancer control. While many studies have focused on cancer screening [ 16 - 18 ] and palliative care [ 19 , 20 ], to our knowledge, no study has looked at what FPs consider important in terms of content, format and dissemination of guidelines related to cancer control activities. The objective of this study was to learn the views of Ontario FPs for the furtherance of cancer control efforts by the provincial cancer agency, Cancer Care Ontario. The information derived from the study is also of benefit for national cancer control efforts through the CSCC. Methods In order to learn FPs' views on guidelines on cancer control, a qualitative methodology was considered the most appropriate. To facilitate FPs' involvement (e.g., eliminate travel and time barriers) and ensure representation of FPs practicing in remote areas, the most feasible option was to hold interviews by teleconference [ 21 ]. Thus, we conducted the process as key informant interviews involving between two and four participants [ 22 , 23 ] and using a semi-structured interview format. With respect to recruitment, FPs from various regions in Ontario identified through the Canadian Medical Directory were recruited via an information letter and follow-up telephone call. In addition, FP colleagues identified potential participants or individuals who might suggest potential participants. Sampling was purposeful [ 23 ]. We felt that it was important to involve FPs from the different regions in Ontario (northern, eastern, central east, southwest and central west), as well as have both urban and rural representation, as needs with respect to guidelines may differ based on these characteristics. Once a commitment to participate was made, a time and date for the teleconference were established. FPs were then faxed a consent form and a list of questions that would be asked during the interview. To allow for in-depth discussion, three cancer disease sites were selected: lung, colorectal and cervix. The first two were included because of high incidence; the latter because conflicting screening guidelines [ 21 ] currently exist [ 24 - 27 ]. Questions focused on preferred topics for guidelines along the cancer control continuum, preferred format and method of dissemination of guidelines, and perspectives on guidelines written for patients. Questions were tested for clarity and coverage of important issues in a pilot session involving four FPs and the moderator (LZ). As a result of the pilot, it was decided that four participants were the maximum number for each session to ensure that the session was not too long – an important consideration for physicians with busy schedules. Five group interviews were held. There was a minimum of 2 and maximum of 4 FP participants in each session. Each of the five interview sessions lasted between 45 and 60 minutes. In order to accommodate FPs' schedules, three of the five interviews were held in the early evening, and two in the morning. FPs were asked to select either a lottery ticket or telephone card as a small acknowledgement of their participation. Each session was led by the same individual (LZ), an experienced qualitative researcher. Interviews were audio-taped. Audio tapes were transcribed immediately after each interview, and underwent a preliminary analysis allowing for emergent issues or ideas to be explored in future sessions. Transcripts were read by one of the researchers (LZ); latent content analysis (coding and classification into themes) was done manually [ 22 ]. As each question addressed a specific topic, the question topics themselves acted as broad organizing categories. Relevant transcript sections were marked and assigned code words. Codes of similar type and content were combined into sub-categories within each question topic [ 28 ]. A second researcher (EG) was given two transcripts to read and code independently using the list of codes and categories. Coding was compared. Where necessary, definitions of code words were refined and categories expanded upon [ 29 ]. Disagreements and differences were resolved through discussion and data was re-examined where necessary [ 30 ]. Representative quotes were selected from the transcripts in order to illustrate key issues raised by the participants. Ethical approval for this study was obtained from the Ottawa Hospital Research Ethics Board. Results Of the 13 physicians participating in the study, seven were male and six were female; five practiced in a rural setting, the remainder in an urban setting. The majority (9/13) were in group practice. Approximately 5% of FPs contacted agreed to participate in the study. Overall, topics raised by urban and rural FPs were similar except with respect to guidelines on cancer treatment. Guideline topics Using lung, colorectal and cervical cancers as exemplars, FPs were asked for which topics along the cancer control continuum (i.e., prevention, screening, diagnosis, treatment, follow-up, or palliation) they most wanted guidelines. Screening was the topic most frequently mentioned, although reasons behind requests for screening guidelines differed for each disease site. For lung cancer, there are currently no evidence-based screening tools or maneuvers, and no screening guidelines. FPs were uncertain whether to routinely screen for lung cancer in their practices. For colorectal cancer, on the other hand, there are a number of screening tools and a number of recommendations made by different organizations. Conflicting guidelines resulted in FPs being uncertain about what to do in practice. I'm certainly much less certain of the area of screening for colorectal cancer. I mean there's a lot of different guidelines out there, it depends on who you read and I regard that as an area very much in flux...I still remain a bit confused as to who should have what. (FP 6 ) FPs also commented that a similar confusion was prevalent with regard to screening guidelines for cervical cancer. After three normal [screens], every two years and discontinue at the age of 70, that's what it says. And then this other one says start at age 18, and after three normals then do it every three years except high risk patients should have annual smears...the American College of OB/GYN recommends its smears always continue annually. The American Cancer Society and the Canadian Task Force recommends screening until age 65 and 69 respectively. So, it's a dog's breakfast. (FP 5 ) However, in comparison to colorectal cancer screening, there was much less ambiguity about what to do in practice. FPs readily adapted cervical cancer screening guidelines to suit individual patient situations or demands. I would say I have people who I am willing to see every 3 years because I feel quite confident that they'll be back; they're good at keeping up and the ones that I'm more uncertain about in terms of their follow up, I'll make sure I do it more frequently just in case...For me it varies very much between 1 and 3 years and it is very much a decision of my own. (FP 11 ) In discussing the issue of conflicting guidelines, FPs raised the point that where more than one guideline exists, the credibility of all come into question. For example, if there are multiple guidelines on a topic, can any be the 'right' one to use? FPs were also very interested in treatment guidelines. Interest centered around two situations: decision making with patients, and dealing with side-effects of treatment. In the first, FPs wanted to know about treatment available to their patients diagnosed with cancer. They saw their role as helping patients and families make informed choices. As such, they wanted information on treatment goals, survival rates for different treatments, quality of life issues as well as risk of and dealing with potential side-effects. A lot of the patients I have who go out to a cancer clinic come back and make sure I agree with what they're choosing and the problem is I don't have the information to be able to even aid them in making their decision....So if we had a little bit more information than those flow charts that are 'yes/no' to say [this] is the most recent information for survival rates...that kind of thing would be helpful. (FP 7 ) The second area regarding treatment was raised by rural physicians who often saw cancer patients in emergency departments when, for example, patients were home between chemotherapy cycles. FPs mentioned the difficulty caring for patients when they knew little about their treatment plan. Areas where we really need specific evidence-based guidelines are in treatment and follow-up. I mean, although the patient may disappear to the cancer clinic...they certainly do show up in emergency, and sometimes the husband or wife calls us as well and says, "Well you know they're getting this drug or they're getting this radiation, they're really sick and what do you think about this?" And if we don't even know what they're getting or what the potential side-effects are, it's really hard to be helpful. So, we need specific guidelines. (FP 5 ) Rural physicians were also interested in guidelines for follow-up. ...we are going to be doing more and more of our own follow-up, that's the trend, that's the next century, so we need good guidelines. (FP 5 ) Guidelines were seen by some FPs as a potential communication device between cancer centres and community-based FPs. They suggested that guidelines could be sent to the FP from the cancer centre and include notations by oncologists regarding individual patients. In this way, FPs would feel they had the tools to provide on-going support and care for patients in the treatment or follow-up stages. Format and dissemination Two themes were identified related to guideline format: format aspects and presentation. Regarding format, Table 1 presents FPs' 'definition' of what attributes comprise a good guideline. The outstanding features requested were a combination of brevity, and formatted in such a way that FPs were able to quickly identify relevant content. Table 1 Components of a 'Good' Guideline Dated Has a clearly defined, reputable source Involves FPs in the creation process to ensure its clinical practicality Not too text based (graphics, tables, flowcharts) Clear, non-ambiguous recommendations Well organized Clearly graded as to levels of evidence One guideline from one authorative body (to reduce confusion) Readable in a few minutes Designed so that FPs will use the guideline frequently and become familiar with it One to two pages long The most popular forms of presentation suggested by FPs were a binder that would be easy to update, and tear-off sheets that could be given to the patient but which also provided a review opportunity for the physician through the act of explaining the guideline to the patient. CD ROMs, posters or software packages (guidelines and a recall system for screening) were other suggestions. I like [the] idea of the tear off sheet to use in discussion with patients. I think guidelines are only useful to the extent that we can go over them again and again and actually be familiar with them ourselves, rather than just having them tucked in a big binder with many other guidelines. So, something that can be used with the patient. (FP 3 ) Computer-based dissemination was acknowledged as the best way of distributing material widely and addressing the difficulties and expense of updating material. Presentation by local leaders, CME, fax, small group meetings, and mail were other suggestions. While computers were mentioned most often, FPs emphasized that information needs to be widely disseminated to all physicians. For this reason, mail was still seen as the most viable form of dissemination. Patient guidelines All FPs agreed that guidelines written for patients would be useful, although there was concern that they should be written very clearly, only be available for topics for which there is good evidence, and not be conflicting. They felt patient guidelines would be useful in that they would act as an added voice, giving weight to the FP's recommendation. Guidelines were also seen as useful in countering misinformation brought in by patients (e.g., from the Internet) to the consultation. On the whole, FPs felt that the more information patients had, the better. Three FPs felt that guidelines would encourage patients to take responsibility for their own care; patients could remind their FP if they were due for screening. So I think the biggest effort is to establish the proper guidelines that are accepted by a group of authorities in Canada and then that would make it easier for me to say, "Well, look, this is the actual guideline that is the result of a great deal of research and in fact you really don't need that mammogram at the age of 40"...I think there has to be an effort to make sure that patients are not given conflicting guidelines. (FP 8 ) In terms of content, FPs felt the guideline should echo the FP message. In addition to the tear-off sheets mentioned earlier, ideas for presentation included an educational message played on the telephone when a patient calls, or video messages broadcast on office televisions. Discussion FPs have traditionally been responsible for prevention, screening, and early detection, and palliative care. Of the topics along the cancer control continuum, screening guidelines were most frequently identified by FPs in this study. Screening for cancer is primarily the responsibility of FPs who need to stay informed of changes or conflicts in recommendations. British general practitioners, when interviewed about use of guidelines, said that they referred to guidelines for cases that they encountered either most commonly or most rarely in practice [ 12 ]. FPs' preferences for screening guidelines addressed two different information needs. In the case of lung cancer, where a familiar screening maneuver was not recommended (i.e., chest x-ray), FPs wanted guidelines that addressed what they should do for routine screening. In the case of colorectal cancer, FPs received conflicting messages about screening, and sought guidance as to which recommendation to use. One barrier to guideline use is if guidelines are considered controversial [ 21 , 31 , 32 ]. FPs identified conflicts in recommendations for colorectal and cervical cancer screening. However, they expressed less difficulty in making decisions regarding cervical cancer screening for their patients in comparison with colorectal cancer screening. This may be because cervical cancer screening is a long established practice with good evidence of benefit. Differences between guidelines for cervical cancer relate to the length of interval between routine screening [ 24 , 25 ]. Conversely, colorectal screening is a new practice for which there have been long standing recommendations against routine screening. Currently, differences in recommendations relate to the type of screening maneuver [ 26 , 27 ]. Decisions regarding screening practice also depended on patient factors such as a patient's motivation to adhere to a screening routine. Physician factors (e.g., perceptions of guidelines and clinical experience) and patient factors have been identified as two of the three determinants of a decision making model for cancer screening in the case of unclear or controversial guidelines. The third determinant was the perceived quality of the physician-patient relationship and the clarity of the recommendation being discussed [ 21 ]. As with Feightner et al.[ 33 ], FPs in this study encouraged development of patient versions of FP guidelines. Guidelines were seen as an opportunity to counter misinformation brought in by patients and also as a means having of both physician and patient participate in the patient's care. The literature with respect to improving cancer screening practice shows that interventions that target both the physician and the patient have the greatest impact [ 34 ]. While the FPs in this study play a role in treatment and follow-up, they do not feel they have the information they need to help their patients. Practice location appears to influence the type of treatment information FPs want, although generalizations can not be made on the basis of the qualitative methods used in this study. In urban settings, FPs' roles along the continuum of cancer care principally involve prevention, screening, diagnosis, and palliation. In addition to the above roles, rural physicians are involved in treatment and follow-up, and this involvement is perceived as increasing in the future. Consequently, they state that guidelines on treatment and follow-up would be helpful. Rural physicians participating in a Canada-wide focus group and interview study on FP-oncologist communication also indicated a need for treatment and follow-up information [ 2 ]. While FPs expressed a need for guidelines on cancer follow-up, several guidelines have been published. In the case of colorectal cancer for example, a guideline has been published in Ontario under the auspices of Cancer Care Ontario [ 35 ]. The challenges with respect to guidelines that are not created specifically for FPs include a need to find the best ways of making FPs aware of the existence of such guidelines, and also to provide implementation strategies geared towards the FP practice [ 36 - 39 ]. FPs in this study preferred guidelines in a paper format disseminated by mail rather than electronically. The 2001 Janus survey conducted by the College of Family Physicians of Canada found that only approximately one quarter of FPs across Canada have access to and use computerized CPGs in their office [ 40 ]. The FPs in this study prefer a format that they could use interactively with patients. Recurrent use with patients was seen as a way of helping FPs assimilate the knowledge. Guidelines as a 'look-up' resource and as a general educational tool could become part of the general practitioner's knowledge base [ 12 ]. The findings of this study are limited by several factors. The purpose of the study was to gather information on FPs' perspectives on guidelines along the cancer control continuum. As with all qualitative research, the results of the study are not generalizable beyond the sample. However, the intent of sampling in qualitative research is to identify key informants who will illuminate particular aspects of the research topic [ 23 ]. FPs agreeing to participate in this study are likely those who have a strong interest in guidelines. The perspectives shared by FPs offer insight into the guideline topics, format, and dissemination of guidelines that FPs consider important in caring for their patients with cancer. Further research should focus on identifying the guideline needs of a larger, nation-wide sample of Canadian FPs to ensure that efforts by initiatives such as the CSCC result in CPGs that will be both viewed positively and adopted by FPs. However, perception of the value of a guideline is not enough to ensure adoption. While much effort has gone into guideline development, the focus on dissemination has been through traditional dissemination strategies (e.g., publication in professional journals). Knowledge transfer is known to be more complex [ 41 ] requiring multifaceted strategies to encourage adoption and taking into account the environment, the potential user and characteristics of the innovation (e.g., guideline) [ 36 , 42 , 43 , 39 , 38 ]. Conclusion Guideline topics, format, dissemination, and patient guidelines discussed by FPs in this study reflect their particular practice situations. FPs' strongest preferences were for guidelines on cancer screening, followed by guidelines on treatment that would help them support and provide care for patients. The conflicting messages of some guidelines did not necessarily make decision making problematic for FPs. Rather, it was the reason behind the conflict that created difficulties. FPs saw patient guidelines as an educational tool for both themselves and their patient. Guidelines on treatment and follow-up are available, although they are generally geared towards specialists not FPs. This suggests a need for FP versions to be created. Challenges for Canadian guideline developers/implementers include not only ensuring that the evolving needs of FPs are met, but also that they address the needs of FPs with respect to how that information is formatted, delivered to FPs and how FPs are supported in its use. Competing interests Dr. Graham is a CIHR New Investigator. Authors' contributions LZ participated in the design of the study, conducted the interviews and analysis, and drafted the manuscript. EG participated in the design of the study, the analysis, and contributed to the manuscript. IG participated in the design of the study, and assisted with the pilot test of interview questions. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC535864.xml

554097

Mammalian cells lack checkpoints for tetraploidy, aberrant centrosome number, and cytokinesis failure

Background Mammalian cells have been reported to have a p53-dependent tetraploidy checkpoint that blocks cell cycle progression in G1 in response to failure of cell division. In most cases where the tetraploidy checkpoint has been observed cell division was perturbed by anti-cytoskeleton drug treatments. However, other evidence argues against the existence of a tetraploidy checkpoint. Cells that have failed to divide differ from normal cells in having two nuclei, two centrosomes, a decreased surface to volume ratio, and having undergone an abortive cytokinesis. We tested each of these to determine which, if any, cause a G1 cell cycle arrest. Results Primary human diploid fibroblasts with intact cell cycle checkpoints were used in all experiments. Synchronized cells exhibited G1 arrest in response to division failure caused by treatment with either cytochalasin or the myosin II inhibitor blebbistatin. The role of tetraploidy, aberrant centrosome number, and increased cell size were tested by cell/cell and cell/cytoplast fusion experiments; none of these conditions resulted in G1 arrest. Instead we found that various drug treatments of the cells resulted in cellular damage, which was the likely cause of the arrest. When cytokinesis was blocked in the absence of damage-inducing drug treatments no G1 arrest was observed. Conclusions We show that neither tetraploidy, aberrant centrosome number, cell size, nor failure of cytokinesis lead to G1 arrest, suggesting that there is no tetraploidy checkpoint. Rather, certain standard synchronization treatments cause damage that is the likely cause of G1 arrest. Since tetraploid cells can cycle when created with minimal manipulation, previous reports of a tetraploidy checkpoint can probably be explained by side effects of the drug treatments used to observe them.

Background Cell cycle checkpoints preserve genome integrity by monitoring the fidelity of DNA replication and segregation. In mammalian somatic cells, the best-characterized checkpoints are the DNA damage/replication checkpoints and the mitotic spindle checkpoint. The DNA damage/replication checkpoints result in cell cycle arrest if DNA is not fully replicated, or is damaged [ 1 ]. The mitotic spindle checkpoint results in cell cycle arrest prior to anaphase if the spindle is not properly assembled [ 2 ]. There is also evidence that defects in events relating to cell division itself can result in cell cycle arrest. Lanni and Jacks [ 3 ] and Casenghi et al.[ 4 ] found that mammalian cells that had adapted to microtubule depolymerization and exited mitosis without undergoing cytokinesis arrested in G1 of the subsequent cell cycle. Kurimura and Hirano [ 5 ] and Andreassen et al. [ 6 ] reported that inhibition of cytokinesis with the actin-depolymerizing drug cytochalasin resulted in a similar arrest in G1 of the subsequent cell cycle. These treatments resulted in cells that were tetraploid, and Andreassen et al. [ 6 ] proposed that the cell cycle arrest was triggered by ploidy, terming this effect a "tetraploidy checkpoint". Other evidence suggests that mammalian cells are not sensitive to tetraploidy. Rao and Johnson used cell fusion to study the regulation of DNA synthesis and mitosis by fusing cells at different cell cycle stages [ 7 , 8 ]. Binucleate tetraploid cells resulting from fusion between cells in different cell cycle stages were able to progress through the cell cycle. Uetake and Sluder ([ 9 ], reviewed in [ 10 ]) reported that inhibition of cytokinesis with a low dose of cytochalasin also allowed cell cycle progression. Most strikingly, there are rare cases of human infants born with fully tetraploid karyotypes [ 11 ]. Although these individuals have severe defects, their existence argues against tetraploidy as a trigger for cell cycle arrest. Here we investigate whether tetraploidy or other cellular defects in binucleate cells lead to cell cycle arrest. We show that neither tetraploidy, aberrant centrosome number, cell size, nor failure of cytokinesis lead to G1 arrest, suggesting that there is no tetraploidy checkpoint. Rather, certain standard synchronization treatments cause DNA damage that is the likely cause of G1 arrest. Results and discussion Immortalized cell lines often have altered checkpoints, therefore we used early passage primary cells to investigate the tetraploidy checkpoint. All experiments were performed with human diploid fibroblasts (HDF) from infant foreskin and used prior to passage 10. We had previously developed methods for synchronizing these cells [ 12 ], and tested them here for the presence of normal checkpoint mechanisms. First, the levels of p53 were determined by western blotting and found to be similar to other p53+/+ cell lines (not shown). Second, we tested for a functional DNA damage response. G1 phase HDF cells were released from serum starvation and irradiated with ultraviolet (UV) light. The cells were then assayed for entry into S phase by 5-bromodeoxyuridine (BrdU) incorporation. The HDF cells exhibited a normal DNA damage response; at a low dose of UV, cells were delayed by about 12 h for entry into S phase, and at a higher dose most cells did not enter S phase even 36 h after irradiation (Figure 1A ). Third, we tested for a functional spindle checkpoint. Exponentially-growing HDF cells were treated with nocodazole for 12 h to depolymerize microtubules, and assayed by light microscopy. Nocodazole treatment caused a 6-fold increase in the mitotic index, indicating that the cells had a functional spindle checkpoint. Figure 1 Cell cycle responses of human diploid fibroblast (HDF) cells. (A) Response to DNA damage. HDF cells were exposed to 0, 10 or 20 J/m 2 ultraviolet light and entry into S phase was assayed by BrdU incorporation. For each bar n ≥ 300 cells. (B) Recovery from nocodazole arrest. HDF cells were arrested in mitosis by double thymidine block followed by nocodazole (left) and released for 30 min. (center and right). DNA, blue; α-tubulin, green; γ-tubulin, red. (C) Example of binucleate cells created by cytochalasin-induced cytokinesis failure. DNA, blue. (D) Cell cycle progression of HDF cells in response to cytokinesis failure induced with 2 μM cytochalasin. Cells were assayed for BrdU incorporation at the indicated times after removal of cytochalasin. "control" cells were not treated with cytochalasin; "cytochalasin (mononucleate)" cells were treated, but completed cytokinesis, and "cytochalasin (binucleate)" cells were treated and failed to divide in cytokinesis. For each bar n ≥ 300 cells. (E) Cell cycle progression of HDF cells in response to cytokinesis failure induced with 12.5 μM blebbistatin. Cells were assayed for BrdU incorporation at the indicated times after removal of blebbistatin. "control" cells were not treated with blebbistatin; "blebbistatin (mononucleate)" cells were treated, but completed cytokinesis, and "blebbistatin (binucleate)" cells were treated and failed to divide in cytokinesis. For each bar n ≥ 300 cells. (F) Cell cycle progression in response to the presence of extra centrosomes. Image shows the product of fusion between a G1 cell and a G1 cytoplast. 24 h after fusion this cell has four centrosomes, indicating that it has undergone centrosome duplication, and has incorporated BrdU, indicating that it has entered S phase. DNA, blue; BrdU, green; pericentrin, red. Punctate blue staining is due to cell surface marker used to identify fusion products [12]. Scale bars represent 10 μm. We first tested HDF cells for the previously described G1 arrest following cytochalasin-induced failure of cytokinesis [ 6 ]. Cells were synchronized in mitosis by double thymidine arrest followed by nocodazole treatment (Figure 1B ), then released for 30 min, after which most cells had a bipolar spindle. Cells were then allowed to proceed into interphase in medium containing BrdU, +/- 2 μM cytochalasin. By 10 h after the addition of cytochalasin, both control and cytochalasin-treated cells had exited mitosis; approximately 30% of the cytochalasin-treated cells had two nuclei (binucleate) (Figure 1C ) and the remainder had a single nucleus (mononucleate), presumably having completed cytokinesis successfully. Thus there were two types of control cells in these experiments: cells that had not experienced the drug, and cells that had experienced the drug, but remained mononucleate. The cultures were washed at this point to remove drug and allowed to proceed in the cell cycle. At 6 h after the removal of cytochalasin, 50% of the untreated control cells had entered S phase, whereas only about 10% of either the mononucleate or binucleate cytochalasin-treated cells had entered S phase (Figure 1D ); these numbers changed only slightly by 12 h. However, at 24 h after cytochalasin removal, 75% of the control cells and 44% of the mononucleate cytochalasin-treated cells had entered S phase, whereas only 11% of the cytochalasin-treated binucleate cells had entered S phase. Similar results were obtained with 5 μM and 10 μM cytochalasin (not shown). Thus, binucleate HDF cells resulting from cytochalasin-induced failure of cytokinesis did arrest in G1, as previously described for other cells [ 6 ]. A potential problem with cytochalasin treatment is that depolymerization of the actin cytoskeleton is likely to have effects other than blocking cytokinesis. Indeed, we found that even at 2 μM, cytochalasin had a strong cytotoxic effect, delaying cell cycle progression significantly, with slow recovery after release (data not shown and [ 13 ]). To determine whether the effect was specific to cytochalasin, the above experiment was repeated using two other drugs that inhibit cytokinesis: blebbistatin and aurora kinase inhibitor-1 (AKI-1). Blebbistatin is an inhibitor of non-muscle myosin II, the motor protein that provides the force for furrow ingression during cytokinesis [ 14 ]. AKI-1 inhibits the aurora family of kinases, which play important roles in mitosis and cytokinesis [ 15 ]. HDF cells were synchronized in mitosis by double thymidine block followed by nocodazole treatment, then released into medium containing BrdU, +/- 12.5 μM blebbistatin. By 10 h after the addition of blebbistatin, most cells had exited mitosis; in the presence of blebbistatin approximately 30% of the cells were binucleate and the remaining cells were mononucleate, presumably completing cytokinesis successfully. Blebbistatin was removed, and cells were assayed for S phase entry over time. At 6 h after the removal of blebbistatin, 44% of the untreated control cells and 53% of the mononucleate blebbistatin-treated cells had entered S phase, whereas only 18% of the binucleate blebbistatin-treated cells had entered S phase (Figure 1E ). By 24 h the fraction of both untreated and blebbistatin-treated mononucleate cells that had entered S phase rose to about 70%, whereas the fraction of binucleate cells that had entered S phase remained at about 20% (Figure 1E ). Similar results were obtained with 25 μM and 50 μM blebbistatin, as well as with 5 μM AKI-1 (not shown). This indicates that synchronized mitotic cells that failed cytokinesis became arrested in G1 regardless of the specific inhibitor used. Cells that have failed to divide after mitosis differ from normal cells in that they have two nuclei, two centrosomes, and a decreased surface area to volume ratio. We tested each of these defects individually for an effect on G1 arrest. To test the role of centrosome number, serum-starved G0 cells were fused with enucleated G0 cytoplasts to create cells with two centrosomes, but only one diploid nucleus (Figure 1F ). The cell-cytoplast fusions were released from G0 into BrdU-containing medium and allowed to proceed through the cell cycle. The fused cells were compared to cells in the population that had experienced the fusion treatment but had not fused. At 24 h after fusion, 66+/-15% of cytoplast-cell fusions with an extra centrosome had entered S phase, and 63+/-11% of unfused control cells had entered S phase. Therefore the presence of an extra centrosome at G1 does not delay S phase entry and is not responsible for the G1 arrest in binucleate cells resulting from cytochalasin-induced failure of cytokinesis. To test the role of tetraploidy, serum-starved HDF cells were fused to create binucleate cells. Creating binucleate cells by fusion avoided disruption of the actin cytoskeleton, allowing us to examine the effect of ploidy alone. The binucleate cells resulting from fusion were both tetraploid and had two centrosomes; we showed above that centrosome number was not a factor in the G1 arrest. As above, unfused cells in the population served as an internal control. At 24 h after fusion, 72+/-2% of the unfused cells and 75+/-1% of the fused, binucleate, cells had entered S phase. Therefore, tetraploidy does not cause the observed G1 arrest resulting from cytochalasin-induced failure of cytokinesis. Cells that fail to divide at cytokinesis are larger than normal cells. Larger cells have a decreased surface area to volume ratio, which might affect the response to perturbation of the cytoskeleton. Thus, the apparent sensitivity to cytokinesis failure might derive directly from a difference in size. To test this, we created large binucleate cells by fusing serum-starved G0 cells to each other. The fusion products were released into growth medium for 3 h to allow for reattachment to the culture substrate. We then added 25 μM blebbistatin, 5 μM cytochalasin, or 5 μM AKI-1 to cells for 10 h, followed by release into BrdU-containing medium. Figure 2 shows that mononucleate and binucleate cells in the control and drug-treated populations entered S phase with similar kinetics. Note that in the cells treated with cytochalasin there was a significant delay in S phase entry, consistent with the cytotoxicity of cytochalasin that we and others have described [ 13 ]. These results demonstrate that binucleate cells are not more sensitive to cytokinesis inhibitors due to their increased size. Figure 2 Cytokinesis inhibitors do not block the G1 to S phase progression of binucleate HDF cells. Serum-starved G0 cells were fused and released into medium containing BrdU and (A) no drug, (B) 25 μM blebbistatin, (C) 5 μM AKI-1, or (D) 5 μM cytochalasin. Some cells remain unfused after the fusion treatment, and were used as mononucleate controls. Time points were taken to assay for S phase entry. If cells were sensitive to failure of cytokinesis, one might expect that the sensitivity would be expressed as a delay in exit from mitosis, a time when a cytokinetic defect could be corrected. This would be similar to the known DNA damage, DNA replication and spindle assembly checkpoints [ 16 ]. We tested in two ways whether mammalian cells delay the exit from mitosis in response to cytokinesis failure. First, HDF cells were imaged by time-lapse microscopy as they progressed through mitosis in the presence or absence of blebbistatin. Cells were synchronized in mitosis by nocodazole treatment, then released for 30 min, when 25 μM blebbistatin was added. Control cells (n = 5) exhibited cytokinetic constrictions beginning about 60 min after release from nocodazole. These cells flattened and began to spread, signaling the end of mitosis, about 85 min after release (Figure 3A ). Blebbistatin-treated cells (n = 9) did not undergo observable cytokinesis, as expected, but did flatten and spread about 110 min after release from nocodazole (Figure 3A ). We also tested for a delay in mitotic exit by staining with MPM-2, and antibody specific for mitotic phosphoepitopes [ 17 ]. Cells that were MPM-2 positive and had condensed DNA were considered to be in mitosis (Figure 3B ). After release from nocodazole arrest, the fraction of mitotic cells declined in both the control and blebbistatin-treated populations with only a slight delay apparent in the blebbistatin-treated cells (Figure 3C ). Both assays showed that blebbistatin treatment resulted in only a brief delay in the exit from mitosis, suggesting that failure of cytokinesis does not trigger a checkpoint-like arrest. Figure 3 Cytokinesis failure does not significantly delay the exit frommitosis. (A) Images from time-lapse series of HDF cells at the indicated times after release from nocodazole-mediated mitotic arrest. "Control" cells were not treated with blebbistatin; "Blebbistatin" cells were treated with blebbistatin beginning at 30 min after release from nocodazole. (B) MPM-2 immunofluorescence as a marker for mitotic exit. Fluorescence image of a mitotic cell with condensed DNA and intense MPM-2 staining (top) and a cytokinetic cell with decondensed DNA and diminished MPM-2 staining (bottom). DNA, blue; MPM-2, green. Scale bar represents 10 μm. (C) Mitotic index of control and blebbistatin-treated cells after mitotic release. "Control" cells were not treated with blebbistatin; "Blebbistatin" cells were treated with blebbistatin beginning at 30 min after release from nocodazole. Mitotic index was determined by MPM-2 staining and DNA morphology, as in (B). For each point n = 100 cells. Since we had ruled out most of the cellular defects associated with division failure as being the cause of the G1 arrest, we attempted to further characterize the arrest. Andreassen et al. [ 6 ] reported that p53 is important in the G1 arrest caused by cytokinesis failure. To test the role of the p53 pathway, we repeated the blebbistatin experiment above with wt, p53 -/-, and p21 -/- mouse embryonic fibroblast (MEF) cells. Wt MEF cells behaved similarly to the HDF cells, arresting in G1 in response to cytokinesis failure (Figure 4A ). However, in p53 -/- and p21 -/- MEF cells both binucleate and mononucleate cells entered S phase with the same kinetics (Figure 4A ). Thus, the p53-p21 pathway is required for the G1 arrest of binucleate cells. Figure 4 Cells arrested in G1 by cytokinesis failure enter p53- and p21-dependent premature senescence. (A) p53 pathway dependence of the G1 arrest following cytokinesis failure. Wt, p53-/- and p21-/- mouse embryo fibroblasts were synchronized and treated with blebbistatin and assayed as in Figure 1E. For each bar n ≥ 200 cells. (B) G1 arrested binucleate cells entered premature senescence as assayed by senescence-associated β-galactosidase activity (SA-β-gal). The dark stain in the binucleate cell shown is the reaction product of Xgal cleavage. Scale bar represents 10 μm. (C) Time course of appearance of senescent cells. Mononucleate and binucleate cells were assayed for SA-β-gal at the indicated days after blebbistatin removal. For each point n = 60–100 cells. p53-p21-dependent G1 arrest often results in either apoptosis or senescence [ 18 , 19 ]. To determine the fate of the G1-arrested products of a failed cytokinesis, binucleate HDF cells were prepared using blebbistatin as described above. Blebbistatin was then removed and the cells were assayed by microscopy. The binucleate cells persisted in the population over the course of more than two weeks, consistent with these cells being permanently arrested in the cell cycle. The binucleate cells did not undergo apoptosis, as assayed by morphology and staining with annexin V, an early marker of apoptosis (not shown). However, the binucleate cells did develop several hallmarks of senescence, including becoming flattened and enlarged, and accumulating senescence-associated β-galactosidase activity (Figure 4B ). As the criteria for defining cellular senescence are not firmly established [ 20 ], we will refer to this phenotype as "senescent-like". At 4 days after blebbistatin removal approximately 35% of binucleate cells and 10% of mononucleate cells were senescent-like; by 12 days virtually all of the binucleate cells, but only 10% of mononucleate cells, were senescent-like (Figure 4C ). We have shown that failure of division of synchronized cells results in a p53-dependent arrest, but that the arrest is not due to ploidy, centrosome number, or cell size, and that the arrest is not preceded by a delay in mitotic exit, suggesting that it is not a classical checkpoint. The characteristics of the arrest are similar to those of the G1 arrest caused by the DNA damage checkpoint in HDF cells, which respond to irreparable DNA damage by entering senescence, instead of apoptosis [ 21 ]. These similarities led us to test whether the binucleate G1 arrest might actually be due to DNA damage suffered during the treatment. Cells were synchronized in mitosis by the double thymidine – nocodazole regimen described above and treated with 25 μM blebbistatin. The binucleate cells were released from blebbistatin for 1 h, 3 days, and 8 days respectively, then stained for γ-H2AX, a marker of DNA damage [ 22 ]. As a positive control for DNA damage, asynchronous cells were treated with 1 mM hydrogen peroxide for 30 min, allowed to recover in medium for 1 h, and stained for γ-H2AX. In the untreated control cells, only 3.4% of cells contained γ-H2AX foci, whereas in the hydrogen peroxide treated cells, 33% of cells contained γ-H2AX foci (Figure 5A ). Remarkably, at 1 h after release of synchronized cells from blebbistatin, 52% of the binucleate cells contained γ-H2AX foci in one or both nuclei (Figure 5A ), suggesting the presence of DNA damage. However, we found that 32% of the mononucleate cells that successfully completed cytokinesis after blebbistatin treatment also contained γ-H2AX foci. This suggested that the observed DNA damage might not be the result of division failure per se, and therefore might have occurred prior to the addition of blebbistatin, possibly during cell synchronization. At 3 days after release from blebbistatin, 30% of the binucleate cells and 11% of the mononucleate cells contained visible γ-H2AX foci. At 8 days after release from blebbistatin, 29% of the binucleate cells and only 6% of the mononucleate cells contained γ-H2AX foci. Most of the binucleate cells also displayed senescent-like phenotypes at 8 days after the removal of blebbistatin (Figure 5B ). Figure 5 Binucleate cells contain nuclear γ-H2AX foci. (A) Binucleate cells were prepared by synchronization and treatment with 25 μM blebbistatin as described above, then stained for the DNA damage marker γ-H2AX. Untreated control cells (top) did not contain any visible γ-H2AX foci, whereas binucleate cells, after released from blebbistatin for 1 h, (bottom) contained γ-H2AX foci that were similar to those of cells treated with H 2 O 2 (middle). Scale bar represents 2.5 μm. (B) Culture was continued for 8 days after release from blebbistatin. Most mononucleate cells lacked γ-H2AX foci (top), whereas approximately 30% of binucleate cells still contained nuclear γ-H2AX foci (bottom). The binucleate cells were also flattened and enlarged, consistent with a senescent-like arrest. Scale bar represents 10 μm. The binucleate cells that persisted in culture were arrested in G1, as they did not incorporate BrdU after the previous round of mitosis, and they did not proceed to the next round of mitosis, as evidenced by the preservation of the binucleate phenotype. The presence and persistence of nuclear γ-H2AX foci in the G1-arrested binucleate cells suggested that DNA damage might be the cause of the arrest. However, not all the arrested binuclear cells contained visible γ-H2AX foci, indicating that γ-H2AX-associated DNA damage might not be the only cause of the arrest. The percentage of binucleate cells with nuclear γ-H2AX decreased from 52% to 29% over 8 days of culturing, possibly indicating that some cells were able to correct the DNA damage after being arrested in G1 for several days. In contrast, the percentage of mononucleate cells that displayed γ-H2AX foci decreased dramatically over 8 days of culturing, however this was likely due to proliferation of normal mononucleate cells in the culture rather than a difference in response of the mononucleate and binucleate cells to the treatment. The presence of γ-H2AX foci in the mononucleate cells that successfully completed cytokinesis after blebbistatin treatment suggested that the DNA damage might have been the result of the synchronization treatments, prior to the addition of blebbistatin. Therefore, we tested whether any of the cell synchronization treatments alone had an effect on cell cycle progression. Asynchronous cells were treated with double thymidine block, nocodazole, or blebbistatin individually, following the same protocols used above in multiple treatments. Cells were released from drug and S phase entry was assayed at time points. Figure 6A shows that none of the drug treatments resulted in a substantial failure of cell cycle progression after release. Most importantly, we found that most of the binucleate cells that resulted from cytokinesis failure with blebbistatin treatment alone were able to enter S phase normally after release. This result indicates that there is no cytokinesis checkpoint, in accord with the results of Uetake and Sluder [ 9 ]. Figure 6 The combination of double thymidine block and nocodazole treatment causes DNA damage in HDF cells (A) Asynchronous HDF cells were treated with either double thymidine, nocodazole, or blebbistatin individually, then released into BrdU-containing growth media and assayed for S phase entry. For each bar n ≥ 200 cells. (B) Cells were subjected to the treatments in the order double thymidine, nocodazole, blebbistatin. Samples of cells were taken after release from each drug, and S phase entry was assayed. For each bar n ≥ 200 cells. (C) Cells were treated with double thymidine block followed by nocodazole, then stained for γ-H2AX. Scale bar represents 2.5 μm. Since none of the single drug treatments resulted in a cell cycle arrest, we reasoned that some combination of the treatments must be responsible. To determine which combination of treatments caused a G1 arrest, cells were subjected to the treatments in the order double thymidine, nocodazole, blebbistatin. Samples of cells were taken after release from each drug, and S phase entry was assayed (Figure 6B ). Although neither double thymidine nor nocodazole arrest and release alone resulted in a G1 arrest, the combination of them did; only about 45% of such cells progressed into S phase. The addition of blebbistatin to the treatment did not cause a further reduction in the fraction of cells entering S phase; about 64% of the mononucleate cells progressed into S phase. However, only about 22% of the binucleate cells that failed cytokinesis after blebbistatin treatment progressed into S phase. This indicated that the G1 arrest in our experiments is due to the double thymidine block, followed by nocodazole treatment, and that binucleate cells are more susceptible to this effect. To determine whether the thymidine – nocodazole combination caused the DNA damage we observed, cells were treated with both drugs as above and stained for γ-H2AX. At 3 h after release from the drug treatments, about 33% of cells had γ-H2AX foci (Figure 6C ). Thus the thymidine – nocodazole synchronization treatment caused the DNA damage that resulted in cells becoming arrested in G1. Given these results we suggest a simple model for the increased susceptibility of binucleate cells: those cells that failed cytokinesis are more likely to become arrested in G1 because they contain two nuclei, and thus have twice the chance of inheriting DNA damage compared with cells that successfully divided. Conclusions We have shown that tetraploidy, aberrant centrosome number, increased cell size, and failure of cytokinesis do not lead to G1 arrest in primary human diploid fibroblasts. Rather, we found that the observed G1 arrest in cells that have failed to divide is likely due to cellular damage caused by standard synchronization treatments. We note that all published observations of a G1 arrest in response to division failure involved extensive manipulation of mammalian cells in culture. It seems likely that these manipulations resulted in DNA damage, or in other damage, that resulted in a G1 arrest, but was not directly associated with division failure. For example, Uetake and Sluder [ 9 ] found that supplementing the culture substrate with fibronectin allowed binucleate cells formed by cytochalasin treatment to progress through the cell cycle, suggesting that cell adhesion was defective in the drug-treated cells. Given that binucleate cells clearly can cycle when formed with minimal manipulation, it is likely that all previous reports of a tetraploidy checkpoint can be explained by side effects of the drug treatments used to observe them. Methods Cell methods Human diploid fibroblasts (HDFs) were from infant foreskin. Wt, p53 -/- and p21 -/- mouse embryo fibroblasts (MEFs) were the kind gift of Laura Attardi (Stanford, CA). HDFs and MEFs were cultured in DMEM (Gibco) with 10% fetal bovine serum. HDFs were used prior to passage 10 and MEFs were used prior to passage 5. HDFs were synchronized in G0 by serum starvation [ 12 ] and S phase by double thymidine block [ 23 ], as described. In the cell fusion experiments, serum-starved G0 cells, or cytoplasts derived from those cells by centrifugation, were fused with serum-starved G0 cells, as described [ 12 ]. Immunocytochemistry was as described [ 12 ]. Live cell imaging was with a Nikon Diaphot microscope equipped with an environmental chamber allowing incubation at 37°. Images were collected with a CCD camera (Photometrics) and processed with Openlab (Improvision) and Photoshop (Adobe) software. Senescence-associated β-galactosidase activity (SA-β-gal) was assayed as described [ 24 ]. Assay for S phase entry by BrdU incorporation Cells were incubated with 20 μM BrdU (Sigma) for indicated times and fixed in -20°C methanol for at least 10 min. Fixed cells were treated with DNase I (Boehringer Mannheim) and exonuclease III (New England Biolabs) to expose the BrdU epitope prior to incubation with anti-BrdU antibodies, as described [ 12 ]. Nuclei were visualized by staining 4',6-diamidino-2-phenylindole (DAPI). Cells were observed with a Zeiss Axioskop microscope with a Zeiss Plan-Neofluar 100/1.3 objective, and images were collected with a cooled-CCD camera (Hamamatsu) controlled by Openlab software. Assay for DNA damage by γ-H2AX staining Cells were fixed with 2% paraformaldehyde at room temperature for 10 min, washed 3× with PBS, then permeabilized with-20°C methanol for 5 min and stained with 27 ng/ml γ-H2AX antibody (Trevigen, MD). Only cells with multiple, clearly labeled foci were counted as being γ-H2AX positive. Drug-induced cytokinesis failure Cells synchronized in S phase by double thymidine block were released from the block for 6 h to allow completion of S phase. Nocodazole (100 ng/ml) (US Biological) was then added for 6 h to arrest the cells in mitosis. Cells were released from the mitotic arrest for 30 min, during which time most cells formed a bipolar mitotic spindle (Fig 1B ). At 30 min after release from mitotic arrest, 20 μM BrdU was added to the medium, together with the indicated concentration of cytochalasin B (Sigma) or (s)-(-)-blebbistatin (Toronto Research Chemicals). Cells were incubated in this medium for 10 h to inhibit cytokinesis, then changed to growth medium containing 20 μM BrdU but no cytokinesis inhibitor, and assayed for S phase entry at the indicated times. Authors' contributions CW participated in the design of the study and carried out the experiments. TS conceived of the study and participated in its design.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC554097.xml

549561

Effects of the search technique on the measurement of the change in quality of randomized controlled trials over time in the field of brain injury

Background To determine if the search technique that is used to sample randomized controlled trial (RCT) manuscripts from a field of medical science can influence the measurement of the change in quality over time in that field. Methods RCT manuscripts in the field of brain injury were identified using two readily-available search techniques: (1) a PubMed MEDLINE search, and (2) the Cochrane Injuries Group (CIG) trials registry. Seven criteria of quality were assessed in each manuscript and related to the year-of-publication of the RCT manuscripts by regression analysis. Results No change in the frequency of reporting of any individual quality criterion was found in the sample of RCT manuscripts identified by the PubMed MEDLINE search. In the RCT manuscripts of the CIG trials registry, three of the seven criteria showed significant or near-significant increases over time. Conclusions We demonstrated that measuring the change in quality over time of a sample of RCT manuscripts from the field of brain injury can be greatly affected by the search technique. This poorly recognized factor may make measurements of the change in RCT quality over time within a given field of medical science unreliable.

Background Considerable effort has been directed toward improving randomized controlled trial (RCT) design, execution, and reporting [ 1 - 6 , 14 ]. Such efforts to define standards of quality for RCTs beg the question: are RCTs improving in quality over time? Many reviews have attempted to answer this question. In general, these reviews measure the presence or absence of several criteria chosen to define quality in a sample of RCT manuscripts that was selected from a parent population of RCT manuscripts. The parent population of RCT manuscripts may be either a field of medical science or a defined part of the medical literature (e.g., RCT manuscripts from a chosen journal). Then, by examining a score of quality as a function of the year-of-publication of the sampled RCT manuscripts, conclusions are drawn as to whether or not quality is changing over time in the parent population of RCTs. If such reviews are to be useful, then, the sample of RCT manuscripts that was chosen for analysis must represent the parent population of RCT manuscripts. As much as the RCT manuscripts published in a single journal or group of journals would provide a well-defined parent population, the RCT manuscripts from a given field of medical science would be difficult to completely identify. Ultimately no search strategy can claim to identify all manuscripts on a given topic that have been published in every book and journal worldwide. Thus, two search techniques might provide considerably different samples of RCT manuscripts from the same field of medical science depending upon how much and / or what parts of the parent population of RCT manuscripts they can access. The current communication empirically demonstrates this point as a potential pitfall in measuring the change in quality over time of RCT manuscripts sampled from a representative field of medical science. Methods Criteria of quality We chose internal validity as a measure of quality according to the definition given by Gehlbach [ 7 ], namely that a RCT is internally valid when "within the confines of the study, results appear to be accurate and interpretation of the investigators is supported". We selected criteria of internal validity according to the recommendations of Moher et al. [ 8 ]. The relevant points are addressed below. I. Definition of the quality construct We intended to measure the presence or absence of various criteria of RCT quality as described in the published manuscript. No attempt was made to contact the authors of a manuscript either to clarify the information provided in the manuscript or to gain additional information about a RCT. We acknowledge that relying on the published manuscript in order to assess the quality of a RCT may be biased (1) against well-designed RCTs that were reported in poorly written manuscripts and (2) in favor of poorly-designed RCTs that were reported in well-written manuscripts [ 9 ]. Thus, our scoring process ultimately measured the quality of the report of the RCT manuscript, rather than the true methodological quality of the trial as it was conducted. However, attempting to obtain an understanding of the true methodological quality of a RCT in a retrospective manner by contacting the authors of the manuscripts would undoubtedly collect more information on recent RCTs because their authors will be more accessible (i.e., less likely to have relocated, retired, or died). Attempting to contact the authors of manuscripts is rarely successful [ 10 ] and, when it is successful, accurate information about the design and conduct of the RCT is not always forthcoming [ 11 , 12 ]. II. Definition of the scope of internal validity and identification of quality criteria Although random allocation and the use of a concurrent control group are the sine qua non of the RCT, additional criteria have been so frequently included in their design and execution that they are now commonly considered as part of quality RCTs. Several sources (themselves located by PubMed MEDLINE and bibliography searches) were used to identify such criteria [ 2 , 9 , 13 - 18 ]. After forming a composite list of internal validity criteria from these sources, we searched the literature (again by means of PubMed MEDLINE and bibliographies) for instances where the presence or absence of each criterion in a RCT affected the results obtained from the RCT. Thus, we identified criteria that were supported by empirical evidence as measures of RCT quality. We identified six criteria that had predominantly supporting evidence in their favor. Subsequently, allocation concealment was included as a separate quality criteria. The quality criteria, with brief descriptions, are listed in Table 1 . Table 1 The quality scale This table lists the criteria of quality that were used to score the RCT manuscripts. Abbreviated definitions for the presence (1 point) or absence (0 points) of each criterion are provided. 1) assessment of the distribution of patient characteristics and prognostic factors between groups present distribution of patient characteristics and prognostic factors assessed without asymmetry between groups absent not mentioned; distribution of patient characteristics and prognostic factors assessed with asymmetry noted between groups 2) prevention of the movement of patients between groups after allocation, and the use of intention-to-treat analysis present use of intention-to-treat analysis; no movement of patients between groups confirmed absent not mentioned; patients known to change groups before analysis 3) the blinding of the patients to the treatment they received present statements of double-blind present; use of a placebo; statements of the treatments being indistinguishable present; patients not aware of study due to clinical condition absent not mentioned; lack of placebo use in control group; readily-distinguishable treatments; blinding breakdown confirmed 4) the blinding of the health care providers to the treatments received by the patients present third-party dispensation of treatments; statements of health care provider blinding present; health care provider identical to outcome observer, and outcome observer is blinded absent not mentioned; health care team aware of patient allocation; lack of placebo in control condition; readily-distinguishable treatments; blinding breakdown confirmed 5) the blinding of the outcome observer to the treatment received by the patient present statements of double blind present; objective outcome; use of standardized tests or questionnaires that do not require an outcome observer; subjective principle outcome but outcome observer blinded to treatment; blinded health care providers performing outcome assessment absent not mentioned; subjective outcome without blinding of the outcome observer; blinding breakdown confirmed 6) completeness of follow-up present no patients lost to follow-up; acute experimental design does not permit loss of patients; analysis of lost patients provided according to randomization groups, with reason for loss absent not mentioned; no analysis of lost patients provided; effect of patient loss to follow-up confirmed 7) allocation concealment present use of consecutive opaque envelopes or pre-ordered treatments; third party assignment of allocation absent not mentioned; repeatable pattern of allocation; use of obvious identifiers for allocation (e.g., birth date, record number); assignment of treatment by treating physician We limited our quality scale to measure criteria that have been demonstrated empirically to be associated with the quality of RCTs. This necessarily excluded many items associated with RCT design and execution that are widely thought to affect quality or that are included in commonly-used quality scales, but it provided us with a defensible "bare minimum" definition of quality. It should be noted that we did not intend our list of criteria to be encompassing of all aspects of quality; our criteria were intended to serve only as a tool for the comparative analysis of the two sets of RCT manuscripts for the purpose of this study. III. Scoring System Each of the seven criteria was scored as being present (1 point) or as absent (0 points) in the RCT manuscript. Definitions of each criterion are shown in Table 1 . If a RCT manuscript did not mention the presence of a criterion, it was considered absent. Conversely, all written statements in the manuscripts were assumed to be accurate both factually and semantically. IV. Criteria Scoring Verification The intra-rater reliability for the scoring of the quality criteria was determined by comparing the individual criteria scores given to n = 16 RCT manuscripts by one of the authors of this communication (MKB) on two occasions separated by 3 weeks. The correlation coefficient (Kappa) measured in this manner was 0.94. Inter-rater reliability was determined by comparing the quality criteria scores given to n = 10 RCT manuscripts by two different examiners. One copy of each manuscript was scored by one of the authors of this communication (MKB) while the other copy was scored by an independent examiner (Dr. Babak Jahromi, Department of Neurosurgery, the University of Toronto) who was provided with a thorough description of the criteria. The correlation coefficient (Kappa) for inter-rater reliability was determined to be 0.74. Manuscript selection and the screening process We chose to evaluate the field of brain injury because two search techniques for sampling the population of these RCT manuscripts were readily available. The first search technique was our own PubMed MEDLINE search. The second search technique was performed by the Cochrane Collaboration Injuries (CIG) Group, and forms the CIG trials registry. Copies of the RCT manuscripts identified by these two search techniques were retrieved through the library holdings and interlibrary loan services of five universities. Next, the manuscripts were read by one of us (CY) to screen-out inappropriately identified manuscripts. Table 2 provides a detailed list of these exclusions. Inherent in the phrase 'randomized controlled trial' is (1) the random allocation of patients into multiple groups for prospective analysis, and (2) the concurrent comparison of at least one group that receives the experimental treatment against another group that does not; manuscripts that did not include random allocation and a concurrent control group were excluded. Furthermore, in order for a manuscript to be considered pertinent to the study of brain injury one of the following conditions had to be met: (1) brain injury had to directly define the patient population; (2) brain injury had to be the cause of a second condition (e.g., seizures) that defined the patient population; or (3) brain injury had to be the outcome measure for the patient population. If none of the above conditions were met the manuscript was discarded from further examination. Duplicate publications, protocol descriptions, abstracts, letters-to-the-editor, and incomplete or preliminary reports were also removed during the screening process. Table 2 Exclusion of manuscripts from the PubMed MEDLINE and CIG Trials Registry groups of manuscripts Manuscripts inappropriately identified by the PubMed MEDLINE search and the CIG trials registry were removed from review during a screening process performed by one of the authors of the current communication (CY). reason for exclusion PubMed MEDLINE search CIG Trials Registry group INITIALLY IDENTIFIED 139 312 libraries unable to locate 0 15 unrelated to brain injury 2 3 duplicate publications 2 8 inaccurately claimed to be a controlled trial 22 47 inaccurately claimed to use randomization 11 30 abstracts / letters-to-the-editor 1 31 protocol descriptions 3 0 incomplete / preliminary reports 0 4 non-human subjects 0 1 TOTAL NUMBER DISCARDED 41 139 REMAINING 98 (70% of initially identified) 173 (55% of initially identified) The design and yield of the two search techniques was as follows: 1) the PubMed MEDLINE search: The first search technique we used to identify RCT manuscripts pertaining to brain injury involved the PubMed search engine of the MEDLINE database. It was designed to represent a typical literature search performed by a North American researcher who is fluent only in English. The search term "brain injuries" (C10.228.140.199) was used with the limitations of (1) randomized controlled trial, (2) human subjects, and (3) publication in English. The PubMed MEDLINE search included manuscripts indexed from January, 1966, up to February, 2001 (the time at which the search was performed). The PubMed MEDLINE search identified n = 139 manuscripts. During the screening process, n = 41          manuscripts from the original 139 (30%) were discarded leaving n = 98 manuscripts (see Table 2 for          a detailed list of the exclusions). 2.) the CIG trials registry: The Injuries Group of the Cochrane Collaboration was kind enough to share their list of RCT manuscripts with us for the purpose of conducting this study. The list of manuscripts they provided was compiled by means of the following three steps: step 1) The CIG trials master list was searched using the keywords "head" or "brain" in conjunction                   with "injur*" or "trauma*". The CIG trials master list is a local database maintained at the London                   School of Hygiene and Tropical Medicine that uses a detailed search strategy to identify RCTs                   from multiple computerized databases (a copy of this search strategy is available from Ms. Fiona                   Renton of the London School of Hygiene and Tropical Medicine Fiona.Renton@lshtm.ac.uk ) as                   well as various hand searches of journals performed during the writing of systemic reviews; it is                   updated quarterly. step 2) MEDLINE, EMBASE, and CENTRAL databases were searched using the exploded                   keyword "head injuries:ME" or "head injuries:TI". EMBASE includes references from 1974                   onward and, while it uses its own database, it is based on an indexing hierarchy which incorporates                   that used by MEDLINE. Here, MEDLINE was searched with the SilverPlatter search engine, not                   with the PubMed Search engine. Manuscripts of the MEDLINE database indexed as early as 1966                   were accessible to the SilverPlatter search engine. The CENTRAL database is a general list of                   clinical trials that is maintained by the collaborative efforts of multiple Cochrane specialty groups. step 3) Manuscripts identified by hand searches of relevant journals and from references provided                   by direct contact with experts in the field of brain injury were also included. The original CIG trials registry was completed in 1998 and was last fully updated in May, 2001; it is that          version which was used in our study. The CIG trials registry included n = 312 manuscripts. During the screening process, n = 139 manuscripts          from the original 312 (45%) were discarded leaving n = 173 RCT manuscripts (see Table 2 for a          detailed list of the exclusions). 3.) overlap between the PubMed MEDLINE search and the CIG trials registry: Of the total unscreened samples of manuscripts identified through each search technique, n = 80 manuscripts were present in both samples; this corresponded to 58% of the sample of manuscripts identified by PubMed MEDLINE search and 26% of the sample of manuscripts from the CIG trials registry. After the removal of inappropriate manuscripts during the screening process, and scoring process only n = 56 manuscripts were identified by both the PubMed MEDLINE search and the CIG trials registry. This corresponded to 57% and 32% of the PubMed MEDLINE search and the CIG trials registry samples, respectively. The scoring process Each of the RCT manuscripts was read by both authors of the current communication (CY and MKB) who, for clarity's sake, will be referred to as "examiners". One examiner ("non-judging examiner": CY) performed the screening process described previously, then recorded the year-of-publication of each manuscript that survived the screening process in a computerized spreadsheet (Microsoft Excel) and marked them with identification numbers. Then, the non-judging examiner hid the names of the authors of the manuscript, the authors' degrees and departmental affiliations, the journal in which the RCT manuscript was published, and the year-of-publication of the manuscript with black marker. This information was covered wherever it was found in the manuscript so that when the manuscript was scored by the second examiner ("judging examiner": MKB) there would be no potential for bias [ 8 , 19 ]. The data collected by the judging examiner was entered into a computerized spreadsheet that was different from the one linking the year-of-publication of the manuscripts with their identification numbers. The two spreadsheets were combined only when all the manuscripts had been read. As mentioned above, allocation concealment was included in the list of quality criteria after the first evaluation of the manuscripts. Accordingly, the judging examiner re-read all the manuscripts specifically to determine the inclusion of allocation concealment. The manuscripts were still blinded as described above, and the data was entered into a third spreadsheet that was subsequently analyzed independently of the preexisting data. Manuscripts in French and Spanish were read without written translation by the judging examiner, whereas written translations were provided to the judging examiner for manuscripts in Japanese (by CY), German and Italian (by Mrs. Margaret K. Borsody), and Chinese (by Language Line, Inc., document translation service). Statistical analysis After completion of the scoring process, statistical analyses were conducted by the judging examiner. The data was considered interval in nature and thus data analysis for discrete variables was used [ 20 ]. Furthermore, since this study was constructed as a longitudinal analysis of the change in quality scores over time, it was necessary to use some form of regression analysis to examine the data. Considering these requirements, binary logistic regression analyses were performed for each individual quality criteria. All statistical analyses were done by SPSS (version 11.5, SPSS Inc.). Scores for the individual quality criteria were examined as dependent variables against the independent variable of year-of-publication. Significance is defined as a P < 0.05. Since the samples of manuscripts from the PubMed MEDLINE search and the CIG trials registry are known to be derived from the same parent population of RCTs (i.e., RCTs in the field of brain injury), it is inappropriate to directly compare them against each other with statistical tests. Rather, it was our goal to analyze the two samples of RCT manuscripts separately, and to make likely conclusions about the parent population from each sample of manuscripts as if there was no other sample of manuscripts available for comparison. Then, knowing that the two samples of RCT manuscripts represent the same parent population, it was our intention to compare the conclusions derived from the separate analyses to determine the impact of the search technique thereupon. Results Regression analysis of the individual quality criteria against the year-of-publication of the RCT manuscripts was performed to determine if the frequency of reporting of each quality criteria changed over time. For the sample of RCT manuscripts identified by the PubMed MEDLINE search, no significant relationship was found for any individual quality criterion (listed in Table 3 with the results from the statistical analysis). The RCT manuscripts identified by the CIG trials registry were also examined in this manner. Analyzing each quality criterion individually as a function of the year-of-publication of the manuscripts in that sample showed that two criteria ("prevention of the movement of patients between groups after allocation, and the use of intention-to-treat analysis"; "the assessment of the distribution of patient characteristics and prognostic factors between groups") and nearly another ("completeness of follow-up") were reported in the manuscripts with increasing frequency over time (Table 4 ). Table 3 Regression analysis of individual quality criteria versus year-of-publication of the manuscripts identified by the PubMed MEDLINE search This table lists the results of the regression analyses comparing year-of-publication against the individual quality criteria. quality criterion regression result (W = Wald stat) Cox & Snell R 2 regression line the assessment of the distribution of patient characteristics and prognostic factors between groups W = 0.96, P = 0.33 0.01 y = -0.075x + 9.45 prevention of the movement of patients between groups after allocation, and the use of intention-to-treat analysis W = 1.63, P = 0.20 0.02 y = 0.088x - 10.30 the blinding of the patients to the treatment they received W = 0.30, P = 0.58 0.003 y = 0.022x - 0.80 the blinding of the health care providers to the treatments received by the patients W = 2.00, P = 0.16 0.02 y = 0.055x - 5.71 the blinding of the outcome observer to the treatment received by the patient W = 0.01, P = 0.93 0.000 y = 0.003x + 0.57 adequacy of follow-up W = 0.03, P = 0.86 0.000 y = -0.006x + 1.18 allocation concealment W = 1.06, P = 0.30 0.011 y = 0.39x - 78.4 Table 4 Regression analysis of individual quality criteria versus year-of-publication of the manuscripts identified by the CIG Trials Registry This table lists the results of the regression analyses comparing year-of-publication against the individual quality criteria. quality criterion regression result (W = Wald stat) Cox & Snell R 2 regression line the assessment of the distribution of patient characteristics and prognostic factors between groups W = 5.53, P = 0.02 0.03 y = 0.072x - 4.71 prevention of the movement of patients between groups after allocation, and the use of intention-to-treat analysis W = 4.74, P = 0.03 0.03 y = 0.054x - 13.25 the blinding of the patients to the treatment they received W = 0.04, P = 0.84 0.000 y = 0.006x + 0.77 the blinding of the health care providers to the treatments received by the patients W = 0.16, P = 0.69 0.001 y = -0.010x + 0.32 the blinding of the outcome observer to the treatment received by the patient W = 0.27, P = 0.60 0.002 y = 0.012x - 0.57 adequacy of follow-up W = 3.25, P = 0.07 0.02 y = 0.043x - 3.25 allocation concealment W = 0.23, P = 0.88 0.000 Y = 0.004x - 7.64 Discussion Many of reviews have attempted to measure the change in RCT quality over time in a field of medical science. It occurred to us that such an analysis could be influenced by the search technique that was used to identify the RCT manuscripts. Based on this concern we hypothesized that two samples of RCT manuscripts taken from the same field of medical science by different search techniques could provide different measures of the change in quality over time. We empirically tested this hypothesis, and by doing so demonstrated that the conclusions made about the change in quality of RCT manuscripts from a representative field of medical science could be significantly influenced by the search technique that was used to sample the field. This demonstration may then bring into question the validity of previous reviews that have claimed to define the change in quality of RCTs over time in various fields of medical science. In our study, the samples of RCT manuscripts provided by the PubMed MEDLINE search and the CIG trials registry had less overlap than we would have expected considering that both search techniques involved the MEDLINE database. In particular, the CIG trials registry identified only about 60% of the RCT manuscripts found by the PubMed MEDLINE search despite involving its own search of the MEDLINE database. This observation may ultimately relate to the use of different search terms to identify manuscripts from the MEDLINE database, and to the use of different search engines of the MEDLINE database (i.e., PubMed, versus SilverPlatter in the CIG trials registry) that themselves can affect the identification of manuscripts from the common database. Whatever may be the cause for the discrepancy between our two samples, it may undermine any claim that a search technique necessarily produces a more representative sample from a field of medical science simply because it identifies a greater number of RCT manuscripts. The two search techniques otherwise differ in several ways. For example, the PubMed MEDLINE search was designed so as to exclude any manuscripts published in a non-English language. This would approximate the typical literature search performed by many researchers in North America, and accordingly all the manuscripts identified by the PubMed MEDLINE search were readily available in local university libraries. Conversely, the CIG trials registry tended to include more references from the non-English language literature (n = 27 manuscripts after the screening process). This inclusiveness of the CIG trials registry seemed to account for the 15 irretrievable manuscripts listed by the CIG trials registry. It is reasonable to state that the non-English language literature is part of medical science and that it should not be discounted solely because of its country-of-origin or the language in which it was written. As another difference, the CIG trials registry involved hand-searches of journals and lists of references provided by authorities in the field of brain injury, which are not features of the PubMed MEDLINE search and which may predispose the CIG trials registry search technique toward recovering more recently-published manuscripts. Recently published manuscripts may be of higher quality, thereby biasing the longitudinal measurement of quality in the RCT manuscript sample provided by the CIG trials registry. Alternatively, such extra efforts would be considered by most to improve on the yield of a search technique by including journals and books that are not indexed by computerized databases. Arguments can be made that either of the search techniques provided a more representative sample of RCT manuscripts from the field of brain injury, but which search technique is superior – if either can be said to be so – is not a concern of the current study. It was solely our intention to compare the findings provided by two commonly-used search techniques to demonstrate that the search technique can in fact influence the measurement of the change in RCT quality over time. We acknowledge a priori that neither of the search techniques we used necessarily sampled RCTs from the field of brain injury in a representative manner. Furthermore, we do not claim to have accurately measured how the quality of RCT manuscripts is changing over time in the field of brain injury with either one of them. This is because we are not confident that either search technique provided a representative sampling of the field of brain injury (i.e., that either search technique had access to all the relevant manuscripts). With regards to the field of brain injury, it is clear from our observations that neither the PubMed MEDLINE search nor the CIG Trials Registry can claim to be complete, as each search technique failed to identify a large number of RCT manuscripts that were found by the other search technique. In other fields of medical science there may be specialized databases or registries that claim to identify all relevant manuscripts (e.g., the Renal Registry in the field of nephrology [F.P. Schena, personal communication]). The authors of the current communication cannot understand how such a claim can be made or proven, since demonstrating the completeness of a search strategy would require proving that there are no relevant manuscripts that it does not identify. Ultimately, proving that something does not exist is scientifically impossible. Alternatively, it might be claimed that a search strategy identifies the majority of relevant manuscripts. This, of course, depends upon (1) the definition of a majority and (2) the assumption that the finding of even a few relevant manuscripts not identified by the search strategy means there are no other such manuscripts outside of the reach of that search strategy. Again, such a claim would depend upon the assumption that the inability to find further relevant manuscripts indicates that no further relevant manuscripts exist; as described above, this is a scientific impossibility. Rather than claiming perfection or near-perfection, it would seem to us to be more appropriate and accurate to claim that a given search strategy has exhausted all options for identifying relevant manuscripts. What, then, should be done to avoid a biasing influence related to the search technique during reviews of RCT quality over time? The simplest means of avoiding a such an influence would have apparently been to use multiple search techniques in order to better sample the parent population of RCT manuscripts in a field of medical science. In general, including multiple techniques into a single 'comprehensive' search would be preferable to a simple search involving only a single technique, but even so this does not ensure that the combination is truly comprehensive (as we have demonstrated with the CIG Trials Registry). Essentially this was the goal of the CIG trials registry, but even it did not completely encompass the sample of manuscripts identified by the PubMed MEDLINE search despite involving a MEDLINE search of its own. Similarly, previous reviews of RCT quality have often involved secondary searches following an initial computerized search, but such efforts certainly cannot match the breadth and thoroughness of that from the Cochrane Collaboration. If such reviews of RCT quality are to judge entire fields of medical science it would seem that the search techniques they employ must be shown to produce a representative sampling of the parent population of RCT manuscripts as well as a high yield from that parent population. We hope that the findings presented here bring more attention to this concern in future reviews of the change in RCT quality over time. Conclusions We demonstrated that measuring the change in quality over time of a sample of RCT manuscripts from the field of brain injury can be greatly affected by the search technique. This poorly recognized factor may make measurements of the change in RCT quality over time within a given field of medical science unreliable. The search strategy should be accurately reported in any study that attempts to follow trends in the quality of RCT manuscripts over time, and its limitation in sampling the RCT manuscripts from a field of medical science should be acknowledged and evaluated. Competing Interests The author(s) declare that they have no competing interests. Authors' contributions Both MKB and CY contributed equally to this study, and the exact nature of their contributions are described in the Methods section. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549561.xml

300886

“Suicide” Proteins Contribute to Sperm Creation

You might say that caspases are obsessed with death. The primary agents of programmed cell death, or apoptosis, caspases kill cells by destroying proteins that sustain cellular processes. Apoptosis, a highly controlled sequence of events that eliminates dangerous or unnecessary cells, contributes to a wide variety of developmental and physiological processes—in a developing embryo, apoptosis creates the space between fingers and adjusts nerve cell populations to match the number of cells they target; in an adult, apoptosis counters cell proliferation to maintain tissue size and density. Now it appears that caspases may also play a role in creating life. As Bruce Hay, Jun Huh, and colleagues report, multiple caspases and caspase regulators are required for the proper formation of free-swimming sperm in the fruitfly Drosophila . Caspases, which typically exist in a quiescent state in nearly all cells, are regulated through a complex network of activators and inhibitors. Once activated, a “caspase cascade” ultimately cleaves and irreversibly alters the function of essential cellular proteins, leading to apoptosis. A few of the dozen-plus known caspases appear to contribute to inflammation responses, but the vast majority are enlisted to kill cells. Not surprisingly, cells keep caspase activation under tight wraps. That's why it's intriguing that multiple caspases normally associated with the induction of cell death participate in this nonapoptotic process. During spermatogenesis, germline precursor cells—the cells that generate sex cells—give rise to 64 haploid spermatids. (Sex cells are haploid, containing half the chromosomes found in body cells.) Spermatids are connected by intracellular “bridges” that, along with most other cytoplasmic components, must be expelled in a process called “individualization” to create terminally differentiated free-swimming sperm. Protein structures known as investment cones surround each spermatid nucleus and sweep out the neighboring cytoplasm, bridges, and organelles, forming a bulge that eventually detaches as a “waste bag” as it reaches the sperm tail. This process—elimination of cytoplasm and membrane packaging of individual spermatids—also occurs in mammals. Many types of human infertility result when it is disrupted. To explore how caspases affect this process, Hay's group studied the consequences of inhibiting caspase activity (or the activity of specific caspase activators) in the male germline cells of fruitflies. In both cases, they observed that the bulges and waste bags were either abnormally small or absent and that the normal path of investment cone movement was disrupted. The researchers also inspected the flies to look for structural differences and found that spermatids in both mutant strains remained connected by cytoplasmic bridges and retained residual cytoplasm. Together, the authors conclude, these results demonstrate that individualization depends on caspase activity. Hay's team went on to characterize the pathways that activate caspases during sperm individualization. They found that in one pathway, two key activators of caspase-dependent cell death—Ark and Hid (both of which have mammalian counterparts)—promote the activity or stabilization of the caspase Dronc. A second caspase, Dredd, and its activator Fadd (which also have mammalian counterparts) were also found to be important. Double mutants that removed both Dronc and Dredd activity had more severe defects in individualization than mutants that removed only one or the other, suggesting that these caspases have distinct roles in this process. Interestingly, Drice—the downstream caspase activated just as individualization begins (downstream caspases are typically activated by upstream caspases such as Dronc and Dredd)—was not affected by inhibition of Dronc and Dredd. This result, along with the fact that Dronc and Drice were activated at different times and places, suggests that some other mechanism activates Drice. Different apoptosis-related caspases and caspase regulators, the authors conclude, are recruited through different pathways at distinct points in time and space to create individually packaged, free-swimming sperm, a distinctly nonapoptotic process. Studies in mice suggest that individualization may occur similarly in mammals, with activation of apoptotic caspase cascades resulting in free-swimming sperm and loss of specific caspase activators causing infertility and defective spermatogenesis. The abnormal differentiation and residual cytoplasm seen in caspase-inhibited Drosophila mutants, for example, resemble “cytoplasmic droplet sperm,” a condition seen in infertile men. Insights into the molecular basis of caspase activation in sperm individuation could provide clues to male infertility and suggest possible treatments. Given the widespread role of programmed cell death in supporting processes fundamental to life, perhaps it's not surprising that the agents of apoptosis also support the creation of life. Developing spermatids in a normal Drosophila testis

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC300886.xml

539247

The unmasking of Pneumocystis jiroveci pneumonia during reversal of immunosuppression: case reports and literature review

Background Pneumocystis jiroveci pneumonia (PCP) is an important opportunistic infection among immunosuppressed patients, especially in those infected with human immunodeficiency virus (HIV). The clinical presentation of PCP in immunosuppressed patients have been well-reported in the literature. However, the clinical importance of PCP manifesting in the setting of an immunorestitution disease (IRD), defined as an acute symptomatic or paradoxical deterioration of a (presumably) preexisting infection, which is temporally related to the recovery of the immune system and is due to immunopathological damage associated with the reversal of immunosuppressive processes, has received relatively little attention until recently. Case presentation We aim to better define this unique clinical syndrome by reporting two cases of PCP manifesting acutely with respiratory failure during reversal of immunosuppression in non-HIV infected patients, and reviewed the relevant literature. We searched our databases for PCP cases manifesting in the context of IRD according to our predefined case definition, and reviewed the case notes retrospectively. A comprehensive search was performed using the Medline database of the National Library of Medicine for similar cases reported previously in the English literature in October 2003. A total of 28 non-HIV (excluding our present case) and 13 HIV-positive patients with PCP manifesting as immunorestitution disease (IRD) have been reported previously in the literature. During immunorestitution, a consistent rise in the median CD4 lymphocyte count (28/μL to 125/μL), with a concomitant fall in the median HIV viral load (5.5 log 10 copies/ml to 3.1 log 10 copies/ml) was observed in HIV-positive patients who developed PCP. A similar upsurge in peripheral lymphocyte count was observed in our patients preceding the development of PCP, as well as in other non-HIV immunosuppressed patients reported in the literature. Conclusions PCP manifesting as IRD may be more common than is generally appreciated. Serial monitoring of total lymphocyte or CD4 count could serve as a useful adjunct to facilitate the early diagnosis and pre-emptive treatment of this condition in a wide range of immunosuppressed hosts, especially in the presence of new pulmonary symptoms and/or radiographic abnormalities compatible with the diagnosis.

Background Pneumocystis jiroveci (Pj) (previously known as Pneumocystis carinii f. sp. hominis ) was first identified as a pathogen in premature infants suffering from interstitial plasma cell pneumonia in European countries during and after World War II, occasionally occurring in epidemics [ 1 - 3 ]. Since then Pneumocystis pneumonia (PCP) had only been reported sporadically in patients with malignancies and solid organ transplantations until the HIV epidemic [ 4 ]. The incidence of PCP increased significantly after the emergence of human immunodeficiency virus (HIV) infection. However, with the identification of CD4 T lymphocyte depletion as an independent risk factor for the development of PCP [ 5 ], widespread use of antimicrobial prophylaxis [ 4 ], and the introduction of highly active antiretroviral therapy (HAART), there has been a steady decline in the incidence of PCP among HIV-infected patients [ 7 , 8 ]. Nevertheless, with the rising number of patients receiving immunosuppressive therapies for malignancies, solid organ transplantations and autoimmune diseases, PCP has been increasingly recognized in non-HIV immunosuppressed hosts [ 9 - 15 ]. For instance, PCP occurs in 3.4% to 43% of solid organ transplant recipients [ 16 ], and it is particularly prevalent among those patients who are put on long-term steroids. In a non-HIV immunosuppressed cohort with PCP, the use of steroids was found to be a contributing factor in 87% of patients [ 17 ]. In another similar cohort of immunosuppressed patients, steroids had been administered systemically in 90.5% within one month before the diagnosis of PCP. Although a median daily dose equivalent to 30 mg of prednisone was administered in most of these patients prior to the development of PCP, up to 25% had received as little as 16 mg of prednisone daily [ 18 ]. Interestingly, PCP has also been reported in patients with endogenous steroid excess due to Cushing's disease [ 19 , 20 ]. Paradoxically, the clinical symptoms of PCP were often unmasked in HIV-negative immunosuppressed patients during the reversal of immunosuppression, often at the time when the dose of steroids was tapered [ 11 , 17 , 21 - 24 ], or when the endogenous steroid production was reduced [ 25 - 27 ]. However, serial changes in the absolute lymphocyte count before and during reversal of immunosuppression were not mentioned in these patients. Recently, paradoxical worsening of clinical symptoms and signs of PCP after initiation of HAART has also been reported in HIV-positive patients [ 28 - 31 ]. The onset of clinical deterioration was associated with an upsurge in the CD4 lymphocyte count and a reduction in the HIV viral load [ 28 - 31 ]. Tissue damage is thought to occur as a result of immune reconstitution in HIV-positive patients. Here, we report two cases and review the literature on this topic from the perspective of immunorestitution disease. Case presentation Case 1 This is a fifty-one year old female patient with history of diabetes mellitus and systemic lupus erythematoses (SLE) complicated by lupus nephritis. Although we have included her case in our previous publication [ 32 ], we have not reported her clinical details at that time. She was put on prednisolone 30 mg and azathioprine 100 mg daily since end of June and mid-July 2002, respectively. She was admitted to Queen Mary Hospital on 11 th August 2002 for investigation of jaundice. Investigations revealed deranged liver function tests with cholestatic pattern. A diagnosis of drug-induced hepatotoxicity was entertained, and azathioprine was stopped after admission. As her autoimmune disease was under control, her steroid dosage was reduced from 25 mg to 15 mg daily within the next 14 days. Her CXR taken on admission was normal. Soon after her immunosuppressive therapy was tapered, she developed fever and non-productive cough. A repeat CXR performed on 9 th Sept revealed new infiltrate over the left mid-zone, suggestive of pneumonia. She was started on intravenous ceftazidime 1 gram eight hourly and oral clarithromycin 500 mg twice daily. Serial CXR performed three days later showed increasing bilateral pulmonary infiltrates and worsening hypoxemia. There was an upsurge of total lymphocyte count from 0.7 × 10 9 /L (total white cell count 7.2 × 10 9 /L) at the time of admission to 5.6 × 10 9 /L (total white cell count 10.8 × 10 9 /L) at the time of clinical deterioration. Bronchoscopy with transbronchial biopsy performed on the same day revealed Pneumocystitis jiroveci by methenamine sliver stain. Workup for other opportunistic pathogens including cytomegalovirus and aspergillus was negative. She was commenced on intravenous pentamidine (4 mg/kg/day) and corticosteroids for severe PCP infection. Despite active treatment she developed progressive respiratory failure and required admission to intensive care unit. She subsequently recovered after a stormy hospital course, and upon discharge from hospital, her total lymphocyte count had returned to her baseline of 0.86 × 10 9 /L. Case 2 A thirty-three year old gentleman initially presented to Prince of Wales Hospital with a diagnosis of SLE/dermatomyositis overlap syndrome. He was treated with steroid and hydroxychloroquine 200 mg twice daily since 1997. He had a flare up of disease in May 1998 with active vasculitis and myositis, for which he was put on prednisolone and azthioprine 50 mg and 100 mg daily respectively. Upon reassessment one month later, disease activity was under control, and the dosage of prednisolone was reduced to 45 mg daily. Twelve day after reducing the immunosuppressive regimen, he was admitted to hospital for treatment of left buttock abscess. The CXR taken on admission was unremarkable. An aspirate of the pus from the lesion grew methicillin-sensitive staphylococcus aureus ; he was treated with cloxacillin 1 g intravenously every 6 hourly, together with incision and drainage of the buttock abscess. In view of the underlying active pyogenic infection, the steroid dosage was rapidly tapered from 45 mg to 15 mg daily within the next four days. However, he was noted to have persistent fever associated with mild unproductive cough. A repeat chest radiograph showed new infiltrates over the right upper and left lower zones, and he was empirically treated with intravenous ceftazidime 1 gram every 8 hours, cloxacillin 1 gram every 6 hours and netimicin 100 mg every 8 hours. As there was no clinical response after 5 days of treatment, bronchoscopy and bronchoalveolar lavage (BAL) was performed, which was positive for Pneumocystis jiroveci . Investigation for the presence of co-existing opportunistic pathogens such as cytomegalovirus and aspergillus was negative. On the day after bronchoscopy, he was commenced on intravenous cotrimoxazole 1.3 grams every 6 hours. He remained stable initially with fever on downward trend. However, on the 3 rd day of treatment, he developed sudden desaturation with resurgence of high fever, and required supplemental oxygen therapy. Repeat chest radiograph showed increased perihilar hazziness in both lung fields. There was also an upsurge of total lymphocyte count from 0.6 × 10 9 /L (total white cell count 11.2 × 10 9 /L) on admission, to 1.3 × 10 9 /L (total white cell count 10.4 × 10 9 /L) at the time of clinical deterioration. He was treated with high dose prednisolone (80 mg daily), and his condition improved promptly afterwards. He was subsequently discharged, and on follow up at the clinic one month later, his total lymphocyte count had returned to his baseline level of 0.6 × 10 9 /L. Immunorestitution disease (IRD) has been described in both HIV and non-HIV immunosuppressed hosts previously [ 27 - 31 ]. In the setting of PCP, it is defined as an acute symptomatic presentation of the disease that is related temporally to the recovery of the immune system, associated with reversal of immunosuppressive processes such as reduction in the dosage of corticosteroids and/or cytotoxic agents or a reduction of HIV viral load due to HAART, which results in the development of immunopathological damage. The preexisting microbial infection could be either asymptomatic or mildly symptomatic. Using this case definition, we attempted to review the English literature for other reported cases of PCP manifesting as IRD. The English-language literature (1966 – 2003) was searched in the Medline database of the National Library of Medicine in October 2003. The keywords " Pneumocystis carinii ", " Pneumocystis jiroveci ", "HIV", immunosuppression", "immunosuppressive", "steroid", and "corticosteroid" were used to select cases. All the case reports and case series with clinical details were included in this study if they fulfilled the above definition of IRD. When appropriate, the cited bibliographies were also retrieved for further analysis. As for statistical analysis, we used the Wilcoxon Signed Rank test, a non-parametric test for comparing paired samples, to analyze the serial changes in lymphocyte counts and HIV viral loads before and during the development of IRD. A two-tailed p-value of less than 0.05 was considered significant. All statistical analyses were performed using SPSS version 11.5 for Windows. Including our present case, a total of 29 cases of PCP in non-HIV immunosuppressed hosts fulfilling our definition of IRD have been reported in the literature (table 1 ) [ 22 - 27 , 32 ]. There were altogether 13 males and 8 females, with a median age of 38 years (range 2 to 75 years). The age and sex were not mentioned in 8 cases. Their underlying immunosuppressive conditions included solid organ tumours (13 cases), haematological diseases (8), autoimmune diseases (4), endogenous Cushing's disease (3), and a solid organ transplant recipient (1). All patients had received steroids or had excessive endogenous steroid production, whereas 18 (62.1%) of them had concomitant cytotoxic therapy for the underlying diseases. The median duration between steroid tapering and clinical manifestations of PCP was 21 days (range 1 to 83 days). Steroids were completely withdrawn at a median of 7.5 days (range 1 to 21 days) before the onset of symptoms in eight patients. Serial lymphocyte counts were only available in eight patients. An upsurge of the absolute lymphocyte counts was observed from the time of reduction of immunosuppression (median 300/μL, range 290 to 600/μL at baseline) to the time of occurrence of IRD (median 1200/μL, range 600 to 5620/μL); the median increase in total lymphocyte count was 800/μL, with a range of 300 to 4880/μL. Comparing the lymphocyte counts before and after reversal of immunosuppressive therapy, the difference was statistically significant (Wilcoxon Signed Rank Test for paired samples; p = 0.012). In addition to our patient, reintroduction or increasing doses of steroids were required in 7 (53.8%) of 13 patients in the acute management of PCP in the literature, at the time when they developed clinical deterioration during antimicrobial therapy [ 24 - 27 ]. Seven (53.8%) of 13 cases had respiratory failure requiring mechanical ventilation. Among these 29 cases, 13 (44.8%) subsequently died of PCP. Among HIV-positive patients, 13 cases with newly diagnosed PCP were reported in the literature, in which IRD occurred shortly after the introduction of HAART (table 2 ) [ 28 - 31 ]. Seven (53.8%) out of 13 cases received steroids as adjunctive therapy in addition to antimicrobials. HAART was given in all cases at a median 18 days (range 1 to 35 days) after the initiation of treatment for PCP. During IRD, recurrence of fever (100%), dyspnoea (100%), and paradoxical worsening of pulmonary infiltrates (58.3%) were observed in these patients [ 28 - 31 ]. IRD occurred at a median of 14 days (range 5 to 17 days) after HAART. An upsurge of the CD4 lymphocyte count was observed before (median 28/μL, range 4 to 290/μL) and during IRD (median 125/μL, range 30 to 564/μL); this was associated with a concomitant reduction of the median HIV viral load from 5.5 log 10 copies/ml (range 5.0 to 5.9 log 10 copies/ml) to 3.1 log 10 copies/ml (range 2.9 to 4.5 log 10 copies/ml) before and during IRD respectively, and the differences observed in both the CD4 counts and viral loads before and during IRD reached statistical significance (Wilcoxon Signed Rank Test for related samples; p = 0.001 and 0.017, respectively). Antimicrobials, steroids, or both for PCP were reintroduced for IRD in 4, 1, and 6 cases respectively. Only 2 cases were treated conservatively. One case required mechanical ventilation for severe respiratory distress. None of the patients died. PCP manifesting as a form of IRD is not a rare phenomenon. As shown in our previous study, it happens in 7 out of 10 (70%) of HIV-negative immunosuppressed hosts infected with Pj [ 32 ]. However, the diagnosis of PCP is usually delayed in this group of patients because of atypical presentation. In this clinical setting, PCP manifesting as IRD often runs an acute and fulminant course, with nonspecific lesions on chest radiographs, and high absolute lymphocyte counts [ 32 ]. In our own reported series, despite the administration of steroid therapy to suppress the immunopathological damage, more than 80% of patients developed acute respiratory failure and required mechanical ventilation. Patients who developed PCP during reversal of immunosuppressive therapy in our series tended to be older, and this might partially explain the increased mortality observed in this group [ 32 ]. Rapid reduction of immunosuppressive therapy such as steroids has been implicated as a predisposing factor for the development of PCP in HIV-negative patients [ 11 , 17 , 23 , 24 ]. In one study, PCP occurred in 79 (70%) of 113 patients during steroid tapering [ 17 ]. Another study suggested that 8 (72.2%) out of 11 episodes of PCP developed when steroid therapy was tapered [ 23 ]. A subsequent study also demonstrated that 43% of patients had a rapid reduction of steroid dosing before the clinical manifestations of PCP [ 11 ]. A similar experience was reported in children, and 17 (89.5%) out of 19 children were diagnosed to have PCP during steroid tapering according to a previous report [ 21 ]. Another series revealed that 7 of 11 patients experienced acutely symptomatic PCP when the dose of steroids was decreased or terminated 5 days to 3 weeks before the diagnosis of PCP [ 22 ]. However, all these cases were not analyzed from a perspective of IRD. Serial changes of the absolute lymphocyte counts or their subsets were either not noted or reported [ 11 , 17 , 21 - 24 ]. Hence we have not included these cases for further analysis in this review. Among HIV positive patients, PCP manifesting acutely during the initiation anti-retroviral therapy is a well-recognized phenomenon. The underlying immunopathological nature of this condition, which is reminiscent to IRD occurring in non-HIV infected patients, has been confirmed by histological examination of the lungs and transbronchial biopsy specimens, which demonstrated mixed inflammatory infiltrates including macrophages, neutrophils, lymphocytes, and plasma cells. Almost all infiltrating lymphocytes found in the tissues were of the T cell lineage, shown by immunophenotyping to be predominantly CD4 and CD8 cells [ 28 ]. In another study [ 30 ], the BAL fluid obtained from one of six patients with an IRD-type presentation of PCP was analyzed. Infiltration of predominantly CD4 and CD8 lymphocytes with the proliferative marker (Ki67) and perforin-positive cell were seen in the BAL specimen. Therefore, it is likely that the phagocytosed Pj is presented by alveolar macrophage to T cells, which trigger the inflammatory response [ 30 ]. In our own experience, as well as from the review of published literature, it appears that a surge of absolute lymphocyte count, especially the CD4 lymphocyte count in HIV-positive patients, could potentially act as a surrogate marker for immunopathological damage during IRD in both HIV-negative and HIV-positive patients. In our recent publication [ 32 ], 7 out of 10 non-HIV immunosuppressed patients demonstrated a consistent rise in the absolute lymphocyte count during tapering of immunosuppression prior to the onset of symptomatic PCP. In this group of patients, the surge in lymphocyte count is likely the result of withdrawal of lymphocytotoxic immunosuppressants such as corticosteroids. Similarly, a rising trend of the CD4 lymphocyte count, consistent with immune reconstitution after HAART, was also observed in 13 HIV-positive cases before and during the development of symptomatic PCP [ 28 - 31 ]. In fact, an upsurge in the absolute lymphocyte count has been shown to be a marker of IRD in our previous publications involving viral and tuberculous infections [ 33 - 36 ]. However, it must be emphasized that the number of circulating lymphocytes may not always correlate with their number in the affected tissues or their in vivo functional activity. This can be exemplified by a case of PCP occurring during steroid withdrawal, in which the lymphocyte counts surged to a very high level and then rapidly dropped to a low level within one day. The migration of lymphocytes from the circulation to tissue might explain this rapid drop in lymphocyte count and the resulting immunopathological damage [ 27 ]. In the future, further studies on lymphocyte subsets and cytokine profiles of susceptible hosts during the development of IRD should be performed to elucidate the underlying immunopathological mechanisms behind this interesting phenomenon. From the result of this review, it appears that HIV-positive patients with PCP are at risk of clinical deterioration due to IRD if HAART therapy is started within 1 to 2 weeks after the initiation of treatment for PCP (table 2 ). With a better understanding of the pathogenetic mechanisms resulting in IRD, we may be able to prevent the occurrence of IRD by delaying the initiation of HAART in HIV-positive patients with PCP. However, in non-HIV immunosuppressed patients, it is even more important to recognize the atypical presentations of PCP in the context of IRD. Since the clinical and/or radiological features alone may not be sufficient for diagnosis, analysis of serial changes in lymphocyte counts in patients undergoing a reduction of immunosuppression can alert the clinician to the possibility of IRD due to occult pathogens such as Pj. To prevent IRD in non-HIV immunosuppressed patients, the use of prophylactic antibiotics against Pj to reduce the microbial load in selected patients remains an important issue. Recently, a multi-center study showed that the CD4 lymphocyte count may be a useful marker to monitor the risk of development of PCP in non-HIV immunosuppressed hosts [ 37 ], and patients with low CD4 lymphocyte counts of less than 300 or 400 may require prophylaxis. In fact, asymptomatic colonization of Pj has been demonstrated in HIV-negative patients when the CD4 lymphocyte count was less than 400 [ 38 ]. Nested polymerase chain reaction (PCR) identified a significant percentage of clinically silent Pj colonization in 20% of non-HIV immunosuppressed patients [ 39 ]. Therefore, early detection of asymptomatic infection of Pj in blood and respiratory specimens before, and during intense immunosuppression may enable selection of cases for pre-emptive treatment of Pj infection in order to prevent the development of IRD during reversal of immunosuppression [ 40 , 41 ]. Conclusions PCP occurring in the context of IRD is not a rare phenomenon and is likely to be under-reported in the literature. In this setting, it may be more common for PCP to manifest acutely with a fulminant clinical course. Clinicians caring for immunosuppressed patients should be alert to this unique phenomenon so as to initiate timely and appropriate investigations and treatment for their patients. Serial monitoring of lymphocyte count, or if possible CD4 count, could serve as a useful adjunct to facilitate the diagnosis and management of this condition in a wide range of immunosuppressed hosts, especially in the presence of new pulmonary symptoms and/or radiographic abnormalities compatible with the diagnosis. Competing interests The author(s) declare that they have no competing interests. Authors' contributions RAL and DSH were involved in the clinical evaluation and treatment of patients. BST and IFH helped with literature searching and review. AKW and VCC drafted and refined the manuscript. KYY conceived the study, participated in its design and coordination, and supervised the preparation of the manuscript. All authors have read and approved the final draft of the manuscript before submission. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539247.xml

526373

Long-term patterns in European brown hare population dynamics in Denmark: effects of agriculture, predation and climate

Background In Denmark and many other European countries, harvest records suggest a marked decline in European brown hare numbers, a decline often attributed to the agricultural practice. In the present study, we analyse the association between agricultural land-use, predator abundance and winter severity on the number of European brown hares harvested in Denmark in the years 1955 through 2000. Results Winter cereals had a significant negative association with European brown hare numbers. In contrast to this, root crop area was positively related to their numbers. Remaining crop categories were not significantly associated with the European brown hare numbers, though grass out of rotation tended to be positively related. The areas of root crop production and of grass out of rotation have been reduced by approximately 80% and 50%, respectively, while the area of winter cereals has increased markedly (>70%). However, European brown hare numbers were primarily negatively associated with the number of red fox. Finally, we also found a positive association between mild winters and European brown hare numbers. Conclusion The decline of Danish European brown hare populations can mainly be attributed to predation by red fox, but the development in agricultural land-use during the last 45 years have also affected the European brown hare numbers negatively. Additionally, though mild winters were beneficial to European brown hares, the increasing frequency of mild winters during the study period was insufficient to reverse the negative population trend.

Background In most countries in Western Europe, the landscape has undergone dramatic changes during the last century due to changes in the agricultural practices. In Denmark, both the uncultivated land and the semi-cultivated land, such as permanent grass areas, have been reduced dramatically, reflecting the general intensification of agriculture [ 1 ]. Additionally, fields have become larger, which has resulted in widespread fragmentation of remaining habitats, and today the landscape appears as a mosaic of natural habitats surrounded by cultivated land [ 1 , 2 ]. These changes in agriculture have affected a number of wildlife species living in this man-made landscape. For instance, the shift in agricultural practice has severely influenced the diversity and abundance of insects with concomitant consequences for the dynamics of a wide range of farmland birds [ 3 , 4 ], including the grey partridge ( Perdix perdix ) [ 5 ]. Among mammals, the European brown hare ( Lepus europaeus ) in particular has experienced a dramatic decline in many European countries [reviewed by [ 6 ]], including Denmark [ 7 , 8 ]. Despite its currently declining numbers, the European brown hare is still common and one of the most important game species throughout the country [ 8 ]. The dynamics of European brown hares seem resilient to even heavy hunting pressure [ 9 ], though local population dynamic data may be needed to ensure sustainable harvest [ 10 ]. In Denmark hunting of European brown hares is generally assumed to be without regulating effect [ 8 ]. The European brown hare is a typical grass steppe herbivore, and inhabits primarily open landscapes, including cultivated farmland [ 11 ], which is the predominant landform in Denmark [ 1 ]. The species is rather sedentary, and has generally small home ranges [e.g. [ 12 ]]. This site fidelity makes European brown hares highly susceptible to changes in their surrounding habitats, and the general decline in the European brown hare populations in Europe is mainly being attributed to changes in agriculture practice and land-use [reviewed by [ 6 , 12 ]]. European brown hares are important prey primarily for mammalian predators. In Northern Europe, the red fox ( Vulpes vulpes ) is the main predator on European brown hares, and foxes have been reported to influence the dynamics of several European brown hare populations substantially [ 13 - 17 ]. Also, infectious diseases such as the European brown hare syndrome virus, pseudotuberculosis, pasteurellosis and coccidiosis are present in many European brown hare populations [ 18 - 20 ]. Haerer et al. [ 19 ], however, concluded that diseases were not responsible for the decline of brown hare populations in Switzerland. Similarly, Frölich et al. [ 20 ] found that compared to red foxes, infectious diseases seemed to play a minor role in the dynamics of European brown hare populations in Germany. An increasing number of papers have documented the importance of climate for a number of life history traits and abundance of many terrestrial vertebrates [ 21 ], and European brown hare populations are affected negatively by cold winters and cold springs [ 12 , 22 - 24 ]. Factors regulating vertebrate populations may, however, exhibit large spatial variation, and even among proximate populations spatial variation and gradients in vertebrate population dynamics may exist [e.g. [ 25 ]]. In this study, we analyse and contrast the simultaneous associations between agricultural land-use, the number of red fox harvested, winter severity and the population dynamics of European brown hares across 11 Danish districts during 45 years. Results In the period 1955–2000 the European brown hare harvest record declined steadily in all the Danish districts but one: On the island Bornholm the European brown hare population declined until the late 1980s, but has increased markedly since then, and has now reached a level higher than that of 1955 (Fig. 1 ). Figure 1 Harvest records of European brown hare (bags per 1000 ha) from 11 Danish districts, 1955–2000. We found statistical significant direct density-dependence ( X t-1 ) in the European brown hare time series (G = 281.4, df = 2, P < 0.0001). Additionally, the effect of district was also significant (G = 769.0, df = 1, P < 0.0001). From 1955 to 2000, agricultural land-use changed markedly in Denmark, resulting in large temporal shifts in the areas covered by the different crop categories (Fig. 2 ). The areas covered with grass and green fodder in rotation and in particular the areas out of rotation have been reduced, the latter by approximately 50% since the mid 1950s. An even more dramatic decline has been observed for the root crops in the same period, a decline by more than 80% (Fig. 2 ). In Storstrøm, however, the area covered with root crops has been relatively stable over the years. In general, cereals were the dominant crop category in 1955 through 2000, with a shift from a predominance of spring cereals to winter cereals in the 1980s (Fig. 2 ). Figure 2 The total areas of the seven crop categories in Denmark 1955–2000. Note that prior to 1960 no data on winter cereals exist. The areas covered with winter cereals had a marked negative association with the number of European brown hares (Table 2 ), whereas root crops had a marked positive relation. Neither spring cereals, nor winter and spring rape seemed to be associated with European brown hare numbers (Table 2 ). Similarly, neither grass areas in or out of rotation were significantly related to European brown hare numbers, though the latter tended to have a positive association (Table 2 ). Table 2 Summarised results from the analysis of the impact of agricultural land-use, red fox and winter climate on European brown hare harvest records from 11 Danish counties, 1955–2000. Also given is the change in model deviance (Δ deviance) explained by the variable when fitted last. Type 3 sums of squares. Variable Coefficient SE F value P value Δdeviance Winter cereals (t) -0.08322 0.03778 4.85 0.0282 37.7 Spring cereals (t) -0.06243 0.04675 1.78 0.1825 2.5 Grass in rotation (t) 0.03217 0.03562 0.82 0.3669 4.1 Grass out of rotation (t) 0.05766 0.03070 3.53 0.0612 1.7 Winter rape (t) -0.03272 0.03094 1.12 0.2909 4.0 Spring rape (t) -0.02704 0.03948 0.47 0.4938 4.2 Root crops (t) 0.10370 0.04753 4.76 0.0297 0.4 Fox (t) -0.01791 0.02024 0.78 0.3767 5.2 Fox (t-1) -0.15720 0.01971 63.62 <.0001 52.8 NAO (t) 0.03732 0.01110 11.31 0.0008 4.0 Note: Due to the standardization of model variables, regression coefficients and Δ deviance values are directly comparable. The number of red foxes harvested in year t-1 had a marked negative effect on the number of European brown hares harvested the following year (Table 2 ), whereas red fox number in year t seemed unimportant for the European brown hare numbers. Mild winters, i.e. high winter state of the large-scale climatic phenomenon the North Atlantic Oscillation [NAO; [ 26 ]], had a significant positive effect on the European brown hare numbers (Table 2 ). Discussion Despite its high reproductive potential [e.g. [ 27 ]], the Danish European brown hare has, according to annual harvest records, declined dramatically since 1955. Studies of Danish European brown hare populations indicate that its reproductive success is low compared to that of con-specifics in other countries [ 27 ], and has, in turn, declined from the 1950s to the 1990s [ 28 ]. Hansen [ 27 ] suggested that the low reproductive success observed in Danish European brown hares might be attributed to the agricultural practice. Using data covering almost half a century, our analyses suggest that the dramatic decline in the Danish European brown hares can be attributed mainly to the negative effect of red foxes, but also to the agricultural land-use. The area of winter cereal production has increased dramatically during the last century, and became the dominating crop in the early 1990s. We also found a significant positive association between root crops, a crop type that has declined dramatically in the second half of the last century, and European brown hare numbers. European brown hares mainly forage on grasses and herbs [ 12 ], and cereals such as wheat are preferred during winter [ 29 , 30 ], which seems to contradict the results of our analyses. However, as European brown hares choose to feed on specific crops according to plant phenology [ 12 , 31 ], cereals, although important in winter, still occupy large areas when no longer of nutritional value to European brown hares, which may result in low availability of food especially during summer. Similarly, rape is avoided in the diet [ 29 ], but European brown hares may spend a substantial fraction of their time in rape fields during winter [ 30 ] prior to the development of glucosinulates [see [ 29 ]]. Apart from winter cereals and root crops, the crop categories did not affect the European brown hare numbers significantly. The lacking effect of grass and green fodder areas, especially those out of rotation, was unexpected as other studies have shown that hares prefer grass areas year-round [ 12 , 30 ]. This, however, may be attributed to the fact that we were unable to separate grass areas into those e.g. with and without cattle, factors that might affect European brown hare use of grass areas [ 12 ]. In a recent study, Fox [ 32 ] showed that farmland birds seemed to benefit from the reduced application of pesticides and inorganic fertilisers seen in Denmark since the early 1980s. The continuing decline in European brown hare numbers in that period therefore suggests that hares are not directly affected by the use of such substances, but rather respond to the loss of suitable habitat and space. European brown hares mainly move along field margins [ 12 , 33 ], and the increasing field size [ 34 ] combined with the general loss of suitable habitats possibly force hares to aggregate in the remaining patches of non-agricultural and non-urbanized land. Here, density-dependence may perpetuate the negative population development as European brown hares aggregate in the few, remaining pockets of suitable habitat. In line with this, Frylestam [ 23 ] reported an inverse relationship between fertility and density in European brown hares, which he attributed to shortage of food at least in some parts of the year, which also has been suggested in a number of other studies [ 12 , 24 , 27 ]. Hence, agricultural land-use, especially the increasing use of winter cereals and the marked reduction in root crops, but probably also grass areas out of rotation, seem to have contributed to the European brown hare decline in Denmark. However, the single-most important parameter for the number of European brown hares was the number of red foxes. Hence, our results are consistent with other studies reporting that red foxes may affect European brown hare populations negatively through predation [ 13 - 17 ]. This relationship is also particular evident from the positive development in the European brown hare population on the island Bornholm following the outbreak of sarcoptic mange there [ 35 ]. To elaborate the fox-hare interaction further, we reran the analyses including all variables but the red fox variables, and added the delayed AR term (i.e. X t-2 ). In seven of the 11 districts the delayed density dependence was significant (P < 0.05), suggesting important inter-specific interactions [ 36 , 37 ]. The significant association with red fox numbers in year t-1 (and not year t ) most likely reflects that compared to adults, juvenile European brown hares suffer more from predation [e.g. [ 19 ]]. Hence, high red fox numbers in year t result in low harvest of European brown hares in year t , which affects the reproductive potential of the populations, and, hence, the number of European brown hares the following year (year t+1 ). Also, the significant effect of district may point to differences in habitat quality, but also differences in the history of the sarcoptic mange, i.e. time since the eruption of the mange, among districts [see [ 35 ]]. Both European brown hare over-winter survival [ 24 ] and reproductive rate [ 23 ] increases with temperature, which may be attributed to improved forage availability following mild winters [see [ 38 , 39 ]]. Our analyses revealed similar results as mild winters affected the European brown hares positively. There may, however, also be a negative effect of mild winters, namely through transmission of diseases and parasites, which may be enhanced under mild climatic conditions [ 18 ]. Nevertheless, the overall effect of mild winters seemed positive. The upward trend in the NAO since the 1960s [ 40 ], however, was not sufficient to reverse the European brown hare decline, but may have decelerated it. Conclusion Our analyses have provided important insight into the structure of the European brown hare dynamics in Denmark, and documented important patterns within the mechanisms regulating European brown hares. Hence, we have documented that the decline of European brown hares in Denmark mainly can be attributed to predation by red fox, but also to changes in agricultural land-use. Additionally, though mild winters were beneficial to European brown hares, the increasing frequency of mild winters during the study period was insufficient to reverse the negative population trend. Methods European brown hare and red fox density indices As indices of European brown hare and red fox density, we used the harvest records from 14 Danish counties from 1955–2000. In 1970, the geographical borders of some of the counties were changed, which in two cases lead to substantial changes in area [ 7 ]. This, together with inconsistency in crop registration, forced us to lump together some counties, and we therefore present analyses of European brown hare harvest records from 11 districts (Fig. 1 ). European brown hare and red fox harvest records were expressed as the number of animals shot per hectare per year. Harvest records may seem rather crude indicators of density compared to e.g. spotlight surveys and line transects [see [ 41 ]]. Unfortunately, no alternative indices of European brown hare density in Denmark are available. Harvest records, however, may still be useful indicators of the long-term trends in European brown hare numbers [ 41 ]. Sarcoptic mange was first encountered in Denmark in the early 1980s, and is now present in seven of the 11 districts [ 35 ]. In one district (Bornholm; Fig. 1 ), the disease has practically eliminated the entire red fox population, and red fox hunting has been prohibited here since 1991. Consequently, we restrict our analyses of the Bornholm district to 1955 through 1990. In all other districts, red fox hunting is open from 1 September to 31 January, and for European brown hares from 1 October to 31 December. Neither red fox nor European brown hare hunting is quota-based. Agricultural land-use Time series quantifying agricultural land-use data, i.e. annual crop areas in hectares, covering the period 1955–81 were obtained from Statistics Denmark. Data for the period 1982–2000 were obtained from the official website of Statistics Denmark . The annual crop data were grouped into seven categories (Table 1 ). Table 1 Description of model variables used in the analyses of European brown hare harvest records from Denmark, 1955–2000. Variable Description Winter cereals Winter cereals, including winter wheat, winter barley, and winter rye Spring cereals Spring cereals, including spring wheat, spring barley, spring rye, and oat Grass in rotation Grass and green fodder in rotation, including grass, clover, maize, and Lucerne Grass out of rotation Grass areas permanently out of rotation Winter rape Winter rape Spring rape Spring rape Root crops Sugar beets for sugar production and fodder, mangolds, turnips, and carrots for feed Fox The number of red fox harvested NAO The winter state of the North Atlantic Oscillation No data on the fraction of winter and spring cereals, respectively, were available for 1955–61, and these data points were omitted in the analyses. Similarly, prior to 1982 no data on the fraction of winter barley and spring barley were available. However up until 1968, winter barley was only sown on a few thousand hectares in Denmark, when it was forbidden due to its function as reservoir for mildew attacking spring barley [ 42 ]. The use of mildew resistant winter barley was permitted in 1982. Hence, from 1955–81 barley was regarded as spring barley. Finally, no data on the fraction of winter and spring rye are available from 1955–61 and 1979–98, but in Denmark spring rye is generally sown on very few hectares only [ 42 ], and we regarded rye as winter rye unless otherwise specified [see also [ 32 ]]. Oat is sown as spring crop only. No data on the areas covered with grass and green fodder in or out of rotation were available for 1994, and the data point was interpolated, i.e. 1994 equalled the average of 1993 and 1995. Climatic variability As an index of winter severity, we used the winter state of the North Atlantic Oscillation [NAO; [ 26 ]]. The climatic phenomenon NAO is defined as the difference in sea-level pressure between Portugal and Iceland, affecting the temperature, precipitation and wind across the Northern Hemisphere [ 26 , 43 , 44 ]. The NAO, thus, integrates the effects of a number of abiotic factors, and therefore seems particularly useful when analysing the dynamics of larger, endothermic animals with relatively large buffer-capacity against climatic variability. The relationship between the winter NAO (December t-1 through March t ) and temperature, precipitation and wind is particularly strong in Northern Europe [ 45 ]. When NAO is in its high state, the winter climate in Northern Europe is warm, wet, and windy, while when in its low state winters are cold and dry [ 44 ]. The NAO winter index has shown an upward trend since the 1960s, which accounts for a substantial fraction of the general warming in the Northern Hemisphere [ 40 ]. The NAO index data are available from the Climate Analysis Section, National Center for Atmospheric Research, USA . Data analyses To remove heteroscedasticity in the system, data were log e -transformed, i.e. X (t) = ln( N (t) + 1), where N (t) is the number of European brown hares harvested in year (t) . To allow the direct comparison of regression coefficients, variables were standardized prior to the analyses (i.e., [ X (t) - µ (1955–2000) ]/ s (1955–2000) ). Regression coefficients therefore express the rate of change in standard deviation units of the independent variable per one standard deviation unit of the dependent variable [ 46 ]. To obtain stationarity in the time series [see [ 47 ] for details], data were detrended following [ 48 ] by including year as covariate in all models. For each district, we tested for the presence of multi-collinearity among parameters prior to the analyses by means of condition indices and variance proportions [ 49 ], but multi-collinearity was not observed in any districts (condition index < 12.47; [see [ 49 ] for details]). The European brown hare time series were tested for non-linearity in X (t) vs. X (t-1) by applying the likelihood ratio test [ 50 ]. In all districts, linearity in X (t) vs. X (t-1) was not rejected λ< 7.76, P > 0.05). We analysed the data using a first-order autoregressive (AR(1)) mixed model approach with district as fixed factor: X (t) ~( µ , V ), where and where LAND_USE represents the seven crop categories (see Table 1 ). Because of the inclusion of red fox numbers, we did not include a delayed AR term (i.e. X t-2 ), as inclusion of both delayed density dependence and predator abundance can be seen as redundant [ 37 ]. All analyses were performed in SAS 8.2, using the PROC MIXED procedure with restricted maximum-likelihood estimation of regression coefficients [ 49 ]. Model reduction was conducted using log-likelihood ratio tests [ 51 ]. Authors' contributions NMS designed the study, participated in data preparation, carried out the statistical analyses, and drafted the manuscript. TA participated in data preparation. MCF supported the data analyses and contributed to the writing. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC526373.xml

546336

The Bottleneck of Central Processing: Clues from Reaction Times

Between stimulus and response lies a black box—the mind—whose inner workings are largely unmapped. One of the essential questions about those inner workings concerns the serial versus parallel nature of their processing capabilities. Parallel processing allows multiple tasks to proceed at once, while serial processing creates a bottleneck through which multiple tasks must pass, one at a time. Any reasonably complex task is likely to involve both parallel and serial components, and parsing a task into its components is a central goal for researchers of cognitive processing. In a new study, Mariano Sigman and Stanislas Dehaene propose a model of cognitive processing for a set of simple tasks in which a bottleneck occurs between initial sensory processing and motor response. They predict that this bottleneck will contribute significantly to variations in response time as the cognitive challenge increases and verify this by testing people on a specific numerical evaluation task. A simple but highly effective technique for examining bottlenecks is to measure variations in response time for a task as the stimulus is varied in some small but cognitively challenging way, or when the stimulus is presented along with a stimulus for a competing task. The task in this study was to determine whether a presented number was greater than or less than 45. The complexity of the task was determined by three variables: notation, distance, and response complexity. Notation was varied by presenting the number either as a numeral or its spelled-out equivalent (for instance, “36” or “thirty-six”). Distance was varied by presenting numbers either closer to or further from 45 (for instance, 31 versus 36), and the required response was either one or two finger taps. In a separate series of experiments, the researchers challenged the subject with an interfering tone-recognition task at the same time as or slightly after the presentation of the numerical task. Comparing human reaction times The use of these two sets of experiments allowed the authors to deduce two different kinds of information about the processing involved in the number task. First, by varying the delay in presentation of the tone-interference task and measuring the delay in response time for the number task, they showed that both number perception and motor response can proceed in parallel with other, competing tasks, while the central component of the number task—determining distance—was processed serially, through a central bottleneck. Next, Sigman and Dehaene turned off the tone, and asked how variations in notation, distance, and response complexity altered the variance in the response time—that is, the spread of values for the same task by the same subject. For all three variables, the more challenging task (numbers presented as words, smaller distance, or two taps) had an increased response time. However, only the calculation of distance increased the spread of values obtained in multiple trials. This further suggests that only the central calculation step—what the authors refer to as stochastic integration—proceeds through a central bottleneck, while the other two components can be processed in parallel. Thus, in both experiments, the task was parsed by the brain into the same components, with the serial component being the one subject to the most variance. Sigman and Dehaene note that the ability of the perceptual and motor parts to be performed without central computation depends on the extreme simplicity of the tasks in this experiment. More complex motor challenges, for instance, undoubtedly would require some central input, and thus proceed through the bottleneck. Similarly, a high degree of training in the numerical distance task would likely increase the automaticity of the response, thus avoiding the central slowing seen in the task-naïve subjects studied here.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC546336.xml

545603

A reverse genetic screen in Drosophila using a deletion-inducing mutagen

A new reverse-genetics mutagenesis method that uses the crosslinking drug hexamethylphosphoramide to introduce small deletions has been used to generate and screen pools of mutagenized Drosophila , identifying two mutants.

Background The fruitfly Drosophila melanogaster has been the prime genetic model organism for almost a century. This success story is mainly founded on countless so-called forward genetic screens designed to elucidate gene functions on the basis of their mutant phenotypes. Many of those screens reached a scale that has been termed 'saturating' as they identify all nonredundant genes involved in a certain phenotypic trait. However, forward genetic screens are limited in that they are only capable of uncovering functions that are easily measurable or visible. Furthermore, genes having a redundant or nonessential role are less likely to be found by forward genetics. The reverse genetic approach to unravel gene function starts with the DNA sequence. Mutations within the gene are induced and identified by various techniques and only subsequently is the mutant phenotype analyzed [ 1 ]. Reverse genetics may be undirected or directed, the undirected approach involving random mutagenesis, commonly by transposable elements or by chemicals, the establishment of mutant collections, and the identification of mutations in the gene of interest [ 2 - 5 ]. In contrast, directed reverse genetics is based on techniques that allow for specific inactivation of a gene. These include specific knockdown of gene activities through RNA-mediated interference (RNAi) [ 6 , 7 ] and targeted gene disruption [ 8 , 9 ]. Both undirected and directed reverse genetic techniques have certain advantages and drawbacks. Transposon-based mutagenesis tends to be nonrandom because of the occurrence of hotspots for transposon integration. The use of transposable elements of different origin, such as P-elements and piggyBac, which exhibit a different insertion bias, can partly circumvent this problem. However despite large-scale efforts, the ultimate goal of covering the whole Drosophila genome by insertion mutagenesis is far from being achieved [ 10 , 11 ]. Moreover, while null mutants of P-element-tagged genes (P-elements have the tendency to integrate 5' to a gene) can easily be generated by imprecise excision, piggyBac transposons only excise precisely [ 10 ]. RNAi and small interfering RNA (siRNA) screens provide a powerful tool to dissect the function of genes at a genome-wide scale [ 12 - 14 ], but the technique is most easily applied to cell cultures and is thus limited to cell-biological problems. Large-scale RNAi screens in multicellular organisms have been done only in C. elegans [ 15 ] and for technical reasons a similar approach in Drosophila is not feasible. Targeted gene knockout in Drosophila allows for generation of both null as well as hypomorphic mutations [ 16 ]. However, the technique is time-consuming and technically challenging and hence not applicable on a large scale. Random mutagenesis in reverse genetics generally relies on well-established techniques and commonly used mutagens, such as ethylmethansulfonate (EMS) [ 5 , 17 ] and N -ethyl- N -nitrosourea (ENU) [ 18 ]. Those chemicals primarily induce single-nucleotide polymorphisms, which can most efficiently be detected by sequencing [ 19 ], by denaturing high-pressure liquid chromatography (DHPLC) [ 5 , 17 ], or by enzymatic cleavage of heteroduplex DNA with single-strand-specific endonucleases such as Cel-I [ 18 , 20 - 22 ]. Mismatch-cleavage analysis and DHPLC require special machinery and DHPLC is not very well suited for high-throughput analysis. Fast neutrons have also been used to introduce small DNA lesions, which can simply be resolved by agarose electrophoresis after PCR amplification [ 23 ]. This kind of mutagenesis may be limited to seeds or to labs in the vicinity of a reactor. We reasoned that it would be worthwhile to establish a generally applicable reverse genetic technique based on an unbiased and practicable random mutagenesis and an efficient mutation-detection performed on standard laboratory equipment. Here we introduce a novel mutagenesis protocol utilizing the cross-linking drug hexamethylphosphoramide (HMPA) [ 24 ], streamlined fly genetics and high-throughput fragment analysis on sequencers to demonstrate the feasibility of our reverse genetics approach. Results and discussion Fly genetics There are two ways to handle mutagenized progeny. Either large collections are established and maintained, which then are systematically and continuously screened for mutations of interest, or mutagenized progeny are screened directly and only animals exhibiting a desired trait are kept. The first method is in practice an F3 screen, which requires balancing of mutagenized chromosomes and maintenance of many stocks. This approach is far more labor-intensive than a simple F1 screen of progeny and thus is more suited to stock centers. Moreover, balancer chromosomes have many DNA sequence polymorphisms to wild-type chromosomes (our unpublished data), which will interfere with detection of mutagen-induced sequence polymorphisms. To circumvent the inherent problems with balancers, we devised an alternative genetic strategy, which had to fulfill the following criteria. First, mutagenized chromosomes have to be passed on in an unrecombined form such that mutations cannot be lost. Second, the mutagenized chromosomes should be brought into an isogenic background for mutation detection. Third, for economic reasons stock-keeping should be kept at an absolute minimum. We generated a fly strain (KNF306) isogenic to our yw wild-type laboratory strain but containing the same dominant marker on the two major autosomes. Both chromosome 2 and chromosome 3 are carrying white + marked P-element insertions, which were chosen because white + expression is restricted to different subregions of the eye (Figure 1a ). Chromosome 2 is marked by an insertion in the CG31666 locus, which results in white + expression only in the posterior part of the eye. Chromosome 3 harbors an insertion in the promoter of CG32111 , and this transgene causes dorsal white + expression. The combined expression patterns of both show a 'pie-slice' eye-color appearance (Figure 1a ). Thus, the same marker permits us to distinguish between linkage on chromosomes 2 or 3. Neither of the transgenes affects viability, and the line can be kept as a homozygous stock. Mutagenized chromosomes of strain KNF306 were passaged only via males, which were mated to the parental yw strain background. Thus, the marked autosomes remained unrecombined and could be unambiguously assigned because of the dominant character of the white + transgenes (Figure 1b ). Mutagen-fed F0 males were mass-mated and F1 males were mated in single crosses (see Materials and methods). After 4 to 5 days, nonsterile F1 males were recovered, pooled in groups of five, and their DNA extracted and analyzed. If a pool gave a positive signal, the crosses were traced back and F2 progeny carrying the mutant chromosome (as judged by the eye-color pattern) of each of the five crosses were individually re-tested. If this re-test was positive, a single F2 male of the respective cross was taken to establish a balanced stock. Non-positive crosses were discarded. Like any other genomic locus, the white + coding regions of both transgenes constitute targets for mutagenesis, and mutagenic events can be easily scored in the F1 progeny as a loss of the characteristic expression pattern. As discussed later, effectiveness of mutagenesis can be assessed from the occurrence of white - progeny, and as an internal control mutation rates at the two loci should be comparable. The crossing scheme and analysis procedure illustrated was optimized for autosomal genetics. We have generated another strain, KNF307, which in addition carries X chromosomes marked by a characteristic enhancer trap insertion at the omb locus (data not shown). However, analysis of X-chromosomal loci would require additional handling of F1 females or mutagenesis of F0 females and hence we did not carry out X-chromosomal screens. Mutagenesis EMS has been used as a deletion-inducing chemical in large-scale screens [ 25 ], but unbiased evaluation of its properties suggests that EMS-induced deletions are exceptional [ 26 ]. On the other hand, the deletions found by Liu et al . [ 25 ] ranged in size between 545 base-pairs (bp) and 1,902 bp and would not have been detected by Greene et al . [ 26 ]. The cross-linking carcinogen hexamethylphosphoramide (HMPA) has been shown to predominately induce deletions that were either in the range 2-315 bp or reached cytologically visible dimensions [ 24 ]. As our analysis method restricted the size of PCR fragments to about 800 bp, we chose HMPA as a mutagen, because EMS-induced deletions are likely to affect at least one of the primer-binding sites and would hence be undetectable. We modified the original HMPA mutagenesis protocol to administer a shorter, but more intense pulse of HMPA ([ 24 ], see also Materials and methods). A dose was applied that causes a similar rate of X-linked recessive lethals as standard EMS treatment, but only moderate male sterility (Table 1 and data not shown). We also did not add N , N -dimethylbenzylamine, which in our hands potentiated the sterilizing activity of HMPA. It has been reported that F1 progeny may exhibit mosaicism for mutagenized tissue [ 27 ]. Mosaic flies could generate a primary positive signal, but might not transmit the mutated gene. We have seen mosaicism at the white + loci and we have found positive F1 pools that did not yield mutant F2 progeny (Table 1 and data not shown). However, we were unable to determine whether some of the primary positives were due to mosaicism or to PCR artifacts. Mutation detection DNA from pools was prepared by a novel high-throughput extraction protocol allowing for up to 2,000 PCRs per pool (see Materials and methods). As HMPA is reported to induce deletions as small as 2 bp and as a mutated allele is diluted 10-fold as a result of our pooling of five flies, we decided to analyze PCR fragments on a sequencer offering maximal resolution and high sensitivity. We have also evaluated the 'poison-primer technique' which is reported to preferentially amplify alleles with a deletion at the poison-primer binding site from large pools [ 28 ]. However, the small deletion alleles we have tested did not outperform the amplification of the wild-type allele to the extent previously reported, implicating that the technique is more suited to large deletions and not generally applicable (data not shown). PCR products were analyzed on either a gel-based or a capillary sequencer (see Materials and methods). To increase efficiency of mutation detection on gels, we pooled up to three PCR products. These were labeled with different fluorescent tags, partly because they were of similar size (Figure 2a ). Screening The efficiency of HMPA mutagenesis could be assessed from the rate of white - mutations at the transgenes on chromosomes 2 and 3. Overall, we found 24 mutations in about 62,700 male and female flies. Two flies were mosaic for the mutations. Given that mosaicism can only be scored in eyes and there only in nonoverlapping expression domains, the mutation rates discussed below may be slightly underestimated (Table 1 ). Male sterility was 25.4 %. Aguirrezabalaga et al . [ 24 ] reported a mutation rate of 2.8 × 10 -4 at the vermilion locus scoring early and late progeny. The rate reached 3.7 × 10 -4 when only late progeny was regarded. After a few rounds of screening we have stopped screening early progeny (brood 1 flies, see Materials and methods), because we did not recover any white - mutation. As sperm development takes up to 10 days [ 27 ], we also consider it unlikely that brood 1 from our crossing scheme will yield appreciable efficiency. Disregarding brood 1, we obtained an average rate of 2.25 × 10 -4 mutations at the white + loci, which are about twice as large as the vermilion locus. Our mutagenesis procedure involving an overnight incubation with HMPA rather than a 3-day incubation with HMPA and N , N -dimethylbenzylamine is therefore not much less efficient than the original protocol. There was no difference in the frequency of induced white - mutations between brood 2 and brood 3 (Table 1 ). The small difference between mutation frequency on the identical mini-white genes located on chromosomes 2 and 3 may be attributed to statistical variance, to position effects, to different size of the enhancers driving white + expression or to systemic errors due to the smaller expression domain of the insertion on chromosome 3. The following additional parameters can be utilized to estimate mutant recovery. The white gene for which the mutation rate has been assessed encodes a protein of 688 amino acids from an open reading frame (ORF) of 2,064 bp. We assume that any deletion within the ORF would generate a null phenotype. Only 14 out of 31 HMPA-induced deletions selected at the vermilion locus would have been scoreable by our PCR approach, because the remaining 17 mutations were caused by large deletions affecting both primer-binding sites [ 24 ]. We designed PCR primers for each gene to be scored such that they encompass the first coding exon and the PCR products are between 450 and 807 bp in size. The average weighted length of our PCR fragments was 710 bp (including two primers of 20 nucleotides each). We thus expect one mutation in 30,317 flies (1/(2.25 × 10 -4 × 14/31 × (710 - 2 × 20)/2,064)) or one mutation in 6,063 pools, respectively. Taking into account the fact that two mosaic flies may not have transmitted (reducing the mutation rate to 2.0 × 10 -4 ), the estimate would be one mutation in 33,883 flies or one in 6,777 pools. We have scored 16,902 F1 males at two to 11 loci and recovered two transmitting mutations from about 20,900 analyzed PCR reactions (see Additional data file 2). According to the estimate we would have expected three. The first mutation detected was a 41-bp deletion in the first exon of CG15000 , which during the course of this study turned out to be the second exon of the dNAB locus (Figure 2c , and see [ 29 ]). The deletion causes a frameshift and very probably constitutes a null mutation. As shown in Figure 2a,b , the mutation was identified on a gel-based sequencer in a pool of PCR products labeled with the fluorophore NED (Applied Biosystems) and propagated in one of the five F2 crosses. The mutant chromosome is currently purified by separating the CG15000/dNAB allele - easily traceable by a restriction-fragment length polymorphism - from the white + marker (P. Geuking and K.B., unpublished work). Second, we detected a mutation in CG17367 on the capillary sequencer (Figure 3a,b ). The net 11-bp deletion (19-bp deletion, 8-bp insertion) is situated in the first intron and 5' to the start codon. The allele is viable over a deficiency uncovering the CG17367 locus. This study focused on implementing HMPA mutagenesis for reverse genetics. As discussed above, HMPA efficacy has been assessed from mutations at the white + loci, which have been selected on the basis of phenotype rather than sequence. Thus, our modified HMPA protocol may also prove to be valuable for forward genetic approaches. At the molecular level we could also identify deletions in the white + genes (data not shown), but we have not systematically investigated all of the white - mutations. Conclusions While the analysis of PCR fragment-length polymorphisms on our sequencers was very efficient, HMPA mutagenesis turned out to be the limiting parameter. It is about 28-fold less efficient than EMS mutagenesis when it is assumed that all HMPA hits are deleterious (3.2 × 10 -3 nucleotide substitutions at the 1 kb awd locus [ 5 ] for EMS compared to 2.25 × 10 -4 deletions per 2 kb white + locus for HMPA), but mutagen dose cannot be increased further because of the concomitant increase in male sterility. The new techniques that we have introduced increase the diversity of the toolkit available to laboratories interested in conducting reverse genetic screens. The pros and cons of the critical parameters are next considered individually. EMS or ENU versus HMPA as mutagen HMPA-induced deletions are very likely to induce null mutations when hitting an exon. EMS, on the other hand, primarily induces GC-to-AT transitions, but is not well suited for introducing small deletions. A considerable fraction of the transitions will not affect protein function. In Arabidopsis , about 44% of the mutations after EMS mutagenesis were silent, 51% were missense mutations and 5% were nonsense mutations [ 26 ]. Similarly, in a zebrafish ENU screen, only 15 out of 270 mutants (5.5%) were truncation mutants [ 18 , 19 ]. Recently, Guo et al . [ 30 ] determined the tolerance of a protein to random amino-acid changes and determined that about two thirds of amino-acid substitutions were neutral and only 34% were disruptive. Assuming that all truncation mutations are deleterious, it can be concluded that about 22% (34% of 51% plus 5%) of EMS-induced mutations negatively influence protein function. Of those amino-acid substitutions an unknown fraction will retain partial function. Thus, allelic series can be generated through EMS [ 22 ] and the generation of partial loss-of-function alleles may be a potential asset of EMS mutagenesis. Overall, HMPA is maximally sixfold less effective at inducing loss-of-function mutations (22% of 28%) than a high dose of EMS, but this disadvantage is compensated for by a more straightforward mutant analysis. Mutant analysis Mutant analysis depends critically on the mutagen and vice versa. Currently, the most effective way to screen for EMS-induced polymorphisms is the TILLING approach, which, however, requires a second round of PCR, specialized chemistry of the secondary primers, and an enzymatic reaction on the secondary product. TILLING cannot easily be performed on standard sequencers: we have tried to analyze Cel-I cleaved fluorescent heteroduplex DNA on an ABI 3730 sequencer, but did not obtain satisfactory sensitivity (data not shown). HMPA induced mutations can be detected by fragment-length analysis of primary PCR products on standard sequencers. Hence, screening for small deletions reduces PCR costs by a factor of 2 and spares the effort of secondary assays. Mutant handling Mutant handling is independent of the mutagenesis protocol and may be combined with either EMS or HMPA mutagenesis. For example, TILLING can be performed both on large mutant collections and on a continuous supply of freshly generated mutants. Finally, given the genotoxic properties of HMPA in both prokaryotes and higher eukaryotes [ 31 , 32 ], both the mutagenesis and the mutation-detection procedures described in this study may be directly transferred to other model organisms. Materials and methods HMPA mutagenesis About 150 1-3-day-old F0 KNF306 ( y , w ; CG31666-white + ; CG32111 - white + ) males were starved for 4 to 6 hours in a plastic bottle containing three layers of water-soaked LS14 filter papers (Schleicher & Schüll). A 1.1 ml sample of HMPA solution (5% sucrose, 0.1 M NaPO 4 , 25 mM HMPA, optional 0.05% bromophenol blue) was carefully applied to the filters using a syringe with a long needle (21G2) inserted through the foam stopper. The starved males were exposed to the HMPA solution overnight. Bromophenol blue does not affect mutagenicity detectably, but stains the guts of the flies blue and thus enables mutagen uptake to be monitored and controlled. Freshly eclosed flies do not ingest enough mutagen. HMPA-contaminated plasticware must be disposed of by thermal waste treatment. Fly work and crossing procedure In six bottles containing standard corn medium, each 25 mutagenized KNF306 F0 males (Figure 1a ) were allowed to mate to 15 to 20 virgin yw females (brood 1). After 2 days males were taken out and crossed to yw virgins in new bottles (brood 2A) and this cross was transferred after 3 days (brood 2B). After another 2 days F0 males were recovered and mated to fresh yw virgins (brood 3A). F1 males of broods 2A, 2B and 3A were collected and mated individually to three yw virgins in about 650 separate crosses per week. Five hundred non-sterile males were removed after 4 to 5 days and five males were pooled for DNA extraction. Fertilized females were returned, and unsuccessful crosses were discarded. If analysis of PCR fragments indicated a primary positive pool, crosses were traced back and kept for further analysis; the other crosses were discarded. From each of the five crosses of primary positive pools a single F2 male or female containing the chromosome of interest as manifested by the typical eye-color pattern was collected for DNA extraction. If PCR analysis yielded a secondary positive result in one of the five F2 flies, a single F2 male containing the chromosome of interest was taken out from the respective cross for balancing (Figure 1b ). The whole crossing scheme requires 6 weeks and was organized such that a mutagenesis was performed every second week (see Additional data file 1). Large-scale DNA extraction, PCR and fragment analysis DNA was extracted in bulk by squishing pools of each five flies through mechanic force in a vibration mill (Retsch MM30) programmed to shake for 20 sec at 20 strokes per second. Flies were placed into wells of a 96-well deep-well plate. Each well was then filled with 500 μl squishing buffer (10 mM Tris-Cl pH 8.2, 1 mM EDTA, 0.2% Triton X-100, 25 mM NaCl, 200 μg/ml freshly added proteinase K) and one tungsten carbide bead (Qiagen). The deep-well plate was then sealed with a rubber mat (Eppendorf) and clamped into the vibration mill. (Tungsten carbide beads can be recycled: after an overnight incubation in 0.1 M HCl and thorough washing in double-distilled water (ddH 2 O) the beads were virtually free of contaminating DNA.) Debris was allowed to settle for about 5 min and each 50 to 100 μl of supernatant were transferred into a 96-well PCR plate. The reactions were incubated in a thermocycler for 30 min at 37°C, and finally for 5 min at 95°C to heat-inactivate proteinase K. A Tecan pipeting robot was used for PCR setup. To 5 μl of template DNA, master-mix was added and PCR was performed on an MJR thermocycler that was integrated into the robot. The master-mix per reaction was composed of 20.48 μl ddH 2 O, 0.6 μl MgCl 2 (25 mM), 0.6 μl dNTPs (10 mM), 0.1 μl fluorescently labeled primer 1 (100 μM), 0.1 μl primer 2 (100 μM), 0.12 μl hot-start Taq polymerase (HotStar, Qiagen, 5 U/μl), 3 μl 10× buffer containing MgCl 2 (Qiagen). Cycling conditions were 95°C 15 min, 35 × (95°C 20 sec, 60°C 30 sec, 72°C 1 min), 72°C 2 min, 4°C. Three differently labeled PCR reactions (oligos were 5' labeled with Applied Biosystems' fluorophors FAM, NED and VIC, respectively) were then pooled. To facilitate sizing of fragments we also added ROX1000 size marker (Applied Biosystems) to five DNA pools. Samples of 1.5 μl pooled DNA were mixed with 1.5 μl loading buffer (consisting of one part 25 mM EDTA pH 8.0 with 50 mg/ml blue dextran and five parts HiDi formamide (Applied Biosystems)). The reactions were incubated for 3 min at 95°C, cooled down, and 1.5 μl each were loaded onto a 96-lane ABI 377 sequencer. Run conditions were as follows: 1 h pre-run at 1,000 V, 35 mA, 51°C and 10 h run at 2,400 V, 50 mA, 51°C. Gel images recorded at four different color channels by the GeneScan software were analyzed visually. Slight modifications to this protocol were introduced for analysis performed on an ABI 3730 capillary sequencer. First, DNA was diluted 20-fold before PCR. Second, after PCR, reactions were diluted 100-fold and 2 μl of diluted PCR products were added to each 15 μl HiDi formamide (Applied Biosystems). PCR product was diluted on a Tecan pipeting robot. Diluted DNA was denatured for 2 min at 95°C before analysis. Sample injection (10 sec) and analysis (12,000 scans) was done according to standard protocols. Identification of deletion fragments was then performed by visual inspection of gel-images generated by the Data Collection Software (Array Viewer option, Applied Biosystems). No internal size standard was used, as deletion fragments were identified relative to wild-type PCR product. Additional data files The following additional files are available with the online version of this paper. Additional data file 1 contains the time schedule of mutagenesis, fly work, and screening. The whole procedure takes six weeks and is organized such that one mutagenesis has to be performed every second week to generate a continuous supply of mutagenized progeny. Additional data file 2 contains information on the 10 other genes scored. Gene names, fluorescent labels, fragment lengths and the number of analyzed F1 flies are given. Labeled primers were ordered from Applied Biosystems. Primer sequences are available upon request. Supplementary Material Additional data file 1 The time schedule of mutagenesis, fly work, and screening Click here for additional data file Additional data file 2 Information on the 10 other genes scored Click here for additional data file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC545603.xml

549039

Postpartum maternal morbidity requiring hospital admission in Lusaka, Zambia – a descriptive study

Background Information on the extent of postpartum maternal morbidity in developing countries is extremely limited. In many settings, data from hospital-based studies is hard to interpret because of the small proportion of women that have access to medical care. However, in those areas with good uptake of health care, the measurement of the type and incidence of complications severe enough to require hospitalisation may provide useful baseline information on the acute and severe morbidity that women experience in the early weeks following childbirth. An analysis of health services data from Lusaka, Zambia, is presented. Methods Six-month retrospective review of hospital registers and 4-week cross-sectional study with prospective identification of postpartum admissions. Results Both parts of the study identified puerperal sepsis and malaria as, respectively, the leading direct and indirect causes of postpartum morbidity requiring hospital admission. Puerperal sepsis accounted for 34.8% of 365 postpartum admissions in the 6-month period. Malaria and pneumonia together accounted for one-fifth of all postpartum admissions (14.5% & 6% respectively). At least 1.7% of the postpartum population in Lusaka will require hospital-level care for a maternal morbidity. Conclusions In developing country urban settings with high public health care usage, meticulous review of hospital registers can provide baseline information on the burden of moderate-to-severe postpartum morbidity.

Background Maternal morbidity refers to complications that have arisen during the pregnancy, delivery or postpartum period. Every year an estimated 50 million women are affected by maternal morbidity. Defining, interpreting and measuring maternal morbidity, however, is recognised to be difficult and the prevalence of such morbidity (both general and specific) has been poorly described [ 1 , 2 ]. Over the past decade, the nature and extent of postpartum maternal morbidity has received increasing interest in both developed and developing countries with a range of research methods of varying sophistication being used to identify long and short-term and acute and chronic morbidity following childbirth [ 1 , 3 - 8 ] The WHO (1998) [ 1 ] defines the postpartum period, or puerperium, as beginning one hour after the delivery of the placenta and continuing until 6 weeks (42 days) after the birth of the infant. As the woman recovers from labour, adapts to her new role and reverts physically to her non-pregnant state, it is a special but critical time for both the mother and her infant [ 9 ]. Many of the complications leading to postpartum maternal morbidity arise during labour and delivery and in the first 1–2 weeks following delivery; for at least 18 million women these morbidities become long-term and are often debilitating [ 1 ]. Major acute obstetric morbidities include haemorrhage, sepsis and pregnancy-related hypertension. Longer-term morbidities include uterine prolapse, vesicovaginal fistulae (VVF), incontinence, dyspareunia and infertility [ 10 ]. Fortney & Smith [ 2 ] have described 6 dimensions to maternal morbidity: aetiology, severity, duration, time of onset, accumulation and sequelae. However, in many developing countries health services data on postpartum morbidity remains extremely limited. In many settings, data from hospital-based studies is hard to interpret because of the small proportion of women that have access to supervised deliveries and medical care. But, as Fortney and Smith suggest, in areas with good access to and uptake of health care the measurement of the type and incidence of postpartum complications severe enough to require hospitalisation may provide useful baseline information on the acute and severe direct morbidity women experience in the early weeks following childbirth. Zambia's capital city, Lusaka, offers such a context. This paper describes the application of a pragmatic and inexpensive approach to the baseline assessment of the burden of moderate-to-severe morbidity in the postpartum period in this urban African setting, with a view to monitoring trends and identifying preventable risk factors. Methods A network of 23 public sector clinics and a single referral hospital, the University Teaching Hospital (UTH), comprise the public health care system for Lusaka's population of approximately 1.5 million. There is also a small private sector (2% of deliveries) [ 11 ] and there are traditional practitioners. An estimated 10.5% of deliveries occur at home. All public clinics provide antenatal care and a postnatal care service at 6 weeks after delivery. In addition, 10 of the clinics provide 24-hour care for labour and delivery and a 1-week postnatal care service. In Lusaka urban district, a recent systematically sampled community survey suggests that there is a relatively high coverage of antenatal, delivery and postnatal services (Table 1 ). Women interviewees also reported good access to medical treatment for serious problems during pregnancy and in the first month postpartum. Hospital admission data, if used as a proxy for moderate-to-severe morbidity, may therefore be considered pragmatically representative of the population health needs in this setting. Table 1 Uptake of maternal health care services, Lusaka Urban community survey data [11] Uptake of maternal health care services % of women reporting (n = 946) % of women with 5 or more antenatal check-ups 73% Proportion of deliveries with a professional attendant 89.5% % of women attending a postnatal check-up in the 6 weeks postpartum 84% % of women reporting a "serious problem" in the antenatal or postpartum period who said that they had been able to get medical attention "as soon as they felt they had needed it" 85% At the University Teaching Hospital, which acts as the district referral hospital for the city, admission and discharge registers are kept in each ward and department. In an earlier retrospective study, analysing referrals for all pregnancy-related complications, undertaken at UTH [ 12 , 13 ], 4% of these cases were identified as referrals in the postpartum period (95/2,892 over a two- month study period). In this study, the hospital registers were used to identify all cases of postpartum morbidity presenting to UTH, and from these to estimate the incidence of, and identify the nature of postpartum morbidity severe enough to require admission for hospital-level treatment. Data collection was carried out by LV between July and September 2000. Ethical clearance for the study was obtained from the University of Zambia Research Ethics Committee. For the purpose of this study, the WHO definition of the postpartum period (from delivery until 6 weeks after delivery) was used as the time period inclusion criteria [ 1 ]. All women identified as having been admitted to UTH for in-patient treatment for morbidity during this period were included for the purpose of the review, whether or not their morbidity was explicitly "obstetric" in origin. Women who were admitted to hospital to accompany and nurse their babies that had neonatal problems were excluded. Retrospective data collection Relevant admission and discharge registers at UTH were reviewed for the six-month period July-December 1999. Dependent on the type, timing and severity of a postpartum problem, women may be referred or may self-refer to one of three different units within the hospital: (i) Women with early postpartum complications, defined as problems occurring within 24 hours of delivery, are admitted through the labour ward admission room; (ii) Women with postpartum problems occurring more than 24 hours after delivery are referred or may present themselves to the gynaecology filter clinic, from where they may then be referred on to the emergency admission ward; (iii) Cases of breast abscess are generally admitted though the surgical unit. From any of the units women may be then admitted onto a longer stay gynaecology ward for further management and treatment. Women admitted to hospital irrespective of the length of stay were included in the data capture. Table 2 outlines the identification process and inclusion criteria that were used in each of the wards. The figures are likely to be an underestimate of the total number of postpartum admissions because as is often found in studies of this nature, diagnosis was frequently poorly recorded in the registers. Only women who could be positively identified as postpartum morbidity admissions were included in the final analysis. Any women admitted to medical wards with non-obstetric conditions – such as malaria – in the later part of the puerperium would have been missed. Using admissions registers to identify postpartum cases also excludes any women who were admitted to hospital prior to the puerperium (for example antenatally, or in labour) and subsequently developed postpartum problems requiring prolonged in patient care. Table 2 Identification of postpartum cases and inclusion criteria Emergency Admission Ward Gynaecology Wards Labour Ward Surgical Unit Source of identification of cases Admission Register Discharge Register Admission Register Discharge Register Inclusion criteria to be defined as postpartum admissions All cases recorded as postpartum. All cases with infected caesarean section wound. All cases with infected episiotomy/perineal tears. All cases recorded as postpartum. All cases with infected caesarean section wound. All cases with infected episiotomy/perineal tears. Any non-obstetric conditions e.g. malaria, PTB, pneumonia, meningitis. All cases recorded as postpartum. All cases of breast abscess/mastitis Cases were crosschecked by name and age between the emergency admission and gynaecology wards and between the short and longer stay surgical wards, to prevent double counting. Cases of breast abscess were also crosschecked between the surgical unit and the gynaecology department to ensure cases had not been referred. Data was entered on Epi. Inf. 6.04 for analysis. Routine health service statistics for the same six-month study period (July-December 1999) were also collated. There were 19,691 deliveries within UTH and the satellite clinics (5,511 and 14,180 deliveries respectively), of which 1,021 were by caesarean section (an average of 39 per week). Women delivering either at UTH or the clinics were instructed to attend the local satellite clinic for postnatal follow-up visits at week 1 after delivery and again at 6 weeks. Some women were seen at UTH for a postnatal visit at week 1 to follow-up on some complication while those who had a caesarean section were reviewed at 6 weeks. Recorded postnatal check-up at the clinics at 1-week was 40% and 21% at 6 weeks; however, this data was incomplete for some clinics. At UTH, an average of 42 women per week attend the postnatal clinic – most of them after a caesarean section. During the same six-month period, there were 93 maternal deaths at UTH. Of these maternal deaths 8 (8.5%) were known to be due to puerperal sepsis. A much larger number of the women who died had sepsis and stigmata of AIDS (Y Ahmed, personal communication). Prospectively identified cases Due to the limitations of the routine data sources used in the retrospective review, a small cross sectional study was also conducted using prospective identification of cases in order to verify the findings. Over a 4-week period, from 14 August to 10 September 2000, all early postpartum admissions to the maternity unit at UTH were identified through the labour ward admission register. Obstetric case notes were sought and reviewed. Late postpartum cases (>24 hours and up to 6 weeks after delivery) for the same time-period were also identified and recorded by the same means as described for the retrospective study. On the gynaecology ward, cases were identified through daily review of the ward round books, to identify new admissions, and through consultation with the senior ward sister and the ward clerk. Results Retrospective data After crosschecking of data to prevent double counting, 365 maternal postpartum admissions to the hospital were positively identified for the 6-month study period July-December 1999. Cases of retained placenta (n = 55), removed in theatre, were not included in the final analysis as it was not possible to differentiate between clinic referrals and UTH deliveries. Referral source and admission status Of the 365 admissions, 236 (65%) cases were identified through the emergency admission ward, 120 were identified on the gynaecology ward and the remaining 9 cases were identified on the surgical wards. More than half of the emergency admission ward cases were referred from the satellite MCH clinics (Table 3 ). Two-thirds of all cases were admitted to either the gynaecology transit ward or one of the longer gynaecology wards (97 [41%] and 61 [26%] respectively). A quarter of women were discharged home later the same day after a period of treatment or observation. Six women (2.5%) were admitted to either the special observation unit or medical intensive care unit; five women (2%) were referred to the general adult "filter clinic". Table 3 Sources of postpartum referrals to the emergency admission ward, University Teaching Hospital Referral source Number (%) (n = 236) Satellite clinics 135 (57%) Self-referral 28 (12%) Out-patient department within hospital 12 (5%) Not documented 61 (26%) Age range The age range for all cases was 15–48 years. Of all postpartum admissions (365 women), more than half were aged between 20–29 years, a reflection of the fact that the largest number of births takes place within this age group. Nature of morbidities requiring hospital admission There were 39 different recorded diagnoses across the 365 cases (Table 4 ). Puerperal sepsis was the most frequent diagnosis, accounting for one-third (34.8%) of all postpartum hospital admissions over the 6-month period. Infection of the reproductive tract, including infected tears and episiotomy, infected caesarean section wound and puerperal sepsis accounted for 47% of all admissions (170/365). Table 4 Identified postpartum admission by diagnosis, University Teaching Hospital, July–December 1999 Diagnosis Cases (n = 365) Percent of all admissions Puerperal sepsis 127 34.8 Malaria 53 14.5 Infected tears/episiotomy 26 7.1 Hypertension 24 6.6 Pneumonia 22 6.0 Infected caesarean section 17 4.7 Anaemia 11 3.0 Breast Abscess 10 2.7 Symphisiotomy 9 2.5 Eclampsia 9 2.5 Puerperal Psychosis 7 1.9 "After pains" 6 1.6 Pulmonary Tuberculosis 5 1.4 Pyrexia (not linked to diagnosis) 5 1.4 Postpartum Haemorrhage 4 1.1 Pre-eclampsia 3 0.8 Retained products of conception 3 0.8 Urinary tract infection 3 0.8 Other (gastroenteritis, meningitis, measles etc) 21 5.8 Malaria was the second most common diagnosis accounting for 14.5% of all cases. Hypertensive disorders including hypertension, pre-eclampsia and eclampsia accounted for 10.9%. Estimating the population burden of moderate-to-severe postpartum morbidity Using MacKeith et al.'s [ 14 ] estimates for public sector coverage of deliveries (87.5%) and the UTH and clinic figures for the period (19,691 deliveries), we estimate that there were approximately 22,000 births in Lusaka during the 6-month period July-December 1999, and 365 hospital admissions for postpartum morbidity were positively identified for the same period. If hospital admission can be taken as a practical proxy measure for the moderate-to-severe end of morbidities, then the burden of moderate-to-severe postpartum morbidity in this urban African population may therefore be estimated to be at least 1.7% (365/22,000 births). Because this incidence refers only to women admitted live to hospital in the postpartum period, the actual incidence will be somewhat higher, with the addition of postpartum morbidities that occurred in women who were already hospital in-patients, any maternal deaths that occurred outside of the hospital, and any missed late postpartum admissions to medical wards. Cross-sectional study of prospectively identified cases Over the one-month period August-September 2000, 64 admissions to the hospital for postpartum maternal morbidity were identified. The routine hospital procedures differentiate between "early" (up to 24 hours postpartum) and "late" (after 24 hours) postpartum admissions. The former are admitted through the labour ward admissions and the latter through the emergency gynaecology ward. Twenty women (31%) were "early" referrals and 44 (69%) were "late". This categorisation of the data therefore concerns time elapsed between delivery and admission to hospital, and not between time of delivery and onset of the condition. Early-postpartum referrals Just over one-third (7/20) of the maternal referrals in the first 24 hours after delivery were for retained placenta and just over one-third were for pregnancy- related hypertension or eclampsia (7/20). Three referrals were for postpartum haemorrhage. There were no maternal deaths in this group. Late-postpartum referrals Referrals from the satellite clinics accounted for 73% of all "late" (>24 hours after delivery) postpartum admissions to UTH; the remaining cases were mainly self-referral. The diagnoses of the 44 cases of late-postpartum admissions in the one-month period are shown in Table 5 . Table 5 Late postpartum referrals (Cross-sectional study of prospectively identified cases) Diagnosis Number (%) (N = 44) Puerperal sepsis 5 (11%) Malaria 5 (11%) Pregnancy-related hypertension 5 (11%) Infected tears/episiotomy 4 (9%) Infected caesarean section 3 (7%) Symphisiotomy 3 (7%) Puerperal psychosis 2 (4.5%) Secondary postpartum haemorrhage 2 (4.5%) Meningitis 2 (4.5%) Retained products 1 (2%) Eclampsia 1 (2%) Breast abscess 1 (2%) Other 10 (23%) Twenty-one cases (47%) were admitted to the longer stay gynaecology ward. More than half of all cases (52%) were in the 20–29 years age group. Among the 44 women admitted to hospital at more than 24 hours postpartum, 4 died; 2 of these deaths occurred on the longer stay ward. Of the 4 deaths, one could be directly linked to obstetric events (anaemia); the others were related to non-obstetric causes (pneumonia, cryptococcal meningitis, and encephalitis). Discussion This hospital study used time-period based inclusion criteria in its identification of cases of maternal morbidity sufficiently severe to require hospital admission. The majority of morbidity thus identified was directly linked to obstetric causes e.g. puerperal sepsis, infected wounds, and pregnancy-induced hypertension. However, non-obstetric conditions, including malaria and pneumonia were found to have accounted for at least one-fifth of all postpartum admissions in the retrospective review. This mirrors the increased role that indirect causes have been found to be playing in maternal mortality rates in countries such as Zambia [ 14 ]. Data accuracy We have already outlined some of the practical difficulties with using routine hospital data sources such as admission registers. Admission rates estimated from the small prospective part of the study do not suggest, however, that many cases were lost in the identification process in the larger retrospective part. The former identified an average of 11 late-postpartum maternal admissions to hospital per week in the month observed, and the latter, an average of 14 late-postpartum maternal admissions per week over the six months reviewed. In both parts of the study, puerperal sepsis and malaria were identified as leading causes of postpartum morbidity requiring hospital admission. In the prospective identification of postpartum morbidity requiring hospital admission, puerperal sepsis, malaria and hypertensive disease each accounted for the same number of admission (5 in each) but the numbers are too small to draw any conclusions. Some inaccuracy in the classification of certain morbidities may be expected due to the use of admission diagnoses. It is also a limitation in the design of this element of our study that data collection did not extend to the medical wards. It is therefore not possible for us to ascertain whether, or to what extent, there are medical ward admissions of women in the late puerperium with non-obstetric conditions such as malaria. Puerperal sepsis Puerperal sepsis was the leading cause of postpartum hospital admissions in this population, accounting for 34.8% of all identified postpartum cases in the retrospective part of this study. Other hospital-based studies as well as surveys of women's self-reports of postpartum morbidity report puerperal sepsis as a leading cause of postpartum morbidity in developing countries [ 5 , 6 , 15 - 18 ]. Puerperal sepsis cases identified through the retrospective data collection part of the study accounted for more than twice as many cases as the second commonest postpartum morbidity requiring hospital admission, malaria. For the retrospective study the overall rate of puerperal sepsis cases requiring hospital-level care and admission was 0.64% of all supervised deliveries in the public sector services, a rate that falls in between those estimates from earlier hospital-based studies in Niger: 0.22% [ 5 ]; and in Nigeria: 1.7% [ 15 ]. However, it should again be remembered that in this study, women who delivered in hospital and developed septic complications before discharge would be excluded from this figure. The overall figure can therefore be expected to be higher. Other obstetric postpartum morbidity A number of other obstetric-related postpartum morbidities including anaemia, breast abscess, symphisiotomy, puerperal psychosis, after-pains, urinary tract infection, secondary postpartum haemorrhage and retained products of conception were also identified as late-postpartum referrals through the retrospective study. Anaemia in the postpartum period is not an uncommon health problem [ 1 ]. Surveys of women's self-reported morbidity frequently cite symptoms in the postpartum period that could be suggestive of or lead to anaemia, including chronic fatigue [ 19 ] and excessive bleeding [ 15 , 20 , 21 ]. Non-obstetric postpartum morbidity Malaria and pneumonia together accounted for one-fifth of all the postpartum hospital admissions that we identified. This finding suggests the usefulness of an approach that employs a "time-period" definition to identify cases rather than simply a set of purely obstetrically-related diagnostic categories. In all, 14.5% of identified postpartum maternal admissions to hospital were due to malaria. While it is widely recognised that the severity and frequency of malaria is greater in pregnant, compared to non-pregnant women, until recently it was generally thought that the importance of pregnancy-related malaria ends with delivery [ 22 ] and malaria is rarely mentioned as an important postpartum morbidity in the obstetric literature. Diagne et al.'s study from Senegal [ 22 ], however, was one of the first to suggest that the increased susceptibility to malaria in pregnancy persists up to 60 days after delivery. They found that compared to the non-pregnant state, the incidence of episodes of malaria increased significantly during the second and third trimesters of pregnancy and reached a maximum during the first 60 days after delivery. A number of factors may modify susceptibility to malaria in the postpartum period. The age at which partial immunity to malaria is acquired is critically dependent on transmission intensity [ 23 , 24 ]. Wide variations are seen in levels of immunity to malaria among Zambian women secondary to geographical and other factors affecting transmission [ 23 ]. Many of the malaria cases identified in the study may be the result of recrudescence rather than new infection particularly because the study took place during the transition between dry and wet seasons and study participants were primarily from urban and peri-urban communities. Susceptibility may also be dependent on haematological and nutritional factors as well as HIV status [ 25 ]. The contribution of HIV/AIDS to maternal postpartum morbidity and mortality Pneumonia and Pulmonary TB were important causes of postnatal morbidity in this study and were likely related to HIV/AIDS. The HIV serostatus was rarely available of the index postpartum cases, though unlinked anonymous testing of HIV in the antenatal population in 4 sentinel sites in Lusaka during 1998 showed a high HIV prevalence of 27.4% [ 26 ]. HIV positive women are more prone to postpartum infections including urinary tract infections, chest, episiotomy and caesarean section wound infections [ 27 , 28 ]. Furthermore, in this study, causes of postpartum morbidity included puerperal psychosis, cerebral malaria and HIV related cerebral complications – all of which can be difficult to diagnose with certainty in a malaria and HIV endemic area [ 29 ]. Of the 93 cases of maternal mortality during the six-month retrospective study period in 1999, almost a third of the cases were attributed to a presumptive diagnosis of HIV/AIDS and had no other direct or indirect cause of maternal mortality (personal communication, Y Ahmed, 2003). Hospital admissions in the first 24 hours following delivery Review of registers on the labour ward, for both the retrospective and prospective aspects of this study suggest that the majority of referrals in the early-postnatal period (first 24 hours) were for infant rather than maternal indications. Of the early postpartum referrals for maternal indications, retained placenta is the leading reason for referral from clinics to the hospital. This reflects the urban context of the study and the relative ease of transportation that permits a district policy of removal of placenta at hospital level rather than by the practitioner with essential obstetric care skills at the delivery clinic. Conclusion The high public sector maternity care usage in this community permits the low-cost review of routine data to be reasonably meaningful. The caveats are those associated with extraction of data from health facility admission registers, which are not always complete, and which cannot take account of changes in diagnosis or subsequently arising complications. In the absence of more robust data, such reviews, if carried out meticulously, do offer the opportunity to identify the extent of moderate-to-severe postpartum morbidity and the principle causes. They thus may provide the groundwork for detailed condition-specific research to take place exploring aetiology, duration, time of onset and outcome, and the implications for health care provision. Competing interests The author(s) declare that they have no competing interests. Authors' contributions LV: Conceived of the study, conducted the field data collection and data analysis and contributed to the paper. SFM: Wrote the first draft of the paper, participated in the study design and advised on the data collection and analysis. YA: Contributed to the study design, advised on the data collection and analysis and contributed to the paper. All authors read and approved the final manuscript. The views expressed in this article are those of the authors and not of their institutions. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549039.xml

539253

The family as a determinant of stunting in children living in conditions of extreme poverty: a case-control study

Background Malnutrition in children can be a consequence of unfavourable socioeconomic conditions. However, some families maintain adequate nutritional status in their children despite living in poverty. The aim of this study was to ascertain whether family-related factors are determinants of stunting in young Mexican children living in extreme poverty, and whether these factors differ between rural or urban contexts. Methods A case-control study was conducted in one rural and one urban extreme poverty level areas in Mexico. Cases comprised stunted children aged between 6 and 23 months. Controls were well-nourished children. Independent variables were defined in five dimensions: family characteristics; family income; household allocation of resources and family organisation; social networks; and child health care. Information was collected from 108 cases and 139 controls in the rural area and from 198 cases and 211 controls in the urban area. Statistical analysis was carried out separately for each area; unconditional multiple logistic regression analyses were performed to obtain the best explanatory model for stunting. Results In the rural area, a greater risk of stunting was associated with father's occupation as farmer and the presence of family networks for child care. The greatest protective effect was found in children cared for exclusively by their mothers. In the urban area, risk factors for stunting were father with unstable job, presence of small social networks, low rate of attendance to the Well Child Program activities, breast-feeding longer than six months, and two variables within the family characteristics dimension (longer duration of parents' union and migration from rural to urban area). Conclusions This study suggests the influence of the family on the nutritional status of children under two years of age living in extreme poverty areas. Factors associated with stunting were different in rural and urban communities. Therefore, developing and implementing health programs to tackle malnutrition should take into account such differences that are consequence of the social, economic, and cultural contexts in which the family lives.

Background Despite improvements in the health of children under five years of age, malnutrition remains as an important public health problem in Mexico [ 1 ] and in other developing countries [ 2 - 7 ]. Malnutrition in young children affects linear and brain growth and intelligence quotient, and is synergistically associated with child morbidity and mortality [ 8 - 13 ]. Previous reports have stated that malnutrition occurs mainly in rural areas and worsens under conditions of extreme poverty [ 4 , 14 , 15 ]. In Mexico, malnutrition is more frequent in the southern states, which are underdeveloped and have a predominantly indigenous population living in poor housing with unsanitary conditions [ 16 ]. The most recent nutritional survey at national level showed that the prevalence of stunting decreased during the last decade from 23% to less than 18%. Stunting is defined as the proportion of children under five years old whose height-for-age is less than -2 standard deviations of the reference population median. Wasting (the proportion of children under five years old whose weight-for-height is less than -2 standard deviations of the reference population median) has declined from 6% to 2%. However, analysis shows that prevalence of stunting is higher in rural (31.7%) than in urban strata (11.6%) [ 17 ]. The extent of malnutrition in extremely impoverished rural areas has fuelled implementation of public health programs aimed at improving children's nutritional status [ 18 ], but the impact of these programs has not been completely evaluated. Poverty is not confined to rural areas, nor is malnutrition; they are also present in urban environments [ 19 ]. Uncontrolled and unplanned growth of cities has fostered the emergence of urban slums that lack basic sanitary services. In such settings, underprivileged families live overcrowded and malnutrition is undoubtedly present. Unfortunately, specific information about nutritional status of children living in urban slums cannot be obtained from existing epidemiological data. The extent of child malnutrition in these areas is probably underestimated because of a flawed health information system. Malnutrition is an outcome of various factors resulting from unfavourable socioeconomic circumstances such as difficulties in obtaining food, unemployment -which determines an irregular income for the family's breadwinner-, limited access to education and health services, or illness caused by unsanitary conditions [ 2 , 5 , 14 , 20 - 24 ]. These circumstances are worsened by unequal access to, and distribution of resources among members of the family. However, some families are able to cope with such adverse environments and to maintain their children in an adequate nutritional status. The purpose of this study was to ascertain whether family-related variables constitute risk factors for stunting among young Mexican children living in extreme poverty, and whether these factors differ between rural and urban settings. Methods Study design A case-control study was conducted from August through December 1998. Two extreme poverty level areas (one urban and one rural) were selected. Each area was analysed separately. Setting The study took place in the south-western State of Guerrero, one of the poorest in Mexico, with a population of nearly three million. The rural area, named Alto Balsas, is located next to the Balsas River; the urban area, named El Sinai, is a poor neighbourhood on the outskirts of the port of Acapulco, which has half a million inhabitants. For the rural area, four villages with a combined population of 4,638 were included in the study. The closest city is approximately 45 miles away. The principal language spoken in these villages is Náhuatl, and agriculture is the main economic activity. Because of seasonal conditions there is a high rate of cyclic migration each year. This migration occurs during harvest season, where farmers migrate from the State of Guerrero to another states. There is one medical doctor and one primary care centre per approximately 2300 people. El Sinai is located in the outskirts of Acapulco and has 6,860 inhabitants, most people speak Spanish and they work in the local industry as unskilled manual workers. In this area, there is one medical doctor and one clinic per 3400 people. In both study areas, sanitation and public services are deficient. Table 1 shows the characteristics of both study areas. Table 1 Characteristics of study areas Alto Balsas a El Sinai Rural area Urban area Population 4,638 6,860 Distance to nearest urban centre 45–60 km ----- Number of schools Kindergarten 4 2 Elementary school 4 2 Secondary school 0 1 Health Facilities (belonging to the Ministry of Health) Rural clinic (2) Health post (2) Urban clinic (1) Health Personnel (employed by the Ministry of Health) Physicians 2 2 Nurses 3 2 Primary care technicians 2 2 Indigenous language 40–70% 4% b Economic activities Agriculture 75–90% 0 Industry 5–20% 66% Services 5–25% 12% Annual migration c 30–50% 10% a Four communities b Estimations based on data from this study c Proportion of families who migrate seasonally every year Study population The units of study were children between 6 and 23 months of age. Children older than 6 months were included on the assumption that this is the minimum amount of time the child is exposed to family-related factors; also, this is the average age in which children are weaning (this is defined as the time when mothers begin to introduce food other than milk into the child's diet). Thus infants become more exposed to environmental causes of under-nutrition. The family was defined as next of kin, or persons sharing the same household and food expenses [ 25 ]. Only one child from each family was included in the study; if there was more than one eligible child, the oldest was selected. Infants with congenital diseases or those with low birth weights (less than 2,500 g) were excluded. Case definition Cases were stunted children. This was ascertained by using the height-for-age indicator [ 26 ]. The criterion for stunting was: Z-value less than -2.00 standard deviations (SDs) below the median height-for-age [ 27 ]. Control definition Controls were children without stunting: equal to or above -2.00 SDs below the median in families that had no stunted children. The sample size was calculated by using the case-control study formula in accordance with the following assumptions [ 28 ]: α = 0.05, β = 0.20, minimum risk to be detected = 2.5, proportion of controls exposed to the least frequent variable (migration) = 0.15, and case-control ratio 1:1. The required sample size was 112 children per group in each area. Study variables A five-dimension framework to identify family-related factors that might have influence on the child's nutritional status was constructed. Each dimension comprises several variables selected to build up a comprehensive scenario. The dimensions are the following: 1. Family characteristics: parents' age and literacy, type and duration of parents' union; family structure (nuclear or extended); presence of both parents (complete or incomplete family); number of members of the family and mother's use of contraceptives. 2. Family income: parents' employment (type of job, time on the same job, per capita family income), unemployment and migration during the past two years. 3. Household allocation of resources and family organisation: time spend by the mother to care for the child and to do domestic activities, and the way the family distributed and shared its income (percentage of income spent in food, clothes, transportation, rent, etc); 4. Social networks: type of networks (within or outside the family), size of networks, frequency of interactions and type of support (economic, child care, etc.). 5. Child health care: patterns of breast-feeding and health-care-seeking for preventive care (immunisations and Well Child Program visits) or curative care. In addition, the following variables were included: child characteristics (age, sex, birth order and birth weight); and housing characteristics (type of construction, crowding conditions, indoor plumbing, sewerage system, and whether the kitchen was in a separate room). Data collection In each study area, a local health worker able to communicate in Náhuatl and Spanish was trained to carry out a census to identify children fulfilling inclusion criteria, to interview the mother and to obtain the anthropometric measurements. Data were collected by using a pro-forma. The mother or caregiver was personally interviewed. During the visit, the interviewer measured and recorded the height and weight of each child aged between 6 and 23 months. The weight was measured using a digital scale with a precision error of ± 1 oz. The mother helped to measure the recumbent length of her child and this was done by using a portable calibrated board. To assure accuracy and reliability, one of the researchers (HR or RC) visited 10% of the households within the following week to confirm the data. There were no inconsistencies in data or anthropometric measures that could affect the results. Mexican Institute of Social Security IRB and Ministry of Health authorities of the State of Guerrero approved the study. Local community leaders accepted and collaborated in the study and each family head as well as child's mother gave their informed consent. Statistical analysis Analysis was carried out separately for each area. Firstly, a bivariate analysis was run; crude odds ratios (OR) and 95% confidence intervals (95%CI) were calculated for each variable in every dimension. The analysis included estimates of correlation and interaction among variables. Secondly, to obtain the best predictive model for stunting, all statistically or conceptually significant variables within each dimension were included in an unconditional multiple logistic regression analysis; the method selected for modelling started from a saturated model until finding the best explanatory model, after assessing the significance of each covariate and adjusting for major potential confounders such as age and sex of the child, literacy of the mother and household income. Once the best explanatory model was found, goodness of fit assessment was performed. The statistical analysis was carried out using SPSS (SPSS Professional Statistics 7.5 SPSS Inc. 1997) and STATA (STATA Statistical Software: Release 5.0 Stata Corporation, 1997). Results Through the census, 326 eligible children aged between 6 and 23 months living in the rural area and 448 in the urban area were identified. Two hundred and forty-seven children (75.8%) from the rural area and 409 (91.3%) from the urban area who fulfilled the inclusion criteria were located. Because of the harvest season, some families living in the rural area migrated temporarily, so the remaining 24.2% of children could not be located. Regarding the children living in the rural area 43.7% were stunted. Therefore, the group was divided into 108 cases and 139 controls; as to the urban area, 48.4% of children were stunted, resulting into 198 cases and 211 controls. Table 2 shows children and housing characteristics of both study groups. Table 2 Children and housing characteristics of study groups Variable Rural Urban Cases n = 108 % Controls n = 139 % Cases n = 198 % Controls n = 211 % Child characteristics Age in months (median, min-max) 16 (6–23) 13 (6–23)** 15 (6–23)| 12 (6–23)** Sex (male) 53.7 53.2 58.1 48.4* First birth order 11.1 20.9* 26.8 35.1* Birth weight (g) mean (SD + ) 2947(444) 2984(355) 3178(485) 3370(484) Housing characteristics Dirt floor 87.0 85.6 40.9 31.1* No indoor plumbing 78.7 71.9 60.6 49.3* No sewerage system 100.0 100.0 39.4 28.0 No separate kitchen 87.0 79.9 57.1 44.1** Overcrowding 79.6 64.7 48.0 33.2** In the rural area, the proportion of first birth order children in the group of cases was lower than in the control group. Housing conditions were similar in both groups. Regarding urban children, cases had poorer housing conditions than controls; also, cases were older than controls in both study areas. Table 3 shows the distribution of variables within each dimension. Table 3 Distribution of variables by dimensions Variable Rural Urban Cases n = 108 % Controls n = 139 % Cases n = 198 % Controls n = 211 % Dimension 1. Family characteristics Mother's age (years) Median (min-max) 30 (18–43) 27 (15–44) 25 (15–43) 24 (16–47) Father's age (years) Median (min-max) 32 (19–45) 30 (15–60) 28 (18–56) 28 (18–60) Illiteracy of mother 50.0 37.4* 62.9 44.8** Illiteracy of father 65.7 53.2* 50.3 46.2 Parents' civil status (married) 85.2 72.7** 63.1 68.2 Duration of parents' union longer than two years 95.4 84.2** 15.6 27.5** Type of family (nuclear) 61.1 57.6 68.2 71.6 Completeness of the family 94.4 93.5 90.9 88.6 Size of the family (number of members) median (min-max) 8 (3–16) 7 (2–16) 5 (3–14) 5 (2–18) Mother's use of contraceptive method 13.9 26.6* 58.7 69.6* Dimension 2. Family income Mother's occupation Housewife 80.6 76.3 86.4 84.8 Other 19.4 23.7 13.6 15.2 Father's occupation Worker 13.0 20.9 27.8 20.4* Farmer 65.721.3 53.2* 17.7 22.3 Other 25.9 55.5 57.3 Time during which father has worked in the same place (months) Median (min-max) 7 (1–30) 8 (0–36) 4 (1–60) 4 (1–30) Per capita family income per month (USD) Mean (SD) 22.8 (11.7) 25.2 (6.0) 39.5 (18.6) 45.2 (24.1)* migration of parents From rural to urban settings 71.2 55.8* From rural to rural 2.8 8.6 Dimension 3. Household allocation of resources and family organisation Child care provided exclusively by the mother 80.6 89.9* 89.9 87.2 Time spend by the mother to care for the child (hours/day) Median (min-max) 5 (3–17) 6 (0–21) 5 (0–21) 5 (1–14) % of family income spent in: Food. Mean (SD + ) 39.0 (13.5) 37.5 (14.9) 52.0 (15.6) 50.7 (14.8) Transportation. Mean(SD + ) 8.5 (5.9) 8.1 (4.1) 12.3 (7.3) 12.7 (7.0) Dimension 4. Social networks Lack of social networks 24.1 24.5 14.1 14.7 Size of network (Small) 63.0 59.0 75.3 63.0** Type of support Child care 64.8 50.7* 62.1 64.5 Economic 62.0 63.3 22.2 14.2** Dimension 5. Child health care Breast feeding (months) Median (min-max) 7 (0–12) 6 (0–17) 4 (2–15) 3 (1–12) Age at weaning (months) Median (min-max) 7 (1–14) 6 (1–13)** 4 (2–15) 4 (1–12) Complete immunisation scheme 83.3 87.8 77.3 82.5 Attendance to the Well Child Program activities (number of visits) Median (min-max) 2 (0–4) 2 (0–9) 2 (0–6) 3 (0–8)** * p < 0.05, **p < 0.01 (between cases and controls within the same area) + Standard deviation In the rural area, the following variables showed statistically significant differences when comparing cases and controls: parents' illiteracy, parents' civil status, duration of parental union, mother's use of contraceptive method, father engaged in farming activities, exclusive provision of care by the mother, social networks for child care, and age at weaning. At the urban area, mother's illiteracy, duration of parental union longer than two years, father's occupation, per capita family income, parents' migration from rural areas, size of social networks and frequency of attendance to the Well Child Program activities, were statistically significant variables. Table 4 shows the results of the crude analysis of each dimension. In the rural area there was at least one significantly associated variable, while in the urban area there were two or more, except in dimension 3 (household allocation of resources and family organisation), in which no statistically significant variables were found. Table 4 Variables associated with stunting, bivariate analysis Variable Odds Ratio Confidence interval (95%) p value RURAL AREA Dimension 1. Family characteristics Duration of the parents' union: longer than two years 3.81 1.38 – 10.53 .01 two years or less 1.00 Mother not using a contraceptive method 2.22 1.13 – 4.34 .02 Mother using contraceptive method 1.00 Dimension 2. Family income Father occupation: Farmer 1.68 1.00 – 2.83 .04 Another type of job 1.00 Dimension 3. Household allocation of resources and family organisation Child care: provided exclusively by mother 0.46 0.22 – 0.96 .03 shared with other caretakers 1.00 Dimension 4. Social networks Family networks for child care 1.79 1.06 – 3.00 .02 Without family networks 1.00 Dimension 5. Child health care Weaning: after six months of age 2.22 1.33–3.70 .002 at/before six months of age 1.00 URBAN AREA Dimension 1. Family characteristics Illiteracy of mother 2.09 1.40 – 2.14 .001 Non-illiteracy of mother 1.00 Duration of parents' union: longer than two years 2.04 1.24 – 3.35 .005 two years or less 1.00 Family with more than 4 members 1.52 1.03 – 2.26 .03 Family with 4 members or less 1.00 Mother not using a contraceptive method 1.61 1.06 – 2.42 .02 Mother using a contraceptive method 1.00 Dimension 2. Family income Per capita family income: below $25 USD per month 1.65 1.03 – 2.64 .03 above $25 USD per month 1.00 Father engaged at the same work place: equal or less than 2 years 1.84 1.19 – 2.82 .005 more than 2 years 1.00 Parents migration: migrant from rural to urban area 1.96 1.28 – 2.99 .002 non migrant 1.00 Dimension 4. Social networks Small networks 1.78 1.16 – 2.73 .008 Other size or without networks 1.00 Family networks for economic support 1.72 1.03 – 2.87 .03 Without family networks 1.00 Dimension 5. Child health care Number of visits to the Well Child Program: Less than two 2.43 1.58 – 3.74 .0001 Two or more 1.00 Breast feeding: longer than six months 2.23 0.98 – 5.10 .05 six or less months 1.00 Additionally, two of the child's characteristics showed significance (data not presented in the table): child's age (rural area, OR 4.5, CI95% 2.59 – 8.11; urban area, OR 1.98, CI95% 1.29 – 2.92), and child's sex (urban area, OR 1.45, CI95% 0.98 – 2.14; rural area not significant) Table 5 presents the final explanatory models for each area after adjusting for some established risk factors (child's age and sex, maternal literacy, and household income). In the rural area, the model included only three variables belonging to dimensions two, three, and four. Dimension two (family income) showed an increased risk when the father's occupation was farmer. Dimension three (household allocation of resources and family organisation) showed that one covariate related to the mother's activities (the child being cared for exclusively by the mother) had a protective effect. Dimension four (social networks) showed that having family networks to provide care for the child entailed a higher risk. Table 5 Variables associated with stunting, multivariate analysis* Variable Odds Ratio Confidence interval (95%) p value RURAL AREA** Father's occupation: farmer 1.77 0.98 – 3.18 .05 Child care provided exclusively by the mother 0.30 0.13 – 0.69 .004 Family networks for child care 2.31 1.28 – 4.15 .005 URBAN AREA** Duration of parents' union longer than two years 1.89 1.01 – 3.54 .04 Parents migration from rural to urban area 1.57 0.95 – 2.59 .07 Father worked at the same place for 2 years or less 3.23 1.88 – 5.56 .0001 Small networks 2.11 1.27 – 3.49 .004 Less than 2 visits to the Well Child Program 2.57 1.54 – 4.30 .0001 Breast feeding longer than six months 1.71 0.62 – 4.73 .29 * Unconditional logistic regression analysis **Adjusted by age and sex of the child, maternal literacy, and per capita family income in both areas The explanatory model for the urban area included several covariates as risk factors in most dimensions. Dimension 1: migration of parents from rural to urban area, and duration of the parents' union longer than two years. Dimension 2: the father being in the same employment for two years or less. Dimension 4: having small family networks. Dimension 5: child having fewer than two visits to the Well Child Program activities during the past 6 months, and breast-feeding longer than six months. Age showed to be a significant confounder, and other characteristics of the child such as sex or birth weight, or those characteristics related to the parents such as literacy or per capita family income did not change the significance of the model when adjusted. Discussion The role of the family as an important influence for nutritional status of children has been increasingly emphasized during the past few years, [ 29 ] and reinforced by the household production function perspective [ 30 ]. This is defined as "a dynamic process that occurs within the household to allow family members to combine their knowledge, resources and patterns of behaviour, either to promote, recover, or maintain health status" [ 31 ]. Nutritional status is an indicator of well being and malnutrition is the result of a complex process within which coexist a number of variables. The results in this study showed that the family related factors for stunting were different in each context -urban or rural-. The model highlighted the importance of identifying, [rather arbitrarily] a number of conceptually grounded dimensions that could be associated with stunting. As mentioned earlier, rural and urban environments are different. Regarding the results of the rural area, within the family income dimension, the variable father's occupation (farming) was found to be a risk factor for child's stunting. In this area, availability of food depends upon local production, which in turn is related to father's occupation. Maize is the staple, and there is lack of food variability or inability to produce sufficient food for the family's nutritional requirements [ 32 ]. In the urban area the instability of the father's employment (working in the same place for less than 2 years) was found to be a risk factor. Lack of stable employment is a common problem among unskilled workers; their income is low and irregular, thus affecting their capacity to purchase goods and food, which in turn affects child nutrition [ 33 ]. Nevertheless, the other important variable included in this dimension, per capita family income, which was statistically significant in the bivariate analysis, did not show significance in the multivariate analysis. This finding could be interpreted as the ability of the family to cope with a difficult environment. Analysis of microeconomic variables shows the need of further studies to confirm plausibility of our findings and its association with malnutrition. The dimension of family structure, which included socio-demographic characteristics of the parents, migration, and literacy attainment, has been repeatedly associated with the nutritional status of children. [ 2 , 5 , 20 , 23 , 34 , 35 ] In the urban area, some indicators of this dimension were associated with higher risk of malnutrition, particularly the longer duration of parents' union. It is possible that the economic and social burden on poor urban families with several children led the mother to give less attention to her younger children, whose nutritional status suffered in consequence. This result highlights the importance of considering the reproductive health sphere when developing health programs. An interesting finding in this study was the relationship between social networks and children's nutritional status. It is often assumed that extended families have extensive social networks, and that this can be advantageous for the care of young children. However, we found quite the opposite in the rural area: the presence of family networks was associated with stunting, while exclusive provision of child-care by the mother showed a protective effect. These findings stress the role of the mother as primary caregiver, at least for young children; such findings also suggest that when a mother is present to care for the child, some of the effects of living in a poor community can be ameliorated. In the urban area, conditions seem to be different, and the sizes of social networks influenced the nutritional status of the children positively. Urban families living in a more aggressive environment might be more likely to obtain health benefits if they have greater family or social support. Use of health services, based on preventive measures from the Well Child Program, was also analyzed. Few visits to the Well Child Program activities were associated with stunting in the urban area. Restricted access to health care services in deprived areas is an important shortcoming with regard to monitoring child's growth and health [ 36 ]. Based on these findings, promotion of comprehensive services for children living in extreme poverty is relevant. Most health workers carry out preventive activities, among which nutritional status monitoring is essential. Therefore, it seems advisable to train these workers to recognise families showing some of the risk factors described in this study, so malnourished or at-risk children can be identified. Migration was also explored. In the final explanatory model addressing the urban area, parents' migration from rural to urban area was found to be a risk factor for stunting. Migration from rural to urban settings is frequent among young people looking for improving their living conditions. However, families' adaptation to the new situation is a long process, given that such families live in a hostile environment and they have limited access to information regarding how to care for the child. In contrast, migration of one member of the family from one urban setting to another did not show a relationship with nutritional status of children. Adult male family members, mainly the heads of household, migrate to other states in Mexico or to the U.S. in search of better income. Perhaps the influence of this variable could be identified by a qualitative approach. Therefore, further research on this aspect is necessary. The study has some limitations. Firstly, the study groups were of unequal size; this is a consequence of including all children aged from 6 to 23 months of age identified by the census. Additionally, since children of the control group were significantly younger than cases, some of them could have become stunted by the time they were of comparable age to the cases. To solve this limitation, the covariate age was adjusted in the multivariate analysis. However, it is possible that an age-matched cases-control design would have been preferable. Secondly, the fact that almost 25% of eligible children in the rural area were not included due to migration of their families could have led to selection bias. Another limitation is the lack of precise information regarding food practices among participating families. The interview allowed identifying general aspects such as duration of breast-feeding and age of weaning. Breast-feeding beyond 6 months of age entailed a risk of stunting, but only in the urban area. This issue has been controversial; some authors suggest that the relationship between prolonged breast-feeding and malnutrition may indicate a maternal decision to continue breast-feeding to a nutritionally disadvantaged child [ 37 - 39 ], rather than being the direct effect of prolonging breast-feeding on the child's nutritional status [ 40 ]. Further research in this area will contribute to the knowledge of cultural preferences about breast-feeding and its consequences on children's nutrition. Conclusions This study suggests the influence of the family on the nutritional status of children under two years of age living in extreme poverty areas. Factors associated with stunting were different in rural and urban communities. Therefore, developing and implementing health programs to tackle malnutrition should take into account such differences that are consequence of the social, economic, and cultural contexts in which the family lives. Competing interests The author(s) declare that they have no competing interests. Authors' contributions HR and AS conceived, designed and co-ordinated the study. RPC participated in the design of the study, statistical analysis and interpretation of data, and drafted the manuscript. RC co-ordinated the acquisition of data and participated in the statistical analysis. SD collaborated in the data analysis and drafted the manuscript. JIS and GG participated in the conception and design of the study, interpretation of data and critical revision for important intellectual content. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539253.xml

534788

Ubiquitin is associated with the survival of ectopic stromal cells in endometriosis

Background Endometriosis is a condition that affects women of reproductive age, where the glandular and/or stromal tissues from the eutopic endometrium implant in ectopic locations. It is well established that the survival of ectopic implants is due to lower levels of apoptosis, but no consensus exists as to which pathway/s this is mediated by. The ubiquitin protein shares a similar sequence homology to an anti-apoptotic protein called BAG-1 and is expressed in the normal endometrium. Currently, no studies have been conducted to determine ubiquitin expression and its possible anti-apoptotic effects in endometriosis. Methods Archived endometrial tissues from endometriosis patients and women undergoing laparoscopic diagnosis (controls) from January 2000 to July 2003 at Westmead Hospital were examined, where 14 cases of endometriosis and 55 controls were included in the study. Results Both the ubiquitin protein and apoptosis were expressed in both glandular and stromal cells throughout the menstrual cycle of the eutopic endometrium, in which ubiquitin exhibited a cyclic expression, reaching a peak in late proliferative phase. In contrast, ubiquitin was predominantly expressed in cells of stromal origin in endometriosis, was no longer regulated by a cyclic pattern and was associated with an aberrant level of cell survival. Conclusions For the first time, this study shows that ubiquitin is expressed in endometriotic cells and may contribute to a reduced sensitivity of ectopic endometrial tissue to apoptosis. These findings also suggest that stromal cells contribute differentially to the development of ectopic endometrial tissue.

Background Endometriosis is a condition that affects 10% of women of reproductive age [ 1 ]. In the condition, endometrial glands and/or stroma from the eutopic endometrium lodge and survive at ectopic sites such as the fallopian tubes, ovaries and peritoneal cavity. The most accepted theory as to how these cells migrate to ectopic areas is Sampson's retrograde menstrual transport theory, where endometrial cells are shed through the fallopian tubes [ 2 ]. However, what is not known is how these foreign cells continue to survive in their new ectopic location. Ubiquitin is a 76 amino acid protein [ 3 ] that is involved in the degradation of short lived, regulatory or misfolded proteins, thus maintaining cellular homeostasis [ 4 ]. Ubiquitin tags these proteins to be taken to the proteasome and in some instances also to the lysosomic machinery to prevent damage to cells. This is the most cited function of ubiquitin, although ubiquitin also shares a very similar sequence homology to the B-cell lymphoma athanogene 1 (BAG-1), which has known anti-apoptotic properties. To our knowledge, the hypothesis that ubiquitin may mediate anti-apoptosis within the endometrium has never been examined and consequently has not been investigated in the area of ectopic endometrial cell survival in endometriosis [ 5 ]. Accumulating evidence suggests that dysregulated apoptotic machinery in endometriosis has a role in its pathogenesis, but no consensus exists as to which pathway/s this is mediated by [ 6 ]. Transformed ectopic endometrial cells are able to survive due to a number of factors, such as down regulation of soluble Fas concentrations and a lack of cell-surface Fas expression on T-cells [ 7 ]. Therefore, this study was undertaken to determine the expression of ubiquitin within endometriosis during the menstrual cycle and to establish whether it is related to a decrease in the incidence of apoptosis in this tissue, thus potentially promoting ectopic endometrial cell survival. Methods Tissues Approval to conduct this study was granted by the Human Research Ethics Committee of the Western Sydney Area Health Service. Archived formalin fixed, paraffin embedded tissues from 14 women with endometriosis and 55 women undergoing laparoscopy for non-endometrial pathologies such as leiomyomata and benign ovarian cyst were studied. Endometriotic implants (n = 20) and endometrial tissues (n = 59) were obtained from the women above, respectively. All tissues were observed from the Department of Pathology at Westmead Hospital between January 2000 and July 2003. Mean ages for endometriosis and control groups were 42.5 ± 2.69 and 36.4 ± 1.17 years, respectively. Women with histological evidence of malignancy, necrosis, active inflammation or hormonal treatment were excluded from the study. Control subjects had normal, regular menstrual cycles and endometrial samples were collected from the proliferative and secretory phases of the cycle. Sectioning Sections (4 μm) of endometriosis and control endometrial tissue blocks were cut on a microtome (Microtome, Stauffenberg, Germany), placed on superfrost slides (Menzel-Glaser, Braunscheig, Germany) and air-dried at room temperature for 24 hours. Localisation of ubiquitin De-waxing was performed with two changes of histolene and absolute ethanol through a single change of 95% (v/v) and 70% (v/v) ethanol and a final change of tap water. All incubations were of 5 min duration. Following the suppression of endogenous peroxidase activity for 10 min with a blocking agent (DAKO Pty Ltd, Botany, Australia) sections were incubated with normal goat serum for 60 min to prevent non-specific binding. A polyclonal rabbit anti-ubiquitin primary antibody (DAKO Pty Ltd, Botany, Australia) prepared at a titre of 1:100 was added for a further 60 min at 37°C. Slides were then washed three times in tris-buffered saline (TBS: 50 mM Tris HCl, 150 mM NaCl, pH 7.5; 5 min) before the biotinylated goat anti-rabbit IgG secondary antibody (DAKO Pty Ltd, Botany, Australia) prepared at 1:200 was applied (30 min). A horseradish peroxidase avidin biotinylated complex kit (HRP-ABC) and diaminobenzidine (DAB) were used according to the manufacturer's instructions (DAKO Pty Ltd, Botany, Australia) to detect the secondary antibody. All washes used TBS except after DAB administration, where water was used. Sections were counterstained with haematoxylin and eosin (H&E) and cover slips were attached using a DAKO aqueous based mounting medium. Negative controls were obtained through the omission of the primary antibody to ubiquitin (that showed the absence of specific staining). Ubiquitin has increased expression in the normal endometrium in both the late proliferative and late secretory phase. In this study, late secretory phase tissue was used as a positive control [ 8 ]. Currently, no single assay exists that detects apoptosis with high specificity and sensitivity [ 9 ]. Thus the TUNEL technique used in this study was used in conjunction with the classical method of H&E staining (which detects nuclear shape changes during the early stages of apoptosis) as positive strand break detection alone (TUNEL) may overestimate the true occurrence of cell death within a given cell population. A presence of DNA strand breaks, for example, may not correlate with nuclear segmentation or may be detected during the late lytic stage of apoptosis, where most cells are no longer viable. TUNEL labelling Sections were deparaffinized and rehydrated with histolene and a graded series of alcohols (absolute, 95%, 90%, 80% and 70%) for 5 min each. This was followed by a 20 min incubation of sections with proteinase K [15 μg/ml in 10 mM Tris/HCl, pH 7.5] at room temperature. Sections were then rinsed twice in phosphate-buffered saline (PBS) and reacted with 50 μl of the TUNEL reaction mixture (Roche Diagnostics, Castle Hill, Australia) for 60 min in a dark, humidified chamber at room temperature. The sections were then rinsed three times in PBS and incubated for a further 30 min with 50 μl of the Converter-POD (Roche Diagnostics, Castle Hill, Australia) followed by 10 min with DAB. This procedure ensures the detection of TUNEL labelled cells. Finally, sections are counterstained and mounted as described for ubiquitin staining. For positive controls, sections were treated with DNAse 1 to induce DNA strand breaks or peroxidase blocking solution was excluded. Negative controls were achieved by omitting terminal deoxynucleotidyl transferase (TdT; Roche Diagnostics, Castle Hill, Australia). H&E stains The Department of Pathology, Westmead Hospital, kindly provided the H&E slides used for this study. The basis for this procedure is to identify cells with cell blebbing, nuclear condensation and cell shrinkage, which are characteristic features of apoptosis. Scoring of ubiquitin sections The ubiquitin staining in the normal and endometriotic tissue was calculated using a semi-quantitative method to determine the average intensity scores of the protein in the nucleus of glands and the stroma using an Optimas Image Analysis program (Silver Spring, USA). Five randomly selected fields were viewed and evaluated with a grading of 0, 1, 2 or 3 (negative, weak, moderate, strong) according to a scale created by Watanabe and colleagues [ 10 ]. Scoring of TUNEL sections TUNEL apoptotic cell numbers were determined by counting darkly labelled cells in five randomly selected fields at X400 and expressed as the apoptotic cell mean/field according to Meresman and colleagues [ 11 ]. Scoring of H&E sections Ten randomly chosen fields at X600 magnification were used to determine the number of apoptotic stromal and glandular cells according to a modified grading scale used by Meresman and colleagues [ 12 ]: (-) < 3 apoptotic cells/field; (+) >3 apoptotic cells/field. This modified grading system was used to make statistical analysis more accurate, as there were only a few samples that exhibited > 8 apoptotic cells/field. Also, apoptotic cells were identified by their characteristic morphological features in H&E-stained late secretory endometrial sections. These included cell shrinkage and chromatin margination or chromatin condensation with formation of apoptotic bodies [ 12 ]. Apoptotic bodies were identified by the presence of one, or several of the following features [ 13 , 14 ]: 1) Single round mass of condensed strongly eosinophilic cytoplasm with a clump of strongly basophilic, homogenous chromatin. 2) Cytoplasm with more than one piece of condensed chromatin. 3) Condensed chromatin fragments without cytoplasm. Statistical analysis All data is expressed as mean ± SEM. Comparisons between experimental groups were performed using the independent t-test and Mann-Whitney U test for parametric and non-parametric analysis, respectively. A P-value less than 0.05 was considered a significant difference between groups. Results Immunohistochemistry of ubiquitin in the control endometrium and endometriotic implants Analysis of the effect of the menstrual cycle phase on glandular epithelial cell ubiquitin expression showed a greater level of ubiquitin during the proliferative phase than the secretory phase in controls (P = 0.032; Figure 1A , 3A,3D,3G,3J , and 3M ) in contrast to ectopic glandular cells of endometriosis tissues where ubiquitin is higher during the secretory phase (P = 0.022; Figure 1A , 4A and 4B ). Figure 1 A. Ubiquitin grades for glandular cells of control endometrium and endometriotic implants during the menstrual cycle . Bars represent mean ± SEM; p < 0.05. P = proliferative and S = secretory. B Ubiquitin grades for stromal cells of control endometrium and endometriotic implants during the menstrual cycle. Bars represent mean ± SEM; p < 0.05. P = proliferative and S = secretory. Figure 3 Immunohistochemical staining of control endometrium. Panels A, D,G,J and M are ubiquitin stained sections where brown cells are ubiquitin labelled (X400). Panels B, E, H, K and N are H&E stained sections where □ are apoptotic cells (X600). Panels C, F, I, L and O are TUNEL stained sections where ▼ are TUNEL positive cells.; Menstrual cycle phases : MP = mid proliferative; LP = late proliferative; ES = early secretory; MS = mid secretory and LS = late secretory. Figure 4 Immunohistochemical staining of endometriotic implants. Panels A and B are ubiquitin stained sections where brown cells are ubiquitin labelled (X400). Panels C and D are H&E stained sections where □ are apoptotic cells (X600). Panel E and F are TUNEL stained sections where ▼ are TUNEL positive cells. Menstrual cycle phases : P = proliferative and S = secretory. Furthermore, the effect of the menstrual cycle phase on stromal cell ubiquitin expression, showed a similar level of ubiquitin between the proliferative and secretory phase of controls (P > 0.05) in contrast to ectopic stromal cells of endometriosis tissues where ubiquitin is higher during the secretory phase (P = 0.020; Figure 1B and 4B ). Detection of apoptotic cells in the control endometrium and endometriotic implants A significantly higher level of apoptosis was observed using H&E in ectopic glandular cells of patients with endometriosis in comparison to controls in the proliferative phase (2.79 ± 0.43 vs 1.90 ± 0.15 respectively; P = 0.04; Figure 2A , 3B,3E and 4C ) whereas no significant difference was seen during the secretory phase (1.29 ± 0.23 vs 1.64 ± 0.19 respectively; P > 0.05; Figure 2A , 3H,3K,3N and 4D ). Figure 2 A. Apoptotic grades for glandular cells of control endometrium and endometriotic implants during the menstrual cycle using H&E. Data are expressed as the apoptotic cell mean/field. Bars represent mean ± SEM; *p < 0.05. Menstrual cycle phases : P = proliferative and S = secretory. B. Apoptotic grades for glandular cells of control endometrium and endometriotic implants during the menstrual cycle using TUNEL Data are expressed as the apoptotic cell mean/field. Bars represent mean ± SEM; Menstrual cycle phases : P = proliferative and S = secretory. C. Apoptotic grades for stromal cells of control endometrium and endometriotic implants during the menstrual cycle using H&E. Data are expressed as the apoptotic cell mean/field. Bars represent mean ± SEM; *p < 0.05. Menstrual cycle phases : P = proliferative and S = secretory. D. Apoptotic grades for stromal cells of control endometrium and endometriotic implants during the menstrual cycle using TUNEL. Data are expressed as the apoptotic cell mean/field. Bars represent mean ± SEM; Menstrual cycle phases : P = proliferative and S = secretory. Similar levels of apoptosis were observed within proliferative phase ectopic stromal cells of patients with endometriosis and controls using the H&E method (4.57 ± 0.85 vs 5.08 ± 0.50 respectively; P > 0.05; Figure 2C , 3B,3E and 4C ). In contrast, a considerably lower level of apoptosis was observed using H&E in secretory phase ectopic stromal cells of patients with endometriosis than in controls (2.73 ± 0.63 vs 4.17 ± 0.33 respectively; P = 0.03; Figure 2C , 3H,3K,3N and 4D ). Apoptotic cells are seen using the TUNEL technique (Figure 3C,3F,3I,3L,3O , 4E and 4F ). However no statistically significant difference in levels of cell death were observed for both ectopic glands of endometriosis patients and controls in either the proliferative (1.73 ± 0.88 vs 1.51 ± 0.38 respectively; P > 0.05; Figure 2B , 3C,3F and 4E ) or secretory phase (1.05 ± 0.48 vs 0.70 ± 0.19 respectively; P > 0.05; Figure 2B . 3I,3L,3O and 4F ). Similar levels of apoptosis were also observed in ectopic stromal cells of endometriosis patients and controls using the TUNEL technique in both the proliferative (4.53 ± 0.93 vs 3.59 ± 0.85 respectively; P > 0.05; Figure 2D , 3C,3F and 4E ) and secretory phase (1.83 ± 0.43 vs 2.08 ± 0.52 respectively; P > 0.05; Figure 2D , 3I,3L,3O and 4F ). Discussion Ubiquitin is implicated in the removal of short lived, regulatory or misfolded proteins but is also widely known to play a part in DNA repair and removal of virus budding [ 4 ]. The expression of ubiquitin during the proliferative phase of control tissues is likely to be modulated by increasing oestrogen since there is an increasing oestrogen level in response to FSH during folliculogenesis. Thus ubiquitin may play a role in supporting the developing endometrium should implantation occur, as its maximal expression correlates to day 14 of the menstrual cycle. In addition, Bebington and colleagues have previously shown an increase in ubiquitin levels during the late secretory phase as an indication that the protein may take part in the differentiation of the endometrium [ 8 ]. In this study, apoptosis was found to be present in endometriotic tissue with varying intensity. A previous study of endometriosis has also shown this result [ 15 ] but the possible biological relationship of ubiquitin to this condition has not previously been elucidated. Our data shows that the expression of ubiquitin is increased during the secretory phase of the menstrual cycle in both glands (Figure 1A ) and stroma (Figure 1B ) in endometriosis tissues compared to controls. In addition, the level of apoptosis observed within the glands was greater during the proliferative phase (Figure 2A ), in contrast with a significant decrease in apoptosis in ectopic stromal cells during the secretory phase (Figure 2C ). This study suggests that increased levels of ubiquitin within ectopic endometrial cells may allow their continued survival, through a yet to be established pathway. The up-regulation of ubiquitin in human endometriotic tissue may facilitate ectopic endometrial cell survival, particularly allowing those of stromal origin to grow, survive and evade T-cell mediated disposal. This finding is of particular interest because ubiquitin has primarily been attributed to the removal of aberrant proteins but seems, from our data, to be also associated with cell survival. This duality in ubiquitin's role may be attributed to the type of lysine residue linkage that occurs during polyubiquitination. Studies by Deng and colleagues [ 16 ] have shown that if ubiquitin proteins are linked to each other through lysine 48, the target protein (in this case proteins on ectopic endometrial cells), will be directed to the proteasome for degradation. However, if the linkage occurs through lysine 63, the target protein associates with other proteins, aiding in its survival. Our data is consistent with the hypothesis that ubiquitin has a protective effect on ectopic endometrial cells, particularly those of stromal origin, as shown by the increased level of ubiquitin with an associated decrease in apoptosis during the secretory phase. However, a different mechanism may apply to glandular endometriosis, where despite a greater ubiquitin expression during the secretory phase, no significant association was found with ectopic glandular cell survival. This suggests that an insufficient ubiquitin tagging during the proliferative phase may cause ectopic glandular cells to undergo apoptosis. Other factors that may explain ectopic endometrial cell survival include the down-regulation of apoptotic receptors [ 15 ], failure of immune cells to recognise and eliminate ectopic cells [ 17 ] and an increase in cytokine levels and growth factors in the peritoneal fluid of women with endometriosis [ 18 ]. The significant increase in age of patients with endometriosis may have an unknown effect on the level of apoptosis as older women may have increased incidence of endometrial cell death due to increased longevity. However, this age difference is consistent with another study that also shows a higher age range in women with the condition [ 19 ]. Our results suggest that ubiquitin is important for endometrial function throughout the menstrual cycle where it may play an important role in the regeneration of the endometrium [ 5 ]. Also, a loss of ubiquitin regulation in the ectopic environment may have an important role in the pathogenesis of endometriosis. In this respect, it is interesting that ubiquitin has recently been shown to exert an immunomodulatory effect in other tissues [ 20 ][ 21 ]. Endometriotic implants may survive at ectopic locations due to a combination of factors, including protection from apoptosis and immune attack. Conclusions In conclusion, this study demonstrates for the first time that ubiquitin is expressed in endometriotic cells. Furthermore, the up-regulation of ubiquitin expression may contribute preferentially to the development of these ectopic endometrial lesions due to their reduced sensitivity to apoptosis. Authors' contributions RI was responsible for carrying out this study as part of her honours degree. CB provided RI with training in the techniques used. SF and CM were responsible for the conception and design of the study, research funding, and for supervision of RI's work. SF, CB and CM read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC534788.xml

539284

Integrating partonomic hierarchies in anatomy ontologies

Background Anatomy ontologies play an increasingly important role in developing integrated bioinformatics applications. One of the primary relationships between anatomical tissues represented in such ontologies is part-of . As there are a number of ways to divide up the anatomical structure of an organism, each may be represented by more than one valid partonomic (part-of) hierarchy. This raises the issue of how to represent and integrate multiple such hierarchies. Results In this paper we describe a solution that is based on our work on an anatomy ontology for mouse embryo development, which is part of the Edinburgh Mouse Atlas Project (EMAP). The paper describes the basic conceptual aspects of our approach and discusses strengths and limitations of the proposed solution. A prototype was implemented in Prolog for evaluation purposes. Conclusion With the proposed name set approach, rather than having to standardise hierarchies, it is sufficient to agree on a suitable set of basic tissue terms and their meaning in order to facilitate the integration of multiple partonomic hierarchies.

Background Introduction As the bioinformatics emphasis has shifted from gene sequence analysis to functional genomics and proteomics, the need to describe gene function in the context of specific tissues of an organism has increased. Hence, in addition to anatomy ontologies built for medical purposes, e.g. GALEN [ 1 ], descriptions of anatomies are now often used to annotate a variety of genetic data, such as gene-expression. (A list of such ontologies for human as well as model organisms, e.g. mouse, Drosophila , zebrafish and C elegans , can be found on the Open Biological Ontologies web site [ 2 ]. An ontology model typically consists of concepts and relationships between these concepts. One of the key relationships in anatomy is part-of . It is possible to distinguish between different kinds of part-of , e.g. structural part-of and functional part-of . Each anatomy ontology may define one or more such part-of relationships. Even for a single type of part-of, there may be more than one correct way to devide the anatomy of an organism into parts and subparts. Hence, multiple valid partonomic (part-of) hierarchies may exist for any organism. This raises the issue of interoperability across such hierarchies: when is a tissue in one hierarchy equivalent to a tissue in another hierarchy, and what are the part-of relationships across these hierarchies? In general, biologists refer to tissues by their names, unlike computers which can easily work with ID numbers. For example the name "/embryo/limb/forelimb bud/ectoderm" is used to describe the ectoderm of the forelimb bud of the limb of the mouse embryo. Although this name uniquely identifies the tissue, it does so by encoding the particular partonomic hierarchy in the name. This causes problems when trying to work with more than one single hierarchy. This paper discusses a naming scheme that preserves the unique identification property of tissue names, without having to restrict it to a particular hierarchy, thus making it easier to integrate multiple partonomic hierarchies. There is a large body of work discussing mereology (part-of relationships) in the biomedical literature. For example, Rogers and Rector [ 3 ] describe their experience of modelling part-of relationships in human anatomy as part of the GALEN project. Aspects of the Digital Anatomist Foundational Model (FMA) are given in [ 4 ]. Partonomies in a 3D model of human anatomy are briefly discussed in [ 5 ]. All of these papers distinguish between different kinds of part-of relationships. An example of an anatomy ontology using only one type of part-of can be found in the Edinburgh Mouse Atlas Project (EMAP). Although EMAP also uses derives-from relationships to capture cell lineage information with respect to embryo development, it is significantly less complex than GALEN and the FMA. Such variation of complexity is common and typically reflects the different purposes for which the ontologies were built. The EMAP ontology is used to label spatial data for the developing mouse embryo, specifically gene expression data [ 6 ]. We are not aware of any previous work dealing specifically with the integration of multiple part-of anatomy hierarchies. However, ontology alignment and integration in general is an active reserach area and has produced tools that aim at helping with the manual alignment of ontologies as well as with the automation of ontology integration. Examples of such tools include OntoMorph [ 7 ], OntoMerge [ 8 ] and the PROMPT tools suite [ 9 ]. Some work has been carried out in trying to use such tools to systematically merge GALEN and FMA, but the results have been rather limited [ 10 , 11 ]. In this paper we are not trying to argue for a general solution to the ontology integration problem, which as the evidence suggests is very hard to achieve. Instead we approach the problem from our specific application experience and seek to find a specific solution for a more limited domain. The remainder of the paper is organised as follows. The next section introduces the Edinburgh Mouse Atlas, which forms the basis of the work presented here. Thereafter, the issue of multiple part-of hierarchies is discussed. The next section introduces the developed name set representation, followed by a discussion that covers the implementation of a Prolog prototype system. The proposed approach is then evaluated in the discussion section, followed by our conclusions. Edinburgh Mouse Atlas The Edinburgh Mouse Atlas (EMAP) and Gene Expression (EMAGE) Database project [ 12 - 16 ] has developed a digital atlas of mouse development which provides a bioinformatics framework to spatially reference biological data. The core databases contain 3D grey-level reconstructions of the mouse embryo at various stages of development, a systematic nomenclature of the embryo anatomy (the anatomy ontology), and defined 3D regions (domains) of the embryo models which map the anatomy onto the spatial models. Through the 3D domains users can navigate from the spatial representation of the embryo to the ontology and vice versa. Data from an in situ gene expression database is spatially mapped onto the atlas allowing the users to query gene expression patterns using the 3D embryo model and/or the ontology as a reference. Following the description of mouse embryo development by Theiler [ 17 ], the anatomy ontology is organised into 26 developmental stages, referred to as Theiler stages (TS1-TS26). Each stage is primarily organised as a structural part-of tree, or partonomic hierarchy. Figure 1 shows the top 3 levels of the tree at TS6. (The browser shown in the figure is available on-line at the Mouse Atlas web site [ 12 ].) The tissues represented by subnodes of a node in the tree are intended to be non-overlapping ( exclusive ) and complete , i.e. they describe all distinct parts of the parent tissue. For example, in Figure 1 , the trophectoderm consists of the mural trophectoderm and the polar trophectoderm, which are distinct from each other and are the only parts of the trophectoderm at that stage. Although this holds for EMAP, it is not a requirement for the proposed approach. (In this paper, the term 'tissue' is used in a very generic way, meaning both: whole anatomical structures as well as specific tissues.) Each tissue can be uniquely identified by its full name . A full name is an n-tuple: ( t 0 , t 1 ,..., t n ). The path name of the tissue is ( t 0 , t 1 ,..., t n -1 ). The component name is t n . For example, given the tissue name (using a file directory style notation): /embryo/branchial arch/3rd arch/branchial pouch/endoderm/dorsal its full name is: (embryo, branchial arch, 3rd arch, branchial pouch, endoderm, dorsal) its path name is: (embryo, branchial arch, 3rd arch, branchial pouch, endoderm) and its component name is: dorsal Although the ontology covers all parts of the mouse embryo, there may not be a single node representing a particular tissue of interest. For example, there is no single node named (embryo, mesenchyme, trunk mesenchyme, paraxial mesenchyme, somite, sclerotome). However, there is a tissue named (embryo, mesenchyme, trunk mesenchyme, paraxial mesenchyme, somite), which has somite 05 to somite 20 as subparts (somite 05 to somite 20 are part of that tissue), and each of those has a subpart with component name sclerotome. The approach taken in EMAP is to introduce a new tissue node, called a group , with the appropriate subparts identified. Figure 2 shows the anatomy part-of graph for this example (at Theiler stage 14). Although adding the notion of groups to EMAP is addressing the need for alternative arrangements of the part-of hierarchy, it does also raise a number of new questions. For example, it requires a suitable algorithm to determine appropriate tissues of which the new group should be part of. Also, some restrictions should be put in place to constrain what new links can be added; for example, if a new group contains all parts of some other tissue, then that tissue itself, rather than all of its parts, should be linked to the group. In other words, we require a mechanism that prevents biologists from adding too many part-of links unnecessarily. Let us assume that a new group needs to be introduced that contains leg as one of its parts. In this case the biologist should introduce a single part-of link between the new group and leg, and not multiple part-of links between the new group and hip, knee, lower leg and upper leg (which are the parts a leg consists of). The fact that these are parts of the group should be deduced from the transitivity property of the part-of relationship. These and other considerations seem representative of the more general problem of trying to integrate multiple part-of hierarchies over the same anatomical space. The remainder of this paper describes a possible solution to this problem. Multiple part-of hierarchies As previously mentioned, there is more than one way to structure the anatomical part-of hierarchy of an organism. The intersection of these hierarchies may occur at any level; they may share some or all of the their leaf nodes, but may also share intermediate nodes. A particular hierarchy may only deal with part of the organism, e.g. brain or heart, while others, such as EMAP, cover the entire organism. The central example we use in this paper is that of somites. The somites are a repeating anatomical structure down the back of the animal. They give rise to the vertebrae, muscles of the backbone, skin and other structures. Each somite is divided into 3 parts: dermomyotome, myocoele and sclerotome. The dermomyotome is a group of cells which form the dermal layer of the skin and muscle tissue. The myocoele is a fluid-filled cavity of the somite, and the sclerotome gives rise to the bone of the vertebrae. Most ontologies require each of their concepts to be uniquely identified by a name. In the context of an anatomical ontology, such as EMAP, it is clearly not enough to simply use the name sclerotome when wanting to refer to the sclerotome of somite 18. In general, the full name of the tissue is required, though in some cases a part of it may be sufficient, e.g. there is only one tissue at Theiler stage 14 that has component name somite 18. Focusing on the somite part of the anatomy given in Figure 2 , we can draw two possible hierarchies, as shown in Figure 3 . (somite, somite 05, dermomyotome) in H1 and (somite, dermomyotome, somite 05) in H2 clearly semantically refer to the same mouse embryo tissue, in spite of using different names. Hence, for an anatomy ontology to embody its particular part-of hierarchy in the naming of its tissues is not helpful when it comes to integrating multiple hierachies. The proposal is therefore to avoid this problem by using name sets to identify a particular tissue. Results Name set representation of part-of hierarchies Basic name sets Each tissue in a part-of hierarchy is identified by the set of component names along the path from the root to the tissue (including the component name of the tissue itself). For example, in H1 the dermomyotome of somites 5 and 20 are represented as {dermomyotome, somite, somite 05} and {dermomyotome, somite, somite 20}, respectively; and in H2 somite 20's dermomyotome is represented as {dermomyotome, somite, somite 20}. Using NS(T) to denote the name set of tissue T , equivalence between two tissues is identified by the equivalence of their name sets: NS ( T i ) = NS ( T j ) → T i = T j Let T i T j denote that T i has T j as a direct subpart, and let T i T j denote that T i has T j as a subpart (direct or indirect) 1 , i.e. T i T j ... T k implies T i T k , then the part-of relationships can be derived from the name sets as follows: NS ( T i ) ⊂ NS ( T j ) → T i T j and T i T j ∧ (¬ ∃ k · T i T k ∧ T k T j ) → T i T j The first line simply states that T i has T j as a subpart, if the name set of the first is a proper subset of the name set of the second. The second line states that T i has T j as a direct subpart (or child tissue) if T i has T j as one of its subparts, and there are no other subparts of T i which themselves have T j as one of their own subparts. In the graph representing the ontology, an arrow is drawn from T i to T j if, and only if, T i T j . The name set representation does not explicitly deal with temporal relationships. For example, changes in the anatomy of the developing embryo must be captured explicitly, i.e. if a particular subpart disappears from one developmental stage to the next, this should be reflected in the lack of that subpart in the ontological representation for the latter stage. Furthermore, the given representation does not explicitly distinguish between classes and instances of tissues. For example, while in general it holds that a leg has a lower leg part, this may not be true in specific instances. The proposed representation does not deal with such instance issues; many of the existing model organism anatomy ontologies used in bioinformatics today similarly do not represent information at the instance level. Rest-of tissues A "rest-of" tissue is a tissue that represents all parts of that tissue other than those which are explicitly represented in a "sibling" of the rest-of tissue. For example, the embryo mesenchyme marked as T 1 in Figure 4 does not include the mesenchyme of the first branchial arch (labeled T 3 ) or any of the other parts of the embryo (not shown in the figure). Looking at the name set representation of T 1 and T 3 (in Figure 4 ), we see that NS ( T 1 ) ⊂ NS ( T 3 ). Based on the definition from above, T 1 T 3 follows. This, however, is not true. In other words, the basic name set representation introduced earlier is not sufficient to cope with rest-of tissues. Positive and negative name sets To deal with "exclusions" such as required for rest-of tissues, we introduce negative name sets ( NS n ) in addition to the name sets we already have (and we shall refere to as positive name sets ( NS p ) from now on). A tissue T r includes in its negative name set the component name of any "sibling" tissue T s , if T s has a subpart with the same component name as T r . For example, branchial arch is added to the negative name set of T 1 because of T 3 (from Figure 4 ). Part-of relationships can now be derived from the name set representation of tissues as follows: NS p ( T i ) ⊂ NS p ( T j ) ∧ NS n ( T i ) ∩ NS p ( T j ) = ∅ → T i T j and T i T j ∧ (¬ ∃ k · T i T k ∧ T k T j ) → T i T j The first line states that T i has T j as a subpart, if the positive name set of T i is a proper subset of the positive name set of T j , and the intersection of the positive name set of T j and the negative name set of T i is empty. The intersection part has been added to enforce the exclusions needed to deal with rest-of cases. The second line's meaning is identical to what it was before. Returning to the example in Figure 4 , T 1 is now represented as NS p ( T 1 ) = {embryo, mysenchyme} and and NS n ( T 1 ) = {branchial arch}, T 3 is represented as NS p ( T 3 ) = {1st arch, branchial arch, embryo, mesenchyme} and NS n ( T 3 ) = {}. Since NS n ( T 1 ) ∩ NS p ( T 3 ) = {branchial arch}, i.e. non-empty, T 3 is not a subpart of T 1 , as required. For exclusions to work properly, negative name sets must be propagated to their subparts, as is implicitly the case for positive name sets already. Hence, T 2 (in Figure 4 ) will also include branchial arch in its negative name set, keeping T 3 from becoming one of its subparts. Integration of multiple part-of hierarchies Assuming that two or more part-of hierarchies are based on the same set of component names, integrating these hierarchies becomes a trivial task. Relationships (identity as well as part-of) between tissues from different hierarchies follow directly from the rules described above. For example, applying these rules to the hierarchies in Figure 3 , the integrated part-of hierarchy of Figure 5 can automatically be generated. Given the integrated name set representation of two or more hierarchies, it is not possible to determine which tissue belongs to which original hierarchy. For example, if asked for the immediate subparts of somite, based on the rules governing the part-of relationship, all of the tissues at the second level of the diagram in Figure 5 would be returned. To address this problem, extra information needs to be captured. This can easily be achieved by adding a view set to each tissue. For example, the view set for somite would be {H1, H2}, as it would be for all leaf node tissues in Figure 5 . The intermediate tissue nodes have either {H1} (left part) or {H2} (right part) as their view sets. Thus, recreating one of the original hierarchies simply becomes a matter of filtering the integrated hierarchy using the view sets. In addition to the reconstruction of the original hierarchies, new views on the integrated hierarchy, or even on the original ones, can easily be created using appropriate name set "queries". Prototype A prototype of the name set representation for the Mouse Atlas anatomy ontology has been implemented in Prolog; an extension of the prototype we developed for our work on the Formalisation of Mouse Embryo Anatomy [ 6 ]. This original prototype included the following two predicates: tissue(S, T, FN). • S: stage ID, e.g. 14 for Theiler stage 14; • T: tissue ID number (accession number); • FN: full name of tissue represented by the list [N1, N2, N3, ...]; hasPart(TID1, TID2). • TID2 is an immediate part of TID1, i.e. TID 1 TID 2; For the evaluation of the name set representation, we use an extended version of the tissue predicate (view handling is ommitted from the protoype description to keep our examples simple) : ext_tissue(S, T, FN, NSp, NSpL, NSn, NSnL). • S, T, FN: as above; • NSp: positive name set of tissue represented by a list [N1, N2, N3, ...]; • NSpL: length of NSp; • NSn: negative name set of tissue represented by a list [N1, N2, N3, ...]; • NSnL: length of NSn; For example, the embryo mesenchyme tissue of Figure 4 is represented as: ext_tissue(14,705, ["embryo", "mesenchyme"], ["embryo", "mesenchyme"], 2, ["branchial arch", "limb", "organ system"], 3). The following Prolog clause is used to determine whether T p T c is true: subPart(Tp, Tc) :- ext_tissue(Sp, Tp,_, NSpp, NSpLp, NSnp,_), ext_tissue(Sp, Tc,_, NSpc, NSpLc,_,_), NSpLc > NSpLp, ord_subset(NSpp, NSpc), ord_disjoint(NSpc, NSnp). Predicates ord_subset and ord_disjoint from the Prolog library were used to implement the set theoretic aspects of the representation. Although these predicates support ordered sets, this is not required for our representation (but there were no unordered set predicates in the library). NSpLc > NSpLp is required to enforce proper subset relationships. The following two Prolog clauses are used to determine whether T p T c is true: not_immediate_subPart(Tp, Tc) :- subPart(Tp, Tm), subPart(Tm, Tc). immediate_subPart(Tp, Tc) :- subPart(Tp, Tc), not not_immediate_subPart(Tp, Tc). The Prolog implementation given is not particularly efficient and there are a number of optimisations that could be put in place. However, as the purpose of the prototype was not to deliver a robust application for end-users, but a reference implementation of the proposed approach for evaluation purposes, it proved entirely sufficient. The paper makes no claims over the relative merits of different implementation strategies for the proposed approach. Alternatives to Prolog include using a relational database system or an ontology language, such as OWL (more details of OWL available from W3C [ 18 ]). The latter is of particular interest as it is gaining wide acceptance in the bioinformatics domain. At the time this work began, tools for developing ontologies using OWL were still in their early stages, and hence, we decided not to use them. In the meanwhile, however, Protege [ 19 ] and OilEd [ 20 ], have matured sufficiently and do provide appropriate alternative implementation platforms. Discussion For evaluation purposes, a number of tests were carried out on the name set representation of the Mouse Atlas anatomy. These are discussed here, together with some general observations about the proposed approach. The first assumption that must hold is that no two tissues (at any given stage) have the same name set representation. This was tested using test1 :- ext_tissue(S, T1,_, NSp,_, NSn,_), ext_tissue(S, T2,_, NSp,_, NSn,_), T1 not = T2. test1 returns no, i.e. no two different tissues with the same name sets were found, as required. To test whether all part-of relationships can be reconstructed from the name set representation, we used test2 :- immediate_subpart(T1, T2), not hasPart(T1, T2). test3 :- hasPart(T1, T2), not immediate_subPart(T1, T2). Both, test2 and test3 return no, i.e. the name set representation does not lead to any part-of relationships that are not intended (test2), and all existing part-of relationships are found through the name sets (test3), as required. The smallest form of part-of hierarchy integration is the addition of a new tissue node, which is equivalent to adding a group in EMAP. A recently identified need for a group has been for all (embryo, mesenchyme, trunk mesenchyme, paraxial mesenchyme, somite, myocoele) tissues at Theiler stage 17. Using predicate immediate_subPart_ns(S, NSp, NSn, T). • S: stage ID; • NSp: positive name set of new tissue node (group); • NSn: negative name set of new tissue node (group); • T: tissue ID of immediate sub-part of tissue identified by name sets and stage; we can write a "query" in the form: immediate_subPart_ns(17, ["embryo", "mesenchyme", "myocoele", "paraxial mesenchyme", "somite", "trunk mesenchyme"],[], T), tissue(_, T, FN), writeName(FN), nl, fail. and obtain the following result: ("embryo", "mesenchyme", "trunk mesenchyme", "paraxial mesenchyme", "somite", "somite 05", "myocoele") ("embryo", "mesenchyme", "trunk mesenchyme", "paraxial mesenchyme", "somite", "somite 06", "myocoele") ... ("embryo", "mesenchyme", "trunk mesenchyme", "paraxial mesenchyme", "somite", "somite 30", "myocoele") Similarly, using predicate immediate_superPart_ns(), we obtain: ("embryo", "mesenchyme", "trunk mesenchyme", "paraxial mesenchyme", "somite") immediate_superPart_ns() is analogous, and its Prolog implementation very similar, to immediate_subPart_ns(). Details are, therefore, omitted. The correctness of these results was confirmed by one of the biologists who created EMAP. Other, similar tests, worked equally well. A constraint put on all of these cases, however, is that the name set of the new group tissue must only contain names that are already used in the existing hierarchy. This raises the question of how to deal with the introduction of new component names. For example, the addition of a group (embryo, head) cannot automatically be carried out, since the existing hierarchy does not use head in its name sets. For the integration to work, it is first necessary to add head to the appropriate name sets in the existing hierarchy. This can be done at the highest appropriate levels, since sub-parts inherit all name set elements from their super-parts, and may therefore not require as much effort as one initially expects. For the head example, however, we did identify two additional problems which are likely to be typical in this context. Firstly, some agreement needs to be reached as to what in fact is considered to be part of the newly introduced tissue. In our example: how much of the neck is anatomically considered to be part of the head? The second problem deals with the fact that an existing tissue may need to be divided further in order to obtain the appropriate subparts for the newly introduced tissue. For example, the carotid artery runs from the head into the body of the mouse embryo, i.e. only a part of carotid artery is actually part of the head. Hence, the carotid artery needed to be divided into two subparts, one for the head section of it, one for the rest. In our name set approach, the former contains head in its positive name set, while the latter contains head in its negative name set. Of course, only the head section part becomes part of the head. Neither of these two problems presents any direct consequences for our approach. When merging ontologies of different granularity, the same principle as before applies: shared component names must be used in a consistent manner. Assuming ontology O 1 includes midbrain as one of the parts of the brain, but no further detail, and O 2 is a brain anatomy ontology that divides the midbrain into cerebral aqueduct, floor plate, lateral wall, etc., then we would find {brain, central nervous system, embryo, midbrain, mouse, nervous system, organ system} as the positive name set for midbrain in O 1 , and {brain, cerebral aqueduct, midbrain} as the positive name set in O 2 , resulting in {brain, central nervous system, cerebral aqueduct, embryo, midbrain, mouse, nervous system, organ system} – the union of these previous two name sets – as the representation of midbrain in the merged ontology. The meaning of the component names in the intersection of the two original names sets, {brain, midbrain} must have been used in a consistent manner for the merger to work, though many of the component names will differ across the ontologies, because of the different levels of granularity, e.g. the terms nervous system and organ system are unlikely to be found in the brain specific ontology. (We omitted the negative name sets from this discussion, but the implications are essentially the same as for the positive name sets.) Taking a closer look at these "basic tissue terms", called component names thus far, shows that some of them have additional structural complexity and if one wishes to take advantage of the semantics of these complexities, the proposed name set representation would need to be extended. For example, at Theiler stage 18 the tissue (embryo, branchial arch, 1st arch, mandibular component, mesenchyme) has two subparts, called (..., mesenchyme derived from head mesoderm) and (..., mesenchyme derived from neural crest). The naming, hence, reflects lineage relationships between tissues, and the identity of a tissue is partially established by that relationship. Although extensions to the name set representation could be developed to allow the inclusion and subsequent reasoning over such information, it would lead to a semantic overloading of the name sets and for simplicity are, therefore, not considered further – the (component) name is treated as an atomic string describing a tissue, while the lineage relationship is modelled externally to the name sets. Theoretically, merging two part-of hierarchies can be accomplished by systematically (top-down) adding each tissue from one hierarchy into another, i.e. conceptually the problem can be reduced to iteratively adding "group nodes" as discussed above. The approach discussed in this paper will not work where there has been no agreement on the basic component terms, and as such is different from already existing work on merging autonomous ontologies. This raises two questions: what is the basis on which these terms should be agreed and what benefits are to be obtained from the proposed solution if such agreement has to be reached before these partonomic hierarchies can be merged. With respect to the first question, if a basic term, for example skin, exists, then it must be possible to dissect the mouse to a level that separates all the corresponding tissue from the rest of the mouse tissues, e.g. separate all skin tissue from the rest of the mouse. Other examples of basic terms are, therefore, head, skeleton, limb and forelimb. At this point scientists are then free to use combinations of these terms (for the positive and negative name sets) to describe the anatomical concepts they are interested in, e.g. {head, skin} to refer to the skin of the head. The different anatomy hierarchies created by different scientists can then be automatically merged using the approach proposed in this paper. Hence, to answer the second question from above, the benefit of our solution lies in the removal of the need for multiple scientists to agree on a single anatomy partonomy where all tissue concepts are defined and their part-of relationships specified. Instead, a much more flexible solution is offered without having to sacrifice the interoperability across multiple data sets annotated with these anatomical concepts. Essentially, the solution is based on the transitivity property of the structural part-of relationship. As such, one could imagine implementations other than the one based on name sets to achieve the same result. The basic idea, however, would be the same. Using the name set concept makes the solution more directly accessible to biologists, who are more familiar with naming anatomical concepts than using computer generated IDs. We believe that the same approach may be applicable in other ontology areas, which have similarly transitive relationships, but since we have not tested this idea, we shall not elaborate on it in this paper. Also, the work described here only deals with the integration of hierarchies that are based on the same type of part-of relationships. Some preliminary studies suggest that where there are different types and these types are organised in an is-a hierarchy, the proposed integration mechanism will still work at the level of the common part-of type. For example, let H1 be a part-of hierarchy based on part-of-type-1, and let H2 be a part-of hierarchy based on part-of-type-2. If both, part-of-type-1 and part-of-type-2, are specialised versions of the more general part-of-type-0, i.e. part-of-type-1 is-a part-of-type-0 and part-of-type-2 is-a part-of-type-0, then we can use the proposed approach to integrate H1 and H2. The integrated hierarchy, however, would only support part-of-type-0 semantics. Our work in this area is still in its early phase and beyond the scope of this paper. Further details will be reported elsewhere. The work presented in this paper has focused on the issue of integrating different partonomic hierarchies in one species, mouse. We note that a similar approach may be useful when trying to integrate partonomic hierarchies across different organisms. This is subject of current research work, however, and will be reported on separately. Conclusions Anatomy ontologies play an important role in bio-medical informatics. One of the key relationships modelled in such ontologies is that of part-of . For any given organism, however, there is more than one way to divide it into parts and subparts, thus leading to more than one valid partonomic hierarchy. To be able to interoperate between bioinformatics resources that make use of these anatomy ontologies, the corresponding hierarchies must be reconciled in some way. The paper addresses the problem that unique identifying names for tissues often reflect the partonomic hierarchies in which they are used. Although these names are in fact ordered sets (the order implying a particular hierarchy) of "component names", the order in these sets is not necessary to uniquely identify any tissue. Also, the sets of components in names can be used to derive all part-of relationships in the hierarchy. Based on these observations, we have developed a name set representation which facilitates integration of different partonomic hierarchies. Although this does not eliminate the requirement to agree on a set of suitable basic tissue terms and their meaning, it does remove the need to standardise the partonomic hierarchies. The proposed approach has been tested for the anatomy ontology of the Edinburgh Mouse Atlas. A Prolog prototype was implemented for evaluation purposes. Note 1 T j is a direct subpart of T i , if T j is part of T i and there is no other tissue T k such that T j is part of T k and T k is part of T i . If such a tissue T k exists, T j is an indirect subpart of T i . Authors' contributions AB developed the name set representation, implemented the prototype and carried out parts of the evaluation. YY provided input with respect to the current implementation of EMAP and EMAGE in relation to the proposed name set representation. DD carried out part of the evaluation process. RB contributed to the development of the name set representation. DD and RB are overall project leaders of EMAP and EMAGE. All authors have contributed to the writing and/or revision of the paper.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539284.xml

509409

Recognition and Accommodation at the Androgen Receptor Coactivator Binding Interface

Prostate cancer is a leading killer of men in the industrialized world. Underlying this disease is the aberrant action of the androgen receptor (AR). AR is distinguished from other nuclear receptors in that after hormone binding, it preferentially responds to a specialized set of coactivators bearing aromatic-rich motifs, while responding poorly to coactivators bearing the leucine-rich “NR box” motifs favored by other nuclear receptors. Under normal conditions, interactions with these AR-specific coactivators through aromatic-rich motifs underlie targeted gene transcription. However, during prostate cancer, abnormal association with such coactivators, as well as with coactivators containing canonical leucine-rich motifs, promotes disease progression. To understand the paradox of this unusual selectivity, we have derived a complete set of peptide motifs that interact with AR using phage display. Binding affinities were measured for a selected set of these peptides and their interactions with AR determined by X-ray crystallography. Structures of AR in complex with FxxLF, LxxLL, FxxLW, WxxLF, WxxVW, FxxFF, and FxxYF motifs reveal a changing surface of the AR coactivator binding interface that permits accommodation of both AR-specific aromatic-rich motifs and canonical leucine-rich motifs. Induced fit provides perfect mating of the motifs representing the known family of AR coactivators and suggests a framework for the design of AR coactivator antagonists.

Introduction The androgen receptor (AR) is the cellular mediator of the actions of the hormone 5-α dihydrotestosterone (DHT). Androgen binding to AR leads to activation of genes involved in the development and maintenance of the male reproductive system and other tissues such as bone and muscle. However, it is the pivotal role of AR in the development and progression of prostate cancer that has led to increasing interest in this nuclear receptor. Presently, hormone-dependent prostate cancer is treated with a combination of strategies that reduce circulating levels of androgens, such as the administration of antiandrogens that compete for the androgen-binding pocket in the core of the C-terminal ligand-binding domain (LBD). The benefits of these treatments are typically transient, with later tumor growth associated with increases in expression levels of AR or its cofactors, or mutations that render AR resistant to antiandrogens ( Gregory et al. 2001 ; Culig et al. 2002 ; Lee and Chang 2003 ). Alternative approaches to inhibiting AR transcriptional activity may therefore lie in disrupting critical protein associations the receptor needs for full function. The precise details of how AR binds the dozens of coregulator proteins reported to associate with different regions of AR in vivo remain poorly understood ( Lee and Chang 2003 ). Many nuclear receptors activate transcription by binding short leucine-rich sequences conforming to the sequence LxxLL (where “x” is any amino acid), termed nuclear receptor (NR) boxes, which are found within a variety of NR coactivators including the p160 family. Hormone binding to the LBD stabilizes the C-terminal helix of the receptor, helix 12, in a conformation that completes a binding surface for these LxxLL motifs ( Darimont et al. 1998 ; Nolte et al. 1998 ; Shiau et al. 1998 ; Bledsoe et al. 2002 ). The structural elements composing this binding interface, consisting of helices 3, 4, 5, and 12 of the receptor, are synonymous with a previously defined hormone-dependent activation function that lies within the LBD termed activation function (AF)–2. Association of p160 coactivators allows the recruitment and assembly of a number of other cofactors that together modulate the state of chromatin and interactions with components of the basal transcription machinery to initiate transcription ( Glass and Rosenfeld 2000 ). AR, however, utilizes multiple mechanisms to activate gene transcription. Generally, AR activity is dependent on contributions from multiple transactivation functions that lie within the N-terminal domain (NTD) collectively called AF-1. Although the AR AF-2 can bind to a restricted set of LxxLL motifs ( Ding et al. 1998 ; He et al. 1999 ; Needham et al. 2000 ) and is relatively potent ( Wang et al. 2001 ), it usually displays weak independent activity at typical androgen-regulated genes, with significant activity observed only in the presence of high levels of p160 coactivators, as detected in some prostate cancers ( He et al. 1999 ; Gregory et al. 2001 ). Instead, the AR AF-2 exhibits a distinct preference among NRs for phenylalanine-rich motifs conforming to the sequence FxxLF ( He et al. 2000 ; He and Wilson 2003 ). Such motifs have been identified in the AR NTD and in an AR cognate family of coactivators that includes AR-associated protein (ARA) 54, ARA55, and ARA70 ( He et al. 2000 , 2002b ; Lee and Chang 2003 ). The NTD FxxLF motif (residues 23–27) mediates a direct, interdomain, ligand-dependent interaction between the NTD and LBD (N/C interaction) that is thought to facilitate dimerization, stabilize androgen binding, and possibly regulate AF-1 and AF-2 activity ( Langley et al. 1998 ; He et al. 2000 ). In addition, the NTD also contains a related hydrophobic motif, WxxLF (residues 433–437), that nucleates formation of an alternative N/C interaction that may serve to inhibit AR activity ( He et al. 2000 , 2002a ; Hsu et al. 2003 ). Presently, how the AR AF-2 surface can accommodate residues with bulky aromatic side chains and distinguish FxxLF motifs from LxxLL motifs is not known. To understand the structural basis of this unusual coactivator recognition preference, we characterized the full repertoire of interacting sequences using phage display to define amino acids preferred at the AR coactivator binding interface. Crystal structures of the AR LBD in complex with several phage display–derived peptides reveal the structural basis of FxxLF motif specificity and an induced fit of the receptor that allows accommodation of other related hydrophobic motifs. Comparisons of the structures suggest strategies for the design of AR coactivator antagonists. Results AR Preference for Aromatic Groups in Coregulator Recognition Phage display has been used to study coactivator recognition specificity and to identify coactivator motif sequence variants preferred by the estrogen receptor (ER), thyroid hormone receptor (TR) β, and most recently AR ( Chang et al. 1999 ; Norris et al. 1999 ; Paige et al. 1999 ; Northrop et al. 2000 ; Hsu et al. 2003 ). Using phage display, we screened more than 2 × 10 10 randomized peptides against DHT-bound AR LBD. Selections identified sequences containing hydrophobic motifs that were primarily aromatic in character, consistent with another recent study ( Hsu et al. 2003 ) ( Figure 1 ). Of these aromatic motifs, FxxLF and related motifs with substitutions of phenylalanine or tryptophan for leucine at positions +1, +5, or both, dominated the selections. (Peptide residues are numbered in reference to the first hydrophobic residue of the core motif, which is numbered +1. Residues preceding the first hydrophobic residue are numbered negatively in descending order starting with −1.) Substitutions of tyrosine at the +5 position were also observed, but to a much lesser extent (unpublished data). At the +4 position, valines, methionines, and even the aromatic residues phenylalanine and tyrosine were observed ( Figure 1 ; unpublished data). In general, LxxLL motifs were not selected. The LxxLL motif shown in Figure 1 was derived from prior phage selections with ER and subsequently demonstrated to bind AR in FRET-based screens in vitro (unpublished data). Figure 1 AR LBD–Interacting Peptides Selected by Phage Display Hydrophobic residues of the core motif are highlighted in yellow. Residues in bold were ordered in electron density maps. Preliminary characterization of the subset of AR-interacting peptides shown in Figure 1 confirmed that each competed for binding of in vitro translated AR cofactors to bacterially expressed AR LBD in pulldown assays, and generally did so with modestly improved efficiency relative to the native FxxLF motif from the AR NTD and significantly greater efficiency than a native LxxLL motif from glucocorticoid receptor-interacting protein 1 (GRIP1) NR box 3 (P. Webb, personal communication). The equilibrium dissociation constants (K d ) were directly determined for the interaction between the AR LBD and FxxLF and LxxLL peptides and one variant tryptophan-containing peptide, FxxLW, using surface plasmon resonance ( Table 1 ). The K d for FxxLF was 1.1 μM, similar to the affinities of physiologically derived FxxLF motifs determined previously by isothermal titration calorimetry ( He and Wilson 2003 ). The affinity of LxxLL was less than 2-fold weaker, with a K d of 1.8 μM, but more than three times stronger than the tightest binding p160-derived LxxLL motif, NR box 3 of transcriptional intermediary factor 2 (TIF2) ( He and Wilson 2003 ). Surprisingly, the affinity of FxxLW, with a K d of 920 nM, was slightly better than FxxLF, in spite of the presence of the tryptophan residue at the +5 position. Together, our results are consistent with the notion that the phage display peptides interact with the same AR surface that binds FxxLF and LxxLL motifs in native cofactors, and that they do so with similar or improved affinities relative to their natural counterparts. Table 1 Rate and Dissociation Constants for the Interaction between the AR LBD and Selected Peptides Surface plasmon resonance data were best fit using the two-state conformational change model ( Warnmark et al. 2001 , 2002 ). Dissociation constants were calculated from rate constants as described previously ( Warnmark et al. 2001 ) One Site Fits All To understand the binding mode of different AR coactivators, we determined the crystal structures of DHT-bound AR LBD without peptide and in complex with each of the seven peptides listed in Figure 1 . All complexes crystallized in the space group P 2 1 2 1 2 1 with one molecule per asymmetric unit and unit cell dimensions similar to those observed in previous AR LBD crystal structures ( Matias et al. 2000 ; Sack et al. 2001 ). Overall structural features of the complexes are shown in Figure 2 . Peptides assumed short α helical conformations centered on the core hydrophobic motif and bound in a solvent channel relatively free of crystal contacts on a groove formed by helices 3, 4, 5, and 12 of the receptor ( Figure 2 A). Detailed data collection and refinement statistics, as well as buried surface areas for each complex, are listed in Table 2 . The structures confirm previous suggestions that AR utilizes a single binding interface for LxxLL and noncanonical aromatic-rich motifs ( He et al. 2000 , 2002a ). Only side chains move to accommodate the array of peptides, sometimes considerably, with the unbranched side chains of Lys720, Met734, and Met894 making the largest conformational changes upon binding of peptide ( Figure 2 B). Figure 2 A Structural Profile of the AR Coactivator Binding Interface AR–peptide complexes are colored as follows: FxxLF, yellow; FxxLW, orange; WxxLF, wheat; WxxVW, purple; FxxYF, green; FxxFF, blue; LxxLL, pink; unbound, grey. (A) Cα trace of the peptides superimposed onto the AF-2. For clarity only the LBD of AR–FxxLF is shown. (B) Superposition of the LBD of the AR–peptide complexes in the region of the coactivator interface. Backbone atoms are shown as a Cα trace. Side chains of residues composing the interface are shown as sticks. (C) Hydrophobic side chains of the core motif superimposed as in (B). Table 2 Summary of Structures and Crystallographic Statistics a Numbers in parenthesis denote values for the highest resolution shell b R sym = Σ| I − < I >| / Σ ( I ) c R cryst = Σ | F o − F c | / Σ | F o |, where F o and F c are observed and calculated structure factors, respectively; R free was calculated similarly with a randomly selected set of reflections consisting of 5% of total reflections that were excluded from refinement d Values for side chain atoms only FxxLF The mechanisms that permit AR to accommodate motifs with bulky phenylalanine residues were assessed in a crystal structure of the AR LBD in complex with the FxxLF peptide. The FxxLF peptide recapitulates the binding mode of p160-derived LxxLL motifs to other nuclear receptors ( Darimont et al. 1998 ; Nolte et al. 1998 ; Shiau et al. 1998 ; Bledsoe et al. 2002 ). The peptide forms a short α helix whose hydrophobic face, composed of Phe+1, Leu+4, and Phe+5, binds an L-shaped groove formed by helices 3, 4, 5, and 12 of the LBD that is composed of three subsites that accommodate each hydrophobic residue ( Figures 2 A and 3 A). The conserved charged residues at either end of the cleft, Lys720 and Glu897, the so-called charge clamp residues, make electrostatic interactions with the main chain atoms at the ends of the peptide helix: Lys720 with the carbonyl group of Phe+5, and Glu897 with the amide nitrogens of Phe+1 and Arg−1 ( Figure 3 C). Glu897 also interacts with the side chain of Arg−1. The two interior residues of the motif, Glu+2 and Ser+3, are solvent exposed and do not interact with the receptor. Figure 3 Interactions of FxxLF and LxxLL with the AR LBD (A and B) FxxLF (A) and LxxLL (B) bound to the AR AF-2 interface. FxxLF and LxxLL are shown as yellow and pink Cα coils, respectively. Helices 3, 4, and 5 of the LBD are shown as blue ribbons; Helix 12 is shown in green. LBD residues interacting with peptides are depicted as white sticks. For clarity only peptide side chains making significant interactions with the LBD are shown. (C and D) Hydrogen-bonding interactions between backbone atoms of FxxLF (C) and LxxLL (D) with Glu897 of the LBD. Peptide alpha carbons are labeled. Comparison of AR alone and AR in complex with FxxLF (and other aromatic-rich peptides described below) reveals that the AF-2 cleft reorganizes to accommodate the bulky peptide side chains (see Figures 2 B and 4 ). The unbranched side chains of Lys720 and Met734 move from an extended conformation over the +5 pocket to one almost perpendicular to the surface of the protein. The pockets for Phe+1 and Phe+5 are arranged in a line, forming a deep, extended cleft on the LBD spanning the length of the two side chains on the face of the peptide helix (see Figures 3 A and 4 B). Phe+1, almost entirely solvent inaccessible, binds face down at the base of this groove, making hydrophobic contacts with Leu712, Val716, Met734, Gln738, Met894, and Ile898, which define the +1 pocket. The top of the groove, composed of Val716, Lys720, Phe725, Ile737, Val730, Gln733, and Met734, narrows to form the +5 pocket. Met734 and the aliphatic portion of Lys720 constrict this subsite, forming van der Waals interactions with opposite faces of the Phe+5 benzyl ring. Together, the +1 and +5 residues are almost entirely solvent inaccessible. In contrast, Leu+4 binds in a shallow hydrophobic patch consisting of Leu712 and Val716 lined at the ridges by Val713 and Met894 and is largely solvent exposed. Figure 4 Induced Fit of the AR AF-2 Interface Surface representations of the AR AF-2 interface. The unbound structure is shown in (A), the FxxLF bound in (B), and the LxxLL bound in (C). Side chains of the hydrophobic residues of the core motifs of FxxLF and LxxLL are shown as spheres. LxxLL The preference of AR for motifs with aromatic groups over leucine-rich motifs was assessed with a crystal structure of the AR LBD in complex with the LxxLL peptide. The structure reveals similarities between the binding modes of the LxxLL and FxxLF motifs to AR, and other LxxLL motifs to other nuclear receptors. The LxxLL motif adopts a helical conformation, and interactions of the motif with the AF-2 cleft are predominantly hydrophobic, with the three leucine residues of the motif contributing most of the interactions. However, significant differences can be seen between the binding mode of the LxxLL motif to AR and that of p160-derived LxxLL motifs to other nuclear receptors. First, flanking residues were largely disordered, with only two N-terminal flanking residues and one C-terminal residue visible in electron density maps (see Figures 1 and 3 B). This contrasts with extended structures seen in the p160-derived LxxLL motifs in complex with their cognate receptors ( Darimont et al. 1998 ; Nolte et al. 1998 ; Shiau et al. 1998 ; Bledsoe et al. 2002 ). Second, the LxxLL peptide backbone forms hydrogen bonds with only one of the two conserved charge clamp residues, Lys720. A shift in the position of the LxxLL peptide helix precludes direct interactions with Glu897 (see Figures 2 A and 3 D). This shift results from changes in the geometry of the +1 and +5 subsites mediated by Met734, which moves 2.5Å toward the +1 pocket (see Figures 2 B and 4 C) and enables binding of a leucine at the +5 subsite by a simultaneous widening and shallowing of the pocket. This movement of Met734 causes displacement of the +1 residue, resulting in a rotation of the peptide helix away from helix 12, toward helix 3. A slight translation of the peptide helix also occurs away from helix 12 because of the shorter side chain length of leucine (see Figure 2 A). Side chains of residues flanking the first leucine of the motif make additional hydrophobic interactions with the AR surface (see Figure 3 B). Trp+2 reaches over Met734, clamping the methionine in between itself and Leu+1. Leu−1 extends over Met894, abutted against Glu893. These interactions likely explain the moderate affinity of AR for this particular LxxLL motif despite suboptimal complimentarity with the residues of the core motif (as discussed below) and the loss of main chain interactions with Glu897. WxxLF, FxxLW, and WxxVW To understand how the AR AF-2 accommodates tryptophan residues, structures of AR in complex with peptides containing tryptophan substitutions at the +1 or +5 position, or both, were determined ( Figure 5 ). Surprisingly, WxxLF, analogous to the only tryptophan-containing motif known in vivo, WHTLF in the AR NTD, was relatively disordered, with the peptide displaying the highest B-factor and least well defined density, suggesting that it binds with the lowest affinity ( Table 2 ). Nonetheless, each of the tryptophan peptides adopted similar helical conformations. As described above for the LxxLL motif, substitutions at the +1 and +5 positions for non-phenylalanine residues result in shifts of the peptide helix (see Figure 2 A). Consequently, backbone interactions with Lys720 are maintained, but interactions with the other charge clamp residue, Glu897, are lost. Once again, however, flanking residues within the peptide make additional contacts with the AR surface, and, unlike the LxxLL peptide, these contacts include Glu897. In FxxLW and WxxVW, the −2 serine ( Figure 6 ) forms a bidentate hydrogen-bonding interaction, making hydrogen bonds to both Glu897 and the backbone amide group of the +2 residue. Ser−2 of WxxLF similarly interacts with Glu897, but is too distant for helical-capping interactions with the +2 amide group. Instead, Glu893, in a more typical interaction with the +1 amide nitrogen, caps the WxxLF helix ( Figure 6 B). Thus, tryptophan substitutions are tolerated, but they induce a shift in the peptide backbone that precludes interactions with one of the charge clamp residues. This suboptimal interaction is compensated partially by interactions of flanking residues with the AR surface. Figure 5 Interactions of the Tryptophan Motifs with the AR LBD FxxLW (A), WxxLF (B), and WxxVW (C) bound to the AR AF-2 interface. FxxLW, WxxLF, and WxxVW are shown as orange, beige, and purple Cα coils, respectively. The LBD is depicted as in Figure 3 . Figure 6 Interactions of Ser−2 with Glu897 Interactions between Ser−2 of the peptides (A) FxxLW, (B) WxxLF, (C) WxxVW, and (D) FxxFF and Glu897 of the LBD. Peptide alpha carbons are labeled. FxxFF and FxxYF Finally, effects of substitutions at the +4 position were assessed in structures of AR in complex with peptides containing FxxFF and FxxYF motifs ( Figure 7 ). Surprisingly, the binding mode of FxxFF to AR resembled that of the tryptophan peptides more closely than the binding mode of FxxLF (see Figures 2 A and 7 B). Like the tryptophan peptides, interactions with Glu897 are mediated by Ser−2 instead of the peptide backbone (see Figure 6 D). Deviations from ideal helical geometry allow Phe+4 to bind facedown in the +4 pocket with the benzyl ring stacked against Val713. Figure 7 Interactions of FxxYF and FxxFF with the AR LBD FxxYF (A) and FxxFF (B) bound to the AR AF-2 interface. FxxYF and FxxFF are shown as yellow and orange Cα coils, respectively. The LBD is depicted as in Figure 3 . By contrast, the conformation of FxxYF was the closest to FxxLF (see Figure 2 A). Other than FxxLF, only FxxYF makes direct backbone interactions with Glu897. Unlike the facedown orientation of Phe+4 observed in the FxxFF peptide, Tyr+4 is bound edgewise into the shallow +4 pocket, making interactions with Val713, Val716, and the aliphatic portion of Lys717. FxxYF was the most ordered of all the peptides, with 12 out of 15 residues observed in the electron density (see Figures 1 and 7 A). Significant interactions were observed involving residues other than hydrophobic residues of the motif. Lys+2 and Met+6 are predominantly solvent exposed, extending out over the protein surface. Met+6 is bound on top of Phe+5, while Lys+2 makes a water-mediated hydrogen bond with Asp731. Thr−3 of the peptide defines a new subsite, with the hydroxyl group forming a hydrogen bond to Gln738 and the methyl group making hydrophobic contacts in a pocket formed by Glu897, Ile898, and Val901. Similar interactions were observed in the glucocorticoid receptor (GR)–TIF2 complex involving the −3 glutamine of the TIF2 NR box 3 motif ( Bledsoe et al. 2002 ). However a valine to asparagine substitution at the residue corresponding to 901 in AR creates a pocket with a more polar character in GR ( Figure 8 ). Figure 8 Sequence Alignment of the AF-2 Region of NRs Residues composing the coactivator interface of AR are highlighted in yellow. The absolutely conserved glutamate and lysine composing the charge clamp are highlighted in pink and blue, respectively. Residue numbering is that of AR. Restrictions of the Three Subsites Together, the structures described above permit an assessment of the way that individual subsites of the AR AF-2 cleft accommodate hydrophobic groups. The indole rings of tryptophan and the phenyl rings of phenylalanine fit into their pockets analogously with the +1 and +5 residues bound facedown and edgewise, respectively, into the AF-2 cleft. On the other hand, the position of the +4 residue is variable, with binding in this shallow pocket largely dictated by the position of the peptide backbone caused by the bound conformations of the +1 and +5 residues (see Figure 2 C). Small shifts in the position of the N-terminal of helix 12 can be seen, which reposition Met894 for more optimal contacts with +4 residues bound at that subsite (see Figure 2 B). The binding mode detected in the +1 pocket is the most conserved of the three hydrophobic subsites (see Figure 2 C). The benzyl moiety of the indole side chains superimpose with the corresponding benzyl side chains of the phenylalanine-rich motifs, effectively mimicking interactions of a phenylalanine residue. However, the presence of a hydrogen-bonding partner on the indole side chain enables an additional polar interaction not seen in the phenylalanine-rich motifs between the indole nitrogen and Gln738 (see Figure 5 B). Unexpectedly, this additional interaction in the +1 pocket does not occur with Trp+1 of WxxVW (see Figure 5 C). While similarly distanced to make the same interaction, the plane of the indole ring is rotated about 20° relative to that of WxxLF, causing it to be at a poor angle for strong hydrogen bonding to Gln738. Binding of tryptophans in the +5 pocket is slightly more variable (see Figure 2 C). Trp+5 of WxxVW is bound similarly to phenylalanine residues at the same position. Only the six-membered ring of the indole group is fully buried in the pocket. The five-membered ring of the indole side chain sticks out, solvent exposed. In contrast, the +5 indole group of FxxLW is rotated almost 90°, resulting in burial of both rings of the indole group, as well as the formation of a strong hydrogen bond between the indole nitrogen and Gln730 (see Figure 5 A). Binding in this orientation appears to be highly favorable, as the FxxLW peptide deviates from helical geometry at the +5 position to do so. Discussion The crystal structures reported here reveal how AR binds coactivator motifs with bulky aromatic hydrophobic groups and permit construction of a profile of the AR coregulator interface (see Figure 2 ). In some ways, this interface resembles those of other nuclear receptors: it is an L-shaped hydrophobic cleft comprised of three distinct subsites that bind hydrophobic groups at the +1, +4, and +5 positions in cognate peptides. Moreover, the so-called charge clamp residues (Lys720 and Glu897) bracket the cleft. Nonetheless, the AR coregulator recognition site is unique in that it rearranges upon motif binding to form a long, deep, and narrow groove that accommodates aromatic residues at the +1 and +5 positions ( Figure 9 ). Sequence alignments of AR with other NRs suggest that a unique combination of substitutions at Val730, Met734, and Ile737 combine to permit the formation of a smoother, flatter interaction surface that displays a higher complimentarily to aromatic substituents than to branched aliphatic (see Figure 8 ). Of these, methionine, the only unbranched hydrophobic amino acid and the most accommodating, at a key position between the +1 and +5 sites, allows the AR AF-2 interface to vary the size and shape of its pockets to associate with a more diverse set of coregulators. GR also contains a methionine residue at this position, raising the possibility that it may also employ induced fit to broaden motif recognition. While naturally occurring mutations in AR have yet to be observed at Met734, it is interesting to note that mutations at Val730 and Ile737 have been reported in patients with prostate cancer and androgen insensitivity, respectively ( Newmark et al. 1992 ; Quigley et al. 1995 ; Gottlieb et al. 1998 ). Figure 9 Surface Complimentarity of Hydrophobic Motifs in the AR, ERα, and GR AF-2 Clefts (A) AR–FxxLF, (B)AR–LxxLL, (C) ERα–GRIP1 (LxxLL) ( Shiau et al. 1998 ), and (D) GR-TIF2 (LxxLL) ( Bledsoe et al. 2002 ). The inside surfaces of the AF-2 cleft in AR, ERα, and GR are depicted. The LBD is additionally shown as a Cα trace with key side chains shown as white sticks. Phenylalanines and leucines of the FxxLF and LxxLL motifs are shown as spheres. The same characteristics that make the AR AF-2 ideal for binding of longer, aromatic side chains also make it less well suited for binding of shorter, branched side chains. Although changes in the position of Met734 widen the groove towards the +5 subsite to permit binding of leucine residues, the gross features of the groove remain largely the same (see Figure 9 B). As a result, the +1 and +5 leucines bind in a smooth, elongated groove and interactions between the +1 and +5 residues on the face of the peptide helix, or with a hydrophobic “bump” present in other receptors caused by a isoleucine to leucine substitution between the +1 and +5 subsites, are absent. Thus, a smaller proportion of the available surface area is available for van der Waals interactions. Unlike the conserved interaction modes of aromatic residues with the +1 and +5 sites, binding interactions at the +4 site are variable and characterized by nonspecific interactions. This finding agrees with the relatively high conservation of residues at the +1 and +5 positions of AR-interacting motifs and suggests that these residues drive peptide interaction with the LBD, whereas the +4 site is less critical. Indeed, the +4 pocket is shallow, surface exposed, and relatively featureless, explaining the assortment of residues selected at the +4 position. It is likely that any hydrophobic residue that does not clash with surrounding residues would be suitable at this subsite. While peptide motif recognition is governed by hydrophobic interactions, polar interactions from backbone atoms and residues outside the core motif also contribute. With the exception of FxxFF, motifs containing phenylalanines at the +1 and +5 positions present canonical main chain interactions with both charge clamp residues, Lys720 and Glu897. This finding stands in contrast to predictions of previous studies ( Alen et al. 1999 ; He et al. 1999 ; Slagsvold et al. 2000 ; He and Wilson 2003 ), which concluded that Lys720 was dispensable for FxxLF binding and that Glu897 was required for binding to FxxLF and LxxLL motifs. Lys720 comprises a significant portion of the +5 subsite, making important van der Waals interactions with the Phe+5 benzyl group in addition to hydrogen bonds to the motif backbone. These results suggest that Lys720 is required for binding of FxxLF motifs. However, it may be that enough binding energy is provided by the other residues of the +5 subsite (i.e., Met734), as well as by the other subsites themselves, such that removal of Lys720 would have little effect on binding. Observations that Lys720 plays a greater role in LxxLL motif binding are likely due to the fact that there is less surface area contributing to van der Waals contacts in LxxLL motifs. Disrupting binding contributions from Lys720 would thus have a more detrimental effect on binding. On the other hand, Glu897 interacts with the FxxLF peptide backbone, but is disengaged from the LxxLL peptide backbone. One possible explanation for the apparent requirement for Glu897 in LxxLL binding is that it might interact with residues outside of the core motif. The corresponding glutamate of GR, Glu 755, forms hydrogen bonds with the −3 asparagine of TIF2 NR box 3 ( Bledsoe et al. 2002 ), and Glu897 of AR participates in noncanonical interactions with the hydroxyl group of a Ser−2 residue that was selected in all of our tryptophan-containing peptides. This is especially intriguing given that the only WxxLF motif known in vivo, located in the AR NTD, also possesses a Ser−2 residue. WxxLF also makes backbone interactions with an alternate charge clamp residue, Glu893, pointing towards adaptability in AR AF-2 charge clamp formation. Sequence alignment of NR coactivator sequences shows that positively charged residues are favored N-terminal to the core hydrophobic motif while negatively charged residues are favored C-terminal to the motif ( He and Wilson 2003 ). Our phage-selected peptides are consistent with this trend. Arginines and lysines were observed at the N-terminal −1 position in all peptides, except for LxxLL, in which Arg was present at the −3 position. Moreover, four out of seven peptides contained negatively charged aspartate or glutamate residues C-terminal to the core motif. While previous studies have shown that complementary interactions between charged residues flanking coactivator signature motifs of coactivators and charged residues surrounding the AF-2 cleft modulated binding to the receptor ( He and Wilson 2003 ), we find that the flanking charged residues are typically disordered in the electron density, with only Arg−1 of FxxLF interacting with Glu897, and Lys+2 of FxxYF forming a water-mediated hydrogen bond to Asp731. Thus, if charge–charge interactions between flanking peptide residues and the AR surface occur, they are too weak to be detected crystallographically. Finally, the AR AF-2 surface is an attractive target for pharmaceutical design. Selective peptide inhibitors that bind the AF-2 surface of liganded ERα, ERβ, and TRβ have been developed ( Geistlinger and Guy 2003 ), and similar α-helix–mediated protein–protein interfaces have successfully been targeted with tight binding small molecule inhibitors ( Asada et al. 2003 ; Vassilev et al. 2004 ). Drugs that directly interfere with coactivator binding or formation of the AR N/C interaction would likely inhibit AR activity, perhaps even in androgen-resistant prostate cancers in which conventional therapies have failed. Strategies for designing AR coactivator antagonists are revealed in spite of the changes to the structure at the interface. Together the +1, +4, and +5 subsites contribute the majority of buried surface area of the peptide–LBD interaction ( Table 2 ). Inhibitors may be designed by varying hydrophobic constituents at these hotspots. The +1 and +5 subsites of AR have a unique preference for aromatic side chains and provide the most viable starting points for designing AR-specific inhibitors. Aromatic groups, possibly with polar constituents to exploit hydrogen bonding interactions with Gln733 and Gln738 in the +1 and +5 subsites, respectively, may provide promising leads. Indeed, initial screens have yielded compounds that bind to the +1 subsite in such a manner (E. Estébanez-Perpiñá, personal communication). Poorly conserved binding and a lack of strong structural features at the +4 subsite suggest that this site may be incorporated for achieving other characteristics important for inhibitors besides fit. Synthetic strategies that link together groups that bind with moderate affinity to the +1, +5, and possibly +4 subsites may yield tight binding inhibitors of AR coactivator association. Materials and Methods Protein purification Expression and purification of the AR LBD for crystallization were performed essentially as described ( Matias et al. 2000 ). The cDNA encoding the chimp AR LBD (residues 663–919—human numbering), which displays 100% identity to the human form in protein sequence, was cloned into a modified pGEX-2T vector (Amersham Biosciences, Piscataway, New Jersey, United States) and expressed as glutathione S-transferase (GST) fusion protein in the E. coli strain BL21 (DE3) STAR in the presence of 10 μM DHT. Induction was carried out with 30 μM IPTG at 17 °C for 16–18 h. E. coli cells were lysed in buffer (10 mM Tris, [pH 8.0], 150 mM NaCl, 10% glycerol, 1 mM TCEP, 0.2 mM PMSF) supplemented with 0.5 μg/ml lysozyme, 5 U/ml benzonase, 0.5% CHAPS, and 10 μM DHT. All buffers for further purification steps contained 1 μM DHT. Soluble cell lysate was adsorbed to Glutathione Sepharose 4 Fast Flow resin (Amersham Biosciences), washed with buffer containing 0.1% n-octyl β-glucoside, and eluted with 15 mM glutathione. After cleavage of the GST moiety with thrombin, final purification of the AR LBD was carried out using a HiTrap SP cation exchange column (Amersham Biosciences). Eluted AR LBD was dialyzed overnight at 4 °C against buffer containing 50 mM HEPES (pH 7.2), 10% glycerol, 0.2 mM TCEP, 20 μM DHT, 150 mM Li 2 SO 4 , and 0.1% n-octyl β-glucoside, then concentrated to greater than 4 mg/ml for crystallization. Purification of AR LBD for use in phage affinity selection was carried out as above without the final dialysis and concentration steps. The expression construct contained the AR LBD as an inframe fusion with GST in a modified pGEX-2T vector containing both a flexible region and an AviTag sequence (Avidity, Denver, Colorado, United States) allowing in vivo biotinylation. The GST–AR LBD fusion expression plasmid was cotransformed with a plasmid-encoding E. coli biotin ligase (Avidity) into BL21 (DE3) STAR cells. Protein expression was carried out as above but with induction supplemented with 50 μM biotin to ensure quantitative biotinylation of AR LBD. Phage affinity selections and peptide identification Phage affinity selections were performed essentially as described ( Paige et al. 1999 ). Biotinylated AR LBD (10 pmol/well) was incubated in streptavidin-coated Immulon 4 96-well plates (Dynatech International, Edgewood, New Jersey, United States) in TBST (10 mM Tris-HCl [pH 8.0], 150 mM NaCl, 0.05% Tween 20) with 1 μM DHT for 1 h at 4 °C. Affinity selections were performed in TBST containing 1 μM DHT. M13 phage distributed among 24 libraries displaying a total of greater than 2 × 10 10 different random or biased amino acid sequences were added to the wells containing immobilized AR LBD and incubated for 3 h at 4 °C. After washing, bound phage were eluted using pH 2 glycine. Enrichment of phage displaying target-specific peptides was monitored after each round of affinity selection using an anti-M13 antibody conjugated to horseradish peroxidase in an ELISA–type assay. Synthetic peptides corresponding to the deduced amino acid sequences from receptor-specific phage were tested for their ability to interact with purified AR LBD using a FRET–based assay format. Peptides were synthesized according to the deduced amino acid sequence displayed on phage with an additional C-terminal amino acid sequence consisting of SGSGK to allow the attachment of a biotin tag (Anaspec, San Jose, California, United States). Flourophor conjugates were prepared by incubating either biotinylated peptides with streptavidin-cryptate (Cis Bio International, Bagnols Sur Ceze Cedex, France), or biotinylated AR LBD with streptavidin-XL665 (Cis Bio). Interaction between peptide and AR LBD was monitored by the ratio of energy transfer by excitation at 320 nm and emission at 625 nm and 665 nm. Surface plasmon resonance Affinities of peptides to the AR LBD were determined with a Biacore (Piscataway, New Jersey, United States) 2000 instrument. A peptide derived from silencing mediator for RXR and TR 2 (SMRT2) served as a negative control. 1 mM peptide stock solutions in DMSO were diluted into HBS-P buffer (10 mM HEPES [pH 7.4], 150 mM NaCl, 0.005% Surfactant P20) to generate 10 μM working solutions. HBS-P buffer was flowed through the cells to achieve a stable baseline prior to immobilization of the biotinylated peptides. To achieve the binding of approximately 250 RU of peptides to individual cells, working solutions of peptides were diluted to 100 nM in HBS-P buffer. Unbound streptavidin sites were blocked by injection of a 1 mM biotin solution at a rate of 10 μl/min. Purified AR LBD was diluted into HBS-P buffer to a concentration of 10 μM and injected into all four Flowcells using the Kinject protocol at a flow rate of 10 μl/min (contact time 360 s, dissociation time 360 s). Following the dissociation phase, the surface of the chip was regenerated to remove residual AR LBD by QuickInject of buffer containing 10 mM HEPES and 50% ethylene glycol (pH 11). Following the establishment of a stable baseline, the same procedure was repeated using a series of AR LBD dilutions (5 μM, 1 μM, and 300 nM) in an iterative manner. Analysis of the data was performed using BIAevaluation 3.0 software (Biacore). The SMRT2 signals were subtracted as background from the three remaining peptide signals. Data were best fit using the two-state conformational change model ( Warnmark et al. 2001 , 2002 ). Crystallization, data collection, and refinement Purified, concentrated AR LBD was combined with 3x to 6x molar excess of peptide and incubated 1 h at room temperature before crystallization trials. Complexes were crystallized using the hanging drop vapor diffusion method. Protein–peptide solution was combined in a 1:1 ratio with a well solution consisting of 0.6–0.8 M sodium citrate and 100 mM Tris or HEPES buffer (pH 7–8). Crystals typically appeared after 1–2 d, with maximal size attained within 2 wk. For data collection, crystals were swiped into a cryo-protectant solution consisting of well solution plus 10% glycerol before flash freezing in liquid nitrogen. The addition of ethylene glycol to a well concentration of 10%–20% was later found to both improve crystal quality and enable the freezing of crystals directly out of the drop. Datasets were collected at 100K at the Advanced Light Source (Lawrence Berkeley Laboratory, Berkeley, California, United States), beamline 8.3.1, with either a ADSC Quantum 315 or Quantum 210 CCD detector. Data were processed using Denzo and Scalepack ( Otwinowski and Minor 1997 ). Molecular replacement searches were performed with rotation and translation functions from CNS ( Brunger et al. 1998 ). Initial searches for AR–FxxLF were performed using the structure of AR–R1881 (PDB: 1E3G) with R1881 omitted from the search model. Subsequent searches for all other complexes were performed using the refined LBD structure from the AR–FxxLF complex. To minimize the possibility of model bias, FxxLF peptide and DHT were omitted from all molecular replacement searches. Protein models were built by iterative rounds of simulated annealing, conjugate gradient minimization, and individual B-factor refinement in CNS followed by manual rebuilding in Quanta 2000 (Accelrys, San Diego, California, United States) using σ A -weighted 2 F o − F c , F o − F c , and simulated annealing composite omit maps. Superposition of structures was performed with LSQMAN ( Kleywegt 1996 ). Buried surface area calculations were performed with CNS. All figures were generated with PyMOL ( DeLano 2002 ). Coordinates and structure factors for all complexes have been deposited in the Protein Data Bank. Accession numbers are listed in Table 2 . Supporting Information Accession Numbers The Swiss-Prot ( http://www.ebi.ac.uk/swissprot ) accession numbers for the gene products discussed in this paper are AR (P10275), ARA54 (Q9UBS8), ARA55 (Q9Y2V5), ARA70 (Q13772), ER (P03372, Q92731), glucocorticoid receptor-interacting protein 1 NR box 3 (Q61026 ), GR (P04150), NR box 3 of TIF2 (Q15596), and TR β (P10828). The Protein Data Bank ( http://www.rcsb.org/pdb ) accession numbers for the structures used in this paper are FxxFF (1T73), FxxLF (1T7R), FxxLW (1T79), FxxYF (1T7M), LxxLL (1T7F), unbound (1T7T), WxxLF (1T74), and WxxVW (1T76).

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC509409.xml

539291

Functional effects of 17alpha-hydroxyprogesterone caproate (17P) on human myometrial contractility in vitro

Background 17alpha-hydroxyprogesterone caproate (17P) administration reportedly improves outcome for women with a previous spontaneous preterm delivery. This study, using in vitro strips of human uterine smooth muscle, aimed to investigate the direct non-genomic effects of 17P on spontaneous and induced contractions in tissues obtained during pregnancy, and in the non-pregnant state. Methods Biopsies of human myometrium were obtained at elective cesarean section, and from hysterectomy specimens, and dissected strips suspended for isometric recordings. The effects of 17P (1 nmol/L -10 micro mol/L) on spontaneous and agonist-induced (oxytocin 0.5 nmol/L for pregnant, phenylephrine 10 μmol/L for non-pregnant) contractions were measured. Integrals of contractile activity, including the mean maximal inhibition values (MMI) observed at the maximal concentration, were compared with those from simultaneously run control strips. Results There was no significant direct effect exerted by 17P on pregnant or non-pregnant human myometrial contractility. The MMI ± SEM for spontaneous contractions in pregnant myometrium was 4.9% ± 7.2 (n = 6; P = 0.309) and for oxytocin-induced contractions was 2.2% ± 1.3 (n = 6; P = 0.128). For non-pregnant myometrium, the MMI ± SEM for spontaneous contractions was 8.8% ± 11.0 (n = 6; P = 0.121) and for phenylephrine induced contractions was -7.9% ± 6.5 (n = 6; P = 0.966). Conclusions The putative benefits of 17P for preterm labor prevention are not achieved, even partially, by a direct utero-relaxant effect. These findings outline the possibility that genomic effects of 17P, achieved over long periods of administration, are required for its reported therapeutic benefits.

Background Preterm delivery constitutes a major problem in obstetric practice because of the large associated contribution to perinatal mortality and morbidity [ 1 , 2 ]. A significant proportion of all preterm deliveries are due to spontaneous preterm labor [ 2 ]. Despite much research effort, until recently, no effective method of preventing or treating preterm labor, and improving neonatal outcome, has been available. Meta-analysis of various tocolytic compounds in clinical practice has revealed that while they resulted in a delay in the interval to delivery, of time periods up to a week, they did not reduce the incidence of preterm delivery at different gestational ages [ 3 ]. More importantly, their use, in comparison to placebo, was not associated with any benefit in terms of objective measures of neonatal wellbeing or morbidity. It has however been recently reported that weekly injections of 17-alpha-hydroxyprogesterone caproate (17P), in women who have had a previous spontaneous preterm delivery, significantly reduces the risk of preterm delivery before 37, 35 and 32 week's gestation [ 4 ]. In addition, infants of women treated with 17P had significantly lower rates of necrotising enterocolitis, intraventricular haemorrhage and the need for supplemental oxygen. While evidence for the use of progestational compounds for prevention of preterm delivery, and recurrent miscarriage, has been conflicting [ 5 - 7 ], meta-analysis restricted to trials of 17P has suggested a significant reduction in the preterm delivery rate. This reported benefit of 17P, while being a welcome development in therapeutic strategies for preterm labor, has raised many questions in relation to whether the same benefit would apply to low risk groups, and the potential effects, if any, from the castor-oil injection of placebo used [ 8 , 9 ]. One of the most important questions relates to the mechanism by which the drug works and there are currently no data to answer this. Progestins have the potential to exert both genomic and non-genomic effects. The aims of this study were focused specifically on the latter mechanism i.e. to investigate the direct effects of 17P on contractions of isolated human myometrium, both spontaneous and agonist-induced, in tissue obtained during pregnancy and in the non-pregnant state. Methods Patient Recruitment and Tissue collection Patient recruitment took place in the Department of Obstetrics and Gynaecology, University College Hospital Galway. Ethical committee approval for tissue collection was obtained from the Research Ethics Committee at University College Hospital Galway and recruitment was by written informed consent. The biopsies were excised from the upper lip of the lower uterine segment incision in the midline i.e. upper portion of lower uterine segment. Women undergoing induction of labor were excluded from the study. For hysterectomy specimens, all women were pre-menopausal and undergoing abdominal hysterectomy without evidence of malignant uterine disease. Biopsies of myometrial tissue from hysterectomy specimens were obtained from the fundus. The biopsies were immediately placed in Krebs-Henseleit physiological salt solution (PSS), pH 7.4, containing: 4.7 mmol/L KCl, 118 mmol/L NaCl, 1.2 mmol/L Mg 2 SO 4 , 1.2 mmol/L CaCl 2 , 1.2 mmol/L KPO 4 , 25 mmol/L NaHCO 3 , and 11 mmol/L glucose. Tissues were stored at 4° C and used within 12 hours of collection. Tissue Bath Experiments Longitudinal myometrial strips were dissected measuring approximately 2 × 2 × 10 mm. The strips were mounted for isometric recording under 2 grams of tension in organ tissue baths, as previously described [ 10 - 12 ]. The tissue baths contained 10 ml of Krebs-Henseleit physiologic salt solution maintained at 37°C, pH 7.4 and gassed continuously with 95%O 2 /5%CO 2 . Myometrial strips were allowed to equilibrate for at least 60 minutes, during which time the Krebs-Henseleit physiologic salt solution was changed every 15 minutes. After the equilibration period, regular spontaneous myometrial contractions were allowed to develop. In separate experiments contractions were induced by the addition of oxytocin (0.5 nmol/L) to pregnant myometrial strips, and phenylephrine (10 μmol/L) to strips obtained from non-pregnant myometrium. Once regular phasic contractions had developed, the integrated tension for the first 20 minutes was calculated, and this value served as a control since no significant spontaneous reduction in myometrial contractility was observed over the duration of experiments in control strips (without addition of vehicle or 17P). The mechanical response of tissues was then measured by calculation of the integral of selected areas for 20 minute periods, corresponding to the cumulative exposure to 17P or vehicle, using the PowerLab hardware data acquisition system and Chart v3.6 software (AD Instruments, Hastings, UK). At 20 minute intervals 17P was added in a cumulative manner at concentrations of 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L (i.e., 1 × 10 -9 - 1 × 10 -5 M). Control strips (i.e. without exposure to 17P) were run simultaneously and for a similar duration, consisting of two separate groups of experiments as follows: 1. exposure of strips to PSS only; 2. exposure of strips to PSS and the vehicle for 17P. The overall duration of experiments was therefore 3 hours which is in accordance with standard in vitro myometrial experiments, allowing for an accurate drug exposure period of 20 minutes. The effects of 17P were assessed by subtracting the integrals of contractility measured after each bath exposure of 17P, from those obtained in control experiments (vehicle only). This allowed for calculation of the net effect of 17P on myometrial contractility. Potential effects of the vehicle were obtained by subtraction of the mean integrals measured after vehicle exposure, from those obtained without addition of vehicle i.e. spontaneous or agonist-induced contractions in PSS only. Percentage contractility was calculated by expressing the net integral measured after each 17P concentration addition, as a percentage of the integral calculated in the 20 minute period prior to any 17P addition. The mean maximal inhibition (MMI) refers to the mean percentage relaxation [i.e. 100% - mean contractility measured for each separate n sample of non-pregnant and pregnant myometrium] observed at the maximal bath concentration of 17P (i.e. 10 μmol/L), or corresponding concentration of vehicle for control strips. Drugs and Solutions Oxytocin, phenylephrine and 17P were purchased from Sigma-Aldrich, Dublin, Ireland. Because 17P is a lipid soluble compound it was necessary to dissolve it in a lipophilic vehicle. The vehicle used consisted of 75% Dimethyl sulfoxide (DMSO): 25% ethanol, to make a 1 mmol/L (10 -3 M) stock solution of 17P. The resultant final tissue bath concentrations of DMSO and ethanol, at the maximum concentration of 17P investigated (i.e. 10 μmol/L or 10 -5 M), were 0.83% and 0.28% respectively, allowing for the cumulative effect at each bath addition. DMSO and ethanol were purchased from Sigma-Aldrich, Dublin, Ireland. Fresh Krebs-Henseleit physiological salt solution was made daily. Fresh 17P solutions were prepared on the day of experimentation and were maintained at room temperature for the duration of the experiment. Statistical Analysis The integrals of contractile activity measured were compared to control values, and expressed as a percentage of that measured before drug addition. The effects of 17P on myometrial contractility were calculated for each 20-minute period of exposure from 1 nmol/L-10 μmol/L, and compared with the measurements obtained from control strips (ie. spontaneous, oxytocin-induced or phenylephrine-induced contractility, in the presence and absence of vehicle). The integrals of contractile activity were compared using a one-way ANOVA, which was followed by a Tukey HSD post-hoc. A value of P < 0.05 was accepted as statistical significance. Results There were 6 women recruited for the study at the time of elective cesarean section. The reasons for cesarean section included previous cesarean section (n = 2), breech presentation (n = 2), unstable lie (n = 1) and postmaturity with poor cervical Bishop score (n = 1). The mean age, ± standard error of the mean (SEM), was 36.3 ± 2.8 years (range 26–46). The median gestation at the time of cesarean section was 39 weeks (range 38–41). In relation to parity, 2 of the women were para 0, 1 was para 1, and 3 were of parity greater than 1, at the time of recruitment. There were 6 women recruited at the time of abdominal hysterectomy. The reasons for hysterectomy included menorrhagia (n = 4), irregular vaginal bleeding (n = 1), and fibroids with an ovarian cyst (n = 1). The mean age ± SEM was 42.2 ± 2.4 years (range 36–50). For non-pregnant myometrial tissue representative recordings are shown in Figure 1 . In Figure 1A a recording of spontaneous myometrial contractions is shown. In Figure 1B , the effects of vehicle for 17P on spontaneous contractions is demonstrated, and in Figure 1C the effects of addition of 17P are shown. The results of the calculated integrals are provided in Table 1 . Bath exposure of the strips from hysterectomy specimens to 17P did not result in any alteration in contractile activity, at any of the bath concentrations studied, in comparison to vehicle only experiments (spontaneous contractions: P = 0.121; phenylephrine-induced contractions: P = 0.966). Figure 1 Effects of 17P on myometrial contractility in non-pregnant tissue. Representative recordings of spontaneous contractions in PSS only (A), the effects of cumulative additions of 17P vehicle (B), and the effects of cumulative additions of 17P (C) are shown. Table 1 Mean Maximal Inhibition Values for Vehicle and 17P. Myometrial Contractility Non-Pregnant Pregnant Net Vehicle Relaxation ± SEM Net 17P Relaxation ± SEM Net Vehicle Relaxation ± SEM Net 17P Relaxation ± SEM Spontaneous -0.6% ± 12.1 (n = 6) 8.8% ± 11.0 (n = 6) 49.7% ± 11.9 (n = 6) ♦ 4.9% ± 7.2 (n = 6) *Agonist-Induced 31.8% ± 5.0 (n = 6) -7.9% ± 6.5 (n = 6) 56.2% ± 2.3 (n = 6) ¶ 2.2% ± 1.3 (n = 6) *Phenylephrine-induced in non-pregnant myometrium and oxytocin-induced in pregnant myometrium. ♦ P = 0.022 in comparison to controls with PSS only ¶ P < 0.001 in comparison to controls with PSS only The results obtained from myometrium during pregnancy are similarly shown in Figure 2 , and in Table 1 . Figures 2A , 2B and 2C demonstrate recordings of oxytocin-induced contractions, the effects of vehicle, and the effects of 17P respectively. There was no significant net relaxant or uterotonic effect exerted by 17P on pregnant myometrial contractility, at any of the bath concentrations studied experiments (spontaneous contractions P = 0.309; oxytocin-induced contractions: P = 0.128). No significant difference was observed between the effects of 17P on contractility in either pregnant or non-pregnant myometrium. Figure 2 Effects of 17P on myometrial contractility in pregnant tissue. Representative recordings of oxytocin-induced contractions in PSS only (A), the effects of cumulative additions of 17P vehicle (B), and the effects of cumulative additions of 17P (C) are shown. The vehicle for 17P independently exerted a uterorelaxant effect on spontaneous and agonist-induced contractions in pregnant myometrial tissue only (spontaneous contractions P = 0.022; oxytocin-induced contractions: P = 0.000), and this was not observed in non-pregnant myometrial tissue (spontaneous contractions P = 0.241; phenylephrine-induced contractions: P = 0.068) (Table 1 ). Discussion These results demonstrate that 17P does not appear to exert a direct relaxant effect on human myometrial contractions in vitro, in tissue obtained during the third trimester of pregnancy, or in the non-pregnant state. These findings are in contrast to the inhibitory effect of progesterone derivatives on spontaneous contractions in animal uterine tissues [ 13 , 14 ], and therefore suggest that the reported benefit of 17P in preventing preterm delivery in women who have had a previous preterm delivery, involves other mechanisms of action, presumably via its genomic effects. It seems likely that prolonged exposure of uterine smooth muscle to 17P, with resultant activation of the progesterone receptor isoforms, has the potential to modify gene transcription in order to maintain physiological uterine quiescence. However no direct effect on contractile activity was observed in our study over a period of hours of exposure. Previous studies, using various progestins, have yielded conflicting results in terms of the direct effects of these metabolites on human myometrial contractility in vitro. Progesterone metabolites, in some studies, have been reported to decrease both the frequency and amplitude of contractions [ 15 - 17 ], while other reports have outlined that progesterone stimulates the frequency and tonus of contractions in term human myometrium [ 18 , 19 ]. It has been hypothesized that progesterone addition to myometrial strips only enhances contractility if the tissue specimen was never deprived of progesterone i.e. placed immediately in a medium containing progesterone [ 19 ]. The studies to date have included various progesterone metabolites but we are unaware of any previous studies evaluating the effects of 17P on human myometrial contractions in vitro. The focus on 17P in this study has arisen from the recently reported randomized clinical trial outlining its benefits clinically for recurrent preterm labor. 17P is a naturally occurring progesterone which has been isolated from corpus luteum and adrenal gland. The synthetic caproate ester, like the naturally occurring compound, is a steroid, is highly lipophilic and is inactive when administered orally. To achieve solubility for clinical studies, the vehicle used for injection was castor-oil [ 4 ]. This led to significant controversy in relation to the potential effect of castor-oil on myometrial contractions in the placebo arm of the recently reported study [ 8 ]. For laboratory, or in vitro studies, achieving the required solubility of progestins can also be a difficult process. Efforts to achieve solubility in previous reports have included the use of Hepes buffer and ultrasound baths [ 19 , 20 ] with a resultant maximum 50% solubility. After numerous attempts at solubility for 17P in our studies, the most appropriate vehicle was a combination of DMSO and ethanol. While both of these compounds may exert an effect on uterine contractility, the maximal bath concentrations of both solvents was 0.83% and 0.28% respectively, which is acceptable for in vitro studies of this nature. In addition, control experiments, with PSS only, and PSS plus vehicle, were simultaneously run, to clearly evaluate any potential effect of vehicle. It is however apparent from our results, that addition of vehicle only did exert a significant relaxant effect on myometrium obtained during pregnancy, but no effect of vehicle was observed in myometrium obtained in the non-pregnant state. For both tissues, addition of 17P did not alter contractile activity. While the vehicle effects observed were sizeable, the experiment design, the numbers of patients recruited, and the reproducibility of the results, clearly indicate that 17P does not exert a direct effect on human myometrial contractility. In addition, while the experiments described here investigated the effects of 17P added cumulatively, separate experiments (data not shown) using a single dose (the highest dose) revealed similar results i.e. there was no evidence of tachyphylaxis over the duration of the experiments. There are some limitations to the methodology used in our study. The tissue biopsies from the non-pregnant uterus were obtained from the body of the uterus, while those obtained at the time of caesarean section were excised from the lower uterine segment. The results were similar for both tissue types, and there is reasonable evidence to suggest that the functional characteristics of lower and upper uterine segment myometrium are similar [ 21 ]. At present there are no data pertaining to differential progesterone receptor expression levels between upper and lower uterine segments. There are also obvious ethical constraints in obtaining biopsies from the upper segment of the uterus at cesarean section, and it is not feasible to dissect strips with certainty from the lower segment of the non-pregnant uterus. Secondly, the tissue samples obtained during pregnancy were all recruited from women at term. It could be that preterm myometrium displays a different responsiveness. Finally, in vivo administration of 17P could potentially result in the formation of a metabolite with a more efficient uterorelaxant effect. Conclusions Whether 17P injections become incorporated for routine use in the management of preterm labor remains to be seen. Questions surrounding its true benefits, the associated difficulties in terms of vehicle solubility and placebo, and the mechanism of action, remain. Our findings highlight the solubility problems for scientific evaluation as occurred in clinical studies. Whatever the possible mechanism of action of 17P in reducing the incidence and adverse sequelae of preterm labor, this study demonstrates that it does not exert a direct relaxant effect, unlike other conventional methods of tocolysis investigated, and raised the likeliehood of a genomic effect secondary to long term administration during pregnancy. Authors' contributions DJS performed the experiments and wrote the manuscript. MWO'R performed the experiments. AMF analysed the data and wrote the manuscript. JJM designed, supervised the study and wrote the manuscript. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539291.xml

509420

Corneal topographic changes following retinal surgery

Background To study the effect of retinal/ vitreoretinal surgeries on corneal elevations. Methods Patients who underwent retinal/ vitreoretinal surgeries were divided into 3 groups. Scleral buckling was performed in 11 eyes (Group 1). In 8 (25%) eyes, vitreoretinal surgery was performed along with scleral buckling (Group 2). In 12 eyes, pars plana vitrectomy was performed for vitreous hemorrhage (Group 3). An encircling element was used in all the eyes. The parameters evaluated were best-corrected visual acuity (BCVA), change in axial length, and corneal topographic changes on Orbscan topography system II, preoperative and at 12 weeks following surgery. Results There was a statistically significant increase in anterior corneal elevation in all the three groups after surgery (p = 0.003, p = 0.008 & p = 0.003 respectively). The increase in posterior corneal elevation was highly significant in all the three groups after surgery (p = 0.0000, p = 0.0001 & p = 0.0001 respectively). The increase in the posterior corneal elevation was more than the increase in the anterior elevation and was significant statistically in all the three groups (group I: p = 0.02; group II: p = 0.01; group III: p = 0.008). Conclusions Retinal/ vitreoretinal surgeries cause a significant increase in the corneal elevations and have a greater effect on the posterior corneal surface.

Background Changes in corneal curvature and axial length have been reported following scleral buckling procedure using keratometer [ 1 - 7 ]. Videokeratography has also been used to assess the corneal surface changes after buckling procedure[ 8 ]. All these studies have reported a change in the corneal curvature in its anterior surface. However, no study has been performed to evaluate the posterior corneal topographic changes with scanning slit topography system following retinal surgeries. Hence, we made an attempt to study the effect of various retinal surgeries on posterior corneal surface using Orbscan II topography system. Methods A prospective study was performed by enrolling patients admitted in the Retina Services of Rajendra Prasad Centre for Ophthalmic Sciences, New Delhi for retinal/ vitreoretinal surgeries. Thirty one eyes of 31 patients who underwent retinal/ vitreoretinal surgery and came for regular follow-up as per schedule during the period between December 2001 and October 2002 were included in the study. Only those patients were included who had not undergone any previous ocular surgery & did not have any corneal pathology. An informed consent was obtained from all the patients. The patients were evaluated preoperatively on parameters of best-corrected visual acuity (BCVA), axial length (AL) measured by Ultrasound A-scan instrument (Sonomed, Inc., NY) and detailed corneal examination on slit-lamp biomicroscope and on Orbscan topography system II (Bausch and Lomb, Salt Lake City, Utah). The parameters that were evaluated by Orbscan topography system were anterior elevation, posterior elevation and simulated keratometry. All surgeries were performed by a single surgeon (LV) under local anesthesia by peribulbar injection of 6 ml of 2% xylocaine and 2 ml of 0.5% bupivacaine. Eyes with fresh retinal detachment with clear media and absence of advanced proliferative vitreoretinopathy underwent scleral buckling procedure (Group I, n = 11). In all these eyes, the break/ s were localized, cryotherapy was performed and subretinal fluid was drained. Only circumferential buckle of silicone of style 276 (Labtician Ophthalmics, Inc., Oakville, Canada) was used and radial buckle or sponge was not used in any eye. The size of the buckle was 90° to 360° depending upon the requirement in individual cases. Encircling element of style 240 (Labtician Ophthalmics, Inc., Oakville, Canada) was used in all the eyes undergoing scleral buckling. In eyes with associated vitreous hemorrhage or advanced proliferative vitreoretinopathy changes along with retinal detachment, vitreoretinal surgery was performed along with scleral buckling (Group II, n = 8). Pars plana vitrectomy was performed and vitreoretinal membranes were removed either by peeling or by segmentation or delamination. Air fluid exchange was performed followed by Air-Silicone oil exchange. In eyes with only vitreous hemorrhage without the presence of retinal detachment, only pars plana vitrectomy was performed (Group III, n = 12). An encircling element of style 240 (Labtician Ophthalmics, Inc., Oakville, Canada) was used in all these eyes to counter the anterior traction that could not be fully released by vitrectomy in order to avoid lens damage. The intraoperative details including the nature of surgery, size of the buckle, encircling element, drainage of subretinal fluid, vitrectomy & use of silicone oil or gas, were noted. Postoperative treatment included topical ciprofloxacin 0.3% QID, topical dexamethasone 0.1% QID and topical Homatropine 2% QID. The patients were evaluated at 12 weeks following surgery on similar (preoperative) parameters. Statistical analysis The data of all the patients were managed on an excel spreadsheet. All the entries were checked for any possible keyboard error. Preoperative and postoperative measurements in the three retinal surgery groups were summarized by mean and standard deviation. Changes following surgery within each group were assessed using paired 't' test. Preoperative and postoperative values in the three groups were compared using one way analysis of variance (ANOVA), followed by bon ferroni correction for multiple comparison. For the three retinal surgery groups, median was computed for increase in various parameters due to surgery. Kruskal Wallis one way analysis of variance was used to compare median increase in the three groups. STATA 7.0 statistical software was used for data analysis. In this study, p-value smaller than 0.05 was considered as statistically significant. Results The mean age of the patients was 45.96 ± 15.17 (range: 18–78) years and majority (83.87%) of the patients (N = 31) were males. Of these, right eye was operated in 19 (61.29%) patients. The mean preoperative best-corrected visual acuity (BCVA) was hand motion close to face in 26 eyes; counting finger near to face in 4 eyes and 1/60 on snellen's acuity chart in 1 eye. The mean decimal postoperative BCVA was 0.20 ± 0.12 at 12 weeks follow-up after surgery. The mean preoperative anterior corneal elevation as recorded by Orbscan II topography system in group I was 0.006 ± 0.007 mm, which increased to 0.024 ± 0.013 mm at 12 weeks after surgery (p= 0.003). In group II, it increased from 0.009 ± 0.006 mm preoperatively to 0.021 ± 0.010 mm at 12 weeks (p= 0.008) and in group III, it increased from 0.003 ± 0.004 mm preoperatively to 0.012 ± 0.007 mm at 12 weeks follow-up (p = 0.003). On comparative evaluation between the groups, the change in anterior corneal elevation was significant between group I and III (p = 0.04). The mean posterior elevation in group I increased from a preoperative value of 0.016 ± 0.010 mm to 0.043 ± 0.007 mm at 12 weeks after surgery (p= 0.0000). In group II, it increased from 0.014 ± 0.006 mm preoperatively to 0.043 ± 0.007 mm at 12 weeks (p= 0.0001) and in group III it increased from a preoperative value of 0.012 ± 0.005 mm to 0.029 ± 0.006 mm at 12 weeks after surgery (p = 0.0001). A comparative analysis between the groups indicated that the increase in posterior corneal elevation between groups I & III and groups II & III were found to be highly significant (I vs III: p= 0.001; II vs III: p= 0.001). Again, the increase in the posterior corneal elevation was greater than the increase in the anterior elevation in all the 3 groups and on comparative evaluation, the difference in the increase in posterior and anterior elevation was significant statistically in each group (group I: p = 0.02; group II: p = 0.01; group III: p = 0.008). The mean corneal astigmatism in group I increased from 0.89 ± 0.54D preoperatively to 2.50 ± 1.39D at 12 weeks follow-up (p= 0.004). In group II, the average corneal astigmatism increased from 0.87 ± 0.30D to 3.38 ± 2.15D at 12 weeks (p= 0.01) and in group III, the mean preoperative and postoperative corneal astigmatism was 0.85 ± 0.55D and 1.37 ± 0.87D respectively (p= 0.02). A comparative analysis of the change in corneal astigmatism following surgery between groups II & III was significant statistically (p= 0.02). The mean preoperative axial length in group I was 23.27 ± 0.79 mm which increased to 23.98 ± 0.76 mm at 12 weeks after surgery (p= 0.009). The mean preoperative and postoperative (12 weeks follow-up) axial length in group II were 23.92 ± 1.32 mm and 25.94 ± 2.96 mm respectively (p= 0.03). The mean preoperative and postoperative axial length in group III were 22.69 ± 0.87 mm and 22.71 ± 0.83 mm respectively (p= 0.79). Comparative analysis of increase in axial length following surgery between groups I & III and groups II & III were found to be significant statistically (I & III: p = 0.003; II & III: p = 0.003). Discussion Retinal surgery with or without the use of encircling and buckling elements for external tamponade can alter the shape of the globe. This may cause changes in the refractive status of the eye. Scleral buckling is known to cause a change in the shape of the sclera and can cause induced refractive changes, including astigmatic and nonastigmatic changes [ 5 - 10 ]. We have used Orbscan slit scanning system II to evaluate the corneal topographic changes following retinal/ vitreoretinal surgeries. The data accumulated by Orbscan may be limited by factors such as the accuracy of the system which is ± 20 μm, the measurement noise which leads to both positive and negative difference in the height of the posterior corneal surface and the necessity of aligning the posterior surface before and after surgery which may be a source of error[ 11 , 12 ]. However, this is the best tool available to study the posterior corneal elevation. In the present study, there was a significant increase in both anterior and posterior corneal elevation as detected by scanning slit topography (Orbscan II topography system) following surgery. This increase in the anterior and posterior corneal elevation is probably due to the use of encircling element and/ or buckle in retinal surgeries resulting in corneal steepening. We noted that the change in the posterior elevation was more significant than anterior elevation. A comparative analysis between the three groups indicated that there was no significant difference in the anterior elevation; however the posterior elevation was significantly more in eyes with buckle. It is possible that the buckle and the encircling element have a greater effect on the posterior corneal surface. The increase in the anterior and posterior corneal elevation might be one of the contributing factors for the non improvement of visual acuity in an eye that has undergone retinal/ vitreoretinal surgery. This change might escape detection by routine videokeratography. The anterior protrusion of the corneal back surface induces an increase in the negative power of the corneal surface. Assessing the corneal surface by keratometry or placido disc videokeratography may provide inadequate information regarding refractive change caused by corneal surface alteration that results in retinal/ vitreoretinal surgery. There was an increase in mean corneal astigmatism following surgery in all the groups. This is due to the effect of encircling element or buckle on the corneal surface. Studies have reported that induced astigmatism has been associated with radial scleral buckles[ 5 , 13 ], circumferential buckles[ 14 ], medial rectus disinsertion[ 15 ], anterior location of the scleral buckle[ 13 ], and use of sponge material rather than hard silicone[ 1 ]. In our study, a comparative analysis between the groups indicated that buckle causes more astigmatic changes than encircling element. In retinal/ vitreoretinal surgeries, the encircling band creates a circular indentation of the eye, thereby increasing its anteroposterior axial length; the myopic shift may be upto 3 diopters (D)[ 9 , 10 , 13 , 16 ]. An increase in axial length by 0.54 mm[ 9 ] and 1.7 mm[ 14 ] has been reported following scleral buckling in two studies. In the present study, there was an increase in the mean axial length of the eyes in all the three groups (Table 1). This increase in axial length may be attributed to an anteroposterior elongation of the eyeball secondary to the transverse compression by the buckle and/ or an encircling element. The postoperative increase in axial length was found to be more pronounced in eyes with buckle. Buckle being wider and thicker results in greater indentation and thereby a greater increase in axial length than encircling element. Conclusions Retinal/ vitreoretinal surgeries result in an increase in the elevation of the corneal surfaces. These changes are more pronounced on posterior corneal surface. Declaration of competing interest None declared. Individual contribution of authors RS designed the study and performed the data collection. NS wrote the manuscript. LKV performed the retinal surgeries. RMP performed the statistical analysis and RBV followed up the patients. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC509420.xml

509421

Brain choline concentrations may not be altered in euthymic bipolar disorder patients chronically treated with either lithium or sodium valproate

Background It has been suggested that lithium increases choline concentrations, although previous human studies examining this possibility using 1 H magnetic resonance spectroscopy ( 1 H MRS) have had mixed results: some found increases while most found no differences. Methods The present study utilized 1 H MRS, in a 3 T scanner to examine the effects of both lithium and sodium valproate upon choline concentrations in treated euthymic bipolar patients utilizing two different methodologies. In the first part of the study healthy controls (n = 18) were compared with euthymic Bipolar Disorder patients (Type I and Type II) who were taking either lithium (n = 14) or sodium valproate (n = 11), and temporal lobe choline/creatine (Cho/Cr) ratios were determined. In the second part we examined a separate group of euthymic Bipolar Disorder Type I patients taking sodium valproate (n = 9) and compared these to controls (n = 11). Here we measured the absolute concentrations of choline in both temporal and frontal lobes. Results The results from the first part of the study showed that bipolar patients chronically treated with both lithium and sodium valproate had significantly reduced temporal lobe Cho/Cr ratios. In contrast, in the second part of the study, there were no effects of sodium valproate on either absolute choline concentrations or on Cho/Cr ratios in either temporal or frontal lobes. Conclusions These findings suggest that measuring Cho/Cr ratios may not accurately reflect brain choline concentrations. In addition, the results do not support previous suggestions that either lithium or valproate increases choline concentrations in bipolar patients.

Background Bipolar disorder is a chronic severe mental illness affecting approximately 1% of the adult population. The most widely used mood stabilizer for this condition is lithium [ 1 ], although the exact mechanism by which it is clinically effective remains undetermined. One suggestion is that it acts via effects on choline metabolism. This is based upon findings that lithium can inhibit the membrane transport of choline in both animals [ 2 ], and human post-mortem brain tissue [ 3 ]. It also increases the accumulation of erythrocyte choline in lithium-treated patients [ 4 - 7 ]. Also of note is that choline concentrations increase significantly in rats following electroconvulsive shock [ 8 ]. Based upon this data choline has been used to treat mania in a some small pilot studies [ 9 ], with one open label study reporting that choline augmentation of lithium treatment helped rapid-cyclers [ 10 ]. Patients treated with choline also had increased basal ganglia concentrations of choline, suggesting that externally administered choline could alter brain concentrations [ 11 , 12 ]. The most appropriate method to measure brain choline concentrations in vivo utilizes proton magnetic resonance spectroscopy ( 1 H-MRS). Previous studies of bipolar patients utilizing this methodology have had mixed findings. Overall, while some studies have suggested there may be increased choline concentrations in specific situations [ 13 - 18 ], more have found no changes [ 19 - 27 ], and one found a trend towards a decrease in concentrations [ 28 ]. In both patients and volunteers lithium also doesn't appear to alter choline/creatine peak ratios concentrations [ 29 , 30 ]. Nonetheless, two reviews concluded that the evidence to date suggests that lithium increases brain choline concentrations [ 31 , 32 ], although as noted in these reviews previous studies have varied considerably in terms of patient populations, brain region studied, medications administered, and MRS methodology. Many studies have also examined differing patients (Type I and Type II) in differing mood states (mixed, depressed, manic, and euthymic). This may partially explain the varied results. Sodium valproate is also widely used as a mood stabilizer, both alone and in combination with lithium [ 33 ]. To date there have been few studies which have examined the effects of sodium valproate on choline concentrations or activity. An in-vitro study suggested that valproate may inhibit choline acetyltransferase activity [ 34 ]. In one study 9 patients taking either lithium or valproate were examined [ 35 ], and increased Cho/Cr ratios were seen in the bipolar patients compared to controls. There were no differences between the lithium and valproate treatment groups, although the sample sizes were small. However, another study in epilepsy patients treated with valproate found no changes in choline concentrations [ 36 ]. Nonetheless, given the lack of studies to date, the possibility that valproate and lithium may both increase choline concentrations warrants further investigation. Most of the previous studies have examined Cho/Cr ratios. However, it should be noted that the "choline" resonance peak seen in 1 H-MRS spectra is composed primarily of phosphocholine and glycerophosphocholine, along with free choline, acetylcholine, and cytidine diphosphate choline. Also, we have shown in animal studies that both lithium and valproate can both decrease creatine concentrations [ 37 ]. Therefore, when using Cho/Cr ratios it is not possible to be certain that any changes in this peak represent changes in brain choline concentrations. We were therefore interested to determine if there were any differences in results when using different methodologies, and more specifically to determine if studies using a ratio methodology may have different results from studies utilizing metabolite concentrations. Methods In the first part of the study patients taking either lithium or valproate were examined using the Cho/Cr ratio method, and both Bipolar Type I and Bipolar Type II patients were included who could also be taking other medications. In the second part of this study only Bipolar Type I patients on valproate monotherapy were included, and quantification of choline concentrations was made. Some of the data from the first part of this study has been reported previously [ 38 ]. Subjects and Study Design All subjects gave full informed consent, and both studies were approved by the ethics committee at the University of Alberta. Healthy controls were examined using a detailed, but non-standardized, psychiatric interview. They were excluded if there was any personal history, or immediate family history, of psychiatric disorder. For patients, diagnoses were made using DSM-IV criteria for Bipolar Disorder Type I or Type II following detailed psychiatric interview, with additional information being available in almost all cases from long-term psychiatric clinic records. They also had to be taking a dose of either lithium or valproate which maintained their blood levels within the ranges of 0.4–1.2 mmol/l for lithium and 200–700 μmol/l for sodium valproate. Serum lithium and valproate levels were also measured on the day of MRS scanning. Other medications taken by the patient were noted. In the second part of the study the same criteria were used, except that only patients meeting diagnostic criteria for Bipolar Disorder Type I were included, and they had to be on sodium valproate monotherapy. This was done to examine Bipolar Type I patients in more detail, and to remove a possible confounding variable. All patients had to be euthymic for the previous 3 months, as determined by interviews with the patient, and additional interviews with their relatives and bipolar clinic records when available. MRS scans were carried out within 24 hours of this interview. Magnetic Resonance Spectroscopy Methodology For both studies magnetic resonance experiments were performed using a Magnex 3 T scanner with 80 cm bore equipped with actively shielded gradient, and spectrometer control was provided by an Surrey Medical Imaging System (SMIS) console. The subjects head was immobilized with a restraint system. Signal transmission and reception were achieved using a quadrature birdcage resonator for 1 H measurements. Part 1 - Magnetic Resonance Spectroscopy Initially, MRI data were acquired using gradient echo imaging sequences to produce multiple slice images along both coronal and transverse planes. This allowed registration of a 2 × 2 × 3 cm volume-of-interest (VOI) to be selected in the temporal lobe. 1 H MR spectra were acquired using the PRESS localization method [ 39 , 40 ], with TE = 32 ms, TR = 3 s, and with 128 averages. Baseline correction and deconvolution of the spectra was accomplished using the Peak Research (PERCH) spectrum analysis software package. The metabolite peaks of interest [choline (Cho) and creatine (Cr)] in each spectrum were fitted to a Gaussian line-shape for peak area estimation. To determine changes in choline concentrations we examined the Cho/Cr ratio. Figure 1 shows an individual 1 H MRS spectra in which all the major metabolite peaks can be seen. Figure 1 A typical 1 H-MRS spectrum of the human brain at 3.0 T. A number of metabolites can be seen. 1: creatine (methylene) + phosphocreatine, 2: glutamate + glutamine, 3: myo -inositol + glycine, 4: taurine, 5: total choline compounds , 6: creatine (methyl) + phosphocreatine, 7: N-acetylaspartate. Study 2 - Magnetic Resonance Spectroscopy To accurately quantify the brain concentration of creatine we used a 125 ml glass sphere containing a solution of 4 mmol creatine as an external standard. The PRESS sequence was used to acquire proton MRS data with TE1 = 25 msec, TE2 = 25 msec, TR = 3000 msec, and 128 scan averages. The MRS data were acquired from three 2 × 2 × 2 cm 3 voxels placed in the cortex of the left frontal lobe, the cortex of the left temporal lobe, and in the external standard solution. The average coordinates [ 41 , 42 ] of the centers of the two brain voxels were determined: x = 0.5 mm (SD = 1.6), y = 63.5 mm (SD = 12.1), z = -25.5 mm (SD = 4.2) in the frontal lobe, and x= 32.2 mm (SD = 6.3), y = 20.5 mm (SD = 3.9), z = 10.7 mm (SD = 2.6) in the temporal lobe. In order to measure T 1 and T 2 values of the metabolites in the brain and external standard solution, MRS data were collected with different TE values at a constant TR and different TR values at a constant TE both for the healthy volunteers and the patients and also from external standard solution [ 42 ]. However, due to these constraints, the fact that the two studies used different populations at different times, and the size of the external 125 ml container (which limited voxel size to 2 × 2 × 2 cm 3 ), it was not possible to exactly match the voxel size or location between the two studies. MRS Data Analysis For quantitative measurement of brain metabolite concentrations we used previously described methodology [ 42 , 43 ]. In this, [Met] b , in millimoles per kg of wet brain, the CSF volume fraction, f csf , in the spectroscopic voxels must be corrected. Thus, brain metabolite concentrations were calculated as described in the following equation: where V voxel is the volume of a 8 cm 3 spectroscopic voxel [ 43 ], and N b represents the number of metabolite molecules per unit voxel in brain. Statistical Analysis for both MRS studies Means ± SEM were used in the statistical analysis. Sex differences were analyzed using chi-squared, and age differences with ANOVA with post-hoc Tukey tests. The MRS data was analyzed using Student's unpaired t -test using a significance level of p < 0.05 comparing diagnostic groups (patients vs controls) in each brain region (frontal and temporal). Results Study 1 Subjects A total of 18 healthy controls, 14 bipolar patients taking lithium, and 11 bipolar patients taking valproate completed this study. Of the 14 bipolar patients taking lithium, 7 were Type I and 7 were Type II. In the valproate group, 7 were Type I and 4 were Type II. These groups were studied both separately and together, but as there were no statistically significant differences between the Type I and Type II patients, the results for both types are presented together. Of the 14 bipolar patients taking lithium 12 patients were taking other psychotropic medications: these were benzodiazepines (7 patients), antidepressants (5 patients), and antipsychotics (2 patients). Of the 11 patients taking sodium valproate 10 patients were taking other psychotropic medications: these were benzodiazepines (5 patients), antidepressants (5 patients), and antipsychotics (4 patients). The mean age for the lithium group was 40.43 ± 2.96 years, for the valproate group 35.47 ± 2.27 years, and for the control group was 31.35 ± 2.89 years. These differences were statistically significant (F = 3.68, df = 2, p = <0.05), which was attributable to the lithium group being significantly older than the control group (Tukey post hoc , p < 0.05). There were no gender differences within the groups: 10 females and 8 males in the control group (χ 2 = 0.167, df 1, p > 0.05), 5 females and 9 males in the lithium group (χ 2 = 1.143, df 1, p > 0.05), and 6 females and 5 males in the valproate group (χ 2 = 0.474, df 1, p > 0.05). Mean serum lithium levels were 0.79 ± 0.06 mmol/l, and the range was 0.46–1.08 mmol/l. The mean serum valproate levels were 508 ± 42 μmol/l, and the range was 210–912 μmol/l. MRS Data 1 H MRS We utilized the ratio of the choline peak to creatine peak (Cho/Cr) as a primary correlate of Choline concentrations. This result has been reported briefly in a previous publication [ 38 ]. The mean Cho/Cr ratio with this measure was 1.46 ± 0.04 for controls, 1.18 ± 0.07 for lithium-treated patients, and 1.12 ± 0.08 for valproate-treated patients. These were statistically significant, with a reduction in ratios occurring in both the control vs. lithium comparison (t = 3.628, df = 30, p = 0.001) and the control vs. valproate comparison (t = 4.248, df = 27, p = 0.002). Study 2 Subjects A total of 11 healthy controls and 9 Bipolar Type I patients taking valproate as monotherapy were entered into this study. The mean age for the control group was 37.3 ± 2.2 years, and for the valproate patients 42.4 ± 3.0 years. These differences were not statistically significant (F = 1.49, df = 1, p = 0.27). There were no gender differences within the groups: 7 females and 2 males in the valproate group and 5 females and 6 males in the control group (χ 2 = 0.474, df 1, p > 0.05). The mean serum valproate levels were 472 ± 36 μmol/l, and the range was 284–728 μmol/l. In the frontal lobe the mean choline concentration for the healthy controls was 2.21 ± 0.17 mmol/kg wet brain and for the patients was 2.38 ± 0.12 mmol/kg wet brain. In the temporal lobe the mean choline concentration for the healthy controls was 2.35 ± 0.14 mmol/kg wet brain and for the patients was 2.40 ± 0.19 mmol/kg wet brain. There were no statistically significant differences between the controls and patients in either the frontal (t = 0.78, df = 18, p = 0.44) or temporal (t = 0.203 df = 18, p = 0.84) lobes (Table 1 ). Table 1 Concentrations (mmol/kg wet brain) and ratios (Cho/Cre) in frontal and temporal lobes in healthy volunteers and in patients chronically treated with valproate (Study #2) Choline (Cho) Creatine (Cre) Cho/Cre Frontal Temporal Frontal Temporal Frontal Temporal Healthy Controls Age Sex 1 50 M 3.51 2.95 6.67 8.53 0.53 0.35 2 45 M 2.19 3.03 10.1 9.11 0.22 0.33 3 43 F 3.01 2.31 9.97 9.52 0.30 0.24 4 39 M 2.11 2.72 7.94 7.60 0.27 0.24 5 37 F 2.47 2.34 9.98 9.89 0.25 0.24 6 36 F 1.91 1.76 8.28 8.19 0.23 0.22 7 35 M 1.76 2.36 7.93 8.36 0.22 0.28 8 32 F 1.88 1.51 9.56 9.56 0.2 0.16 9 32 M 1.94 2.14 7.04 7.79 0.28 0.28 10 30 F 1.82 2.52 7.8 8.63 0.23 0.29 11 28 M 1.72 2.23 7.16 8.51 0.24 0.26 Mean 37.00 2.21 2.35 8.40 8.70 0.27 0.26 Valproate Treated Patients 1 58 F 2.72 2.1 9.16 10.13 0.30 0.21 2 50 M 2.61 3.42 8.17 10.53 0.32 0.33 3 49 F 2.03 1.79 8.56 7.48 0.24 0.24 4 48 F 2.44 1.88 9.93 8.19 0.25 0.23 5 36 M 2.60 2.53 7.84 7.51 0.33 0.34 6 35 F 2.07 2.77 9.26 10.39 0.22 0.27 7 35 F 2.78 1.89 8.35 9.79 0.33 0.19 8 34 F 1.76 2.93 7.26 8.01 0.24 0.37 9 34 F 2.43 2.27 7.75 7.23 0.31 0.31 Mean 42.11 2.38 2.40 8.48 8.81 0.28 0.28 The Cho/Cr ratios in the frontal lobes were 0.27 ± 0.028 in controls and 0.28 ± 0.015 in patients. In the temporal lobes the Cho/Cr ratios were 0.26 ± 0.021 in controls and 0.28 ± 0.016 in patients. There were no statistically significant differences between the controls and patients in either the frontal (t = 0.367, df = 18, p = 0.72) or temporal (t = 0.539, df = 18, p = 0.59) lobes (Table 1 ). Discussion The results from the present study vary considerably between the two sections utilizing differing methodologies. This is despite the fact that both studies were carried out by the same group on the same scanner with bipolar patients coming from the same patient pool. This strongly suggests that the methodology used to determine choline concentrations can considerably alter the results. In the first part of the study we found that both the lithium-treated and valproate-treated patients had significantly reduced Cho/Cr peak ratios compared to controls. This is similar to the findings from one previous study which also suggested that there may be a trend towards decreased choline in grey matter [ 28 ]. This study was a frontal lobe study that measured metabolite concentrations in a 1.5 T scanner in bipolar type I patients hospitalized for manic (n = 9) or mixed (n = 8) states. In this study most patients were being treated with valproate and an atypical antipsychotic. These findings, however, differ from those in the second part of the present study in which we found no differences in choline concentrations between valproate-treated patients and controls in either frontal or temporal lobes. This second part of the study was much better controlled in terms of the patients receiving valproate monotherapy, only including bipolar Type I patients, and in using an external choline solution to accurately quantify choline concentrations. This finding of a lack of change is also in keeping with most previous studies. Several studies which have also previously measured metabolite concentrations with 1.5 T scanners also found no changes. These include a study of the hippocampus in 15 euthymic bipolar type 1 patients, of whom 10 were taking either lithium or valproate [ 19 ], a study of basal ganglia in 8 rapid cycling patients on lithium [ 22 ], a study of the anterior cingulate in 10 bipolar children [ 23 ], and a study in frontal lobes of 23 euthymic bipolar patients of whom 13 were on lithium [ 25 ]. Several other studies have examined metabolite ratios, mostly in patients on lithium, and those also found no changes in choline concentrations [ 20 , 21 , 26 , 27 ]. In a study using metabolite ratios in bipolar children who were off medication for at least one week there was also no change in choline concentrations [ 24 ]. In a double-blind placebo-controlled human volunteer study before and after one week of lithium administration we also found no changes in cholinein 10 volunteers [ 30 ], which is similar to a patient study which compared 7 patients on lithium to 6 non-lithium treated controls and in which no differences were seen [ 29 ]. In contrast, animal studies have suggested that lithium may increase brain choline concentrations, and in lithium-treated patients it also increases the accumulation of choline within erythrocytes [ 4 - 7 ]. Nonetheless, 1 H-MRS studies in patients examining this possibility is mixed. To date 6 studies have suggested some support for this [ 13 - 18 ], but in none of these studies were metabolite concentrations measured, and most of the studies measured choline/creatine ratios [ 14 - 18 ], the other one measuring metabolite intensity/tissue volume [ 13 ]. The first study to examine brain choline in basal ganglia studied only 4 patients, all of whom were on lithium [ 18 ]. Another study examined 19 euthymic inpatients and found increased choline/creatine ratios in basal ganglia, but only 10 of these patients were receiving lithium [ 17 ]. The third study to report an increase in this ratio (in this case in the left subcortical region) was in a mixed group of patients receiving a wide range of medications [ 16 ]. Two other studies have reported increased choline concentrations, but only in limited circumstances. In one study in 11 bipolar children patients were examined before and after lithium administration [ 14 ]. There were no significant findings before or after lithium administration, although there was a trend towards increased choline/creatine ratios in the patients before lithium treatment. This latter finding does not suggest that in patients lithium significantly alters the choline/creatine ratio. The final study examined 15 euthymic males who were on either lithium or valproate [ 13 ]. This study found that thalamic choline concentrations, determined by measuring metabolite intensity/tissue volume ratios, were significantly increased only if the right and left hemisphere were compared separately, but not if they were compared together. It is also conceivable that both lithium and valproate may increase Choline concentrations, but that the differences were not large enough for us to detect, or that without lithium or valproate treatment patients would have lower Choline concentrations. The cross-sectional nature of this study does not allow this to be examined. It is also important to recognize other limitations of the present study. Firstly, these MRS studies are not pre- and post-treatments, so may not accurately reflect changes that occur in individual patients. Secondly, part of the study used a ratio-method to assess choline concentrations, the limitations of which are increasingly clear (particularly since creatine concentrations may be altered by medication [ 37 ]). Thirdly, the sizes of all groups are small and it therefore possible that a larger study may have been fully powered to identify differences between groups. Fourthly, several patients in the first study (but not the second study) were on other drugs which may have affected the results of this study. Fifthly, we have not determined if age affects the results, and in the first part the groups were not all matched for age. In addition, the voxel locations were not the same in both studies due to the reasons discussed in the methodology section. Nonetheless, despite these limitations we believe the results add significantly to the literature in this under-researched area. We conclude that, taking all current evidence together including the findings from the present study, it is unlikely that either lithium or valproate significantly alter brain choline concentrations. However, given the large differences in patients populations, medications received, and MRS methodologies it is difficult to directly compare all these studies. In addition, the methodology used to measure choline concentrations can significantly alter the results. Future MRS studies in bipolar patients should, therefore, examine metabolite concentrations rather than a ratio of choline compared to other metabolites. Competing interests None declared.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC509421.xml

521178

Human MicroRNA Targets

MicroRNAs (miRNAs) interact with target mRNAs at specific sites to induce cleavage of the message or inhibit translation. The specific function of most mammalian miRNAs is unknown. We have predicted target sites on the 3′ untranslated regions of human gene transcripts for all currently known 218 mammalian miRNAs to facilitate focused experiments. We report about 2,000 human genes with miRNA target sites conserved in mammals and about 250 human genes conserved as targets between mammals and fish. The prediction algorithm optimizes sequence complementarity using position-specific rules and relies on strict requirements of interspecies conservation. Experimental support for the validity of the method comes from known targets and from strong enrichment of predicted targets in mRNAs associated with the fragile X mental retardation protein in mammals. This is consistent with the hypothesis that miRNAs act as sequence-specific adaptors in the interaction of ribonuclear particles with translationally regulated messages. Overrepresented groups of targets include mRNAs coding for transcription factors, components of the miRNA machinery, and other proteins involved in translational regulation, as well as components of the ubiquitin machinery, representing novel feedback loops in gene regulation. Detailed information about target genes, target processes, and open-source software for target prediction (miRanda) is available at http://www.microrna.org . Our analysis suggests that miRNA genes, which are about 1% of all human genes, regulate protein production for 10% or more of all human genes.

Introduction The Functions of MicroRNAs In the past three years, several hundred novel genes encoding transcripts containing short double-stranded RNA hairpins, named microRNAs (miRNAs), were identified in plants and animals ( Lee et al. 1993 ; Reinhart et al. 2000 , 2002 ; Lagos-Quintana et al. 2001 , 2002 , 2003 ; Lau et al. 2001 ; Lee and Ambros 2001 ; Llave et al. 2002a ; Mette et al. 2002 ; Mourelatos et al. 2002 ; Park et al. 2002 ; Ambros et al. 2003b ; Aravin et al. 2003 ; Brennecke et al. 2003 ; Dostie et al. 2003 ; Grad et al. 2003 ; Houbaviy et al. 2003 ; Lai et al. 2003 ; Lim et al. 2003a , 2003b ; Palatnik et al. 2003 ). More recently, miRNAs have also been identified in a large DNA virus, the Epstein Barr virus, and are likely to be found in other viruses ( Pfeffer et al. 2004 ). The cellular functions of most animal miRNAs are unknown. More than ten years after the discovery of the first miRNA gene, lin-4 ( Chalfie et al. 1981 ; Lee et al. 1993 ), we know that miRNA genes constitute about 1%–2% of the known genes in eukaryotes. Investigation of miRNA expression combined with genetic and molecular studies in Caenorhabditis elegans, Drosophila melanogaster, and Arabidopsis thaliana have identified the biological functions of several miRNAs (recent review, Bartel 2004 ). In C. elegans, lin-4 and let-7 were first discovered as key regulators of developmental timing in early larval developmental transitions ( Ambros 2000 ; Abrahante et al. 2003 ; Lin et al. 2003 ; Vella et al. 2004 ). More recently lsy-6 was shown to determine the left–right asymmetry of chemoreceptor expression ( Johnston and Hobert 2003 ). In D. melanogaster, miR-14 has a role in apoptosis and fat metabolism ( Xu et al. 2003 ) and the bantam miRNA targets the gene hid involved in apoptosis and growth control ( Brennecke et al. 2003 ). In mouse, miR-181a modulates hematopoietic differentiation ( Chen et al. 2004 ), and miR-196 directs the cleavage of Hox-B8 transcripts ( Yekta et al. 2004 ). miRNAs have specificity. In a range of organisms, miRNAs are differentially expressed in developmental stages, cell types, and tissues ( Lee and Ambros 2001 ; Lagos-Quintana et al. 2002 ; Sempere et al. 2004 ). In particular, differential expression has been observed in mammalian organs ( Lagos-Quintana et al. 2002 ; Krichevsky et al. 2003 ; Sempere et al. 2004 ) and embryonic stem cells ( Houbaviy et al. 2003 ). Estimates in worm show that there are approximately 1,000 molecules of miRNA per cell, with some cells exceeding 50,000 molecules ( Lim et al. 2003b ). This dynamic range of regulation of miRNA expression underscores the regulatory functional importance of miRNAs. The Mechanism of miRNA Action How do miRNAs pair with their target messages? miRNAs cause the translational repression or cleavage of target messages ( Doench and Sharp 2004 ). Some miRNAs may behave like small interfering RNAs (siRNAs) that direct mRNA cleavage between the nucleotide positions 10 and 11 in the siRNA:mRNA target duplex ( Tuschl et al. 1999 ; Zamore et al. 2000 ; Elbashir et al. 2001 ; Hutvágner and Zamore 2002a ; Llave et al. 2002b ; Martinez et al. 2002 ; Bartel 2004 ; Yekta et al. 2004 ). It appears that the extent of base pairing between the small RNA and the mRNA determines the balance between cleavage and degradation ( Hutvágner and Zamore 2002a ). Recent examples of cleavage of target messages are, in mouse, mir-196 guiding cleavage of Hox-B8 transcripts ( Yekta et al. 2004 ) and, in Epstein Barr virus, miR-BART2, a virus-encoded miRNA, guiding the cleavage of transcripts for virus DNA polymerase (gene BALF5 ) ( Pfeffer et al. 2004 ). While cleavage of mRNA is a straightforward process, the details of the mechanism of translational repression are unknown. The following rules for matches between miRNA and target messages have been deduced from a range of experiments. (1) Asymmetry: experimentally verified miRNA target sites indicate that the 5′ end of the miRNA tends to have more bases complementary to the target than its 3′ end. Loopouts in either the mRNA or the miRNA between positions 9 and 14 of the miRNA have been observed or deduced ( Brennecke et al. 2003 ; Johnston and Hobert 2003 ; Lin et al. 2003 ; Vella et al. 2004 ). Recent experiments show some correlation between the level of translational repression and the free energy of binding of the first eight nucleotides in the 5′ region of the miRNA ( Doench and Sharp 2004 ). However, confirmed miRNA:mRNA target pairs can have mismatches in this region ( Moss et al. 1997 ; Johnston and Hobert 2003 ). (2) G:U wobbles: wobble base pairs are less common in the 5′ end of a miRNA:mRNA duplex, and recent work shows a disproportionate penalty of G:U pairing relative to standard thermodynamic considerations ( Doench and Sharp 2004 ). (3) Cooperativity of binding: many miRNAs can bind to one gene ( Reinhart et al. 2000 ; Ambros 2003 ; Vella et al. 2004 ), and the target sites may overlap to some degree ( Doench and Sharp 2004 ). Given the overlap between the siRNA and miRNA pathways, it is reasonable to assume that rules of regulation in the siRNA pathway will partly apply to miRNA target recognition ( Hutvágner and Zamore 2002b ; Boutet et al. 2003 ; Doench et al. 2003 ). Lately, detailed characteristics associated with siRNA functionality were identified: low G/C content, a bias towards low internal stability at the3′ terminus, lack of inverted repeats, and strand base preferences (positions 3, 10, 13, and 19) ( Jackson et al. 2003 ; Reynolds et al. 2004 ). These observations may provide clues for better quantitative description of miRNA:mRNA interaction. Regions adjacent or near to the target site can be important for miRNA specificity. In lin-41, a 27-nucleotide (nt) intervening sequence between two consecutive let-7 sites is necessary for its regulation ( Vella et al. 2004 ). Because of lack of conservation of this 27-nt intervening sequence in C. briggsae, incorporation of a corresponding rule is premature. Maturation of miRNAs and Assembly in RNA-Induced Silencing Complex miRNAs are transcribed as longer precursors, termed pre-miRNAs ( Lee et al. 2002 ), sometimes in clusters and frequently in introns (25% of human miRNAs; Table S1 ). Upon transcription, miRNAs undergo nuclear cleavage by the RNase III endonuclease Drosha, producing the 60–70-nt stem-loop precursor miRNA (pre-miRNA) with a 5′ phosphate and a 2-nt 3′ overhang ( Lee et al. 2003 ). The pre-miRNA is subsequently transported across the nuclear membrane, dependent on the protein exportin 5 ( Lund et al. 2003 ; Yi et al. 2003 ). Dicer cleaves the pre-miRNA in the cytoplasm about two helical turns away from the ends of the pre-miRNA stem loop, producing double-stranded RNA. A helicase unwinds the cleaved double-stranded RNA in a strand-specific direction ( Khvorova et al. 2003 ; Schwarz et al. 2003 ). One of the unwound strands is subsequently incorporated into a ribonuclear particle (RNP) complex, RNA-induced silencing complex (RISC) ( Hutvágner and Zamore 2002a ; Martinez et al. 2002 ). Every RISC contains a member of the Argonaute protein family, which tightly binds the RNA in the complex ( Hammond et al. 2001 ; Hutvágner and Zamore 2002a ; Martinez et al. 2002 ; Mourelatos et al. 2002 ). There are at least eight members of the Argonaute family in mammals ( Sasaki et al. 2003 ), and only a small subset has been functionally characterized. The Argonautes and Dicer bind single-stranded RNA via their PAZ domains ( Lingel et al. 2003 ; Sasaki et al. 2003 ; Song et al. 2003 ; Yan et al. 2003 ), and the known structures of the PAZ domains may have implications for prediction of miRNA targets ( Lingel et al. 2003 ; Song et al. 2003 ; Yan et al. 2003 ). Association of mRNAs and miRNAs with Fragile X Mental Retardation Protein Among the prime candidates for miRNA control are the genes that are posttranscriptionally regulated. The mRNA-binding protein fragile X mental retardation protein (FMRP) is involved in the regulation of local protein synthesis ( Antar and Bassell 2003 ) and binds 4% of mRNAs expressed in the rat brain, as tested in vitro ( Brown et al. 2001 ). The loss of function of FMRP causes fragile X syndrome, the most prevalent form of mental retardation (one in every 2,000 children). Over the past three years a number of different groups have identified in vivo mRNA cargoes of FMRP. The Warren and Darnell laboratories have identified ligands by co-immunoprecipitation followed by microarray analysis, complemented by extraction of polyribosomal fractions ( Brown et al. 2001 ). They discovered that FMRP and one of its three RNA-binding domains specifically binds to G-rich quartet motifs ( Brown et al. 2001 ; Darnell et al. 2001 ; Denman 2003 ; Miyashiro et al. 2003 ). Three more studies found that mRNAs containing U-rich motifs bind recombinant FMRP in vitro and associate with FMRP-containing mRNPs in vivo ( Chen et al. 2003 ; Denman 2003 ). Lastly, antibody-positioned RNA amplification as a primary screen followed by traditional methods identified over 80 FMRP-regulated mRNAs, with a combination of G-quartet and U-rich motifs in their mRNA sequences ( Miyashiro et al. 2003 ). Independently, FMRP has been shown to be associated with RISC components and miRNAs ( Jin et al. 2004 ). The Drosophila homolog of FMRP (FXR) and the Vasa intronic gene were identified as components of RISC ( Caudy et al. 2002 ). More recent studies have proved that mammalian FMRP interacts with miRNAs and with the components of the miRNA pathways including Dicer and the mammalian orthologs of Argonaute (AGO) 1 ( Ishizuka et al. 2002 ; Jin et al. 2004 ). Given the association of FMRP with Argonaute-containing complexes, we propose and investigate the hypothesis that the cargoes carried by FMRP are also miRNA targets, and we derive hypotheses of specific pairing interactions. Here, we predict miRNA targets in five vertebrate genomes as a way of facilitating experiments and exploring a number of open questions. What proportion of all genes is regulated by miRNAs? How many genes are regulated by each miRNA? Are specific cellular processes targeted by specific miRNAs or by miRNAs in general? What is the extent of cooperativity in miRNA:mRNA binding? Results Prediction of miRNA Targets Using currently known mammalian miRNA sequences, we scanned 3′ untranslated regions (UTRs) from the human (Homo sapiens), mouse (Mus musculus), and rat (Rattus norvegicus) genomes for potential target sites. The scanning algorithm was based on sequence complementarity between the mature miRNA and the target site, binding energy of the miRNA–target duplex, and evolutionary conservation of the target site sequence and target position in aligned UTRs of homologous genes. We identified as conserved across mammals a total of 2,273 target genes with more than one target site at 90% conservation of target site sequence (Tables S2 and S3 ) and 660 target genes at 100% conservation. We also scanned the zebrafish (Danio rerio) and fugu (Fugu rubripes) fish genomes for potential targets using known and predicted miRNAs ( Figure 1 ; Tables S4 and S5 ) and identified 1,578 target genes with two or more conserved miRNA sites between the two fish species. Figure 1 Target Prediction Pipeline for miRNA Targets in Vertebrates The mammalian (human, mouse, and rat) and fish (zebra and fugu) 3′ UTRs were first scanned for miRNA target sites using position-specific rules of sequence complementarity. Next, aligned UTRs of orthologous genes were used to check for conservation of miRNA–target relationships (“target conservation”) between mammalian genomes and, separately, between fish genomes. The main results (bottom) are the conserved mammalian and conserved fish targets, for each miRNA, as well as a smaller set of super-conserved vertebrate targets. In addition to the analysis of 3′ UTRs, we also scanned all protein-coding regions for high-scoring miRNA target sites. For convenience, these results are reported both as hits in cDNAs (coding plus noncoding; Table S6 ) and as hits in coding regions ( Table S7 ), with cross-references in the UTR target tables (number of hits in the coding region for each UTR in Tables S2 and S3 ). The algorithm and cutoff parameters were chosen to provide a flexible mechanism for position-specific constraints and to capture what is currently known about experimentally verified miRNA target sites: (1) nonuniform distribution of the number of sequence-complementary target sites for different miRNAs; (2) 5′–3′ asymmetry (the complementary pairing of about ten nucleotides at the 5′ end is more important than that of the ten nucleotides at the 3′ end [ Doench and Sharp 2004 ], and the matches near the 3′ end can to a limited extent compensate for weaker 5′ binding); and (3) influence of G:U wobbles on binding ( Doench and Sharp 2004 ). In choosing these parameters, we drew on experience from careful analysis of target predictions in Drosophila ( Enright et al. 2003 ) as well as proposed human targets of virus-encoded miRNAs ( Pfeffer et al. 2004 ). To facilitate evaluation of predicted targets and design of new experiments, we provide methods and results in a convenient and transparent form. We make the miRanda software freely available under an open-source license, so that researchers can adjust the algorithm, numerical parameters, and position-specific rules. We also provide web resources, including a viewer for browsing potential target sites, conserved with or without positional constraints, on aligned UTRs, with periodic updates ( http://www.microrna.org , as well as links to these targets from the miRNA registry site RFAM ( http://www.sanger.ac.uk ; Griffiths-Jones 2004 ). We provide both high-scoring targets, as strong candidates for validation experiments, and lower-scoring targets, which may have a role in broader background regulation of protein dose. Expression information (see Table S3 ) for miRNAs and mRNAs provides an additional filter for validation experiments, in addition ranking target sites by complementarity and evolutionary conservation. Validation of Target Predictions Only a small number of target sites of target genes regulated by miRNAs have been experimentally verified, so we sought direct and indirect evidence to help validate or invalidate the proposed set of mammalian targets. (1) We compared predicted targets with experimentally verified targets in mammals, C. elegans, and D. melanogaster , as well as their mammalian homologs. (2) We compared predicted target numbers from real and shuffled miRNA sequences and estimated the rate of false-positive predictions. (3) We assessed the enrichment of miRNA targets in mRNAs that are known cargoes of FMRP, an RNA-binding protein known to be involved in translational regulation. Agreement with known targets We previously used known miRNA sites for the let- 7 and lin-4 miRNAs in Drosophila to develop the target prediction method and check for consistency ( Enright et al. 2003 ). More recent experimental target identification provides independent control data. Recent work in C. elegans ( Vella et al. 2004 ) has narrowed the originally reported list of six target sites for let-7 in the UTR of lin-41 down to three elements, two target sites, and a 27-nt intervening sequence (a possible binding site for another factor). The surviving two target sites have high alignment scores, S = 115 and S = 110, while the other four sites are below threshold ( Enright et al. 2003 ), fully consistent with the experimental results. As one of the confirmed sites has a single-residue bulge, target prediction methods that require a perfect run of base pairs near the 5′ end of the miRNA would not detect it, while our method does. lsy-6, a recently experimentally identified miRNA in C. elegans, controls left–right neuronal asymmetry via cog-1, an Nkx-type homeobox gene; the cog-1 gene has a target site in its 3′ UTR, which also has a high score ( S = 125) and passes the conservation filter. Experiments in D. melanogaster have identified six new miRNA–target gene pairs: miR-7 targets the notch signaling genes HLHm3, HLHm4, and hairy, and miR-2b targets the genes reaper , grim, and sickle ( Stark et al. 2003 ). Consistent with these experiments, our target predictions in D. melanogaster ( Enright et al. 2003 ) ranked HLHm3, hairy, and HLHm4 at positions 1, 3, and 7, respectively, in the list of 143 target genes for miR-7 ( Enright et al. 2003 ). Similarly, our predictions ranked reaper, grim, and sickle at positions 3, 11, and 19, respectively, among the other 120 predicted target genes for miR-2c . We also predicted miR-6 to target this group of pro-apoptotic genes, with sites that have lower scores than the miR-2 family but are conserved in D. pseudoobscura. Unfortunately, one cannot in general use these predicted and then validated target sites ( Stark et al. 2003 ) for the derivation of new prediction rules, as the set of targets tested is limited to the type predicted and is not exhaustive. Indirect validation comes from the prediction that mammalian orthologs of some of the known miRNA targets in C. elegans and D. melanogaster are miRNA targets. An example is the proposed conservation of the miRNA–target relationship lin-4:lin-28 (we use the notation miRNA:mRNA for a miRNA–target pair), first discovered in worm ( Moss and Tang 2003 ): we detect target sites in human lin-28 for the lin-4 miRNA homolog miR-125 . We also confirm the human analog of a let-7:lin-28 relation predicted in C. elegans ( Reinhart et al. 2000 ). In summary, the predicted target sites on human lin-28 are miR-125 (1 site), let-7b (2 sites; Moss and Tang 2003 ), miR-98 (2 sites), and miR-351 (1 site). Another known lin-4 and let-7 target in C. elegans is lin-41 . The human homolog of lin-41 (sequence provided by F. J. Slack, personal communication) and another closely related gene (encoding Tripartite motif protein 2) are predicted as high-ranking targets of let-7 and miR-125 (the human homolog of lin-4 ) (see Tables S2 and S3 ). Another known instance of miRNA target regulation in worms is the regulation of cog-1 by the lsy-6 miRNA ( Johnston and Hobert 2003 ). Although there is no obvious homolog of lsy-6 in mammals, the vertebrate homolog of the target gene cog-1, nkx-6.1, is a conserved target for five different miRNAs in our predictions (see Table S2 ). The comparison of our results with known targets shows that our method can detect most (but not all) known target sites and target genes at reasonably high rank. However, given the small number of experimentally verified miRNA–target pairs, additional validation tests are desirable, such as statistical tests using randomization of miRNA sequences to estimate false positives. Estimate of false positives As a computational control of the validity of the prediction method, one can perform a statistical test that attempts to estimate the probability that a predicted site is incorrect. Here, a “false positive” is a predicted target site of a real miRNA on a real mRNA that has passed all relevant thresholds but is incorrect in that it is not biologically meaningful. The statement “not biologically meaningful” is rarely clearly defined, but can reasonably be taken to mean that no functionally effective miRNA:mRNA interaction occurs under conditions of co-expression at physiological concentration, where “functionally effective” is defined in terms of detectable changes of phenotypic attributes. Technically, an estimate of the false-positive rate can be obtained by computing (directly or via randomization) the background distribution of scores for biologically non-meaningful miRNA target sites and then deriving the probability that a non-meaningful target site passes all score thresholds, i.e., for a single aggregate score, that the incorrect site has a score T > T c , where T c is a fixed threshold that may be, in general, different for each miRNA. We chose to estimate the background distribution using shuffled miRNAs obtained by swapping randomly selected pairs of bases of each given miRNA 1,000 times, keeping the nucleotide composition constant. The shuffled miRNA sequences were scanned against human, mouse, and rat 3′ UTR sequences exactly as for the prediction procedure for real miRNA sequences. In the procedure, a miRNA:mRNA match site is predicted to be a target site if it passes three thresholds, S > S c for match score, |Δ G | > |Δ G c | for free energy of duplex formation, and C > C c for conservation, where C reflects a binary evaluation of orthology of mRNAs, similarity of position of the site on the mRNA, and a threshold percentage of conserved residues in the two mRNA target sites. Finally, the predicted target sites for a set of shuffled miRNAs are counted and then averaged over a total of ten randomized runs. The percentage of false positives for target transcripts with more than two, three, and four sites is 39%, 30%, and 24%, respectively, using a non-permissive conservation threshold of 100% for target site sequences ( Figure 2 ). In addition, the false-positive rate for single sites with a score of more than 110 is approximately 35%. Figure 2 Distribution of Transcripts with Cooperativity of Target Sites and Estimated Number of False Positives Each bar reflects the number of human transcripts with a given number of target sites on their UTR. Estimated rate of false positives (e.g., 39% for ≥2 targets) is given by the number of target sites predicted using shuffled miRNAs processed in a way identical to real miRNAs, including the use of interspecies conservation filter. To provide a realistic estimate of false positives using randomization, the distribution of scores from random trials (“random-false”) should be similar to the distribution of incorrect (non-meaningful) hits from real trials (“real-false”). The difference between these two distributions is difficult to compute in principle, as very few validated correct predictions are known at present. For human sequences, without any conservation filter, we obtained a total of 2,538,431 predicted target sites for real miRNAs, and, for shuffled miRNAs, on average, 2,033,701 (± 82,172) target sites—a difference of 20%. This difference may be indicative of a biological signal in the raw score (S) and energy ( Δ G) calculated by the miRanda algorithm or may be due to different polynucleotide compositions of shuffled miRNAs compared to real miRNAs. Even if this difference represents a real effect, by far the most predictive criterion for accurate target detection is conservation of target sites across species, and not alignment scores or energies (20% compared to a factor of three, see Figure 2 ; Table S8 ). As a consequence, the current set of predicted targets rests heavily on the criterion of conservation of miRNA:mRNA match between different species. We believe this to be essentially true for all currently published target prediction methods. Indirect experimental support: FMRP-associated mRNAs An excellent opportunity to test our target predictions comes from experiments showing the association of mRNAs and miRNAs with proteins involved in translational control, even if these experiments do not provide information on specific miRNA:mRNA pairings. In particular, FMRP, which may regulate translation in neurons, not only associates with hundreds of mRNAs ( Brown et al. 2001 ; Chen et al. 2003 ; Denman 2003 ; Miyashiro et al. 2003 ; Waggoner and Liebhaber 2003 ) and with miRNAs ( Jin et al. 2004 ), but also associates with components of the miRNA processing machinery, Dicer, and the mammalian homologs of AGO1 and AGO2 ( Jin et al. 2004 ). If all FMRP-bound mRNAs are regulated by miRNAs, one should see a large enrichment of predicted targets among such mRNAs. We tested this hypothesis with 397 FMRP-associated mRNAs taken from a number of recent experiments ( Brown et al. 2001 ; Chen et al. 2003 ; Denman 2003 ; Miyashiro et al. 2003 ; Waggoner and Liebhaber 2003 ). Are FMRP-bound messages enriched in predicted targets? Using five different datasets ( Table S9 ), we predicted that 74% of FMRP-associated messages are miRNA target genes (294 of 397 mRNAs). This corresponds to an enrichment factor of about five compared to the 59 targets one would expect from our analysis in a randomly chosen set of 397 mRNAs, where 59/397 equals 4,462/29,785 (4,462 predicted mammalian target mRNAs pass the 90% conservation filter for one or more sites per transcript out of a total of 29,785 transcripts). This suggests that in the 397 FMRP target genes, 59 should pass the filters. The enrichment factor does not vary much with the cutoff parameters used in target prediction (data not shown), but is subject to some uncertainty because of potential false-positive predictions. The enrichment of miRNA:FMRP interaction is consistent with the hypothesis that translational control involving FMRP protein is executed in a complex that involves one or more miRNAs interacting with transcripts at specific sites. Note that this analysis supports the validity of target gene prediction, not the identity of the controlling miRNA or the accuracy of specific sites. An additional validation test involved FMRP cargoes that had been identified in more than one study, using independent experimental methods. For example, the mRNAs of 14 genes ( Brown et al. 2001 ) were overrepresented in both the polyribosome fraction of mouse fragile X cells and in co-immunoprecipitation with mouse brain mRNPs that contain FMRP. Almost all of the 14 genes are predicted targets with more than one conserved site (11 of 12 annotated UTRs; Table S9 ). In some cases, expression data provide additional support: postsynaptic density protein 95 (PSD95)–associated (SAPAP4), a neuron-specific protein, is regulated by many miRNAs highly expressed in rat brain primary cortical neurons ( Kim et al. 2004 ). In summary, the three validation approaches (retrospective, statistical, and indirect experimental) suggest that the current version of the miRanda algorithm, in spite of clear limitations, can predict true miRNA targets at reasonable accuracy, provided that (1) the targets are detected as conserved and (2) the gene contains more than one miRNA target site or a single high-scoring site ( S > 110, approximately, including sites with almost perfect complementarity suggestive of mRNA cleavage). Overview of Mammalian miRNA Target Genes More than 2,000 mammalian targets. We predicted 2,273 genes as targets with two or more miRNA target sites in their 3′ UTRs conserved in mammals at 90% target site conservation (see Tables S2 and S3 ). This means we predicted approximately 9% of protein-coding genes to be under miRNA regulation. In addition, we predicted another 2,128 genes with only one target site, but the false-positive rate for these is significantly higher ( Figure 2 ). Of these, the top-scoring 480 genes ( S > 110) have an estimated false-positive rate comparable to that of genes with multiple sites and thus also are good candidates for experimental verification. Some of the genes with single sites may contain additional sites that we cannot detect for a number of reasons, including truncated UTRs. A significant subset of the total number of single-site target genes (7%) has near complementary single sites. These near complementary sites may indicate cleavage, for which additional sites may not be necessary. The targets listed in Table 1 were selected for variety of function, variation in number of sites, and varied extent of conservation (some are also conserved in fish). Somewhat surprisingly, the number of predicted targets per miRNA varies greatly, from zero (for seven miRNAs) to 268 (for let-7b ), but the distribution is nonuniform (mean = 7.1, standard deviation = 4.7; Figure 3 ). This indicates a range of specificity for most miRNAs and suggests that regulation of one message by one miRNA is rare. Figure 3 Multiplicity and Cooperativity in miRNA–Target Interactions One miRNA can target more than one gene (multiplicity) (A), and one gene can be controlled by more than one miRNA (cooperativity) (B). The distributions are based on ordered (ranked) lists and decay approximately exponentially (approximate straight line in log-linear plot). (A) Some miRNAs appear to be very promiscuous (top left), with hundreds of predicted targets, but most miRNAs control only a few genes (bottom right). (B) Some target genes appear to be subject to highly cooperative control (top left), but most genes do not have more than four targets sites (bottom right). Although specific values are likely to change with refinement of target prediction rules, the overall character of the distribution may well be a biologically relevant feature reflecting system properties of regulation by miRNAs. Table 1 Selection of Predicted miRNA Targets Add “ENSG00000” to the beginning of the identifiers to derive Ensembl identifiers. All miRNA–target relationships shown here are conserved in mammals, i.e., homologous miRNAs target transcripts of homologous genes at similar UTR positions with similar local sequence. Genes that are predicted to be targets in both mammals and fish are in bold. Where the miRNA–target relationship is also conserved in non-mammalian vertebrates, the miRNA is in bold a Contains conserved CPE motif N.A., not available Functional analysis We analyzed the distribution of functional annotation for all targets of all miRNAs using Gene Ontology (GO) terms (see Materials and Methods ; Table S10 ) and domain annotations from InterPro ( Mulder et al. 2003 ). The target genes reflected a broad range of biological functions ( Figure S1 ). The most enriched GO term was “ubiquitin-protein ligase activity,” with 3.3-fold enrichment ( Table S10 ). Since ubiquitination is a process controlling the quantity of specific proteins in a cell at specific times, miRNA regulation of components of the ubiquitin pathway could increase protein levels. Other overrepresented functional terms were “neurogenesis” (3.2-fold), “protein serine/threonine kinase” (2.5-fold), and “protein-tyrosine kinase activity” (2.5-fold). The four domains most overrepresented in predicted targets relative to all genes were Homeobox domain, 5.3-fold; KH domain, 4.0-fold; and Guanine-nucleotide dissociation stimulator CDC25 domain, 3.4-fold ( Figure S1 ; Table S10 ). Interestingly, KH domains are RNA-binding domains found in a wide range of proteins such as hnRNPk, FMR1, and NOVA-1. In addition to the Homeobox domain, other DNA-binding domains and domains associated with chromatin regulation were also enriched, suggesting that miRNAs in animals target the transcription machinery disproportionately, as they do in plants. Another overrepresented domain was semaphorins (3.0-fold). The semaphorins and plexins (semaphorin receptors) are involved in axon guidance, angiogenesis, cell migration, the immune system, and the adult nervous system ( Pasterkamp and Verhaagen 2001 ). Many semaphorins and their receptors are predicted targets of brain-expressed miRNAs (e.g., let-7c, miR-125b, miR-153, miR-103, miR-323, miR-326, and miR-337 ). The plexins dimerize with Neuropilin (NP1) to signal the Semaphorin ligand attachment; neuropilin is a predicted high-ranking target of let-7g and miR-130, both brain-expressed miRNAs. A significant proportion of ephrin receptors (seven out of ten genes) and ephrin ligands (five out of seven genes) are predicted targets. The family of ephrins is involved in boundary formation, cell migration, axon guidance, synapse formation, and angiogenesis, and the ephrin ligand, EphA2, which contains a conserved cytoplasmic polyadenylation element (CPE) motif, is considered to be under translational regulation in axon growth cones ( Steward and Schuman 2003 ). Although many members of the ephrin family are predicted targets of brain-expressed miRNAs, they appear to be targeted by different miRNAs, consistent with differential regulation. In Drosophila, both ephrin and EphR, closest to the mammalian B class of the ephrin family, also are predicted miRNA targets. Do specific miRNAs target particular functional groups? We analyzed each miRNA individually for GO term and domain enrichment ( Table S10 ). The targets of some miRNAs were strongly enriched in certain categories, e.g., miR-105 in “small GTPase mediated signal transduction” (5-fold), miR-208 in “transcription factor” (6-fold), and miR-7, which lies in the intron of the hnRNPk (an RNA-binding protein) gene, in “RNA binding proteins.” Neuronal differentiation of embryonic carcinoma cells by retinoic acid in both mice and humans is coupled to induction of let-7b, miR-30, miR-98, miR-103, and miR-135 ( Sempere et al. 2004 ), and their targets are enriched in “neurogenesis” (3.5-fold). miR-124a and miR-125, both highly and specifically expressed in brain, preferentially target RNA-binding proteins. Thirty-one new miRNAs (miR - 322 – miR - 352) cloned from rat neuronal polyribosomes have a large number of neuronal target genes and share many targets, e.g., miR-352 and miR-327 target 5HT-2c, and miR-340, -328, -326, -331, and -333 potentially target beta-catenin, which is implicated in various stages of neural differentiation. Two highly expressed miRNAs in the thymus, miR-181a and miR-142–3p are key components of a molecular circuitry that modulates hematopoietic lineage ( Chen et al. 2004 ). Ectopic expression of miR-181a causes a 2-fold increase in the cells of the B cell lymphoid lineage. Some of our high-ranking targets for miR-181a may provide clues for the mechanism of this effect. Germ cell nuclear factor GCNF (NR6A1) (the second-highest-ranked target for miR-181a ) is expressed in the thymus and bone marrow. miR-181a itself is encoded on the antisense strand of an intron of GCNF . We also predict that the gene Bcl11b, known to affect B cell growth, is a target of miR-181a, ranking third, as well as Lim/homeobox protein LHX9, recently found expressed in developing thymus ( Woodside et al. 2004 ). FMRP cargo mRNAs regulated by miRNAs FMRP is composed of several RNA-binding domains (two KH and one RRG) that bind messages. The specific binding motifs for FMRP on messages are incompletely known, but are thought to include G-quartet patterns and/or U-rich sequences ( Dolzhanskaya et al. 2003 ; Ramos et al. 2003 ). We predicted 294 mRNAs known to be FMRP cargoes as miRNA targets (see Table S9 ). The most reliable of these ( Table 2 ) reflect high confidence in experimental identification of FMRP association or conservation of target site between mammals and fish. Table 2 Selected FMRP Cargoes Predicted as miRNA Targets Transcripts for genes (Gene and ID) are described as FMRP cargoes in several studies (DR) and predicted here as targets of specific miRNAs (miRNA). Selected from a total of 294 such targets a Reference from which data was extracted b Homologous miRNA–mRNA pair conserved in fish c Additional miRNAs are predicted to target the gene (number in parentheses): APP (9), BASP1 (4), Capicua (2), DLG3 (7), and DLG4 (5) d The miRNA has multiple target sites on the gene e The 3′ UTR of the gene contains a CPE motif (Table S11) Alzheimer's disease amyloid protein Amyloid precursor protein (APP) is an FMRP-bound protein that is translationally regulated. The APP transcript contains a 29-nt motif at position 200 in the 3′ UTR that is known to aid destabilization of the APP mRNA in certain nutrient conditions and that binds nucleolin, a protein associated with RNPs containing FMRP ( Rajagopalan and Malter 2000 ). In addition, there is an 81-nt sequence at position 630 in the APP 3′ UTR that is required for the TGFbeta-induced stabilization of the APP mRNA ( Amara et al. 1999 ). We predicted APP as a target, with a total score of S = 708 with a minimum of eight miRNA sites, including two let-7 top-ranking sites that are conserved in human, mouse, and rat. One of the predicted miRNA target sites in the APP UTR lies in the 81-nt region ( Figure 4 ), and another is within 30 nt of the motif at position 200. Figure 4 Potential miRNA Target Sites in the 3′ UTRs of Selected Genes Nucleotide sequence conservation between the 3′ UTRs of human and the closest mouse or rat orthologous genes is averaged for each block of 40 base pairs (long rectangles; white indicates 0% identical nucleotides, black indicates 100% identical nucleotides, and grey indicates intermediate values). The positions of target sites for specific miRNAs (triangles above rectangles, with numbers indicating miR miRNAs, e.g. “130” is “mir-130”) are, in general, distributed nonuniformly. Sequence motifs other than target sites (triangles below rectangles) are mRNA stability elements (APP), a G-quartet (DLG4), and an AU-rich element (ELAVL1), representing possible protein-binding sites. Detailed alignments between the miRNA and the predicted target sites (arbitrary selection) illustrate, in general, stronger match density at the 5′ end of miRNAs than at the 3′ end, as required by the algorithm and as observed in experimentally validated targets. The nonconserved nucleotides in the target sites are highlighted in red. Gene names map to the following Ensembl identifiers (142192 is ENSG00000142192, etc.): APP, 142192; CPEB2, 137449; DLG4, 132535; EFNB1, 090776; EIF2c1, 092847; ELAVL1, 066044; EPHB1, 154928; EPHB3, 182580; FMR1, 102081; FMR2, 155966; FXR1, 114416; FXR2, 129245; and PTEN, 171862. Other APP-interacting proteins, APP-binding family B member 1 (mir-9, miR-340, and miR-135b), APP-binding family member 2 (let-7 and miR-218), and APP-binding family 2 (miR-188 and miR-206) were also predicted targets, some of which had near exact target site matches. In summary, the APP gene appears to be subject to translational regulation by the combinatorial control of a number of different miRNAs. PSD95 and synaptic processes PSD95 and similar scaffolding molecules, link the NMDA receptor with intracellular enzymes that mediate signaling; this process is involved in the development and maintenance of synaptic function and synaptic plasticity, and interference in this process is implicated in schizophrenia and bipolar disorder ( Beneyto and Meador-Woodruff 2003 ). FMRP binds PSD95 and is required for mGluR-dependent translation of PSD95 ( Todd et al. 2003 ). PSD95 is a high-ranking target of miR-125, miR-135, miR-320, and miR-327, all of which are either exclusively expressed in brain or enriched in brain tissue ( Lagos-Quintana et al. 2002 ; Krichevsky et al. 2003 ; Sempere et al. 2004 ). In particular, large transcript numbers of miR-125b are found copurified with polyribosomes in rat neurons in ( Kim et al. 2004 ). PSD95 has one reported G-quartet in its 3′ UTR at position 648 ( Todd et al. 2003 ), further suggesting it as an in vivo FMRP target. We predicted an additional G-quartet site at position 205–235 in the 3′ UTR of PSD95 . One of the miRNA (miR-125) target sites overlaps with the G-quartets, raising the possibility that miRNAs directly compete with FMRP to bind the message in this location. Likewise, NAP-22, which has three miRNA target sites (see Table S9 ), has a miR-207 target site that overlaps with a G-quartet ( Darnell et al. 2001 ). Other PSD95 family members are also involved in synaptic processes, in particular, in the integration of NMDA signaling in the synaptic membrane. All PSD95 family members in mammals (also known as discs large 1–5 ), SAP90, and CamKII are predicted miRNA targets (see Table S9 ), as well as mGluR, the protein product of which is an agonist that induces the rapid translation of PSD95 ( Todd et al. 2003 ) and three NMDA receptor subunits (see Table S9 ). These results suggest that miRNAs may be involved in NMDA and glutamate receptor signaling to coordinate and integrate information, with specificity achieved through the combinatorial action of different miRNAs. Components of RNPs Regulated by miRNAs FMRP-associated proteins FMRP binds its own mRNA, implying negative feedback if the binding inhibits FMRP production ( Ceman et al. 1999 ) . The fact that miRNAs target transcripts for FMRP and FMRP-binding proteins suggests another negative feedback loop in which high levels of these proteins inhibit their own production (depending, of course, on the concentration of miRNAs and mRNAs) ( Figure 4 ). The genes for six FMRP-associated (not associated at the same time) proteins, hnRNP A1, Pur-alpha, Pur-beta, Staufen, AGO-2, and PABP, are predicted miRNA targets. This indicates that FMRP-containing RNPs are under miRNA regulation. FXR2, a gene similar to FMR1 is also a miRNA target in human, mouse, rat, and fish. Details of the implied feedback regulation and differential control of RNP action remain to be determined. RISC. Our data suggest that the RNAi–miRNA machinery itself is under miRNA regulation; for example Dicer appears to be controlled by let-7 and miR-15b; Ago-1 by let-7 and miR-29b/c; Ago-2 by miR-138; Ago-3 by miR-138, miR-25, and miR-103; and Ago-4 by miR-27a/b. Dicer and two of the Argonautes also are predicted to be targets in both zebrafish and fugu. The let-7 sites on the 3′ UTR of Dicer and Ago-1 ( Figure 4 ) will accommodate most of the let-7 variants with similar scores. The variants of let-7 are expressed in a wide range of tissues and developmental stages, suggesting broad regulation of Dicer and Ago-1 by miRNAs. In contrast, the only miRNA that targets Ago-2 is miR-138, which has so far been cloned only once in the cerebellum ( Lagos-Quintana et al. 2002 ). The target site for miR-138 has only one mismatch at position 8; this may induce a siRNA-like cleavage of the message ( Hutvágner and Zamore 2002a ; Doench et al. 2003 ). Ago-3 is also a top target for miR-138, with only two mismatches in its site. We suggest that some miRNAs targeting this machinery (e.g., let-7, miR-27, miR-29, and miR-103) are expressed fairly widely, while others (e.g., miR-138 and miR-25 ) have lower and more restricted expression. Other RNPs. The highly conserved RNA-binding proteins, ELAV-like proteins (HuR, HuB, HuC, and HuD), contain three RNA-recognition motifs, which bind AU-rich elements in 3′ UTRs of a subset of target mRNAs ( Good 1995 ). These AU-rich elements increase the proteins' cytoplasmic stability and increase translatability ( Perrone-Bizzozero and Bolognani 2002 ). Experiments have identified 18 mRNAs bound to HuB in retinoic-acid-induced cells; of the 14 we were able to map unambiguously, 12 are predicted miRNA target genes: Elavl1 (known to regulate its own mRNA), Gap-43, c-fos, PN-1, Krox-24, CD51, CF2R, CTCF, NF-M, GLUT-1, c-myc, and N-cadherin ( Tenenbaum et al. 2000 ). Three of the ELAV-like genes themselves are also targets of a large number of miRNAs (see Tables S2 and S3 ; Figure 4 ). This is yet another example of miRNAs predicted to target the bound messages of RNA-binding proteins and of the regulation of RNA-binding genes by miRNAs. Cytoplasmic Polyadenylation Binding Proteins Regulated by miRNAs We predicted all four human cytoplasmic polyadenylation binding proteins (CPEBs) known in mammals as miRNA targets ranked within the top 170 target genes with 6–20 sites in their UTRs ( Figure 4 ; Table S11 ). Indeed, CPEB2 is the highest-ranking gene of all transcripts . The orthologs to CPEB1 in fish and fly (known as orb in D. melanogaster ) are also predicted as targets. CPEB is an RNA-binding protein first shown to activate translationally dormant mRNAs by regulating cytoplasmic polyadenylation in Xenopus oocytes ( Hake and Richter 1994 ). It also regulates dendritic synaptic plasticity ( Mendez and Richter 2001 ; Richter 2001 ) and dendritic mRNA transport ( Mendez and Richter 2001 ; Huang et al. 2003 ) and facilitates transport of mRNAs in dendrites together with kinesin and dynein in RNPs ( Huang et al. 2003 ). CPEB binds to its target message through the CPE motif (UUUUAU), which must be within a certain distance of the hexanucleotide AAUAAA. CPEB keeps messages in their dormant state until phosphorylated, after which it activates polyadenylation ( Mendez et al. 2000 ), thereby activating translation or degradation ( Mendez et al. 2002 ). In addition, CPEB co-fractionates with the postsynaptic density fraction in mouse synaptosomes, consistent with translation of stored mRNAs in dendrites being part of the mechanism of synaptic plasticity. We have three more lines of evidence suggesting the notion that translational regulation by CPEB is linked to miRNA regulation. First, our target list and the list of genes regulated by CPEB significantly overlap. There are nine genes known to be CPEB-regulated, seven of which are predicted targets: alpha-CAMIIK, Map 2, Inositol 1, 4–5-Triphosphate Receptor type 1, Ephrin A receptor class A type 2, SCP-1, and CPEB3 ( Mendez and Richter 2001 ). Second, CPEB is known to self-regulate in D. melanogaster ( Tan et al. 2001 ). The CPEB1 homolog in fly, orb, and CPEBs in vertebrates are predicted miRNA targets. Third, the gene most correlated in expression to the CPEB homolog in D. melanogaster is a Piwi protein (Sting), a member of the Argonaute family ( Pal-Bhadra et al. 2002 ; Stuart et al. 2003 ) that is involved in translational regulation and in the RISC. Among the predicted miRNA targets, 115 genes also contained CPE motifs, which were conserved in at least two mammals in the same positions in the UTRs and are therefore candidates for CPEB regulation ( Table S11 ; see Materials and Methods ). Our predictions include HuB, HuR, Eif-4 gamma, DAZ associated protein 2, VAMP-2 (known to be posttranscriptionally regulated), Presynaptic protein SAP102, and brain-derived neurotrophic factor precursor. Taken together these data suggest that the CPEB genes, the known CPEB-regulated genes, and the predicted CPEB-regulated genes are strong miRNA target candidates and provide rich ground for experimentation. Targets of Cancer-Related miRNAs Deregulated expression of certain miRNAs has been linked to human proliferative diseases such as B cell chronic lymphocytic leukemia ( Calin et al. 2002 ; Lagos-Quintana et al. 2003 ) and colorectal neoplasia ( Michael et al. 2003 ). Recent analysis of the genomic location of known miRNA genes suggested that 50% of miRNA genes are in cancer-associated genomic regions or in fragile sites ( Calin et al. 2004 ). The miRNAs miR-15 and miR-16 are located within a 30-kb region at Chromosome 13q14, a region deleted in 50% of B cell chronic lymphocytic leukemias, 50% of mantle cell lymphomas, 16%–40% of multiple myelomas, and 60% of prostate cancers ( Calin et al. 2002 ). Furthermore, miR-15 and miR-16 are down-regulated, or their loci lost, in 68% of B cell chronic lymphocytic leukemias ( Calin et al. 2002 ). Similarly, miR-143 and miR-145 are down-regulated at the adenomatous and cancer stages of colorectal neoplasia ( Michael et al. 2003 ), and miR-155 is up-regulated in children with Burkitt lymphoma ( Metzler et al. 2004 ). Our method predicted cancer-specific (by annotation) gene targets of miR-15a, miR-15b, miR-16, miR-143, miR-145, and miR-155 . The target genes and their miRNA regulators are as follows: (1) CNOT7, a gene expressed in colorectal cell lines and primary tumors ( Flanagan et al. 2003 ) (miR-15a); (2) LASS2, a tumor metastasis suppressor ( Pan et al. 2001 ) (miR-15b); (3) ING4, a homolog of the tumor suppressor p33 ING1b, which stimulates cell cycle arrest, repair, and apoptosis ( Shiseki et al. 2003 ) (miR-143); (4) Gab1, encoding multivalent Grb2-associated docking protein, which is involved in cell proliferation and survival ( Yart et al. 2003 ) (miR-155); and (5) COL3A1, a gene up-regulated in advanced carcinoma ( Tapper et al. 2001 ) (miR-145). miR-16 has a tantalizing number of high-ranking targets that are cancer associated and specifically involved in the Sumo pathway There is increasing evidence that Sumo controls pathways important for the surveillance of genome integrity ( Muller et al. 2004 ). The first- and fifth-highest-ranked targets of miR-16 are Sumo-1 activating and conjugating enzymes, respectively. The top two single-site targets for miR-16 are an Activin type II receptor gene (TGFbeta signaling) and Hox-A5, both known to be dysregulated at the level of protein expression in colon cancers ( Wang et al. 2001 ). Both of these sites show near perfect complementary matching between miR-16 and the target genes (indicating possible cleavage). Both of these target genes are also targets for another cancer related miRNA, miR-15 . Targets Conserved between Mammals and Fish Roughly 55 miRNAs have identical mature sequences in fugu and mammals, and 80 have very similar sequences in the two species; additional fish miRNA sequences can be predicted with confidence based on sequence similarity. It is therefore reasonable to expect that the targets of these probably functionally homologous miRNAs are orthologous genes in the different species. To follow up on this hypothesis, we assessed conservation of mammalian miRNA–target pairs between the 2,273 mammalian and 1,578 fish miRNA targets (with more than one target site per UTR). The analysis yielded 240 target genes conserved between mammals and fish. The number 240 is probably an underestimate because of several factors, including: (1) unfinished annotation of genomes, particularly rat and fugu; (2) ambiguity in assigning orthologs; and (3) lack of UTR information . The full set of conserved target genes between fish and mammals indicates a wide functional range of conserved targets ( Table S12 ). Many Hox genes are conserved as targets, including the miR-196 targets, Hox-A4:miR-34a, Hox-C9:let-7b (near prefect complementary match), and Hox-B5:miR-27b . Examples from the notch signaling pathway include miR-30:hairy enhancer of split 1 (Hes1) and miR-152:noggin. Targets Conserved between Vertebrates and Flies Twenty-eight of the 78 identified miRNAs in flies have apparent mammalian homologs. Based on this remarkable conservation across hundreds of millions of years, it is reasonable to expect that there is some conservation of target sites, target genes, and target pathways between flies and humans. Most strikingly we can identify hox genes and axon guidance genes as common targets between vertebrates and flies, e.g., capicua and sex combs reduced (one of the vertebrate homologs of Hox-A5 ). The hox gene cluster in Drosophila contains high-ranking predicted targets ( Enright et al. 2003 ) of miR-10 and miR-iab-4, and the hox gene cluster in mammals contains high-ranking targets of miR-196 . These miRNAs are themselves located in the hox gene region. We predicted miR-iab-4–3p to target abd-B in Drosophila, a gene related to the ancestral hox-7 cluster, the ancestral parent of many of the predicted targets of miR-196 . Axon guidance receptors and ligands conserved as targets include Lar, ephrins, and slits . Human slit1 is a top target of miR-218, which itself is transcribed from the intron of slit2, suggesting down-regulation of slit1 by transcription of slit2 . We expect that there are many more conserved targets but we are hindered by the difficulty of mapping orthologous genes between human and fish. Future work will elucidate to what extent there are common pathways regulated by common miRNAs between vertebrates and invertebrates. Target Sites in Protein-Coding Sequences Experiments suggest that miRNA target sites in metazoans are preferentially in UTRs, not in coding regions. If this is true, a correct target site prediction method should predict a larger number of targets in UTRs than in coding regions. Alternatively, target sites in coding regions may so far have escaped experimental verification, especially in plants, in which targets of miRNAs in coding regions are the rule, not the exception. To investigate this issue we computed the average density of target sites for high-scoring targets ( S > 130) and before application of conservation filters. The statistical assessment of the influence of conservation filters in coding regions would have raised complicated issues, as nearly two-thirds of nucleotides in coding regions are conserved between mammalian genomes to preserve amino acid sequences. Interestingly, we found, on average, 11 pre-conservation target sites per 1 million nucleotides in coding regions, versus 15 such target sites per 1 million nucleotides in UTRs. This is consistent with a stronger “raw” prediction signal in UTRs and may indicate a lower number of biologically relevant target sites in coding regions in mammals, consistent with early experimental findings. As a guide to experimentation, we report all sites in coding regions with an alignment score above 110 for miRNAs of length up to 20 nt and an alignment score above 130 for miRNAs longer than 20 nt (scores depend on the length). These cutoff scores approximately correspond to a 75% complementary match between miRNA and target, leaving open the question of how many match pairs are needed to lead to translational inhibition in coding regions, by any mechanism. We identified 942 genes that contained such sites in their coding regions. Strikingly, there was only one site with a perfect match, and this was for the imprinted miR-127, known to be antisense to the reciprocally imprinted retrotransposon-like gene on the opposite strand ( Seitz et al. 2003 ). Of the 942 genes, 25% have been otherwise identified as targets based on conserved target sites in their UTRs. However, only five genes have targets sites in their UTRs complementary to the same miRNA that targets the coding region (see Table S3 , columns H and I). For example, miR-211 has a near perfect complementary site in the coding region of a gene of unknown function (Ensembl ID ENSG00000134030, containing an Eif-4 gamma domain) and also has two conserved “normal” sites in the UTR. Similarly, miR-198 has a site in the coding region, as well as conserved sites in the UTR region, of a sodium and chloride GABA transporter (Ensembl ID ENSG00000157103). However, we see no trend for miRNAs that have conserved sites in UTRs to have additional sites in the coding region; rather, stronger target sites for a given miRNA tend to be confined either to the UTR or the coding region and are rarely in both. Target Sites with Near Perfect Matches in cDNAs We scanned all cDNAs for high-scoring matches without using conservation to check for high-scoring targets, which we may have missed through strict conservation rules (see Table S6 ). Over 40 genes contain sites that have near perfect complementarity to a miRNA ( S >120), and these target genes may be cleaved rather than translationally repressed as in the case of miR-196 and Hox-B8 . For example miR-298, an embryonic-stem-cell-specific miRNA ( Houbaviy et al. 2003 ), has a near match with MCL-1, and miR-328 (neuronally expressed) has a near match with LIMK-1, which is known to be involved in synapse formation and function. miR-129, expressed in mouse cerebellum, has a near perfect complementary match with Musashi-1, which is an RNA-binding gene essential for neural development, regulated in the cerebellum, and up-regulated in medulloblastoma ( Yokota et al. 2004 ). Comparison of miRNA Target Prediction Methods Recently, several computational methods for the prediction of miRNA targets have been developed ( Enright et al. 2003 ; Lewis et al. 2003 ; Rajewsky and Socci 2003 ; Stark et al. 2003 ; Kiriakidou et al. 2004 ; Rehmsmeier et al. 2004 ). Two of these have been applied to mammalian miRNAs, as described in Lewis et al. (2003) and Kiriakidou et al. (2004) . We now compare and contrast these two methods with each other and with the current version of our method, as further developed from miRanda 1.0 and as applied to mammalian and vertebrate genomes ( Enright et al. 2003 ). We compare algorithms and target lists, as an aid to the design of experiments. The three prediction methods share the goal of identifying mRNAs targeted by miRNAs. All three use sequence complementarity, free energy calculations of duplex formation, and evolutionary arguments in developing a scoring scheme for evaluation of potential targets. Results are reported as lists of target sites and lists of target genes containing such sites. The three methods differ, however, in important technical details, such as the datasets of miRNA and UTR sequences and the algorithm and scoring scheme, as well as the report format. We now summarize these technical differences and compare the lists of resulting target genes for a common subset of miRNAs. The interpretation of such comparisons is hampered by the fact that selection criteria and the use of numerical cutoffs differ conceptually, and genomic coverage is nonuniform. In the first method, Lewis et al. used 79 miRNAs in human, mouse, and rat, seeking targets in a UTR dataset extracted from the June 2003 version of the Ensembl database. The UTR dataset had 14,300 ortholog triplets conserved between human, mouse, and rat and 17,000 ortholog pairs between human and mouse. All annotated UTRs were extended by 2 kb of 3′ flanking sequence. The algorithm required exact complementarity of a 7-nt miRNA “seed” sequence, defined as positions 2–8 from the 5′ end of the miRNA, to a potential target site on the mRNA, followed by optimization of mRNA–miRNA duplex free energies between an extended window of 35 additional bases of the mRNA and the rest of the miRNA. Target genes were ranked using a composite scoring function, which took into account all sites for a particular miRNA on a given mRNA. Conserved miRNA:mRNA pairs were required to involve orthologs of miRNA and mRNA in human, mouse, and rat, but there was no requirement for conservation of target site sequence (beyond the seed match) or position on the mRNA. Using shuffled miRNA sequences, with the constraint that shuffled controls match real miRNAs in relevant sequence properties, the false-positive rate of predictions was estimated to be 50% for target genes conserved between mouse and human, 31% for target genes conserved in human, mouse, and rat, and 22% for target genes identified in fugu as well as mammals. As a final result, Lewis et al. reported 400 conserved target genes for the 79 miRNAs. Among these targets, 107 genes were reported as conserved in the fish fugu. In the second method, Kiriakidou et al. used 94 miRNAs in human and mouse, seeking targets in a dataset of 13,000 UTRs conserved in mouse and human (from Ensembl, date not given). The algorithm used a 38-nt sliding mRNA window and calculation of miRNA–mRNA duplex free energies, keeping duplexes with energies below −20 kcal/mol. The duplexes were further filtered using a set of requirements regarding matches and loop lengths in certain positions, as derived and extrapolated from experimental tests involving a predicted target site for let-7b miRNA on the UTR of the human homolog of worm lin-28 . The target site sequence was engineered into a Luciferase reporter, followed by sequence variation of the target site and test of an initial set of 15 predictions in the same reporter assay. Using shuffled miRNA sequences, and applying the same rules and parameters, the false-positive rate of predictions was estimated to be 50% for targets conserved between human and mouse. As a final result, Kiriakidou et al. reported 5,031 human targets, with 222 reported as conserved in the mouse. In the third method (this work), we used 218 mammalian miRNAs and 29,785 transcripts derived from Ensembl ( Table 3 ) and, as a final result, report 4,467 target genes. What are the main differences between these three prediction methods? Comparison of the total number of predicted target genes is not very informative, as different datasets and cutoffs were used. We attempted to remove one of the technical differences, by explicitly comparing reported targets for the same set of 79 miRNAs used by Lewis et al. (although significant differences remained in the sets of UTR sequences used): the overlap of target genes between Kiriakidou et al. (out of 189) and Lewis et al. (out of 400) was 10.6%; the overlap between Lewis et al. (out of 400) and this work (out of 2,673) was 46%; and the overlap between Kiriakidou et al. (out of 189) and this work (out of 2,673) was 49%. In each case the totals (“out of”) are the number of target genes for the common set of 79 miRNAs and the percentage is relative to the smaller set of two compared. The obvious reason for the larger overlap with our results, 46% and 49% respectively, is the larger number of targets in our predictions, which in turn is primarily the result of choice of cutoff. Table 3 Number of Genes and 3′ UTR Sequences Used for Target Prediction 3′ Direct comparison of the three prediction methods is complicated by the fact that the noticeable differences between the target lists of the three methods are due to the aggregate effects of datasets, algorithm, including selection rules, use of conservation, and cutoffs. The following characteristics of the three methods underlie these differences and should be taken into consideration when choosing targets for experimentation. (1) As to UTR datasets, Lewis et al., with the earliest published report, used a smaller set of UTRs, with some likelihood of false positives as a result of UTR extension. The UTR sets used in this work, the third in terms of publication date, are the most comprehensive and plausibly the most reliable (as of February 2004). (2) As to miRNA datasets, there was an increase from 79 for Lewis et al. to 94 for Kiriakidou et al. to 218 miRNAs used in this work. (3) As to the cooperativity of binding, the scoring system of Lewis et al. evaluated cooperativity of multiple target sites by the same miRNA on a target gene, but disregarded multiple target sites from different miRNAs on one gene; that of Kiriakidou et al. focused on single sites; and that of this work gave high scores to multiple hits on a target gene, no matter whether these hits involved the same miRNA or different miRNAs. These tendencies are not exclusive where scores involve functions of several real numbers, with cutoffs applied to the aggregate score; e.g., our method also allows strong single target sites. (4) As to assessment of false positives using statistical methods based on shuffling, the comparison of percentages is inconclusive, as the statistics of the background distribution of true negatives is not well known. It appears certain, however, from both Lewis et al. and this work, that statistical confidence increases with the extent of conservation among increasingly distant species. (5) As to validation experiments, each of the methods used a different type and set, with mixed overall conclusions. On the reassuring side, there was direct validation of some of the predicted target sites of Lewis et al. and of Kiriakidou et al. using reporter constructs in cell lines. We found some agreement between the sites validated in this way and our predicted targets (details in Table S13 ), but in some cases we predicted different details of target sites for a given experimentally tested miRNA:mRNA pair. Also, Kiriakidou et al. used a series of such experiments to extrapolate from a set of specific sequence variants to general rules for identification of target sites. However, serious doubts about the validity of any set of rules persist as there is very little in vivo validation in which native levels of specific miRNAs are shown to interact with identified native mRNA targets with observable phenotypic consequences under normal physiological conditions. (6) As to differences in algorithm, one can state opinions about the strengths or weaknesses of each particular algorithm, but the relationship between each prediction method and the actual in vivo process by which miRNAs have functional interactions with their target mRNAs remains unclear or, at best, unproven. In summary, in our view, each of the three methods, including the one in this work, falls substantially short of capturing the full detail of physical, temporal, and spatial requirements of biologically significant miRNA–mRNA interaction. As such, the target lists remain largely unproven, but useful hypotheses. The predicted targets are useful in practice for the design of experiments as they increase the efficiency of validation experiments by focusing on target lists significantly enhanced in likely targets, compared to random. It is plausible that targets near the top of lists are the most likely to lead to successful experiments. Task-specific filtering of target lists for particular planned experiments is recommended, especially with respect to cooperativity of binding (more than one site for one or more miRNAs on one gene transcript) and coincidence of expression, as new data on expression patterns of miRNAs and mRNAs in different tissues become available. For example, a recommended conservative approach to the design of experiments would use all available expression information and restrict the predicted target genes to those with two or more target sites at normal threshold ( S > 90) or one target site with a higher threshold ( S > 110), counting only sites with up to one G:U pair in residues 2–8 counting from the 5′ end of the miRNA. To take into account the rapid development of this field and the likely close interaction of theory and experiment, we plan to periodically update our prediction method and parameters and make revised target lists available on http://www.microrna.org . Next, we discuss some conceptual consequences of the composition of our target list. Discussion How Widespread Is the Regulation of Translation by miRNA? With plausible parameters, we have predicted that close to 9% (2,273 out of 23,531) of all mammalian genes have more than one miRNA target site in their 3′ UTRs, with 1,314 being stronger candidates with more than two target sites. This could well be an underestimate of the total number of genes subject to miRNA regulation, as we have used a conservative conservation filter. On the other hand, not all predicted miRNA–mRNA pairs would have a biological consequence unless both miRNA and mRNA are expressed at the same time in the same cell and at sufficient concentration. The human genome has about 250 miRNA genes, compared to about 35,000 protein genes. Thus, the the determination that about 1% of genes (miRNAs) control the expression of more than 10% of genes is a reasonable first order estimate. It is currently not known if any miRNAs control the expression of miRNA genes, i.e., the progression from miRNA transcript to mature miRNA. How Conserved in Evolution Are miRNA Targets? As many miRNA sequences are detectably conserved across large evolutionary distances, they must be subject to strong functional constraints. These constraints are unlikely to come from single-site interactions with the target, as experimentally validated animal miRNAs rarely have perfectly matched target sites. Plausibly, the evolution of miRNAs is constrained by functional interactions with multiple targets. As a consequence, any compensatory mutation in the miRNA in response to mutations in a target site would be disruptive to the miRNA's interaction with other target sites. Co-evolution of the miRNA sequence and all of its target sequences is therefore a rare event. With these assumptions, the constraints on the local mRNA sequence of individual target sites are weaker than those on the miRNA sequence. We were therefore surprised to observe a substantial number of cases (28.6% of the 2,273 targets) with 100% conservation of target site sequence and with the target sites being within ten nucleotides of each other on the globally aligned UTRs of orthologous genes between mammals. Lacking more detailed knowledge of miRNA evolution, we draw two operational conclusions. (1) Conservation of target site sequence and position is a practical information filter for predicted target sites, reducing the rate of false positives. (2) It is very likely that new miRNAs have continuously appeared in evolution ( Lai 2003 ) at some non-negligible rate and that the set of targets for any given miRNA has lost or gained members, even between species as close as human and mouse. It is therefore important to develop prediction tools that do not rely on conservation filters or at least allow us to make them weaker. Work on this is in progress. Multiplicity and Cooperativity Regulation by miRNAs is obviously not as simple as one miRNA–one target gene, as perhaps the early examples (lin-4 and let-7) seemed to indicate. The distribution of predicted targets reflects more complicated combinatorics, both in terms of target multiplicity (more than one target per miRNA) and signal integration (more than one miRNA per target gene). The distribution of the number of target genes (and target sites) per miRNA is highly nonuniform, ranging from zero for seven miRNAs to 268 for let-7b, with an average of 7.1 targets per miRNA. It is difficult to describe in detail, beyond the examples discussed in this text and beyond the annotation of target genes in Figure 2 and Table S3 , which specific processes appear to be regulated by each miRNA or each set of co-expressed miRNAs. Groups of targets may reflect a reaction, a pathway, or a functional class (see Results ). Although all miRNA–target pairs are subject to the condition of synchrony of expression, it is likely that typically one miRNA regulates the translation of a number of target messages and that, in some cases, the target genes as a group are involved in a particular cellular process. This was already known for the case of lin-4 ( Ambros 2003 ). The number of miRNA target sites per gene is also nonuniform, with a mean of 2.4. Although we do list target genes with single miRNA sites, there is increasing evidence that, in general, two or more sites are needed in the context of repression of translation. Although the details of these distributions (see Figure 2 and Table S3 ) depend on technical details, such as uniform cutoff for all miRNAs and evaluation in terms of a particular, imperfect scoring system, the general features of the distributions (see Figure 3 ) may be generally valid. We conclude that multiplicity of targets and cooperative signal integration on target genes are key features of the control of translation by miRNAs. Neither multiplicity nor cooperativity is a novel feature in the regulation of gene expression. Indeed, regulation by transcription factors appears to be characterized, at least in eukaryotes, by analogous one-to-many and many-to-one relations between regulating factor and regulated genes ( Kadonaga 2004 ). We are, of course, aware that the control cycles and feedback loops involving miRNAs cannot be adequately described without more detailed knowledge of the control of transcription of miRNA genes, about which little is known at present. Mechanisms of miRNA Action The role of a few animal miRNAs as posttranscriptional regulators of gene expression and, in particular, as inhibitors of translation is well established. However, the molecular mechanism of action is not well understood. Posttranscriptional control of protein levels can be achieved, for example, by cleaving the mRNA, by preventing RNP transport to ribosomes, by stalling or otherwise inhibiting translation on ribosomes, or by facilitating the formation of protein complexes near ribosomes that degrade nascent polypeptide chains. What do our results imply regarding the mechanism of action? In analogy to plant miRNAs that have near perfect sequence complementarity and facilitate mRNA degradation, our predicted targets with near perfect complementarity between miRNA and mRNA plausibly are involved in mRNA cleavage (e.g., miR-196 and miR-138; see Results ). Most of these would involve single target sites. In the case of Hox-B8, cleavage has been experimentally shown in mammalian cells ( Yekta et al. 2004 ). We estimate that fewer than 5% of miRNA targets are cleaved as a result of miRNA binding. Multiple target sites of lesser complementarity are consistent with RNP formation leading to translational inhibition, not mRNA degradation. Although we did predict single miRNA target sites for some genes, most target genes have multiple sites, indicating that cooperative binding ( Doench and Sharp 2004 ) may be essential for formation of inhibitory RNP complexes. An interesting and somewhat paradoxical feature is seen with mRNAs bound by FMRP, some of which increased and some of which are decreased in polysome fractions in FMRP knock-out mice ( Brown et al. 2001 ). We see no bias in which of these two sets is most enhanced as predicted miRNA targets. This ambiguity not only raises questions about details of FMRP regulation but also raises the possibility that miRNA targets may not always be translationally repressed and may instead be translationally enhanced. Improvement of Prediction Rules Current methods for predicting miRNA targets rely on conservation filters to reduce noise. Although the miRNA–mRNA pairings of experimentally validated targets were carefully used to define prediction rules ( Enright et al. 2003 ; Lewis et al. 2003 ; Stark et al. 2003 ), the information content in sequence match scores and free energy estimates of RNA duplex formation appears to be low. What is missing? Perhaps the fine details of experimentally proven target site matches are incorrect, although in some experiments mismatches and insertions have been tested. More plausibly, the rules do not yet capture additional functionally relevant interactions of miRNAs, such as in maturation and transport. Such additional interactions remain to be described in molecular detail, such as interactions with the small RNA processing machinery (Drosha and Dicer) and with the components of RNPs (AGO and FMRP). A first step in this direction is the very recent analysis of the crystal structure of a PAZ domain of a human Argonaute protein, eIF2c1, complexed with a 9-mer RNA oligonucleotide in dimer configuration, which may represent three-dimensional interactions for the 3′ end of a miRNA (and siRNA) complexed, e.g., with Dicer or AGO ( Ma et al. 2004 ). In this structure, each PAZ domain makes close binding contact with nine nucleotides of a single-stranded RNA. The two 3′ terminal nucleotides bind in a pocket through RNA backbone and other contacts. The remaining seven nucleotides bind PAZ through a series of backbone contacts such that nucleotides 3 to 9 are in an RNA helical conformation with bases exposed for base pairing to the second single-stranded RNA. If a 20–21-nt single-stranded RNA is bound to a PAZ domain in the same fashion, the 5′ end would be free for other interactions, such as binding to another protein domain in the RISC or base-pairing to mRNA. The conformational entropy that results when the 3′ end binds to PAZ, because the RNA helix is pre-formed, is consistent with weaker base pairing between miRNA and mRNA at the 3′ end of the miRNA, and stronger base pairing at the 5′ end. The dimeric structure of the PAZ domain ( Ma et al. 2004 ) also raises the tantalizing possibility of cooperative binding of a dimer of two miRNA–PAZ combinations to two target sites on one or more mRNAs. In such an arrangement, seven residues at the 3′ ends of the two miRNAs (residues 3–9, but not the terminal two nucleotides) are paired in antiparallel fashion, with near perfect complementary pairing. As more details of molecular contacts become available, prediction rules will evolve and improve in accuracy. The following elements are worth considering in the next generation of target prediction rules: (1) details of strand bias as deduced from siRNA experiments ( Khvorova et al. 2003 ), (2) contribution of sequences outside of the mRNA target sites, (3) refinement of position-dependent rules, including different gap penalties for the mRNA and the miRNA, (4) energetics of miRNA–protein binding, starting with PAZ domain interaction, and (5) translation of systematic mutational profiling experiments into scoring rules ( Doench and Sharp 2004 ). Principles of Regulation by miRNAs Although the predicted targets are subject to error (see estimate of false positives) and the prediction rules in need of improvement, several general principles of gene regulation by miRNAs are emerging. (1) Except in cases where a highly complementary match causes cleavage of the target message, miRNAs appear to act cooperatively, requiring two or more target sites per message, for either one or several different miRNAs. (2) Most miRNAs are involved in the translational regulation of several target genes, which in some cases are grouped into functional categories. (3) miRNAs carried in the context of RNPs appear to be sequence-specific adaptors guiding RNPs to particular target sequences. miRNA regulation of cellular messages may therefore range from a switch-like behavior (e.g., cleavage of mRNA message) to a subtle modulation of protein dosage in a cell through low-level translational repression ( Bartel and Chen 2004 ). These aspects of miRNA regulation complicate the design of experiments aiming at testing target predictions, or, more generally, at discovering biologically meaningful targets. Straightforward experiments that test one target site for one miRNA on one UTR will not be able to disentangle the effects of multiplicity or cooperativity. Tests for multiple sites on one UTR for one miRNA capture aspects of cooperativity ( Doench and Sharp 2004 ), but still do not capture signal integration by diverse miRNAs. The most complicated situation is one in which multiple miRNAs affect multiple genes in combinatorial fashion, with fine-tuning depending on the state of the cell. We look forward to the results of ingenious experiments designed to deal with the complexity of miRNA regulation. The results of this genome-wide prediction for mammals and fish are meant to be a guide to experiments that will in time elucidate the genetic control network of regulators of transcription, translation/maturation, and degradation of gene products, including miRNAs. Materials and Methods miRNA sequences Mature human and mouse miRNA sequences were obtained from the RFAM miRNA registry ( Griffiths-Jones 2004 ). To cover cases of incomplete data, any mouse miRNA sequence not (yet) described in humans was assumed to be present in human, with the same sequence, and vice versa. Similarly, all mouse miRNAs were assumed to be identical and present in the rat genome. These assumptions are reasonable as sequence identity for known orthologous pairs in human and mouse is, on average, 98% (with 110 out of 146 orthologous sequences being identical). In total, 218 mammalian miRNAs were used. For human target searches, 162 native miRNA sequences were available plus 17 mouse and 39 rat miRNA sequences; for mouse, 191 native, 14 human, and 13 rat sequences; and for rat, 45 native, 159 mouse, and 14 human miRNA sequences. Mature miRNA sequences for zebrafish and fugu were predicted starting from known human and mouse miRNA precursor sequences ( Ambros et al. 2003a ). Each precursor sequence was used, in a scan against the zebrafish supercontigs (release 18.2.1) using NCBI BLASTN (version 2.2.6; E-value cutoff, 2.0) ( Altschul et al. 1990 ), to identify a sequence segment containing the potential zebrafish miRNA. The mammalian and fish segments were then realigned using a global alignment protocol (ALIGN in the FASTA package, version 2u65; Pearson and Lipman 1988 ). After testing the potential fish miRNA precursors for foldback structures ( Zuker 2003 ), the final set of 225 predicted zebrafish miRNAs was selected. The same set of sequences was used for fugu. 3′ UTR sequences The Ensembl database ( Birney et al. 2004 ) served as the source of genomic data. The Ensembl BioPerl application user interface was used to generate 3′ UTR sequences for all transcripts of all genes from each genome. Some transcripts are alternatively spliced from the same gene, so the total number of genes is smaller than the number of transcripts ( Table 3 ). When no Ensembl annotated 3′ UTR sequences were available, we predicted 3′ UTRs by taking 4,000 bp of genomic sequence downstream of the end of the last exon of a transcript ( Table 3 ). If this predicted region overlapped coding sequence on either strand, we halted 3′ UTR extension at that point. UTR orthology and alignment Orthology mappings between genes from different genomes were obtained using “orthologue tables” from the EnsMart ( Kasprzyk et al. 2004 ) feature of the Ensembl database. Pairs of orthologous UTRs were aligned with each other using the AVID ( Bray et al. 2003 ) alignment algorithm to facilitate analysis of conservation of position and sequence of target sites. In total, 26,205 human transcripts, representing 15,869 genes, were mapped to both mouse and rat transcripts. For zebrafish, 11,442 transcripts, representing 10,909 genes, were mapped to fugu transcripts and 11,306 transcripts mapped to human transcripts (10,063 genes). miRNA target prediction The miRanda algorithm (version 1.0; Enright et al. 2003 ) was used to scan all available miRNA sequences for a given genome against 3′ UTR sequences of that genome derived from the Ensembl database and—tabulated separately—against all cDNA sequences and coding regions. The algorithm uses dynamic programming to search for maximal local complementarity alignments, corresponding to a double-stranded antiparallel duplex. A score of +5 was assigned for G:C and A:T pairs, +2 for G:U wobble pairs, and −3 for mismatch pairs, and the gap-open and gap-elongation parameters were set to −8.0 and −2.0, respectively. To significantly increase the speed of miRanda runs, in calculating the optimal alignment score at positions i, j in the alignment scoring matrix, the gap-elongation parameter was used only if the extension to i, j of a given stretch of gaps ending at positions i–1 , j or j–1 , i (but not of stretches of gaps ending at i–k, j or j, i–k for k > 1) resulted in a higher score than the addition of a nucleotide–nucleotide match at positions i, j . Removal of this restriction with the availability of more computing power would result in a moderate increase in average loop length, but the advantages of this would probably be superceded by overall refinement of target prediction rules. Importantly, complementarity scores at the first eleven positions, counting from the miRNA 5′ end, were multiplied by a scaling factor of 2.0, so as to approximately reflect the experimentally observed 5′–3′ asymmetry; for example, G:C and A:T base pairs contributed +10 to the match score in these positions. The value of the scaling factor at each position is an adjustable parameter subject to optimization as more experimental information becomes available. Because of the ongoing discussion about the rules for target prediction, target genes (a total of 490) that contained target sites with more than one G:U wobble in the 5′ end are flagged in the Table S2 . The thresholds for candidate target sites were S > 90 and Δ G < −17 kcal/mol, where S is the sum of single-residue-pair match scores over the alignment trace and Δ G is the free energy of duplex formation from a completely dissociated state, calculated using the Vienna package as in Enright et al. (2003) . After finding optimal local matches above these thresholds between a particular miRNA and the set of 3′ UTRs in each genome, we asked whether target site position and sequence for this miRNA were conserved in the 3′ UTRs of orthologous genes, i.e., between human and mouse or rat, or between fugu and zebrafish. The alignments of target sites were generated transitively (UTR→miRNA→UTR) via a shared (or homologous) miRNA. We required that the positions of pairs of target sites in two species fall within ±10 residues in the aligned 3′ UTRs. Conserved target sites with sequence identity of 90% or more (human versus mouse or rat) and 70% or more (zebrafish versus fugu) were selected as candidate miRNA target sites and stored in a MySQL database. Using human as the reference species, we predicted 10,572 conserved target sites (conserved in either mouse or rat) in 4,463 human transcripts, of which 2,307 transcripts of 2,273 genes contained more than one target site. Similarly, using zebrafish as a reference species, we predicted 7,057 conserved target sites (conserved in fugu) in 4,820 zebrafish transcripts. To focus on the strongest predictions, conserved target sites for each miRNA were sorted according to alignment score, with free energy as the secondary sort criterion. In cases where multiple miRNAs targeted the same site on a transcript (or within 25 nt of a site), only the highest scoring, lowest energy miRNA was reported for that site. Functional analysis of targets. To facilitate surveys of target function and analysis of functional enrichment, InterPro domain assignments ( Mulder et al. 2003 ) and GO (molecular function hierarchy) mappings ( Ashburner and Lewis 2002 ) for all human genes were obtained using EnsMart. For each functional class derived from either source, we calculated its degree of under- or overrepresentation, F class , using the log-odds ratio of the fraction of annotated target genes with the same class (F 1 ) and the fraction of all annotated Ensembl human genes with that class (F 2 ): Here, N represents the number of genes of a given functional class for either target genes (N tar ) or all genes (N all ), and C represents the total number of functional classes. To eliminate bias from small counts we did not report assignments that were present in less than 1% of all annotated target genes ( F 1 ≤ 0.01 or F 2 ≤ 0.01). Randomized trials For each random experiment all miRNAs were shuffled by randomly swapping two bases of a miRNA 1,000 times. These shuffled sequences were then searched against human, mouse, and rat 3′ UTR sequences in the same way described for the main analysis, including analysis of conservation of target site sequence and position in orthologous 3′ UTRs. A total of ten randomized experiments were performed. Counts were averaged across all experiments, and the standard deviation and other statistical measures were calculated. Analysis of FMRP-associated mRNAs We compiled a list of 464 gene identifiers of FMRP-associated mRNAs from five different publications ( Brown et al. 2001 ; Chen et al. 2003 ; Denman 2003 ; Miyashiro et al. 2003 ; Waggoner and Liebhaber 2003 ). Among the 464 gene identifiers, 397 identifiers were mapped to the corresponding genes in our 3′ UTR dataset. The remaining 67 genes were not mapped because their published identifiers were obsolete, primarily because of their Affymetrix probeset identification numbers. To identify miRNA regulation of the 397 FMRP-associated mRNAs, these genes were then compared with the set of predicted miRNA targets. CPE motif prediction. We predicted CPE motifs in human, mouse, and rat UTRs. We used a search pattern using four criteria: (1) presence of the CPE motif UUUUAU, (2) presence of the hexanucleotide AAUAAA, (3) the CPE and the hexanucleotide motif being within 100 nucleotides of each other, and (4) the conservation of these motifs and the positions of the motifs in the mouse ortholog ( Mendez and Richter 2001 ). Supporting Information Figure S1 Overrepresentation of the GO and Interpro Domains (347 KB PDF). Click here for additional data file. Table S1 Human miRNAs in Introns (25 KB XLS). Click here for additional data file. Table S2 Predicted Mammalian miRNA Targets by Gene (8.0 MB XLS). Click here for additional data file. Table S3 Predicted Mammalian miRNA Targets by miRNA (17.0 MB XLS). Click here for additional data file. Table S4 Predicted Fish Targets by Gene (5.6 MB XLS). Click here for additional data file. Table S5 Predicted Fish Targets by miRNA (9.8 MB XLS). Click here for additional data file. Table S6 High-Scoring miRNA Matches in Human cDNAs (601 KB XLS). Click here for additional data file. Table S7 High-Scoring miRNA Matches in Human Coding Regions (512 KB XLS). Click here for additional data file. Table S8 Estimate of False Positives (23 KB XLS). Click here for additional data file. Table S9 Predicted Targets That Are Associated with FMRP (678 KB XLS). Click here for additional data file. Table S10 Function of Targets by Interpro and GO Mapping (357 KB XLS). Click here for additional data file. Table S11 Target Genes That Contain Predicted CPE Motifs (529 KB XLS). Click here for additional data file. Table S12 Conserved Vertebrate Target Genes (621 KB XLS). Click here for additional data file. Table S13 Overlap of the Predicted Targets with Validated Gene Targets from Lewis et al. (2003) and Kiriakidou et al. (2004) (68 KB XLS). Click here for additional data file.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC521178.xml

522869

Global nucleosome occupancy in yeast

A genome-wide study of nucleosome occupancy at yeast promoters shows that promoters that regulate active genes, contain multiple conserved motifs, or contain Rap1 binding sites tend to be depleted of nucleosomes.

Background Global gene-expression patterns are established and maintained by the concerted actions of transcription factors and the proteins that constitute chromatin. The global network of interactions between transcription factors and promoters in yeast is increasingly being characterized [ 1 ]. The role of chromatin in gene regulation is less clear, however. For example, the distribution of nucleosomes, the fundamental units of chromatin, is poorly understood on a gene-specific basis, much less a global basis [ 2 ]. The nucleosome consists of approximately 146 base-pairs (bp) of DNA wrapped around an octamer of histone proteins - two each of histones H2A, H2B, H3 and H4. Eukaryotic genomes are packaged into repeating units of nucleosomes separated by around 10-80 bp of linker DNA. High occupancy by nucleosomes is thought to be generally repressive [ 3 ], and extensive remodeling (and loss) of nucleosomes occurs in the promoters of genes undergoing activation [ 4 ]. In the case of the PHO5 promoter in yeast, this remodeling proceeds until essentially no nucleosomes are detected across a region of several hundred base-pairs [ 5 , 6 ]. Transcription factors and chromatin proteins each form complex regulatory networks that interact in a variety of ways [ 1 , 7 ]. Transcription factors modify chromatin structure by recruiting enzymes that remodel nucleosomes or posttranslationally modify histones (by acetylation or methylation, for example) [ 8 - 10 ]. The modifications can be maintained through cell division and propagated to proximal nucleosomes by positive-feedback mechanisms [ 7 , 11 , 12 ]. Hence, a signal such as the activation of a transcription factor can be temporally and spatially transmitted through chromatin. Conversely, chromatin can influence transcription factor function by modulating the accessibility of target binding sites in the DNA [ 13 , 14 ]. We used chromatin immunoprecipitation (ChIP) and DNA microarrays to evaluate nucleosome occupancy levels for essentially all promoters in yeast. Promoters that regulate active genes, contain multiple conserved motifs or recruit Rap1 tend to be relatively nucleosome-depleted. We also used real-time PCR and micrococcal nuclease digestion to show that nucleosomes are depleted in the vicinity of Rap1 consensus sites. This depletion can be partially reversed by the actions of the small molecule rapamycin or by removing Rap1-binding sites. We suggest that other transcription factors have less robust nucleosome-depleting activities than Rap1 and must therefore act collaboratively to gain access to their cognate sites in the DNA. Results ChIP-based assay for nucleosome occupancy Histones are essential components of the nucleosome and efficiently cross-link to nucleosomal DNA. Antibodies against invariant portions of histones have been used previously in ChIP assays to follow nucleosome loss at the yeast PHO5 promoter [ 5 , 6 ]. We extended this approach to evaluate relative nucleosome occupancy at essentially all promoters and other intergenic regions in yeast. DNA associated in vivo with histone H3 was isolated by ChIP using antibody against the carboxy terminus of histone H3 (no posttranslational modifications are thought to occur in this region). ChIP DNA and unenriched control DNA were amplified by in vitro transcription and evaluated using microarrays. DNA associated with histone H2B was evaluated in a similar fashion using anti-FLAG antibody and a FLAG-H2B strain. H3 and H2B datasets were compiled by averaging four and three independent biological experiments, respectively. These datasets are remarkably similar as shown by a genome-wide correlation of 0.83 (Figure 1a-c ). This correlation is comparable to that observed when comparing replicate H3 datasets (or H2B datasets), and suggests that both assays measure similar phenomena. In the H3 and H2B datasets, respectively, there are 347 and 214 regions depleted at least 1.5-fold relative to the average over all intergenics. In contrast, there are just 84 and 6 regions in the respective datasets enriched at least 1.5-fold relative to this average. The relatively narrow range of ChIP enrichment and the negative skew of the data (Figure 1b ) are consistent with the conventional view that the majority of the genome is packaged into nucleosomes with intervening stretches of free DNA such as the activated PHO5 promoter [ 5 , 6 ]. Despite these consistencies, a possible caveat to using ChIP to evaluate nucleosome occupancy is that immunoprecipitation efficiency can depend on epitope accessibility. Rather than having low occupancies, genomic regions depleted in the H3 ChIP might be inaccessible as a result of association with large protein complexes in chromatin. To investigate this possibility, we examined a published chromatin fractionation dataset in which cross-linked chromatin fragments were subjected to phenol-chloroform extraction and DNA that partitioned into the aqueous phase was quantified by microarrays [ 15 ]. Given the polar nature of DNA and the hydrophobic nature of denatured protein, aqueous extraction should generally enrich for free DNA. We found that regions depleted in the H3 ChIP assay overlap extensively with regions enriched by aqueous extraction, but not with regions depleted by aqueous extraction (Figure 1d ). Overall, there is a negative correlation of -0.54 between the H3 ChIP and aqueous-extraction datasets. Although the fractionation data may partially reflect differential cross-linking of lysines in the histone tails [ 15 ], this analysis suggests that regions depleted in the H3 ChIP experiment are relatively protein-free, as would be expected of non-nucleosomal DNA. Nucleosome occupancy correlates inversely with promoter strength As previous studies show that PHO5 activation is accompanied by marked nucleosome loss in the promoter region [ 5 , 6 ], we sought to determine whether nucleosome depletion is a general attribute of active promoters. A total of 4,365 intergenic regions that reside immediately upstream of one or more validated yeast genes were assigned as promoters. Relative transcription rates were determined for each yeast gene from transcript levels measured by array and previously collected mRNA half-life data [ 16 ]. We found an inverse correlation of -0.39 between the enrichment of promoters in the H3 and H2B ChIP assays and the transcription rates of downstream genes (Figure 2a ). Under the conditions examined, PHO5 is not induced and its promoter has an average nucleosome occupancy according to these datasets. To evaluate further the relationship between nucleosome depletion and transcription, we collated a set of 308 nucleosome-depleted promoters on the basis of their relative depletion across the replicate H3 and H2B experiments. Of these nucleosome-depleted promoters, 42% regulate highly active genes (Figure 2b ). These data suggest that there is a systematic relationship between promoter strength and nucleosome depletion. However, as this correspondence is not complete there are likely to be other determinants of nucleosome occupancy. Transcription factor binding motifs are over-represented in nucleosome-depleted promoters To identify additional determinants of occupancy, we sought sequence elements associated with nucleosome depletion. Specifically, we carried out an unbiased search for elements up to 10 bp in length that occur with higher frequency in nucleosome-depleted promoters. Two distinct categories of sequences emerged (Figure 3a ). The first includes poly(dA.dT) elements. Stretches of 10 or more dA.dT nucleotides appear in 38% of depleted promoters, compared with 26% of promoters overall (hypergeometric p < 10 -5 ). dA.dT stretches destabilize nucleosome formation in vitro and in vivo [ 17 , 18 ]. The enrichment of poly(dA.dT) elements in nucleosome-depleted promoters probably reflects, at least in part, this destabilizing influence. As a high proportion of the poly(dA.dT) elements identified in nucleosome-depleted promoters are more than 10 bp long (30% are at least 14 bp), these data do not address the minimum length required for destabilization. However, in vitro studies show that a 16-bp insertion leads to a 1.7-fold increase in accessibility of nucleosomal target sites [ 18 ]. The second sequence element enriched in nucleosome-depleted promoters corresponds to the consensus motif for the Rap1 transcription factor. This motif commonly occurs in the promoters of ribosomal proteins genes and is required for Rap1 binding in vitro and in vivo [ 19 , 20 ]. Some variant of this motif appears in 22% of nucleosome-depleted promoters, compared with just 8% of promoters overall (hypergeometric p < 10 -5 ). Furthermore, multiple Rap1 sites are found in 19% of nucleosome-depleted promoters with Rap1 sites, compared to 8% of promoters with Rap1 sites overall (hypergeometric p < 10 -3 ). These data suggest that Rap1 recruitment may lead to nucleosome loss. Because only the Rap1 consensus site was identified in an unbiased search, we sought to identify additional sequence motifs by incorporating species conservation data. Specifically, we evaluated a set of 71 conserved motifs identified by Kellis and colleagues, a majority of which function in transcription factor recruitment [ 21 ]. Nearly half of these 71 motifs are over-represented in nucleosome-depleted promoters relative to promoters overall, as defined by a hypergeometric p < 0.001. However, many of the implicated motifs appear in the same promoters. For example, nine of the over-represented motifs are associated with filamentation gene promoters [ 21 ]. We therefore considered the possibility that the total number of conserved motifs might be a more relevant predictor of nucleosome depletion. Indeed, we found that 31% of nucleosome-depleted promoters contain at least eight motifs, compared with 11% of promoters overall (hypergeometric p <10 -5 ; Figure 3b ). Furthermore, nucleosome-depleted promoters contain an average of 6.1 motifs, whereas the average promoter contains 3.1 (permutation p < 0.001; Figure 3c ). Next, we sought motifs associated with nucleosome depletion in the absence of multiple motifs, by confining our analysis to promoters containing a maximum of four motifs. This analysis identified just two over-represented motifs, which correspond to the Rap1 and Swi4 binding sites. Hence, although a large number of conserved motifs are enriched in nucleosome-depleted promoters, most appear to be relevant mainly when occurring in combination. Functionally cooperative transcription factors associate with nucleosome-depleted promoters As a majority of the conserved motifs recruit transcription factors [ 21 ], we examined the relationship between transcription factor binding and nucleosome occupancy more directly. Lee and colleagues combined ChIP and microarrays to identify target promoters for essentially all yeast transcription factors under the same conditions used here to evaluate nucleosome occupancy [ 1 ]. For each factor, we determined the significance of overlap between its target promoters and the set of nucleosome-depleted promoters. Of the 113 transcription factors in their database, 31 tend to associate with nucleosome-depleted promoters as defined by a hypergeometric p < 0.001. Rap1 has the most significant association (Figure 4a ), consistent with the enrichment of its binding motif (see above). Other top-ranked factors include Fhl1, which associates with many Rap1-bound promoters, and Swi4, whose binding motif is also enriched (Table 1 ). We sought an underlying binding mechanism or function common to the transcription factors we had identified. However, these factors utilize a variety of binding domains, regulate different pathways, and only a minority have significant associations with promoters of highly active genes. Nonetheless, a commonality does emerge when transcription factor cooperativity is considered. A recent informatics study by Banerjee and Zhang identified 31 functionally cooperative transcription factor pairs (representing a total of 33 factors) on the basis of comprehensive binding and expression data [ 22 ]. Only a fraction of these are known to interact physically, suggesting that other mechanisms also confer cooperative function. There is a remarkable correspondence between these functionally cooperative factors and those that preferentially associate with nucleosome-depleted promoters (see Table 1 ). Of the 31 factors we found to associate with nucleosome-depleted promoters, 17 were found to be functionally cooperative by Banerjee and Zhang ( p < 10 -5 ). Furthermore, an evaluation of nucleosome occupancy at promoters bound by both members of a cooperative pair revealed a significant association with nucleosome-depletion for 18 of the 31 pairs (hypergeometric p < 0.01). Together, these findings suggest that binding motifs and transcription factors act in combination to deplete nucleosomes and suggest a role for nucleosomes in transcription factor cooperativity [ 23 - 25 ]. Conditional nucleosome depletion at Rap1 consensus motifs Although a number of transcription factors appear to act in defining promoter nucleosome occupancy, only the Rap1 consensus motif was identified in an unbiased search of nucleosome-depleted promoters. Furthermore, there is a highly significant association between nucleosome-depleted promoters and promoters bound by this factor in vivo [ 1 ] (Figure 4a ). To investigate the relationship between Rap1 recruitment and nucleosome depletion further, we used ChIP and real-time PCR to evaluate nucleosome occupancy at several Rap1 binding sites in ribosomal protein promoters. We found that these regions are depleted 3- to 10-fold in H3 and FLAG-H2B ChIP assays, relative to a control promoter ( TUB2 ) with average occupancy by global analysis (Figure 4b ). We also used an orthogonal approach in which micrococcal nuclease digestion [ 26 ] was used to probe for nucleosomes at the TUB2 , RPS11B and RPS15 promoters (Figure 4c ). A pattern of nuclease protection indicative of a regular nucleosome array is evident at the TUB2 promoter, consistent with the average nucleosome occupancy attributed to this promoter by global ChIP analysis. In contrast, nuclease protection is not evident at the RAP1 sites in the RPS15 promoter, consistent with the marked nucleosome-depletion attributed to this region by global ChIP and real-time PCR analysis. The region surrounding the RAP1 sites in RPS11B exhibits weak nuclease protection, consistent with the modest nucleosome-depletion attributed to this region by global ChIP and real-time PCR. Although these focused analyses specifically addressed Rap1 sites in ribosomal protein genes, our global analyses indicate that approximately 30% of nucleosome-depleted promoters containing Rap1 motifs do not regulate ribosomal protein genes. Together these data confirm that nucleosomes are markedly depleted in the vicinity of Rap1 consensus sites in vivo , and thus extend previous studies showing that Rap1 induces local alterations in chromatin structure that, for example, result in increased nuclease sensitivity [ 27 - 29 ]. To gain further insight into the relationship between Rap1 and nucleosome depletion, we examined a mutant RPS11B promoter lacking its Rap1 consensus sites. We found that removal of these sites, which completely abrogates Rap1 binding [ 30 ], causes nucleosomes to return to the region, as reflected by a greater than twofold change in H3 ChIP enrichment (Figure 4d ). We also examined the effect of rapamycin treatment on nucleosome occupancy in the vicinity of these consensus sites. Although ribosomal protein gene expression is dramatically reduced by rapamycin [ 31 , 32 ], Rap1 remains bound to its target promoters ([ 30 , 33 ], and B.B., E.P. and S.S., unpublished results). We found that rapamycin treatment causes nucleosomes to return to the vicinity of Rap1 sites, as reflected by twofold and greater increases in H3 ChIP enrichment (Figure 4e ). Together these data show that Rap1 consensus sites are required for conditional nucleosome depletion at ribosomal protein gene promoters. Discussion To gain further insight into the role of nucleosomes in gene regulation, we systematically evaluated promoter nucleosome occupancy in yeast by immunoprecipitating nucleosomal DNA and quantifying enrichment with microarrays. Promoters that are inefficiently immunoprecipitated by general anti-histone antibodies, and are therefore presumed to be relatively nucleosome-depleted, tend to regulate active genes (Figure 2 ). This is consistent with the previous observation that the activated PHO5 promoter is largely devoid of nucleosomes [ 5 , 6 ]. However, as not all nucleosome-depleted promoters regulate active genes, there are most likely to be additional determinants of depletion. An unbiased search for sequence elements enriched in nucleosome-depleted promoters revealed poly(dA.dT) elements, previously shown to destabilize nucleosome formation [ 17 , 18 ], and the Rap1 consensus motif. By incorporating sequence conservation data [ 21 ], more than 30 other enriched motifs could be identified. However, most of these appear to be relevant mainly when occurring in combination. When we limited this analysis to promoters containing four or fewer motifs, all but two of these additional motifs drop out (only the Rap1 and Swi4 consensus sites remain). As the majority of conserved motifs incorporated in this analysis recruit transcription factors [ 21 ], these data suggest that multiple transcription factors act in combination to deplete nucleosomes. This possibility is further supported by our finding that functionally cooperative transcription factors tend to bind nucleosome-depleted promoters. These associations may reflect a mechanistic model in which transcription factors compete collaboratively to displace nucleosomes in order to gain access to target sites in the DNA [ 23 ]. This model was formulated to explain why certain pairs of transcription factors bind cooperatively to proximal target sites in vivo and on a chromatin template, but not to naked DNA [ 23 - 25 ]. This view invokes a broad role for nucleosomes as ubiquitous negative regulators of transcription factor binding and function. We speculate that by promoting synergy among multiple transcription factors and impeding the activities of individual ones, nucleosomes facilitate threshold behavior and filter noise (for example, genetic variation in motif sequence) in the transcriptional regulatory network. Although many factors appear to act in defining promoter nucleosome occupancy, our data indicate that Rap1 has a uniquely important role. Rap1 and its consensus motif are both markedly enriched in nucleosome-depleted promoters. Follow-up studies using real-time PCR and micrococcal nuclease digestion also demonstrate marked nucleosome depletion in the vicinity of Rap1 sites in the promoters of ribosomal protein genes. Moreover, nucleosomes appeared to return when the Rap1 consensus sites in one of these promoters were removed. These findings are consistent with previously described roles for Rap1 in opening chromatin and altering nucleosome positioning [ 27 , 28 ]. However, Rap1 recruitment is not equally associated with nucleosome depletion under all conditions. We find that nucleosomes partially return to the vicinity of Rap1 sites during a rapamycin-induced starvation response [ 34 ], even though Rap1 remains bound ([ 30 , 33 ], and B.B, E.P. and S.S., unpublished results). Hence, the nucleosome loss associated with Rap1 recruitment is most likely to require additional proteins, such as Esa1, a histone acetyltransferase recruited by Rap1 under exponential growth conditions but released in stress [ 30 ]. These findings may also offer insight into the barrier activity previously documented for Rap1 [ 35 ]. Heterochromatin propagation involves the sequential modification of histones in adjacent nucleosomes through positive-feedback mechanisms [ 7 , 11 ]. Certain factors such as Rap1 are able to block this propagation by largely unknown mechanisms [ 36 ]. One model speculates that these barriers create nucleosome-free 'holes' lacking the histone substrate required for heterochromatin propagation [ 29 , 35 ]. By identifying such a 'hole' in the vicinity of Rap1-binding sites in vivo our data support this model. Remarkably, the nucleosomal hole and the barrier function ascribed to Rap1 may be conditional, as nucleosomes return following treatment with the small molecule rapamycin, which activates a starvation response. Heterochromatic silencing has been shown previously to moderate under these conditions [ 37 ]. Hence, we speculate that dynamic influences on nucleosome occupancy may enable Rap1 to define chromatin domains and vary them in response to environmental cues. More broadly, the widespread nucleosome loss observed in the promoters of active genes provides a general caveat for ChIP studies examining posttranslational histone modifications, as a decrease in signal for a histone modification at a promoter undergoing activation may actually reflect nucleosome loss. Similarly, regions that appear relatively hypo-modified by ChIP may actually be nucleosome-depleted. However, this is not the case for low levels of acetylation [ 38 ] and H3 lysine 4 methylation [ 39 ] observed at yeast telomeres, as these regions have high occupancy. The data also provide insight into the maintenance of epigenetic information by histone modifications. Whereas epigenetic memory of a repressed state can be maintained on histones in promoters, memory of an activated state must be maintained on histones outside the promoters, for example in transcribed regions, which may not undergo significant nucleosome loss during activation [ 5 , 6 ]. Methylation of histone H3 at lysines 4 and 36, targeted to transcribed regions in yeast via interactions between RNA polymerase and the methylases [ 39 - 47 ], may represent such 'activating' marks. Materials and methods Chromatin immunoprecipitation (ChIP) DNA associated with histone H3 in vivo was immunoprecipitated with antibodies against the invariant H3 carboxy terminus using a ChIP protocol described previously [ 39 , 48 , 49 ]. Briefly, 45 ml log-phase w303a yeast (OD 600 ~ 1.0) growing in yeast extract/peptone/dextrose (YPD) were cross-linked in 1% formaldehyde for 15 min, washed twice in PBS, resuspended in 400 μl lysis buffer (50 mM Hepes-KOH pH 7.5, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate) and lysed with glass beads. The resulting extract was sonicated to fragment chromatin (4 × 20 sec burst/30 sec rest with a Branson Sonifier 250 at 70% duty, power 3) and centrifuged for 15 min. Solubilized chromatin was then immunoprecipitated with polyclonal antibodies against the carboxy terminus of histone H3 (Abcam or Cell Signaling). A unenriched whole-cell extract sample (WCE) was also retained as a control. After enrichment, cross-links were reversed by incubating samples in 10 mM Tris-HCl pH 8.0, 1 mM EDTA, 1.0% SDS, 150 mM NaCl at 65°C overnight. DNA was purified from ChIP and WCE samples by proteinase K treatment, phenol/chloroform extraction, ethanol precipitation, and incubation with RNAse. DNA associated with histone H2B in vivo was isolated in a similar manner from yeast containing epitope-tagged H2B [ 50 ] using anti-FLAG M2 monoclonal antibodies (Sigma). DNA amplification and hybridization To obtain sufficient quantities for hybridization, immunoprecipitated DNA (from approximately 10 8 cells) and whole-cell extract DNA (unenriched control) were amplified in a linear fashion as described [ 51 ]. Briefly, terminal transferase was used to add poly(T) tails to DNA fragment and a T7-poly(A) adaptor primer was used to incorporate T7 promoters. The reaction products were used as template for an in vitro transcription reaction carried out with the T7 Megascript Kit (Ambion) and RNA samples were purified using an RNeasy Mini Kit (Qiagen). Amplified RNA was reverse-transcribed, incorporating amino-allyl dUTP, and the resulting DNA was fluorescently labeled by incubation with monofunctional reactive Cy5 (enriched sample) or Cy3 (unenriched control) dye as described [ 52 ]. Microarrays containing 6,438 PCR-amplified intergenic regions were prepared as described previously [ 39 , 53 , 54 ]. Mixed Cy5-/Cy3-labeled probe was hybridized to intergenic microarrays for 12-14 h at 60°C, washed and then scanned using a GenePix 4000A scanner with GenePix Pro software (Axon Instruments) as described [ 55 ]. In addition, transcript levels were determined by hybridizing Cy5-labeled mRNA extracted from log phase w303a yeast against Cy3-labeled genomic DNA on microarrays containing 6,218 open reading frames (ORFs), as described previously [ 16 ]. Microarray data processing Cy5 and Cy3 fluorescence were integrated for each feature using GenePix Pro Software (Axon). Data were processed and composite Cy5:Cy3 ratios determined according to protocols at the Stanford Microarray Database [ 56 ]. Correlations between replicate datasets were ~0.8 for all experiments. Composite datasets were log 2 transformed and zero centered before further analysis. The histone H3 ChIP dataset was determined from four independent immunoprecipitations and hybridizations (two each using antibodies from Cell Signaling or Abcam). The FLAG-H2B ChIP dataset was determined from three independent immunoprecipitations and hybridizations. The mRNA dataset was determined from three independent extractions and hybridizations of mRNA against genomic DNA. Relative transcription rates were determined by dividing transcript levels by half-life data collected by Wang and colleagues [ 16 ]. A set of activated promoters was defined as those in the top 10% by mRNA expression level of associated gene, with divergent promoters assigned to the more highly expressed gene. Complete datasets are available online [ 57 ]. Analysis of nucleosome-depleted promoters Z-scores were assigned to each intergenic that reflect depletion across the four H3 and three H2B ChIP experiments, using the formula Z = (x - μ)/σ where x is the average of the replicate measurements, μ is the average of all intergenics and σ is the standard error of the replicate measurements. We defined as nucleosome-depleted the 410 features with the highest Z-scores. This set, which includes 308 promoters, contains nucleosome-depleted outliers and is not inclusive of all promoters that immunoprecipitate with average or lower efficiency. The average aqueous enrichment ratio [ 15 ] for these 308 depleted promoters is 1.7-fold, significantly higher than expected by chance (permutation p < 0.001), consistent with the premise that these promoters are relatively free of nucleosomes. Sequence elements common to nucleosome-depleted promoters were identified by searching between 10 and 500 bp upstream of gene start sites for over-represented sequences up to 10 bp in length using the GeneSpring program suite (Silicon Genetics). Enrichment was confirmed by evaluating the significance of overlap between the set of nucleosome-depleted promoters and the set of promoters containing Rap1 consensus motifs (ACACCCATACAT with up to two mismatches) or poly dA.dT stretches at least 10 bp in length (identified using PatMatch, Saccharomyces Genome Database [ 58 ]). Statistical significances of overlaps between sets are expressed as P -values calculated by a hypergeometric probability model. The P -values reflect the extent to which observed overlaps exceed that expected under the null hypothesis that there is no relationship between the sets [ 59 ]. Where specified, permutation analyses were carried out by generating 1,000 random but representative promoter sets with an Excel macro and used to confirm statistical significance. Lists of promoters containing the 71 conserved motifs [ 21 ] were collated from gene sets available online [ 60 ]. Lists of promoters bound by transcription factors at a significance of p < 0.001 [ 1 ] were collated from data available at [ 61 ]. Real-time PCR Regions approximately 200 bp in size that span one or more Rap1 consensus sites in ribosomal protein gene promoters were amplified from ChIP and unenriched control samples using SYBR green PCR mix (Qiagen) in an MJ Research real-time PCR machine according to the manufacturers' instructions. Fold-ratios that reflect relative enrichment or depletion of a given region in the H3 or FLAG-H2B ChIP assays were determined using the 2 -ΔΔC T method described in the Applied Biosystems User Bulletin. For each region examined, the TUB2 promoter was used as the normalizer (this promoter is used as a control because its occupancy approximates that of the average promoter by global analysis), and the unenriched control sample was used as the calibrator. Each reported ratio represents the average of three independent ChIP experiments analyzed in duplicate by real-time PCR. The following primer pairs were used: RPS22A promoter: 5'-GCCTAAAACGCCCATAAGTT-3' and 5'-ACTGCAAACCCATATTCAAGA-3' RPS15 promoter: 5'-TACACCGCGCGTATAAATCA-3' and 5'-CCCAGCAAGGAGTTTCTCAG-3' RPS11B promoter: 5'-GAAGAAATATTTCCTTGCTGCACC-3' and 5'-AAGGGAAACGTAAAGCTATTGGAC-3' RPL23A promoter: 5'-ATTAACATCTGTACACCCCCAACT-3' and 5'-TACAGTTCGTTTCCTGCC ATATTA-3' TUB2 promoter: 5'-GGCCTAACAGTAAAGATATCCTCC-3' and 5'-GTTGTAGTAGCTGCTATGT CACTC-3' Centromeric vectors containing either a mutant RPS11B promoter lacking the two Rap1 consensus motifs [ 30 ] or an essentially wild-type allele were transformed into wild-type yeast and used in an H3 ChIP assay to evaluate the consequence of removing Rap1 binding sites on nucleosome occupancy. Enrichment was evaluated by real-time PCR using the following primer pair that selectively amplifies the plasmid alleles but not the endogenous RPS11B promoter: 5'-CTGGAAGAAATATTTCCTT GCTCTAG-3' and 5'-AAGGGAAACGTAAAGCTATTGGAC-3'. Micrococcal nuclease assay Log-phase cultures of W303a yeast grown in 450 ml YPD to OD 600 of 1.0 were spheroplasted with zymolase (10 mg in 40 ml volume of 1 M sorbitol, 50 mM Tris pH 7.4, 10 mM β-mercaptoethanol (β-ME), at 30°C for 38 min shaking at 300 rpm), divided into five aliquots, and digested with increasing concentrations (20 U to 320 U) of micrococcal nuclease (Worthington Biochem) in 600 μl 0.5 mM spermidine, 1 mM β-ME, 0.075% NP-40. DNA from digested samples was extracted with phenol twice and chloroform once and precipitated in ethanol. Samples were washed, resuspended in 10 mM Tris pH 7.5, subjected to RNAse treatment, cleaned up with the MinElute kit (Qiagen) and run out in a 1% agarose gel. Following depurination, denaturation and neutralization of the gel, DNA was transferred onto nylon membranes by capillary action and covalently linked to the membranes by UV irradiation. Southern blotting was carried out using a DIG Luminescent Detection Kit (Roche) and DIG-labeled probe generated by PCR using the TUB2 , RPS11B and RPS15 primers described above.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC522869.xml

534789

CP-31398, a putative p53-stabilizing molecule tested in mammalian cells and in yeast for its effects on p53 transcriptional activity

Background CP-31398 is a small molecule that has been reported to stabilize the DNA-binding core domain of the human tumor suppressor protein p53 in vitro . The compound was also reported to function as a potential anti-cancer drug by rescuing the DNA-binding activity and, consequently, the transcription activation function of mutant p53 protein in mammalian tissue culture cells and in mice. Results We performed a series of gene expression experiments to test the activity of CP-31398 in yeast and in human cell cultures. With these cell-based assays, we were unable to detect any specific stimulation of mutant p53 activity by this compound. Concentrations of CP-31398 that were reported to be active in the published work were highly toxic to the human H1299 lung carcinoma and Saos-2 cell lines in our experiments. Conclusion In our experiments, the small molecule CP-31398 was unable to reactivate mutant p53 protein. The results of our in vivo experiments are in agreement with the recently published biochemical analysis of CP-31398 showing that this molecule does not bind p53 as previously claimed, but intercalates into DNA.

Background The tumor suppressor protein p53 protects organisms from malignancy by either inducing programmed cell death or by arresting the cell cycle in response to cellular stress. The intracellular concentration of p53 is tightly regulated at the posttranslational level and the protein is very unstable under physiological conditions. Upon stress, p53 is stabilized and can act as a potent transcription factor that activates a plethora of downstream target genes [ 1 , 2 ]. The p53 target genes can be grouped into classes according to their effect on a cell. One class is represented by p21 CIP , a cyclin dependent kinase inhibitor that is a potent inhibitor of the cell cycle. Another class of p53 target genes, of which bax is the most known representative, mediates p53-induced apoptosis. Other p53 target genes prevent the process of angiogenesis [ 2 ]. Not surprisingly, p53 is inactivated in a wide variety of human cancers [ 1 , 3 ]. Most mutations found in cancers are mis-sense mutations mapping to the central core domain of p53, which confers sequence-specific DNA binding activity to the protein. These mutations can cause destabilization of the core domain and loss of the DNA binding function. Thus, most mutant p53 proteins lack the ability to bind the DNA control elements of their target genes and fail to activate their expression. As a consequence, cells lacking functional p53 are unable to arrest the cell cycle or to undergo apoptosis in response to genotoxic stress. Since lack of p53 function plays such a central role in cancer development and in resistance to treatment, there has been much interest in the search of means and molecules to reactivate mutant forms of p53 [ 4 - 9 ]. A report by Foster et al. [ 7 ] generated special interest since it reported the discovery of a class of small molecules that was able to stabilize the p53 core domain. Not only were these compounds reported to stabilize the active conformation of wild type p53 but they were also shown to stabilize mutant p53 forms and enable them to activate transcription of p53 target genes. While the initial screening was conducted by an in vitro assay, activity of these compounds was subsequently confirmed in cell culture experiments and in a xenograft tumor mouse model [ 7 ]. One of their compounds, termed CP-31398, was reported to increase reporter gene activation by mutant p53 proteins about tenfold in the human p53-null lung carcinoma cell line H1299. We tested CP-31398 in a yeast cell-based assay and in human tissue culture cells. We could not detect any reactivation of mutant p53 in these cellular assays. Our results are in agreement with, and provide support to the results obtained by Rippin et al. [ 10 ], which indicate that CP-31398 intercalates with DNA rather than binding p53. Results The yeast Saccharomyces cerevisiae does not contain p53 homologous proteins. However, it has been demonstrated that p53 expressed in yeast can function as a potent transcriptional activator of artificial genes bearing its specific recognition sequence [ 11 ]. To test different mutant forms of p53 and the potential effect of various molecules on the activity of such mutants, we constructed a yeast strain carrying an integrated bi-directional reporter gene construct in which a single p53 binding site from the human p21 CIP1 promoter [ 12 ] was inserted between the divergent HIS3 and lacZ genes (figure 1A ). The p53-dependent expression of the yeast marker gene HIS3 allows growth selection on media lacking histidine and containing 3-amino-triazole (3-AT), which is a competitive inhibitor of the HIS3 gene product. The p53-dependent activation of this reporter gene is convenient for library screening, while expression of the bacterial lacZ gene allows verification and quantitation of the transcriptional activity of the various p53 forms and putative modulators. Figure 1 Human p53 protein activates transcription from a reporter construct in Saccharomyces cerevisiae . (A) Schematic representation of our yeast reporter construct integrated into our yeast strain. The black circle represents a single p53 responsive element from the human p21 promoter. (B) β-galactosidase assay to measure activation of the lacZ reporter gene. Wild type p53 and the indicated point mutant variants were transformed into the p53 responsive reporter strain and β-galactosidase activity in solution was determined. The activity of wild type p53 was arbitrarily set to 100%. p53R282W and p53V173A showed about 40% of activation compared to wild type p53. No activation of the reporter gene was detected in yeast cells containing the other point mutant variants. Average and standard deviation were determined from three independent experiments. (C) Growth on selective plates containing 20 mM 3-AT depends on expression of the HIS3 reporter gene and correlates with the activation of the lacZ reporter gene. Control plates consist of standard drop-out plates lacking the corresponding growth marker without 3-AT. Growth under selective conditions was dependent on activation of the p53 dependent reporter gene. Transformation of this strain with an episomal plasmid expressing human wild type p53 led to activation of the integrated lacZ and HIS3 reporter genes, which resulted in increased β-galactosidase activity (figure 1B ) and cell growth on plates lacking histidine and containing 20 mM 3-AT (figure 1C ). In contrast, expression of three mutant forms of p53 [ 1 ] with point mutations in their DNA-binding domain that completely abolish sequence-specific DNA-binding activity (p53R175H, p53R248W, p53R273H) did not activate transcription of the reporter genes (figure 1B and 1C , and data not shown). Expression of mutant forms that retain some DNA-binding activity in vitro and in mammalian cells [ 13 ] led to reduction of reporter gene expression compared to wild type p53 (figure 1B ). All p53 variants were expressed to comparable levels, as verified by western blot analysis (data not shown). Thus, the results of these transcriptional assays, taken together with published results of experiments performed in mammalian cells, indicate that the relative transcriptional activity of wild type p53 and the tested derivatives is comparable in yeast and in human cells. Since lack of p53 function plays such a central role in cancer development and in resistance to chemotherapeutic treatment, many efforts have been directed towards trying to reactivate mutant forms of p53 [ 4 - 9 , 14 ]. The report by Foster et al. [ 7 ] generated special interest since it presented the discovery of a small molecule (CP-31398) that was able to stabilize the core domain of p53 in vitro . In addition, this compound was reported to enable some otherwise silent p53 mutants to activate transcription from target gene promoters in cell culture experiments. We tested the effect of CP-31398 on human p53 activity in our p53-responsive yeast strain. Yeast cells expressing either wild type p53 or the mutant p53R173A were grown in media containing increasing concentrations of CP-31398. Activation of transcription of the p53-dependent reporter gene was assessed by measuring β-galactosidase activity in extracts from these cells (figure 2 ). No significant difference in lacZ reporter gene expression was observed between untreated cells and cells that were incubated with increasing concentrations of the compound. Very high concentrations of CP-31398 (500 μg/ml) reduced reporter gene activity, both in the case of wild type p53 expression and in the case of p5R173A expression. Results of growth assays on selective plates to indirectly measure HIS3 expression paralleled our data from the lacZ experiments (data not shown). Figure 2 Treatment with the p53 stabilizing compound CP-31398 shows no effect on reporter gene activity in yeast. Yeast cells expressing wild type p53 (lanes 1–5), p53V173A (lanes 6–10) or empty vector (-, lanes 11–15, white bars) were incubated with the concentrations of CP-31398 indicated (0–500 μg/ml) and expression of β-galactosidase was determined. β-galactosidase activity of wild type p53 without CP-31398 treatment was arbitrarily set to 100%. Yeast cells were treated with CP-31398 for 16 hours. Since these negative results regarding the lack of expected effects of CP-31398 on p53 could be due to our assay system in yeast, we tested CP-31398 in experiments with human tissue culture cells. We transfected the human p53-null H1299 lung carcinoma cell line that was also used for some of the experiments described by Foster et al. [ 7 ] with plasmid DNA expressing either human wild type p53 or the p53R173A mutant together with a reporter plasmid carrying a p53-responsive luciferase gene [ 12 ]. When we treated these cells with CP-31398 in concentrations that were shown to be effective by Foster et al. (5–20 μg/ml and higher concentrations), reporter gene signals decreased and massive cell death was observed (figure 3A and data not shown). Lower concentrations that showed no obvious toxicity to the cells had no significant effect on reporter gene activity. We observed very similar effects when we performed corresponding experiments in the osteosarcoma cell line Saos-2 (p53 null cell line) (data not shown). Figure 3 Treatment of H1299 lung carcinoma cells with CP-31398 provokes massive cell death and p53 independent decline of luciferase reporter gene activity. (A) H1299 cells were transfected with expression constructs for wild type p53 (lanes 1–3) and p53V173A (lanes 4–6). All the samples were cotransfected with a p53-responsive luciferase reporter (p21 luciferase, containing a single p53 responsive p53 binding site from the human p21 promoter, termed WWP-luc, see material and methods) and a constitutive reference β-galactosidase construct (CMV- lacZ ) for normalization. These cells were subsequently incubated with 0, 10, 15 μg/ml CP-31398 respectively and relative luciferase activities were determined. (B) H1299 cells were transfected with an expression construct for the synthetic activator GAL4 -VP16. All samples were cotransfected with a gal4p responsive luciferase reporter (UAS G luciferase) and a reference β-galactosidase plasmid (CMV- lacZ ) for normalization. The control cells were transfected with CMV- lacZ and UAS G luciferase only. These cells were subsequently incubated with 0, 10 and 15 μg/ml CP-31398 and relative luciferase activities were determined. The cells were treated with CP-31398 for 16 hours. Cell death and decreased reporter gene activity was not dependent on the expression of p53 since treatment with CP-31398 of the same cell lines expressing the unrelated activator GAL4-VP16 co-transfected with the respective reporter construct caused similar toxicity and lower reporter gene activity (figure 3B ). We next tested whether CP-31398 might have an effect on endogenous wild type p53 in the human cell line HeLa. These cells express wild type p53 protein, but p53 levels are low because of the presence of the viral HPV E6 protein, which targets p53 for degradation [ 15 ]. We transfected HeLa cells with the same p53-dependent luciferase reporter construct that was used with the other cell lines and treated the cells with increasing concentrations of CP-31398 (figure 4 ). To our surprise, there was a strong increase in reporter gene activation. When we expressed additional human wild type p53 from a transfection plasmid, the signal increased even more (data not shown). In contrast to the previous effect on other cell lines described above, we did not observe any significant cell death in the case of HeLa. Figure 4 Treatment of HeLa cervical carcinoma cells with CP-31398 leads to p53 dependent induction of the luciferase reporter. HeLa cells were transfected with a p53 responsive reporter gene (WWP-luc) and a reference β-galactosidase plasmid (CMV- lacZ ) for normalization. Control cells were transfected with CMV- lacZ alone. The cells were subsequently incubated with CP-31398 (0–10 μg/ml) and relative luciferase activities were determined. Cells were treated for 16 hours. We subjected extracts from HeLa cells treated with CP-31398 to western blot analysis. The p53 signals correlated with increasing CP-31398 concentrations, whereas the actin control signals did not (figure 5A ). These results are consistent with a classical response to genotoxic stress by compounds causing stabilization of p53 [ 16 ]. Figure 5 Western Blot analysis of HeLa cells treated with CP-31398 and daunorubicin. (A) HeLa cells were treated with increasing concentrations of CP-31398 and protein extracts were subjected to SDS-PAGE and subsequent detection with an anti-p53 antibody (DO-1). (B) HeLa cells were treated with the established p53 inducing agent daunorubicin. Protein extracts were subjected to SDS-PAGE and subsequent detection with an anti-p53 antibody (DO-1). Expression of actin is detected as a loading control in experiments 5A and B. We also measured changes in p53 levels in HeLa cells after treatment with increasing concentration of daunorubicin, a known anticancer agent that is highly cytotoxic by a number of proposed mechanisms – intercalation into DNA among them [ 17 ]. We found, as expected, that daunorubicin treatment led to a progressive stabilization of p53 in HeLa cells comparable to the response when cells were treated with CP-31398 (figure 5B ). Discussion We assessed the proposed p53 stabilizing action of CP-31398 in yeast cells and in human cells. CP-31398, a compound isolated in an antibody-based in vitro screen, was reported to stabilize the p53 DNA-binding core domain and to reactivate mutant p53 in vivo [ 7 ]. We were unable to detect any effect of CP-31398 on p53-dependent reporter gene activation by a mutant form of human p53 neither in human cells nor in yeast cells. In our hands, CP-31398 did not stabilize mutant p53 proteins so as to show differences in activation of p53-dependent reporter genes in yeast and in mammalian cells. In addition, concentrations that were shown to be effective in cell culture by Foster et al. [ 7 ] led to extensive cell death. Most importantly, such cell death was independent of p53 expression. The p53 protein expressed within yeast cells functions as a potent transcriptional activator. Reconstitution of transcriptional activation by p53 in a heterologous, yet cellular system such as a yeast cell should be suitable to assess DNA-binding and transcriptional activation activity regardless of posttranslational modifications and other influences that are inevitable when p53 is studied in the context of its regulatory network in mammalian cells. It has been proposed that such posttranslational modifications like acetylation and phosphorylation activate the latent DNA binding activity of p53 by allosteric mechanisms [ 18 ]. However, more recent in vivo and in vitro studies question whether DNA binding itself is regulated at all and suggest that induction of p53 activity primarily occurs at the level of increasing protein concentration within the nucleus [ 16 , 19 , 20 ]. The evident p53 activity in yeast cells, in which the proposed mammalian-specific p53 modifying enzymes are missing, seems to be more readily consistent with the conclusions of such studies. With our system in yeast, we should be able to detect stabilization of the p53 core domain as long as this leads to increased binding of p53 to its specific DNA recognition sequence and subsequent activation of reporter gene expression. Therefore, our yeast system provides a convenient means to screen compound libraries for identifying molecules that can reactivate mutant p53 proteins in a cellular environment. Thanks to the easy genetic malleability of yeast and the lack of endogenous p53-related pathways, cellular screens with this organism should allow not only identification of compounds that can permeate cellular membranes and be active in an intracellular environment but also rapid exclusion of molecules that are not specific for the chosen target. In contrast to the results obtained with the exogenous expression of wild type p53 in yeast cells or with the H1299 and Saos-2 human cells, we observed a strong increase in wild type p53-dependent reporter gene activation in HeLa cells. These cells showed no apparent cell death after treatment with CP-31398. Wang et al. [ 21 ] reported stabilization of wild type p53 and an increase in p53 levels in other cell lines. These observations are consistent with the results we obtained in HeLa cells. These authors also reported that ubiquitination and degradation of wild type p53 is blocked by CP-31398. This effect seems to be specific to mdm2-mediated p53 degradation since HPV (human papilloma virus) E6-mediated degradation of p53 was unaffected. We do not know why we do not see any stabilization of exogenous p53 in H1299 or Saos-2 cells, but it is possible that unspecific toxicity induced by CP-31398 masks the increasing p53-dependent reporter signal. While these results indicate that CP-31398 might stabilize wild type p53, they do not explain the mechanism. Direct interaction and stabilization of p53 is not excluded. However, other explanations seem plausible. Stabilization of the core domain structure by CP-31398 as proposed in the original article should presumably have no effect on p53 protein levels. But p53 levels increase after treatment with CP-31398. Such a response is in line with a classical stabilisation of p53 after genotoxic stress. In contrast, Wang et al. reported that no serine 15 or 20 phosphorylation was detected in their cells after treatment with CP-31398. Interaction with mdm2 was unaffected, but p53 degradation was nevertheless blocked [ 21 ]. Therefore, it remains unclear by which mechanism CP-31398 stabilizes p53; it seems unlikely that core domain stability and DNA binding are influenced by CP-31398 directly. It is interesting to note that CP-31398 can intercalate into DNA as reported by Rippin et al. [ 10 ]. This intercalation is probably toxic to the cell and likely induces a classical p53 response, similar to the known p53 inducer daunorubicin. Our results strongly suggest a classical p53 stabilization through reduced degradation due to genotoxic effects caused by CP-31398. In fact, wild type p53 levels changed quite dramatically in HeLa cells, which are resitant to the apoptotic effects of p53, whereas the other human cell lines did not survive the treatment, probably because they underwent apoptosis in response to CP-31398 [ 22 ]. In support to this interpretation, our control substance daunorubicin showed very similar and expected results as those obtained with CP-31398. Conclusions In contrast to the results reported by Foster et al. [ 7 ], we did not detect any stimulation of mutant p53 activity in vivo by CP-31398, a potential anti-cancer compound. Concentrations of CP-31398 that were reported to be active in the published work were highly toxic to human cells in our experiments. The results of our in vivo experiments are in agreement with the recently published biochemical analysis of CP-31398, which shows that this molecule does not bind p53 as previously claimed, but rather intercalates into DNA. Methods Yeast strains The yeast strain used in our experiments is a derivative of the S. cerevisiae strain JPY5 [ 23 ] ( MAT ura3-52 his3Δ200 leu2Δ1 trp1Δ63 lys2Δ385 ). The p53 responsive yeast strain was constructed by integration of the reporter construct described in the result section and in figure 1A into the HIS3 locus by homologous recombination. The integrating p21 reporter plasmid was linearized with AflII that cuts in the 3' untranslated region (3'UTR) of the S. cerevisiae HIS3 gene. Yeast growth and manipulations Yeast genetic techniques and media were as described in [ 24 ]. For selection of plasmids, dropout media containing all except the specified amino acids were used. Yeast transformation was performed by the lithium acetate procedure [ 25 ]. Recombinant plasmids All p53 forms tested in yeast were expressed from the vector pGAD424 (Clontech, Inc). Wild type p53 was subcloned from a mammalian expression vector with primers containing HinDIII restriction sites by polymerase chain reaction (PCR). The PCR product was introduced into the HinDIII sites of pGAD424, removing the GAL4 AD ORF from pGAD424. All the point mutant p53 variants were generated by assembled PCR with mismatched primer pairs and subsequent cloning into pGAD424 analogous to wild type p53. The yeast reporter plasmid was derived from pDE96 (yeast integrating plasmid, bi-directional HIS3 , lacZ ) [ 26 ] by introduction of a hybridised double stranded oligo containing the p53 responsive element from the p21 CIP1 promoter (p21_sense_SalI 5'-TCG AGC CGT CAG GAA CAT GTC CCA ACA TGT TGA GCT G-3' and p21_anti_XbaI 5'-CTA GCA GCT CAA CAT GTT GGG ACA TGT TCC TGA CGG C-3') into the XbaI and SalI sites of the vector backbone. The plasmid WWP-luc is described in [ 12 ]. The mammalian p53 expression plasmids were constructed by subcloning the HinDIII p53 fragments from the yeast expression vectors into the GAL4 expression plasmid pSCETV- GAL4 (1-93)RV, this resulted in p53 expression under the control of the CMV promoter. The mammalian GAL4 dependent reporter Gal5-luc contains five GAL4 responsive binding sites in front of the luciferase cassette [ 27 ]. Gal4-VP16 is described elsewhere [ 28 ]. Yeast β-galactosidase assay Yeast β-galactosidase assays in solution using permeabilized cells were performed as described in [ 24 ]. Activity was normalized to the number of cells assayed. Mammalian cell culture Cells were obtained from ATCC (American Type Culture Collection, Manassas, Virginia, USA) and cultured according to the recommendations of ATCC. Transient transfection and luciferase assays We used Polyfect ® transfection reagent (Qiagen, Inc) according to manufacturers recommendations for transfection of all cell lines. Cells for luciferase assays and western blotting were harvested by scraping 48 hours after transfection and subjected to three freeze thaw cycles in 100 mM potassium phosphate pH 7.8 1 mM dithiothreitol buffer. Supernatants were clarified by centrifugation (5 min, 13000 rpm) and resuspended in 100 μl extraction buffer. 10 μl of extract was mixed with 100 μl luciferase assay solution (Promega) and analyzed in a luminometer (EG&G Berthold Lumat LB 9507). β-galactosidase assays were performed according to standard methods using 50 μl of the extract and luciferase units were normalized according to β-galactosidase values. All measurements were performed from at least two independent transfections experiments. Western blot analysis and antibodies Protein extracts were prepared as described above. Proteins were separated by SDS-PAGE, electrophoretically transferred to nitrocellulose membranes, and western blotting was performed according to standard procedures. Anti-p53 antibody DO-1 (Santa Cruz Biotechnology, Inc) reacts with an amino terminal epitope mapping between amino acid residues 11–25 of wild type and mutant p53. Anti-actin antibody (I-19; Santa Cruz Biotechnology, Inc) is an affinity purified goat polyclonal antibody raised against a peptide mapping to the carboxy terminus of human actin. Authors' contributions All experimental work was carried out by ST. AB conceived of the study and participated in its design and coordination. Both authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC534789.xml

539290

Antigenized antibodies expressing Vβ8.2 TCR peptides immunize against rat experimental allergic encephalomyelitis

Background Immunity against the T cell receptor (TCR) is considered to play a central role in the regulation of experimental allergic encephalomyelitis (EAE), a model system of autoimmune disease characterized by a restricted usage of TCR genes. Methods of specific vaccination against the TCR of pathogenetic T cells have included attenuated T cells and synthetic peptides from the sequence of the TCR. These approaches have led to the concept that anti-idiotypic immunity against antigenic sites of the TCR, which are a key regulatory element in this disease. Methods The present study in the Lewis rat used a conventional idiotypic immunization based on antigenized antibodies expressing selected peptide sequences of the Vβ8.2 TCR ( 93 ASSDSSNTE 101 and 39 DMGHGLRLIHYSYDVNSTEKG 59 ). Results The study demonstrates that vaccination with antigenized antibodies markedly attenuates, and in some instances, prevents clinical EAE induced with the encephalitogenic peptide 68 GSLPQKSQRSQDENPVVHF 88 in complete Freunds' adjuvant (CFA). Antigenized antibodies induced an anti-idiotypic response against the Vβ8.2 TCR, which was detected by ELISA and flowcytometry. No evidence was obtained of a T cell response against the corresponding Vβ8.2 TCR peptides. Conclusions The results indicate that antigenized antibodies expressing conformationally-constrained TCR peptides are a simple means to induce humoral anti-idiotypic immunity against the TCR and to vaccinate against EAE. The study also suggests the possibility to target idiotypic determinants of TCR borne on pathogenetic T cells to vaccinate against disease.

Introduction Experimental allergic encephalomyelitis (EAE) is an experimentally induced autoimmune disease mediated by T cells. It can be induced in susceptible animals either by immunization with myelin basic protein (MBP) or proteolipid protein PLP, or by immunization with synthetic peptides from the MBP sequence [ 1 ]. EAE can also be initiated by the passive transfer of encephalitogenic, MBP-specific T cell lines or clones [ 2 , 3 ]. In the Lewis rat, EAE is characterized by a self limiting, ascending, hind limb paralysis. Histologically, EAE is hallmarked by perivascular and submeningeal infiltration of inflammatory cells within the brain and spinal cord [ 4 ]. After recovery, animals become refractory to further induction of paralysis by immunization with MBP. Owing to similarities in clinical expression and histopathology, EAE has long been recognized as an animal model for multiple sclerosis, a demyelinating chronic inflammatory disease in humans of unknown origin. For this reason, studies on EAE are thought to elucidate aspects of the pathogenesis and indicate possible ways of immune intervention. EAE is mediated by MHC class II -restricted, MBP-specific CD4 + T lymphocytes bearing an antigen receptor (TCR) variable (V) regions belonging to a limited set of TCR V region gene families [ 5 , 6 ] and restricted Vα-Vβ gene combinations [ 7 ]. Several rational approaches have been used to prevent EAE, including passive transfer of monoclonal antibodies that interfere with the recognition of the MHC, TCR and MBP peptide complex [ 8 , 9 ], antibodies against CD4 [ 10 ] and T regulatory cells [ 11 - 14 ]. Active immunity against attenuated encephalitogenic T cells was shown to prevent the induction of disease [ 15 , 16 ] and vaccination with synthetic peptides of the complementarity-determining regions (CDR) of the TCR of ecephalitogenic T cells, confer resistance to EAE in the rat [ 17 - 20 ]. Together these facts indicated that T cells are crucial to the pathogenesis of EAE and, in converse, immunity to idiotypic determinants of the TCR of encephalitogenic T cells may be protective. Approaches to directly target the TCR of pathogenetic T cells are an attractive direction for therapy and immunointervention as well as an opportunity to further understand the immunological events involved in protection in vivo . However, limitations exist to methods available for TCR vaccination. Vaccination using attenuated encephalitogenic T cells requires that these are specifically expanded in vitro and can only be used in an autologous system. Synthetic peptides, albeit successful in several instances [ 17 - 20 ], offer no tri-dimensional conformation and may even yield to opposite effect, e.g ., worsening of disease [ 21 , 22 ]. Similarly, vaccination with single chain TCR was shown to either prevent or exacerbate EAE in mice [ 23 ]. In previous work from this laboratory we demonstrated the induction of anti-receptor immunity using immunoglobulins (Ig) expressing discrete peptide portions of human CD4 [ 24 ]. We refer to such Ig as antigenized antibodies, i.e ., Ig molecules in which foreign peptide sequences are conformationally-constrained and expressed in the complementority-determining region (CDR) loops [ 25 ]. Immunization with antigenized antibodies is an efficient method to focus the immune response against defined epitopes of foreign antigens. If CDR sequences of TCRs are functionally comparable to Ig idiotypes, antigenized antibodies provide a tool to induce anti-idiotypic responses against TCR. Here, we used antibodies antigenized with TCR sequences as vaccines to control disease. We engineered two antibodies encompassing in the CDR3 of the heavy (H) chain two synthetic peptides from the sequence of rat Vβ8.2 gene product, 39 DMGHGLRLIHYSYDVNSTEKG 59 (CDR2) and 93 ASSDSSNTE 101 (CDR3, VDJ junction), both reported to confer protection against EAE in the Lewis rat [ 17 - 20 ] when used as vaccines. The results show that vaccination with antigenized antibodies expressing sequences of encephalitogenic T cells induces anti-idiotypic immunity against the TCR and high level resistance against EAE. Material and Methods Animals Eight week old, weight-matched female Lewis rats were purchased from Charles River Laboratories (Wilmington, MA). Animals were housed (three rats per cage) in the animal facility of the Universitiy of California, San Diego. Food and water were provided at libitum . Antigenized antibodies The peptide sequences 93 ASSDSSNTE 101 and 39 DMGHGLRLIHYSYDVNSTEKG 59 were engineered into the CDR3 loop of the murine V H 62 gene [ 26 ] according to our published methods [ 27 ]. The antigenized V H was then ligated in plasmid vector containing a human γ1 constant (C) region gene. Transfection of the plasmid DNA was performed on murine J558L cells, a H-chain defective variant of myeloma J558, carrying the rearrangement for a λ1 light (L) chain [ 28 ]. The resulting antigenized antibodies were termed γ1TCR-I and γ1TCR-II, respectively (Figure 1 ). Wild-type transfectoma antibodies γ1WT and γ2bWT [ 26 ] engineered to have the same C and V regions, but lacking the TCR peptides in the CDR3 of the H chain, served as controls. Transfected cells were incubated without selection for 24 hours and then selected in the presence of 1.2 mg/ml G418 (GIBCO). G418-resistant clones secreting high level of Ig were identified by enzyme-linked immunosorbent assay (ELISA) using horseradish peroxidase (HRP)-conjugated goat antibody to human Ig (Sigma) [ 29 ]. Cultures secreting 10–20 μg/ml were selected, expanded, and their supernatants precipitated by (NH 4 ) 2 SO 4 . Antibodies were purified by affinity chromatography on a Protein A-Sepharose column (Pharmacia-LKB, Alameda, CA) equilibrated with 3 M NaCl/1M glycine, pH 8.9. Elution was performed using glycine 0.1 M- HCl/0.5 M NaCl pH 2.8. The eluted fractions were neutralized using 1 M Tris-HCl, pH 8.0, and dialyzed against 0.15 M phosphate-buffered saline (PBS) pH 7.3. The purity of the antibodies was assessed by electrophoresis on a 10% Sodium Dodecyl Sulfate (SDS)-Polyacrylamide Gel (PAGE). Figure 1 Schematic representation of the two V regions antigenized with TCR sequences. In each case the H chain of the antigenized antibody is formed of a murine V H 62 region in which the CDR3 has been engineered to express either 93 ASSDSSNTE 101 or 39 DMGHGLRLIHYSYDVNSTEKG 59 sequence between two Val-Pro (VP) doublets of the unique cloning site in the CDR3 loop of V H 62 . The complete H chain is the product of the fusion of the antigenized V H region with a human γ1C region. The light (L) chain (not shown) is the murine λ1 which is provided by the J558L host cell. (H chain not to scale). Synthetic peptides Synthetic peptide GSLPQKSQRSQDENPVVHF corresponding to amino acid residues 68–88 of guinea-pig MBP [ 30 ], DMGHGLRLIHYSYDVNSTEKG corresponding to amino acid residues 39–59 of rat Vβ8.2 (CDR2 peptide), ASSDSSNTE corresponding to amino acid residues 93–101 of rat Vβ8.2 (CDR3 peptide) rat [ 17 , 18 ], and the (NANP) 3 peptide of Plasmodium falciparum parasite [ 31 ] were all synthesized in the Peptide Synthesis Facility of the Universitiy of California, San Diego. After synthesis peptides were analyzed by HPLC for purity. Peptide KKSIQFHWKNSNQIKILGNQGSFLTKGPS corresponding to residues 21–49 of the extracellular domain of human CD4 was described previously [ 32 ]. Enzyme-linked immunosorbent assay (ELISA) Serum antibodies against antigenized antibodies and their control were determined by ELISA on 96-well polystyrene microtiter plates (Costar, Cambridge, MA) coated (5 μg/ml – 50 μl/well) with γ1TCR-I, γ1TCR-II, γ2bTCR-I proteins in 0.9% NaCl by drying at 37°C. The wells were blocked with a 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS), and then incubated overnight at +4°C with individual rat sera diluted in PBS containing 1% BSA and 0.05% Tween 20 (PBSA). After washing, the bound antibodies were detected by adding peroxidase-conjugated goat antibodies to rat IgG (γ specific) (Biomeda, CA) at 1:500 dilution in PBSA for 1 hour at room temperature. After washing, the bound peroxidase was measured by adding o-phenylenendiamine (100 μl/well) and H 2 O 2 . After 30 minutes, the plates were read in a micro-plate reader (Vmax, Molecular Devices) at 492 nm. Tests were done in duplicate. Antibodies to TCR peptides were detected in ELISA on 96-well polystyrene microtiter plates coated (10 μg/ml) with the Vβ8.2 synthetic peptides 39 DMGHGLRLIHYSYDVNSTEKG 59 and 93 ASSDSSNTE 101 in 0.1M carbonate buffer, pH 9.6, by overnight incubation at +4 C. After blocking unreactive sites, sera (1:25 dilution in PBSA) were added to plates and incubated overnight at +4°C. The bound antibodies and reactive peroxidase were detected as detailed above. Ig reactive with synthetic peptide 21 KKSIQFHWKNSNQIKILGNQGSFLTKGPS 49 of human CD4 were determined on 96-well polystyrene microtiter plates coated (2.5 μg/ml) with peptide 21–49 in 0.9% NaCl by drying at 37 C as previously established [ 32 ]. Briefly, sera (1:400 dilution in PBSA) were incubated overnight at +4°C. After washing, the test was continued as specified above. Plates were read in a micro-plate reader (Vmax, Molecular Devices) at 492 nm. FACS analysis Autoantibodies reactive with the Vβ8.2 + TCR were sought by flowcytometry on the S23B1E11 T cell hybridoma [ 33 ], derived from the fusion of Vβ8.2 + CD4 T lymphoblasts specific for MBP with the murine TCR α/β - BW1100.129.237 thymoma cell line [ 33 ]. For FACS analysis the following procedure was utilized. 10 6 hybridoma T cells in 100 μl of RPMI-1640 containing 1% egg albumin, 0.01% NaN 3 and 10 mM Hepes, were incubated with rat sera (1:10 dilution) for 90 minutes at +4°C. Cells were washed three times with cold RPMI-1640 and subsequently incubated with a fluorescein-isotyocianate (FITC)-conjugated goat antibody (0.5 μg/10 6 cells) to rat Ig (H+L) (Caltag, So. San Francisco, CA) for 20 minutes at +4°C. After incubation, the cells were washed twice, resuspended in 1% paraformaldehyde, and analyzed in a FACS Scan (BD Biosciences). To stain for dead cells, 20 μl of propidium iodide in PBS were added to unfixed cells before FACS analysis. R-phycoerythrin conjugated mouse monoclonal antibody R78 (IgG1, k) specific for the rat Vβ8.2, the kind gift of Pharmingen (San Diego, CA), was used to control for the expression of the Vβ8.2 TCR on S23B1E11 hybridoma cells. In vitro proliferative response Poplyteal, inguinal and paraortic lymph nodes were removed from immunized animals at different times, dissociated and washed in RPMI-1640. Lymph node cells were plated in round-bottom 96-well plates at 2.5 × 10 5 cells/well in the presence of various (10–100 956;g/ml) amounts of antigen in 200 μl of RPMI containing 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, 4 mM glutamine, 0.1 mM non-essential aminoacids, 1 mM sodium pyruvate and 0.5 μM 2-β mercaptoethanol. Cultures were incubated for 72 hours in a 10% CO 2 atmosphere. The evening before harvest 1 μCi/well of [ 3 H]-thymidine was added to each well. Cells were harvested onto glass fiber filters and counted on an automatic Beckman LS 6000IC β-counter. Vaccinations and immunizations schedule Animals were vaccinated with antigenized antibodies (100 μg/rat) in complete Freunds' adjuvant (CFA) divided equally between the posterior paws (25 μl each) and two points in the back subcutaneously. A booster injection (50 μg/rat) in incomplete Freunds' adjuvant (IFA) was given subcutaneously on day 21. EAE was induced on day 28 by immunization with MBP peptide 68 GSLPQKSQRSQDENPVVHF 88 (30 μg/rat) in the anterior paws (25 μl each) in CFA (H37RA 10 mg/ml). Control rats were similarly injected with transfectoma antibody γ1WT or γ2bWT. Rats inoculated with Freunds' adjuvant only served as additional control. Serum samples were collected from the retro-orbital sinus on day 0 before vaccination, day 21 before booster injection, day 28 before EAE induction, and day 50 after recovery from disease. Sera were stored at -20 C until use. Clinical evaluation of EAE EAE was monitored daily by two operators for clinical signs using the following scale: grade 0 = no appreciable symptoms; grade 1 = tail atony; grade 2 = paraparesis; grade 3 = paraplegia; grade 4 = paraplegia with forelimb weakness, moribund state. Typically symptoms of disease began to appear on day 11–13 from the injection of the encephalitogenic peptide. The Disease Index was calculated according to the formula: [(Maximum Score) × (Duration of Disease) × (Incidence)]. Statistical Methods Statistical analyses was performed using the Fisher's test. Results Vaccination with antigenized antibodies and effect on EAE Two antigenized antibodies were engineered to express the CDR3 93 ASSDSSNTE 101 and CDR2 39 DMGHGLRLIHYSYDVNSTEKG 59 sequences, and were termed γ1TCR-I and γ1TCR-II, respectively (Figure 1 ). Rats were vaccinated with an individual antigenized antibody and received a booster injection 21 days later. EAE was induced on day 28 by immunization with the encephalitogenic MBP peptide 68 GSLPQKSQRSQDENPVVHF 88 . As shown in Table 1 , vaccination with both γ1TCR-I and γ1TCR-II reduced disease severity. Rats immunized with γ1TCR-I (group I) had a disease index of 1.8. Within this group, two out of six rats (33%) did not develop disease, one had grade 1 and three had grade 2. None proceeded through grade 3 or 4. Rats immunized with γ1TCR-II (group II) had a disease index of 4.9. Within this group two out of ten rats (20%) did not develop the disease, two had grade 1, four had grade 2 and two had grade 3. In contrast, all fifteen control rats vaccinated with γ1WT or given CFA only (groups III and IV) developed EAE with a disease index ranging between 11.3 and 22.4. Unmanipulated rats immunized with the MBP peptide (group V) developed EAE with a disease index of 25.2. There was a direct correlation between the severity of the disease and its duration. In rats immunized with γ1TCR-I, the disease lasted on average for 2.5 days and in rats immunized with γ1TCR-II 3.8 days. In contrast, in all the other groups (groups III-V) the duration of the disease was significantly longer (6–7 days). Of note, although group III rats had an overall lower score than unmanipulated rats, they differed from rats in group I or group II by the above mentioned parameters and these difference were statistically significant (Table 1 ). CFA did not confer protection. Taken together, these data indicate that active immunity elicited with antigenized antibodies expressing rat Vβ8.2 TCR peptides was effective in markedly reducing the severity of EAE in the Lewis rat. Table 1 Vaccination against antigenized antibodies expressing TCR peptides protects from EAE Severity of Disease* Group No. Rats Immunogen Incidence Max Score (mean ± SD) Duration (mean ± SD) Disease Index I 6 γ1TCR-I 4/6 1.1 ± 0.9 a 2.5 ± 2.2 b 1.8 II 10 γ1TCR-II 8/10 1.6 ± 1.0 c 3.8 ± 2.3 d 4.9 III 10 γ1WT 10/10 2.2 ± 0.9 5.1 ± 1.0 11.3 IV 5 CFA 5/5 3.4 ± 0.9 6.6 ± 1.3 22.4 V 6 - 6/6 3.5 ± 0.5 7.2 ± 1.3 25.2 * EAE was scored according to incidence, severity and duration. Disease index was calculated as follows: Mean Maximum Score × Mean Duration Disease × Incidence. Significance: ( a ) Group I vs Group III p = 0.04 and Group I vs Group V p = 0.0002; ( b ) Group I vs Group III p = 0.009 and Group I vs Group V p = 0.0001; ( c ) Group II vs Group III p = 0.16 and Group II vs Group V p = 0.0005; ( d ) Group II vs Group III p = 0.12 and Group II vs Group V p = 0.001. Antibody responses after vaccination Antibodies in response to the immunogen were assessed by solid-phase ELISA at various times after immunization. As shown in Table 2 , antibody titers against the immunogen developed in each group (group I-III) irrespective of which antibody was used to detect the antibody response in sera. This suggests that the human constant region of the antigenized antibodies is immunogenic in the rat. Antibody titers increased after the booster immunization and after challenge with the encephalitogenic MBP peptide. Control rats (group IV-V) did not mount any antibody response. No reactivity was found on the 19 mer MBP peptide (GSLPQKSQRSQDENPVVHF) used as a control. Anti-TCR (anti-idiotypic) antibodies were tested using two approaches. In the first case, sera of immunized animals were tested on Vβ8.2 synthetic peptides by ELISA. A weak but distinct response was detected in both instances starting on day 21 or 28 (Figure 2 ). Sera from control animals did not react with TCR peptides. Together with the fact that these were tested at a 1:25 dilution it appears that the anti-idiotypic response is weak. In the second case, we tested anti-idiotypic antibodies for their reactivity with the TCR in its native configuration. This was done by flowcytometry using the Vβ8.2 + T cell hybridoma S23B1E11 as the cell substrate. Two out of six rats in group I had a bright cellular staining (Figure 3 ). Reactive antibodies were detectable on day 21, 28 and day 50. Rats immunized with γ1TCR-II (group II) as well control rats (group III-V) were negative. Interestingly, the two rats whose sera reacted with TCR by flowcytometry did not develop symptoms of EAE. Table 2 Detection of antibodies against γ1TCR-I and γ1TCR-II in vaccinated Lewis rats Days After Vaccination a Immunogen Rats (No.) Responders (No.) 0 21 28 50 γ1TCR-I 6 6/6 ≤2.3* 3.9 ± .2 4.2 ± .2 4.5 ± 0.2 γ1TCR-II 10 10/10 ≤2.3 3.7 ± 0.2 4.1 ± 0.1 4.5 ± 0.2 γ1WT 10 10/10 ≤2.3 3.2 ± 0.4 3.9 ± 0.2 4 ± 0.2 CFA 5 0/5 ≤2.3 ≤2.3 2.6 ≤2.3 - 6 0/6 - - ≤2.3 ≤2.3 b γ1TCR-I 6 6/6 ≤2.3 4 ± 0.2 4 ± 0.2 4.6 ± 0.3 γ1TCR-II 10 10/10 ≤2.3 4.1 ± 0.3 4.4 ± 0.3 4.7 ± 0.5 γ1WT 10 10/10 ≤2.3 3.3 ± 0.3 4 ± 0.2 4.2 ± 0.2 CFA 5 0/5 ≤2.3 ≤2.3 ≤2.3 ≤2.3 - 6 0/6 - - ≤2.3 ≤2.3 * Antibody titers are expressed in log10. Sera were tested on microtiter plates coated with each of the TCR antigenized antibody γ1TCR-I (panel a) or γ1TCR-II (pane b). End point dilutions were determined as the last serum dilution binding with an OD ≥ 0.200. Figure 2 Antibody response to TCR peptides following vaccination with antigenized antibody γ1TCR-I or γ1TCR-II 39 DMGHGLRLIHYSYDVNSTEKG 59 tested on the ASSDSSNTE (panel a) or (panel b). The number of rats in each group is that indicated in Table 1. Results are expressed as Log2 ± SD. Figure 3 Sera from rats vaccinated with γ1TCR-I bind Vβ8.2 + T cells by flowcytometry. Vβ8.2 + S23B1E11 T cell hybridoma were used as substrate. Sera were tested at 1:25 dilution. Bound antibodies were revealed using a FITC-conjugated goat antibody to rat Ig. Vaccination with a murine antigenized antibody To explore the importance of foreigness of the constant region on the immunogenicity of the Vβ8.2 peptides we engineered an antigenized antibody with a murine γ2b constant region. Homology search using the BLAST program of the NCBI gene bank indicated that the murine γ2b C region is 56.7% identical to the rat γ2b C region, with a homology of 71% between residues 106 and 333. Because significant protection was found in rats vaccinated with the antibody expressing the 93 ASSDSSNTE 101 peptide (γ1TCR-I), we engineered an antibody with the same V region (γ2bTCR-I). Rats vaccinated with γ2bTCR-I and subsequently immunized with MBP peptide, were protected only partially compared to rats vaccinated with γ1TCR-I (10.2 vs. 1.9) (Table 3 ). Notably, within the six rats immunized with γ2bTCR-I, two were grade ≤ 2 and four developed a grade 3 for an average of two days. On the other hand, three out of six rats immunized with control antibody γ2bWT proceeded through a grade 4 disease. Similarly, all five control rats (group III and IV) developed a grade 4 disease. Of note, although the severity of the disease in group I rats was less than in control group II, the difference was not statistically significant (Table 3 ). All rats developed antibodies to the respective immunogen. However, when compared with the total antibody titer of rats immunized with γ1TCR-I and γ1TCR-II the titers were on average lower at single time points (Table 4 ). All sera reacted with the synthetic peptide 93 ASSDSSNTE 101 starting from day 21 with a progressive increase over time (Figure 3 ). Table 3 Protection against EAE by vaccination with antigenized antibodies with a murine γ2b constant region Severity of Disease* Group No. Rats Immunogen Incidence Max Score (mean ± SD) Duration (mean ± SD) Disease Index I 6 γ2bTCR-I 6/6 2.5 ± 0.8 a 4.2 ± 0.4 b 10.5 II 6 γ2bWT 6/6 3.0 ± 1.3 5.7 ± 1.7 17.1 III 2 CFA 2/2 4 6 24 IV 3 - 3/3 4 7.3 ± 0.6 29.2 * EAE was scored according to incidence, severity and duration. Disease index was calculated as follows: Mean Maximum Score × Mean Duration of Disease × Incidence. Significance: ( a ) Group I vs Group II p = 0.394 and Group I vs Group IV p = 0.051; ( b ) Group I vs Group III p = 0.05 and Group I vs Group IV p = 0.026 . Table 4 Detection of antibodies against γ2bTCR-I in vaccinated Lewis rats Days After Vaccination Immunogen Rats (No.) Responders (No.) 0 21 28 50 γ2bTCR-I 6 6/6 ≤2.3 3.2 ± 0,2 3.7 ± 0.1 4 ± 0.3 γ2bWT 6 6/6 ≤2.3 3 ± 0.3 3.4 ± 0.3 3.6 ± 0.3 CFA 2 0/2 ≤2.3 ≤2.3 ≤2.3 ≤2.3 - 3 0/3 - - ≤2.3 ≤2.3 * Antibody titers are expressed in log10. Sera were tested on microtiter plates coated with γ2bTCR-I. End point dilutions were determined as the last serum dilution binding with an OD ≥ 0.200. Serum antibodies of vaccinated rats bind a synthetic peptide of human CD4 In the attempt to correlate the antibody response after vaccination with protection, the sera of vaccinated rats and their controls were tested on a synthetic peptide corresponding to amino acid residues 21–49 of the first extra-cellular domain of human CD4. This peptide binds Ig irrespective of antigen specificity and heavy chain isotype with an affinity of 10 -5 M (26). It also binds antigen:antibody complexes formed at molar equivalence with an affinity about 100 fold higher [ 31 ]. When the sera of vaccinated rats were assayed on plates coated with the synthetic peptide of human CD4, strong binding was observed by sera from all rats immunized with γ1TCR-I whereas sera from rats immunized with γ1TCR-II or γ1WT bound much less (Figure 5a ). Control sera of groups IV and V did not bind. Binding could be attributed either to a differential property of the two antigenized V regions or to differences in the immune response triggered by the V regions themselves. To distinguish between the two possibilities two experiments were performed. First, we assessed binding of γ1TCR-I and γ1TCR-II on the CD4 peptide. Both bound equally at saturating and non-saturating concentrations (data not shown). Second, we tested sera of rats immunized with γ2bTCR-I considering that, if the effect was due to the immune response against 93 ASSDSSNTE 101 , we would have found similar results. As shown (Figure 5b ), the sera of γ2bTCR-I vaccinated rats all bound to the CD4 peptide comparably to rats vaccinated with γ1TCR-I. This suggests that binding to the CD4 peptide may reflect differences in the type of V regions utilized by the antibodies generated in vivo in response to immunization with the TCR peptide 93 ASSDSSNTE 101 as compared with the TCR peptide 39 DMGHGLRLIHYSYDVNSTEKG 59 or the wild type V region. Further studies will be needed to clarify this issue. Figure 5 Sera from rats vaccinated with γ1TCR-I or γ2bTCR-I bind synthetic peptide 21–49 of human CD4. Proliferative response Spleen cells and draining lymph nodes of rats tested 15 or 30 days after the initial immunization were tested in a proliferative assay against the Vβ8.2 peptides. No proliferative response was detected (data not shown). Discussion In this report we demonstrate that the severity of EAE in the Lewis rat can be greatly attenuated, and in some instances completely prevented, by active immunization with antigenized antibodies expressing amino acid sequences of the rat Vβ8.2 gene product. Immunity against synthetic peptides of the TCR has been shown to be effective in preventing or reducing the severity of EAE in the rat [ 17 - 20 ], suggesting that autoimmunity against the TCR reacting with encephalitogenic sequences of MBP is key to immunoregulatory events. The control of pathogenetic T cells may involve both T cells and antibodies. Autoregulation via T cells in EAE is well established. Thus, spontaneous recovery from EAE is impaired by splenectomy or thymectomy [ 34 ] and EAE can be prevented by vaccination with "attenuated" pathogenic T cells [ 15 ]. Autoregulation in EAE may involve both CD4 + and CD8 + T cells [ 35 - 38 ] as well as suppression by cytolytic T-T interactions [ 39 ]. A prevailing idea has been that in the rat [ 40 ] and in the mouse [ 41 ] idiotypic determinants of the TCR may be autoimmunogenic and contribute to mechanisms of immune regulation leading to protection. On the other hand, at least in a few instances, monoclonal antibodies against these TcR Vβ region [ 9 , 42 , 43 ] or against TCR idiotype [ 44 ] have been shown to block or attenuate disease. Here we show that immunity against idiotopes of antibodies engineered to express TCR peptides is effective in generating anti-idiotypic immunity directed against rat Vβ8.2 TCR gene product. Importantly, this type of immunity protected from EAE. The new approach used herein to induce anti-TCR immunity is based on conventional idiotypic immunization in which antigenized antibodies mimic the immunogenic properties of soluble TCR functioning as a surrogate internal image [ 45 ] in much the same way as previously demonstrated for a non-self antigen [ 31 ]. The present approach is reminiscent of experiments in which induction of anti-idiotypic immunity against TCR with specificity for MHC was obtained by immunization with soluble alloantibodies of relevant specificity [ 46 , 47 ], or by immunization with autologous idiotype positive molecules that are shed from the cell surface in the serum [ 48 ]. Thus, antibodies purposely modified to express selected loops of the TCR obviate the necessity to purify the receptor, isolate idiotypic TCR molecules from the serum, or use antigen-specific T cell blasts. Antibodies reacting with TCR peptides were detected in every vaccinated rat indicating that immunization with antigenized antibodies is an efficient method to induce an anti-idiotypic response specific for a somatic receptor. The fact that only two out of sixteen vaccinated rats developed antibodies against the native receptor detectable by flowcytometry on Vβ8.2 + T cells suggests that cross-reactive anti-idiotypic antibodies may be very low titer. Alternatively, they may be adsorbed on T cells in vivo precluding their detection in the serum. The first possibility is consistent with the self nature of TCR peptides and a predicted paucity of self reactive clonotypes within the natural B cell repertoire. Interestingly, we noted that the anti-idiotypic response against a non-self peptide expressed in an antigenized antibody [ 31 ] is much greater than the one observed here against a self peptide. That only rats vaccinated with the antigenized antibody expressing the 93 ASSDSSNTE 101 sequence developed flowcytometry-reactive autoantibodies could reflect difference in conformation once the two peptides are embedded in the CDR3 loop of an antigenized antibody. For instance, 93 ASSDSSNTE 101 could be better surface exposed and more stably expressed 39 DMGHGLRLIHYSYDVNSTEKG 59 . A computer-assisted comparison of hydrophilicity profiles [ 49 ] of the 93 ASSDSSNTE 101 peptide in the parental. TCR Vβ8.2 gene product and in the antibody V region shows that in both instances the peptide is highly hydrophilic (Figure 5 ). On the other hand, the Vβ8.2 CDR2 region shows a highly hydrophilic profile alternating with large hydrophobic regions of poorly exposed amino acid residues, both in the parental TCR and in the antibody CDR (data not shown). Our data show that although the process of antibody antigenization allows one to conformationally-constrain and express discrete peptide sequences of somatic receptors, the induction of anti-receptor antibodies is not directly predictable. Previously, we demonstrated flowcytometry-reactive antibodies to human CD4 in a high proportion (75 %) of cases [ 24 ]. We conclude that the physical characteristics of a given receptor peptide (e.g., length, hydrophilicity, etc.) likely determine its ability to induce antibodies cross-reactive with the native receptor. Interestingly, rats immunized with the antigenized antibody expressing 93 ASSDSSNTE 101 but not 39 DMGHGLRLIHYSYDVNSTEKG 59 reacted immunologically with a synthetic peptide of human CD4 previously described to bind Ig [ 32 ]. Because the two antigenized antibodies reacted equally with the CD4 peptide and only differ by the composition of their CDR3, we suggest that binding to CD4 by anti- 93 ASSDSSNTE 101 serum antibodies underscores qualitative differences of the immune response between rats immunized with γ1TCR-I and 1TCR-II, respectively. Thus, it appears as if 93 ASSDSSNTE 101 induced a different immune response than 39 DMGHGLRLIHYSYDVNSTEKG 59 . Furthermore, since vaccination with γ1TCR-I also promoted greater protection from EAE, it is tempting to speculate that a component of the anti-idiotypic response against γ1TCR-I is associated with protection. In conclusion, three points have emerged from this study. First, antigenized antibodies expressing conformationally-constrained loops of the Vβ8.2 TCR can be used as vaccines in the prevention of EAE in the Lewis rat. Our new approach to generate anti-TCR immunity, confirms the relevance of anti-idiotypic regulation in controlling rat EAE [ 17 , 18 , 20 ]. Second, since a weak antibody anti-idiotypic response in the apparent lack of a cell proliferative response was associated with protection, it appears as if a humoral anti-TCR response is relevant to protection from disease. Although this contrasts the relevance of T cell immunity in the regulation of EAE in the rat, reports exist to support the idea that humoral immunity is also important [ 20 , 50 , 51 ]. EAE was shown to be prevented or attenuated by passive transfer of serum from rats recovering from EAE [ 52 ], or by passive transfer of monoclonal antibodies against these TCR Vβ region and its idiotypes [ 9 , 42 - 44 ]. However, whether anti-idiotypic antibodies against the TCR predispose to anergy, apoptosis or killing of pathogenetic T cells remains to be determined. Finally, our study indicates that antigenized antibodies can be used as vaccines in conditions where immunopathology and disease involve receptors on somatic cells, and anti-receptor immunity alone could prevent or mitigate a pathological condition. Competing interests The authors declare that they have no competing interests Figure 4 Antibody response to TCR peptide 93 ASSDSSNTE 101 following vaccination with antigenized antibody γ2bTCR-I. The number of rats in each group are not indicated in Table 3. Results are expressed as means of Log2 ± SD. Figure 6 Hydrophilicity profiles of TCR peptides-containing V regions. Hydrophilic profile of the rat Vβ8.2 TCR, amino acid residues 80–130, inclusive of the CDR3 sequence 93 ASSDSSNTE 101 . Figure 7 Hydrophilic profile of the mouse VH62, amino acid residues 80–125, engineered with the 93 ASSDSSNTE 101 peptide of the rat Vβ8.2 TCR-CDR3.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539290.xml

539285

Anti-Plasmodium activity of ceramide analogs

Background Sphingolipids are key molecules regulating many essential functions in eukaryotic cells and ceramide plays a central role in sphingolipid metabolism. A sphingolipid metabolism occurs in the intraerythrocytic stages of Plasmodium falciparum and is associated with essential biological processes. It constitutes an attractive and potential target for the development of new antimalarial drugs. Methods The anti- Plasmodium activity of a series of ceramide analogs containing different linkages (amide, methylene or thiourea linkages) between the fatty acid part of ceramide and the sphingoid core was investigated in culture and compared to the sphingolipid analog PPMP (d,1-threo-1-phenyl-2-palmitoylamino-3-morpholino-1-propanol). This analog is known to inhibit the parasite sphingomyelin synthase activity and block parasite development by preventing the formation of the tubovesicular network that extends from the parasitophorous vacuole to the red cell membrane and delivers essential extracellular nutrients to the parasite. Results Analogs containing methylene linkage showed a considerably higher anti- Plasmodium activity (IC 50 in the low nanomolar range) than PPMP and their counterparts with a natural amide linkage (IC 50 in the micromolar range). The methylene analogs blocked irreversibly P. falciparum development leading to parasite eradication in contrast to PPMP whose effect is cytostatic. A high sensitivity of action towards the parasite was observed when compared to their effect on the human MRC-5 cell growth. The toxicity towards parasites did not correlate with the inhibition by methylene analogs of the parasite sphingomyelin synthase activity and the tubovesicular network formation, indicating that this enzyme is not their primary target. Conclusions It has been shown that ceramide analogs were potent inhibitors of P. falciparum growth in culture. Interestingly, the nature of the linkage between the fatty acid part and the sphingoid core considerably influences the antiplasmodial activity and the selectivity of analogs when compared to their cytotoxicity on mammalian cells. By comparison with their inhibitory effect on cancer cell growth, the ceramide analogs might inhibit P. falciparum growth through modulation of the endogenous ceramide level.

Background Sphingolipids are essential components of eukaryotic cell membranes, predominantly found in the outer leaflet. Sphingosine and ceramide (Figure 1 ) are the two simplest molecules structurally, which belong to the sphingolipid family. Sphingosine represents the sphingoid backbone, and ceramide has a fatty acid linked in a amide bond to sphingosine. Sphingolipid species have two types of functional groups linked to the 1-position, i.e. sphingomyelin (SPM) (Figure 1 ) having a phosphorylcholine group, and a variety of glycolipids having either glucose, galactose, galactosyl-sulfate or oligo-glycosides linked to the sphingosine moiety of ceramide. Figure 1 structures of sphingolipids and analogs Until recently, sphingolipids were primarily considered to be structural components of membranes. However, data accumulated during the last decade have expanded the view of their biological functions. They are now also considered to be key molecules which regulate many functions essential to eukaryotic cells [ 1 - 5 ]. They are involved, for example, in the regulation of membrane fluidity and are part of discrete membrane microdomains or rafts implicated in signalling and trafficking in cells [ 4 , 6 - 8 ]. Interest in sphingolipids was strengthened by an increasing body of evidence demonstrating their role as secondary messengers for intracellular signal transduction pathways that regulate many cellular processes. For example, ceramide accumulates in response to several different inducers such as cytokines, cytotoxic agents or to stressful conditions, which lead to cell cycle arrest or to apoptosis [ 9 ]. Sphingosine is a protein kinase C inhibitor [ 10 ] that inhibits growth or stimulates proliferation, depending upon the cell type [ 11 , 12 ]. Ceramide plays a central role in sphingolipid metabolism [ 13 ]. It can be converted into SPM through transfer of the choline phosphate group from phosphatidylcholine or serves as a precursor for complex sphingolipids (cerebrosides which possess sugar residues and gangliosides which contain sialic acid residues in addition to the carbohydrate units). Moreover, ceramide can be phosphorylated by a distinct kinase and can also be produced by enzymatic hydrolysis of complex sphingolipids. In turn, ceramide can be hydrolyzed to sphingosine and fatty acid by ceramidases. In contrast to yeast and mammalian cells, the current understanding of sphingolipid metabolism and the biological role of sphingolipids in the development of Plasmodium falciparum , the causative agent of malaria, is still limited. Gerold et al. [ 14 ] provided evidence that de-novo synthesis of sphingolipids occurs in the intraerythrocytic stages of the human malaria parasite P. falciparum and can be inhibited by the well established inhibitors of de-novo ceramide biosynthesis, fumonisin B1, cyclo-serine and myriocin [ 15 , 16 ]. However, these compounds are weak inhibitors of parasite growth. Evidence was provided that another pathway for the synthesis of glycosylated sphingolipids exists in P. falciparum [ 14 , 17 ]. The importance of sphingolipid metabolism for parasite development was demonstrated by Haldar's work showing that: ( i ) The parasite contains two distinct forms of SPM synthase, one sensitive to sphingolipid analogs, d,1-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP) or d,1-threo-1-phenyl-2-palmitoylamino-3-morpholino-1-propanol (PPMP) (Figure 1 ), known to inhibit the synthesis of glucosylceramide in mammalian cells [ 18 ], and the second insensitive to them [ 19 ]; ( ii ) These analogs blocked the parasite proliferation in culture by preventing the formation of the tubovesicular network (TVN) that extends from the parasitophorous vacuole to the red cell membrane and delivers essential extracellular nutrients to the parasite [ 20 - 22 ]. Neutral magnesium-dependent sphingomyelinase activity was also identified in P. falciparum [ 23 - 25 ], indicating that a sphingomyelin cycle (ceramide-SPM conversion) exists in Plasmodium . Recently, an increase in the intracellular ceramide content and an activation of parasite sphingomyelinase(s) were found to be associated with the parasite death process as induced by artemisinine and mefloquine [ 26 ]. Given the importance of sphingolipids in many cellular functions and the central role of ceramide in sphingolipid metabolism, the anti- Plasmodium activity of non-natural analogs of ceramides was investigated on the intraerythrocytic development of P. falciparum . Interestingly, a series of analogs containing a methylene (CH 2 -NH) linkage between the fatty acid and the sphingoid-analog core showed considerably higher anti- Plasmodium activity than their counterparts with a natural amide (CO-NH) linkage or than PPMP. The methylene analogs irreversibly blocked parasite development in contrast to PPMP whose effect is cytostatic. Their efficiency in inhibiting parasite growth did not correlate with their potential to inhibit parasite SPM synthase activity, indicating that SPM synthase is not their primary target. Possible mechanisms of action are discussed. Methods Materials D,1-threo-1-phenyl-2-palmitoylamino-3-morpholino-1-propanol-HCl (D,1-threo-PPMP) was purchased from Matreya (Pleasant Gap, PA). 6-((N- (7-nitrobenz-2-oxa-1, 3-diazol-4-yl) amino) hexanoyl sphingosine (NBD-C 6 -ceramide) and N- (4,4-difluoro-5, 7-dimethyl-4-bora-3a, 4a-diaza-s-indacene-3-pentanoyl) sphingosyl phosphocholine (BODIPY-FL-C 5 -ceramide) were obtained from Molecular Probes, Inc. (Eugene, OR). The compounds of Figure 3 and Figure 4 were synthesized according to the procedure described by Dagan et al [ 27 ], using specific starting materials for each analog. The compounds of Figure 2 were synthesized by linking specific fatty acids to the amino group of substituted 1,3-dihydroxy-2-aminophenyl derivatives. The full description of the synthesis of each specific analog will be described in a separate publication. Figure 2 Anti- P. falciparum activity of ceramide analogs having an amide linkage (series A). Figure 3 Anti- P. falciparum activity of ceramide analogs having a methylene or a thiourea linkage (series B). Figure 4 Anti- P. falciparum activity of selected derivatives P. falciparum culture and synchronization P. falciparum strains (FcB1/Colombia, K1/Thailand, F32/Tanzania, W2/Indochina) were maintained in continuous culture on human erythrocytes in RPMI medium containing 7% (v/v) heat-inactivated human serum under an atmosphere of 3% CO 2 , 6% O 2 , 91% N 2 , at 37°C, as described by Trager and Jensen [ 28 ]. Parasite synchronization was performed successively by treatment with 5% (w/v) sorbitol and by concentration in gelatin solution as previously described [ 29 ]. Anti- Plasmodium activity Drug susceptibility assays were performed using a modification of the semi automated microdilution technique of Desjardins et al . [ 30 ]. Stock solutions of test compounds were prepared in DMSO. Drug solutions were serially diluted twofold with 100 μl culture medium in 96-well plates. Asynchronous parasite cultures (100 μl, 1 % parasitemia and 1 % final hematocrite) were added to each well and incubated for 24 hours at 37°C prior to the addition of 0.5 μCi of [ 3 H] hypoxanthine (Amersham, France, 1 to 5 Ci.mmol/ml) per well. After a further incubation of 24 hour, plates were frozen and thawed. Cell lysates were then collected onto glass-filter papers and counted in a liquid scintillation spectrometer. The growth inhibition for each drug concentration was determined by comparison of the radioactivity incorporated in the treated culture with that in the control culture (having the same final % of DMSO) maintained on the same plate. The concentration causing 50% growth inhibition (IC 50 ) and 90% growth (IC 90 ) were obtained from the drug concentration-response curve and the results were expressed as the means ± the standard deviations determined from several independent experiments. The DMSO concentration never exceeded 0.1% (v/v) and did not inhibit the parasite growth. Cytotoxicity test upon human embryonic cells A human diploid embryonic lung cell line (MRC-5, Bio-Whittaker 72211D) was used to assess the cytotoxic effects towards eukaryotic host cells. MRC-5 cells were seeded at 5,000 cells per well in 100 μl. After 24 hours, the cells were washed and two-fold dilutions of the drug were added in 200 μl standard culture medium (RPMI medium + 5% fetal calf serum) and maintained for five days under 5% CO 2 atmosphere. The final DMSO concentration in the culture remained below 0.1%. Untreated cultures were included as controls. The cytotoxicity was determined using the colorimetric MTT assay according to the manufacturer's recommendations (Cell proliferation kit I, Roche Applied Science, France) and scored as a percentage of reduction in absorption at 540 nm of treated cultures versus untreated control cultures. IC 50 values were obtained from the drug concentration-response curve. The results were expressed as the mean ± the standard deviations determined from several independent experiments. The index of selectivity was defined as the ratio of the IC 50 value on MRC-5 to that of P. falciparum . Parasite stage-specific inhibitory effects and reversibility Synchronized cultures (1–2% parasitemia) at the ring stage (0–10 hours old parasites), the trophozoite stage (25–35 hours old parasites) and the schizonte stage (40–48 hours old parasites) were maintained in the presence of drug concentrations in the vicinity of IC 50 values. Aliquots were removed at the indicated times, washed three times with culture medium and maintained in culture in the absence or in the presence of a given drug. Parasite morphology was determined on Giemsa-stained smears defined according to the following criteria: the ring stage, when parasites exhibited a peripheral cytoplasm stained by Giemsa and a unstained intraparasitic vacuole; the trophozoite stage, when parasites showed a fully stained cytoplasm, haemozoin crystals and one nucleus; the schizont stage, when parasites presented several distinctive nuclei. Parasitaemias were determined by counting 3,000 cells for each sample. Controls consisted of parasites incubated with DMSO instead of drugs processed in the same way. Sphingomyelin synthase activity assays SPM synthase activity was measured as described by Haldar et al. [ 31 ]. Briefly, assays were performed on P. falciparum cultures at the trophozoite stage (20–30 h old parasites). 400 μl of culture (1 × 10 8 parasites) were incubated for 60 min at 37°C with 10 μM NBD-C 6 -ceramide and 0 to 500 μM PPMP or AD2646. Cells were then lysed by freezing and thawing of the culture. Lipids were extracted by a modification of the method of Bligh and Dyer [ 32 ]. To each sample, three volumes of a CH 3 OH/CHCl 3 mixture (1:2, v:v) were added and the mixture vortexed for one min. Organic and aqueous phases were separated by centrifugation (12,000 × g, five min) and the organic phase was dried. Lipids were dissolved in 15 μl ethanol and analysed by thin layer chromatography on HPTLC plates (Silica gel 60 F 254 , Merck, Darmstadt, Germany) in CH 3 OH/CH 3 Cl 3 /NH 4 OH (75:25:4, v:v:v). For qualitative analyses, the fluorescent lipids were detected under UV and for quantitative analyses, the fluorescent lipid spots were scraped, eluted in one ml methanol and quantified at an excitation of 470 nm and an emission of 530 nm in a spectrofluorometer. The percentage of SPM synthase activity for each drug concentration was determined by comparison of the fluorescence quantified in the analog-treated culture with that in the control culture (without drug). Labelling of infected red blood cells and fluorescence microscopy Infected erythrocytes treated with or without ceramide analogs were incubated for 30 min, at 37°C, in culture medium containing 10 μM BODIPY-FL-C 5 -ceramide, washed three times with culture medium without serum and fixed overnight, at 4°C, in 3.7% formaldehyde/0.05% glutaraldehyde. Cells were mounted on poly-L-lysine coated slides and viewed using a Nikon Eclipse TE 300 DV inverted microscope with an 100X oil objective mounted on a piezzo electric device using appropriate fluorescence emission filters. Image acquisition (z-series) was performed with a back illuminated cooled detector (CCD EEV: NTE/CCD-1024-EB, Roper Scientific, France) using a 0.2 μm step. Data acquisition and image deconvolution process were performed with Metamorph software (Universal Imaging Corporation, Roper Scientific, France). The images presented correspond to the maximum intensity projection of the deconvoluted z-series. Results and Discussion Anti-Plasmodium activity of non-natural ceramide analogs Non-natural analogs of ceramides were synthesized comprising two functional groups [ 27 ] : 1) A phenyl group substituted on carbon 3 of a sphingoid-like backbone; with the phenyl group replacing the sphingosine acyl chain [ 33 , 34 ] to which were linked nitro or amine groups, or carbon chains of varying lengths; and 2) a fatty acid with an amide (CO-NH) linkage (series A, Figure 2 ), a methylene (CH 2 -NH) or a thiourea (CS-NH) linkages (series B, Figure 3 ) on carbon 2. Analogs in which the alkyl group replaces the amide were investigated because the carbonyl group of ceramide was shown not to be necessary for triggering apoptosis in mammalian cells. In fact, replacement of the carbonyl group of ceramide by a methylene group substantially reduced the time required for cell death [ 35 ]. Only D / L - threo enantiomers were investigated on P. falciparum since reports demonstrated that D / L-erythro enantiomers of ceramide analogs were less efficient in inhibiting glucosylceramide synthase in mammalian cells [ 18 ] and did not inhibit SPM synthase activity in P. falciparum [ 19 ]. Figure 2 and Figure 3 show the IC 50 values obtained for the different compounds on the development of the chloroquine-resistant strain FcB1 of P. falciparum in culture (IC 50 value for chloroquine = 115 ± 25 nM, n = 3). Interestingly, the nature of the linkage considerably influences the anti- Plasmodium activity. Analogs with amide linkage were found to inhibit parasite growth with IC 50 values in the micromolar range (Figure 2 ). Best IC 50 values were similar to that obtained with the ceramide-related compound PPMP (IC 50 = 9.0 ± 1.7 μM, n = 3). However, this IC 50 value for PPMP differed from the previously reported value (IC 50 = 0.85 μM) [ 19 ]. The discrepancy may be due to drug susceptibility assay conditions which were performed on synchronized cultures at the ring stage for Lauer et al. [ 19 ] and on asynchronous cultures in the present study. Analogs with methylene linkages were more efficient than the amide analogs in killing parasites with IC 50 values in the nanomolar range (Figure 3 ). For the D-threo nitro phenyl analogs of series A, no particular increase of the inhibitory activity was observed with the increase of the N -acyl chain length (IC 50 values ranging from 10.8 to 40.4 μM, Figure 2 ). For the series B, best activities were observed for N -alkyl chain length of 12–16 carbons (IC 50 values ranging from 17 to 42 nM for the series B, Figure 3 ). In both series, substitution of the nitrophenyl group by an aminophenyl group instead of nitro group decreased the anti- Plasmodium activity significantly (compare compounds AD2495 and AD2623 of series A, Figure 2 ; and compounds AD2646 and AD2672 of series B, Figure 3 ). Increase of the analog hydrophobicity by substitution of the nitro group of the phenyl ring by alkyl chains seems to decrease the anti- Plasmodium activity of compounds of both series (compare compounds AD2583 and AD2603-7, Figure 2 and compounds AD2646 and AD2677-78-80, Figure 3 ). Surprisingly, in the B series, the anti- Plasmodium activity was restored in compounds with symmetrical alkyl chains of 6–8 carbon length (compounds AD2651 and AD2670, Figure 3 ). No systematic difference in anti- Plasmodium activity was observed between D-threo and L-threo enantiomer of a same analogue: e.g. the enantiomers AD2646 and AD2645 of the B series showed similar activity (Figure 3 ). It can also be noted that ceramide analogs containing a thiourea linkage also showed a significant anti- Plasmodium activity (Figure 3 , compounds AD2215-17) with, however, a less pronounced inhibitory effect than analogs with a methylene linkage. Inhibition of parasite growth by the methylene analog AD2646 was observed having similar IC 50 values on the P. falciparum strains K1 (IC 50 = 45 nM), F32 (IC 50 = 21 nM) and W2 (IC 50 = 28 nM), suggesting that the drug is not restricted to a specific strain and acts through a conserved mechanism in malarial parasites. Furthermore, analysis of drug combination with antimalarial drugs showed that AD2646 has a non-synergistic and non-antagonistic effect with CQ on the CQ-resistant strain K1, and with mefloquine and with artemether on the FcB1 strain (data not shown). Compound AD2646 (Figure 1 ) was selected to further investigate the biological effects of methylene analogs on parasite development. Structure-activity relationship around AD2646 showed that the presence of a nitro group linked to the phenyl is not essential for anti- Plasmodium activity (Figure 4 , compare IC 50 values of compounds AD2646 and AD2730) nor hydroxylation on carbon 1 (compare compounds AD2730 and AD2724). In contrast, hydroxylation of carbon 3 is important for anti- Plasmodium activity since removal of the hydroxyl group reduced the activity 13.5 times (compare compounds AD2730 and AD2729). Cytotoxicity on human cells MRC-5 of ceramide analogs in methylene linkage The cytotoxicity of methylene analogs upon human MRC-5 cells (diploid embryonic lung cell line) was evaluated (Table 1 ). Derivatives tested showed IC 50 values in the micromolar range, from 5 to 8 μM (except for AD2619), which are similar to the IC 50 value of PPMP. No major difference of toxicity was observed between D - and L-threo enantiomers (compare AD2646 and AD2645). In contrast to what was observed for P. falciparum , hydroxylation of the sphingosine carbon 3 does not seem important for cytotoxicity since similar IC 50 values were measured for AD2646 and AD2729, suggesting different mechanism(s) of action for AD2646 on MRC-5 cells and P. falciparum . AD2646 and 4 derivatives show high selectivity for P. falciparum as illustrated by the high index of selectivity of these compounds ranging from 160 to 624. The index of selectivity was defined as the ratio of the IC 50 value on MRC-5 cells to that on P. falciparum . It can be noted that no selectivity was observed for PPMP. A similar range of growth inhibition was measured on P. falciparum (Figure 2 ) and HL-60 cells [ 36 ] with ceramide analogs in amide linkage supporting a weak selectivity of these analogs for P. falciparum . Table 1 Cytotoxicity of methylene analogs and PPMP on human MRC-5 cells Compounds IC 50 (μM) IC 90 (μM) Index of selectivity AD2646 (-) 4.9 7.5 160 AD2645 (+) 6.1 10.3 161 AD2672 (-) 3.7 5.9 2 AD2730 (-) 6.1 9.9 322 AD2729 (-) 5.8 9.8 22 AD2619 (-) 26.7 42.3 624 PPMP 7.5 12.4 0.8 IC 50 and IC 90 values are the mean of three independent experiments. The S.E. were within 10% of the mean. (-): D-threo , (+): L-threo . Index of selectivity is defined by the ratio of the IC 50 value on MRC-5 cells to that on P. falciparum . It must be emphasized that the amide linkage of ceramide analogs is not required for activating apoptosis in cancer cells [ 35 ]. An increase of cytotoxicity of ceramide analogs in methylene linkage compared to their counterparts in amide linkage was also observed on the human histolytic lymphoma U937 [ 35 ] and the human leukaemia HL-60 cells [ 27 ] however, with higher IC 50 values than that observed with P. falciparum . Stage-specific inhibitory effects of AD2646 and reversibility To investigate the cytostatic or cytotoxic effects of AD2646 on the parasite development, cultures at the ring stage (0–10 hours), the trophozoite stage (25–35 hours) and the schizonte stage (40–48 hours) were incubated with 30, 100 or 250 nM of AD2646 for 24.5 hours for the ring stage, for 11 hours for the trophozoite stage, and for 14 hours for the schizonte stage. Aliquots were then taken, washed and incubated in the absence or the presence of drug for a further 13 hours to 24 hours depending upon the parasite stage tested (see Figure 5 ). Parasitaemia and parasite stages were determined on Giemsa-stained smears at time of aliquot removal and after the subsequent incubation. Figure 5 P. falciparum stage sensitivity to AD2646 . Parasites at the ring (A), trophozoite (B) and schizonte (C) stages were maintained in the presence of 30 nM (square) or 100 nM (triangle) AD2646 for 24 h30, 11 h and 14 h, respectively. Aliquots were then taken, washed and maintained in culture in the absence (open symbol) or in the presence (full symbol) of the same concentration of analog. Controls were cultures maintained in the absence of drug (full circle) and processed as the treated cultures. Parasitemia and parasite morphology were determined on Giemsa-stained smears at the indicated time. Each value is the mean of two independent experiments. Development of the ring stage was slightly affected by a continuous incubation with 30 nM AD2646. In contrast, when incubated with 100 and 250 nM, parasite growth was irreversibly blocked at the young trophozoite stage and the parasite degenerated. Drug removal after 24 hours of incubation did not allow a recovery of parasite growth (Figure 5A ). The trophozoite stages were more sensitive to AD2646 since a continuous incubation with 30 nM completely blocked development. Parasites did not enter into division and then degenerated. Only a partial recovery of parasite growth was observed when drug was removed after 11 hours of incubation. A more marked effect was observed with 100 nM AD2646 with degenerated parasites already observed after only 11 hours. No recovery of parasite growth was then observed after drug removal (Figure 5B ). The schizont stage appeared less sensitive than the trophozoite stage since a slight effect was only observed on the parasite development with 30 nM AD2646. However, parasite growth was irreversibly blocked by an incubation with 100 nM AD2646 and parasites degenerated (Figure 5C ). Similar results were observed for the methylene analogs AD2651 and AD2670, the trophozoite stage being the most sensitive with a complete inhibition of parasite development for 250 nM (data not shown). It can be noted that, in contrast to methylene analogs, addition of PPMP to parasite culture led to a preferential and reversible arrest of parasite development at the ring stage. The schizont stage (>30 hours old parasites) was insensitive to this concentration of drugs [ 14 , 19 ]. A cytostatic effect of PPMP on the ring-stage was effectively observed : rings blocked by a 24 hours incubation with 5 μM PPMP recovered to a normal growth after drug removal (data not shown). Blockage of parasite development was associated with the inhibition of a sensitive SPM synthase and TVN formation that delivers extracellular nutrients to the parasite [ 20 - 22 ]. Inhibition of sphingomyelin synthase activity and tubovesicular network formation of P. falciparum by compound AD2646 Figure 6 reproduces the inhibitory effects of PPMP and the methylene analogue AD2646 on the SM synthesis activity of young trophozoite (20–30 hours)-infected erythrocytes maintained in culture. As previously reported [ 19 ], no SPM synthase activity was measured in non-infected red blood cells and a biphasic inhibition curve was observed with PPMP in infected erythrocytes. Two pools of SPM synthase activity are present in parasites with respect to their inhibition by the ceramide analogue, one very sensitive to the drug and the second only inhibited by high concentrations of drug. The biphasic inhibition curve that superimposes on the PPMP inhibition curve was also recorded for AD2646 indicating that PPMP and AD2646 inhibit the SPM synthase activity of infected-red blood cells in a similar way. Figure 6 Inhibition of P. falciparum sphingomyelin synthase activity by AD2646 and PPMP . Trophozoite cultures (20–30 hours aged parasites) were incubated with 0–500 μM PPMP (full square) or AD2646 (open square) and 10 μM NBD-C 6 -ceramide for 60 min, at 37°C. SPM synthase activity was measured as described by Lauer et al. [19]. The percentage of SPM activity was determined by comparison of the activity measured in control cultures maintained without the analogs. Each value is the mean of triplicate experiments. In contrast, PPMP and AD2646 have completely different effects on the TVN formation for drug concentrations that block parasite growth. After 24 hours of incubation, ring development was totally inhibited by 5 μM PPMP and no TVN was observed as previously described [ 20 ] (Figure 7C ). As in controls maintained without drug (Figure 7A ), TVN was distinctly observed after 24 hours of incubation of rings with 60 nM AD2646 (Figure 7B ). This concentration blocks irreversibly the parasite development indicating that AD2646 has no major effect on TVN formation. Figure 7 Effects of AD2646 and PPMP on the formation of the tubovesicular network of P. falciparum . Infected erythrocytes at the ring stage were incubated for 24 hours in presence of 60 nM AD2646 (B) or 5 μM PPMP (C). TVN formation in treated cells and untreated cells (A) was evaluated by membrane staining using BODIPY-Fl-C5-ceramide. Arrow: TVN. Bar: 5 μm. These data do not support the hypothesis of parasite growth inhibition due to an inhibition of the parasite SPM synthase activity as was demonstrated for PPMP [ 19 - 22 ] : 1) The anti- Plasmodium activity of AD2646 does not correlate with its inhibitory activity on the SPM synthase. Although AD2646 and PPMP showed similar inhibitory activity on this enzymic activity in parasites in cultures, AD2646 is about 300 times more efficient in inhibiting parasite development than PPMP; 2) In contrast to PPMP which inhibits the parasite development preferentially and reversibly at the ring stage [ 19 ], AD2646 inhibited parasite development preferentially and irreversibly at the trophozoite stage (Figure 5 ); 3) Inhibition of the SPM synthase activity by PPMP is associated with an inhibition of the TVN formation [ 19 - 22 ]. This was not observed in the presence of AD2646 (Figure 7 ). What could be the mechanism(s) of action of ceramide analogs in methylene linkage on P. falciparum ? By their lipidic nature, these analogs might act through a detergent effect that could lead to lysis or modification of the integrity of infected-erythrocyte membranes. This apparently is not the case. No significant lysis of normal erythrocytes was observed after 48 hous of incubation with concentrations of analogs up to 10 μM (data not shown). Furthermore, no preferential lysis of infected-erythrocytes was observed on Giemsa-stained smears of infected cultures maintained 48 hours with 250 nM AD2646, a concentration inhibiting parasite growth totally. Interestingly, the absence of a fatty acyl carbonyl group (methylene linkage) in our ceramide analogs is a critical factor for the efficacy of their antiplasmodial activity. Sphingolipids preferentially interact with cholesterol in membranes, especially in detergent-resistant microdomains (DRMs or rafts). Rafts have been described in Plasmodium and are involved, at least, in the uptake of erythrocyte raft proteins and maintenance of the parasitophorous vacuole containing the parasite, inside the erythrocyte [ 37 ]. This interaction implies : 1) van der Waals interactions between the saturated acyl chain and sphingoid moiety of sphingolipids and the rigid planar tetracyclic rings of cholesterol [ 38 ] and 2) hydrogen bonds between the 3-β hydroxyl group of cholesterol and the fatty acyl carbonyl group resulting from the amide linkage with the sphingoid moiety [ 39 ]. The amide-linked fatty acid function seems to have a profound stabilizing effect on cholesterol-sphingolipid interactions [ 40 ]. It could be hypothesized that in a membrane context, methylene analogs might have a destabilizing effect on the cholesterol-sphingolipid interactions and, in consequence, modifications of membrane properties. Indeed, P. falciparum growth is characterized by a setting up of new permeabilities of the infected-erythrocyte membrane [ 41 ]. Although the biochemical nature of these new permeabilities is still unknown, they have been characterized from an electrophysiological point of view and involve a malaria-induced anion channel [ 42 , 43 ]. The effect of ceramide analogs was investigated on the properties of this channel. A 24 hours-incubation of infected-erythrocytes with 250 nM AD2646 or 10 μM PPMP did not modulate significantly the induced channel activity measured in the whole-cell configuration of the patch-clamp technique (S. Egee, unpublished data), suggesting that these ceramide analogs do not inhibit parasite growth through modifications of infected-erythrocyte membrane permeabilities. Ceramide is at the parting of different ways of sphingolipid metabolism. Analogs have the potential to inhibit different ceramide-metabolizing enzymes and then might have a pleiotropic effect. Ceramide analogs in amide linkage were described as potent inhibitors of alkaline ceramidase in HL60 human myeloid leukemic cells [ 44 , 45 ]. Methylene analogs inhibit the biosynthesis of SPM and glycosphingolipids in HL60 cells, and acid ceramidase in vitro [ 10 ]. When applied to cancer cells, such analogs induced an elevation of the endogenous level of ceramide with the consequent effects of growth suppression and cell death by apoptosis [ 44 , 45 ]. In contrast to what was observed for cancer cells [ 27 ], preliminary results suggest that the ceramide analog AD2646 induced non-apoptotic death of P. falciparum . Parasites exposed to 1 μM AD2646 for up to 36 hours failed to exhibit characteristic apoptosis, as determined by terminal deoxynucleotidyl transferase DNA fragmentation assay and DNA fragmentation using both gel electrophoresis and fluorescence microscopy methods, although the nucleus appeared highly condensed (M. Dellinger, unpublished data). Apoptosis in P . falciparum is still controversial although some characteristics of apoptosis has been described in Plasmodium [ 46 ]. Recently, an increase in the intracellular ceramide content and an activation of parasite sphingomyelinase(s) were found to be associated with a non-apoptotic parasite death process as induced by artemisinine and mefloquine [ 26 ]. The hypothesis that AD2646 induced parasite death through modulation of endogenous ceramide level, as observed for cancer cells, is under investigation. Authors's contribution ML and PG carried out the in vitro inhibition assays on P. falciparum and MRC-5 cells. MG and MD performed the fluorescence microscopy and apoptosis investigations on P. falciparum , respectively. SE and ST carried out electrophysiological studies on the malaria-induced anion channel. AD, CW and SG participated in the design and synthesis of ceramide analogs. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539285.xml

549038

Cross-talk between phosphatidic acid and ceramide during ethanol-induced apoptosis in astrocytes

Background Ethanol inhibits proliferation in astrocytes, an effect that was recently linked to the suppression of phosphatidic acid (PA) formation by phospholipase D (PLD). The present study investigates ethanol's effect on the induction of apoptosis in astrocytes and the formation of ceramide, an apoptotic signal. Evidence is presented that the formation of PA and ceramide may be reciprocally linked during ethanol exposure. Results In cultured rat cortical astrocytes, ethanol (0.3–1 %, v/v) induced nuclear fragmentation and DNA laddering indicative of apoptosis. Concomitantly, in cells prelabeled with [ 3 H]-serine, ethanol caused a dose-dependent, biphasic increase of the [ 3 H]-ceramide/ [ 3 H]-sphingomyelin ratio after 1 and 18 hours of incubation. As primary alcohols such as ethanol and 1-butanol were shown to inhibit the phospholipase D (PLD)-mediated formation of PA, a mitogenic lipid messenger, we tested their effects on ceramide formation. In astrocytes prelabeled with [ 3 H]-serine, ethanol and 1-butanol, in contrast to t-butanol, significantly increased the formation of [ 3 H]-ceramide. Moreover, exogenous PA, added to transiently permeabilized astrocytes, suppressed ethanol-induced [ 3 H]-ceramide formation. Vice versa, addition of C 2 -ceramide to astrocytes inhibited PLD activity induced by serum or phorbol ester. Conclusion We propose that the formation of ceramide in ethanol-exposed astrocytes is secondary to the disruption of phospholipase D signaling. Ethanol reduces the PA:ceramide ratio in fetal astrocytes, a mechanism which likely participates in ethanol-induced glial apoptosis during brain development.

Background The proliferation of astrocytes is stimulated by polypeptide growth factors such as PDGF, EGF, bFGF and IGF-1 acting on cellular signaling pathways which involve tyrosine kinases, protein kinase C, and the Ras-Raf-MAP kinase pathway [ 1 , 2 ]. Astroglial proliferation is also stimulated by neurotransmitters such as acetylcholine and glutamate [ 3 , 4 ], by direct stimulation of protein kinase C with phorbol ester [ 5 , 6 ], and by peptides such as endothelin and prolactin [ 7 , 8 ]. Astroglial proliferation is prominently inhibited by ethanol both in vivo and in vitro [ 9 - 11 ], and this interference likely contributes to the development of the fetal alcohol syndrome (alcoholic embryopathy) (reviewed in [ 12 ]). Ethanol has been shown to potently antagonize proliferative effects of several individual astroglial mitogens including PDGF, IGF-1, acetylcholine and prolactin [ 8 , 13 - 15 ]. The molecular target of ethanol's antimitogenic actions in astroyctes is not known with certainty, but inhibitory interactions of ethanol with lipid signaling pathways have been implicated [ 15 ]. Our group has recently reported strong evidence that the growth-inhibitory effect of ethanol in astrocytes is caused by the disruption of the phospholipase D (PLD) signaling pathway [ 16 , 17 ]. Under physiological conditions, PLD catalyzes the hydrolysis of phosphatidylcholine (PC) to yield phosphatidic acid (PA) and choline. In the presence of ethanol, however, PLD forms phosphatidylethanol (PEth), a non-physiological phospholipid, at the expense of PA. This PLD-specific phenomenon of transphosphatidylation is the reason why downstream events mediated by PLD activation and PA formation are dose-dependently inhibited in the presence of ethanol (or other primary alcohols such as 1-butanol). In our previous work, we have found that astroglial PLD is activated by mitogenic factors including fetal calf serum (FCS), PDGF, and phorbol ester, and we observed that ethanol reduced both astroglial proliferation and PA formation in a parallel manner. 1-butanol reduced PA formation and DNA synthesis with the same potency while t-butanol was inactive for both effects [ 16 ]. More recently, we demonstrated that exogenous PLD as well as PA, when introduced into the cytosol by transient permeabilization, stimulated astroglial cell proliferation. Importantly, the action of PLD was suppressed in the presence of ethanol (0.3 %, v/v) while the mitogenic effect of PA was not affected [ 17 ]. Thus, disruption of the PLD signaling pathway by ethanol is sufficient to suppress astroglial cell proliferation. Recent findings from other groups are also compatible with a central role for the PLD signaling pathway in ethanol toxicity in astrocytes. First, several mitogenic factors including those that are known to be particularly sensitive to ethanol activate PLD activity in astrocytes. This holds true for PDGF [ 16 ], acetylcholine [ 5 , 18 ], glutamate [ 19 ], phorbol esters [ 5 , 6 ], endothelin [ 20 , 21 ], and prolactin [ 22 ]. In fact, disruption of PLD signaling by ethanol was recently found to be responsible for ethanol's inhibitory effect on astroglial DNA synthesis induced by muscarinic agonist [ 23 ]. Second, PLD is activated via the mitogenic Ras-Ral pathway in many cell types [ 24 ], and PA, the immediate product of PLD activity, interacts with and activates proteins such as Raf kinase, protein kinase Cζ, and mTOR which are known to be central to mitogenic signaling (reviewed in [ 25 , 26 ]). In addition, PA is a precursor of diacylglycerol (DAG), the endogenous activator of classical PKC's, and of lyso -PA, a potent mitogen in many cell types [ 25 , 26 ]. Taken together, current evidence suggests that intact PLD signaling is a prerequisite for the proliferative effects of several mitogens, and that disruption of the PLD pathway by ethanol may be a common theme in ethanol-induced inhibition of astroglial proliferation. The present study in fetal astrocytes was motivated by recent reports that ethanol induces apoptosis in astrocytes, an effect that was accompanied by activation of the sphingomyelinase pathway and formation of ceramide [ 27 , 28 ]. Apoptosis denotes an active cellular program causing cellular death upon contact with toxicants. Apoptotic cell death in the CNS has been under intensive study in recent years and involves several intracellular reaction cascades linked by the activation of caspases (reviewed in [ 29 ]). Apoptosis is almost universally accompanied by the formation of ceramide which may occur through de novo -synthesis, inhibition of ceramide breakdown, or activation of (acidic and/or neutral) sphingomyelinase (SMase), an enzyme which catalyzes the hydrolysis of sphingomyelin to ceramide and phosphocholine (see [ 30 ] for review). Ceramide has emerged as a second messenger for apoptotic pathways targeting kinases and phosphatases which are required for the execution of apoptotic cell death (reviewed in [ 31 , 32 ]). In cerebellar astrocytes and in glioma cells, ceramide levels were found to be reciprocally related to cell proliferation [ 33 , 34 ]. For the present study, we developed the hypothesis that the formation of ceramide may be secondary to the inhibition of PLD signaling which we had described earlier (see above). We now report that ethanol-induced ceramide formation in astrocytes is mimicked by 1-butanol, but not by t-butanol, and that PA, the product of PLD activity, antagonizes ethanol-induced formation of ceramide. We also found that ceramide is a potent inhibitor of stimulated PLD activity. Thus, we obtained evidence of a cross-talk between PA and ceramide, two lipid messengers with opposite effects on cellular proliferation. Results Markers of apoptosis When primary astrocyte cultures were exposed to ethanol (0.3–1%, v/v), staining of the cells with Hoechst 33258, a dye intercalating into DNA, revealed condensation and fragmentation of the nuclei which was visible after 16 hrs; the maximum effect was observed after 21 hours (Fig. 1A ). Higher magnification demonstrated the presence of "apoptotic bodies" in the nuclei (Fig. 1B ). A similar effect was observed after treatment of the cells with the well-known apoptogen, staurosporine (1 μM), or with C 2 -ceramide (50 μM) but not with t-butanol (1%, v/v) (not illustrated). In parallel experiments, ethanol caused fragmentation of nuclear DNA in serum-starved astrocytes which is reflected by "DNA laddering" on agarose electrophoresis (Fig. 2 ). Serum withdrawal alone was not effective while incubation with C 2 -ceramide (50 μM) mimicked the effect of ethanol. Ethanol at 0.3% (v/v) was almost as effective as 1 % (Fig. 2 ). Figure 1 Nuclear fragmentation of astrocytes after treatment with ethanol. In this experiment, astrocytes were incubated with ethanol (1 %, v/v) for 21 hours, fixed in methanol/acetic acid (3:1) and stained with bisbenzimide (Hoechst 33258, 1 μg/ml). The characteristic condensation and fragmentation of nuclei indicates apoptosis. Enlargement: left picture, 400 fold; right picture, 1,000 fold. The experiment was repeated three times with similar results. Figure 2 DNA fragmentation in astrocytes after treatment with ethanol. Astrocytes were incubated with the compounds given: lane 1, control (serum-free medium); lane 2, serum-containing medium; lane 3, C 2 -ceramide (50 μM) in serum-free medium; lanes 4–7, ethanol in serum-free medium (concentrations and times as given); lane 8, size markers. After incubation, cells were lysed, DNA was purified, separated on a 3 % agarose gel and stained with ethidium bromide. The experiment was repeated three times with identical results. Effects of ethanol on sphingomyelin hydrolysis Formation of [ 3 H]-ceramide was measured after labeling sphingomyelin with [ 3 H]serine. Under basal conditions, the ratio of [ 3 H]-ceramide to [ 3 H]-sphingomyelin ("C/S ratio") was approximately 1:30. This ratio was not significantly changed during serum withdrawal (Figs. 3 and 4 ). The incubation of astrocytes with ethanol (1 %, v/v) in serum-free medium caused an increase of [ 3 H]-ceramide (cpm per dish) but did not significantly change the total labeling of the large pool of [ 3 H]-sphingomyelin by [ 3 H]-serine (data not shown). As the total incorporation of [ 3 H]-serine into [ 3 H]-sphingomyelin was somewhat variable between individual preparations, we used the C/S ratio to calculate ethanol-induced changes. As shown in Figs. 3 and 4 , ethanol caused a significant increase of the astroglial C/S ratio in a biphasic and dose-dependent manner. Figure 3 Formation of ceramide in ethanol-treated astrocytes: time course. Astrocytes were labeled with [ 3 H]-serine for 72 hours, washed and treated with ethanol (0.3 %, v/v) in serum-free medium. At the indicated time points, the cells were extracted with methanol/chloroform (2:1), lipid extracts were separated by TLC, and radioactivity associated with [ 3 H]-ceramide and [ 3 H]-sphingomyelin was determined by liquid scintillation counting. Data (N = 3–7) are means ± S.E.M. and are expressed as [%] ceramide/sphingomyelin. Statistics: ANOVA, F 1,53 = 2.28, p = 0.02. *, p < 0.05 vs. control at time zero (Dunnett's post test). Figure 4 Formation of ceramide in ethanol-treated astrocytes: concentration dependence. Astrocytes were labeled with 3 H-serine for 72 hours, washed and treated with ethanol (0.1–1 %, v/v). After (A) 1 hour and (B) 18 hours, the cells were extracted with methanol/chloroform (2:1), phospholipids were separated by TLC, and the radioactivity associated with ceramide and sphingomyelin was determined by liquid scintillation counting. Data (N = 5–6) are means ± S.E.M. and are expressed as [%] ceramide/sphingomyelin. Statistics: one-way ANOVA for repeated measurements, (A) F 3,19 = 3.98, p = 0.03; (B) F 3,23 = 4.88, p = 0.02. *, p < 0.05 vs. controls (Dunnett's post test). The rapid and transient phase of ceramide formation occurred within 15 min and reached a maximum at 1 hour after addition of ethanol (1 h value without ethanol: 2.93 ± 0.35%; 1 h value with 0.3% ethanol: 3.92 ± 0.60%; p = 0.02). A second increase gradually developed after 4 hours and reached a maximum at 18 hours of ethanol exposure (18 h value without ethanol: 3.46 ± 0.41 %; 18 h value with ethanol: 4.18 ± 0.51 %; p = 0.005). At this later time point, staurosporine (1 μM) caused an increase of the C/S ratio to 7.97 ± 1.78 % (p < 0.01; not illustrated). Inhibition of ceramide formation by PLD activity and phosphatidic acid To investigate whether ceramide formation is secondary to a disruption of PLD signaling, we used the isomeric alcohols, 1-butanol and t-butanol. 1-Butanol – but not t-butanol – is a substrate of PLD for transphosphatidylation and leads to the formation of phosphatidyl-1-butanol at the expense of PA. The correlations between butanol exposures and disruption of PLD signaling (i.e., suppression of PA formation) in astrocytes were documented in detail in our previous work [ 16 ]. In the present experiments we measured the differential effects of 1- and t-butanol on the formation of ceramide (C/S ratio) in astrocytes. As shown in Fig. 5 , there was a tendency for an increased level of the C/S ratio after addition of 1-butanol at 0.1% while highly significant increases were observed with 0.3%. In contrast, t-butanol (0.1 and 0.3%) had no effect. This pattern was identical at the 1 hour and 18 hours time points (Fig. 5A and 5B ). This pattern suggested a change of the C/S ratio that was secondary to the disruption of the PLD pathway. Figure 5 Effects of isomeric butanols on ceramide formation in astrocytes. Astrocytes were prelabeled with [ 3 H]-serine for 72 hours, washed and treated with 1-butanol ("1-But") or t-butanol ("t-But"). After (A) 1 hour and (B) 18 hours, the cells were extracted with methanol/chloroform (2:1), phospholipids were separated by TLC, and the radioactivity associated with ceramide and sphingomyelin was determined by liquid scintillation counting. Data (N = 8–10) are means ± S.E.M. and are expressed as [%] ceramide/sphingomyelin. Statistics: Repeated measures ANOVA, (A) F 4,49 = 7.2, p = 0.0002; (B) F 4,39 = 25.0, p < 0.0001. *, p < 0.05; **, p < 0.01 vs. controls ("Ctr"). #, p < 0.05; ##, p < 0.01 vs. effect of 1-butanol (Tukey-Kramer multiple comparisons test). To obtain more direct evidence for this hypothesis, we tested the influence of exogenous PA, the product of PLD, on the C/S ratio. For this purpose, we used a permeabilization procedure which makes use of an oxygen-insensitive mutant (C530A) of streptolysin-O (SL-O) to introduce the membrane-impermeable PA into the astroglial cytosol [ 17 ]. In preliminary experiments, we found that transient permeabilization with SL-O for 15 min by itself did not affect [ 3 H]-ceramide levels (data not shown). Basal ceramide levels were also unchanged if exogenous PA was added to the astrocytes in the absence (data not shown) or presence of SL-O (Fig. 6 ). However, the ceramide formation evoked by ethanol (0.3 %) was significantly reduced in the presence of PA (Fig. 6 ). At the 1 h timepoint, the ethanol-induced effect (C/S ratio: 5.14 ± 0.53 %) was reduced by PA pretreatment to 3.98 ± 0.33 %, a relative reduction by 71 percent (Fig. 6A ). At the 18 h timepoint, PA pretreatment reduced ethanol-induced ceramide formation by 61 percent (Fig. 6B ). Figure 6 Effects of phosphatidic acid on ceramide formation in astrocytes. Astrocytes were prelabeled with [ 3 H]-serine for 72 hours, washed and treated with PA (200 μM) or ethanol (EtOH, 0.3 % v/v) during transient permeabilization with streptolysin-O (144 ng/ml) in calcium-free medium. After 15 min, the cultures were washed and re-exposed to calcium-containing medium to initiate pore repair. After (A) 1 hour and (B) 18 hours, the cells were extracted with methanol/chloroform (2:1), phospholipids were separated by TLC, and the radioactivity associated with ceramide and sphingomyelin was determined by liquid scintillation counting. During the experiments, PA was only present for 15 min during cell permeabilization whereas ethanol was present throughout the incubation period. Data (N = 9) are means ± S.E.M. and are expressed as [%] ceramide/sphingomyelin. Statistics: Repeated measures ANOVA, (A) F 3,35 = 5.52, p = 0.005; (B) F 3,35 = 14.5, p < 0.0001. **, p < 0.01 vs. controls ("Ctr"). #, p < 0.05; ##, p < 0.01 vs. effect of ethanol (Tukey-Kramer multiple comparisons test). Inhibition of phospholipase D activity by ceramide As the previous experiments indicated an inhibitory effect of the PLD pathway on ceramide formation, the following experiment tested a possible effect of ceramide on PLD activity measured by the transphosphatidylation assay. In these experiments, the membrane-permeable C 2 -ceramide (50 μM) slightly but non-significantly reduced basal PLD activity by 30% (Fig. 7 ). However, when PLD activity was stimulated by addition of serum (Fig. 7A ) or by PDB, a phorbol ester and stimulator of protein kinase C (Fig. 7B ), C 2 -ceramide at both 10 and 50 μM strongly and significantly reduced PLD activity. Interestingly, serum-induced PLD stimulation was more sensitive to ceramide than PDB-induced PLD activity; the relative inhibitions for 10 and 50 μM C 2 -ceramide were (for serum stimulation) 84 and 93 % and (for PDB stimulation) 53 and 64 %, respectively. Figure 7 Inhibition of phospholipase D activity by ceramide. PLD activity in serum-starved, [ 3 H]-glycerol-labeled astrocytes was determined by the transphosphatidylation assay; in the presence of ethanol, PLD converts [ 3 H]-phosphatidylcholine (PC) into [ 3 H]-phosphatidylethanol (PEth) which reflects PLD activity. In (A), PLD activity was stimulated by addition of medium containing 10% fetal calf serum (FCS) for 5 min. In (B), PLD activity was stimulated by 4β-phorbol-12β,13α-dibutyrate (PDB; 1 μM) for 30 min. C 2 -ceramide ("C 2 ") was added to the cultures in concentrations of 10 and 50 μM 45 min before the addition of FCS and PDB, respectively. Data are means ± S.E.M. of 7–8 experiments and are expressed as [%] PEth/PC. Statistics: ANOVA, (A) F 4,36 = 14.0, p < 0.0001; (B) F 4,37 = 75.2, p < 0.0001. **, p < 0.01 vs. controls. ##, p < 0.01 vs. stimulated PLD activity (Tukey-Kramer multiple comparisons test). Discussion We have previously documented that ethanol suppresses signaling through the mitogenic phospholipase D (PLD) pathway, and we and others have provided evidence that this effect may be responsible for the antiproliferative actions of ethanol in astrocytes [ 16 , 17 , 23 ]. The present study was motivated by findings that ethanol induces astroglial apoptosis via activation of the sphingomyelinase pathway [ 27 , 28 ]. We confirm these earlier reports by showing the induction of apoptotic markers and ceramide formation in ethanol-treated astrocytes. The novel finding of our study is that ethanol-induced formation of ceramide is reciprocally regulated by phosphatidic acid (PA) and the phospholipase D pathway which is itself inhibited by ceramide. We used three different approaches to demonstrate that apoptotic cell death in astrocytes can be induced by exposure to ethanol. First, we report that ethanol can induce nuclear condensation and degradation (Fig. 1 ). Second, application of ethanol to serum-starved astroglial cultures caused "DNA laddering", a typical hallmark of apoptotic degradation of nuclear DNA (Fig. 2 ). The effect of ethanol was mimicked by a cell-permeable ceramide, C 2 -ceramide (Fig. 2 ), and by staurosporine (not illustrated). Third, ethanol induced an increase of ceramide in astroglial cultures (Figs. 3 and 4 ). Our findings clearly confirm that ethanol can induce apoptosis and ceramide formation in astrocytes, a finding which is in agreement with some [ 28 ] but not all [ 35 ] previous studies. We performed a time course of [ 3 H]-ceramide formation after ethanol exposure and found that it was maximal at 1 and 18 hours. Biphasic formations of ceramide such as those observed here have been described previously in a range of peripheral cell types although their significance is unclear; they may reflect different modes of ceramide formation, or different pools of ceramide [ 36 ]. At both time points of maximum ceramide formation (1 hour and 18 hours), we observed the same ethanol-evoked enhancement of ceramide formation. Importantly, apoptotic cell death and ceramide formation were induced by ethanol levels as low as 65 mM which corresponds to blood alcohol levels (0.3%) which are found in heavy drinkers. It should be noted that the present experiments do not unequivocally identify the mechanism of ceramide formation. We used [ 3 H]-serine to pre-label sphingomyelin for 72 hours, removed the precursor, and measured formation of ceramide as an increase of the [ 3 H]-ceramide/ [ 3 H]-sphingomyelin ratio during incubations with ethanol. This ratio most likely reflects the activity of sphingomyelinase(s), and sphingomyelinase activity was actually shown to be responsible for ethanol-induced ceramide formation in a recent study [ 28 ]. However, our present data do not exclude alternative pathways of increased ceramide formation such as de novo -synthesis of ceramide or inhibition of ceramidase. The important findings of this study relate to the interaction between lipid second messenger pathways. It was known from previous work (see Introduction) that PA, the product of phospholipase D (PLD), mediates mitogenic stimulation in astrocytes whereas formation of ceramide by sphingomyelinase activation accompanies apoptosis. We now tested the hypothesis that ethanol causes astroglial apoptosis by inhibiting PLD and, as a consequence, stimulates the sphingomyelinase pathway. The results shown in Figs. 5 and 6 are evidence of a direct inhibitory influence of the PLD pathway on ceramide formation. First, we observed that the addition of 1-butanol, a primary alcohol which suppresses PLD signaling, caused an increase of ceramide levels (Fig. 5 ). This effect was not seen with the inactive isomer t-butanol which does not interfere with PLD signaling in astrocytes (see our previous study [ 16 ]). Second, we used transient permeabilization of astrocytes by streptolysin-O to introduce PA, the product of PLD, into the astroglial cytosol [ 17 , 37 ]. We found that exogenous PA almost completely prevented the ethanol-induced increase of ceramide at early (1 hr) and delayed (18 hrs) phases of ceramide formation (Fig. 6 ). The fact that ethanol and 1-butanol, but not t-butanol increase ceramide formation, whereas PA antagonized this effect, gives strong evidence that PLD-mediated formation of PA keeps ceramide levels low under basal conditions (Fig. 5 ), and that PLD activity antagonizes ceramide formation under the influence of toxicants (Fig. 6 ). Unfortunately, we could not determine the effect of exogenous PA on astroglial apoptosis because the permeabilization procedure was found to induce a delayed apopototic response in astrocytes (not shown). We also probed the reciprocal effects of ceramide signaling on PLD activity. The results shown in Fig. 7 demonstrate that C 2 -ceramide inhibits PLD activity at a concentration as low as 10 μM. Basal PLD activity was only slightly inhibited, but the increases of PLD activity induced by addition of serum or phorbol ester were strongly antagonized. This finding in astrocytes corroborates previous reports that ceramide can inhibit PLD signaling in peripheral cell types [ 38 , 39 ]. It remains a matter of speculation why serum-induced PLD was somewhat more susceptible to inhibition by ceramide than phorbol ester-stimulated activity. Growth factors in serum and phorbol ester may affect different signaling pathways leading to PLD activation, and we previously presented inhibition data with bacterial toxins which supported this idea for astroglial PLD [ 40 ]. Interestingly, ethanol was also observed to inhibit serum- and growth factor-mediated astroglial proliferation more effectively than phorbol ester-induced proliferation [ 6 , 16 ]. At this time, we cannot distinguish which isoform of PLD is responsible for PA formation; previous attempts to selectively down-regulate astroglial PLDs failed due to the long biological half-lives of the proteins [ 41 ]. The molecular target of ceramide for inhibiting PLD also remains to be identified; previous work has implicated direct inhibition of PLD by ceramide as well as upstream molecules such as protein kinase C which activate PLD [ 42 , 43 ]. Conclusions In summary, the present results give evidence of a cross-talk between lipid-signaling pathways in astrocytes such that the product of PLD, namely PA, inhibits ceramide formation whereas ceramide inhibits PLD activation (Fig. 8 ). The experimental evidence suggests that the ratio "PA:ceramide" contributes to the decision whether astrocytes proliferate or undergo apoptosis. Our data suggest that ethanol induces astroglial apoptosis during brain development by disrupting PLD signaling, thereby reducing PA and increasing ceramide formation. This effect likely contributes to the microencephaly and delay of brain development observed in fetal alcohol syndrome. Figure 8 Hypothetical cross-talk between phospholipase D and sphingomyelin pathways in astrocytes, and effect of ethanol. Phosphatidic acid (PA), the product of phosphatidylcholine (PC) hydrolysis by phospholipase D (PLD), inhibits hydrolysis of sphingomyelin (SM) by sphingomyelinase (SMase). Vice versa, ceramide (Cer), the product of SM hydrolysis by SMase, inhibits activation of PLD. Ethanol induces apoptosis by disrupting the mitogenic PLD signaling pathway thereby decreasing the PA:Cer ratio and disinhibiting the pro-apoptotic SMase pathway. Methods Materials [ 3 H]-Serine and [ 3 H]-glycerol were from Biotrend (Köln, Germany). Ceramide (from bovine brain), C 2 -ceramide (N-acetyl-D- erythro -sphingosine), and staurosporine were from Alexis (Lausen, Switzerland). Hoechst 33258, L-α-phosphatidic acid (sodium salt, from egg yolk) and 4β-phorbol-12β,13α-dibutyrate (PDB) were from Sigma (Deisenhofen, Germany); most other chemicals and TLC plates were obtained from Merck (Darmstadt, Germany) or Roth (Karlsruhe, Germany) at the highest purity available. Fetal calf serum (South America) was from Invitrogen, cell culture materials were from Sarstedt (Nürnbrecht, Germany). Phosphatidylethanol (PEth) standard was synthesized as described [ 16 ]. Recombinant, oxygen-insensitive streptolysin-O (C530A) was prepared as described [ 44 ]. Cell culture Astrocyte-rich cultures were prepared from newborn rat pups. Cerebral hemispheres were collected, meninges and blood vessels were removed, the brain tissue was dissociated by trituration, passed through a 50 μm nylon mesh, and the cells were seeded onto plastic culture dishes (14,000 cells per cm 2 ). The growth medium was DMEM containing 10% fetal calf serum (FCS), 2 g/l NaHCO 3 , 100 U/ml penicillin and 100 μg/ml streptomycin. The cells were incubated at 37°C in a 95/5% mixture of air and carbon dioxide. For the experiments, astrocytes were grown for two weeks in culture and were used when they reached confluency. As judged by GFAP immunostaining, these cultures contained >90% astrocytes. Fluorescence microscopy Cells were seeded on microplates (5,000 cells per 200 μl) and incubated with different apoptogens (ethanol 0.3 and 1 %, C 2 -ceramide 50 μM or staurosporine 1 μM) for 24 hrs in serum-free medium. Subsequently, cells were washed and fixed with ice-cold methanol/acetic acid (3:1). Dried and re-hydrated cells were stained with bisbenzimide (Hoechst 33258, 1 μg/ml) solution for 10 min, washed and sealed in gelatin. Photographs were obatined using a Leica Leitz DMRB fluorescence microscope and a Nikon Digital Camera DXM. DNA fragmentation The test was carried out as described [ 45 ] with some modifications. Briefly, cells were incubated with apoptogens (Ethanol 0,3%, 1%; C2-Ceramide 50 μM) in serum-free medium. Then, the cells were transferred and resuspended in Tris-EDTA buffer containing 0.5% Igepal CA 630. Further lysis was performed in buffer containing RNAse A (100 μg/μl), proteinase K (0,5 μg/ml) and SDS (1.2 %). After 5 minutes, the clear solution was mixed with 3 M CsCl in acetate buffer. Precipitated debris and chromosomal DNA was removed by centrifugation, and the supernatant was loaded onto a QIAprep column (twice), centrifuged and eluted by hypotonic Tris-EDTA buffer. The eluate was analyzed by electrophoresis on a 3 % agarose gel and visualized with ethidium bromide. Measurement of [ 3 H]-ceramide formation Phospholipids were labeled by addition of [ 3 H]-serine (1 μCi/ml) to astrocytes kept in growth medium (DMEM plus FCS) for 72 h. After washing, cells were incubated in serum-free medium with ethanol, butanol, C 2 -ceramide or staurosporine (see Results). C 2 -ceramide or staurosporine were added in DMSO; the final DMSO concentration was < 0.1 %. At the end of the incubation, cells were fixed with methanol, transferred and extracted first with with chloroform: methanol (1:1), then with chloroform: methanol: water (10:10:9) to separate water and lipid phases. After addition of ceramide and sphingomyelin standards, the lower (lipid) phase was evaporated, taken up in chloroform:methanol (3:2) and separated by thin-layer chromatography (TLC). For the determination of [ 3 H]-ceramide, we used one-dimensional TLC (HP-TLC plates Merck 11845; eluent: chloroform/acetic acid 9:1). For the determination of [ 3 H]-sphingomyelin, we used two-dimensional TLC (TLC plates Merck 1.05721); solvent I was chloroform/methanol/25% aqueous ammonia (13:7:1), solvent II was chloroform/ methanol/water/acetic acid (30:30:2:5). 3 H-lipids were visualized by using iodine vapor, spots were scraped and suspended in scintillation cocktail (Lumasafe Plus), and radioactivity was measured by liquid scintillation counting (Packard 1600 CA). Formation of [ 3 H]-ceramide was calculated as percentage of radioactivity found in [ 3 H]-sphingomyelin. Cell permeabilization For the introduction of PA, astrocytes were transiently permeabilized with streptolysin-O [ 37 ]. Briefly, astrocytes were washed and exposed to an oxygen-insensitive mutant of streptolysin-O (C530A) in calcium-free HBSS buffer (prepared by dissolving 8 g NaCl, 0.4 g KCl, 60 mg KH 2 PO 4 , 60 mg Na 2 HPO 4 × 2 H 2 O, 100 mg glucose in 1 L of sterile water). PA (sodium salt; 200 μM) was added as a suspension in buffer prepared by sonication. After 15 min, the cells were washed and incubated in serum-free, calcium-containing DMEM. Formation of [ 3 H]-ceramide was determined as described above. Ethanol, if present, was present throughout the incubation period. Using 144 ng/ml of streptolysin-O, this procedure yielded transient permeabilization of > 80% of astrocytes followed by repair of the pore in > 80% of permeabilized cells. The procedure was previously found to allow the entry of approx. 10 6 molecules of PA per cell [ 17 ]. Determination of phospholipase D activity Phospholipase D activity was determined using the transphosphatidylation assay [ 46 ]. For this purpose, astrocytes were kept in serum-free medium containing [ 3 H]-glycerol (1 μCi/ml) for 24 h in order to label phospholipids. More than 60% of the phospholipid label was associated with phosphatidylcholine (not illustrated). Subsequently, the cells were washed and re-exposed to medium containing ethanol (2 %) and PDB (1.0 μM) or FCS (10 %, v/v) as stimulators. PDB and C 2 -ceramide, when used, were dissolved in DMSO (end concentration of DMSO < 0.1 %). After 30 min of reaction time, the cells were washed in cold phosphate-buffered saline and extracted as described above for ceramide determinations. After addition of phosphatidylethanol (PEth) and PA standards, aliquots of the lipid phase were separated by two-dimensional TLC (TLC plate Merck 1.05721) using chloroform/methanol/25 % aqueous ammonia (13:7:1) for the first run and the upper phase of ethylacetate/isooctane/ acetic acid/water (13:2:3:10) for the second run. Individual phospholipids were stained by iodine, and the spots corresponding to PEth, PA and phosphatidylcholine (PC) were isolated and counted for radioactivity in a scintillation counter. To determine PLD activity, formation of [ 3 H]- PEth was calculated as percentage of [ 3 H]-PC. Statistics Data are shown as means ± SEM of N experiments whereby N refers to the number of different astroglial preparations from different animals. Results were obtained from two replicate dishes which were pooled to represent a single experiment. Statistical calculations were performed by GraphPad InStat 3.0 program package, using analysis of variance (ANOVA) of paired or unpaired data as indicated in text and figure legends. Abbreviations Cer, ceramide; FAS, fetal alcohol syndrome; FCS, fetal calf serum; PA, phosphatidic acid; PC, phosphatidylcholine; PDB, 4β-phorbol-12β,13α-dibutyrate; PEth, phosphatidylethanol; PKC, protein kinase C; PLD, phospholipase D; SL-O, streptolysin-O; SM, sphingomyelin. Author's contributions B.S. carried out the cell culture experiments and phospholipid measurements and participated in the experiments concerning apoptotic markers. S.J. contributed substantially to the apoptotic marker experiments. K.L. participated in the design of the study and the final draft of the manuscript. J.K. conceived and supervised the study, performed the statistical analyses and drafted the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549038.xml

549213

Inter-species horizontal transfer resulting in core-genome and niche-adaptive variation within Helicobacter pylori

Background Horizontal gene transfer is central to evolution in most bacterial species. The detection of exchanged regions is often based upon analysis of compositional characteristics and their comparison to the organism as a whole. In this study we describe a new methodology combining aspects of established signature analysis with textual analysis approaches. This approach has been used to analyze the two available genome sequences of H. pylori . Results This gene-by-gene analysis reveals a wide range of genes related to both virulence behaviour and the strain differences that have been relatively recently acquired from other sequence backgrounds. These frequently involve single genes or small numbers of genes that are not associated with transposases or bacteriophage genes, nor with inverted repeats typically used as markers for horizontal transfer. In addition, clear examples of horizontal exchange in genes associated with 'core' metabolic functions were identified, supported by differences between the sequenced strains, including: ftsK , xerD and polA . In some cases it was possible to determine which strain represented the 'parent' and 'altered' states for insertion-deletion events. Different signature component lengths showed different sensitivities for the detection of some horizontally transferred genes, which may reflect different amelioration rates of sequence components. Conclusion New implementations of signature analysis that can be applied on a gene-by-gene basis for the identification of horizontally acquired sequences are described. These findings highlight the central role of the availability of homologous substrates in evolution mediated by horizontal exchange, and suggest that some components of the supposedly stable 'core genome' may actually be favoured targets for integration of foreign sequences because of their degree of conservation.

Background Helicobacter pylori is a bacterial pathogen associated with gastritis, peptic ulcers, gastric adenocarcinoma, and rare lymphomas [ 1 ]. It has a highly panmictic population structure in which homologous recombination makes the predominant contribution to sequence differences within a highly diverse population structure [ 2 ]. The acquisition of genes from other strains and species is by far the most rapid evolutionary process. This occurs frequently without loss of existing functions, is central to the evolution of niche-adaptive and pathogenic characteristics of bacteria, and greatly influences inter-strain differences in gene complement [ 3 - 5 ]. In this context, it is notable that none of the traits typically used to differentiate E. coli from Salmonella can be attributed to point mutation genes but are broadly attributable to horizontal exchange [ 6 ]. H. pylori is relatively unusual in that it is a naturally transformable Gram-negative species that does not appear to have a species-specific DNA uptake sequence and appears to rely upon its niche separation as a transformation barrier [ 7 ]. Disease associated H. pylori strains have been divided into two types, type I being those that carry the cag pathogenicity island [ 8 ] ( cag PAI), which has a foreign species origin, and are associated with more severe disease. Dinucleotide composition is highly stable within a genome and can distinguish between sequences from different species. Based upon its constancy the species composition is referred to as a 'genome signature' [ 9 , 10 ]. This characteristic has been applied to assessments of DNA metabolic processes such as methylation and base conversion, DNA structure, and evolutionary relationships. It has also become established as a method for the identification of sequences that have been acquired by inter-species horizontal transfer. For example, lateral transfer has recently been shown using these methods for a tryptophan pathway operon [ 11 ], the gain of additional metabolic functions in Pseudomonas putida [ 12 ], a determination that many gain of function genes have been acquired by E. coli rather than lost from S. typhi [ 13 ], and more recently developed Bayesian methods based upon similar premises have been used to assess global signatures and determine the origins of some lateral transfer events [ 14 , 15 ]. However there are problems associated with this and other methods that use progressive 'walking windows', and the larger the window the greater the problems. These result from the inclusion of intergenic sequence, the inability to distinguish divergences due to a single highly divergent gene from that from a cluster of less divergent ones, and an inability to identify the limits of the abnormal regions. In practice additional features are necessary to determine the ends of such regions, such as the location of repeats typical of pathogenicity islands in H. pylori [ 16 ], or comparisons with other sequences as in N. meningitidis strain MC58 [ 17 ]. In addition, divergence scores are influenced by the size of the sampling window used such that sampling effects limit analysis of sequences shorter than about 800 bp (data not presented), and the need to use fixed window sizes prevents gene by gene studies. We describe the use of a linear implementation of signature analysis that can efficiently address a range of walking window sizes using dinucleotide signatures (DNS) and longer signatures. In addition, use of a new approach based upon classical text analysis that allows analysis of genomes gene-by-gene is described. Analysis of H. pylori sequences, combined with comparisons of the identified genes between genomes, reveals complex changes that influence both niche-adaptive and core functions illustrating a previously unpredicted range of functions which are continuously undergoing variation and selection. Results and discussion Genes were ranked on the basis of their divergence from the mean genome composition. The degree of divergence that is indicative of acquisition from other species is not an absolute. The frequency with which genes are acquired, the untypicality of the donated material, and the rate at which they are ameliorated to the host sequence composition influence it. Strains J99 and 26695 had 53 (Table 1 ) and 60 (Table 4 ) genes respectively with DNS that were >2 SD from the mean. Those with annotated functions included genes from the cag pathogenicity island (6 and 5), vac and related toxins (3 and 4), and restriction-modification genes (2 and 4). On the basis of the similarities determined in the H . pylori strain J99 sequence annotation, 7 of the most divergent genes as determined by DNS are not present in strain 26695. Likewise, 2 of the 50 most divergent genes in strain 26695 are not present in strain J99. This is consistent with the identification of genes acquired from other species that have not extended to both sequenced strains. It also suggests that a significant proportion of the 6 to 7% of genes unique to one or other strain [ 18 ] are inherent to the Helicobacter gene pool, but are variably present in different strains rather than reflecting recent foreign origins. Comparisons of a selection of identified orthologous genes in the two strains are shown in Figure 1 . Table 1 The 53 most divergent (>2 SD) genes in H. pylori strain J99 by DNS showing their ranking in strain 26695 and in TNS and HNS analysis DNS order JHP # annotation 26695 # 26695 DNS order TNS order HNS order 1 JHP0952 hypothetical protein HP0427 14 3 1355 2 JHP0476 cag pathogenicity island protein (cag7) HP0527 1 2 2 3 JHP0556 vacuolating cytotoxin (vacA) paralog HP0609/10 4/13 5 4 4 JHP0274 vacuolating cytotoxin (vacA) paralog HP0289 2 6 5 5 JHP0305 hypothetical protein HP0322 3 8 10 6 JHP0942 hypothetical protein HP0996 5 13 27 7 JHP0856 vacuolating cytotoxin (vacA) paralog HP0922 6 9 6 8 JHP0050 hypothetical protein HP0058 88 7 84 9 JHP1300 hypothetical protein HP1408 15 1 1 10 JHP1044 hypothetical protein HP1116 8 14 8 11 JHP0928 hypothetical protein NAH - 12 9 12 JHP0074 hypothetical protein HP0080 9 32 125 13 JHP0440 hypothetical protein HP0488 7 16 17 14 JHP1042 hypothetical protein HP1115 20 25 694 15 JHP1321 histidine and glutamine-rich metal-binding protein HP1432 46 4 49 16 JHP0934 hypothetical protein NAH - 15 95 17 JHP0495 cag island protein (cagA) HP0547 31 20 12 18 JHP0931 topoisomerase I (topA 3) NAH - 18 20 19 JHP0693 hypothetical protein HP0756 24 59 1490 20 JHP0632 N-methylhydantoinase HP0696 19 44 36 21 JHP0471 cag pathogenicity island protein (cag3) HP0522 11 35 62 22 JHP0438 outer membrane protein HP0486 26 67 145 23 JHP0026 hypothetical protein HP0030 45 36 64 24 JHP1084 outer membrane protein (omp26) HP1157 34 17 24 25 JHP0481 cag island protein (cagT) HP0532 23 70 558 26 JHP0052 hypothetical protein HP0059 43 24 120 27 JHP0336 hypothetical protein HP1089 12 51 54 28 JHP1426 iron(III) dicitrate transport protein (fecA) HP1400 32 78 111 29 JHP0174 hypothetical protein HP0187 / 8 / 6 47&1127&596 88 90 30 JHP1297 type III restriction enzyme (res) NAH - 63 28 31 JHP0953 hypothetical protein NAH - 26 1463 32 JHP0067 urease beta subunit (urea amidohydrolase) (ureB) HP0072 21 37 70 33 JHP0941 integrase/recombinase (xerD) HP0995 25 100 541 34 JHP0548 flagellin A (flaA) HP0601 33 40 154 35 JHP0299 hypothetical protein HP061/2 230&765 11 275 36 JHP1033 hypothetical protein HP1106 59 262 342 37 JHP1409 type II restriction enzyme (methyltransferase) NAH - 55 15 38 JHP0626 iron(III) dicitrate transport protein (fecA) HP0686 62 89 47 39 JHP0940 hypothetical protein NAH - 53 393 40 JHP1253 hypothetical protein HP1333 40 75 384 41 JHP0132 cytochrome oxidase (cbb3 type) (fixN) HP0144 27 206 209 42 JHP0842 hypothetical protein HP0906 42 29 21 43 JHP0925 hypothetical protein NAH - 130 990 44 JHP0613 hypothetical protein HP0669 69 42 33 45 JHP0565 DNA mismatch repair protein (mutS) HP0621 22 227 82 46 JHP1363 DNA polymerase I (polA) HP1470 30 81 46 47 JHP0489 cag island protein (cagH) HP0541 71 137 398 48 JHP1260 siderophore-mediated iron transport protein (tonB) HP1341 85 1260 402 49 JHP0492 DNA transfer protein (cagE) HP0544 104 95 50 50 JHP1121 DNA-directed RNA polymerase, beta subunit (rpoB) HP1198 84 23 16 51 JHP1434 DNA repair protein (recN) HP1393 35 177 160 52 JHP0491 cag island protein (cagF) HP0543 82 170 828 53 JHP0191 hypothetical protein HP0205 57 33 7 Genes with > 2 SD divergence indicated in bold NAH indicates No Annotated Homologue in the other sequence Table 4 Top 60 most divergent (>2 SD) genes by DNS in H. pylori strain 26695 plus those additional genes in the top 50 genes from TNS and HNS DNS order annotation HP# J99 # J99 DNS order TNS order HNS order 1 cag pathogenicity island protein (cag7) HP0527 JHP0476 2 1 1 2 vacuolating cytotoxin (vacA) paralog HP0289 JHP0274 4 2 4 3 poly E-rich hypothetical protein HP0322 JHP0305 5 8 5 4 hypothetical protein HP0609 JHP0556* 3 6 9 5 hypothetical protein HP0996 JHP0942 6 14 46 6 vacuolating cytotoxin (vacA) paralog HP0922 JHP0856 7 5 3 7 hypothetical protein HP0488 JHP0440 13 10 12 8 hypothetical protein HP1116 JHP1044 10 11 13 9 hypothetical protein HP0080 JHP0074 12 18 122 10 hypothetical protein HP0489 JHP0441 115 36 582 11 cag pathogenicity island protein (cag3) HP0522 JHP0471 21 48 100 12 hypothetical protein HP1089 JHP0336 27 67 59 13 vacuolating cytotoxin (vacA) paralog HP0610 JHP0556* 3 12 17 14 hypothetical protein HP0427 JHP0952 1 3 737 15 hypothetical protein HP1408 JHP1300 9 4 738 16 type III restriction enzyme R protein (res) HP0592 NAH - 30 35 17 hypothetical protein HP0119 NAH - 7 2 18 vacuolating cytotoxin (vacA) HP0887 JHP0819 59 25 34 19 N-methylhydantoinase HP0696 JHP0632 20 35 43 20 hypothetical protein HP1115 JHP1042 14 33 866 21 urease beta subunit (urea amidohydrolase) (ureB) HP0072 JHP0067 32 38 87 22 DNA mismatch repair protein (MutS) HP0621 JHP0565 45 137 64 23 cag island protein (cagT) HP0532 JHP0481 25 87 693 24 hypothetical protein HP0756 JHP0693 19 71 1548 25 integrase/recombinase (xerD) HP0995 JHP0941 33 39 448 26 outer membrane protein HP0486 JHP0438 22 147 142 27 cytochrome oxidase (cbb3 type) (fixN) HP0144 JHP0132 41 102 168 28 type IIS restriction enzyme R and M protein (ECO57IR) HP1517 NAH - 42 14 29 DNA transfer protein (cagE) HP0441 JHP0492 49 51 22 30 DNA polymerase I (polA) HP1470 JHP1363 46 77 54 31 cag island protein (cagA) HP0547 JHP0495 17 15 7 32 iron(III) dicitrate transport protein (fecA) HP1400 JHP1426 28 99 129 33 flagellin A (flaA) HP0601 JHP0548 34 40 180 34 outer membrane protein (omp26) HP1157 JHP1084 24 17 25 35 DNA repair protein (recN) HP1393 JHP1434 51 154 207 36 type I restriction enzyme R protein (hsdR) HP0464 NAH - 90 26 37 cell division protein (ftsK) HP1090 JHP0335 67 181 90 38 hypothetical protein HP1003 NAH - 61 170 39 histidine-rich, metal binding polypeptide (hpn) HP1427 NAH - 26 1449 40 hypothetical protein HP1333 JHP1253 40 53 296 41 hypothetical protein HP0788 JHP0725 68 72 256 42 hypothetical protein HP0906 JHP0842 42 22 16 43 hypothetical protein HP0059 JHP0052 26 21 320 44 GMP reductase (guaC) HP0854 JHP0790 107 169 451 45 hypothetical protein HP0030 JHP0026 23 24 39 46 histidine and glutamine-rich metal-binding protein HP1432 JHP1321 15 9 1432 47 hypothetical protein HP0186 JHP0174 29 130 276 48 fucosyltransferase HP0651 JHP0596 105 43 75 49 translation elongation factor EF-Tu (tufB) HP1205 JHP1128 81 64 166 50 virulence associated protein homolog (vacB) HP1248 JHP1169 79 164 160 51 hypothetical protein HP0449 NAH - 81 449 52 type III restriction enzyme R protein HP1371 JHP1285 55 119 23 53 virB4 homolog (virB4) HP0459 NAH - 49 28 54 2',3'-cyclic-nucleotide 2'-phosphodiesterase (cpdB) HP0104 JHP0096 56 73 68 55 hypothetical protein HP1479 JHP1372 135 153 127 56 RNA polymerase sigma-70 factor (rpoD) HP0088 JHP0081 62 55 31 57 hypothetical protein HP0205 JHP0191 53 78 8 58 hypothetical protein HP1143 JHP1071 78 29 41 59 hypothetical protein HP1106 JHP1033 36 272 277 60 cag pathogenicity island protein (cag13) HP0534 JHP0482 71 225 1021 63 DNA topoisomerase I (topA) HP0440 NAH - 149 24 68 outer membrane protein (omp3) HP0079 JHP0073 796 45 99 69 hypothetical protein HP0669 JHP0613 44 60 42 74 cag pathogenicity island protein (cag8) HP0528 JHP0477 72 50 27 75 hypothetical protein HP0453 NAH - 58 10 84 DNA-directed RNA polymerase, beta subunit (rpoB) JHP1121 50 23 19 91 hypothetical protein HP1142 JHP1070 60 19 6 97 multidrug resistance protein (spaB) HP0600 JHP0547 75 41 30 103 type I restriction enzyme R protein (hsdR) HP1402 JHP1424 195 86 21 109 adenine/cytosine DNA methyltransferase HP0054 NAH - 120 20 119 preprotein translocase subunit (secA) HP0786 JHP0723 159 176 49 121 hypothetical protein HP0058 JHP0051 394 16 53 122 hypothetical protein HP0513 JHP0462 104 28 15 125 type I restriction enzyme M protein (hsdM) HP1403 JHP1423 299 340 44 132 hypothetical protein HP0731 JHP0668 110 80 32 139 hypothetical protein HP0508 JHP0458 84 32 77 142 hypothetical protein HP1187 JHP1113 274 31 38 167 hypothetical protein HP1520 NAH - 20 33 179 hypothetical protein HP0118 JHP0110 64 27 36 195 type III restriction enzyme R protein (res) HP1521 JHP1410 161 210 18 209 outer membrane protein (omp17) HP0725 JHP0662 257 47 101 224 hypothetical protein HP0733 JHP0670 769 222 48 230 hypothetical protein HP0611 JHP0299 35 37 1129 249 hypothetical protein HP0345 NAH - 46 1338 283 hypothetical protein HP0120 NAH - 44 50 291 translation initiation factor IF-2 (infB) HP1048 JHP0377 330 332 45 297 DNA polymerase III alpha-subunit (dnaE) HP1460 JHP1353 509 219 47 342 type I restriction enzyme R protein (hsdR) HP0846 JHP0784 244 101 37 363 adenine specific DNA methyltransferase (mod) HP1522 JHP1411 857 207 11 410 secreted protein involved in flagellar motility HP1192 JHP1117 614 13 1256 593 hypothetical protein HP1516 NAH - 34 1090 631 hypothetical protein HP0586 JHP0534 577 163 29 1080 type II restriction enzyme (methyltransferase) HP0478 JHP0430 953 220 40 * probably frame shifted components of the same vacA related gene Genes with > 2 SD divergence in each analysis are indicated in bold NAH indicates No Annotated Homologue in the other sequence Figure 1 Comparisons using LAlign between a representative selection of orthologous genes with divergent DNA present in both H. pylori strains J99 and 26695 (presented in descending order of divergence as determined in strain J99). It cannot be assumed that all genes identified in this manner have been recently acquired. It is necessary to assess the nature of the sequence to determine if its divergence might be accounted for on the basis of features of the encoded protein. For example, JHP0476/HP0527, JHP1300/HP1408 and JHP0074/HP0080 include repetitive sequences likely to account for their DNS divergence. This type of analysis cannot be used to determine the possible foreign origin of such genes. Notably, the most divergent cag PAI gene (the 1 st and 2 nd most divergent gene in the whole genomes of strain 26695 and J99 respectively, JHP0476/HP0527) has a highly complex repetitive structure and the size of the large divergent peak associated with this island using previous methods is largely due to the presence of this gene. While a significant proportion of the genes identified in this analysis are associated with regions including several such genes and which share characteristics of islands of horizontal transfer or pathogenicity islands, this is far from universally true. There are many instances of single genes or small numbers of genes that are present that are not associated with any features that might otherwise have been used as indicators of horizontal acquisition such as transposases and flanking repeats. Our initial goal was to identify recently acquired and exchanged genes as candidates likely to be important in niche-adaptation, host interactions, and alterations in bacterial fitness. It has been argued that essential genes are unlikely to be transferred successfully since recipient taxa would already bear functional orthologues, which would have experienced long-term co-evolution with the rest of the cellular machinery. In contrast, it is proposed that those under weak or transient selection – like those associated with nonessential catabolic processes, new operons, and those providing new niche-adaptive changes are likely to be successfully transferred and retained [ 19 ]. This leads to a model in which a stable 'core genome' comprised of essential metabolic, regulatory, and cell division genes provides a stable context for the more labile non-essential and niche adaptive genes. On this basis such genes are used for phylogenetic studies and are thought to provide a relatively constant background in which species evolution occurs. Many of the genes identified for which functions are known affect virulence or niche adaptive genes, including: the vacuolating cytotoxin and related toxins (2 and 3), urease and flagellar components, and genes involved in iron acquisition. However, we also find clear evidence, confirmed by differences between the two genome sequences, that recent, and therefore relatively frequent, horizontal transfer is not limited to genes associated with niche adaptation and virulence. Amongst the core function genes identified were mut S, fts K, xer D, and pol A. The comparisons of the latter three between the sequence strains are shown in Figure 1f,g & 1j . These comparisons support the results suggesting that these genes have been the substrates for horizontal exchange between species. Tetranucleotide composition has been used for the consideration of the presence of palindromic sequences that might be substrates for restriction systems and Chi sites and the presence of unstable repeats mediating phase variation [ 10 ], but the use of longer component signatures has not been used to identify horizontally acquired regions in bacterial genomes. Following analysis of eukaryotic sequences it was concluded that DNS captures most of the departure from randomness in DNA sequences and that longer component lengths correlate highly with the DNS results [ 20 ]. Also, analysis of dinucleotides separated by no, one, or two other nucleotides showed that separated pairs are more nearly random than adjacent pairs and were concluded to be relatively uninformative [ 9 ]. However, in preliminary analyses, while results using the typically long walking windows gave concordant results as previously reported, we found that the use of smaller walking windows generated progressively more different patterns of divergence with other length components. Using tetranucleotide (TNS) and hexanucleotide (HNS) signature analysis we find that, while in some instances there is significant overlap between the genes identified using the different component lengths, there are substantial differences that indicate additional horizontally transferred genes not identified by DNS alone (Tables 2 to 6 ). Table 2 Top 50 most divergent genes by TNS in H. pylori strain J99 plus those additional genes > 2 SD greater than the mean by DNS and the 50 most divergent by HNS TNS order Annotation JHP # 26695 # DNS order HNS order 1 hypothetical protein JHP1300 HP1408 9 1 2 cag pathogenicity island protein (cag7) JHP0476 HP0527 2 2 3 hypothetical protein JHP0952 HP0427 1 1355 4 histidine and glutamine-rich metal-binding protein JHP1321 HP1432 15 49 5 vacuolating cytotoxin (vacA) paralog JHP0556 HP0609/10 3 4 6 vacuolating cytotoxin (vacA) paralog JHP0274 HP0289 4 5 7 hypothetical protein JHP0050 HP0058 8 84 8 hypothetical protein JHP0305 HP0322 5 10 9 vacuolating cytotoxin (vacA) paralog JHP0856 HP0922 7 6 10 type I restriction enzyme (hsdS) JHP1422 NAH 319 3 11 hypothetical protein JHP0299 HP061/2 35 275 12 hypothetical protein JHP0928 NAH 11 9 13 hypothetical protein JHP0942 HP0996 6 27 14 hypothetical protein JHP1044 HP1116 10 8 15 hypothetical protein JHP0934 NAH 16 95 16 hypothetical protein JHP0440 HP0488 13 17 17 outer membrane protein (omp26) JHP1084 HP1157 24 24 18 topoisomerase I (topA 3) JHP0931 NAH 18 20 19 hypothetical protein JHP0318 NAH 286 293 20 cag island protein (cagA) JHP0495 HP0547 17 12 21 hypothetical protein JHP0110 HP0118 64 19 22 hypothetical protein JHP1208 HP1288 91 830 23 DNA-directed RNA polymerase, beta subunit (rpoB) JHP1121 HP1198 50 16 24 hypothetical protein JHP0052 HP0059 26 120 25 hypothetical protein JHP1042 HP1115 14 694 26 hypothetical protein JHP0953 NAH 31 1463 27 hypothetical protein JHP1070 HP1142 60 14 28 hypothetical protein JHP1113 HP1187 274 39 29 hypothetical protein JHP0842 HP0906 42 21 30 type II restriction enzyme JHP0630 NAH 173 588 31 histidine-rich, metal binding polypeptide (hpn) JHP1320 HP1427 70 1404 32 hypothetical protein JHP0074 HP0080 12 125 33 hypothetical protein JHP0191 HP0205 53 7 34 hypothetical protein JHP0376 HP1049 235 1128 35 cag pathogenicity island protein (cag3) JHP0471 HP0522 21 62 36 hypothetical protein JHP0026 HP0030 23 64 37 urease beta subunit (urea amidohydrolase) (ureB) JHP0067 HP0072 32 70 38 hypothetical protein JHP0939 HP0991 116 156 39 multidrug resistance protein (spaB) JHP0547 HP0600 75 18 40 flagellin A (flaA) JHP0548 HP0601 34 154 41 hypothetical protein JHP1071 HP1143 78 61 42 hypothetical protein JHP0613 HP0669 44 33 43 hypothetical protein JHP0623 HP0682 231 1186 44 N-methylhydantoinase JHP0632 HP0696 20 36 45 hypothetical protein JHP1049 NAH 278 470 46 vacuolating cytotoxin (vacA) JHP0819 HP0887 59 38 47 putative restriction enzyme JHP0164 NAH 88 43 48 type I restriction enzyme R protein (hsdR) JHP0784 HP0846 244 35 49 hook assembly protein, flagella (flgD) JHP0843 HP0907 103 175 50 hypothetical protein JHP0458 HP0508 84 44 51 hypothetical protein JHP0336 HP1089 27 54 53 hypothetical protein JHP0940 NAH 39 393 54 hypothetical protein JHP0462 HP0513 104 11 55 type II restriction enzyme (methyltransferase) JHP1409 NAH 37 15 58 hypothetical protein JHP1285 HP1371 55 25 59 hypothetical protein JHP0693 HP0756 19 1490 62 cag pathogenicity island protein (cag8) JHP0477 HP0528 72 31 63 type III restriction enzyme (res) JHP1297 NAH 30 28 64 hypothetical protein JHP0668 HP0731 110 32 67 outer membrane protein JHP0438 HP0486 22 145 70 cag island protein (cagT) JHP0481 HP0532 25 558 71 RNA polymerase sigma-70 factor (rpoD) JHP0081 HP0088 62 37 75 hypothetical protein JHP1253 HP1333 40 384 78 iron(III) dicitrate transport protein (fecA) JHP1426 HP1400 28 111 81 DNA polymerase I (polA) JHP1363 HP1470 46 46 85 type I restriction enzyme (hsdS) JHP0414 NAH 275 30 88 hypothetical protein JHP0174 HP0187/8/6 29 90 89 iron(III) dicitrate transport protein (fecA) JHP0626 HP0686 38 47 95 DNA transfer protein (cagE) JHP0492 HP0544 49 50 100 integrase/recombinase (xerD) JHP0941 HP0995 33 541 104 type III restriciton enzyme (mod) JHP1411 HP1522 857 13 105 type I restriction enzyme R protein (hsdR) JHP0416 HP0464 63 29 122 adenine specific DNA methyltransferase (mod) JHP0244 HP0260 236 48 130 hypothetical protein JHP0925 NAH 43 990 137 cag island protein (cagH) JHP0489 HP0541 47 398 138 type I restriction enzyme (hsdR) JHP1424 HP1402 195 22 158 hypothetical protein JHP0540 NAH 674 26 170 cag island protein (cagF) JHP0491 HP0543 52 828 177 DNA repair protein (recN) JHP1434 HP1393 51 160 190 type III restriction enzyme (mod) JHP1296 NAH 121 34 196 role in outermembrane permeability (imp) JHP1138 HP1215/6 208 45 206 cytochrome oxidase (cbb3 type) (fixN) JHP0132 HP0144 41 209 227 DNA mismatch repair protein (mutS) JHP0565 HP0621 45 82 230 hypothetical protein JHP0534 HP0586 577 40 258 type III restriction enzyme (res) JHP1410 HP1521 161 23 262 hypothetical protein JHP1033 HP1106 36 342 281 translation initiation factor IF-2 (infB) JHP0377 HP1048 330 42 290 type II restriction enzyme (methyltrasferase) JHP1284 NAH 750 41 1260 siderophore-mediated iron transport protein (tonB) JHP1260 HP1341 48 402 Genes with > 2 SD divergence in each analysis are indicated in bold NAH indicates No Annotated Homologue in the other sequence Table 3 Top 50 most divergent genes by HNS in H. pylori strain J99 plus those additional genes >2 SD greater than the mean by DNS and top 50 by TNS HNS order J99 annotation JHP # 26695 # DNS order TNS order 1 hypothetical protein JHP1300 HP1408 9 1 2 cag pathogenicity island protein (cag7) JHP0476 HP0527 2 2 3 type I restriction enzyme (hsdS) JHP1422 NAH 319 10 4 vacuolating cytotoxin (vacA) paralog JHP0556 HP0609/10 3 5 5 vacuolating cytotoxin (vacA) paralog JHP0274 HP0289 4 6 6 vacuolating cytotoxin (vacA) paralog JHP0856 HP0922 7 9 7 hypothetical protein JHP0191 HP0205 53 33 8 hypothetical protein JHP1044 HP1116 10 14 9 hypothetical protein JHP0928 NAH 11 12 10 hypothetical protein JHP0305 HP0322 5 8 11 hypothetical protein JHP0462 HP0513 104 54 12 cag island protein (cagA) JHP0495 HP0547 17 20 13 type III restriciton enzyme (mod) JHP1411 HP1522 857 104 14 hypothetical protein JHP1070 HP1142 60 27 15 type II restriction enzyme (methyltransferase) JHP1409 NAH 37 55 16 DNA-directed RNA polymerase, beta subunit (rpoB) JHP1121 HP1198 50 23 17 hypothetical protein JHP0440 HP0488 13 16 18 multidrug resistance protein (spaB) JHP0547 HP0600 75 39 19 hypothetical protein JHP0110 HP0118 64 21 20 topoisomerase I (topA 3) JHP0931 NAH – check 18 18 21 hypothetical protein JHP0842 HP0906 42 29 22 type I restriction enzyme (hsdR) JHP1424 HP1402 195 138 23 type III restriction enzyme (res) JHP1410 HP1521 161 258 24 outer membrane protein (omp26) JHP1084 HP1157 24 17 25 hypothetical protein JHP1285 HP1371 55 58 26 hypothetical protein JHP0540 NAH 674 158 27 hypothetical protein JHP0942 HP0996 6 13 28 type III restriction enzyme (res) JHP1297 NAH 30 63 29 type I restriction enzyme R protein (hsdR) JHP0416 HP0464 63 105 30 type I restriction enzyme (hsdS) JHP0414 NAH 275 85 31 cag pathogenicity island protein (cag8) JHP0477 HP0528 72 62 32 hypothetical protein JHP0668 HP0731 110 64 33 hypothetical protein JHP0613 HP0669 44 42 34 type III restriction enzyme (mod) JHP1296 NAH 121 190 35 type I restriction enzyme R protein (hsdR) JHP0784 HP0846 244 48 36 N-methylhydantoinase JHP0632 HP0696 20 44 37 RNA polymerase sigma-70 factor (rpoD) JHP0081 HP0088 62 71 38 vacuolating cytotoxin (vacA) JHP0819 HP0887 59 46 39 hypothetical protein JHP1113 HP1187 274 28 40 hypothetical protein JHP0534 HP0586 577 230 41 type II restriction enzyme (methyltrasferase) JHP1284 NAH 750 290 42 translation initiation factor IF-2 (infB) JHP0377 HP1048 330 281 43 restriction enzyme JHP0164 NAH 88 47 44 hypothetical protein JHP0458 HP0508 84 50 45 role in outermembrane permeability (imp) JHP1138 HP1215/6 208 196 46 DNA polymerase I (polA) JHP1363 HP1470 46 81 47 iron(III) dicitrate transport protein (fecA) JHP0626 HP0686 38 89 48 adenine specific DNA methyltransferase (mod) JHP0244 HP0260 236 122 49 histidine and glutamine-rich metal-binding protein JHP1321 HP1432 15 4 50 DNA transfer protein (cagE) JHP0492 HP0544 49 95 54 hypothetical protein JHP0336 HP1089 27 51 62 cag pathogenicity island protein (cag3) JHP0471 HP0522 21 35 64 hypothetical protein JHP0026 HP0030 23 36 70 urease beta subunit (urea amidohydrolase) (ureB) JHP0067 HP0072 32 37 82 DNA mismatch repair protein (mutS) JHP0565 HP0621 45 227 84 hypothetical protein JHP0050 HP0058 8 7 90 hypothetical protein JHP0174 HP0187/8/6 29 88 95 hypothetical protein JHP0934 NAH 16 15 111 iron(III) dicitrate transport protein (fecA) JHP1426 HP1400 28 78 120 hypothetical protein JHP0052 HP0059 26 24 125 hypothetical protein JHP0074 HP0080 12 32 145 Outer membrane protein JHP0438 HP0486 22 67 154 flagellin A (flaA) JHP0548 HP0601 34 40 160 DNA repair protein (recN) JHP1434 HP1393 51 177 209 cytochrome oxidase (cbb3 type) (fixN) JHP0132 HP0144 41 206 275 hypothetical protein JHP0299 HP061/2 35 11 342 hypothetical protein JHP1033 HP1106 36 262 384 hypothetical protein JHP1253 HP1333 40 75 393 hypothetical protein JHP0940 NAH 39 53 398 cag island protein (cagH) JHP0489 HP0541 47 137 402 siderophore-mediated iron transport protein (tonB) JHP1260 HP1341 48 1260 541 integrase/recombinase (xerD) JHP0941 HP0995 33 100 558 cag island protein (cagT) JHP0481 HP0532 25 70 694 hypothetical protein JHP1042 HP1115 14 25 828 cag island protein (cagF) JHP0491 HP0543 52 170 990 hypothetical protein JHP0925 NAH 43 130 1355 hypothetical protein JHP0952 HP0427 1 3 1463 hypothetical protein JHP0953 NAH 31 26 1490 hypothetical protein JHP0693 HP0756 19 59 Genes with > 2 SD divergence in each analysis are indicated in bold NAH indicates No Annotated Homologue in the other sequence Table 5 Top 50 most divergent genes by TNS in H. pylori strain 26695 plus those additional genes > 2 SD greater than the mean by DNS and the 50 most divergent by HNS TNS order annotation HP# J99 # DNS order HNS order 1 cag pathogenicity island protein (cag7) HP0527 JHP0476 1 1 2 vacuolating cytotoxin (vacA) paralog HP0289 JHP0274 2 4 3 hypothetical protein HP0427 JHP0952 14 737 4 hypothetical protein HP1408 JHP1300 15 738 5 vacuolating cytotoxin (vacA) paralog HP0922 JHP0856 6 3 6 hypothetical protein HP0609 JHP0556* 4 9 7 hypothetical protein HP0119 NAH 17 2 8 poly E-rich hypothetical protein HP0322 JHP0305 3 5 9 histidine and glutamine-rich metal-binding protein HP1432 JHP1321 46 1432 10 hypothetical protein HP0488 JHP0440 7 12 11 hypothetical protein HP1116 JHP1044 8 13 12 vacuolating cytotoxin (vacA) paralog HP0610 JHP0556* 13 17 13 secreted protein involved in flagellar motility HP1192 JHP1117 410 1256 14 hypothetical protein HP0996 JHP0942 5 46 15 cag island protein (cagA) HP0547 JHP0495 31 7 16 hypothetical protein HP0058 JHP0051 121 53 17 outer membrane protein (omp26) HP1157 JHP1084 34 25 18 hypothetical protein HP0080 JHP0074 9 122 19 hypothetical protein HP1142 JHP1070 91 6 20 hypothetical protein HP1520 NAH 167 33 21 hypothetical protein HP0059 JHP0052 43 320 22 hypothetical protein HP0906 JHP0842 42 16 23 DNA-directed RNA polymerase, beta subunit (rpoB) HP1198 JHP1121 84 19 24 hypothetical protein HP0030 JHP0026 45 39 25 vacuolating cytotoxin (vacA) HP0887 JHP0819 18 34 26 histidine-rich, metal binding polypeptide (hpn) HP1427 NAH 39 1449 27 hypothetical protein HP0118 JHP0110 179 36 28 hypothetical protein HP0513 JHP0462 122 15 29 hypothetical protein HP1143 JHP1071 58 41 30 type III restriction enzyme R protein (res) HP0592 NAH 16 35 31 hypothetical protein HP1187 JHP1113 142 38 32 hypothetical protein HP0508 JHP0458 139 77 33 hypothetical protein HP1115 JHP1042 20 866 34 hypothetical protein HP1516 NAH 593 1090 35 N-methylhydantoinase HP0696 JHP0632 19 43 36 hypothetical protein HP0489 JHP0441 10 582 37 hypothetical protein HP0611 JHP0299 230 1129 38 urease beta subunit (urea amidohydrolase) (ureB) HP0072 JHP0067 21 87 39 integrase/recombinase (xerD) HP0995 JHP0941 25 448 40 flagellin A (flaA) HP0601 JHP0548 33 180 41 multidrug resistance protein (spaB) HP0600 JHP0547 97 30 42 type IIS restriction enzyme R and M protein (ECO57IR) HP1517 NAH 28 14 43 fucosyltransferase HP0651 JHP0596 48 75 44 hypothetical protein HP0120 NAH 283 50 45 outer membrane protein (omp3) HP0079 JHP0073 68 99 46 hypothetical protein HP0345 NAH 249 1338 47 outer membrane protein (omp17) HP0725 JHP0662 209 101 48 cag pathogenicity island protein (cag3) HP0522 JHP0471 11 100 49 virB4 homolog (virB4) HP0459 NAH 53 28 50 cag pathogenicity island protein (cag8) HP0528 JHP0477 74 27 51 DNA transfer protein (cagE) HP0441 JHP0492 29 22 53 hypothetical protein HP1333 JHP1253 40 296 55 RNA polymerase sigma-70 factor (rpoD) HP0088 JHP0081 56 31 58 hypothetical protein HP0453 NAH 75 10 60 hypothetical protein HP0669 JHP0613 69 42 61 hypothetical protein HP1003 NAH 38 170 64 translation elongation factor EF-Tu (tufB) HP1205 JHP1128 49 166 67 hypothetical protein HP1089 JHP0336 12 59 71 hypothetical protein HP0756 JHP0693 24 1548 72 hypothetical protein HP0788 JHP0725 41 256 73 2',3'-cyclic-nucleotide 2'-phosphodiesterase (cpdB) HP0104 JHP0096 54 68 77 DNA polymerase I (polA) HP1470 JHP1363 30 54 78 hypothetical protein HP0205 JHP0191 57 8 80 hypothetical protein HP0731 JHP0668 132 32 81 hypothetical protein HP0449 NAH 51 449 86 type I restriction enzyme R protein (hsdR) HP1402 JHP1424 103 21 87 cag pathogenicity island protein (cag12) HP0532 JHP0481 23 693 90 type I restriction enzyme R protein (hsdR) HP0464 NAH 36 26 99 iron(III) dicitrate transport protein (fecA) HP1400 JHP1426 32 129 101 type I restriction enzyme R protein (hsdR) HP0846 JHP0784 342 37 102 cytochrome oxidase (cbb3 type) (fixN) HP0144 JHP0132 27 168 119 type III restriction enzyme R protein HP1371 JHP1285 52 23 120 adenine/cytosine DNA methyltransferase HP0054 NAH 109 20 130 hypothetical protein HP0186 JHP0174 47 276 137 DNA mismatch repair protein (MutS) HP0621 JHP0565 22 64 147 outer membrane protein HP0486 JHP0438 26 142 149 DNA topoisomerase I (topA) HP0440 NAH 63 24 153 hypothetical protein HP1479 JHP1372 55 127 154 DNA repair protein (recN) HP1393 JHP1434 35 207 163 hypothetical protein HP0586 JHP0534 631 29 164 virulence associated protein homolog (vacB) HP1248 JHP1169 50 160 169 GMP reductase (guaC) HP0854 JHP0790 44 451 176 preprotein translocase subunit (secA) HP0786 JHP0723 119 49 181 cell division protein (ftsK) HP1090 JHP0335 37 90 207 adenine specific DNA methyltransferase (mod) HP1522 JHP1411 363 11 210 type III restriction enzyme R protein (res) HP1521 JHP1410 195 18 219 DNA polymerase III alpha-subunit (dnaE) HP1460 JHP1353 297 47 220 type II restriction enzyme (methyltransferase) HP0478 JHP0430 1080 40 222 hypothetical protein HP0733 JHP0670 224 48 225 cag pathogenicity island protein (cag13) HP0534 JHP0482 60 1021 272 hypothetical protein HP1106 JHP1033 59 277 332 translation initiation factor IF-2 (infB) HP1048 JHP0377 291 45 340 type I restriction enzyme M protein (hsdM) HP1403 JHP1423 125 44 * probably frame shifted components of the same vacA related gene Genes with > 2 SD divergence in each analysis are indicated in bold NAH indicates No Annotated Homologue in the other sequence Table 6 Top 50 most divergent genes by HNS in H. pylori strain 26695 plus those additional genes > 2 SD greater than the mean by DNS and the 50 most divergent by HNS HNS order annotation HP# J99 # DNS order TNS order 1 cag pathogenicity island protein (cag7) HP0527 JHP0476 1 1 2 hypothetical protein HP0119 NAH 17 7 3 vacuolating cytotoxin (vacA) paralog HP0922 JHP0856 6 5 4 vacuolating cytotoxin (vacA) paralog HP0289 JHP0274 2 2 5 poly E-rich hypothetical protein HP0322 JHP0305 3 8 6 hypothetical protein HP1142 JHP1070 91 19 7 cag island protein (cagA) HP0547 JHP0495 31 15 8 hypothetical protein HP0205 JHP0191 57 78 9 hypothetical protein HP0609 JHP0556* 4 6 10 hypothetical protein HP0453 NAH 75 58 11 adenine specific DNA methyltransferase (mod) HP1522 JHP1411 363 207 12 hypothetical protein HP0488 JHP0440 7 10 13 hypothetical protein HP1116 JHP1044 8 11 14 type IIS restriction enzyme R and M protein (ECO57IR) HP1517 NAH 28 42 15 hypothetical protein HP0513 JHP0462 122 28 16 hypothetical protein HP0906 JHP0842 42 22 17 vacuolating cytotoxin (vacA) paralog HP0610 JHP0556* 13 12 18 type III restriction enzyme R protein (res) HP1521 JHP1410 195 210 19 DNA-directed RNA polymerase, beta subunit (rpoB) HP1198 JHP1121 84 23 20 adenine/cytosine DNA methyltransferase HP0054 NAH 109 120 21 type I restriction enzyme R protein (hsdR) HP1402 JHP1424 103 86 22 DNA transfer protein (cagE) HP0441 JHP0492 29 51 23 type III restriction enzyme R protein HP1371 JHP1285 52 119 24 DNA topoisomerase I (topA) HP0440 NAH 63 149 25 outer membrane protein (omp26) HP1157 JHP1084 34 27 26 type I restriction enzyme R protein (hsdR) HP0464 NAH 36 90 27 cag pathogenicity island protein (cag8) HP0528 JHP0477 74 50 28 virB4 homolog (virB4) HP0459 NAH 53 49 29 hypothetical protein HP0586 JHP0534 631 163 30 multidrug resistance protein (spaB) HP0600 JHP0547 97 41 31 RNA polymerase sigma-70 factor (rpoD) HP0088 JHP0081 56 55 32 hypothetical protein HP0731 JHP0668 132 80 33 hypothetical protein HP1520 NAH 167 20 34 vacuolating cytotoxin HP0887 JHP0819 18 25 35 type III restriction enzyme R protein (res) HP0592 NAH 16 30 36 hypothetical protein HP0118 JHP0110 179 27 37 type I restriction enzyme R protein (hsdR) HP0846 JHP0784 342 101 38 hypothetical protein HP1187 JHP1113 142 31 39 hypothetical protein HP0030 JHP0026 45 24 40 HP0478 JHP0430 1080 220 41 hypothetical protein HP1143 JHP1071 58 29 42 hypothetical protein HP0669 JHP0613 69 60 43 N-methylhydantoinase HP0696 JHP0632 19 35 44 type I restriction enzyme M protein (hsdM) HP1403 JHP1423 125 340 45 translation initiation factor IF-2 (infB) HP1048 JHP0377 291 332 46 hypothetical protein HP0996 JHP0942 5 14 47 DNA polymerase III alpha-subunit (dnaE) HP1460 JHP1353 297 219 48 hypothetical protein HP0733 JHP0670 224 222 49 preprotein translocase subunit (secA) HP0786 JHP0723 119 176 50 hypothetical protein HP0120 NAH 283 44 53 hypothetical protein HP0058 JHP0051 121 16 54 DNA polymerase I (polA) HP1470 JHP1363 30 77 59 hypothetical protein HP1089 JHP0336 12 67 64 DNA mismatch repair protein (MutS) HP0621 JHP0565 22 137 68 2',3'-cyclic-nucleotide 2'-phosphodiesterase (cpdB) HP0104 JHP0096 54 73 75 fucosyltransferase HP0651 JHP0596 48 43 77 hypothetical protein HP0508 JHP0458 139 32 87 urease beta subunit (urea amidohydrolase) (ureB) HP0072 JHP0067 21 38 90 cell division protein (ftsK) HP1090 JHP0335 37 181 99 outer membrane protein (omp3) HP0079 JHP0073 68 45 100 cag pathogenicity island protein (cag3) HP0522 JHP0471 11 48 101 outer membrane protein (omp17) HP0725 JHP0662 209 47 122 hypothetical protein HP0080 JHP0074 9 18 127 hypothetical protein HP1479 JHP1372 55 153 129 iron(III) dicitrate transport protein (fecA) HP1400 JHP1426 32 99 142 outer membrane protein HP0486 JHP0438 26 147 160 virulence associated protein homolog (vacB) HP1248 JHP1169 50 164 166 translation elongation factor EF-Tu (tufB) HP1205 JHP1128 49 64 168 cytochrome oxidase (cbb3 type) (fixN) HP0144 JHP0132 27 102 170 hypothetical protein HP1003 NAH 38 61 180 flagellin A (flaA) HP0601 JHP0548 33 40 207 DNA repair protein (recN) HP1393 JHP1434 35 154 256 hypothetical protein HP0788 JHP0725 41 72 276 hypothetical protein HP0186 JHP0174 47 130 277 hypothetical protein HP1106 JHP1033 59 272 296 hypothetical protein HP1333 JHP1253 40 53 320 hypothetical protein HP0059 JHP0052 43 21 448 integrase/recombinase (xerD) HP0995 JHP0941 25 39 449 hypothetical protein HP0449 NAH 51 81 451 GMP reductase (guaC) HP0854 JHP0790 44 169 582 hypothetical protein HP0489 JHP0441 10 36 693 cag island protein (cagT) HP0532 JHP0481 23 87 737 hypothetical protein HP0427 JHP0952 14 3 738 hypothetical protein HP1408 JHP1300 15 4 866 hypothetical protein HP1115 JHP1042 20 33 1021 cag pathogenicity island protein (cag13) HP0534 JHP0482 60 225 1090 hypothetical protein HP1516 NAH 593 34 1129 hypothetical protein HP0611 JHP0299 230 37 1256 secreted protein involved in flagellar motility HP1192 JHP1117 410 13 1338 hypothetical protein HP0345 NAH 249 46 1432 histidine and glutamine-rich metal-binding protein HP1432 JHP1321 46 9 1449 histidine-rich, metal binding polypeptide (hpn) HP1427 NAH 39 26 1548 hypothetical protein HP0756 JHP0693 24 71 * probably frame shifted components of the same vacA related gene Genes with > 2 SD divergence in each analysis are indicated in bold NAH indicates No Annotated Homologue in the other sequence The 50 most divergent J99 ORFs by HNS included 26 (52%) that were not in the 53 (>2 SD) most divergent by DNS, these included 11 restriction-modification system genes and 6 others that were not annotated within the strain 26695 genome sequence. The identification of genes of a type known to be horizontally exchanged, and different between the gene complements of the strains, is strong corroboration for the foreign origin of the additional genes identified by HNS. In several instances (Tables 2 to 6 ) the DNS did not detect these genes at all e.g. restriction enzymes that were the 3 rd , 13 th and 41 st most divergent genes by HNS, were 319 th , 857 th and 750 th most divergent by DNS, respectively. In some instances the TNS gave intermediate results and in others identified other genes as more divergent than the other methods. The TNS was most sensitive for the detection of rpoB (HP1198 / JHP1121) which is associated with a significantly different gene length in the two strains (Figure 1h ). One explanation for this observation is that while the DNS may initially be the most sensitive indicator of horizontal exchange it may become ameliorated to the new sequence characteristics more rapidly that the longer component features, which are probably detecting qualitatively different sequence characteristics. The differences in the analyses using different length components, and a comparison of the results from the two sequenced strains, suggest a complex evolutionary history for the cag pathogenicity island. These suggest that it probably has mosaic structure including sequences from more than one species background, in addition to sequence that is entirely typical of H. pylori . It is normally impossible to determine the chronology of events to distinguish insertions and deletions when comparing strains. In strain 26695 there are two open reading frames that are both good candidate coding sequences. There is only one gene in this location in strain J99 composed of the 5' gene from strain 26695 and the 3' end of the subsequent gene. This could have arisen from either a deletion or an insertion event. However, the normal DNS of the J99 gene (JHP0073, 799 th in divergence) and the 5' 26695 gene (HP0079, 751 st in divergence), and the high divergence of the 3' 26695 gene (HP0078, 68 th in divergence), indicate that the most likely event is an insertion into strain 26695 (Figure 1l ). Likewise HP0119 is likely to contain an insertion and JHP1113 probably reflects the original sequences (Figure 1k ). The inclusion of two DNA metabolism genes associated with recombination and repair is notable. Both mutS and recN were identified in both strains (22 nd and 35 th , and 45 th and 51 st most divergent genes by DNS in strains 26695 and J99 respectively). When the homologous genes were compared between the strains, extensive divergences were evident between more than one region of each protein. That these genes have divergent signatures in both strains suggests that neither has a wholly native composition. This observation is consistent with the models of rapid evolution which suggest that transient competitive advantages are enjoyed by organisms that are hypermutators under conditions of environmental stress and transitions, and that these states which can be produced by mutations in DNA repair genes [ 21 - 26 ]. However, such states have to be reversed so that an unsustainable mutational burden is not attained, and it has been proposed that this reversal is mediated by repair following horizontal transfer and homologous recombination, and that such strains are hyper-recombinogenic [ 27 - 29 ]. The untypicality of mutS and recN suggest that H. pylori is another species that can make use of this strategy for diversification under stressful conditions. The identification of RNA polymerase genes, with associated differences between the strains, is striking. The divergence of phylogenetic trees based upon different sequences has been highlighted, and particularly the differences between the trees associated with RNA polymerase genes and rRNA [ 30 , 31 ]. It has been argued that RNA polymerase is as essential to cell function as is rRNA and that there is no compelling reason to chose rRNA as the more reliable marker [ 32 ]. While the DNS analysis does not address the stability of rRNA (and specifically excludes the rRNA sequences because their differing coding requirements and evolutionary pressures generate a divergent signature for other reasons), it does indicate that RNA polymerase can be a substrate for horizontal transfer, and that trees based upon this gene, or other essential genes, need not necessarily be considered a challenge to rRNA based phylogenies. Conclusions The spectrum of recently horizontally acquired sequences identified emphasizes the two driving forces of horizontal exchange: the transfer of a phenotype which alters or enhances bacterial fitness resulting in increased competitive fitness or altered niche adaptation, and the presence of a substrate for homologous recombination. Because of the focus upon, and relative ease of identifying, large islands associated with readily identifiable features and phenotypes, the importance of the latter component has perhaps been underestimated. The genes that have been considered to code for 'core metabolic' 'house-keeping' functions are amongst those most likely to be changed by horizontal transfer events because of the presence of homologous substrates, and changes are likely to persist even when the change is phenotypically neutral. Equally, changes in the genes involved in core functions such as gene expression and DNA metabolism may have pleotropic effects and there may be significant differences in strain behaviour, that are not simply the consequence of differences in their respective gene complements. The selection of genes for phylogenetic analysis on the basis of their coding for conserved core functions is also problematic because these are also frequently the genes most likely to share the high homology that facilitates recombination and horizontal exchange. Methods A traditional nucleotide signature is generated by segmenting a sequence of DNA into k equal-sized subsequences (or 'windows'). The mathematical basis for the signature is an odds ratio – p i – calculated by dividing the frequency of a length- L oligonucleotide by its expected frequency. The odds ratios for each of the 4 L oligonucleotides in each window ( w ) are compared with the odds ratios for the overall sequence ( s ) [ 9 , 10 , 33 ]. The normalized difference δ is plotted and thus a nucleotide signature consists of a k -length sequence of δ values: δ ( w , s ) = (1/4 L )Σ(4 L , i : x )| p i ( w ) - p i ( s )|, where x is the set of all permutations of length L and i is one such permutation. There are interesting parallels between signature-style genome analysis and stylometric techniques previously used to determine the authorship of controversial literary texts. This is analogous with the biological problem and it is from this that our method is derived. Rather than using a fixed-window signature, signature scores are calculated for each coding open reading frame (ORF) and weighted with variance estimates so that the scores for shorter ORFs confer with their longer counterparts. Bissell's weighted cusum (cumulative sum) [ 34 ], , is modified so that n denotes the number of ORFs in the genome, X i the number of oligonucleotides in ORF i , and w i the number of nucleotides in ORF i . The results are scaled according to ORF size using the standard error σ = √( *# ORF ). In this way false positives are abrogated by normalizing for over-representation of lower order peptides. The method is implemented in Java and efficiency is maintained through an O ( N ) ( N = sequence length) refinement: probabilities for the complete sequence are calculated in O ( N ) steps for any length- L oligonucleotide, and maintain O ( N ) when 4 L > N through a hashing function; the second part of the program calculates σ for each ORF using a loop flattening technique, thereby avoiding the program having to recalculate overlapping sub-expressions. The program is available from and . Sequence alignments, as shown in Figure 1 , were performed and displayed using the programs: Lalign and viewed using Lalignview [ 35 ]. Abbreviations ORF, Open Reading Frame; DNS, Dinucleotide Signature; TNS, tetranucleotide signature; HNS, hexanucleotide signature. Authors' contributions NJS initiated the project, performed the genome sequence analyses, compared the two strains, interpreted the results, and prepared the biological aspects of the manuscript. PB was a DPhil student who worked on the coding aspects of the new methodology. JFP contributed to the bioinformatics discussions and planning stage of this project. SAJ directed and primarily developed the analysis strategy and the implementation of the new computational basis of the methodology, and prepared the computational aspects of the manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549213.xml

509408

Amplification of Trial-to-Trial Response Variability by Neurons in Visual Cortex

The visual cortex responds to repeated presentations of the same stimulus with high variability. Because the firing mechanism is remarkably noiseless, the source of this variability is thought to lie in the membrane potential fluctuations that result from summated synaptic input. Here this hypothesis is tested through measurements of membrane potential during visual stimulation. Surprisingly, trial-to-trial variability of membrane potential is found to be low. The ratio of variance to mean is much lower for membrane potential than for firing rate. The high variability of firing rate is explained by the threshold present in the function that converts inputs into firing rates. Given an input with small, constant noise, this function produces a firing rate with a large variance that grows with the mean. This model is validated on responses recorded both intracellularly and extracellularly. In neurons of visual cortex, thus, a simple deterministic mechanism amplifies the low variability of summated synaptic inputs into the large variability of firing rate. The computational advantages provided by this amplification are not known.

Introduction In the primary visual cortex (V1), different trials of presentation of an identical stimulus yield highly variable firing rates ( Heggelund and Albus 1978 ). This trial-to-trial variability is not inherited from subcortical inputs, as these respond in a much more consistent fashion ( Kara et al. 2000 ). Instead, variability has been related to spontaneous variations in cortical state ( Arieli et al. 1996 ; Buracas et al. 1998 ; Tsodyks et al. 1999 ; Kenet et al. 2003 ). These variations may reflect the perceptual effects associated with a stimulus, rather than the presence of the stimulus itself ( Ress and Heeger 2003 ). A key property of trial-to-trial variability is that it depends on the strength of the stimulus: Response variance across trials is approximately proportional to response mean ( Tolhurst et al. 1981 ). An example of this effect can be seen in the responses of a cell in cat V1 to drifting gratings ( Figure 1 A– 1 C). Different trials of an identical stimulus elicit firing rates that vary greatly ( Figure 1 A). As a result, the standard deviation of the firing rates is roughly comparable to their mean amplitude ( Figure 1 B and 1 C). The ratio of variance to mean is close to the value predicted for a Poisson process ( Figure 2 A, dashed line). For a Poisson process, the variance of the spike counts is equal to the mean. Once spike counts are converted to firing rate by binning in 10-ms windows (i.e., at 100 Hz), the ratio of variance to mean becomes 100. The Poisson-like behavior of firing rates is well known, although reports differ on the exact value of the ratio of variance to mean ( Tolhurst et al. 1981 ; Bradley et al. 1987 ; Vogels et al. 1989 ; Geisler and Albrecht 1997 ; Gur et al. 1997 ; Reich et al. 1997 ; Buracas et al. 1998 ; Kara et al. 2000 ). Figure 1 Variability in the Responses of a Simple Cell (A) Firing rate in response to a cycle of an optimal drifting grating. Three trials are shown. (B) Firing rate averaged over seven trials. Shaded area indicates 2 s.d. (C) Same, for three other stimuli: a grating drifting in the orthogonal direction (top), a grating drifting in the opposite direction (middle), and a blank stimulus (bottom). (D) Membrane potential trace measured for the first cycle. Dashed line is resting potential V rest . Dotted line is firing threshold V thresh (from [H]). (E–G) As in (A–C), for coarse potential. (H) Relation between firing rate and coarse potential. Curve is fit of rectification equation. (I–K) As in (A–C), for predictions of rectification model. Figure 2 Relation between Response Variance and Mean for Three Cells (A) Variance versus mean for firing rate of the simple cell in Figure 1 measured with 13 stimuli (the four in Figure 1 plus nine additional orientations). Line is linear regression. Diagonal line is prediction for a Poisson process. (B) Variance versus mean for coarse potential. Error bars are 2 s.d. Curve is linear fit to standard deviation versus mean. Dashed line is resting potential V rest . Dotted line is firing threshold V thresh . (C) Variance versus mean for firing rate predicted by the rectification model. Details as in (A). (D–F) As in (A–C) for a complex cell. (G–I) As in (A–C) for a third neuron, whose behavior is intermediate between those of simple cells and complex cells. Because the production of firing rates within a neuron introduces remarkably little noise ( Calvin and Stevens 1968 ; Mainen and Sejnowski 1995 ; Carandini et al. 1996 ), trial-to-trial variability is thought to arise from the membrane potential fluctuations that result from summated synaptic input ( Calvin and Stevens 1968 ; Stevens and Zador 1998 ). I have tested this hypothesis by considering membrane potential responses recorded intracellularly in vivo. Results From traces of membrane potential obtained at high temporal resolution ( Figure 1 D), I obtained an estimate of overall synaptic drive by removing the action potentials and low-pass filtering the resulting traces ( Carandini and Ferster 2000 ; Volgushev et al. 2000 ). The outcome of this procedure ( Figure 1 E) is a coarse potential (or “generator potential”; Lankheet et al. 1989 ) that approximates the synaptic current ( Anderson et al. 2000a ). This technique allows one to estimate synaptic currents while concurrently recording firing rates. Variability of Coarse Potential during Visual Stimulation We can now consider the mean and variance across trials for coarse potential. The mean, V mean , is the “signal” reflecting the stimulus-driven synaptic input to the neuron ( Figure 1 F and 1 G, traces). The variance, instead, is the “noise” reflecting the synaptic input's trial-to-trial variability ( Figure 1 F and 1 G, shaded areas). The variability of potential depended only slightly on stimulus strength. Variance was slightly higher when the stimuli depolarized the cell than when they hyperpolarized it ( Figure 1 F and 1 G). For the example simple cell in Figure 1 , standard deviation of potential was 2.8 ± 1.2 mV (s.d.) for V mean between –70 and –65 mV, and 4.0 ± 1.7 mV for V mean between –55 and –50 mV. The relation between standard deviation of potential and V mean can be described by a regression line ( r = 0.27 ± 0.04, s.d., bootstrap) whose slope is 0.08 ± 0.01 and whose intercept at V rest = −60.4 mV is 3.3 ± 0.1 mV. Similar values were obtained in the rest of the population (e.g., Figure 2 E and 2 H): correlation coefficient was r = 0.40 ± 0.19 (s.d., N = 22), intercept at V rest was 3.3 ± 1.4 mV, and mean slope was a shallow 0.14 ± 0.09. In occasional cells (such as that of Figure 2 H), the standard deviation of potential did not grow monotonically with V mean . The ratios of variance to mean seen in membrane potentials were negligible when compared to those seen in firing rate. For the example simple cell, over the entire range of mean potentials the variance of potential grew by less than a factor of four ( Figure 2 B). By contrast, over the entire range of firing rates the variance of firing rate grew by a factor of almost 100 ( Figure 2 A). Similar results were obtained in the remaining cells, such as the complex cell of Figure 2 D and 2 E and the intermediate cell of Figure 2 G and 2 H. In the last cell, the difference between potential and firing rate was particularly striking, as the former shows a downward slope that is clearly absent in the latter. These differences in variability are meaningful because potential and firing rate were recorded from the same responses to the same set of stimuli. They are not simply due to differences in time scale ( Buracas et al. 1998 ; Kara et al. 2000 ) because firing rate and potential were sampled at the same resolution (100 Hz). Accounting for the Variability of Firing Rate The origin of the large variability in firing rate lies not in an unforeseen source of noise, but rather in a deterministic mechanism, the nonlinear transformation of potentials into firing rates. This transformation ( Figure 1 H) can be fitted by a simple rectification model ( Granit et al. 1963 ) describing how firing rate R grows with potential V once this potential is above a threshold V thresh . As expected ( Anderson et al. 2000b ; Carandini and Ferster 2000 ), this rectification model captures the relation between potential and firing rate ( Figure 1 H, curve) and can be used to predict the rough features of firing rate both in individual trials (compare Figure 1 A and 1 I) and in averages across trials (compare curves in Figure 1 B and 1 C with those in Figure 1 J and 1 K). Of course, rectification is not a full account of the transformation between synaptic inputs and firing rates. Indeed, the relationship between firing rate and potential exhibits substantial error bars ( Figure 1 H). These error bars do not denote noise involved in generating spikes, which is negligible ( Calvin and Stevens 1968 ; Mainen and Sejnowski 1995 ; Carandini et al. 1996 ). They simply indicate that (as evident in the Hodgkin–Huxley equations) instantaneous potential is only one of the determinants of firing rate; additional determinants include the membrane potential's recent history ( Azouz and Gray 1999 ) and frequency content ( Carandini et al. 1996 ; Volgushev et al. 2002 ). Despite its simplicity, the rectification model is sufficient to predict the large variability of firing rate, and the increase of firing-rate variance with firing-rate mean. The predicted standard deviation resembles the measured one both in amplitude and in time course (compare shaded areas in Figure 1 B and 1 C with those in Figure 1 J and 1 K). Indeed, a plot of variance versus mean for the predicted firing rate ( Figure 2 C) indicates almost as much variability as that seen for the actual firing rate ( Figure 2 A). Similar results were obtained in the other example cells (compare Figure 2 D to 2 F, and 2 G to 2 I) and in the rest of the population ( Figure 3 A and 3 B). While the rectification model often underestimated the vertical intercept of the line relating mean and variance ( Figure 3 A), it generally captured the line's slope ( Figure 3 B). The model, therefore, accounts for the growth of firing-rate variance with the mean. Figure 3 Performance of the Rectification and Gaussian–Rectification Models in Predicting Firing-Rate Variability Distributions of firing-rate variance versus firing-rate mean were fitted with a line in logarithmic scale, corresponding to the equation variance = a mean b , where a is the intercept of the line and b is the slope of the line. Fitting was performed on the measured distributions (e.g., Figure 2 A), on the distributions predicted by the rectification model (e.g., Figure 2 C), and on those predicted by the Gaussian–rectification model (e.g., Figure 6 B). Dashed lines indicate predictions for a Poisson process. (A) Comparison of measured intercept versus predicted intercept. Diagonal line indicates equality between measured and predicted values. (B) Same, for the slope. (C and D) Same as in (A) and (B), for the predictions of the Gaussian–rectification model. The reason why the rectification model explains the large variability of firing rate is rather intuitive. Trial-to-trial fluctuations in potential are critical to obtain spikes, because many visual stimuli (such as the 210° grating in Figure 1 ) elicit a mean potential that barely reaches the firing threshold ( Anderson et al. 2000b ). Therefore, small fluctuations in membrane potential make the difference between a trial with few or no spikes and one with plenty of spikes. In other words, the firing threshold amplifies small fluctuations in potential into large fluctuations in firing rate. Perhaps less intuitive is the reason why the rectification model explains the growth of firing-rate variance with firing-rate mean. One may think that a necessary condition for this effect is the growth in potential variance observed with increasing mean potential ( Figure 2 B). This is not the case: The variance of potential could stay constant or even decrease (as it does for the cell in Figure 2 H), and the variance of firing rate would still grow with the mean ( Figure 2 G). Predicting the Variability of Firing Rate An intuition and a quantitative account for these properties can be obtained by applying the rectification model to an idealized random distribution of potentials, which we take to be Gaussian. Such a Gaussian–rectification model has been used to explain the dependence of mean firing rate on mean synaptic input ( Anderson et al. 2000b ; Hansel and van Vreeswijk 2002 ; Miller and Troyer 2002 ). It resembles a model proposed by Abeles (1982 , 1991 ) to study neuronal integration time. In the Gaussian–rectification model, the stimulus determines the mean of the Gaussian ( Figure 4 B), and the portion of Gaussian that crosses threshold determines the distribution of firing rates ( Figure 4 A). The mean of the Gaussian is the average potential V mean evoked by the stimulus at that instant ( Figure 4 B). The rectification function ( Figure 1 H) operates on this distribution and determines the distribution of firing rates ( Figure 4 A): Each potential contributes a firing rate given by the rectification function, with a probability given by the value of the Gaussian at that potential. When mean potential V mean is low, the Gaussian lies mostly below the threshold V thresh , so the predicted firing rate is mostly zero ( Figure 4 A, a ). When V mean is higher, however, the tail of the Gaussian that lies above threshold becomes substantially larger, and the distribution of firing rates reaches higher rates ( Figure 4 A, e ). The large peak at 0 spikes/s corresponds to the area of the Gaussian that lies below V thresh . Figure 4 The Gaussian–Rectification Model (A and B) Distributions across trials of model potential V (B) and of model firing rate R (A) for five values of the mean potential V mean . Firing rate is obtained from potential by applying the rectification model ( Figure 1 H). The value for R = 0 is shown at 1/3 of veridical height. (C and D) Mean (data points) and standard deviation (error bars) for the distributions in (A) and (B) as a function of mean potential V mean . Curve and shaded area indicate model predictions for the full range of mean potentials. Arrows indicate the five mean potentials (± 2 mV) used in (A) and (B). Throughout, dashed lines indicate resting potential V rest and dotted lines indicate firing threshold V thresh . Such a simple model is sufficient to predict that the variance of firing rate should increase with mean firing rate. As mean potential V mean increases, the distribution of firing rate becomes broader ( Figure 4 A), increasing not only in mean but also in standard deviation ( Figure 4 C). This phenomenon occurs even though in the model the standard deviation of potential is the same at all mean potentials ( Figure 4 D). The main assumption of the model, that of a Gaussian distribution of potentials, is generally borne out by the data. In most cells, the distribution of potential is close to a Gaussian, especially at the lowest values of mean potential, where spiking seldom occurs ( Figure 5 B). For the example simple cell, the distribution of z -scores (the difference between potential and mean potential, normalized by the standard deviation at that potential) appears remarkably Gaussian ( Figure 6 A). Similar results were obtained in the other cells (e.g., Figure 6 C and 6 E), although in some cells the tails of the distributions exceeded those of a Gaussian, and a large skewness clearly favored the more depolarized tails (not shown). A Gaussian distribution of potentials is commonly predicted in the theoretical literature (e.g., Svirskis and Rinzel 2000 ; Amemori and Ishii 2001 ; Rudolph and Destexhe 2003 ). It would be expected in a passive membrane summating many independent, high-rate presynaptic spike trains ( Rice 1944 ; Tuckwell 1988 ). Figure 5 Application of the Gaussian–Rectification Model to the Data from the Example Simple Cell (A and B) Distributions across trials of potential V (B) and of firing rate R (A) for five values of the mean potential V mean . Curves are best-fitting Gaussians (B) and predicted distributions of firing rate (A). Bin for R = 0 is shown at 1/3 of veridical height (and is three times wider than the others so that area is veridical). (C and D) Mean (data points) and standard deviation (error bars) for the distributions in (A) and (B), as a function of mean potential V mean . Curve and shaded area indicate model predictions for the full range of mean potentials. Arrows indicate the five mean potentials (± 2 mV) used in (A) and (B). Even a reduced model with constant standard deviation of potential (D, shaded area) predicts a growing standard deviation (A, shaded area). Figure 6 Variability of Potential and Predictions of the Gaussian–Rectification Model for Three Cells (A) Distribution of normalized deviations from the mean ( z -scores) for the potential of the simple cell in Figure 1 and Figure 2 A– 2 C. These were computed by subtracting from each potential the corresponding mean potential V mean (the abscissa in Figure 2 B) and dividing by the standard deviation (the square root of the ordinate in Figure 2 B). The results were cumulated. The curve is a normal Gaussian. (B) Variance versus mean for firing rate for the same cell and its prediction by the Gaussian–rectification model. Data points are same as Figure 2 A. Red curve: prediction of Gaussian–rectification model Shaded area: region where the Gaussian–rectification model predicts the occurrence of 75% of the points. Line is linear regression. (C and D) Same as (A) and (B) for the complex cell in Figure 2 D– 2 F. (E and F) Same, for the intermediate cell in Figure 2 G– 2 I. The Gaussian–rectification model has four parameters. Three of these parameters describe the rectification stage and are thus fully constrained by the measured relationship between potential and firing rate ( Figure 1 H). The remaining parameter, the standard deviation of the Gaussian, σ, was obtained from maximum likelihood estimation, i.e., by searching for the standard deviation that maximized the probability of observing the distributions of firing rate ( Figure 5 A). The result, σ = 4.6 mV, slightly overestimates the standard deviation observed for low mean potentials, but correctly estimates it at higher mean potentials ( Figure 5 D, compare shaded area to error bars). The model predicts the main features of the distributions of firing rate ( Figure 5 A). It predicts that when mean potential is low (e.g., V mean = −64 mV; Figure 5 A, a ), the firing rate is always zero, whereas larger mean potentials yield a distribution of firing rates that spans values from 0 to 300 spikes/s (e.g., V mean = −54 mV; Figure 5 A, d ). Deviations from the predictions are largest where they are least significant, i.e., at high firing rates for the high values of V mean (e.g., V mean = −50 mV; Figure 5 A, e ). These high values were achieved seldom; for example, only 21 data points were obtained at V mean = −50 mV ( Figure 5 A, e ), compared to 273 at V mean = −54 mV ( Figure 5 A, d ) and 1,575 at V mean = −64 mV ( Figure 5 A, a ). In fact, the model closely predicts both the firing rate's mean and standard deviation ( Figure 5 C). It predicts the two key effects of increasing mean potential: (1) an increase in the firing rate's mean (as a power law; Anderson et al. 2000b ; Hansel and van Vreeswijk 2002 ; Miller and Troyer 2002 ), and (2) an increase in the firing rate's standard deviation. Crucially, the model closely predicts how firing-rate variance depends on firing-rate mean ( Figure 6 B, red curve). Because of noise in the estimation of variance from a limited number of measurements (in this experiment, seven trials), the data are not expected to fall exactly on the model's prediction; Monte Carlo simulations with a matched number of trials determined the area in which 75% of the observations are predicted to fall ( Figure 6 B, gray area). Similar results were obtained in the remaining cells of the population, except that the model has a mild tendency to underestimate the intercept and overestimate the slope of the relation between variance and mean ( Figure 3 C and 3 D). Overall, the Gaussian–rectification model applied to the trace of mean potential performed as well as the rectification model applied to the individual traces of potential. Both models underestimated the intercept of the lines fitted to the relationship between firing-rate variance and mean: the rectification model by 25 ± 42% ( Figure 3 A), and the Gaussian–rectification model by 44 ± 26% ( Figure 3 B). Both models correctly estimated the slope of the line (the growth in variance with increasing mean), with insignificant errors of 0.10 ± 0.25 for the rectification model ( Figure 3 C), and −0.01 ± 0.22 for the Gaussian–rectification model ( Figure 3 D). This performance is remarkable, given that the Gaussian–rectification model replaces detailed knowledge of potential in individual trials with just one free parameter, the standard deviation σ of potential. These results illustrate how the key element in producing the steep growth in firing-rate variance observed with growing stimulus strength is the nonlinear transformation between potential and firing rate ( Figure 1 H). Indeed, the model was intentionally implemented with the constraint that the standard deviation of potential, σ, be constant. This constraint serves to demonstrate that a mild growth in variance of potential ( Figure 5 D, error bars) is not necessary to produce the steep growth in firing-rate variance ( Figure 5 C, error bars). Variability of Responses to Current Injection The predictions of the Gaussian–rectification model apply to any neuron that meets minimal criteria: a relationship between synaptic input and firing rate that is monotonic and includes a threshold, and noise in the input that has a Gaussian distribution. As an example, let us consider a neuron that is closer to biological reality than the Gaussian–rectification model, one that receives currents (not potentials) in its input and produces individual spikes (not continuous firing rates). In particular, consider an enhanced integrate-and-fire neuron, where each spike is accompanied by a temporary increase in spike threshold and by the entry of calcium, which in turn determines an after-hyperpolarization potassium current (see Materials and Methods ). To ensure realism, I fitted the model parameters to responses to injected currents of a regular spiking neuron. This neuron was recorded in vitro in the visual cortex of the guinea pig, in the near absence of synaptic inputs ( Carandini et al. 1996 ). The injected currents include sinusoids ( Figure 7 A, top four panels) and approximately Gaussian-distributed noise ( Figure 7 A, bottom panels). Once its parameters are appropriately tailored, the enhanced integrate-and-fire model accurately predicts the cell's responses, both in the subthreshold membrane potential waveforms and in the timing of individual spikes ( Figure 7 B and 7 C). Figure 7 Responses of a Regular-Spiking Neuron in the Visual Cortex to Current Injection, and Predictions by an Enhanced Integrate-and-Fire Model Neuron (A) Injected currents were sinusoids or noise waveforms. Noise was obtained by summing eight sinusoids with incommensurate frequencies. (B) Membrane potential responses of a regular-spiking neuron (cell 19s2, experiment 4) recorded with sharp electrodes in a study of guinea pig visual cortex in vitro ( Carandini et al. 1996 ). (C) Predictions of an enhanced integrate-and-fire neuron model fine-tuned to resemble the responses of the cell. Just as predicted, this spiking neuron responds to noisy injected currents with a firing rate whose variance grows with the mean ( Figure 8 ). To simulate the synaptic drive to a simple cell recorded in vivo ( Figure 1 A– 1 D) I injected sinusoidal currents, to which I added Gaussian noise. The model responses ( Figure 8 A– 8 D) resemble those seen in vivo ( Figure 1 A– 1 D). The firing rate is highly variable ( Figure 8 B), with a standard deviation that is roughly comparable to the mean ( Figure 8 C and 8 D), even though the standard deviation of the injected current is constant ( Figure 8 H and 8 I). In fact, for firing rate the variance grows proportionally to the mean ( Figure 8 E), even though for injected current the variance is constant ( Figure 8 J). Figure 8 Variability in the Responses of the Spiking Model Neuron (A) Response of the model neuron to a 0.6-nA sinusoidal current in the presence of Gaussian noise (s.d. 0.25 nA). (B) Corresponding firing rate. Three trials are shown. (C) Firing rate averaged over 16 trials. Shaded area indicates 2 s.d. (D) Same, for three other stimuli: a 0.4-nA sinusoid (top), a 0.2-nA sinusoid (middle), and noise alone (bottom). (E) Variance versus mean for firing rate. Diagonal line is prediction for a Poisson process. Red curve: prediction of Gaussian–rectification model, with no parameters allowed to vary to fit the data. Shaded area: region where the Gaussian–rectification model predicts the occurrence of 75% of the points. (F) Relation between firing rate and injected current. Curve is fit of rectification equation. (G–I) As in (B–D), for injected current. (J) Variance versus mean for injected current. The Gaussian–rectification model captures the essence of this behavior. Once it is given the standard deviation of the noise and the relationship between injected current and firing rate ( Figure 8 F), the Gaussian–rectification model makes a parameter-free prediction of the relationship between variance and mean ( Figure 8 E, curve). This prediction is not perfect (it consistently underestimates firing-rate variance), but it does capture the most important behavior: that variance grows with the mean for firing rate ( Figure 8 E, curve) but not for injected current ( Figure 8 J, horizontal line). Similar results were obtained when the stimulus parameters were changed to simulate synaptic inputs to a complex cell, or when the parameters of the spiking neuron were changed to simulate other cells measured in vitro, or even chosen randomly within reasonable bounds. As predicted, as long as the relationship between synaptic input and firing rate involved a threshold and the input noise was Gaussian, the variance grew with the mean for firing rate but not for injected current. Role of Firing-Rate Encoder Having validated the Gaussian–rectification model, we can now investigate the role of its parameters in determining the curves relating firing-rate variance and mean ( Figure 9 ). The model has four parameters (see Materials and Methods ): (1) the standard deviation σ of potential, (2) the firing threshold, V thresh , (3) the gain k of the relationship between firing rate and potential above threshold, and (4) the exponent n of this relationship. For the purpose of studying the model, we can assume, without loss of generality, that potential is unitless and has standard deviation σ = 1. Then, because V thresh can only determine the range of firing rates that is achieved, only k and n control the shape of the variance versus mean curves ( Figure 9 ). Figure 9 Role of Parameters of Gaussian–Rectification Model The standard deviation of potential was set to σ = 1, so that the shape of the curves relating firing-rate variance to firing-rate mean depends entirely on the gain k and the exponent n of the curves relating firing rate to membrane potential. The effects of these two parameters are explored: varying n (columns) and varying k (rows). Red curves: predictions of the Gaussian–rectification model; shaded areas: regions where the model predicts the occurrence of 75% of the points. Insets illustrate the corresponding curves relating firing rate to membrane potential. The gain k controls curve position, and the exponent n controls curve shape ( Figure 9 ). Increasing the gain k lifts the curves upward by twice as much as it shifts them rightward ( Figure 9 , rows). These shifts occur because variance grows with k 2 and mean grows with k. Decreasing the exponent n causes the curves to saturate ( Figure 9 , columns): The variance saturates to a plateau if n = 1 ( Figure 9 , middle), and it reaches a maximal value and then decreases if n < 1 (as in Figure 9 , left). Saturation occurs because when potential goes well above threshold, increases in mean potential cease to reveal ever larger portions of the Gaussian. If the curves relating firing rate to potential saturate ( n < 1), variations in potential are compressed into proportionally ever smaller variations in firing rate; the opposite occurs if the curves expand ( n > 1). This analysis predicts that it should be fairly common for the firing-rate variance to saturate at high firing rates, possibly showing a plateau or even a decrease. Indeed, in the sample of V1 neurons recorded intracellularly, exponents are typically close to unity ( n = 1.1 ± 0.6). Accordingly, a mild form of saturation is common in the plots of firing-rate variance and mean ( Figure 6 B). To quantify the saturation, however, one needs reliable estimates of firing-rate variance. These estimates are not very reliable in the intracellular sample, which typically involves only a few hundred spikes per cell, leading to large clouds of points in the scatters of variance versus mean ( Figure 6 B). Variability of Extracellularly Recorded Firing Rates To test the model's prediction rigorously, I considered a set of V1 responses obtained with extracellular recordings. Thanks to the large number of spikes (commonly >4,000 per cell), measurements in this dataset yield more precise estimates of firing-rate variance over a wider range of firing rates than are available in the intracellular sample. An analysis of firing-rate variance versus mean for these extracellularly recorded cells supports the predictions of the model ( Figure 10 ). Extracellular data do not afford independent estimates of gain k and exponent n of the transformation of potential into firing rate. I thus first computed the model predictions for a variety of combinations of k and n (such as those shown in Figure 9 ). I then made Bayesian estimations of the values of k and n that maximize the likelihood of the data, while imposing a broad prior for n = 1.1, the median value measured intracellularly. The quality of these two-parameter fits was excellent ( Figure 10 ), of higher quality than could be obtained by fitting a line, the two-parameter “model” commonly used to describe data of this kind ( Figure 2 A). Moreover, a number of cells exhibited the saturation in variance predicted by the model. The eight representative cells shown in Figure 10 are arranged in order of increasing exponent n. The first three ( n = 0.9 to 1.0) show evident saturation in firing-rate variance as mean firing rate increases. The remaining five show a milder saturation, as expected from their higher exponents ( n = 1.1 to 1.2). Saturation was common, as the median n was 1.06, with n < 1 in 13/37 cells. Yet to my knowledge, except for an anecdotal account ( Mechler 1997 ), this common property had not been previously reported. It constitutes further support for the usefulness of the Gaussian–rectification model. Figure 10 Relationship between Variance and Mean for Eight Cells Recorded Extracellularly in Cat V1, and Fits by the Gaussian–Rectification Model For each mean firing rate, data point and error bars indicate mean ± 1 s.d. of the observed variance. Red curves and shaded areas are predictions of the model. Values of exponent n and gain k are reported next to each graph. Cells are arranged in order of increasing exponent n. Discussion We have seen that a large amplification takes place between the trial-to-trial variability of synaptic input and that of firing rate: The variance of synaptic input is small compared to the dynamic range, and it is roughly constant. The amplification of variability arises from the threshold in the transformation of synaptic input into firing rate. A Gaussian–rectification model attributes this amplification to very simple causes: approximately constant Gaussian noise in the input, and rectification due to threshold in the output. It indicates that firing-rate variance would grow with the mean even if the variance of synaptic input were constant. Both of the assumptions of the model, constant Gaussian noise and rectification, are borne out by the data. These assumptions are rather minimal, so they are naturally satisfied by more realistic models. For example, a realistic integrate-and-fire model behaves as predicted: Once it is given constant Gaussian noise in the input, it produces a firing-rate variance that grows with the firing-rate mean. Further support for the Gaussian–rectification model comes from its novel, and correct, prediction that firing-rate variance should saturate at high firing rates. In confirming this prediction I showed that the model can be used to account for variability in firing rate without knowledge of cellular properties. The extension to extracellular data is important because extracellular methods constitute the norm in visual neurophysiology, especially in awake animals, and are the ones used in previous studies of firing-rate variability. These results further lengthen a list of properties of V1 neurons that are simply explained by the firing threshold. In addition to the amplification of trial-to-trial variability demonstrated here, these include the sharpening of tuning for stimulus direction and orientation ( Jagadeesh et al. 1993 ; Carandini and Ferster 2000 ; Volgushev et al. 2000 ), the power-law behavior of firing rate at low contrast ( Heeger 1992 ; Anderson et al. 2000b ; Hansel and van Vreeswijk 2002 ; Miller and Troyer 2002 ), and even the establishment of the dichotomy between simple and complex cells ( Carandini and Ferster 2000 ; Mechler and Ringach 2002 ; Priebe et al. 2004 ). It is remarkable that a mechanism as simple as the firing threshold can determine phenomena that might prima facie require more complex explanations at the level of the network. Limitations of the Approach One limitation of this study lies in the use of coarse potential. Coarse potential is not completely independent of firing rate: Even when spikes are removed and the traces smoothed, there is still a likely contribution of active conductances that has not been removed. Fortunately, this limitation strengthens my observation that coarse potential is not nearly as variable as firing rate: Any unwanted remaining echo of the spikes would make coarse potential more similar to firing rate and, thus, more variable. Therefore, in reality the variance of the actual synaptic input might be even less dependent on the mean than appears, for example, in Figure 2 B, 2 E, and 2 H. A partial control for these effects would be to perform some of the measurements while blocking spikes. However, blocking spikes would prevent the key measurements of this study, which require concurrent measurement of firing rate and estimation of synaptic input. Another limitation of the approach is that I have mostly considered firing rates, not individual spikes. Unlike firing rates, individual spikes can occur only in integer numbers and are separated by refractory periods. These properties can become relevant to response variability, for example, if firing rates become so high that refractory period becomes a limiting factor ( Kara et al. 2000 ). Such concerns are assuaged by the realistic integrate-and-fire model ( Figure 7 ), which shows an increase of firing-rate variance with the mean similar to that predicted by the Gaussian–rectification model. As to the saturation in firing-rate variance that was observed in some neurons, it invariably occurred at firing rates much lower than predicted from the refractory period. A more serious limitation of coarse potentials and firing rates is that they make sense only in a limited range of time windows. The windows should be long enough to be able to contain more than one spike, and short enough that mean potential is approximately constant within the window. An informal analysis of the effect of time window indicates that a range of 5–20 ms is satisfactory. This range, however, might be appropriate only for V1 neurons; further investigations are required before applying these methods elsewhere. Finally, a broader limitation of this work is that it concentrates on variability across trials, with little bearing on another form of variability, the one observed within trials in the irregularity of spike trains ( Softky and Koch 1993 ; de Ruyter van Steveninck et al. 1997 ; Reich et al. 1997 ; Troyer and Miller 1997 ; Buracas et al. 1998 ; Shadlen and Newsome 1998 ; Stevens and Zador 1998 ). Thanks to recent advances, however, the cellular origins of this form of variability have been largely explained ( Reich et al. 1997 ; Stevens and Zador 1998 ). In particular, it is now clear that high variability within trials is to be expected if neurons receive synaptic inputs with slow temporal correlation ( Svirskis and Rinzel 2000 ). In fact, variability within trials is most evident with visual stimuli that provide a roughly stationary response, being greatly diminished with richer stimuli, which elicit highly precise responses ( Bair and Koch 1996 ; Reich et al. 1997 ; Buracas et al. 1998 ). Conversely, trial-to-trial variability is endemic, being present regardless of type of visual stimulus ( Reich et al. 1997 ; Buracas et al. 1998 ). Implications for Cortical Processing What computational advantage might cortical neurons derive by amplifying the variability that they receive in their input? Why reduce the signal/noise ratio? To answer these questions, it might help to clarify the sources of “signal” and “noise.” The main source of variability in synaptic inputs to a V1 neuron is likely to be intracortical because thalamic responses are half as variable ( Kara et al. 2000 ). Variability thus results largely from ongoing cortical activity ( Arieli et al. 1996 ; Buracas et al. 1998 ; Tsodyks et al. 1999 ; Kenet et al. 2003 ). It appears to us as noise simply because it is not synchronized with stimulus onset. By contrast, the mean across trials of potential or firing rate constitutes a signal that is driven entirely by the stimulus. The results of this study suggest that threshold affects the interaction between stimulus-driven activity and ongoing activity, turning it from additive to multiplicative. At the level of firing rates, this interaction is largely multiplicative because the variance of firing rate grows proportionally to the stimulus-driven mean firing rate. At the level of synaptic inputs, instead, this interaction is nearly additive because the variance of potential barely depends on the stimulus-driven mean potential. Indeed, additivity has been seen between local field potentials and ongoing voltage-sensitive dye signals ( Arieli et al. 1996 ). We have seen that the rectification due to firing threshold is single-handedly responsible for the variability of firing rate and is, thus, responsible for turning a largely additive interaction into a multiplicative interaction. It is thus conceivable that the computational role of firing threshold is to multiply stimulus-driven responses by ongoing cortical activity, i.e., to multiply what we call “signal” by what we call “noise.” What may appear as lowering the signal/noise ratio can in fact be seen as a useful process, one that progressively amplifies the ongoing activity that ultimately guides our actions. Materials and Methods Data acquisition in vivo Measurements in vivo were obtained in paralyzed, anesthetized cats. Methods for animal preparation and maintenance have appeared elsewhere ( Carandini and Ferster 2000 ) and were approved by the Animal Care and Use Committees at Northwestern University and at the Smith-Kettlewell Eye Research Institute. The 22 cells recorded intracellularly belong to a sample that has been analyzed in two previous studies by Carandini and Ferster (2000) and by Anderson et al. (2000a) . These studies describe in detail the recording methods, which involved the whole-cell patch technique. The electrical noise associated with this technique is commonly <0.1 mV (as judged from records obtained after losing the patch). From the sample I excluded a few cells that produced less than ten spikes per block of stimuli, or that failed to satisfy other minimal requirements (firing rate >2 spikes/s, spike height >10 mV). Stimuli were optimal gratings drifting in 12 directions in 30° intervals, and a blank screen of uniform gray. The resting potential V rest was taken as the mean potential measured with the blank screen. Coarse potential traces were obtained from traces of membrane potential sampled at 4 kHz by removing spikes ( Lankheet et al. 1989 ) and by applying a low-pass filter with a cutoff of 50 Hz ( Carandini and Ferster 2000 ; Volgushev et al. 2000 ). The same low-pass filter was applied to spike trains sampled at 4 kHz to yield firing rate. Both coarse potential and firing rate were subsampled at 100 Hz. The 37 neurons recorded extracellularly are part of a study of the organization of receptive fields and suppressive surrounds in area V1 ( Bonin et al. 2003 ). This dataset was chosen because it involved lengthy experiments that yielded many thousands of spikes per cell at a variety of firing rates. Recordings were made with quartz-coated platinum/tungsten microelectrodes; methods for data acquisition and animal maintenance have been described by Freeman et al. (2002) . Stimuli were drifting gratings presented at the optimal orientation, spatial frequency, and temporal frequency, and enclosed in one of 66 possible windows. The windows were stationary square gratings with variable period and orientation. Stimuli typically lasted 2 s, and each block of stimuli was typically repeated three to six times. Firing rates were extracted from the spike train by low-pass filtering at 50 Hz and were subsampled at 100 Hz. Data acquisition in vitro Measurements in vitro were made with sharp intracellular electrodes from slices of guinea pig visual cortex. Methods for this preparation were approved by the Animal Care and Use Committee at New York University. The cells are part of the dataset presented by Carandini et al. (1996) ; the cell in Figure 7 is the one whose responses are extensively illustrated in that study (cell 19s2). Rectification model The relation between potential V and firing rate R (e.g., Figure 1 H) was fitted with an extension of the rectification model ( Mechler and Ringach 2002 ), where R ( V ) = k [ V − V thresh ] n + , with [.] + indicating rectification, k a proportionality factor, and n an exponent. Fitted parameters were V thresh = −55.3 mV, k = 16.7, and n = 1.2 for the simple cell in Figure 1 , and V thresh = −46.6 ± 10.5 mV, k = 12.4 ± 7.9, and n = 1.1 ± 0.6 for the whole intracellular population ( N = 22). The distance between V thresh and V rest was 5.1 mV for the simple cell in Figure 1 , and 8.0 ± 4.2 mV for the population. Gaussian–rectification model The mean potential V mean in response to a stimulus was defined as the mean across trials of coarse potential. In the Gaussian–rectification model, the probability of observing a firing rate r ( Figure 4 A) given a mean potential V mean is where R ( V ) is the relation between firing rate and potential V ( Figure 1 H), and N [ V mean , σ ] is the probability distribution of potential ( Figure 5 B), a Gaussian with mean V mean and standard deviation σ. The value of p ( r ) depends on whether r is zero or positive: for r > 0, and for r = 0. The first expression is simply the value of the Gaussian for V = R −1 ( r ). The second expression is the area of the portion of Gaussian that is below threshold ( erf is the error function). These expressions allow maximum likelihood estimation of model parameters from measured firing rates. When parameters of the relation between firing rate and potential R ( V ) are obtained independently (in intracellular recordings; Figure 1 H), the only free parameter was the standard deviation σ of potential. Across the intracellular population, the average value of σ obtained by the fits was 5.4 ± 2.0 mV (s.d., N = 22). The required σ was always larger (by 2.1 ± 1.5 mV) than the standard deviations observed when V mean = V rest , but it was comparable (larger by only 0.9 ± 1.6 mV) to the standard deviations observed when V mean = V thresh . Statistics Let V mean ( t ) be the mean potential at time t from stimulus onset. Because the sample rate is 100 Hz, each time sample corresponds to a 10-ms interval. Of course, V mean ( t ) depends on the stimulus. To simplify the notation, however, consider the case of a single stimulus. Distributions for potential at a given mean potential ( Figure 5 B) were computed as follows: (1) Select a value of interest, v (e.g., v = −55 mV; Figure 5 B, a ); (2) find the times ( t k ) when the mean potential V mean ( t k ) is within 2 mV of v; (3) pooling across trials j, look at the distribution of potential [ V j ( t k )] (e.g., Figure 5 B, a ). Distributions for z -scores (normalized deviations from the mean) of potential ( Figure 6 A, 6 C, and 6 E) were computed as follows: (1) Divide the range of values of V mean in 1-mV intervals, with centers ( v i ); (2) for each interval i, find the set of times ( t ik ) when the mean potential is in the i -th bin; (3) pooling across trials j, compute σ i , the standard deviation of V j ( t ik ); (4) transform each V j ( t ik ) into a z -score: z ijk = [ V j ( t ik ) − v i ]/ σ i ; (5) look at the distribution of (z ijk ) (e.g., Figure 6 A). Enhanced integrate-and-fire model The enhanced integrate-and-fire model was derived in collaboration with Davide Boino (2000) by simplifying a model by Wang (1998) . The model neuron has a single compartment with membrane equation where the currents are: with t spike the time of the last spike. Ca ( t ) is the (unitless) calcium concentration: where the sum extends over all spikes with t spike < t. Spikes result from stereotyped conductances g Na ( t ) and g K ( t ) derived from Hodgkin–Huxley equations and are scaled to approximate the spikes from the recorded neuron. They occur when V m exceeds a threshold, which depends on the time since the last spike: The reversal potentials for sodium and potassium were set to V k = −80 mV and V Na = 55 mV. Passive parameters of the membrane ( C m = 120 pF, g leak = 12.4 nS, V leak = −60.3 mV) were obtained by fitting the membrane potential responses to sinusoids. The remaining parameters ( g AHP = 23.0 nS, Ca 50 = 10, τ Ca = 200 ms, V thresh = −43.5 mV, τ thresh = 36 ms) were obtained by a search algorithm aimed at maximizing the quality of the predictions for firing rate.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC509408.xml

526372

Surveillance of antimicrobial resistance at a tertiary hospital in Tanzania

Background Antimicrobial resistance is particularly harmful to infectious disease management in low-income countries since expensive second-line drugs are not readily available. The objective of this study was to implement and evaluate a computerized system for surveillance of antimicrobial resistance at a tertiary hospital in Tanzania. Methods A computerized surveillance system for antimicrobial susceptibility (WHONET) was implemented at the national referral hospital in Tanzania in 1998. The antimicrobial susceptibilities of all clinical bacterial isolates received during an 18 months' period were recorded and analyzed. Results The surveillance system was successfully implemented at the hospital. This activity increased the focus on antimicrobial resistance issues and on laboratory quality assurance issues. The study identified specific nosocomial problems in the hospital and led to the initiation of other prospective studies on prevalence and antimicrobial susceptibility of bacterial infections. Furthermore, the study provided useful data on antimicrobial patterns in bacterial isolates from the hospital. Gram-negative bacteria displayed high rates of resistance to common inexpensive antibiotics such as ampicillin, tetracycline and trimethoprim-sulfamethoxazole, leaving fluoroquinolones as the only reliable oral drugs against common Gram-negative bacilli. Gentamicin and third generation cephalosporins remain useful for parenteral therapy. Conclusion The surveillance system is a low-cost tool to generate valuable information on antimicrobial resistance, which can be used to prepare locally applicable recommendations on antimicrobial use. The system pinpoints relevant nosocomial problems and can be used to efficiently plan further research. The surveillance system also functions as a quality assurance tool, bringing attention to methodological issues in identification and susceptibility testing.

Background Exaggerated and irrational use of drugs, availability of antibiotics without prescription, the use of pharmaceuticals of doubtful quality and the HIV epidemic may all contribute to the current worldwide surge in antimicrobial drug resistance. Emerging resistance to antimicrobial drugs increases morbidity and mortality by hampering the provision of effective chemotherapy, and makes treatment more costly [ 1 - 3 ]. The surge in antimicrobial resistance seen in many low-income countries is potentially disastrous because of the lack of resources for purchasing expensive second-line drugs [ 4 ]. It is widely held that surveillance of antimicrobial susceptibility is fundamental to combat the emergence of resistance [ 5 ]. Surveillance must be global since resistant bacteria can be transferred between countries, but it must also be local, since countries have very different resistance patterns and different treatment practices [ 6 ]. The primary task of a surveillance system is to provide locally applicable data to guide empiric therapy. Furthermore, surveillance may help assessing the magnitude of the resistance problem locally, nationally and internationally, monitoring changes in resistance rates and detecting the emergence and spread of new resistance traits. A well-functioning surveillance system is also necessary to measure the impact of any interventions. Surveillance systems also functions as a quality assurance tool and may help improving the quality of the susceptibility testing. This paper describes the experience with the implementation of a computerized surveillance system for antimicrobial drug susceptibility at Tanzania's major referral hospital, and its use to analyze the susceptibility patterns of 7621 consecutively recorded clinical bacterial isolates. Methods Setting The study was performed at Muhimbili National Hospital (MNH), Dar es Salaam, Tanzania. With more than 1000 beds, MNH is the largest hospital in the country and serves as a national referral and university teaching hospital, as well as a primary and referral hospital for a population of approximately 3.6 million in the Dar es Salaam area. The Department of Microbiology and Immunology at MNH examines specimens from inpatients and outpatients at MNH, and from a number of nearby hospitals. Bacteriological cultures are performed on more than 23,000 specimens per year. The surveillance system A free-of-charge software for the surveillance of antimicrobial resistance (WHONET, World Health Organization, Geneva, Switzerland) [ 7 ] was implemented at MNH in 1998. Currently a total of 880 microbiology laboratories in 76 countries use this software, however, among these are only 41 laboratories in four countries on the African continent (data from 2002, personal communication from John Stelling, author of the WHONET software). The software has three main parts, a laboratory configuration file which can be used to customize it to the particular laboratory, an interface for data entry and a part for analysis and reporting of resistance data. At our hospital, all bacterial isolates of clinical significance from specimens received during the period July 1 st 1998 to December 31 st 1999 were recorded and analyzed. The specimens examined included urine, pus/secretions (swabs from skin, surgical and traumatic wounds, burns, umbilical cords, throat, nose, eye and ear discharge and genital swabs), blood, cerebrospinal fluid, other body fluids, stools and other specimens. Mycobacteria and anaerobic bacteria were not included in the study. Apart from the WHONET software, we used Stata 8.0 for Macintosh (Stata Corporation, College Station, Texas, USA) to evaluate differences of proportions by Fisher's exact test (2-tailed, cut-off point for statistical significance at p-value of 0.05). Laboratory methods The specimens were cultured and the bacterial isolates identified using standard microbiological methods as described in Mackie & McCartney Practical Medical Microbiology [ 8 ]. Susceptibility testing was performed by Stokes' method [ 9 ] on Iso-Sensitest (Oxoid Limited, Basingstoke, UK) agar plates. This method, developed by Dr Joan Stokes half a century ago, was designed to monitor for both disc and agar quality in that both the clinical isolate and a control strain were tested on every plate. The clinical isolate is swabbed onto the middle of the agar plate and the control strain at the periphery. The antibiotic disk is placed precisely at the interface between the surface areas inoculated with the clinical isolate and the control strain. After overnight incubation, the relative size of the inhibition zones of the clinical isolate and the control strains are compared. The test results are classified as susceptible (S), intermediate (I) or resistant (R) by evaluation of the difference between the inhibition zones of the clinical isolate and the control strain. The control strains used in our lab are S. aureus NCTC 6571, E. coli NCTC 10418 or Pseudomonas aeruginosa NCTC 10662. The isolates showing intermediate resistance were few and were grouped together with sensitive isolates for the purpose of data analysis. Either methicillin or oxacillin disks were used to test for methicillin-resistance in S. aureus , the results being considered equivalent and interchangeable in the data analysis. ß-Lactamase testing was not routinely performed. The susceptibility of pneumococci to penicillin was examined by the use of penicillin 2 μg disks. Commercially produced antibiotic disks, mostly obtained from Oxoid Limited, were used, however, in some instances, antibiotic disks, prepared locally were used due to financial constraints. The Department of Microbiology and Immunology participates in an external quality assessment program in bacteriology led by the World Health Organization-collaborating centre, the National Institute for Communicable Diseases (NICD), Johannesburg, South Africa. The Department of Microbiology and Immunology at our hospital receives bacterial strains from NICD, performs species identification and antimicrobial susceptibility testing, and report the results back to NICD. Evaluation of the surveillance system We evaluated the strengths and shortcomings of the surveillance system in our setting, particularly in terms of how well it performed in its main application areas, providing locally applicable data to guide empiric therapy, monitoring antimicrobial susceptibility trends, detecting the emergence and spread of new resistance traits and as a tool for quality assurance. We also assessed the cost-implications of implementing the surveillance program in our setting. We considered direct costs, such as the purchase of equipment, and indirect costs, such as those related to the running of the laboratory, including human resources. We also comment on the benefits of the surveillance system related to both direct patient care and long-term implications of containing antimicrobial resistance. Results Bacterial isolates A total of 7617 bacterial isolates were registered during the study period, of which 67.4% (n = 5134) were Gram-negative and 32.6% (n = 2483) Gram-positive. Table 1 shows the most frequently encountered bacteria, overall and from various specimen types. The majority of the isolates were obtained from pus (44.3%), urine (43.5%) and blood cultures (10.1%). Cerebrospinal fluid accounted for 0.4% of the isolates. Among the 2034 blood cultures, 15.9% (n = 323) yielded growth of a total of 326 pathogenic bacterial isolates and 447 Coagulase-negative staphylococci (CoNS) as shown in Table 1 . CoNS are potential pathogens and are increasingly considered as a cause of blood-stream infections. However, in many cases they are merely contaminants, i.e. bacterial isolates present on the skin surface, which are introduced in the blood specimen and grow in the blood culture, but do not produce disease in the patient. For CoNS isolates to be considered a probable pathogen, it is commonly required that they are recovered from two separate blood cultures. Since multiple blood cultures were not routinely taken from the same patient in the hospital, the susceptibilities of these isolates were not evaluated further. CoNS and various other Gram-positive probable contaminants, mostly Bacillus spp. were recovered from 22.0% (n = 447) and 6.9% (n = 141) of the blood cultures, respectively. Furthermore, five Candida spp. isolates and one Cryptococcus neoformans were recovered. Among the 49 Salmonella isolates, two were identified as S . Typhi, 16 as S . Typhimurium, 16 as S . Paratyphi B and one each as S . Paratyphi C, S . Enteritidis and S . Arizonae. Twelve Salmonella isolates were not serotyped. Among the 41gonococcal isolates, 28 (68.3%) were from genital swabs. Eleven (26.8%) gonococcal isolates were obtained from the neonatal ward, out of which 4 were specified as from eye discharge. Table 1 Frequency of pathogenic a bacterial isolates from different specimen types at Muhimbili National Hospital, Tanzania Organism Blood (%) Spinal fluid (%) Urine (%) Pus b (%) Other (%) Overall (%) Gram-negative isolates E. coli 27 (3.5) 0 (0.0) 1466 (44.2) 417 (12.3) 26 (21.8) 1936 (25.4) Klebsiella spp. 91 (11.8) 8 (23.5) 1036 (31.3) 603 (17.9) 33 (27.7) 1771 (23.3) Pseudomonas spp. 10 (1.3) 2 (5.9) 52 (1.6) 531 (15.7) 9 (7.6) 604 (7.9) Proteus spp. 7 (0.9) 0 (0.0) 121 (3.7) 249 (7.4) 3 (2.5) 380 (5.0) Enterobacter spp. 4 (0.5) 0 (0.0) 97 (2.9) 1 (0.0) 0 (0.0) 102 (1.3) Salmonella spp. 37 (4.8) 2 (5.9) 6 (0.2) 0 (0.0) 4 (3.4) 49 (0.6) N. gonorrhoeae 0 (0.0) 0 (0.0) 0 (0.0) 41 (1.2) 0 (0.0) 41 (0.5) Haemophilus spp. 1 (0.1) 5 (14.7) 0 (0.0) 0 (0.0) 0 (0.0) 6 (0.1) Other GNR 32 (4.1) 7 (20.6) 12 (0.4) 184 (5.4) 10 (8.4) 245 (3.2) Subtotal, Gram-negative isolates 209 (27.0) 24 (70.6) 2790 (84.2) 2026 (60.0) 85 (71.4) 5134 (67.4) Gram-positive isolates Staphylococcus aureus 72 (9.3) 1 (2.9) 362 (10.9) 1120 (33.2) 12 (10.1) 1567 (20.6) Streptococcus pyogenes 1 (0.1) 0 (0.0) 0 (0.0) 160 (4.7) 2 (1.7) 163 (2.1) Other streptococci c 39 (5.0) 3 (8.8) 52 (1.6) 58 (1.7) 13 (10.9) 165 (2.2) Enterococci 3 (0.4) 0 (0.0) 64 (1.9) 3 (0.1) 1 (0.8) 71 (0.9) S. pneumoniae 2 (0.3) 6 (17.6) 0 (0.0) 11 (0.3) 6 (5.0) 25 (0.3) CoNS a 447 (57.8) ... 45 (1.4) ... ... 492 (6.5) Subtotal, Gram-positive isolates 564 (73.0) 10 (29.4) 523 (15.8) 1352 (40.0) 34 (28.6) 2483 (32.6) Total 773 (100.0) 34 (100.0) 3313 (100.0) 3378 (100.0) 119 (100.0) 7617 (100.0) GNR, Gram-negative rod-shaped bacteria, not further identified; CoNS, coagulase-negative staphylococci; "...", not applicable. a CoNS from blood and urine specimens are reported as possible pathogens, although many may be contaminants. CoNS from other specimen types are considered contaminants and not reported. b Pus includes swabs from skin, surgical and traumatic wounds, burns, umbilical cords, throat, nose, eye and ear discharge and genital swabs. c Streptococci other than S. pyogenes and S. pneumoniae , and streptococci not identified below genus level. Specimens from inpatients and outpatients contributed to 53.2% and 31.9% of the isolates, respectively. A further 6.0% were obtained from specimens from other hospitals in Dar es Salaam, while 8.8% were obtained from other or unknown locations. Among the isolates from inpatients, 36.5% were obtained from the Department of Pediatrics, 28.4% from the neonatal section and 8.1% from the other pediatric wards. The other isolates came from the Departments of Surgery (22.4%), Internal Medicine (16.6%), Obstetrics and Gynecology (9.8%), the Intensive Care Unit (4.9%) and other locations (9,8%). For 4900 isolates, the age or the estimated age group of the patient was known. Of these, 23.6% (n = 1155) were from neonates (≤ 1 month old), 6.8% (n = 335) from children aged one month to seven years, and 69.6% (n = 3410) from adults or children older than 8 years. Antimicrobial susceptibility Tables 2 and Table 3 show the antimicrobial susceptibility patterns of the most frequently isolated Gram-negative and Gram-positive bacteria, respectively. There were no clear-cut differences in the antimicrobial susceptibilities among the various serotypes of Salmonella isolates (data not shown). The majority of Pseudomonas aeruginosa isolates was susceptibility-tested to gentamicin only, to which 4.3% (15/350) were resistant. Among the isolates of Neisseria gonorrhoeae , 70.0% were resistant to penicillin, 45.2% to tetracycline, 59.3% to trimethoprim-sulfamethoxazole, 5.9% to erythromycin and none was resistant to spectinomycin, fluoroquinolones or amoxicillin-clavulanate (data not shown). Table 2 Percentage of Gram-negative bacterial isolates resistant to antimicrobial agents (number of tested isolates in brackets) Drug E. coli Klebsiella spp. Proteus spp. Enterobacter spp. Salmonella spp. GNR Ampicillin 80% (1761) 85% (1572) 60% (331) 72% (86) 70% (46) 56% (204) Amoxicillin- clavulanate 28% (1292) 32% (1153) 17% (247) 32% (78) 52% (23) 31% (124) Ceftazidime 5% (788) 6% (605) 2% (95) 10% (51) 0% (8) 14% (35) Tetracycline 77% (1223) 66% (1016) 77% (211) 72% (54) 42% (12) 45% (153) Gentamicin 8% (1634) 14% (1538) 7% (343) 15% (91) 9% (23) 8% (217) Trimethoprim- sulfamethoxazole 76% (1313) 69% (1174) 57% (224) 70% (56) 73% (44) 51% (172) Sulfonamides 84% (174) 84% (231) 74% (46) 100% (14) 95% (22) 62% (34) Nitrofurantoin 32% (929) 53% (652) 72% (71) 48% (48) ... ... Chloramphenicol 45% (250) 51% (372) 55% (132) ... 20% (41) 57% (138) Fluoroquinolones 13% (432) 6% (343) 3% (65) 6% (32) 0% (20) 15% (40) Nalidixic acid 28% (509) 16% (334) 18% (22) 31% (16) ... ... GNR, Gram negative rod-shaped bacteria, not further identified; "...", not tested. Table 3 Percentage of Gram-positive bacterial isolates resistant to antimicrobial agents (number of tested isolates in brackets) Drug S. aureus CoNS Enterococci S. pneumoniae S. pyogenes Other strept. a Penicillin 97% (1521) 93% (42) 67% (9) 4% (23) 0% (163) 23% (98) Ampicillin ... ... 6% (66) ... ... 13% (83) Methicillin/ cloxacillin 2% (1556) 21% (47) ... ... ... ... Tetracycline 49% (1042) 90% (39) 76% (51) 8% (13) 47% (131) 61% (90) Erythromycin 29% (1543) 69% (48) 26% (65) 6% (18) 7% (161) 26% (156) CoNS, Coagulase-negative staphylococci; "...", not tested. a Streptococci other than S. pyogenes and S. pneumoniae , and streptococci not identified below genus level. Comparison of resistance patterns of isolates obtained from inpatients and outpatients at MNH did not show large differences. However, ampicillin resistance was more frequent in urinary isolates of E. coli from inpatients than in those from outpatients as shown in Table 4 . Likewise, urinary isolates of Klebsiella spp. from inpatients were more frequently resistant to gentamicin and trimethoprim-sulfamethoxazole than isolates from outpatients. Table 4 Percentage of urinary E. coli and Klebsiella spp. isolates from inpatients and outpatients resistant to antimicrobial agents E. coli Klebsiella spp. Drug Inpatients Outpatients P a Inpatients Outpatients P a Ampicillin 87.2 82.7 0.036 a 92.2 91.1 0.624 Amoxicillin- clavulanate 31.4 28.3 0.344 37.7 33.9 0.327 Ceftazidime 4.9 5.6 0.731 7.6 6.0 0.577 Tetracycline 83.1 81.7 0.648 82.0 75.2 0.053 Gentamicin 8.6 7.7 0.572 14.9 5.4 <0.001 a Trimethoprim- sulfamethoxazole 86.0 81.3 0.067 82.7 74.2 0.012 a Sulfonamides 92.1 87.8 0.510 95.2 100.0 0.553 Nitrofurantoin 33.7 33.1 0.881 52.1 58.0 0.157 Fluoroquinolones 17.8 12.7 0.217 7.2 6.7 1.000 Nalidixic acid 29.0 28.2 0.913 14.0 18.8 0.334 a P < 0.05 (Fisher's exact test, 2-tailed) indicates statistical significance of the differences in resistance rates. Comparison of resistance patterns in isolates blood cultures with those from other specimen types showed apparent great differences for some drugs, however, in most cases the number of blood culture isolates were few and did not show statistically significant differences. However, as shown in Table 5 , blood culture isolates of Klebsiella spp. were indeed more frequently resistant to gentamicin than those from other specimen types. A significantly greater proportion of blood culture isolates of S. aureus were resistant to tetracycline than among those from other specimen types, whereas for penicillin the isolates from blood cultures were resistant in a lower proportion than the others. Table 5 Percentage of bacterial isolates from different specimen types resistant to antimicrobial agents E. coli Klebsiella spp. S. aureus Drug Blood Other P a Blood Other P a Blood Other P a Penicillin ... ... ... ... ... ... 91.5 96.9 0.028 a Ampicillin 84.0 79.4 0.803 84.3 85.3 0.759 ... ... ... Amoxicillin- clavulanate 40.0 27.8 0.383 29.8 31.9 0.873 ... ... ... Methicillin ... ... ... ... ... ... 1.4 2.2 1.000 Ceftazidime 0.0 5.3 1.000 5.7 6.0 1.000 ... ... ... Tetracycline 54.5 77.3 0.139 66.7 66.4 1.000 84.6 48.3 <0.001 a Erythromycin ... ... ... ... ... ... 21.1 29.0 0.179 Gentamicin 13.0 7.7 0.416 41.3 12.3 <0.001 a ... ... ... Trimethoprim- sulfamethoxazole 72.0 76.3 0.636 63.0 69.1 0.297 ... ... ... Sulfonamides 83.3 84.0 1.000 86.8 83.4 0.809 ... ... ... Chloramphenicol 58.3 43.8 0.199 57.9 49.3 0.200 ... ... ... Fluoroquinolones 40.0 13.1 0.136 0.0 6.5 0.381 ... ... ... "...", not applicable. a P < 0.05 (Fisher's exact test, 2-tailed) indicates statistical significance of the differences in resistance rates. Evaluation of the surveillance system A great number of bacterial isolates were recorded in the system. All age groups and both inpatients and outpatients were represented in the study. More than a third of the isolates were from outpatient populations from the Dar es Salaam area, however we cannot exclude the possibility of a selection bias in favour of patients with infections caused by resistant organisms, since many patients get treatment at primary health facilities before reaching MNH. We do not know how well the rural population is represented in this material, but we assume that the outpatients in the study are mostly from the Dar es Salaam area. Ten percent of the isolates represented systemic infections, i.e. isolates from blood cultures and spinal fluid. The susceptibility test results were recorded as interpreted values (i.e. "R" (resistance), "I" (Intermediate) or "S" (susceptible)) and not as inhibition zone diameters. In this study, no molecular techniques were available for the detection of resistance genotypes and evaluation of genetic relatedness of bacterial isolates. The direct cost of implementing the surveillance system was limited to the purchase of a computer at approximately 1000 Euro. However, less expensive second-hand computers would be sufficient. The software was downloaded free of charge from the WHO website. The indirect costs of running this surveillance program are related to human sources for operating the software, including data entry and analysis, and the costs of the susceptibility testing activities. It is difficult to separate these indirect costs from the costs of running the daily laboratory activities. In our setting, a laboratory technologist from the department took on the task of operating the software in addition to her regular duties. In our experience, for a hospital of our size, it is recommendable to allocate approximately 50% of a laboratory technologist position to operating the surveillance software. In our setting, this would translate into a monthly cost of approximately 100 Euro for the department. The surveillance system is dependent on susceptibility testing of acceptable quality. The susceptibility testing incurs costs related to human resources and the purchase of laboratory reagents including antimicrobial disks and agar media. Implementing a surveillance system may increase these costs by focusing on the importance of quality reagents. However, since the susceptibility testing activities are an integral activity of the department, which would have been performed regardless of the surveillance system, we choose not to attribute their costs to the surveillance system in this context. The benefits of a surveillance system are difficult to quantify, but are of potentially great magnitude. Foremost, surveillance data may improve empiric therapy for infections and thus save lives and reduce suffering. It may reduce treatment costs by enabling the use of the least expensive effective drugs. Additionally, surveillance systems may contribute to containing or reducing antimicrobial resistance, which in the long term perspective may have great benefits in reducing morbidity and mortality, and diminish the need for expensive second-line antimicrobial agents. The strengths and weaknesses are elaborated on in the Discussion part. Discussion Resistance patterns and implications for therapy Experience from the World Health Organization's External quality assurance system for antimicrobial susceptibility testing has shown that disk diffusion testing is suitable for routine surveillance [ 10 ]. However, disk diffusion is not optimal for testing of certain important resistance traits, such as penicillin-resistance in pneumococci. The lack of international standardization of methods and interpretive criteria causes concern, but there are indications that routine susceptibility testing data are suitable for surveillance even if obtained with different methods [ 11 ]. Consistent with observations from a number of other countries in the region [ 12 - 15 ] and elsewhere [ 16 ], Gram-negative bacilli displayed high rates of resistance to common inexpensive antibiotics. Reasonably priced antibiotics such as ampicillin, tetracycline, trimethoprim-sulfamethoxazole and sulfonamides are now of limited benefit in the treatment of infections caused by important Gram-negative bacteria such as E. coli , Klebsiella spp., Proteus spp. and Salmonella . Chloramphenicol may fail to cure as much as a quarter of infections caused by Salmonella and half or more of infections caused by E. coli , Klebsiella spp. and Proteus spp. Fluoroquinolones appear to be the only reliable drugs for oral treatment of infections caused by common Gram-negative bacilli, whereas gentamicin and third-generation cephalosporins remain useful for parenteral therapy. The study showed a very low prevalence of methicillin-resistant S. aureus , consistent with previous data from the same hospital [ 17 , 18 ]. While the results should be interpreted with some caution since confirmatory nucleic acid based techniques were not available, the data support the current use of isoxazolyl penicillins, such as cloxacillin for the treatment of staphylococcal infections at the hospital. There were few isolates of enterococci compared to studies from high-income countries [ 19 ]. It is reassuring that the current study showed a low rate of ampicillin-resistant enterococci, indicating that nosocomial infections caused by these micro-organisms is a minor problem compared with many high-income countries. Low consumption of broad-spectrum antibiotics such as third-generation cephalosporins, fluoroquinolones, imipenem and vancomycin may explain this finding [ 19 - 21 ]. While other countries in the region have been affected by penicillin-resistant pneumococci [ 22 , 23 ], the current study indicates that pneumococcal disease in Dar es Salaam can safely be treated with penicillin or erythromycin. However, the results should be interpreted with some caution since the number of isolates tested was small. More than a quarter of the gonococcal isolates (11/41) were obtained from the neonatal ward, and most or all of these isolates probably represent gonococcal conjunctivitis. Amoxicillin-clavulanate, spectinomycin, fluoroquinolones and erythromycin appear to be good alternatives for the treatment of gonococcal infections. An apparent increase in resistance to trimethoprim-sulfamethoxazole (from 18% to 59%) is noted since the study by Mbwana [ 24 ] from 1993 to 1995, however, this may be due to the use of different methodology for susceptibility testing. Applicability of data to guide treatment of serious infections Recommendations for antibiotic treatment of serious bacterial infections such as bloodstream infections and meningitis should preferably be based on knowledge of the prevalence and antimicrobial susceptibility patterns of pathogens isolated from blood and spinal fluid. While a fair number of bacterial isolates were tested in the current study, the number of blood culture isolates was limited (n = 329, excluding the CoNS isolates). As shown in Table 5 , there appears to be differences in resistance between isolates obtained from blood cultures and those from other specimen types, but these are difficult to assess because of the low number of blood culture isolates. Thus, the data from the current surveillance should be interpreted with caution with regards to the treatment of serious infections. The CoNS isolates obtained from blood were recorded in the WHONET database, since they may represent clinically important infections such as bacteremia in patients with compromised immunity, patients with indwelling intravascular devices and the newborn [ 25 ]. The study showed that a high proportion (21.9%) of blood culture bottles yielded CoNS isolates. However, the conventional way to distinguish pathogenic isolates of CoNS from contaminants, by requiring growth of a similar CoNS isolate in a separate blood culture, could rarely be used, since follow-up cultures were seldom available. Consequently, the susceptibilities of these isolates were not evaluated further. Relevance of data for outpatient and rural populations It is important to specify for which population the surveillance data are valid. At our hospital, specimens from both inpatients and outpatients were examined. The hospital is to a great extent used as a primary hospital for the population in the Dar es Salaam area. However, among the cases coming to the hospital, there may be a degree of selection of patients with infections caused by resistant microbes, since many patients rely on health centers and pharmacies to cure simple ailments, and only come to the hospital when primary treatment fails. The study found that a few resistance traits, such as ampicillin resistance in E. coli and gentamicin and trimethoprim-sulfamethoxazole resistance in Klebsiella spp. were more frequent in urinary isolates from inpatients than from outpatients. Apart from that, there were no dramatic differences between isolates from inpatients and outpatients. The data from the study should be representative for both the hospital setting and to some degree the population in Dar es Salaam. However, the majority of the population of Tanzania lives in rural areas, where resistance patterns may be substantially different. Thus one should be cautious to extrapolate the results of the current study to be valid for populations in the countryside. Ability to monitor trends of antimicrobial susceptibility Certain trends in antimicrobial susceptibility could be identified by comparison with data from other studies. While resistance to ampicillin, tetracycline and sulfonamides in Gram-negative bacteria was frequent already in the seventies [ 26 , 27 ], it is worrying that resistance to trimethoprim-sulfamethoxazole, chloramphenicol, nitrofurantoin, nalidixic acid and amoxicillin-clavulanate appear to have increased compared to previous studies [ 27 , 28 ]. The extensive use of chloramphenicol for the treatment of presumed cases of typhoid fever and the use of trimethoprim-sulfamethoxazole for the ambulatory treatment of chest infections, malaria and other infections, may have contributed to the high prevalence of resistance to these two drugs. Although still low, it is of concern that the rate of gentamicin-resistance in E. coli has increased from zero in 1978–79 [ 27 ] to 2% in 1995 [ 28 ] and 8% in the current study. In neighboring Kenya, the rate of gentamicin-resistance in E. coli has increased from 2% in the late seventies [ 29 ] to 20% and above in recent studies [ 12 ]. Resistance to gentamicin is common in Gram-negative bacteria with extended-spectrum beta-lactamases (ESBL), sometimes in as much as 96% of isolates [ 30 ]. Such an association cannot be investigated in the current study, since less than half of the isolates of E. coli and Klebsiella spp. were tested for susceptibility to third-generation cephalosporins and other methods for detection of ESBL (double disk synergy test, Etest, PCR) were not available. Also in P. aeruginosa the rate of gentamicin-resistance has increased, from zero in the seventies [ 27 ] to 4% in the current study. Resistance to penicillin and erythromycin was common among S. aureus isolates in this study. However, the rate of tetracycline resistance (49%) was lower than reported from the same hospital in 1979 (57%) and 1982 (74%) [ 17 ]. In the late seventies, tetracycline was used in great quantities in Tanzania to prevent and treat cholera; as much as 1788 kilograms of the drug were used during a period of only 5 months [ 31 ]. Due to the rapid emergence of tetracycline-resistant Vibrio cholerae , the use of the drug was subsequently greatly reduced, and this may have contributed to a concurrent decline in the rate of tetracycline-resistance in an unrelated species such as S. aureus . For meaningful comparison of data from different studies, whether from the same or different laboratories, the same method of susceptibility testing should preferably be employed. In our laboratory, the same method has been used for a number of years. The WHONET software features a number of sophisticated ways to analyze susceptibility information based on the measurements of inhibition zone diameters. Recording the diameter of the inhibition zones in disk diffusion testing is generally recommended [ 32 ], and may increase the accuracy of results and enable the detection of gradual shifts in antibiotic susceptibility over time. It also makes the data independent of the current breakpoints. With the WHONET software, data can easily and rapidly be re-analyzed with reference to new breakpoints. However, the Stokes' method for susceptibility testing [ 9 ], which is used in our laboratory, is based on visual interpretation of the difference in inhibition zones between the clinical isolate and the control strain. The interpretation is recorded as interpreted values, i.e. either susceptible "S", intermediate susceptible "I" or resistant "R". The WHONET software also accepts susceptibility data to be entered and analyzed as "interpreted values", i.e. "S", "I" and "R". The use of such interpreted values enables most of the analysis features of WHONET, but not all. Foremost, analyzing data based on zone diameters (or MIC values) is superior for the early detection of subtle shifts in antimicrobial resistance over time, which may alert clinicians about emerging resistance trends at an early stage. However, one asset of the Stokes' method, particularly under tropical conditions, is that unsuspected poor antibiotic disk quality will be discovered quickly since a control strain is tested on every plate. Furthermore, variations over time in the battery of antibiotics tested makes comparison of data less useful. Laboratories in low-income countries are sometimes vulnerable to this because of unreliable supplies of antibiotic discs. Ability to detect emerging resistance traits Disk diffusion testing may give indications of emerging resistance traits such as methicillin-resistance in S. aureus and ESBL in Gram-negative bacteria. The current surveillance indicated that methicillin-resistance is rare in S. aureus at the hospital. Ideally this should be confirmed with PCR-based methods to detect the mec A gene. Likewise, the disk diffusion testing showed the presence of resistance to ceftazidime in Gram-negative isolates, albeit at a low rate, which calls for further investigation with regard to the possible presence of ESBL. Our laboratory did not employ molecular methods for detection of resistance genes on a routine basis, but a recent study showed low prevalence of methicillin-resistant S. aureus (MRSA) [ 18 ]. Resistance surveillance should be coupled with awareness of signs of various resistance traits and, preferably, the possibility of using molecular methods to verify emerging resistance traits. Ability to detect nosocomial problems The WHONET software is well suited to analyze antibiograms in order to detect suspicious nosocomial outbreaks. These functions too are dependant on the use of a consistent battery of test drugs, and also works better when results are entered as actual values for MIC or zone diameters, as opposed to the interpreted value ("S", "I" or "R"). In our hospital, comparison of resistance rates did not show dramatic variation between isolates from inpatients and outpatients. The exception was a trend for more frequent gentamicin-resistance in inpatient isolates of Gram-negative bacteria, particularly Klebsiella spp., which may suggest possible nosocomial spread. The analysis of antibiograms did not produce convincing evidence of clonal patterns spread of bacterial isolates, possibly partly due to the variations in the battery of antibiotics tested. Molecular methods for the evaluation of the genetic relatedness of bacteria were not available in this study. Suitability for international comparison of resistance data In 2002 a total of 880 laboratories in 76 countries across the world used the software, including 41 laboratories in 4 African countries. The WHONET system has been implemented at MNH since 1998. Unfortunately, there is no international consensus on a recommended method for antimicrobial susceptibility testing. Worldwide at least twelve different in vitro methods are followed, and only in Europe the number is at least ten [ 5 ]. Furthermore, there are ongoing changes in the interpretive criteria for susceptibility testing [ 10 ]. In addition to this, there is an abundance of molecular methods to describe various genetic markers of resistance. In vivo clinical assessment is of great importance in understanding bacterial drug resistance and the gold standard for evaluating resistance in malaria parasite. The multitude of methods employed for antimicrobial susceptibility testing has to some extent hampered the meaningful sharing and comparison of resistance data among countries. Recently, much work has been done in Europe to harmonize resistance surveillance efforts across country borders [ 33 , 34 ]. While many laboratories record inhibition zones for disk diffusion results, interpretation is usually according to national guidelines. Thus, susceptibility patterns from different countries must be compared prudently. The lack of standardization in methods is a problem that must be addressed at an international level. The surveillance system as a quality assurance tool The implementation of the surveillance system brought focus on methodological issues, including microbial identification and susceptibility. The WHONET software has built-in functions to alert the operator if isolates with unexpected resistance patterns are entered. During the surveillance exercise in our laboratory, it was discovered that four isolates of Streptococcus pyogenes were reported as resistant to penicillin. This was subsequently double-checked, and consulting the laboratory bench-book we found that clerical errors were the explanation for this. The use of the surveillance software enabled the easy detection, investigation and correction of such errors, and consequently may contribute to increase the attention to quality issues and generally improve the performance of the lab. The current surveillance project highlighted some methodological issues, most of which were caused by budgetary limitations, such as the occasional use of locally made antibiotic disks and limitations in the identification of organisms due to lack of reagents. Impetus for further research Routine surveillance makes use of available large data sets at little additional cost and may be representative for a greater part of the population. However, often it is necessary to supplement the routine surveillance with ad hoc studies aimed at investigating particular problems. While ad hoc studies generally are more expensive to conduct, they allow for the use of more advanced and expensive laboratory methods and are better at targeting particular populations of interest. The current surveillance study identified a need for more data from bloodstream infections in order to provide reliable guidance for the treatment of serious bacterial infections. As a consequence of this, we started a study of bloodstream infections with the pediatric department at the hospital. Another laboratory-based research was started to ascertain the finding that methicillin-resistance in staphylococci is still relatively infrequent at this hospital. Influencing popular opinion on antimicrobial resistance issues Resistance surveillance is a platform from which to promote focus on antimicrobial resistance issues, both within the hospital and the medical community, but also among the general population. In conjunction with the surveillance exercise, we have highlighted issues regarding antimicrobials and resistance in local newspaper letters [ 35 ], and there is ongoing work to establish a chapter of APUA (Alliance for the prudent use of antibiotics, ) in Tanzania. Cost considerations and human resources The study suggests that laboratories, which perform susceptibility testing, can gain useful information on antimicrobial susceptibility with a minimal budget. As appropriate software can be obtained free-of-charge, the main cost of the surveillance system is associated with purchasing a computer. However, there are other, indirect costs, which may be attributed to the surveillance program depending on the situation of the laboratory, such as running costs for microbiologic procedures, including susceptibility testing. Particularly, it is important to ensure availability of antimicrobial discs of satisfactory quality. A susceptibility surveillance system also implies the need for some additional human resources for data entry and analysis. In our experience, it is recommendable to allocate approximately 50% of a laboratory technologist position to this task. While the WHONET program is excellent for entry, analysis and reporting of resistance data, the software is not intended to function as a complete patient management system for the laboratory. Data can be transferred from other databases into WHONET by the use of a complementary software called BacLink (also free-of-charge). However, in laboratories such as ours, where the management of patients' laboratory tests (i.e. receipt of specimens and laboratory forms, inscription in registers, return of test results, etcetera) is handled manually via register-books, the data must be punched into WHONET by hand. Since the WHONET database is not used directly for patient management, the surveillance activity tends to become less integrated in the clinical routine work than it should. Thus, although the program performs its task very well, in a long-term perspective, a surveillance system that is integrated with a patient management system might be more sustainable. It is difficult to quantify the potential benefits of a well-functioning surveillance system. However, we are fully convinced that the modest costs of the surveillance program are highly justified since the data generated may improve empiric therapy, help contain or prevent the further emergence of antimicrobial resistance, decrease the need for expensive second-line antimicrobial drugs and, ultimately, save lives and reduce suffering. Conclusions It is imperative to preserve the effectiveness of common antibiotics by promoting rational use of antibiotics based on sound knowledge of local resistance patterns. In a hospital with bacteriology services, the implementation of a computerized surveillance system is a low-cost tool to make use of available resistance data. In our hospital, the resistance surveillance system has generated information on resistance patterns that is useful as guidance for empiric therapy of infections. It can help alert clinicians of trends of antimicrobial resistance, guide drug-policy decisions and facilitate rational use of drugs to prevent the further emergence of antimicrobial resistance. The surveillance system has also served as a quality assurance tool and led to increased focus on antimicrobial resistance and prudent use of drugs. There is need for more data from blood cultures for reliable guidance for the treatment of severe, systemic bacterial infections. For antibiotic policy recommendations to be applicable for the general population, more information is needed from outpatients and rural areas. There is limited information on antimicrobial resistance trends on the African continent. Only four African countries use the WHONET system for antimicrobial resistance surveillance, although some countries may use other similar software. Recently much work has been done to establish consensus and a more standardized approach to resistance surveillance in Europe [ 34 ]. Susceptibility data based on recorded zone diameters, instead of interpreted values ("S", "I" and "R"), would make the surveillance system more effective in detecting subtle changes in antimicrobial resistance. We believe there is a need for a standardized approach to antimicrobial resistance surveillance also in the African region, as well as globally. This would facilitate liaisons and sharing of information among countries. Competing interests The authors declare that they have no competing interests. Authors' contributions BB was the principal investigator, participated in the planning and execution of the study, performed data entry and data analysis, and was the main responsible author. DSMM, WU, SYM and AD participated in the planning of the study and contributed to the writing process. MM contributed to designing the WHONET database, performed data entry and microbiological work, and contributed to the writing process. SH participated in the writing. NL was the project coordinator and participated in planning, data analysis and writing. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC526372.xml

549212

In vitro bioassay as a predictor of in vivo response

Background There is a substantial discrepancy between in vitro and in vivo experiments. The purpose of the present work was development of a theoretical framework to enable improved prediction of in vivo response from in vitro bioassay results. Results For dose-response curve reaches a plateau in vitro we demonstrated that the in vivo response has only one maximum. For biphasic patterns of biological response in vitro both the bimodal and biphasic in vivo responses might be observed. Conclusion As the main result of this work we have demonstrated that in vivo responses might be predicted from dose-effect curves measured in vitro .

Background In vitro bioassay is very useful in biomedical experiments. It has the potential to yield very important data about molecular mechanism of action of any biologically active compounds. However, the major challenge for such experiments is extrapolation to in vivo responses. Unfortunately, there is a substantial discrepancy between in vitro and in vivo experiments, and there is a paucity of work directed to prediction of in vivo response from in vitro bioassay. So, the purpose of the present work was development of a theoretical framework to enable improved prediction of in vivo response from in vitro bioassay results. Results A survey of literature revealed that most cases of dose-effect curves for in vitro experiments fall into three classes. They are: • monophasic response; • biphasic pattern; • bimodal or polymodal dose-effect curve. MONOPHASIC RESPONSE is the form most commonly reported in articles on in vitro bioassay. In these cases, with increasing dose of biologically active substance (BAS), the cellular response increases to a maximum (dose-response curve reaches a plateau). The most general schemes exhibiting this class of response can be classified as 3 classes: (I) BAS regulation of enzyme activity, (II) Ligand interaction with one type of receptor, and (III) Ligand interaction with negatively cooperative receptors. We will consider these three classes: (I): BAS might regulate enzyme activity. It might be: • substrate: E+S ←→ ES → E+P → cell response ,     (scheme 1) where E is enzyme, S is substrate, ES is enzyme-substrate complex, P is product. Cellular response is suggested to be proportional to product concentration. Scheme (2) approximates the classic Michaelis scheme [ 1 ]. • enzyme activator ( A ) E+S ←→ ES → E+P → cell response E+A ←→ EA (scheme 2) EA+S ←→ EAS → EA+P → cell response increasing , Scheme (3) is characteristic of many BAS. The majority of these groups are vitamins and minerals, which are known to be enzyme cofactors and serve to increase enzyme activity. • enzyme inhibitor ( I ) E+S ←→ ES → E+P → cell response E+I ←→ EI → no cell response ,     (scheme 3) For example, there is the large class of drugs, whose action can be described with the help of scheme (4). This class is called "inhibitors of angiotensin-converting enzyme". These drugs are commonly used for hypertension treatment and prevention [ 2 ]. (II) Ligand interaction with one type of receptors: L+R ←→ LR → cell response (scheme 4) where L is ligand (BAS), R is receptor, LR is ligand-receptor complex. Scheme (4) is "classic" receptor theory as described by Clark (1937) [ 3 ]. For example, kinetic schemes of such type were proved in the case of estrogen regulation of gene expression [ 4 ], apolipoprotein AI, CII, B and E synthesis [ 5 ]. (III) Ligand interaction with negative cooperative receptors L+R ←→ LR L+LR ←→ L 2 R → cell response (5) where L 2 R is complex ligand-receptor complexes. Scheme (5) is characteristic for insulin receptors [ 6 ]. Kinetic equations for schemes (1)–(5) are well known [ 7 ]. They include "classic" Michaelis [ 1 ] and Clark [ 3 ] equations. It can be shown, due to the first order Taylor series, equations for the schemes (1)–(4) can be re-formulated from particle counter theory as: y = B * x /( 1 + A * x )     (6) and for scheme (5): y = B * x 2 /( 1 + A * x 2 )     (7) where x is incoming signal ( x is BAS concentration). For scheme (1) x is substrate concentration, for scheme (2) it is activator concentration, for scheme (3) it is inhibitor concentration, for schemes (4) and (5) it is ligand concentration. y is cellular response for the in vitro system. A and B are scaling coefficients. The BAS concentration in the whole organism changes as a function of time according to equation (14) (see Methods.) i.e. x(t) = C(t) = C 0 [exp(-k el γt)-exp(-k 1 t)] (8) We used equation (8) as the incoming signal, substituted this into equations (6) and (7) and solved analytically using Math Cad 8 graphing software (MathSoft Inc., Cambridge, MA, USA) to predict in vivo responses for monomodal in vitro dose-effect curves for schemes (1)–(5). We used illustrative values from works [ 8 , 9 ] and demonstrated that for such in vitro dose-effect curves, the in vivo response has only one maximum (fig. 1 ). Figure 1 In vivo response for monophasic dose-effect curves measured in vitro . B = 1. a) equation (6), b) equation (7). k el = 0.0714 1/min, k 1 = 0.0277 1/min, C 0 = 1 nM, γ = β. Illustrative values for fig. 1, 2, 4 taken from Veldhuis et al., (1993) [8] and similar to those measured by Baumann et al., (1987) 9 for the clearance of growth hormone (GH). We define β (degree of conjugation) as the proportion of BAS that is free of binding proteins and is available to interact with cognate receptors. The larger is β, the larger the proportion of "free" BAS (see Methods). For equation (6) the value of this maximum is increasing as β increases; for equation (7) this value is maximum for mid-range β values. BIPHASIC PATTERNS OF BIOLOGICAL RESPONSE In this case, in in vitro experiments the low doses of BAS stimulate cellular response, and the high doses inhibit it. So, a maximum is observed on the dose-response curve. The most common kinetic schemes for such response are: • Negative back loop (substrate and product inhibition): a) E+S ←→ ES → E+P → cellular response ES+S ←→ ES 2 (9) b) E+S ←→ ES → E+P → cellular response ES + P ↔ ESP Such schemes are characteristic of glucose metabolism [ 1 ]. • Presence of two receptor types: one type stimulates cellular response, another type inhibits it. L+R ←→ LR → "positive" cellular response L+R' ←→ LR' → "negative" cellular response (10) where R are receptors of the first type, R' are receptors of the second type, LR , LR ' are ligand-receptor complexes with different receptor types. This mechanism has been proven for estrogen regulation of nitric oxide synthase (activity in the rat aorta [ 10 ]; protein pS2 expression in hormone-dependent tumors [ 11 ] and so on. • Desensitization of cellular receptors L+R ←→ LR → positive cellular response LR → decrease in receptor number (11) It has been suggested, that mechanism (11) is basic for drug tolerance [ 7 ]. For example, this mechanism was described for uretal cell stimulation by 17-β-estradiol. Before estradiol treatment, expression of estrogen receptors mRNA in cells was much higher then after 12-days estradiol administration [ 12 ]. It is well known that endogenous opioid receptors become down regulated after chronic exposure to exogenous opioids [ 13 ] and receptor down-regulation has often been observed to follow acute exposure to hormones including growth hormone [ 14 ]. • Change of effector's molecule conformation: "Active" conformation + ligand suplus ←→ "Passive" conformation (12) Scheme (12) was suggested by Bootman and Lipp (1999) [ 15 ] for Ca ++ regulation of 1,4,5-trisphosphate activity. The authors suggested that Ca ++ surplus induces a change in Ca ++ -channel conformation from "open" or "active" to "closed" or "passive" [ 15 ]. For schemes (9)–(12), due to the first order Taylor series, this kinetic equation can be derived: y = A*x*exp(-B*x) (13) Using equation (13), we obtained a prediction of in vivo biphasic dose-effects curves (fig. 2 ). As is apparent from the figure, the magnitude and the analytical appearance of in vivo response is affected by the dose of BAS and its degree of conjugation (β). Both the bimodal and biphasic in vivo responses might be observed for biphasic dose-effect curves. Changes of dose of BAS concentration or its conjugation with blood proteins (or their concentration) might dramatically change the form of in vivo response. For the simulations shown in Figure 2 we used values for k el and k 1 and blood volume (4.9 liters) based on measurements by Baumann et al. (1987) [ 9 ] and Veldhuis et al. (1993) [ 8 ] for growth hormone secretion, clearance and pulsatility. Polymodal biological responses are commonly observed in biological systems. It has been demonstrated, that in some experimental systems, administration of a single, bolus dose of hormone produces a polymodal response [ 16 ]. Figure 2 In vivo response for biphasic dose-effect curves measured in vitro . B = 1. a) variation of β, C 0 = 1 nM, b) variation of C 0 , β = .388. k el = 0.0714 1/min, k 1 = 0.0277 1/min, γ = β. Bimodal dose-effect curves are usually observed for BAS with regulatory activity [ 17 , 18 ]. The mechanism of their formation is still unclear. From our point of view, bimodal dose-response curve might be described by superposition of two biphasic dose-effect curves with different B value. This might be observed in cascade system of signal transduction and amplification. If x regulate intermediate z formation in biphasic way with B 1 , and z has biphasic response on y formation with B 2 , then if B 1 < B 2 , summary dose-effect curve ( y concentration from x ) is bimodal (fig. 3 ). Differences in B 1 and B 2 value define the maximum points. For example, with B 2 increasing, the interpeak distance will also increase. Figure 3 Possible mechanism of bimodal dose-effect curve formation for in vitro systems. a) intermediate z formation as function of x concentration, B 1 = 1, b) final product y formation as function of z concentration, B 2 = 5, c)summary dose-response curve. See comments in the text of the article. For systems, which have bimodal dose-effect curve in vitro , the polymodal response in vivo is observed (fig. 4 ). The form of this response might be change to "seems constant" due to BAS concentration of β value. The differences of maximum values are observed, this differences is time-dependent: the highest maximum is observed with the longest observation. It might be demonstrated, that with change of B 2 value to 20, only bimodal in vivo response will be observed. So, the form and the value of maximums are dependent from the dose of BAS and degree of conjugation. Figure 4 In vivo response for bimodal dose-effect curves measured in vitro . B 1 = 1 , B 2 = 5 . a) variation of β, C 0 = 1 nM, b) variation of C 0 , β = .388. k el = 0.0714 1/min, k 1 = 0.0277 1/min, γ = β. Discussion Analogues of hormones are commonly used in medicine for hormone replacement therapy (for example in post-menopausal women), for oral contraception, as anabolic drugs, for asthma therapy and so on [ 2 ]. But engineered modifications of hormones, growth factors or their analogs are likely to differ from the native analogues in their affinity for binding proteins. In view of this, an important practical consequence of our simulations results are that the testing of newly designed hormones in in vivo systems (with endogenous binding proteins) will require measurements of acute biological response at multiple concentration and time points. For longer-term responses requiring protein synthesis (such as a secretion of body mass or longitudinal bone growth), it could be argued that such multiple time point studies would not be as important. However, in so far as long term biological responses are the consequence of critical initial events which may require threshold concentrations of free hormone, or repeated patterns of hormone exposure over prolonged periods [ 16 , 19 ], this assumption may not be justified. Another application of our work may be the study of hormone functions in glandular tumour disorders. With these disorders, there is usually serious metabolic or hormonal dysfunction. From our point of view, it may be not only due to gland biosynthesis of abnormal hormone. Tumour-produced hormones may not differ structurally from their normal analogues. The dysfunctional occurs due to abnormal concentrations of hormones, which are synthesised by tumours. As it follows from our results, changes in concentrations can dramatically change the form and value of biological response. On the other hand, in many tumour disorders the concentrations of binding proteins are changed. For example, in ovarian carcinoma the changes of sex binding protein and ratio free/bound sex hormones (β) are observed [ 20 ]. As follows from our results, this can dramatically change the biological response to such hormones, i.e. apparent biological functions. So with testing in vitro such hormones seems to be normal (and they may be normal), but in vivo they may have abnormal effects due to changes of their binding protein concentration, or ratio free/bound hormone. Conclusion So, as a result of this work we have demonstrated that in vivo responses might be predicted from dose-effect curves measured in vitro . For monophasic curves, in vivo response is proportional to BAS concentration. For the most complex in vitro curves, the value and the form of in vivo response depends in a predictable way on the dose of BAS and its degree of conjugation. Methods To obtain the discussed results we used linear pharmacokinetics model: where: m 1 (t) mass of biologically active substance (BAS) in the place of infusion, m 2 (t) mass of BAS in compartment (blood), k 1 , k el constants of hormone diffusion from place of infusion to blood and excretion form blood (accordingly). Many of biologically active substances are conjugate into complexes with blood proteins (for example: GH, nerves growth factor, IGF-1): B+P ⇔ K HP (15) where B is BAS, P is blood protein, BP is BAS-protein complex, K is dissociation constant. For many BAS, concentration of free (not bound with blood proteins) BAS is equal to: [B] ≈β [B 0 ] (16) where β is constant ("degree of conjugation"), [ B ] is concentration of free BAS, [ B 0 ] is initial concentration of BAS. If β = 1 then BAS dose not conjugate with protein. If β = 0 then all BAS is in conjugate form. It may be that only conjugate BAS (for example, bilirubin), or only unconjugated BAS can be excreted form the blood (for example, sex hormones). This means that for scheme (14) the law of mass action will be written in the next way: dm 1 /dt = - k 1 m 1 , m 1 (0) = M dm 2 /dt = k 1 m 1 - γ k el m 2 , m 2 (0) = 0 (17) where γ is a constant. γ = 1-β if only conjugate form of BAS can be excreted and γ = β if only unconjugated form is excreted. But γ is a constant with respect to t : γ = const(t) . This means that solution of system (17) is: C(t) = C 0 [exp(-k elγ t)-exp(-k 1 t)] (18) where C(t) is BAS concentration in the blood compartment ( C = m 2 / V , V = const (about 5 liters) is blood volume), C 0 is seems initial BAS concentration ( C 0 = M/V ).

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549212.xml

499543

Structural characterization of genomes by large scale sequence-structure threading: application of reliability analysis in structural genomics

Background We establish that the occurrence of protein folds among genomes can be accurately described with a Weibull function. Systems which exhibit Weibull character can be interpreted with reliability theory commonly used in engineering analysis. For instance, Weibull distributions are widely used in reliability, maintainability and safety work to model time-to-failure of mechanical devices, mechanisms, building constructions and equipment. Results We have found that the Weibull function describes protein fold distribution within and among genomes more accurately than conventional power functions which have been used in a number of structural genomic studies reported to date. It has also been found that the Weibull reliability parameter β for protein fold distributions varies between genomes and may reflect differences in rates of gene duplication in evolutionary history of organisms. Conclusions The results of this work demonstrate that reliability analysis can provide useful insights and testable predictions in the fields of comparative and structural genomics.

Background Recent advances in networks theory have demonstrated a key role of uneven distributions occurring in many natural processes. It has been found that seemingly unrelated systems such as economic, professional, sexual and social networks, airline routing, power lines connections, language networks and internet hyperlinks all exhibit a power law decay of the cumulative distribution P x ≈ x - γ , where x is the number of links connected to each network node and γ is the value of the exponent typically varying in the range of 2–3 [ 1 ]. The heterogeneous architecture of scale-free networks imparts a robustness and error-tolerance from random perturbation and is often viewed as a possible common blueprint for naturally occurring large-scale networks. The critical role of the power law distribution has also been acknowledged in many areas of life sciences: metabolic and other cellular networks, proteins interaction maps, brain cellular organization, food and ecological webs all have been described as scale-free systems. It would be fair to say that the advances in the scale free network studies have revitalized the original Pareto's inequality law introduced more then a century ago [ 2 ]. The applicability of the scale free networks has been examined in numerous structural genomics studies. It has been proposed that the genomic occurrence of protein families, superfamilies and folds can follows an asymptotic power law: SDF(GO) = aGO - b (1) , where SDF(GO) is survival distribution function of genomic occurrence GO of a certain protein family, superfamily and fold. These findings have laid the foundation for characterizing the evolution of the protein universe in terms of a growing scale-free system in which individual genes are represented as nodes of a propagating network [ 3 - 7 ]. In our previous work [ 9 ], we have used the large-scale sequence-structure threading to assign protein folds to 33 genomes from all three superkingdoms of life. It has been found that more then 60% of the studied eukaryotic, 68% of archaeal and 70% of bacterial proteomes could be assigned to defined protein folds by threading. The estimated results have been used to analyze the distribution of protein architectures, topologies and domains (or homologous superfamilies according to the CATH classification [ 8 ]). Thus, we have found that the frequencies of genomic occurrence of assigned protein domains (homologous superfamilies) and topologies can be described by a power function (1) with moderate accuracy. According to the formalism of network theory, such a power law representation of the cumulative distribution of node connections governs a scale-free character of the system [ 10 ]. At the same time we have noted that the values of the power exponent b estimated in the study generally fall below the 2–3 range typical for scale-free systems (analogous observations could also be noted in a number of similar investigations [ 3 - 5 ]). Table 1 (see Additional file 1 ) features the estimated parameters a and b along with the corresponding correlation coefficients r 2 reflecting the goodness of fit of experimental data with the logarithmic linear plots (1) (Table 1 also reflects the total number of the analyzed ORF-s in each genome and the corresponding number of proteins for which the THREADER has confidently assigned certain fold). The established lowered values of the power exponent and modest accuracy of the power law dependences (1) encouraged us to seek alternative approaches to more accurately describe protein folds distributions. Results Weibull (reliability) analysis The Weibull distribution is a general-purpose statistical function defined within Extreme Value Theory [ 11 ] and widely used in reliability engineering to model material strength, durability of electronic and mechanical components or equipments. In the most common case the probability density distribution is described by a two-parameter Weibull distribution , where α is a scaling factor and β is a shape parameter also known as the slope [ 12 ]. The Weibull analysis operates on life data, i.e it utilizes time-to-failure (or time under the testing stress) to assess the reliability of a system and to forecast its stability through parameters of the characteristic life span α and shape β . A typical Weibull experiment is based on application of disruptive stress to multiple samples representative of the population until the tested objects achieve a state of failure and produce time-to-failure numbers. The corresponding time-to-failure values form heterogeneous Weibull distributions described by (2). Application of Weibull function to genomic analysis The distribution of protein folds in a genome can be viewed much like the behavior of a mechanical system under disruptive testing. It is feasible to stipulate that the increase of genomic abundance of any protein fold occurs under evolutionary pressure. Some folds are able to expand their genomic occurrence over a course of evolution others have higher probability to be lost through genetic drift and other random events, i.e. to fail. Considering these analogies, we anticipated that the Weibull logistic can provide some natural explanations for highly heterogeneous abundance of protein folds in genomes. To test this hypothesis we used two independent approaches to examine whether the genomic occurrence of protein topologies and domains can indeed be adequately described by the Weibull function. First of all, we employed the maximum likelihood (ML) method [ 13 ] to fit the survival distribution function SDF(x) of the genomic occurrences of protein topologies and homologous superfamilies into the Weibull dependence (2). The corresponding Weibull shape parameters have been established by solving the equation while the scaling factors have been calculated as . The ML method allowed very accurate description of the distribution of protein folds among the genomes. Figures 1a and 1b feature the survival distributions of CATH topologies and homologous superfamilies among all the studied genomes in combined (these experimental (observed) data curves are marked in red). On the same graphs we have plotted the SDF(GO) parameters reproduced within equation (2) through α and β values estimated by the ML approach. It is obvious that these computed blue curves labeled as 'Weibull analytical' resemble the experimental distributions (marked in red) very precisely. The corresponding α and β values estimated by the ML approach have been collected into Table 2 (see Additional file 2 ). The second way of examining applicability of the Weibull function (2) was based on notion that the double logarithmic transformation of the SDF(x) leads to the equation of a straight line: log(- log( SDF ( x )) = β log( x ) - β log( α )     (3) We performed the transformation (3) on the experimental SDF(GO) data to estimate the Weibull coefficients α and β and squared correlation coefficients r 2 which all have also been collected into Table 2 (marked as 'Weibull by plotting'). The values of the estimated squared correlation coefficients r 2 from Table 2 demonstrate very high accuracy of the linear dependences (3) established from the survival distributions of CATH folds in the studied genomes. These parameters also allow comparing the accuracy of double logarithmic dependences (3) with accuracy of simple logarithmic dependences derived from the power law model (1): log ( SDF ) = a - b * log ( GO )     (4) As it has been mentioned earlier, the dependences (4) have been estimated for the SDF(GO) functions for individual genomes, superkingdoms and for the combined set of proteins. The comparison of r 2 values from Table 2 and Table 1 established from the linear functions (3) and (4) respectively, reveals that for all studies cases (individual genomes, superkingdoms, total dependences) the statistical quality of Weibull dependences (3) is much better than of power law function (4). Figures 1a and 1b feature the Weibull distributions estimated by plotting (double logarithmic transformation) which reproduce the experimental SDF(x) curves with remarkable accuracy. Apparently, the distributions calculated from (3) (labeled as 'Weibull by plotting') are much closer to the experimental distributions than the power law curves (labeled as 'power law') computed within the conventional power function (1). Apparently, that the Weibull distributions established by the double logarithmic representations (4) (marked on Figure 1 'Weibull by plotting') are very close to those calculated by the ML method ('Weibull analytical'). It should be mentioned, however, that despite close resemblance between the Weibull distributions established by the analytical ML method and the 'double logarithmic' approach, the corresponding values of α and β parameters from Table 2 differ (due to the different data fitting algorithms employed by two methods) and the preference should, perhaps, be given to more stringent ML-derived data. Characteristic conditions for the Weibull distribution Although the estimated statistical criteria clearly demonstrate the suitability and superiority of a Weibull function over a power function in describing protein fold distributions, we decided to examine several additional criteria characteristic of the Weibull distribution. As it has been suggested by Romeu [ 14 ] there are four such characteristic properties immanent for the Weibull function. The double logarithmic plot of life data (also called 'a Weibull paper') should be linear As it can be seen from Table 1 the estimated r 2 values from the columns marked as 'Weibull by plotting' are all contained within the range ~ 0.95 – 0.98 what demonstrates that the 'Weibull papers' do indeed describe protein folds distribution in the studied genomes with high accuracy. Figures 2a,2b feature the 'Weibull papers' for the distribution of protein topologies and domains among all the studied species and illustrate that deviations from linearity are very insignificant. The slope of the 'Weibull paper' is an alternative estimator of β The data from Table 2 demonstrate that the estimated slopes of the 'Weibull papers' are very close to the values of β derived by analytical maximum likelihood approach. The x β transformation should yield an exponential distribution with mean α β The genomic occurrences of protein topologies and domains in the genomes and superkingdoms have been transformed into GO β distributions through the power factors β . The exponential character of the resulting distribution has been examined by several statistical tests and in all cases has been confirmed. The observed medians of the exponential distributions GO β accumulated in Table 3 (see Additional file 3 ) demonstrate strong correlations with the calculated α β values. Characteristic life α of the Weibull distribution lies approximately at the 63% of the population The values of the Weibull characteristic life at 63% of distributions have been calculated and collected in Table 3. It is obvious that these parameters closely match values α defined by plotting. Thus, all four specific criteria studied indicate that the genomic occurrence of protein topologies and domains can be characterized as true Weibull distributions. To support this notion further we have also considered another important property of the he Weibull distribution – the dependence of its median ( MDN ) from shape and scale parameters [ 13 ]: To assess the applicability of this condition, we calculated Weibull medians using sets of α and β parameters – estimated by graphical (double logarithmic transformation) and analytical (ML) approaches. The corresponding ' MDN Calctd ' values have been collected into Table 3 along with the observed medians of the Weibull distributions (marked ' MDN Obsrvd '). The estimated high quality linear dependences between the theoretical and observed medians are present on figures 3a and 3b for topologies and domains distributions respectively. The graphics demonstrate that calculated and observed median values are virtually the same what unanimously confirms validity of the condition (5). Thus, multiple independent tests have demonstrated that occurrence of protein folds in genomes obey the Weibull distribution and therefore can be interpreted in terms of the reliability theory what can provide additional insight into folds evolution. Discussion Interpretation of the Weibull parameters The very fact that we were able to assign the Weibull character to the distributions of the CATH protein topologies and homologous superfamilies within genomes ultimately implies that parameters of genomic occurrence can be classified as extreme values. According to the Extreme Values Theory the Weibull distribution will successfully model life systems for which many competing similar failure processes are "racing" to failure and the first to reach it produces the observed failure time [ 15 ]. In regard to genomic occurrence this may suggest that protein folds increase their genomic occurrence in a competitive manner and that those folds having a greater potential to duplicate, will continue to duplicate at the cost of less abundant folds which may ultimately disappear from genome. On another hand, according to reliability theory a Weibull distribution with β > 1 characterizes a life system that increasingly deteriorates. If the shape parameter is smaller then unity ( β <1), there is a reliability growth as the failure rate of the system decreases with time [ 14 ]. It is not clear at the moment, whether a reliability criterion is directly applicable to protein folds distributions. However, β does indeed describe the "skewdness" of the fold distribution, for example Caenorhabditis elegans has the lowest calculated value β among the studied organisms, whilst this genome has also been characterized for its recent expansion and duplication of several gene families [ 16 ]. Presumably, many of these folds are present at lower abundances in other genomes. It could be proposed that such a low β (according to the reliability theory characterizing the genome of C. elegans as the most stable amongst the studied) may reflect the fact that chances of loosing some lower abundant fold families are lower for C. elegans (considering that >70% of the translated ORFs C. elegans genome have been covered by the sequence-structure threading we have assumed that the recently duplicate genes are accordingly represented in the results). In this context, the reliability of a proteome can be viewed as its ability to maintain and expand its composition without loss of protein folds. We can speculate that life systems that enjoy evolutionary success will tend to minimize β <1 i.e. to have more balanced (less heterogeneous) folds representation in their genomes. The fact that most β values presented in Table 2 fall below the unity threshold demonstrates that, in general, the reliability of genome fold composition increases with time, i.e. less protein folds reach the failure state (termination of multiplication and, likely, following evolutionary extinction) as an organism evolves. Interestingly, little difference is observed has been found between the β shape parameters for topologies distributions across the three superkingdoms. All three linear dependences ln(- ln( SDF ( GO ))) ~ ln( GO ) for Bacteria, Eukaryote and Archaea presented on Figures 4a,4b appear very similar. As it has been already mentioned above, it is difficult to decide at this point whether the observed Weibull character of protein folds distribution can be placed in a larger context. We can only speculate that protein folding preferences may lead to a greater abundance of favorable protein configurations and to extinction of those folds which are less favorable. Such selection may represent a mechanism of evolutionary quest for searching for better protein folds. In any case, the observed phenomenon illustrates the act of natural selection in determination of the protein fold repertoires and that the propagation of protein folds in a genome occurs in a competitive manner, i.e. more abundant folds tend to expand their genomic presence even further causing lesser abundant folds to extinct. It also remains to be seen whether some other properties of genomes and proteomes can also be described by the Weibull statistics. In our studies we plan to use the Weibull approach to examine other distributions such as genomic occurrence of transcriptional promoters and regulatory elements, levels of gene expression and occurrence of protein domains per gene, among others. Another possible development for the reliability analysis in structural genomics might be to investigate whether the standard libraries of proteins folds themselves can be adequately described by the Weibull function. As it has been stipulated, in the study we have used the CATH standard library of protein folds, which is one of the most accepted and used protein folds classifications. Ii is not unfeasible, that the representation of protein architectures, topologies, homologous superfamilies, etc in the CATH can be adequately described by the Weibull law. Thus, it has been previously demonstrated that another widely used folds library – SCOP does indeed obey the power low [ 4 ]. Such observations would not necessarily contradict the uneven character of the fold distributions in individual proteomes or superkingdoms as a given protein fold library should reflect the proportion of protein folds occurrence in nature. At the same time, we anticipate that the analysis of the standard fold libraries in terms of the Weibull distributions may bring an additional insight into the field and will be carried out in the near future. To summarize the current work, it is possible to conclude that the use of the Weibull distribution allows more accurate description of protein topologies and domains distributions within and among genomes than power function used in conventional structural genomic studies. In addition, we were able to establish the Extreme Values relationships for protein folds distributions to demonstrate that the protein fold repertoire of an organism most likely occurs as a result of the competition amongst folds. This may reflect a mechanism of natural selection searching for an optimal protein structures when more evolutionary favorable folds tend to populate the entire genomic space and cause the extinction of lesser favorable protein configurations. Conclusions Use of a Weibull function allows describing cumulative distribution of protein topologies and domains within individual genomes and superkingdoms with higher accuracy compared to the conventional power function used in the related studies. The developed approach may be applied to quantification of the distribution of different properties of genomes and can be particularly useful for assessing and comparing fold distributions between different organisms and possible impact of the "reliability" of organisms due to a higher redundancy in their fold composition. In general, the results of investigation demonstrate the feasibility and importance of using the reliability analysis to improve the bioinformatics analysis of proteomes. Methods Assignment of protein folds The prediction of the protein folds has been conducted using the THREADER2 program [ 17 ]. The CATH homologues superfamily representative has been assigned to a given protein sequence if the THREADER2 produced an output above 2.9 for the Z score for the threading energy. After a certain CATH entry has been assigned to a protein sequence, it has also been associated with the corresponding higher level CATH representations: class, architecture and topology. The translated protein sequences for 33 complete genomes downloaded from the NCBI and ENSEMBL databases have been processed in this manner. The threading computations have been paralleled for processing on a Beowulf cluster consisting of 52 dual processor blades (2 × 1 GHz, 1 G RAM). The automated control was implemented by our own PVM-supporting Perl scripts enabling to distribute and query the individual threading processes over multiple computer servers. Survival distribution calculation After the occurrences of distinct classes, architectures, topologies and homologue family representatives have been established within the individual genomes, superkingdoms and in total, the corresponding survival distributions have been computed. First of all, we have established the counts of protein architectures, topologies and domains (homologues families) with a given genomic occurrence GO . At the next step these numbers have been converted into the fractional values. After that the survival distribution functions SDF(x) have been computed for genomes, superfamilies and for the combined set of proteins. The SDF(GO) numbers have been calculated for each integer GO value in the range from 0 to the maximal GO estimated within the set (genome/superfamily/total). Statistical analysis The fitting of the SDF ( GO )~ GO functions has been conducted by the SAS 9.0 statistical package (SAS Inc.). The power law dependences SDF(GO) = aGO - b have been analyzed as a logarithmic transforms Log( SDF(GO) = a - b * GO where the fitting has been conducted for the linear function. The Weibull – like dependences SDF(GO) = exp(-aGO b ) have been fitted using both non-linear approximation (by maximum likelihood method) and by the linear fitting of the double logarithmic transform: log(- log( SDF ( x )) = β log( x ) - β log( α ) The calculation of median valued for the survival distribution has been done by the 'R-project' open source statistical package. Supplementary Material Additional File 1 Parameters of power – law dependences for the survival distribution of genomic occurrences SDF(GO) = a GO b . Click here for file Additional File 2 Parameters α , β and medians (calculated and observed) of Weibull distribution of survival functions of genome occurrences established by maximum likelihood and plotting methods. Click here for file Additional File 3 Statistical parameters for 'Weibull papers' for genomic occurrences of protein topologies and domains. Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC499543.xml

212703

Large-Scale Association Study Confirms Genetic Complexity Underlying Type 2 Diabetes

A leading cause of death and disability, diabetes affects some 16 million Americans and up to 135 million people worldwide. While environmental factors such as diet and lifestyle are known to influence an individual's risk of getting adult-onset, or Type 2, diabetes, there is also a substantial inherited component, though many of the genetic pathways involved remain unidentified. The challenge of defining these genetic pathways lies in the fact that diabetes is what is known as a “complex trait”: not only is it likely that variations in many different genes or some combination of genes contribute to an increased risk, but there are probably different genes associated with diabetes in different populations. Tackling the monumental task of deciphering this genetic puzzle in the largest known study of its kind, Inês Barroso and colleagues confirm the genetic complexity of the disease and clearly demonstrate that untangling the genetics will require even larger studies. In diabetes, defects in both the secretion and function of insulin—which is produced by the pancreas—impair the body's ability to metabolize glucose. Based on what is known about the biology of pancreatic function and diabetes, the researchers chose 71 potential suspect genes that could reasonably be expected to contribute to the disease if they malfunctioned. Some of these genes are involved in the function of pancreatic beta-cells, which secrete insulin; a second group influences the function of insulin and glucose metabolism; and a third plays a broader role in energy metabolism. The results show that none of the genes on their own had a large effect, though a number of gene variations increased risk slightly. They also suggest the existence of variations in several genes that influence the risk of Type 2 diabetes. The dataset will be valuable for future studies of diabetes and supports the view that variation in genes affecting insulin production as well as insulin action can influence the risk of Type 2 diabetes. However, the genetic complexity of the disease—with a number of genes conveying a slightly increased risk— demands additional studies of larger populations to reliably identify the genes involved and the genetic variations that, alone or in combination, increase or lower an individual's risk of developing the disease.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC212703.xml

539246

Tissue engineering, stem cells, cloning, and parthenogenesis: new paradigms for therapy

Patients suffering from diseased and injured organs may be treated with transplanted organs. However, there is a severe shortage of donor organs which is worsening yearly due to the aging population. Scientists in the field of tissue engineering apply the principles of cell transplantation, materials science, and bioengineering to construct biological substitutes that will restore and maintain normal function in diseased and injured tissues. Both therapeutic cloning (nucleus from a donor cell is transferred into an enucleated oocyte), and parthenogenesis (oocyte is activated and stimulated to divide), permit extraction of pluripotent embryonic stem cells, and offer a potentially limitless source of cells for tissue engineering applications. The stem cell field is also advancing rapidly, opening new options for therapy. The present article reviews recent progress in tissue engineering and describes applications of these new technologies that may offer novel therapies for patients with end-stage organ failure.

Introduction The goal of tissue engineering is to repair organ pathologies such as those acquired congenitally or by cancer, trauma, infection, or inflammation. It is based upon the foundations of cell transplantation and materials science. Tissue can be engineered 1) in vivo - by stimulating the body's own regeneration response with the appropriate biomaterial, or 2) ex vivo - cells can be expanded in culture, attached to a scaffold and then reimplanted into the host. Cells may be heterologous (different species), allogeneic (same species, different individual), or autologous (same individual). Autologous cells are preferred because they will not evoke an immunologic response and thus the deleterious side effects of immunosuppressive agents can be avoided. The ideal autologous cells can often be found within the organ itself. These cells (committed precursors) may be isolated, expanded and transplanted back into the same patient, thus representing an autologous transplantation resource. Previously, urothelial cells could be grown in the laboratory setting with only limited expansion. Several protocols were developed over the last 20 years which identified the undifferentiated cells and kept them undifferentiated during their growth phase [ 1 - 4 ]. Using such cell culture methods it is now possible to expand a urothelial strain from a single specimen which initially covered a surface area of 1 cm 2 to one that covers a surface area of >4000 m 2 (an area equivalent to one football field) within 8–14 weeks. These studies indicate the possibility of collecting autologous bladder cells from human patients, expanding them in culture, and returning them to the human donor in sufficient quantities for reconstructive purposes [ 1 , 3 - 11 ]. Major advances have been achieved within the past decade regarding possible expansion of several primary human cell types with specific techniques that employ autologous cells for clinical application. While autologous cells are recognized as the ideal transplantation resource, many patients with end-stage organ disease are unable to yield sufficient cells for expansion and transplantation. Furthermore, some primary autologous human cells cannot be expanded from particular organs ( i.e . pancreas, liver). Stem cells are envisioned as being an alternate source of cells from which the desired tissue can be derived. Human embryonic stem cells (HESC) can be derived from discarded non transferred embryos and have the advantage of being pluripotential (the ability to differentiate into all tissues of the embryo) and able to self-renew indefinitely. However, their clinical application is limited because they represent an allogeneic resource and thus their use would require high dose immunosuppressant therapy. New stem cell technologies such as somatic cell nuclear transfer (therapeutic cloning) and parthenogenesis offer an exciting alternative to create an inexhaustible supply of ESC that can differentiate into all cell types of the embryo, while not being rejected by the patient's immune system. Although many tissues have been created with ESC, they are not used clinically because of an inability to control differentiation. Hence, their ability to form multiple tissue types also becomes their limitation. New genomics and bioinformatics technologies have and will continue to offer new insights into the understanding of ESC growth and differentiation and their application to engineering tissues. In the near future, these new technologies will allow for the generation of an unlimited supply of any cell type in the body. Stem cells The political and ethical controversy surrounding stem cells began in 1998 with the creation of HESC derived from discarded, non-transferred human embryos[ 12 ]. The HESC were isolated from the inner cell mass of a blastocyst (5 days post-fertilization embryo) using an immunosurgical technique whereby the blastocyst was incubated with antibodies specific to trophectoderm. Complement proteins then resulted in lysis of the trophectoderm so that the only surviving cells were the inner cell mass [ 13 ]. Given that some cells can not be expanded ex vivo , ESC can potentially be the ideal resource for tissue engineering because of two fundamental properties, 1) the ability to self-renew indefinitely, and 2) the ability to differentiate into all three germ layers. With the current restrictions surrounding HESC work, many proponents of stem cell research have sought to modify the ban to incorporate the thousands of non-transferred frozen embryos resulting from IVF to be used for the creation of more HESC lines. A SART-RAND study identified approximately 400,000 frozen embryos in storage since the late 1970s [ 14 ]. However, only 2.8% of these have been designated for research. Of the 11,000 embryos designated for research, only 65% of these ( n = 7,334) are expected to survive the freeze/thaw process. From this, 25% are expected to develop to blastocyst stage ( n = 1, 834). If one assumes a 15% efficiency rate for establishment of a HESC line from blastocysts (as suggested by previous studies [ 12 , 15 ]), it may be estimated that approximately 275 HESC could be created from excess frozen embryos. However, the real number of HESC line generated would actually be much lower since not all frozen embryos allocated for research would be used to create HESC lines. Furthermore, even if the maximum possible number of HESC lines could be derived from human frozen embryos, the clinical application of such cells would be limited by the potential rejection from another individual's immune system. New stem cell technologies (such as somatic cell nuclear transfer and parthenogenesis) promise to overcome this limitation. Somatic cell nuclear transfer (therapeutic cloning) Somatic cell nuclear transfer (SCNT) entails the removal of an oocyte nucleus followed by its replacement with a nucleus derived from a somatic cell obtained from that patient. Activation with chemicals or electric shock stimulates cell division up to the blastocyst stage at which time the inner cell mass is isolated and cultured, resulting in ESC. This approach is distinct from reproductive cloning because the blasotcyst is not transplanted back to the uterus. Hence, development does not proceed beyond the 100 cell stage. This process also differs from fertilization since no sperm is used in this process. The resulting ESC are perfectly matched to the patients immune system and no immunosuppressants would therefore be required to prevent rejection. While interest in the field of nuclear cloning remains high since the birth of Dolly (1997), the first successful nuclear transfer was actually reported over fifty years ago by Briggs and King [ 16 ]. Cloned frogs, which were the first vertebrates derived from nuclear transfer, were subsequently reported by Gurdon in 1962 [ 17 ] although the nuclei were derived from non-adult sources. Indeed, in just the past six years alone important advances in nuclear cloning technology have been reported – a pace of discovery that betokens the relative immaturity of this research arena. In fact Dolly was not the first cloned mammal to be produced from adult cells. Live lambs were produced in 1996 using nuclear transfer and differentiated epithelial cells, although these were derived from embryonic discs [ 18 ]. To be sure, the significance of the Dolly report was that this described the first mammal to be derived from an adult somatic cell using nuclear transfer [ 19 ]. Subsequently, animals from several species have been grown using nuclear transfer technology, including cattle [ 20 ], goats [ 21 , 22 ], mice [ 23 ], and pigs [ 24 - 27 ]. A better understanding of the differences between reproductive cloning and therapeutic cloning may help alleviate some of the controversy surrounding these technologies [ 28 , 29 ]. Banned in most countries for human applications, reproductive cloning is used to generate an embryo that has the identical genetic material as its cell source. Such an embryo could then be implanted into the uterus of a female to give rise to a liveborn infant that is a clone of the donor. In contrast, therapeutic cloning is used to generate only ESC lines whose genetic material is identical to that of its source. These autologous stem cells have the potential to become almost any type of cell in the adult body, and thus would be useful in tissue and organ replacement applications [ 30 ]. Therefore, therapeutic cloning (SCNT) may provide an alternative source of transplantable cells. Figure 1 shows the strategy of combining therapeutic cloning with tissue engineering to develop tissues and organs. It has been estimated that approximately 3,000 people die every day in USA of diseases that could have been treated with stem cells-derived tissues [ 31 ]. With current allogeneic tissue transplantation protocols, rejection is a frequent complication because of immunologic incompatibility and thus immunosuppressive drugs are generally required to manage host-versus-graft disease [ 30 ]. The use of transplantable tissue and organs derived from therapeutic cloning could obviate unwanted immune responses typically associated with transplantation of non-autologous tissues [ 32 ]. Figure 1 Strategy for therapeutic cloning and tissue engineering While promising, somatic cell nuclear transfer technology has certain limitations requiring further improvement before it can be applied widely in clinical practice. Currently, the efficiency of the overall cloning process is quite low as the majority of embryos derived from animal cloning do not survive after implantation [ 33 - 35 ]. In practical terms, multiple nuclear transfers must be performed in order to produce one live offspring for animal cloning applications. The potential for cloned embryos to grow into live offspring ranges between <1 and 18% for sheep, pigs, and mice [ 36 ]. However, greater success (~ 80%) has been reported in cattle [ 37 ], a result which may in part be due to availability of advanced laboratory technologies specifically developed for this species for agricultural/breeding purposes. To improve cloning efficiencies, further improvements are required in the multiple complex steps of nuclear transfer such as enucleation and reconstruction, oocyte activation, and synchronization of cell cycle between donor cells and recipient oocytes [ 38 ]. It must be noted that abnormalities have been found in liveborn clones including macrosomia with an enlarged placenta ("large-offspring syndrome") [ 39 ], respiratory distress, defects of the kidney, liver, heart, and brain [ 40 ], obesity [ 41 ], and premature death [ 42 ]. These may be related to epigenetics of cloned cells which involve reversible modifications of DNA, while the original DNA (genetic) sequences remain intact. Faulty epigenetic modulation in clones may result from altered DNA methylation and/or histone modifications causing the overall chromatin structure of somatic nuclei not to be reprogrammed to an embryonic pattern of expression [ 30 ]. Reactivation of key embryonic genes at the blastocyst stage usually does not occur in embryos cloned from somatic cells, while embryos cloned from embryos consistently express early embryonic genes[ 43 , 44 ]. Proper epigenetic reprogramming to an embryonic state may help to improve the cloning efficiency and reduce the incidence of abnormal cloned cells. Novel applications of somatic cell nuclear transfer (therapeutic cloning) We applied principles of both tissue engineering and therapeutic cloning in an effort to produce genetically identical renal tissue in an animal model ( Bos taurus ) [ 45 ]. Bovine skin fibroblasts from adult Holstein steers were obtained by ear notch and single donor cells were isolated and microinjected into the perivitelline space of donor enucleated oocytes (nuclear transfer). The resulting blastocysts were transferred to the uterus of progestin-synchronized recipients permit further in vivo growth. After 12 weeks cloned renal cells were harvested, expanded in vitro , then seeded onto biodegradable scaffolds. The constructs (consisting of cells + scaffolds) were then implanted into the subcutaneous space of the same steer from which the cells were cloned to allow for tissue growth. The kidney is a complex organ with multiple cell types and a complex functional anatomy rendering it one of the most difficult organs to reconstruct [ 46 , 47 ]. Previous efforts in tissue engineering of the kidney have been directed toward development of extracorporeal renal support systems made of biological and synthetic components [ 48 - 54 ]. Although ex vivo renal replacement devices are known to be life-sustaining, there are obvious benefits for patients with end-stage kidney disease if such devices could be implanted long-term without the need for an extracorporeal perfusion circuit or immunosuppressive drugs. Cloned renal cells were seeded on scaffolds consisting of three collagen-coated cylindrical polycarbonate membranes (figure 2 ). The ends of the three membranes of each scaffold were connected to catheters terminating in a collecting reservoir. This created a renal neo-organ with a mechanism for collecting the excreted urinary fluid (figure 3 ). Scaffolds with the collecting devices were transplanted subcutaneously into the same steer from which the genetic material originated and retrieved 12 weeks after implantation. Figure 2 Combining therapeutic cloning and tissue engineering to produce kidney tissue, an illustration of the tissue-engineered renal unit. Figure 3 Renal unit seeded with cloned cells, three months after implantation, showing the accumulation of urinelike fluid. Chemical analysis of the urine-like fluid (for urea nitrogen/creatinine levels, electrolyte levels, specific gravity, and glucose concentration) revealed that the implanted renal cells possessed filtration, reabsorption, and secretory capabilities. Histological examination of the retrieved implants revealed extensive vascularization and self-organization of the cells into glomeruli- and tubule-like structures. A clear continuity between glomeruli, tubules, and the polycarbonate membrane was noted that allowed the passage of urine into the collecting reservoir (figure 4 ). Immunohistochemical analysis with kidney-specific antibodies revealed the presence of renal proteins, and RT-PCR analysis confirmed the transcription of renal specific RNA in the cloned specimens. Western blot analysis confirmed the presence of elevated renal-specific protein levels. Figure 4 Clear unidirectional continuity between the mature glomeruli, their tubules, and the polycarbonate membrane. As previous studies have confirmed bovine clones harbor mitochondrial DNA (mtDNA) of strictly oocyte origin [ 55 - 57 ], the donor egg's mtDNA was thought to be a potential source of immunologic incompatibility. Differences in mtDNA-encoded proteins expressed by cloned cells could stimulate a T-cell response specific for mt-DNA-encoded minor histocompatibility antigens when cloned cells are implanted back into the original nuclear donor [ 58 ]. We used nucleotide sequencing of the mtDNA genomes of the clone and fibroblast nuclear donor to identify potential antigens in the muscle constructs. Only two amino acid substitutions were noted to distinguish cells from the clone and the nuclear donor. Since peptide-binding motifs for bovine MHC class I molecules remain poorly understood, there is no reliable method to predict the impact of these amino acid substitutions on bovine histocompatibility. Oocyte-derived mtDNA was also considered to be a potential source of immunologic incompatibility in cloned renal cells. Maternally transmitted minor histocompatibility antigens in mice have been shown to stimulate both skin allograft rejection in vivo and cytotoxic T lymphocytes expansion in vitro [ 58 ] that could prevent the use of these cloned constructs in patients with chronic rejection of major histocompatibility-matched human renal transplants [ 59 , 60 ]. We tested for a possible T-cell response to the cloned renal devices using delayed-type hypersensitivity testing in vivo and Elispot analysis of interferon-gamma secreting T-cells in vitro . Both analyses revealed that the cloned renal cells showed no evidence of T-cell response, suggesting that rejection will not necessarily occur in the presence of oocyte-derived mtDNA (figure 5 ). This finding may represent a step forward in overcoming the histocompatibility problem of stem cell therapy [ 47 ]. Figure 5 Elispot analyses of the frequencies of T-cells that secrete IFN-gamma after primary and secondary stimulation with allogeneic renal cells, cloned renal cells, or nuclear donor fibroblasts. These studies demonstrated that cells derived from nuclear transfer can be successfully harvested, expanded in culture, and transplanted in vivo with the use of biodegradable scaffolds on which the single suspended cells can organize into tissue structures that are genetically identical to that of the host. These studies were the first demonstration of the use of therapeutic cloning for regeneration of tissues in vivo . Others in the field have created mouse SCNT derived c-kit-positive stem cells to restore infarcted myocardium [ 61 ], dopaminergic neurons to correct the phenotype of a mouse model of Parkinson disease [ 62 ]. The first HESC line derived from SCNT was created in February, 2004 [ 63 ]. Parthenogenesis Parthenogenesis (< Gr . "virgin birth") is production of offspring by a female with no genetic contribution from a male and without meiotic chromosome reduction. The process is common reproductive strategy among insects such as aphids, flies, ants, and honeybees, but is also known to occur in vertebrates including lizards, snakes, fish, birds, and amphibians. The first demonstration of artificially-stimulated parthenogenesis in vitro was made by Jacques Loeb (1899), who was able to activate oocytes from sea urchins and frogs by pricking them with a needle or by changing the ambient salt concentration. Pincus (1939) demonstrated parthenogenetic activation of mammalian eggs using temperature and chemical stimuli. Thus far, parthenogenetic activation of eggs has been studied in a variety of mammals including mice, goats, cows, monkeys, and humans. Plachot et al . described parthenogenesis in humans by examining 800 human oocytes and showed that 12 activated parthenogenetically and four underwent normal cleavage[ 64 ]. Although there have been no reports of naturally-occurring human parthenotes, a human parthenogenetic chimera has been described [ 65 ]. The juvenile patient presented with developmental delay, apparent sex reversal, and entirely parthenogenetic blood leukocytes. This finding confirmed the viability of chimeras in higher mammals as presaged by successful murine experiments over the previous two decades (see below). There is no confirmed example of de novo mammalian parthenogenetic reproduction, but mammalian oocytes can be artificially induced to undergo parthenogenesis in vitro by a two-step protocol involving electroporation and/or treatment with a chemical agent (ionomycin, ethanol, or inositol 1,4,5-triphosphate) to elevate Ca 2+ levels transiently, followed by application of an inhibitor of protein synthesis (cycloheximide) or protein phosphorylation (6-dimethylaminopurine). Success rates and viability appear to be organism dependent. Mouse parthenotes are capable of developing beyond the post-implantation stage in vivo [ 66 , 67 ]; porcine parthenotes have developed up to post-activation day 29 (limb bud stage, past the early heart beating stage); rabbit parthenotes until day 10–11 [ 68 ]; primates ( Callithrix jacchus ) have only been shown to implant [ 69 ]. The reason for this arrested development is believed to be due to genetic imprinting. In normal zygotes maternal and paternal haploid genomes are epigenetically distinct, and both sets are required for successful development [ 70 , 71 ]. Indeed, unstable chromosome modifications in the form of DNA methylation or histone modification are distinctly different in human sperm, compared to eggs. Therefore each gamete carries unique patterns of gene expression into the embryo. Since all genetic material in parthenotes is of maternal origin, there is no paternal imprinting component and this prevents proper development of extraembryonic tissues whose expression is regulated by the male genome [ 72 ]. In most mammals – including primates – oocytes are arrested at metaphase II just before ovulation. Cytogenetic microscopy shows the presence of a 2n polar body under the zona pellucida and a 2n protonucleus in the cytoplasm. After chemical activation to mimic the effects of sperm penetration on changes in cellular Ca 2+ gradient, the cell fails to complete meiosis II. Instead, the second polar body is never extruded, resulting in a diploid protonucleus derived from two sets of sister chromatids. These chromatids then begin to undergo mitosis resulting in a parthenote manifesting uniparental disomy. Although the derivation of embryonic-like stem cells from oocytes (parthenogenetic stem cells, PSC) is relatively inefficient (perhaps due to complexities of genomic imprinting), when they are differentiated into adult tissues, they appear fully functional. In spite of non-viability of monkey parthenotes, the extracted stem cells seem to assume the morphology and functional behavior of HESC and express appropriate ESC markers. They have embryonic-like replicative ability and have been propagated in vitro in an undifferentiated state for up to 14 months. In vitro , they have been differentiated into cardiomyocyte-like cells, smooth muscle, beating ciliated epithelia, adipocytes, several types of epithelial cells, as well as dopaminergic and serotoninergic neurons. Almost all of these neurons express TUJ1 (beta-tubulin III), and up to 25% of the TUJ1+ cells co-express tyrosine-hydroxylase. This latter enzyme marker is considered diagnostic for catecholaminergic neurons (dopamine, norepinephrine, and epinephrine [ 73 ]). Furthermore, HPLC analysis of culture media following a depolarizing KCl-buffer identifies the release of the neurotransmitters dopamine and serotonin from the cells. Ater two weeks of differentiation, about half of the cells demonstrate neuronal morphology and begin to express voltage-dependent sodium channels that can be blocked by tetrodotoxin. These observations are recapitulated in vivo , since injection of monkey PSC into immunocompromised mice induces formation of benign teratomas containing tissue derivatives from all three germ layers (ectoderm, endoderm and mesoderm) including cartilage, muscle, bone, neurons, skin, hair follicles, and intestinal epithelia [ 74 , 75 ]. Of particular note is the apparent tendency of these cells to differentiate into neuronal tissues, as has been noted by chimera studies [ 67 ]. The reasons for this underlying preference are not well understood although one possible explanation is that it is a consequence of purely maternal genomic imprinting, reflecting a lack of epigenetic balance that would be conferred by paternally-imprinted genes. To be sure, parthenotes are not free from ethical controversy and are viewed by some in society as artificial entities that in some sense represent 'tampering with nature.' Since a parthenote is analogous to a mature ovarian teratoma (a spontaneous in vivo tumorigenic event) the de facto acceptance of experiments using teratoma tumor tissue lends some legitimacy to experimentation on parthenotes. These contradictions await reconciliation in a comprehensive ethical framework. Stem cell genomics The pluripotentiality of stem cells is also their limitation, and explains why they are not used clinically today. Although ESC can be differentiated into skin, neurons, blood, cardiac cells, cartilage, endothelial cells, muscle, hepatocytes, and pancreatic cells, the efficiency can be quite limited for certain cell types. Another difficulty is studying the quality of differentiation: are the neurons derived from stem cells bona fide neurons, or merely neuronal-like cells? To address this question we developed high throughput methodologies using microarrays to evaluate new stem cell derivatives [ 76 ]. We differentiated HESC into retinal pigmented epithelial cells (RPE) (the site of lesions in macular degeneration and retinitis pigmentosa) and used microarrays to identify their genetic signature. We then compared their gene signature to those derived from two established RPE cell lines (one of which has been successfully used clinically). A bronchial epithelial cell line served as a negative control and a freshly isolated human RPE served as a positive control. We demonstrated similarity between our HESC derived RPE and the freshly isolated RPE. The bronchial epithelial and two other established RPE lines were less similar. Interestingly, the data set that represented the genes common to freshly isolated RPE and HESC derived RPE (but not in the two established lines), contained many retinal specific genes. This finding provided further support of the benefits of HESC: the ability to generate a limitless number of HESC with the potential to differentiate along specific lineages to allow creation of RPE cells in quantities necessary for clinical transplantation. The next step would be to couple this technology to ESC derived from SCNT (or parthenogenesis) to create the ideal treatment for macular degeneration and retinitis pigmentosa. Another technology currently under development at our institution is "genomics guided tissue engineering." Here we perform microarrays periodically during stem cell differentiation. For example, microarrays are performed on undifferentiated monkey PSC, PSC derived neural precursors (PSC-NP), and NP that were further differentiated for 8 days (PSC-neurons). We have identified numerous targets such as receptors and ligands present at each of these distinct time points, and are modifying our culture system in order to improve the quality and quantity of differentiation. Furthermore, we are comparing the gene expression profiles of PSC derived neurons to gene expression profiles of reference neurons. Not only will this provide new insight into the type of neurons that may be generated, but it offers clues into what our stem cell derived neurons might be lacking. We can then go back to the culture system and try to target these specific genes/signaling pathways. Further study of stem cell genomics will give additional insight into pluripotentiality. An understanding of pluripotentiality might allow for a somatic cell to be de-differentiated into an intermediate stage, which could then be expanded, differentiated and transplanted back into the patient. We are presently characterizing the genetic signature of pluripoteniality by analyzing gene expression among primate stem cells derived from a variety of methods (IVF, parthenogenesis, and adult stem cells). By identifying "stemness" genes by comparing undifferentiated stem cells to their differentiated counterpart, and comparing this to stem cells of different origins, a core set of pluripotential target genes may be mapped. Of particular interest are the 1,075 genes that are similarly down-regulated in IVF derived human ESC and monkey PSC. Furthermore, we have detected paternally imprinted genes in our HESC but not in our PSC data sets. From this we conclude that paternal imprinting might not be necessary for pluriopotentiality. Conclusion Our systems biology approach incorporates the fields of genomics, cell biology, nuclear transfer, and materials science, and utilizes personnel who have mastered the techniques of bioinformatics, cell harvest, culture, expansion, transplantation, as well as polymer design essential for the successful application of these technologies. Experimental efforts are currently underway involving virtually every type of tissue and organ of the human body. Various tissues are at different stages of development with some already being used clinically, a few in pre-clinical trials, and some in the discovery stage. Recent progress suggests that engineered tissues may have an expanded clinical applicability in the future and may represent a viable therapeutic option for those who require tissue replacement or repair.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539246.xml

539252

The lack of public health research output from India

Background Systematic assessment of recent health research output from India, and its relation with the estimated disease burden, is not available. This information would help understand the areas in health research that need improvement in India to enhance the health of India's population. Methods The health research output from India during 2002, which was accessible in the public domain, was assessed by searching PubMed and other internet health literature databases, and was related to the disease burden suggested by the Global Burden of Disease Study. The main outcome measures were number of health papers with abstracts in basic, clinical and public health sciences; quality-adjusted research output based on the impact factors of journals in which the papers were published; classification of papers in disease/condition categories and comparison of research output with the estimated disease burden in each category. Comparison of the health papers from India during 2002 included in PubMed was done with those from Australia during one quarter of 2002. Results Of the 4876 health papers from India in 2002 in PubMed, 48.4%, 47.1% and 4.4% were in basic, clinical and public health sciences, respectively. Of the 4495 papers based on original research, only 3.3% were in public health. Quality-adjusted original research output was highest for non-communicable diseases (62% of total). Of the total quality-adjusted original research output, the proportions in injuries (0.7%), cardiovascular diseases (3.6%), respiratory infections (0.2%), diarrhoeal diseases (1.9%), perinatal conditions (0.4%), childhood cluster diseases (0.5%), unipolar major depression (0%), and HIV/AIDS (1.5%) were substantially lower than their proportional contribution to the disease burden in India. Human resources, health policy, health economics, and impact assessment of interventions were particularly poorly represented in public health research. The Australia-India ratio for quality-adjusted health research output per unit gross domestic product was 20 and for public health research output was 31. Conclusions Good-quality public health research output from India is grossly inadequate, and strategic planning to improve it is necessary if substantial enhancement of population health were to be made possible. There is inordinately low relative research output in several diseases/conditions that cause major disease burden in India.

Background India suffers a large proportion of the disease burden of the world, which has been estimated to be more than its 16.8% share of the world's population [ 1 , 2 ]. One of the vital elements in improving this situation is the need for a comprehensive and relevant evidence base that would equip India to take informed actions. A systematic assessment of recent health research output from India is not available. Without objective information about the current deficiencies and strengths in the health research output from India, it is difficult to plan substantial improvements in health research output that could enhance India's health status. We analysed the health research output from India in 2002 and related it with the estimated disease burden to identify areas that require particular attention to facilitate effective action to reduce disease burden in this world's second most populous country. Methods Health research output was defined as tangible research information related to human health that was readily accessible in the public domain. PubMed [ 3 , 4 ] (which includes MEDLINE) of the US National Library of Medicine, the most widely used online health literature search database in the world, and websites of major academic institutions in India, international agencies, and publishing houses, were searched to ascertain the health research output from India in the year 2002. PubMed was searched for papers published from India in 2002 using "India" in the author affiliation option in PubMed for all journals, and also by searching the Indian journals in PubMed as several papers in these journals mention only city and state but not India in the author affiliation. Only papers with abstracts were included, as the aim was to review all abstracts and classify the papers in various categories, including type of research, type of paper, disease/condition covered, allopathic or traditional system of medicine, and type and location of first author's institution. PubMed gives institutional affiliation and its location only for the first author. Papers that showed the first author affiliation with an Indian institution were considered as research output from India. Definitions were used to classify the Indian papers located in PubMed. Health research was defined as research that could be related to health. Basic research was considered either pure or applied , pure being experimental or theoretical work to advance health knowledge without a defined specific application and applied having such an application. Clinical research was categorised as patient series / management if the paper was about clinical cases or issues in management of patients, laboratory if it dealt mainly with laboratory analysis of patient specimens, clinical trial if it was a trial in the clinical setting, and clinical epidemiology if it was about distribution and determinants of disease assessed in the clinical setting. Public health research was categorised into epidemiology , environment/ social , and health systems / policy . Epidemiology included population epidemiology that dealt with study of distribution and determinants of disease and health in the population, and biostatistics / methods that dealt with methodological issues in epidemiology. Environment/social included environmental sciences that dealt with environmental influences on health, and social aspects that dealt with social dimensions of health. Health system/policy included health services that dealt with aspects of health service provision, and health policy that dealt with concepts and frameworks related to the health system. A paper was classified as original research if it had original data collection and its analysis, and review / viewpoint if it was not based on original data. An attempt was made to classify each paper under the disease/condition that it covered, according to the listing used in the Global Burden of Disease Study [ 2 ]. If a paper covered generic issue(s) which could not be classified under a particular disease/condition, it was considered unclassifiable for disease/condition. The 2002 impact factor of the journal, in which each paper was published, was used as a measure of the quality of each paper [ 5 , 6 ]. The proportion of papers and the quality-adjusted output for the diseases/conditions were related to the proportion of burden caused by each disease/condition in India as estimated for 2000 by the Global Burden of Disease Study [ 2 ]. The publications of 2002 were related to the disease burden of 2000, as research initiation to publication may take on an average a couple of years. Percent quality-adjusted research output was calculated for papers in the categories of several classifications as follows: IndMED [ 7 ], an online database of the Indian Medlars Centre, which covers several Indian biomedical journals was also searched. However, this database could not be included in the study, as the abstracts/papers for all the months of 2002 were not included in this database with substantial portions missing. As reports on commissioned research in public health may be available on the websites of agencies/organisations, the websites of several international agencies (DFID, European Commission, UNAIDS, USAID, WHO, World Bank), twelve academic institutions of India involved with public health, and sixteen publishing houses, were searched to locate research on public health reported in the public domain from India in 2002. For comparison of the Indian health research output with a developed country, a PubMed search was also done for papers published from Australia during the April–June 2002 quarter, using "Australia" OR "names/abbreviations of the states and territories of Australia [ 8 ]" in the author affiliation option in PubMed for all journals as several papers in Australian journals mentioned only city and state but not Australia in the author affiliation. Data were entered in an MS Access database and analysed using SPSS software. Results Of the 5718 papers with abstracts located on PubMed that were published from India in 2002, 842 (14.7%) papers were considered as non-health papers as they were on pure botany, chemistry, physics or zoology that could not be related to human health, and the other 4876 were health papers. The distribution of the types of research and the types of papers for the health papers is shown in Table 1 . The basic and clinical science papers predominated, with public health papers comprising a very small fraction (4.4% of the total). The proportion of papers based on original research was substantially lower for public health (68.5%) than for basic sciences (94.4%) and clinical sciences (92.3%); of the total 4495 original research papers, public health made up only 3.3%. 4700 (96.4%) of the total health papers were on the allopathic system of medicine and 176 (3.6%) on the traditional systems of medicine in which the majority were on ayurveda (144 [81.8%]). Table 1 Distribution of the types of health research and papers from India in 2002 included in PubMed Type of research No. (%)* of papers Type of paper Original research No. [%]† (%)‡ Review / Viewpoint No. [%]† (%)‡ Basic science 2358 (48.4) 2227 [49.6] (94.4) 131 [34.3] (5.6) Pure 525 (10.8) 518 [11.5] (98.7) 7 [1.8] (1.3) Applied 1833 (37.6) 1709 [38.0] (93.2) 124 [32.5] (6.8) Clinical science 2296 (47.1) 2119 [47.2] (92.3) 177 [46.3] (7.7) Patient series / management 1805 (37.0) 1639 [36.5] (90.8) 166 [43.5] (9.2) Laboratory 283 (5.8) 277 [6.2] (97.9) 6 [1.6] (2.1) Clinical trials 155 (3.2) 153 [3.4] (98.7) 2 [0.5] (1.3) Clinical epidemiology 53 (1.1) 50 [1.1] (94.3) 3 [0.8] (5.7) Public health 216 (4.4) 148 [3.3] (68.5) 68 [17.8] (31.5) Epidemiology 85 (1.7) 72 [1.6] (84.7) 13 [3.4] (15.3) Social / environmental 38 (0.8) 31 [0.7] (81.6) 7 [1.8] (18.4) Health systems / policy 93 (1.9) 45 [1.0] (48.4) 48 [12.6] (51.6) Other§ 6 (0.1) 0 [0.0] (0.0) 6 [1.6] (100.0) Total 4876 (100) 4494 [100.0] (92.2) 382 [100.0] (7.8) *Percent of the total 4876 papers † Percent of total in each type of paper ‡ Percent of total in each type of research §Papers that could not be classified in the above categories of type of research; these mostly consisted of biographies of persons or organizations Table 2 shows the distribution of the diseases/conditions covered by the original research papers from India as compared with the estimated disease burden. A large proportion of the basic science papers (49%) were not classifiable into specific disease/condition categories, as they were generic in nature, as compared with 2.9% papers in clinical science and 13% in public health. Overall, the relative proportion of quality-adjusted original research output for non-communicable diseases was higher than their relative contribution to the disease burden, and this was most marked for clinical sciences. However, some major categories/sub-categories within non-communicable diseases were not covered adequately, as a fairly large proportion of research output was on conditions or issues that were not contributing as much to the disease burden. For example, cardiovascular diseases with a disease burden of 11.4% of the total in 2000 had a relatively low quality-adjusted research output of 3.6% of the total. The estimated disease burden due to neuro-psychiatric conditions was 9.6% of the total and the quality adjusted original research output in this category was relatively fair at 8.8%, but the two major sub-categories of unipolar major depression and biopolar disorder that made up 5.2% of the total disease burden had only 0.2% of the total quality-adjusted original research output. A similar mismatch was seen for infectious & parasitic diseases and respiratory infections that had 33.3% of the total quality-adjusted original research output for 33.9% of the total disease burden, but the six major sub-categories under this group contributing 30.1% of the total disease burden had only 11.8% of the total quality-adjusted original research output (Table 2 ). Table 2 Distribution of original research health papers from India as compared with the estimated disease burden Disease / Condition* % DALY loss in 2000* % DALY loss in 2010* No. (%) of original research health papers† % quality-adjusted output for original research health papers‡ No. (%) of original research basic science papers§ % quality-adjusted output for original research basic science papers¶ No. (%) of original research clinical science papers# % quality-adjusted output for original research clinical science papers** No. (%) of original research public health papers†† % quality-adjusted output for original research public health papers‡‡ Communicable, Maternal, Perinatal and Nutritional Conditions 44.2 34.1 950 (28.6) 37.4 397 (34.9) 42.9 484 (23.5) 29.1 69 (53.9) 59.4 Infectious & parasitic diseases 25.9 22.7 762 (22.9) 33.1 358 (31.5) 40.2 362 (17.6) 23.6 42 (32.8) 48.6 Tuberculosis 6.8 7.0 143 (4.3) 7.4 49 (4.3) 7.2 87 (4.2) 5.6 7 (5.5) 29.2 STDs excluding HIV 1.5 1.1 13 (0.4) 0.3 1 (0.1) 0.1 12 (0.6) 0.6 0 (0.0) 0.0 HIV 3.3 6.0 48 (1.4) 1.6 14 (1.2) 1.5 29 (1.4) 1.8 5 (3.9) 1.2 Diarrhoeal diseases 6.7 4.2 34 (1.0) 1.9 17 (1.5) 2.2 16 (0.8) 1.8 1 (0.8) 0.3 Childhood cluster diseases 4.1 2.5 12 (0.4) 0.5 4 (0.4) 0.4 5 (0.2) 0.6 3 (2.3) 0.0 Respiratory infections 8.0 5.0 18 (0.5) 0.2 2 (0.2) 0.1 15 (0.7) 0.4 1 (0.8) 0.0 Lower respiratory infections 7.7 4.9 8 (0.2) 0.1 2 (0.2) 0.1 6 (0.3) 0.2 0 (0.0) 0.0 Maternal conditions 1.4 0.6 84 (2.5) 1.8 17 (1.5) 1.1 60 (2.9) 2.4 7 (5.5) 2.4 Perinatal conditions 6.1 3.9 25 (0.8) 0.4 1 (0.1) 0.1 23 (1.1) 0.8 1 (0.8) 1.0 Nutritional deficiencies 2.9 1.8 45 (1.4) 1.4 8 (0.7) 0.5 22 (1.1) 1.8 15 (11.7) 7.1 Protein energy malnutrition 1.2 0.7 6 (0.2) 0.2 1 (0.1) 0.0 1 (0.0) 0.0 4 (3.1) 3.4 Iron deficiency anaemia 1.5 1.0 10 (0.3) 0.2 1 (0.1) 0.3 6 (0.3) 0.2 3 (2.4) 0.0 Noncommunicable diseases 38.7 47.5 2344 (70.6) 62.0 732 (64.4) 56.5 1555 (75.6) 70.1 57 (44.5) 40.2 Malignant neoplasms 3.8 5.4 370 (11.1) 11.2 118 (10.4) 9.1 251 (12.2) 14.4 1 (0.8) 2.9 Diabetes mellitus 0.8 0.8 129 (3.9) 3.2 64 (5.6) 3.6 57 (2.8) 2.3 8 (6.3) 8.5 Neuro-psychiatric conditions 9.6 11.5 248 (7.5) 8.8 112 (9.9) 10.5 124 (6.0) 6.6 12 (9.4) 11.6 Unipolar major depression 4.0 5.0 0 (0.0) 0.0 0 (0.0) 0.0 0 (0.0) 0.0 0 (0.0) 0.0 Bipolar disorder 1.2 1.4 3 (0.1) 0.2 0 (0.0) 0.0 1 (0.0) 0.1 2 (1.6) 3.5 Sense organ diseases 1.5 2.1 185 (5.6) 4.8 25 (2.2) 2.1 148 (7.2) 7.5 12 (9.4) 8.8 Cataract 1.2 1.7 25 (0.8) 0.9 6 (0.5) 0.3 16 (0.8) 1.3 3 (2.3) 2.5 Cardiovascular diseases 11.4 14.6 203 (6.1) 3.6 38 (3.3) 2.3 159 (7.7) 5.2 6 (4.7) 0.7 Ischaemic heart disease 5.3 7.1 56 (1.7) 0.9 11 (1.0) 0.7 41 (2.0) 1.2 4 (3.1) 0.2 Cerebrovascular disease 2.1 2.7 20 (0.6) 0.3 3 (0.3) 0.2 17 (0.8) 0.4 0 (0.0) 0.0 Respiratory diseases 3.7 4.8 68 (2.0) 1.5 15 (1.3) 1.1 45 (2.2) 1.6 8 (6.3) 3.2 Chronic obstructive pulmonary disease 1.4 2.0 2 (0.1) 0.0 0 (0.0) 0.0 2 (0.1) 0.0 0 (0.0) 0.0 Digestive tract diseases 2.3 2.4 198 (6.0) 5.3 52 (4.6) 3.7 143 (6.9) 7.5 3 (2.3) 1.5 Cirrhosis of liver 1.1 1.2 12 (0.4) 0.6 1 (0.1) 0.6 11 (0.5) 0.6 0 (0.0) 0.0 Congenital anomalies 3.4 3.5 105 (3.2) 1.6 2 (0.2) 0.2 103 (5.0) 3.3 0 (0.0) 0.0 Injuries 17.2 18.4 28 (0.8) 0.7 7 (0.6) 0.6 19 (0.9) 0.8 2 (1.6) 0.4 Unintentional injuries 15.0 15.9 24 (0.7) 0.6 7 (0.6) 0.6 15 (0.7) 0.6 2 (1.6) 0.4 Road traffic injuries 3.5 5.1 2 (0.1) 0.0 0 (0.0) 0.0 2 (0.1) 0.0 0 (0.0) 0.0 Falls 3.5 3.0 0 (0.0) 0.0 0 (0.0) 0.0 0 (0.0) 0.0 0 (0.0) 0.0 Fires 2.1 2.0 4 (0.1) 0.1 2 (0.2) 0.1 2 (0.1) 0.2 0 (0.0) 0.0 Intentional injuries 2.1 2.5 1 (0.0) 0.1 0 (0.0) 0.0 1 (0.0) 0.1 0 (0.0) 0.0 Self-inflicted injuries 1.4 1.7 1 (0.0) 0.1 0 (0.0) 0.0 1 (0.0) 0.1 0 (0.0) 0.0 Total 100 100 3322 (100) 100 1136 (100) 100 2058 (100) 100 128 (100) 100 *According to the Global Burden of Disease Study [2]; only diseases/conditions with disease burden of >1% of the total are listed, plus diabetes mellitus; since all diseases/conditions are not listed, the sum of sub-categories shown may not add up to the total for their categories; DALY is disability-adjusted life year † Denominator for this percent calculation is 3322, which excludes 1172 papers that were not classifiable into specific disease/condition categories ‡ Based on the total impact factor of 3456.262 for the 3322 original health research papers included in this table §Denominator for this percent calculation is 1136, which excludes 1091 papers that were not classifiable into specific disease/condition categories ¶Based on the total impact factor of 1757.491 for the 1136 original basic health research papers included in this table #Denominator for this percent calculation is 2058, which excludes 61 papers that were not classifiable into specific disease/condition categories **Based on the total impact factor of 1557.811 for the 2058 original clinical health research papers included in this table †† Denominator for this percent calculation is 128, which excludes 20 papers that were not classifiable into specific disease/condition categories ‡‡Based on the total impact factor of 140.96 for the 128 original public health research papers included in this table Overall, the diseases/conditions that were substantially underrepresented in the relative proportion of quality-adjusted original research output as compared with their contribution to the disease burden were injuries, cardiovascular disease, respiratory infections, diarrhoeal diseases, perinatal conditions, childhood cluster diseases (including measles and tetanus), unipolar major depression, and HIV/AIDS (Table 2 ). As the research output was least in public health, a brief description follows to understand this deficiency better. Figure 1 shows the diseases/conditions that were estimated to contribute more than 4% of the total disease burden in 2000 or 2010, and for which the original research output in public health was less than one-third of their proportional contribution to the disease burden estimated for 2010, suggesting that these diseases/conditions needed particular attention. Table 3 shows the distribution of original research in the various areas of public health, which suggests that original research in human resources, health policy, and health economics is relatively more deficient within the already low public health research output. Only six of the original public health research papers were on assessing interventions across the various areas, suggesting that the existing public health research in India has not yet evolved to the stage of methodically assessing the impact of public health interventions, which is a necessary step in the evolution of effective public health action. Figure 1 Diseases/conditions poorly represented in original public health research relative to their contribution to the disease burden in India. Table 3 Distribution of the types of original public health research from India Type of public health research No. (%) of original research public health papers Total impact factor of original research public health papers in each type % quality-adjusted output for all original research public health papers† Epidemiology 72 (48.6) 59.6 38.9 Population epidemiology 69 (46.6) 56.5 36.9 Biostatistics / Methods 3 (2.0) 3.0 2.0 Environment / Social 31 (20.9) 29.3 19.1 Environmental sciences 14 (9.5) 16.7 10.9 Social aspects 17 (11.5) 12.7 8.3 Health Systems / Policy 45 (30.4) 64.4 42.0 Health services 42 (28.4) 63.3 41.3 Health economics* 8 (5.4) 5.7 3.7 Training / human resources* 5 (3.4) 0.6 0.4 Health policy 3 (2.0) 1.1 0.7 Total 148 (100) 153.3 100 *Health economics and training / human resources are sub-categories of health services † Based on the denominator of 153.3 Of the total 4876 health papers from India in PubMed for 2002, 1300 (26.7%) were published in Indian journals, but these papers accounted for only 1.5% of the total impact factor of all health papers from India due to the very low impact factors of Indian journals. Among the public health papers 44.4% were published in Indian journals, for clinical sciences papers this was 39.7%, whereas this proportion was much smaller for basic sciences (12.4%). The highest proportion of quality-adjusted basic research output was by university departments, institutions affiliated with the Council of Scientific and Industrial Research, and technical institutions; the predominant proportion of clinical research was by medical colleges / hospitals; and public health research by medical colleges / hospitals, government departments (due to one paper in a very high impact factor journal), and institutions affiliated with the Indian Council of Medical Research (Table 4 ). The National Capital Territory of Delhi accounted for the highest health research output among all states / union territories or cities (Table 5 ). The top ten research producing cities, with 6% of the population of India, produced 75.6% of the quality-adjusted research output, suggesting a concentration of quality research activity in parts of the country. Table 4 Distribution of health research output from various types of institutions in India Type of institution All health Basic science Clinical science Public health No. (%) of health papers % quality-adjusted output for health papers No. (%) of basic science papers % quality-adjusted output for basic science papers No. (%) of clinical science papers % quality-adjusted output for clinical science papers No. (%) of public health papers % quality-adjusted output for public health papers Medical college / Hospital 2571 (52.7) 33.4 407 (17.3) 11.9 2044 (89.0) 82.1 119 (55.1) 45.9 Indian Council of Medical Research* 159 (3.3) 4.0 52 (2.2) 2.3 78 (3.4) 6.7 28 (13.0) 13.1 Council of Scientific and Industrial Research † 387 (7.9) 13.3 361 (15.3) 18.9 21 (0.9) 1.7 5 (2.3) 0.6 Technical institutions ‡ 281 (5.8) 12.2 268 (11.4) 17.7 7 (0.3) 0.6 5 (2.3) 2.3 Paramedical college/institution 164 (3.4) 2.9 154 (6.5) 4.1 8 (0.3) 0.4 2 (0.9) 0.9 University department 813 (16.7) 16.4 743 (31.5) 23.0 53 (2.3) 2.5 16 (7.4) 4.7 NGO / Foundation / Society 69 (1.4) 1.2 12 (0.5) 0.6 33 (1.4) 1.6 22 (10.2) 8.0 Government department 4 (0.1) 0.6 0 (0.0) 0.0 0 (0.0) 0.0 4 (1.9) 17.9 § Industry 31 (0.6) 0.6 28 (1.2) 0.9 3 (0.1) 0.1 0 (0.0) 0.0 Other 397 (8.1) 15.4 333 (14.1) 20.6 49 (2.1) 4.3 15 (5.9) 6.7 Total 4876 (100) 100 2358 (100) 100 2296 (100) 100 216 (100) 100 *Institutions affiliated with the Indian Council of Medical Research [15] † Institutions affiliated with the Council of Scientific and Industrial Research [16] ‡ Indian Institutes of Technology, Indian Institute of Science, and other technical institutions § Percentage high due to one paper in a very high impact factor journal The total of basic, clinical and public health papers does not add up to the "all health" papers in all rows, as 6 "other" papers that could not be classified as basic, clinical or public health (Table 1) are not included in this table Table 5 Distribution of health research output from states and cities in India State / Union Territory* Population in millions† No. (%)‡ of health papers No. of health papers per million population Total impact factor of health papers % quality-adjusted health research output§ Total impact factor per million population National Capital Territory of Delhi 13.8 1014 (20.8) 73.5 1216.0 20.8 88.1 Karnataka 52.7 491 (10.1) 9.3 786.8 13.5 14.9 Maharashtra 96.7 573 (11.8) 5.9 710.1 12.2 7.3 Uttar Pradesh 166.0 484 (9.9) 2.9 591.0 10.1 3.6 West Bengal 80.0 362 (7.4) 4.5 510.6 8.7 6.4 Tamil Nadu 62.1 476 (9.8) 7.7 476.7 8.2 7.7 Andhra Pradesh 75.7 299 (6.1) 3.9 461.1 7.9 6.1 Union Territory of Chandigarh 0.9 364 (7.5) 404.4¶ 336.2 5.8 373.6¶ Kerala 31.8 183 (3.8) 5.8 177.5 3.0 5.6 Punjab 24.3 105 (2.2) 4.3 123.4 2.1 5.1 Gujarat 50.6 74 (1.5) 1.5 68.4 1.2 1.4 Madhya Pradesh 60.4 71 (1.5) 1.2 61.9 1.1 1.0 Union Territory of Pondicherry 1.0 66 (1.4) 66.0 53.7 0.9 53.7 Haryana 21.1 95 (1.9) 4.5 44.6 0.8 2.1 Orissa 36.7 41 (0.8) 1.1 43.2 0.7 1.2 Rajasthan 56.5 62 (1.3) 1.1 40.8 0.7 0.7 Jammu And Kashmir 10.1 20 (0.4) 2.0 28.6 0.5 2.8 Assam 26.6 18 (0.4) 0.7 26.0 0.4 1.0 Uttaranchal 8.5 21 (0.4) 2.5 25.7 0.4 3.0 Meghalaya 2.3 9 (0.2) 3.9 13.7 0.2 6.0 Himachal Pradesh 6.1 12 (0.2) 2.0 13.1 0.2 2.1 Andaman & Nicobar Islands 0.4 6 (0.1) 15.0 9.9 0.2 24.8 Goa 1.3 6 (0.1) 4.6 8.6 0.1 6.6 Bihar 82.9 5 (0.1) 0.1 5.5 0.1 0.1 Jharkhand 26.9 5 (0.1) 0.2 3.4 0.1 0.1 Sikkim 0.5 3 (0.1) 6.0 1.6 0.0 3.2 Manipur 2.4 4 (0.1) 1.7 1.5 0.0 0.6 Chhattisgarh 20.8 3 (0.1) 0.1 1.3 0.0 0.1 Arunachal Pradesh 1.1 2 (0.0) 1.8 0.9 0.0 0.8 Tripura 3.2 2 (0.0) 0.6 0.3 0.0 0.1 Top fifteen cities (State / Union Territory)* Delhi (National Capital Territory of Delhi) 13.8 1014 (20.8) 73.5 1216.0 20.8 88.1 Bangalore (Karnataka) 8.4 258 (5.3) 30.7 598.2 10.2 71.2 Mumbai (Maharashtra) 11.9 393 (8.1) 33.0 499.4 8.5 42.0 Kolkata (West Bengal) 4.6 299 (6.1) 65.0 463.6 7.9 100.8 Hyderabad (Andhra Pradesh) 3.7 233 (4.8) 63.0 404.2 6.9 109.2 Chandigarh (Union Territory of Chandigarh) 0.9 364 (7.5) 404.4¶ 336.2 5.8 373.6¶ Lucknow (Uttar Pradesh) 3.7 272 (5.6) 73.5 332.1 5.7 89.8 Chennai (Tamil Nadu) 4.2 246 (5.0) 58.6 267.6 4.6 63.7 Pune (Maharashtra) 7.2 108 (2.2) 15.0 163.0 2.8 22.6 Varanasi (Uttar Pradesh) 3.1 87 (1.8) 28.1 138.8 2.4 44.8 Thiruvananthapuram (Kerala) 3.2 121 (2.5) 37.8 135.7 2.3 42.4 Mysore (Karnataka) 2.6 74 (1.5) 28.5 96.2 1.6 37.0 Vellore (Tamil Nadu) 3.5 84 (1.7) 24.0 95.6 1.6 27.3 Pondicherry (Union Territory of Pondicherry) 0.7 66 (1.4) 94.3 53.7 0.9 76.7 Visakhapatnam (Andhra Pradesh) 2.2 34 (0.7) 15.5 38.5 0.7 17.5 * Listed in descending order of total impact factor of health papers; the states / union territories of Mizoram, Nagaland, Dadra & Nagar Haveli, Daman & Diu, and Lakshadweep had no publications in PubMed in 2002 † Population for 2001 from the Census of India [14] ‡ Percent of the total 4876 health research papers from India in 2002 § Percent of the total impact factor of 5842.055 for all 4876 health research papers from India ¶This high per capita output is likely related to the small population of Chandigarh and the high concentration of academic institutions Search of websites of major academic institutions in India, international agencies, and publishing houses revealed that substantial original public health research output that was accessible in the public domain was not readily available from these sources. Among the major academic institutions in India involved with public health research, only one was found to have a few reports on health research accessible on its website [ 9 ] and another had some health research abstracts on its website [ 10 ]. The international agencies had some reports on their websites on India-related health research that were mostly authored by non-Indian authors. In the April-June quarter of 2002, 1905 health papers published from Australia were located on PubMed, of which 722 (37.9%) were in basic sciences, 954 (50.1%) in clinical sciences, and 229 (12%) in public health. Taking into account the population and total gross domestic product (GDP) adjusted for purchasing power parity (PPP) of Australia and India [ 1 ], the quality-adjusted health research output and public health research output were 19.6 and 31 times higher from Australia than India, respectively, per unit GDP adjusted for PPP (Table 6 ). Table 6 Comparison of health research output from India and Australia in 2002 India Australia Australia-India ratio Total Per million population* Per billion GDP-PPP† Total‡ Per million population* Per billion GDP-PPP† Per million population Per billion GDP-PPP No. of health papers 4876 4.72 1.66 7620 392.78 15.49 83.2 9.3 Impact factor for health papers 5842 5.65 1.99 19231 991.27 39.10 175.3 19.6 No. of basic science papers 2358 2.28 0.80 2888 148.87 5.87 65.2 7.3 Impact factor for basic science papers 3944 3.82 1.35 10598 546.31 21.55 143.1 16.0 No. of clinical science papers 2296 2.22 0.78 3816 196.70 7.76 88.5 9.9 Impact factor for clinical papers 1698 1.64 0.58 7624 393.01 15.50 239.1 26.7 No. of public health papers 216 0.21 0.07 916 47.22 1.86 225.9 25.3 Impact factor for public health papers 193 0.19 0.07 1008 51.95 2.05 277.6 31.0 *Based on 1033.4 million population for India and 19.4 million for Australia in 2001 [1] † Based on the gross domestic product adjusted for purchasing power parity (GDP-PPP) of US$ 2930 billion for India and US$ 491.8 billion for Australia in 2001 [1] ‡ Based on multiplying the number of papers and their total impact factor for the April-June 2002 quarter by four to obtain the estimate for the year 2002 The total of the number and impact factor for basic, clinical and public health papers does not add up to that for the health papers, as 6 "other" papers (with total impact factor 6.012) that could not be classified as basic, clinical or public health (Table 1) are not included in this table Discussion The data presented in this paper suggest that the health research output from India is not commensurate with the magnitude and distribution of disease burden. The research output in public health is particularly meagre, which is a major concern as public health sciences are a necessary tool to facilitate improvement in population health. Within this low research output, several diseases/conditions contributing substantially to the disease burden and several major areas of public health importance have relatively less representation. Without dynamic, relevant, good quality and adequate original research in the various aspects of public health it is difficult to imagine how the sub-optimal health status of the Indian population would improve on rhetoric or theoretical concepts alone [ 14 , 15 ]. In this paper we used impact factors for journals as a measure of the quality of papers published in those journals. Although impact factors are not without their limitations, they still offer a tangible, and perhaps the best available, option to compare the quality of publications in journals [ 6 ]. We explored several sources where information about health research output from India could be available in the public domain, as the utilisation of research findings is facilitated most if they are readily accessible in the public domain. However, we did not find any source that would add substantially to the information available in the PubMed database. Indeed, there are more Indian health journals than are included in PubMed, but their quality in general is not as high as those included in PubMed with none of them having an impact factor above zero. Non-inclusion in our analysis of the papers published in these journals, therefore, did not bias our assessment of quality-adjusted research output based on impact factors. The relative low quality and impact factor of a large proportion of Indian journals has been discussed previously [ 16 , 17 ]. PubMed lists affiliation of the first author only, and therefore, the analysis presented in this paper includes only those publications in which the first author had Indian affiliation. There would be other publications with non-Indians as first author and Indians as co-author(s), which we estimate to be a very small fraction of those with Indians as first author. In the general context, the PubMed/MEDLINE database has been used previously to assess the health research output from several countries [ 18 - 25 ]. We used the disease burden in India as estimated by the Global Burden of Disease Study [ 2 ]. Although the limitations of this Study have been debated previously in the literature, we could not find a better alternative for use for our study, as these were the most comprehensive estimates available for India. In any case, these estimates can be taken only as indicative, and therefore, we highlight only gross deviations of health research output from these trends. There has been a previous attempt to assess the health research output from India using the Science Citation Index of 1981–85 and relating the number of papers published in journals of various medical/health specialities with the perceived areas of major disease burden [ 26 ]. However, review of all published abstracts to classify each paper in various categories, the approach used by us, has not been used previously to assess health research output from India to our knowledge. Systematic tracking of health research output, and its relation to the estimated trends in disease burden, are necessary for guiding further appropriate development of health research in India. In addition to the overview of research needs identified in this paper, more in-depth assessment of research needs for major diseases/conditions would also be necessary, as was reported recently for the evidence base needed to control HIV/AIDS in India [ 27 ]. Since public health sciences seem to be the weakest link in improving health in India currently, it is imperative that a strategic framework for developing original public health research in India be evolved. To do so, the demand , supply and environment issues would have to be addressed: • Demand. Among the multitude of factors that influence the demand for relevant public health research, the role of policy makers and senior health academics is of particular importance. This is seriously sub-optimal in India at present. Political compulsions push many policy makers into short-term gains instead of investments in comprehensive research for long-term benefits. Although there has recently been an increasing trend in India towards commissioned research by government and international agencies in some aspects of public health, this by itself is not enough to boost comprehensive public health research in India, and the reports of such studies are many times not available in the public domain which reduces the chance of their widespread utilisation. Many senior health academics in India continue to disregard public health research as a less-respectful cousin of basic and clinical research. Systematic efforts are needed to demonstrate to these groups the linkages between all aspects of health research (basic, clinical and public health), and the linkages between public health research and improvements in population health, in order to boost the demand for relevant and good-quality public health research in India. • Supply. Enhancing the output of public health research will require effort on various fronts. Establishing schools of public health and other institutions to train quality scientists in public health is a priority, as India has a surprisingly few number of institutions that can provide proper training in public health research. Another area that needs quick attention is to make public health exposure in medical and paramedical colleges more practical to encourage hands-on investigative thinking, as currently it is so theoretical that it rarely inspires enthusiasm in young professionals towards public health research. Setting higher standards for the research dissertations currently required for post-graduate degrees in preventive and social medicine would also encourage better quality and practically relevant public health research. It is also necessary to systematically develop performance-based opportunities to public health research scholars for career enhancement. Another element that would help develop public health research capacity in India is evolving mechanisms to encourage contribution to this effort by the many Indian public health researchers living abroad. • Environment. A conducive environment is necessary for the demand and supply of public health research to function optimally. Efforts are needed to develop this by attempting to develop broad-based coalitions, that include health care providers, civil society and non-governmental sector, for-profit private sector and industry, and national and international agencies providing financial support, which would understand and support the need for vibrant public health research as a vital element of societal development. This is a necessary element that has so far received scant attention, which must be addressed if sustainable development of public health research to improve population health is to become possible in India. An environment of good-quality and comprehensive public health research in India would also infuse the much-needed originality in teaching public health sciences and their practical application to the local context. Evolving such frameworks would require building up a critical momentum for this effort through perseverance and wisdom. One such opportunity is provided by the recent initiative of the Indian Ministry of Health and Family Welfare to develop more effective institutes of public health in India, with relevant public health research and its utilisation an important key to improving population health [ 28 ]. The recent attention towards revitalising the academic aspects of health care / medicine through evidence [ 29 ] and evidence-based global health [ 30 ] is particularly relevant for developing nations. Evolving a strong, dynamic and locally-relevant evidence base is even more important for developing nations as this is likely to yield relatively higher returns by contributing to improvements in the health, lives and economy of a larger proportion of the world's population. For this to happen, theoretical concepts alone would obviously be not enough. The practical solutions for this effort would have to be developed wisely. The data and its interpretation presented in this paper are, we hope, an example of how the deficiencies in the evidence-base needed for adequate health care in developing nations can be understood objectively in order to plan its strengthening. Conclusions • Publications from India in PubMed were 11 times less in public health than those in basic sciences and in clinical sciences in 2002. • Injuries, cardiovascular diseases, respiratory infections, diarrhoeal diseases, perinatal conditions, childhood cluster diseases, unipolar major depression, and HIV/AIDS had substantially less proportion of quality-adjusted original research output in India as compared with their contribution to the disease burden. • India produced 20 times less quality-adjusted health research output than Australia per unit gross domestic product adjusted for purchasing power parity, and this ratio for public health research output was even higher at 31 times. • Good-quality public health research output from India is grossly inadequate, and strategic planning to improve it is necessary if substantial enhancement of population health were to be made possible. Competing interests The author(s) declare that they have no competing interests. Authors' contributions LD conceived the idea of this study, guided the design, data collection and analysis, and wrote the initial draft of this paper. YSS contributed to the design, data collection and analysis. MNJ contributed to data collection and analysis. VSUB contributed to data management and analysis. RD contributed to the idea of this study, design and data analysis. All authors read and approved the final manuscript. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539252.xml

545602

Genome-wide mutagenesis of Zea mays L. using RescueMu transposons

The authors describe a large-scale transposon-tagging study in maize using the RescueMu system. The study provides a large resource of tagged and sequenced maize alleles, as well as some insights into the biology of RescueMu .

Background MuDR/Mu transposable elements are widely used for mutagenesis and as tags for gene cloning in maize [ 1 , 2 ]. The high efficiency of Mu insertional mutagenesis regulated by MuDR in highly active Mutator lines reflects four features of this transposon family. First, a plant typically has 10-50 copies of the mobile Mu elements [ 3 ], although some plants have over 100 copies. Second, they insert late in the maize life cycle, generating diverse mutant alleles transmitted in the gametes of an individual Mutator plant [ 1 ]. Third, they exhibit a high preference for insertion into genes [ 1 ]. And fourth, most maize genes are targets as judged by the facile recovery of Mu insertion alleles in targeted screens [ 1 , 4 - 6 ]. In directed tagging experiments, the frequency of Mu -induced mutations for a chosen target gene is 10 -3 -10 -5 [ 7 ]. Interestingly, a bronze1 exon [ 8 ] and the 5' untranslated region of glossy8 [ 9 ] contain hotspots for Mu insertion in specific regions, which may explain the higher frequency of mutable allele recovery for these genes. Somatic mutability, visualized as revertant sectors on a mutant background, is indicative of transposon mobility. By monitoring maintenance of a mutable phenotype, it was established that the Mutator transposon system is subject to abrupt epigenetic silencing, which affects some individuals in most families [ 10 , 11 ]. A molecular hallmark of silencing is that both the non-autonomous Mu elements and the regulatory MuDR element become hypermethylated [ 12 , 13 ]. Without selection for somatic instability of a visible reporter allele and/or hypo-methylation, Mutator lines inevitably lose Mu element mobility. The high efficiency of Mu mutagenesis has been exploited in several reverse genetics strategies. The first protocol described used PCR to screen plant DNA samples to find Mu insertions into specific genes using one primer reading out from the conserved Mu terminal inverted repeats (TIRs) and a gene-specific primer [ 14 - 17 ]. Alternatively, survey sequencing of maize genomic DNA flanking Mu insertions yields a list of tagged genes in each plant [ 18 , 19 ]. A third method uses RescueMu , a Mu1 element containing a pBluescript plasmid, to conduct plasmid rescue by transformation of Escherichia coli with total maize DNA samples. To identify insertions in genes of interest, RescueMu plasmids can be screened or the contiguous host genomic DNA can be sequenced using primers permitting selective sequencing from the right or left TIRs of Mu1 [ 20 ]. Here we describe the initial results of a large scale RescueMu tagging effort conducted by the Maize Gene Discovery Project. The tagging strategy employed grids of up to 2,304 plants organized into 48 rows and 48 columns. Plasmid rescue was undertaken from individual pools of up to 48 plants per row or column. Genomic sequences next to RescueMu insertion sites were obtained for all the rows and for a subset of columns of six grids. Maize genomic sequences were subsequently assembled into 14,887 unique genomic loci using computational approaches. These loci were analyzed for gene content, the presence of repetitive DNA and correspondence to mapped maize genes and ESTs. Gene models were built by co-assembling the genomic sequence with ESTs and cDNAs by spliced alignment and by ab initio gene prediction. Identified gene models were tentatively classified using gene ontology terms of potential homologs [ 21 ]. Many features of Mu element behavior have been examined previously using hundreds of tagged alleles or by analyzing the population of Mu elements in particular plants and a few descendants. With single founder individuals for the analyzed tagging grids, we could examine the distribution of new insertion sites of RescueMu in large progeny sets. The contiguous genomic sequences were analyzed to determine if there were insertion hotspots, preferential insertion site motifs, routine generation of the expected 9-base-pair (bp) direct target sequence duplication (TSD) and evidence of pre-meiotic insertion events. Like other Mu elements, RescueMu exhibits a strong bias for insertion into or near genes, as few insertions were recovered in retrotransposons or other repetitive DNA. In addition, for the set of RescueMu insertions into confirmed genes, a bias for insertions into exons (rather than introns) was observed, consistent with the well-established use of Mutator as a mutagen. The gene-enrichment exhibited by RescueMu was compared against two physical methods of gene enrichment, methyl filtration [ 22 ] and high C 0 t genome fractionation [ 23 ]. Results RescueMu transposition in active Mutator lines In standard Mutator lines, Mu1 elements maintain copy number through successive outcrosses, indicating that some type of duplicative transposition occurs [ 24 ] in the absence of genetic reversion [ 25 ]. Most new mutations are independent and occur late in the life cycle [ 26 , 27 ]. Consequently, a single pollen donor can be used to generate thousands of progeny with diverse Mu insertion events (Figure 1 ). Initially RescueMu germinal insertions were sought by direct mobilization of elements from transgene arrays containing multiple copies of the original 35S:RescueMu:Lc plasmid and the plasmid conferring resistance to the herbicide Basta used for selection of transformed callus [ 20 ]. Using eight different transgene arrays crossed with diverse active Mutator lines, the average germinal transposition frequency through pollen was only 0.07 (Table 1 , grid A); lines with a single MuDR element had no transposed RescueMu ( trRescueMu ). Materials were selected from the progeny of grid A plants for grids B through E using two criteria: there were visible seedling mutations in around 10% of progeny characteristic of a very active Mutator line [ 26 ] and the presence of trRescueMu . By DNA blot hybridization of individuals within grids B through E, the RescueMu transposition frequencies ranged from 0.1 to 0.26 (Table 1 ). By sequence analysis after plasmid rescue, trRescueMu were identified that had inserted into likely maize genes and generated the diagnostic 9-bp TSD characteristic of Mu transposition (data not shown). There were also events initially scored as transposition by blot hybridization that represented RescueMu rearrangements within the transgene array, and deleted forms of RescueMu were detected by blot hybridization and gel electrophoretic sizing of rescued plasmids (data not shown). Although RescueMu insertion frequency was low, overall Mu movement was very high in these grids; visible, independent seedling mutations were identified in 10.1-28.3% of the selfed progeny (Table 1 ), as high as the most active Mutator lines described to date [ 28 ]. In an effort to increase transposition frequency, lines with trRescueMu but no transgene array were selected. Plants with a verified trRescueMu were crossed to r-g and colorless kernels selected - these lack red spotting from RescueMu somatic excision from the 35S:RescueMu:Lc transgene. During subsequent plant growth Basta-sensitivity was scored as a second indicator that the transgene array was absent [ 20 ] and DNA blot hybridization then confirmed that a trRescueMu but not the Basta-resistance transgene was present in the plant. To guard against Mutator silencing, plants were also screened by DNA blot hybridization to verify that they contained unmethylated Mu1 and MuDR elements after digestion of genomic DNA with the methylation-sensitive enzymes Hin fI and Sst I, respectively (data not shown). Four plants each with a single trRescueMu were identified by these criteria and crossed to r-g . A DNA blot hybridization screen was conducted on 393 progeny of these four individuals. Seven progeny were identified with two new trRescueMu , seven plants were identified with three events, and 33 plants had a single trRescueMu ; the original, parental trRescueMu elements were shown to segregate as Mendelian factors in the populations screened (data not shown). The 14 plants with two or three new trRescueMu were each crossed by an anthocyanin tester and also crossed multiple times as pollen parents to tester lines to generate sufficient progeny to construct one grid from each founder plant. Inexplicably, in sampling seedling progeny from each outcross ear, some lineages had very few new trRescueMu . The lines with the highest transposition frequencies had two trRescueMu and were used in grids G through J; DNA blot hybridization analysis of 30-200 grid plants was used to estimate transposition frequencies within each grid, which ranged from 0.38 to 0.66 (Table 1 ), with an average of 0.58 per plant and 0.29 per parental RescueMu element. The two parental trRescueMu elements were shown to be segregating 1:1 and independently (Figure 2 for grid G, and data not shown for other families). Subsequently, surveys within each grid were used to identify plants with two or three newly trRescueMu and no evidence of Mutator silencing for construction of the next tagging populations. In this manner, the frequency of trRescueMu was increased in some grids to 1.0-1.4 per plant (Table 1 ) reflecting a frequency of 0.5-0.7 per parental element. Library plate preparation and gene representation As shown schematically in Figure 1 , the trRescueMu insertion sites have been immortalized by preparing libraries from each of the row and column leaf pools from 16 grids, with three additional grid libraries under construction (Table 1 ). Briefly, total maize DNA was digested with Bam HI and Bgl II, both of which recognize sites outside of RescueMu , and the fragment mixture was used to transform E. coli (see Materials and methods). The resulting library plates contain 56-96 individual row and column libraries representing the diversity of germinal trRescueMu and a sampling of somatic events present in the harvested leaf tissue (each well in a library plate is a pool of 20-48 plants from a row or column). The parental RescueMu insertion sites inherited from the grid founder(s) are present in every library. Library plates contain a high diversity of genomic sequences. In a row of 48 plants, assuming random insertion, two segregating founder elements and a transposition frequency of 1.0, there will be 50 different plasmid types in the heritable class. Including heritable and somatic insertions, we estimate that each row or column library contains about 100-200 distinct plasmid types. Given these parameters, a library plate from a 48 row × 48 column grid with an average of 150 somatic plasmids per row or column library would contain 14,400 somatic insertion sites plus 2,304 germinal events and the two parental insertion sites. Because RescueMu shows a strong bias for insertion into genes [ 20 ], each library plate contains a substantial fraction of the predicted 50,000 genes of maize [ 29 ], provided the insertion sites are random. Ultimately, library plates for 19 grids derived from 33,000 plants and containing an estimated 30,108 heritable trRescueMu insertion sites (grid size × transposition frequency from Table 1 ) will be available online from the Maize Gene Discovery project through MaizeGDB [ 30 ]. Plasmid recovery analysis and identification of probable germinal insertions (PGIs) Based on gel electrophoretic analysis of nearly 1,000 rescued plasmids, the genomic DNA flanking RescueMu averaged 3.5 kilobases (kb), with a range of 0.4-15 kb (data not shown). To accommodate the large size of some plasmids, a PCR template preparation protocol was devised to amplify genomic inserts of up to 16 kb for high-throughput sequencing [ 31 ]; primers were designed to amplify from within the right and left TIRs reading outward into the maize genomic DNA such that high quality sequence would be available to identify the TSDs flanking RescueMu insertion sites. Plasmids from all rows plus several columns of a grid were sequenced, with a routine yield of 80-92% success. A subset of plasmids could not be bidirectionally sequenced, because they lacked the TIRs at one or both ends. Deleted forms of trRescueMu were detected in several percent of the individuals surveyed by DNA blot hybridization (see Figure 2 for an example). If such derivatives retained the origin of replication and ampicillin-resistance marker, they could be cloned by plasmid rescue; if the TIRs were absent, they could not be sequenced. Previous analysis of trRescueMu demonstrated that somatic insertion events, typically found in a tiny leaf sector, were sequenced just once from a leaf DNA sample while multiple instances of the germinal events could be recovered [ 20 ]. Out of 28,988 non-parental plasmids sequenced, 41% (11,749) were recovered once (new trRescueMu somatic plus germinal insertion events) for each grid, and 59% (17,239) were recovered multiple times (probable new trRescueMu germinal insertion events). In addition, a total of 24,875 parental plasmids were transmitted from the founder plants. The percentage of parental plasmids within each grid varied from 17% for grid G to 61% for grid P. Some grids had more parentals than other grids and some parental plasmids were preferentially sequenced for unknown reasons. The parental insertion sites include the two or three known parental sites that each segregated into 50% of the progeny. Somatic sectors in the tassel or ear of the parental plant that generated plasmids found in multiple individuals within the grid are analyzed in a later section. Grid sequence data were used to cross-check the transposition frequency estimated from DNA blot hybridization (Table 1 ) using both a row and column matching method and a more general multiple recovery method. Analysis of 80 individuals from six contributing outcross ears in grid G identified 54 that were newly trRescueMu , equivalent to a frequency of 0.68 new insertions per plant. Using a Poisson model based on this transposition frequency for an individual grid (Table 1 ), the sequencing goal was established to reach a depth sufficient to insure that with 95% confidence, each probable germinal insertion would be recovered at least once. In the Poisson model, the 5% probability for the zero class (in other words, the 95% probability of finding all PGIs at least once) occurs when the observed mean is -ln(0.05) or approximately 3. After sequencing several rows and at least one column for a grid, multiple occurrences of PGIs were counted and used to project the sequences required to obtain the desired average of 3 occurrences of each PGI. As a cross-check of this coverage using the row and column matching method, the sequenced row plasmids were compared to the sequences available from four columns of grid G and 149 matches were found. This is equivalent to a transposition frequency of 0.81 based on 149/(4 × 46 plants per row), somewhat higher than the estimate of 0.68 based on blot hybridization analysis of individual plants. Recovery in both a row and a column is highly indicative of a probable germinal insertion because the row and column plasmids were obtained from different leaves and only germinal insertions would be found throughout a plant. The results for each analyzed grid are shown in Table 2 . The low column sampling in grid K (only 192 plasmids were attempted for each of three columns) and grid M (96 plasmids for two columns and 192 plasmids for a third column) resulted in a lower than expected number of germinal insertions. Grid P had a low germinal insertion count with this method because a portion of the column sequences was from rows generated from different parental plants and subsequently excluded from the analysis. Analysis of the row and column sequence data within grids demonstrates that the row sequencing was too shallow to recover some probable germinal insertions more than once and that a fraction of germinal insertions were not sequenced. For example, within grid G, 385 plasmids were identified twice in the available column data but were missing from the row sequences; this is over twice the number of plasmids identified by row and column matching. From the number of plasmids successfully sequenced per row within grid G, we estimated a 70-95% probability of sequencing the likely germinal insertion events at least once in the rows. For other grids, the sampling efficiency ranged from 30 to 95% per row. Grids in which some rows had sampling efficiency less than 60% are listed as partial in Table 1 ; sequencing was terminated in portions of these grids because of technical difficulties such as an excess representation of a parental insertion site, a large number of rearranged RescueMu elements that could not be sequenced with the standard protocol, or poor yield of RescueMu plasmids for unknown reasons. The second method of identifying probable germinal insertions includes plasmids that were recovered multiple times, regardless of whether a column sequence was present. Almost all somatic insertions should only be recovered once due to their occurrence in just a few cells. The results using this method for each grid are shown in Table 3 . What these data mean in practice is that the 3,138 probable germinal insertions identified after sequencing the same RescueMu plasmid at least twice is not a comprehensive list of the heritable insertion events. On the basis of the number of grid plants and estimated transposition frequencies (Table 1 ), 8,311 probable germinal insertions were expected from the six grids (see Table 3 ). From this we estimate that the majority of the heritable insertion events are represented by only a single sequenced RescueMu plasmid. It is likely that nearly half of the plasmids recovered just once represent a germinal insertion (0.44 = (8,311-3,138/11,749)). By PCR screening of library plates containing the immortalized row and column plasmids, plants containing a specific insertion event can be verified (Figures 1 and 3 ). Selection against specific plasmids in E. coli probably contributed to non-recovery of certain insertion sites as sequencing templates, and these plasmids may also be under-represented in library plates. Verification of germinal transmission Individual grid plants with probable germinal insertions were identified on the basis of recovery of the same plasmid in both a row and a column. In addition, library plates containing all of the row and column libraries can be screened using PCR, with one primer designed to the Mu1 TIRs present in RescueMu and a second primer in the gene of interest, as illustrated in Figure 3 . A probable germinal insertion plasmid should yield the same size product in at least one row and one column library of that grid plate; the row and column identifiers specify the address of the plant(s) containing this insertion. To test this method, 11 instances of duplicate plasmid recovery in grid G (N. Arnoult and G-L.N., unpublished data) and 14 such cases in grid H (K. Goellner and V.W., unpublished data) were verified to be represented in both a row and a column library by PCR screening of the corresponding library plate. Seedling progeny from the identified row and column plants were evaluated for the presence of the expected RescueMu insertion site. A germinal insertion was verified for 16/16 cases examined by DNA blot hybridization and/or PCR of individual progeny plants in the family (see Additional data file 2 for methods and for plants used to verify germinal transmission [ 31 ]). Mutational spectrum of RescueMu As shown in Figure 4 , RescueMu insertions occur in diverse gene types. Illustrating the utility of Mu tagging, insertions are found in housekeeping genes, such as actin, as well as in regulatory genes such as those for transcription factors and protein kinases. Using the database of mapped maize genes and expressed sequence tags (ESTs) [ 30 ], RescueMu insertions are identified in genes on all 10 maize chromosomes [ 32 ]. These data confirm earlier studies tracking Mu insertions using DNA blot hybridization that established that these elements insert throughout the genome and do not show a measurable bias for insertion locally [ 1 ]. In addition, about 85% of RescueMu insertion sites that match maize ESTs correspond to genes of unknown function, suggesting the discovery of novel genes. Of the 14,887 RescueMu insertion sites identified in six grids (multiple insertions into a gene from the same grid being counted only once because the majority are the same insertion event), 88% represent single instances of transposon insertion locations. There were 596 instances of a specific genomic sequence having two or more RescueMu insertion events. If the maize genome contains 50,000 distinct genes that are targets of Mu insertional mutagenesis, then far fewer cases of duplicate recovery would be expected by chance alone, given the number of events analyzed ( p < 0.001); therefore, RescueMu exhibits some preference for particular genes. To determine if there were 'hotspots' for RescueMu insertion within particular genes, data were compared between grids with independent founder individuals. As summarized in Table 4 , 90% of the RescueMu insertion sites were found in just one grid. This was true for both probable germinal insertion events (plasmids found two or more times within a grid) as well as for singlet sites (a mixture of germinal and somatic events). The 10% of insertion sites found in two or more grids represent independent recovery of a RescueMu insertion into the same locus. In addition to the computational comparison in which an overlap of 50 bases (95% identity) was scored as insertion into the same gene, over 730 insertion sites were examined manually for 250 cases of genes with insertions from more than one grid. Of these insertion sites, 80% were at different locations within the same locus; we found 85 cases of insertions within a 1-10 bp region and 67 cases of insertions at the same base. Previously, Dietrich et al . [ 9 ] reported that 62 of 75 Mu insertions at glossy8 were in the 5' untranslated region, with 15 insertions at the same base; similarly, the beginning of exon 2 within bronze1 is the most frequent site of Mu insertion in that gene [ 8 ]. One RescueMu contig from the Genomic Survey Sequencing (GSS) section of GenBank, ZM_RM_GSStuc03-10-31.4765 [ 33 ], is a hotspot for RescueMu insertion, with six plasmids sequenced from row 42 of grid G and one each from grids H, I, and M. Insertion sites were identical across the grids. Sequences generated to both the left and right of the RescueMu element were aligned as demonstrated in Figure 3a . Many maize ESTs matching a maize acetohydroxyacid synthase were found near this insertion site; the closest (GenBank GI: 4966438) is less than 50 bp away. Because this RescueMu insertion site was recovered multiple times in grid G, a heritable insertion may exist. After PCR screening of grid G plasmid libraries, summarized in Figure 3a , the plant at row 42, column 22 was identified. To assess heritability of this RescueMu insertion site, total leaf DNA was extracted from selfed seed of this plant, namely G 42-22, obtained from the Maize Genetics Cooperation Stock Center. PCR screening of the DNA (Figure 3c ) indicated that plant 5 is homozygous for the insertion and plant 7 is homozygous wild type. DNA blot hybridization with a 0.6-kb purified PCR probe amplified with primer pair 1 + 5 confirmed plant 5 to contain the homozygous insertion allele, plant 7 to be wild-type, and the rest to be heterozygous for the insertion (Figure 3e ). Various mutant phenotypes were observed in plant 5 (Figure 3f ), including retarded seedling growth, reduced plant height, discolored streaks on adult leaves and sterile tassel and ear. Because there are multiple Mu elements in this line, further characterization of selfed progeny of its heterozygous siblings will be performed to determine the true phenotype caused by this insertion. Analysis of 9-bp TSD and insertion site preferences Because a 9-bp TSD is characteristic of Mu insertion events, the 9 bp next to the left and right TIRs of an individual RescueMu plasmid were used to join the right and left flanking sequence provided they were complementary (Figures 1 , 3 ); note that the sequences are complementary because they were generated from different strands. For non-parental plasmids, left and right sequence data were available for 13,966 plasmids, and the 9 bp was readily identified computationally for 47.2% (6,596) of these. The remaining non-parental plasmids did not have both right and left sequence data and/or the 9-bp motif could not be verified; 5.7% (1,816) contain only post-ligation sequences. Possible explanations for incomplete sequencing results include deletions next to Mu1 elements that remove a portion of the TIR as well as flanking host sequence [ 34 , 35 ]; these events occur with about a 10 -2 frequency at existing insertion sites and if they occurred during or subsequent to RescueMu insertion they would preclude identification of the 9-bp repeat. Alternatively, the lack of a 9-bp TSD could reflect sequencing error. Manual inspection of 300 of the unmatched cases indicated that for nearly 90% there was an 8/9-base repeat match with the mismatch being an undetermined base (an 'N') or a single missing or additional base. Given that all sequences were single pass but of high average quality (phred 35, equivalent to one base-calling error in 3,160 bases), we consider that 9-bp TSDs exist in virtually all trRescueMu insertion sites. A few cases showed anomalies in the TSDs, which probably reflect rearrangements near RescueMu . Several groups have reported weak consensus insertion site preferences for Mu based on smaller data sets [ 9 , 18 , 20 ]. We have derived a site-specific frequency profile of the bases from 3,999 RescueMu insertion regions [ 32 ]. The profile is in agreement with what has been reported earlier by Dietrich et al . [ 9 ], showing a strong bias for high G/C content in the 9-bp TSD within a flanking dyad-symmetrical consensus: CCT-(TSD)-AGG. The non-random insertion pattern strongly suggests that RescueMu targeting is at least partially dependent on sequence features. In addition, we have compared the profiles derived independently from insertion sites within confirmed exons, introns and uncharacterized regions, respectively, and found the same base preferences in all three sets (data not shown). Of 14,887 genomic loci, 62% matched maize or other plant EST/cDNAs. As more genomic sequence becomes available that can be assembled with ESTs to annotate the non-coding portions of maize genes, it will be interesting to determine if the RescueMu insertion sites that do not match an EST or gene in another species represent introns or other non-coding genic regions. On the basis of the gene structure annotated by maize EST matching, we have located 968 TSD sites within genes. Of these, 849 are inside exons. To check if RescueMu has a preference for insertion into exons (that is, the above observed high frequencies of exon insertions is not the result of potential high exon proportion in the maize genes), a standard binomial test with normal approximation was performed. On the basis of the matching to ESTs, the lengths of all exons and introns observed from all RescueMu contigs were counted as 2,182,954 bp and 439,403 bp, respectively. Assuming that RescueMu does not have a preference to insert into exons (null hypothesis), the probability of observing an exon insertion event is proportional to the length of exons (single binomial trial probability 0.832). The probability of observing at least 849 exon insertion events was calculated (less than 0.001; reject the null hypothesis). This result suggests that RescueMu has some preference to target exon regions within genes. As outlined in Materials and methods, the RescueMu GSS sequences were scanned and masked for repetitive elements as collected in The Institute for Genomic Research (TIGR) Cereal Repeat Database [ 36 ]. The repeat content was compared with results for GSS sequences derived by methylation filtration (MF) and high C 0 t selection (HC) using the same repeat-masking criteria [ 36 ]. The percentage of masked nucleotides was 16.5, 24.5 and 16.2% for RescueMu , MF and HC, respectively. Therefore, on the nucleotide level, RescueMu shows similar repeat content as the physical enrichment methods. However, after we assembled the RescueMu GSS sequences to remove redundancy, only about 3% of the RescueMu loci are composed of repetitive DNA (equal or greater than 75% masked, Table 5 ). If the maize genome is two-thirds retroelements [ 37 ], then there is an approximately eightfold insertion bias by RescueMu against this component of the genome. We also downloaded the latest MF and HC contigs (version 3.0) from TIGR [ 38 ] and applied the same repeat masking on those contigs. Our results show that 28% of the MF and 6% of HC contigs are repetitive DNA. Thus, RescueMu and HC have similar bias against repetitive DNA, superior to the MF bias. It should be noted, however, that the MF and HC GSS sequencing has generated, on average, much longer contigs than RescueMu (see Additional data file 2). In addition, only 0.4% of the RescueMu insertions were found in either the approximately 10,000 copies of the 9.1 kb 28S + 18S rRNA genes [ 39 ] comprising 3.6% of the 2.5 gigabase (Gb) maize genome, or in the large number of tRNA and 5S rRNA genes in the maize genome (Table 5 ). These results demonstrate a strong bias against insertion into genes transcribed by RNA polymerases I and III. Also shown in Table 5 , about 62% of the RescueMu loci match strongly to maize or other plant ESTs or appear to encode proteins with high similarity to known proteins. In addition, about another 5% of the loci were predicted to be genic regions with high stringency by ab initio gene prediction programs. As a control, we matched ESTs to contigs assembled from unfiltered (random) maize GSS sequences [ 38 ]. From about 33,000 of those unfiltered contigs, less than 20% of them show significant matching to ESTs. This shows that RescueMu contigs contain more than threefold enrichment of genic regions than random sequencing. This is consistent with our expectation that RescueMu preferentially inserts into genes. It is worth pointing out that plant EST collections contain ESTs from repetitive elements. Although we masked contigs using the annotated TIGR repeat database [ 38 ], it is possible that some contigs still contain unidentified repetitive elements, which might overestimate the number of genic regions by matching the same ESTs to different copies of repetitive elements. In particular, 18% of the EST matched regions show high similarity to transposon coding regions based on BLAST searches against the GenBank nucleotide and protein databases, suggesting that at most 14% of unfiltered contigs include protein-coding genes. The numbers of genic sequences from MF and HC was reported to be 27% and 22%, respectively [ 36 ]. However, these numbers are not directly comparable to our RescueMu results, because these authors used much higher stringency for the EST spliced alignments with the BLAT program [ 40 ], requiring 95 and 80% identity, respectively, when matching to the TIGR maize gene index or other plant indices. We used the GeneSeqer program for spliced alignment of the RescueMu data, which tolerates less sequence matching without compromising gene structure prediction accuracy [ 41 ]. The results using GeneSeqer for RescueMu , MF, and HC are very similar (data not shown). Palmer et al . [ 42 ] evaluated the gene discovery rates of MF, EST sequencing and RescueMu by comparing the respective sequence sets to rice gene models. They concluded that unique gene discovery is most efficient with MF at a sequencing depth when EST sampling saturates. However, their reported low gene discovery rate for RescueMu does not reflect the RescueMu insertion bias, because their dataset included all sequences deposited in GenBank. That is, they did not remove the redundancy resulting from multiple sequencing of parental insertions. Multiply recovered RescueMu insertion sites in the progeny of a single founder plant Probable germinal insertions involve plasmids recovered several times within a sequenced row and/or column, but in addition, some RescueMu insertion sites were found in two or more row libraries (Table 6 ). Although these could represent hotspots for Mu insertion at exactly the same base, we consider it more likely that they reflect the known ability of Mu elements to insert pre-meiotically, resulting in several progeny with the same newly generated mutation present as a sector on an ear indicative of a single insertion event [ 43 , 44 ]. Robertson estimated that 20% of Mu transpositions occur pre-meiotically, 60% occur during meiosis or immediately afterwards, and 20% occur after the mitosis that separates the two sperm in haploid pollen [ 1 ]. We infer that multiple row recovery of the same insertion site within a grid was indicative of a likely pre-meiotic insertion; in contrast, authentic hotspots have the same insertion site among grids. A second line of evidence is that DNA blot hybridization surveys to calculate transposition frequency within a grid identified many instances of a particular fragment size shared in two or more progeny (data not shown). Finally, phenotypic screening of grid progeny families identified numerous instances of identical phenotypes segregating in related families [ 45 ]; each such phenotypic class was counted just once in calculating the percentage of families with a new visible phenotypic mutation (Table 1 ). To calculate the extent and timing of pre-meiotic sectors, the sequenced plasmids from grids G, H, I, K, M and P were classified as occurring in a single row or in multiple rows. The development of the tassel and ear must be considered when evaluating these data. An insertion event that occurs during meiosis can be represented in two haploid cells. During microgametophyte (haploid plant) ontogeny, both of these cells survive, resulting in two pollen grains with the same event. In contrast, only one megagametophyte develops after megaspore meiosis; therefore, female meiotic and subsequent events in the haploid megagametophyte are always represented in just one progeny plant. Most grid plants resulted from male transmission of RescueMu and a minority (about 10%) from female transmission. Given that the founder plants produced copious pollen, there is a low probability that two grains carrying the same meiotic insertion will both result in seed; therefore, the same RescueMu insertion site found in two rows should usually be from a pre-meiotic transposition event. For all events found in three or more rows, the insertion event must be pre-meiotic. The 103 insertions sites found in three or more rows of grid G must be pre-meiotic events (see Table 6 ). They represent 9% of the probable germinal insertion events (103/1,091) identified by the criterion of recovery of the same plasmid twice or more (see Table 3 ). The percentage was similar for all six grids: there were 321 events identified in three or more rows out of 3,138 probable germinal insertions. Surprisingly, 138 contigs were found in four or more rows in these six grids, including 34 events in 10 or more rows (Table 6 ). Therefore, occasionally there is a RescueMu insertion event very early in the somatic development of the inflorescence or in the apical meristems. The majority of trRescueMu insertion sites are found in only one row (92% of germinal plus somatic insertion sites, Table 6 ). As a cross-check on the analysis of pre-meiotic events presented in Table 6 , we evaluated the actual number of individual plants containing the same insertion site for a subset of each grid, using the sequence data from columns. Using this method we confirmed that among 184 plants in grid G with both row and column sequence data, there were 65 cases of insertion sites found in two or more rows or in two or more columns (Table 7 ). Similar results were obtained for the other five grids. From these calculations and the data in Table 6 it appears that RescueMu insertions must occur routinely before meiosis and that, although rare, there are a significant number of early somatic insertion events that are transmitted to multiple progeny. Discussion RescueMu was introduced into maize by particle bombardment resulting in complex transgene loci containing multiple copies of the transposon and the Basta-resistance plasmid used for selection of transgenic lines [ 20 ]. After crossing with an active Mutator line, RescueMu exhibited somatic excision from a 35S:Lc reporter allele resulting in a red-spotted aleurone but the heritable insertion frequency was very low. Progeny screening identified individuals containing two or three trRescueMu elements lacking the original transgene array by genetic segregation and unmethylated Mu1 and MuDR elements. Some of these individuals and subsequent derivatives with the same characteristics were used as founder plants to construct grids of plants organized into rows and columns for efficient generation and analysis of germinal mutations. Tagging maize sequences with RescueMu followed by plasmid rescue and sequencing of the flanking host DNA has identified 3,138 insertion locales from 17,239 plasmids (see Table 3 ). These plasmids represent 59.5% (17,239/28,988) of the total non-parental plasmids of the genomic loci found in each grid. Because sequencing depth was too shallow to identify all likely germinal insertions, the 40.5% of non-parental plasmids recovered just once (11,749 from Table 3 ) represent a mixture of somatic and germinal events. On the basis of the estimation of germinal insertion frequency from DNA blot hybridization, the six grids should contain more than 8,000 heritable trRescueMu insertion sites, but the sequencing depth was too shallow to identify all of these by multiple recovery of the same plasmid two or more times. RescueMu is suited for both reverse and forward genetic strategies. Given the genomic sequence contiguous to any trRescueMu , a PCR screen can be designed to identify which plant contains the insertion of interest using 96-well plates containing the immortalized collection of row and column rescued plasmids. The row and column plant address can be used to order seed for further genetic and phenotypic analysis as illustrated by the RescueMu insertion into the acetolactate synthase gene (Figure 3 ). Alternatively, the phenotype database, which is organized by individual plant, can be searched to identify individuals segregating for mutations of interest. Active Mutator lines with multiple mobile Mu elements were used so most mutations will be caused by these Mu elements because they increase mutation frequency 50-100-fold above spontaneous levels [ 1 ]. The high forward mutation frequency reflects the copy number of the elements and their preference for insertion into or near transcription units [ 1 ]. From the DNA hybridization blots (data not shown) used to verify that grid founder plants had unmethylated Mu elements, the copy number of unmethylated Mu elements was estimated at 20-40 per founder; therefore, two mobile RescueMu elements would be expected to account for 5-10% of the newly generated mutations. Seed was ordered through the Maize Genetics Cooperation Stock Center [ 46 ] for further characterization. RescueMu insertions were found in genes and ESTs mapped to all 10 maize chromosomes [ 31 ], and were found in all of the gene classifications for maize (Figure 4 ). These data confirm the empirical observations of maize geneticists that MuDR/Mu transposons are general and efficient mutagens for maize genes [ 1 ]. Analysis of 14,887 loci defined by RescueMu insertions demonstrates that transposition is highly preferential for RNA polymerase II transcription units: about 62% of the sites match maize or plant ESTs. Because the EST collections are incomplete and lack intron and promoter sequences, it is likely that an even higher proportion of RescueMu insertion sites are in or near genes but cannot be currently assigned to a specific gene. Given the current efficiency, large tagging populations in excess of 200,000 plants would be required in order to recover RescueMu mutations in all maize genes (estimation is based on the calculation method in [ 47 ]). The numerous grids evaluated for phenotypic characteristics should approach saturation of visible mutations, although most of the mutations are caused by standard Mu elements. Given that the maize genome comprises approximately 70% retrotransposons and other highly repetitive sequences, including around 10,000 copies of the rRNA genes [ 37 ], these components of the maize genome are significantly under-represented in RescueMu insertion sites. Only about 8% of the RescueMu insertion sites match repetitive elements and few insertions (0.4%) were recovered in genes transcribed by RNA polymerase I or III. These results suggest that a chromatin component associated with polymerase II transcription units or the absence of a structure in other classes of genes is important in targeting RescueMu and other Mu elements to maize genes. Similarly, recombination during meiosis and transcription per se is targeted to genes. It is likely that the parasitic Mu elements exploit an element of host gene packaging that evolved for other reasons to facilitate transposition into genes. The biological specificity for maize genes exhibited by RescueMu is close to methyl filtration and high C 0 t fractionation. The probable germinal insertion class defines a collection of mutations of enormous potential for the phenotypic characterization of maize with specifically disrupted functions. However, the low cost of template production is a distinct advantage of both physical enrichment methods compared to the high cost of designing, sampling and self-pollinating tagging grids. Current levels of sample sequencing from the physical enrichment templates highlight the desired redundancy of the RescueMu method, which is important for distinguishing somatic from germinal insertions at individual loci. The physical enrichment methods are considerably below one times coverage of the transcriptome of around 250 Mb; hence the current efficiency of generating novel sequence (the likelihood that the next clone sequenced is new) is much higher with these methods than with RescueMu . Using the RescueMu insertion site data, several parameters of Mu transposition behavior were investigated. We confirm that a 9-bp TSD is characteristic of virtually all Mu insertion sites. We confirm that a small percentage of trRescueMu suffer deletions, including loss of a TIR, as noted in previous studies of Mu1 [ 35 ]. Through evaluation of several hundred Mu insertion sites [ 9 , 18 ], consensus motifs have been proposed for insertion sites. The sequence profile derived from the much larger population of RescueMu insertion sites is consistent with the previously proposed motifs. A bias exists for G+C-rich sequence, reflecting the composition of maize exons. We confirm that there are hotspots for Mu insertion, identified by finding identical trRescueMu insertion sites in independent grids. A few loci were recovered in four or more of the six grids analyzed, and many more in two (1,295 genes) or three (233 genes) grids. There is no strong DNA consensus motif at these hotspots, and we consider it more likely that a specific DNA structure or a protein associated with genes establishes conditions for efficient Mu insertion at particular sites. It is important to note that active transcription is not a requirement for Mu element insertion; otherwise Mu would preferentially insert into genes active late in floral development and in gametophytes. The trRescueMu insertion sites represent a mixture of non-heritable somatic insertions present in leaves, germinal insertions in single grid individuals, insertion events in pre-germinal sectors within flowers, and parental elements. Parental elements identified in a grid founder plant segregated 1:1 in the progeny as expected. In addition, some insertion events were found in three or more grid rows, and hence in three or more individuals, and must be pre-meiotic transposition events in the founder. This class represented 10.2% (321/3,138) of all the likely germinal insertions identified (calculated from Table 6 ). Given the clonal analysis model of the pattern of cell divisions establishing the ear and tassel of maize [ 48 - 50 ], the earliest events within the apical meristem could affect up to half of the ear or tassel, with subsequent events affecting progressively narrower portions of the inflorescence. The majority of the pre-meiotic events are consistent with RescueMu transposition in the floral cells a few cell divisions before the onset of meiosis, that is, in precursor cells that are still proliferating and could generate at least two and up to approximately 50 meiocytes. A smaller fraction of new insertions events occurred early enough to be represented in many progeny of a particular plant. These rare, early transposition events generate very large sectors within the developing inflorescence. Mu transposon mutagenesis is highly efficient, primarily because the transposon targets genes and it is usually found in 10-50 copies per genome. How does the plant tolerate the large number of mutations generated by this agent? Within the diploid somatic tissues, most new mutations lack a phenotype; however, the haploid gametophytes are subject to stringent selection. Unlike animals, in which the phenotypes of the sperm and egg are set by previous gene activity in the parent, many characteristics of the haploid phase of the plant life cycle reflect haploid genetic activity, which requires overlapping but distinctive suites of genes in the mega- and microgametophytes [ 51 ]. Consequently, the late timing of new Mu insertions generates gamete diversity, but the unfit genotypes are culled from the population before fertilization. Coe et al . [ 52 ] describe the general problem that lethals occur much more frequently in pollen than in the megagametophyte. Any method that relies on pollen transmission will therefore fail to recover certain types of mutations that would be recovered through female transmission. For this reason, a subset of maize genes required in both types of gametophyte is refractory to knockout mutagenesis. Conclusions A public resource of transposon-tagged maize alleles was constructed and evaluated. RescueMu is an efficient tag for mutagenizing and cloning maize genes, because 66% of insertion sites appear to be in genes. Sequencing from immortalized plasmid libraries organized into row and column plates reflecting the organization of fields of plants permit identification of probable germinal insertions; the library plates can be searched by PCR to verify germinal insertions and subsequently acquire seed of the corresponding plant. Alternatively, a searchable database of segregating plant phenotypes in seed, seedling, or adult tissues can be used to find plants carrying mutations of interest. Although RescueMu can target most, if not all, RNA polymerase II transcription units in the nuclear genome, the transposon does exhibit hotspots in particular genes. Neither the hotspots nor other insertion sites contain a motif(s) defining predictable insertion locations. RescueMu properties confirm attributes established with smaller populations of standard Mu elements. Materials and methods Biological materials RescueMu contains all of Mu1 plus a 400-bp segment of Sinorhizobium meliloti and pBluescript (Stratagene), as described previously by Raizada et al . [ 20 ]. The complete sequence of RescueMu was obtained in this study using PCR primers to amplify overlapping sections of the element [ 31 ] for bidirectional sequencing (GenBank accession AY301066). In the construct used to make transgenic plants, the RescueMu transposon was placed in the 5' untranslated region of a 35S:Lc expression plasmid where it blocked expression [ 20 ]. Lc is a member of the R family of transcriptional regulators of the anthocyanin pathway [ 53 ]. Transgenic maize lines in the A188 × B73 ( r-r/r-g , A1 , Bz1 , Bz2 ) hybrid background were crossed to r-g testers and subsequently with r-g Mutator lines containing multiple copies of MuDR to visualize RescueMu somatic excision as red anthocyanin sectors in an otherwise white aleurone. The tagging populations used here were developed by screening for transposition of RescueMu from the original, complex transgene arrays to diverse genomic locations. Using DNA blot hybridization, these once-transposed RescueMu ( trRescueMu ) were closely monitored for subsequent transposition, and lines were monitored for Mu1 and/or MuDR methylation in the TIRs, a sign of incipient Mutator silencing. Details of line development and evaluation, including DNA blot hybridization methods, will be presented elsewhere. The anthocyanin tester lines (recessive for r-g , a1 , bz1 or bz2 ) were in inbreds W23, K55, A188, or hybrid combinations of these lines. Some RescueMu lines used in tagging grids were crossed to inbreds A619 or B73, which are both r-g , A1 , Bz1 , Bz2 . Grid backgrounds are presented in detail at [ 31 ]. Plasmid rescue and DNA sequencing Detailed protocols are presented at [ 54 ], and a schematic is provided in Figure 1 . Briefly, leaf tissue was collected from all plants in each row and from a different leaf in each column of a grid. A separate plasmid rescue library was constructed after Bam HI plus Bgl II digestion of the genomic DNA preparations. These libraries were immortalized in library plates available from the project [ 31 ]. Plated colonies were picked, grown overnight in liquid media, and sequencing templates prepared by a direct PCR method suitable for amplifying genomic inserts of up to 16 kb. Cycle sequencing was performed using Big Dye Terminator chemistry to read out from a position around 110 within the left or right terminal inverted repeat (TIR) of RescueMu ; although the primers were selective for one TIR, there was some cross-priming. All grid rows plus several columns were sequenced. Three 96-well plates were normally sequenced for each row or column to obtain sequence information for a desired minimum of 200 plasmids; additional sequencing reactions were conducted if necessary. Matches of row and column sequences are designated as probable germinal insertions, because they represent an insertion site present in two leaves of that plant (designated by its row and column address); when only row sequences were available from a particular plasmid, probable germinal insertions were designated after recovery of the same sequence two or more times. Plasmid types recovered just once are a mixture of heritable and strictly somatic insertions. Parental RescueMu insertion sites present in a grid founder plant segregated in the grid progeny, and these insertion sites were expected to be found in all rows and columns. In some cases, particular parental plasmids were over-represented in the sequenced plasmid population. To reduce their contribution and increase recovery of new insertion sites, a rare-cutting restriction enzyme site was identified in the parental plasmid and the corresponding enzyme was included in the genomic DNA preparation to bias against recovery of that parental plasmid. PCR screening of a library plate to quantify a RescueMu insertion hotspot Six gene primers plus one RescueMu left readout primer were used in this study: 1. 5'-TTGGGAGGTTGAAGGTAAAGACAT-3' 2. 5'-GTGCTG GATTGGTTACTCCG-3' 3. 5'-CGATGATTCTAGTTGAGCGTCTG-3' 4. 5'-ACTCGCACCAACATGAATACC-3' 5. 5'-GTTTCCGAGGACGCAGAGG-3' 6. 5'-AGCGCCAGGGCCAGGGGATTC-3' L. 5'-CAT TTC GTC GAA TCC CCT TCC-3' ( RescueMu ) Locations and directions with respect to the insertion site of RescueMu are shown in Figure 3a . PCR conditions were as follows: 5-20 ng of each plasmid library, 2.0-2.5 mM Mg 2+ , 0.4 mM dNTPs, 0.8-1.0 μM gene primer and 4-5 μM RescueMu L primer in a 50 μl reaction was first denatured for 2 min at 95°C followed by 35 cycles of 30 sec at 95°C, 30 sec at 55°C and 2 min at 72°C, and a final 2 min extension at 72°C. The same PCR conditions were used for screening using 5-100 ng samples of maize total genomic DNA. DNA blot hybridization Total genomic DNA was extracted from leaf tissues using a modified urea method [ 55 ]. After overnight digestion, the restricted DNA was separated on a 0.8% agarose gel and transferred onto Hybond-N+ membrane (Amersham Biosciences) in 0.4 M NaOH. Blots were hybridized with non-radioactive probes labeled with AlkPhos DIRECT system (Amersham Biosciences) for chemiluminescence detection on X-ray film. Initial clustering and assembly of genomic sequences The sequences were screened to remove the TIR sequences using the program crossmatch [ 56 ] and then trimmed to achieve a minimum phred score >15 in sliding windows over 40 bases. Overall the quality scores averaged phred >35, or less than one error in 3,160 bases. The average length of the trimmed, high quality genomic sequence entering the assembly was 378 bases. The right-TIR primer yielded 22% more successful sequence than the left-TIR primer resulting in an excess of right side sequences. Trimmed sequences were then assembled into contigs using phrap [ 56 ] with the following parameters: -minmatch 35 -minscore 30 -node_seq 14 -node_space 9. The member sequences for each contig were extracted from the phrap output files and assigned to a row or column of a grid. Within each contig, only a single sequence from a plasmid was used to determine the row and column representation. For example, if both the left- and right-flanking sequence from a plasmid assembled into one contig, this was considered one recovery of the plasmid. If the left-flanking sequence from one plasmid and the right-flanking sequence from a separate plasmid assembled into the same contig, this was considered two independent recoveries of the same genomic locus. In the latter case, if the right- flanking sequence was from a different row, then the sequence was recovered in multiple rows as well. All sequences were deposited into the Genomic Survey Sequencing (GSS) section of GenBank [ 57 ]. Assembly of RescueMu -derived genomic sequence data As shown in Figure 1 , using the 9-bp TSD characteristically generated during Mu element insertion [ 1 ], the sequences to the right and left of a particular RescueMu element can be assembled into a continuous sequence. To do this, trimmed RescueMu GSS sequences were downloaded from GenBank [ 58 ], for comparison to raw GSS sequences containing the Mu1 TIR sequences. The TIRs were masked by the cross_match program [ 56 ] to determine the flanking 9-bp TSD sequences. The TSDs are the end-overlaps between GSS sequences generated from the left and right side of RescueMu insertion. Merging through TSDs using the reverse-complementary strand of the left and right sequences recovers the original genomic sequences flanking the RescueMu insertion. A special consideration in the assembly of the genomic sequences flanking the right- and left-TIRs of RescueMu is the presence of a GGATCC ( Bam HI), AGATCT ( Bgl II), or a GGATCT ( Bgl II/ Bam HI) or AGATCC ( Bam HI/ Bgl II) motif. The two restriction digestion sites represent a true ligation site of sequence that was non-contiguous in the maize genome, but the post-restriction site sequences can unambiguously be assigned to the right or the left of RescueMu . On the other hand, the GGATCT or AGATCC motif could be contiguous genomic sequence or could have been generated during the ligation step of the plasmid rescue. Consequently, assignment of the position of the sequence beyond the GGATCT or AGATCC motif is ambiguous. If the RescueMu insertion site matched EST sequence across and beyond the GGATCT or AGATCC motif, the post-ligation sequence could be properly assigned (Figure 1 ). In the RescueMu plasmid sequences considered here, the average number of sequences reported to GenBank was 2.3 (131,364/57,022) per plasmid. The 131,364 RescueMu GSS sequences deposited at GenBank were screened for vector sequences against the UniVec database at the National Center for Biotechnology Information (NCBI) [ 59 ] using the crossmatch program: -mismatch 12 -penalty -2 -minscore 20. The resulting 130,861 vector-trimmed sequences were then screened against the maize repeat database annotated by TIGR [ 60 ] using the Vmatch program [ 61 ] with the parameters -l 50 -h 3 -identity 95. The 127,708 repeat-free sequences were then used to identify parental insertions. Any given RescueMu -transformed plant contains the parental RescueMu elements that were recovered at a high frequency during sequencing (from every sequenced row or column). Because our goal is to analyze the gene discovery by newly inserted RescueMu (that is, we are interested in where those non-parentals inserted into the maize genome), we decided to filter out the parental sequences as much as possible. We used Vmatch to cluster near-identical left and right sequences for each grid. A parental cluster contains sequences from nearly all the row or column sequences. A total of 59,069 parental sequences were identified and were excluded from the subsequent assembly. All the non-parental sequences were first preassembled for each plasmid using the left and right 9-bp TSD overlap. The merged GSSs were first clustered by PaCE [ 62 ] (minimum exact match 36 bp, minimum score threshold 30%) and then consensus sequences (contigs) for each cluster were generated by CAP3 [ 63 ] (overlap 40 bp; 90% identity cutoff). Because PaCE and CAP3 only pair sequence with the minimal overlap required to establish statistically significant identity, the number of contigs is probably an overestimate of the number of independent RescueMu insertion sites. For the particular case where TSDs were not recovered during sequencing, the left and right sequences could not be assembled together, even though they were from the same plasmid. Therefore, a Perl script was developed to conduct single-linkage clustering based on clone-pair constraints to assemble the GSS to the same 'genomic loci' if they were derived from the same plasmid clone. Classification of insertion site context To be successful as a gene-discovery tool, the transposon insertions must be predominantly into the genic regions of the maize genome. To quantify the potential enrichment of the RescueMu flanking sequences for genic regions, we matched all assembled contig sequences against various classes of known repetitive sequences, including retrotransposons, DNA transposons, centromeric and telomeric repeats, rRNA genes and plastid DNA. For this analysis, the non-parental sequences were used in their original form, with only vector sequences but not repeat sequences trimmed. The sequences previously discarded for analysis because they consist almost entirely of repetitive elements were assembled using the same procedure as described above for the repeat-trimmed sequences. Note, however, that this number of loci is unreliable and probably an underestimate of the true number of loci recovered because of the intrinsic difficulty with assembling repetitive DNA. To identify the repetitive elements in the contigs, Vmatch (-seedlength 14 -hxdrop3 -l 30 -identity 70) was used in combination with the TIGR cereal repeat database (version 2 consisting of maize, rice, barley, sorghum and wheat repeats). The contigs were also scanned from tRNA genes by tRANscan-SE program [ 64 ] with its default parameters. Gene discovery in GSS contigs Both similarity-based and ab initio approaches have been used to detect gene structures of the GSS contigs. For the similarity-based approach, GeneSeqer [ 65 ] programs were used to match plant EST contigs and cDNAs to GSS contigs. The plant EST contigs were regularly assembled by PlantGDB [ 66 ]. For the ab initio prediction, GENSCAN [ 67 ] (with default parameter settings for maize) was used and only high exon score predications (≥0.90) were selected. The GSS contigs were compared against SPTR [ 68 ], a nonredundant protein data set collected by the European Bioinformatics Institute (EBI), using BLASTX [ 69 ] with an E-value ≤ e-20. The BLASTX top protein hits were used to assign putative functions to the unique regions and for classification into functional categories based on annotation in the Gene Ontology [ 21 ] database. The genetically mapped maize ESTs were retrieved from MaizeGDB [ 70 ]. These ESTs were spliced-aligned to GSS contigs using GeneSeqer as described above. The matched GSS contigs were then plotted on the maize IBM Neighbor genetic map [ 30 ]. Analysis of 9-bp TSD and insertion site preferences For the analysis of RescueMu target sites, we retrieved the 9-bp TSD sequences from the confirmed insertion sites where both the left and right sequences match on the 9-bp TSD. We also retrieved the 20 bp up- and downstream sequences around the TSD. Then a 15-base long profile (9-base TSD and its three up- and downstream neighbors) was derived from the sequences and their reverse-complement orientation determined using the Expectation Maximization Algorithm [ 71 ]. Analysis of tentative unique contigs containing GSS sequences from multiple grids The GSS seqeunces present in each tentative unique contig (TUCs) were extracted from [ 31 ] and assigned to a row or column within a grid. A sample of TUCs with GSS sequences from multiple grids was then selected for detailed analysis. For each GSS in the TUC (excluding post-ligation sequences), the exact location of the TSD was determined by visual examination of the sequence alignment file for the TUC and the untrimmed GSS sequence data. The number of GSS sequences for each grid at each transposition site was recorded. Phenotypic analysis Grid plants were self-pollinated unless male or female-sterile. The resulting F1 families were evaluated by inspection of ears and kernels, at weekly intervals for five weeks after germination in a sand bench in a greenhouse, and at weekly intervals throughout the life cycle in the field. Phenotypes observed were recorded and are assembled into a searchable database at [ 31 ]. Unique phenotypes were documented with a digital image, and there are links to corresponding RescueMu flanking sequences where established. Instructions on how to obtain seed of grid plants is also provided. Additional data files The following additional data are available with the online version of this article: a table listing the internal primers used in sequencing RescueMu (Additional data file 1 ), supplementary material for this paper, including details of methods (Additional data file 2 ). Supplementary Material Additional data file 1 A table listing the internal primers used in sequencing RescueMu Click here for additional data file Additional data file 2 Supplementary material for this paper, including details of methods Click here for additional data file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC545602.xml

548130

Identification of novel prognostic markers in cervical intraepithelial neoplasia using LDMAS (LOH Data Management and Analysis Software)

Background Detection of Loss of Heterozygosity (LOH) is one of the most common molecular applications in the study of human diseases, in particular cancer. The technique is commonly used to examine whether a known tumour suppressor gene is inactivated or to map unknown tumour suppressor gene(s). However, with the increasing number of samples analysed using different software, no tool is currently available to integrate and facilitate the extensive and efficient data retrieval and analyses, such as correlation of LOH data with various clinical data sets. Results An algorithm to identify prognostic disease markers is devised and implemented as novel software called LDMAS. LDMAS is a software suite designed for data retrieval, management and integrated analysis of the clinico-pathological data and molecular results from independent databases. LDMAS is used in stratification of disease stages according to clinical stage or histological features and correlation of various clinico-pathological features with molecular findings to obtain relevant prognostic markers such as those used in predicting the outcome of cervical intraepithelial neoplasia (CIN). This approach lead to the identification of novel prognostic cervical cancer markers and extraction of useful clinical information such as correlation of Human Papilloma Virus (HPV) status with CIN lesions. Conclusions A novel software called LDMAS is implemented and used to extract and identify prognostic disease markers. The software is used to successfully identify 4 novel prognostic markers that can be used to predict the outcome of CIN. LDMAS provides an essential platform for the extraction of useful information from large amount of data generated by LOH studies. LDMAS provides three unique and novel features for LOH analysis : (1) automatic extraction of relevant data from patient records and reports (2) correlation of LOH data with clinico-pathological data and (3) storage of complex data in flexible format. The first feature automates the creation of database of clinically relevant information from huge amount of data, the second feature extracts useful biomedical information such as prognostic markers in CIN and the third feature simplifies the statistical analyses of the data and allows non-statisticians to carry out the analysis. Additionally, LDMAS can be used to extract clinically useful markers from other diseases and interface to high throughput genotyping analysis software such as GDAS used to generate LOH data from Affymetrix ® GeneChip Mapping arrays.

Background Detection of LOH is one of the most common molecular applications in the study of human diseases, in particular cancer. It is commonly used to examine whether a known tumour suppressor gene is inactivated or to map unknown tumour suppressor gene(s). Detection of LOH not only helps in understanding the molecular mechanisms underlying the development of cancer, but also provides important information useful for disease diagnosis and prognosis. LOH detection is commonly carried out by the analysis of microsatellite markers using an automated DNA sequencer. With the raw data from the sequencer being stored in one file per lane together with corresponding clinical information and patient follow up data, each LOH study [ 1 , 2 ] generates hundred of files that need to be organised and related in a structured format. However, with the increasing number of samples analysed using different software, no tool is currently available to facilitate the extensive and efficient data retrieval and analyses, such as correlation of LOH data with various clinical data sets. We have developed a novel software package: LOH Data Management and Analysis Software (LDMAS) in order to satisfy these needs. LDMAS can retrieve LOH data from automated DNA sequencer platform and clinical data from any patient record system and correlate different data sets according to the user's choice. Here we present how LDMAS interfaces to Genotyper software (ABI, Foster City, CA) which is used to determine the presence of LOH, and the patient record system SunQuest (San Francisco, CA), facilitating the identification of LOH markers associated with the development of CIN [ 3 ]. CIN show variable clinical behaviour despite morphological homogeneity within each subgroup. Clinically, it is vital to distinguish CIN lesions with different behaviour and identify those likely to persist and progress despite treatment. Implementation System architecture LDMAS package is composed of three modules: (1) MRES (Medical Report Extractor Software) which parses patient report files, extracts the information of interest and organises it into a structured format, applicable to LDAS (2) LDAS (LOH Data Analysis Software) which obtains LOH data from Genotyper (Applied Biosystems, California) or GDAS (Affymetrix, California) software and correlates it to clinical data obtained from MRES (3) LDMS (LOH Data Management Software) which is used to gather patients' clinico-pathological data and extract significant relationship between the various data sets LDAS and LDMS work synergistically to manage and analyse LOH data. The MRES source code for automatic parsing of patient reports is written in C++ using C++ Builder 5.0 (Borland Software Corporation, Scotts Valley, CA), LDAS is written in Visual Basic for Application as an Excel 2000 add-in and LDMS is written as Visual Basic for Application modules embedded within Access 2000 as fully functional software. These modules can be run independently and used for applications other than LOH. LDMAS runs on Microsoft ® Windows 2000 or Windows XP operating system. Figure 1 shows LDMAS architecture. Type of input data for LDMAS modules The MRES module takes its input from any patient report file containing clinical details such as diagnosis, stage of the disease, treatment and follow up results, parses and formats patient's data into a structured format that can be saved as Excel spreadsheet. In this case, data were taken from SunQuest patient record system and MRES converted and produced the data as: (1) Hospital Number (2) Hospital ID (3) Patient Name (4) Date of Birth (5) Pathological specimen Number (6) Date of Diagnosis (7) Histological diagnosis. The user can manually check the data and use it as template for analysis. Data analysis is carried out using LDAS which obtains LOH data from Genotyper and correlates it to the clinical data obtained from MRES. LDAS obtains data in plain text format and can thus be easily interfaced to any LOH platform generating software such as Genotyper and GDAS. Finally all data are entered into LDMS for storage and intelligent mining of data. Database query results can be exported back to LDAS allowing correlation between LOH and various clinico-pathological parameters such as age, histological grade, treatment modality and their responses, and HPV status as well as carry out multivariate analysis to determine the sensitivity and specificity of the markers involved in the LOH study. A more detailed description of all the modules is provided in LDMAS user's guide [Additional file 1] and the example below. Advantages of LDMAS LDMAS offers several advantages to users. It is user friendly and its architecture is modular allowing versatility of use. It enforces the standardisation of procedures for studies involving large cohorts of individuals. The data is well organized since LDMAS systematically assigns LOH results of each case to its corresponding clinical information. Additionally, LDAS standardizes LOH data analysis implicitly and allows the user to edit the data manually if needed. Microsoft Excel has been chosen to implement LDAS because of its wide use, versatility and convenient statistical analysis features facilitating the implementation of multivariate analysis and correlation testing between LOH and clinico-pathological parameters. Results and discussion LDMAS application in identification of LOH markers associated with persistence / progression of cervical intraepithelial neoplasia We divided the CIN groups into disease free indicating cases that become CIN free after treatment, and disease persistence/progression indicating cases that develop show progression or persistence of CIN despite treatment. We used LDMAS to retrospectively examine the prognostic value of LOH at 12 microsatellite markers including 10 from 3p14, 3p22-21, 6p21 and 11q23 which are frequently deleted in cervical cancer [ 3 , 4 ], in 164 cases of CIN lesions using archival cytological/histological specimens. LOH was further correlated with high risk HPV infection. Initially MRES was used to automatically parse 4300 patient records and extract clinico-pathological data including age, diagnosis, method of treatment and treatment response during follow up. Out of those, 164 cases with follow up of 3 or more years were chosen for the study and their clinico-pathological information was imported into LDAS. Initially, 71 out of the 164 selected cases were examined for LOH using 12 fluorescent microsatellite markers ran on ABI377 DNA Sequencer. LDAS was then used to identify the microsatellite markers for which LOH was significantly associated with disease persistence/progression of CIN using two tailed student t-test. Figure 2 generated using LDAS shows that microsatellite markers D3S1300 (3p14.2), D3S1260 (3p22.2), D11S35 (11q22.1) and D11S528 (11q23.3) have the highest LOH in CIN lesions displaying persistence/progression than those who were disease free during follow up [ 5 ]. Validation of prognostic markers associated with persistence / progression of CIN To validate this finding, LOH at these four markers was investigated in a further series of 93 cases. Compatible results were obtained from these additional cases. The two sets of data were combined and further compared using LDMS. Methodologies included : 1) comparison using χ 2 (chi-squared) test of LOH at each of the four microsatellite markers with age, various methods of treatment, different subtypes of HPV infection and between CINs showing disease free or disease persistence/progression. 2) correlation of LOH data with histological grade of CIN, treatment response and various HPV subtypes. Through such complex analysis, we showed that concurrent LOH at two of the four microsatellite markers could identify 47% of CINs that showed disease persistence/progression with 100% specificity [ 5 ]. Furthermore, LOH at D3S1300 was found to be significantly associated with HPV16 infection. Part of this data analysis is supplied in the LDMAS guide [see Additional file 1]. More detailed analysis of this study is described in [ 5 ]. Algorithm for identifying prognostic disease markers Based on the above example, an algorithm can be developed to extract prognostic markers for other diseases. The algorithm can be summarised in the following pseudocode : (1) Divide the disease in groups according to the pathology staging (2) Parse patient data from clinical records and use the groups defined in part (1) (3) FOR each microsatellite marker carry out a two tailed student t-test between the disease groups using LOH data IF t-test p ≤ 0.05 Marker is significant in prognosis of the disease ELSE Marker is not significant in prognosis of the disease (4) Validate the prognostic markers using χ 2 (chi-squared) test of LOH with clinico-pathological data and correlation of LOH data with histological grade of CIN, treatment response and various HPV subtypes. LDMAS has been implemented using the above pseudocode. Conclusions We have devised an effective algorithm to identify and extract useful markers that can be used to predict the outcome of disease and used the algorithm to successfully identify 4 novel prognostic markers that can be used to predict the outcome of CIN. The algorithm was implemented in a novel software called LDMAS which provides an essential platform for the extraction of useful information from large amount of data generated by LOH studies. Furthermore, LDMAS is used to efficiently store, manage and track the data. Its flexible nature allows the easy manipulation of data facilitating complex analysis as demonstrated in the current study. The various modules of LDMAS can be easily adapted and used with other applications such as high throughput LOH and genotyping using SNPs on Affymetrix ® GeneChip Mapping arrays and fingerprinting studies. Modules such as MRES can be used independently to parse medical records facilitating extraction of specific clinical information of interest. Additionally, LDMAS can be used to extract clinically useful markers for other diseases. Availability and requirements The source code and executable files for LDMAS modules as well as user manual including examples from real study data are freely available and can downloaded from our website at : Additionally examples of input files are provided from our website for users to test the software and assess its functionality. Authors' contributions RH designed and developed and implemented LDMAS software and the web site. AE and RH did the experimental work to generate the data necessary to test and validate LDMAS. MD supervised the study and designed the experimental work using CIN biopsies and smears. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548130.xml

549560

Specialized microbial databases for inductive exploration of microbial genome sequences

Background The enormous amount of genome sequence data asks for user-oriented databases to manage sequences and annotations. Queries must include search tools permitting function identification through exploration of related objects. Methods The GenoList package for collecting and mining microbial genome databases has been rewritten using MySQL as the database management system. Functions that were not available in MySQL, such as nested subquery, have been implemented. Results Inductive reasoning in the study of genomes starts from "islands of knowledge", centered around genes with some known background. With this concept of "neighborhood" in mind, a modified version of the GenoList structure has been used for organizing sequence data from prokaryotic genomes of particular interest in China. GenoChore , a set of 17 specialized end-user-oriented microbial databases (including one instance of Microsporidia, Encephalitozoon cuniculi , a member of Eukarya) has been made publicly available. These databases allow the user to browse genome sequence and annotation data using standard queries. In addition they provide a weekly update of searches against the world-wide protein sequences data libraries, allowing one to monitor annotation updates on genes of interest. Finally, they allow users to search for patterns in DNA or protein sequences, taking into account a clustering of genes into formal operons, as well as providing extra facilities to query sequences using predefined sequence patterns. Conclusion This growing set of specialized microbial databases organize data created by the first Chinese bacterial genome programs (ThermaList, Thermoanaerobacter tencongensis , LeptoList, with two different genomes of Leptospira interrogans and SepiList, Staphylococcus epidermidis ) associated to related organisms for comparison.

Background We are facing a deluge of genome sequences. As of January 14th, 2005, the GOLD site identified 1248 completed or ongoing genome programs , and this certainly reflects only a partial view of the existing programs. While this shows that we implicitely possess an enormous wealth of information about the functions carried out by genes and genomes, the very fact that this amount is enormous makes it extremely difficult to mine that information easily. The role of specialized databases is to make this task easier for end-users. Many types of microbial genome databases exist. Most of them have been developed in a context of bioinformatics centres or laboratories purely favoring in silico research rather than the coupling between experiments using computers and experiments at the bench, and this is reflected in the structure and aims of the databases [ 1 - 8 ]. In contrast, at the onset of genome programs, we decided to set up a data structure for bacterial genomes that would help experimentalists to access knowledge on genes and genomes in an end user-oriented fashion. This was first the aim of the Colibri project, with the goal to organize Escherichia coli genome data, well before the whole genome sequence was known [ 9 ]. Later on, the SubtiList database was at the core of the Bacillus subtilis genome program data access [ 10 ]. Many databases constructed on the GenoList data schema were subsequently constructed ( , [ 11 ]). However, with the exponentially growing set of genome sequences, it became important to divide up the work while maintaining the main goal of the project, that of being end-user-driven and of course, user-friendly. While an ongoing effort aims at integrating all bacterial genomes within the GenoList frame into a single database, it is important to create individual databases that could be regularly updated by a selected team of scientists (preferably those that initiated the corresponding genome program). This is particularly important for countries that are beginning to develop at a high speed into the genomics era. We took the opportunity of the creation of the HKU-Pasteur Research Centre in Hong Kong (China) to set up genome databases for the microbial sequencing projects developed in China (with databases for related organisms for comparison). Within this economic context, it was also important to take into account the cost of development. The paradigm GenoList databases are based on commercial DataBase Management Systems (DBMS) [ 11 ] and we decided to shift from a commercial DBMS to a non-commercial one, providing more freedom for the future of the project. In the present set of databases (GenoChore), emphasis is placed on retrieval of information centered on the gene as the central object, with exploration methods that query simple properties of the gene products (such as molecular mass or isoelectric point) in addition to more complex features such as the class of codon usage bias used in the gene [ 12 ]. Furthermore, queries can be made on the sequence itself using large scale analyses such as BLAST, and search for word patterns present in DNA and protein sequences. Construction and content Data schema Because we wished to shift from a commercial DBMS to an open-source one, there were some applications that could not be implemented readily due to the lack of certain advantages possessed by the commercial DBMSs. Hence, we had to alter the data structure in order to cope with this situation. The core data schema used in this work was that of GenoList version 3.1 [ 11 ], with slight modifications (Figure 1 ). Figure 1 Data Schema of the Databases. The core object of the schema is the Genomic_object, as in GenoList. It uses pointers in the sequence that delimits several categories of objects, including protein Coding DNA Sequences (CDSs), RNAs and other objects such as transcription terminators or riboswitches. Database management system In the present GenoList databases, the DBMS used is Sybase™. While this is convenient because of excellent stability and maintenance, this may pose problems in terms of commercial policies, especially if the structure has to be exported. We therefore decided to rewrite the management of the GenoList structure using MySQL . Most function transfers were straightforward. However some functions such as nested subquery that were not available in MySQL had to be dealt with indirectly. The nested subquery has been entirely circumvented in the PERL code and is dealt with in the Extended Search algorithm by concatenating different SQL queries simply using the "AND" or "OR"connector. Data input Sequence and annotation data were parsed from the files extracted from the International Nucleotide Sequences Database (INSD: DDBJ/EMBL-EBI/GenBank [ 13 , 14 ]) with the following procedure. To get access to the INSD, the authors of a genome sequence must follow the specification of the Feature Table Definition (FTD) jointly issued by the INSD partners . The current version is Version 6.2 Oct 15, 2004. While this specification is rigid, there is still a significant degree of freedom in annotation, so that a large number of individual situations have to be taken care of semi-automatically. The basic idea of the parser is firstly to read through the input file at the INSD and check file formats. Subsequently, the information is collected and distributed into several temporary files using a set of predefined keywords and their qualifiers (i.e. those characterizing the data schema). Subsequently, a check process is initiated to identify all situations that do not fit the specifications, so that they can be corrected manually. Usually, most of the process of creating tables is automatic and only a few exceptions have to be corrected individually. A second type of input is also provided as an interactive interface to tell the database curator what information has to be collected: once collected this information can be loaded into the databases directly (Figure 2 ). Teraprot data are obtained from Infobiogen . Figure 2 Implementation of a Database Curator Page. In order to help users who would participate in the improvement of the database annotation a Curator Page is provided permitting input of updated information. It is available to users after acceptation of their collaboration, through a password protected access. Once data consistency has been verified the new annotations are implemented in the current database. Query methods and interface We kept the interface of GenoList as published, except that a box providing access to protected curation of annotations is now provided, aiming initially at helping the first party (sequencing teams) annotators. The front page is made of three frames. Briefly, the vertical frame on the left contains the controls necessary to get access to the content of the database. The upper part of this frame contains text fields for querying the database according to five types of queries: gene name(s), chromosome region around a gene, chromosome region defined by positions, free text, functional classification (more detailed information about each type of query can be obtained by clicking on the question mark near the query title). The "Extended Search" button gives access to a search form allowing the user to perform multicriteria searches on all the database fields. The lower part of this frame allows one to launch sequence analysis tools: BLAST and FASTA database searches (on the sequence data), and DNA or protein pattern searches. In the former case, the user can choose to explore sequences located upstream of putative operons. In the latter case, the user can search for patterns anywhere in proteins, but also restrict the search to the beginning or end of the protein. The upper frame on the right can contain various types of information, depending on the genome and on the query. It can contain a graphical representation of a chromosome region, that can be obtained in several ways: usually from a gene in the bottom frame. This frame may contain launch forms and result lists from the sequence analysis tools available (pattern search, BLAST or FASTA scanning). The bottom frame on the right always contains detailed information about one given gene, including regularly updated BLAST searches and Teraprot reports as well as related bibliographic references. The original package managing the interface of GenoList databases was written in C/C++, following the first database schema [ 10 ], that had been adapted for use with the Sybase™ DBMS (the initial platform was using the DBMS 4 th Dimension™). The modification of the database schema needed for using MySQL required additional adaptations of the application interfaces. Using the original package would have required iterative work that was systematically adding complexity into the system. Current best Web interfaces and application interfaces (i.e. friendly for sharing parties) are often based on Perl scripts. For this reason a new core management script was recreated, written in Perl, while keeping the package architecture and the Web interfaces. Among other languages that have comparable functionalities, the choice of Perl to create the system was motivated by its powerful capability to glue different programs or scripts together. In addition it is widely used by the INSD, and at the European Bioinformatics Institute in particular within the BioSapiens program . Furthermore, this choice allowed us to keep the optimized fast C code that has been constructed for searching pattern (strings of symbols) inside the DNA or amino acids sequences. The GenoList C/C++ package chose to use the GD library for generating graphic representations of genome regions. The GD graphics library is an open source library which allows programmers to easily generate PNG, JPEG, and WBMP images from many different programming languages. We used here a newer version of the same library (perl module perl-GD version 2.11) to make use of its improvements in creating dynamic pictures. In rewriting the core of the program we used the Perl module DBI . A DBI is a middle layer between the outside applications and the communicator (DBD). Different DBMSs have their own communication mechanism to talk with outside applications, and in the present version the choice of the DBI module has been implemented in such a way that we could change the DBMS if necessary with minimal work. In this way, when changing the DBMS, it will only be necessary to tell the DBI about the specifications of the new DBMS without having to modify any other code. Finally, we used the Perl module CGI to facilitate the production of the WebPage interfaces. As a consequence further developments of the GenoChore package should be performed with minimal effort. Utility and discussion Data schema In the original GenoList structure, the central table corresponding to genomic objects carried all relevant features that are associated to genes and gene products. For the sake of future developments and to accommodate new feature annotation present in genome flatfiles, we separated this table into several gene product tables, specific for RNAs and proteins. The current data structure remains open to include tables for other types of data, such as regulation properties annotations when they will become available. Figure 1 displays a diagram of the current generic database schema (we did not show tables that remain empty for want of annotation data). As expected for a database meant to provide knowledge from genome programs, the central tables are focussing on genomic objects, the main one corresponding to protein Coding DNA Sequences (CDSs). To match this structure, the information present in the flat files created by the sequencing consortia, and present in the INSD, is split into three parts, namely, a) genomic objects, i.e. what we see in a chromosome, at precisely identified positions in the genome sequence (depending on the annotation tools available to the consortia), such as a CDS, a promoter, a terminator, a tRNA, an sRNA etc.; b) genome annotations, i.e. protein, RNA and other bio-molecules' products, functions, comments and so on; c) relations between genomic objects: e.g. the typical concept of gene requires its association to a promoter, a terminator and usually a CDS. In this representation, a set of genome objects' identifiers (ids) is used to represent a gene. This facilitates the association of genomic objects together with much more sophisticated relationships into more complex structures, when required. It is important here to notice that, in contrast to a rather ubiquitous practice, we explicitely separate between Open Reading Frames (ORFs) that are simply sequences multiple of 3 between two termination codons (TAA, TAG and TGA) and CDSs, that begin with a specific codon, usually ATG (in the DNA text), preceded by a ribosome binding site (RBS), typically AAGGAGGT in many bacterial genomes. One must remember that in most genomes the beginning of CDSs has not been experimentally identified. Identification of CDS starts is however much easier in low G+C Firmicutes that do not possess a counterpart of ribosomal protein S1 found in gamma proteobacteria [ 15 ]. In the same way, G+C-rich organisms have usually long ORFs, but the CDSs they harbour are usually highly enriched in A+T at the third codon position. Some caution, therefore, should be exerted by users when using the information collected in the databases about the beginning of proteins in these organisms (for example in the Streptomyces coelicolor database, CoeliList). Nomenclature: naming genes Users know that the system used for naming genes in genome databases is extremely unwieldy and completely lacks standardization. This is usually because genes are simply labelled in databases by access numbers corresponding to the annotation phase of the relevant genome program (e.g. PA3004 for a gene found in the genome of Pseudomonas aeruginosa ). In the absence of knowledge of a gene name it takes some time to identify it (often using BlastP), for example when aiming at the study of its neighborhood ((i.e. proximity of an object or a relationship with others sharing the same conceptual space, including presence in a common article [ 12 ]). Naturally, because most genes have never been experimentally identified in the majority of the newly sequenced genomes, this approach is certainly safer than giving a name without proper identification criteria. However it is extremely useful for scientists studying a genome to start from "islands of knowledge", with genes with a known background, reflected by a known gene (and a gene name has usually been coined by experimentalists for that gene). For this reason, we decided to use a conservative approach, using bidirectional best Blast hits of the genome of interest with model genome ( Escherichia coli K12 and Bacillus subtilis 168). Orthologues were identified as reciprocal best hits [ 16 ] (using a global alignment where the gaps on the edges of the largest sequence are ignored) with at least 50% identity in amino acid sequence and less than 20% difference in protein length. When possible, in order to increase the likelihood of the putative identification we used a second well known representative of the genome under study and looked for orthologues between every pair of each of the two triplets (i.e. between each pair of the three organisms: the organism of which the database is constructed, B. subtilis for Firmicutes and another organism of the same family, such as Listeria monocytogenes , and E. coli for gamma-proteobacteria, with another one of the same family, such as Photorhabdus luminescens ). Finding putative orthologues in the three organisms was considered as substantiating evidence for the use of a gene name. Then, in each triplet, we did not transfer the model organism gene name to all orthologues that were not simultaneously present in the three genomes or that gave different correspondences in different comparisons. In another comparison where the orthologues were found with at least 50% similarity, the model organism gene names to be transferred were preceded by the letter ' y '. In order to help users recognize gene names (and all the knowledge they associate with those names) we used as reference names those in the model bacteria, trying to comply as much as possible with the names used at SwissProt in the HAMAP project [ 17 ]. This allows the users to have "anchor" points to start to use the databases in a more efficient way. Naturally, the names previously used in the corresponding genome programs are kept as synonyms, so that access to the sequences with these names is still allowed. For example, in AeruList, gene rpsA can be accessed directly or using its synonym PA3162 : it is then found downstream of cmk (a context similar to that found in many Gram negative bacteria) and upstream of himD . We are aware that some erroneous identification (or propagation of erroneous identifications) must have occurred in some cases, but we think that this is a trade-off (which will be continuously corrected) for a much more user-friendly usage of the databases. A ' y ' letter starting a gene name indicates that it has not been experimentally identified, nor convincingly identified after in silico analysis yet. We provide curation pages (see below) to help users to correct annotation errors and improve annotation in a continuous way. Functional categories and bioprocesses An important feature for allowing users to explore biological functions is to investigate the genes neighborhoods [ 12 ]. Related functions are often coded by genes in close vicinity in the chromosome. We therefore used the GenoList table for functional categories, that allows the user to make links with the roles of proteins in the cell. The functional classification used in some of the present databases has been created by superimposing the functional classification (ontology) created for SubtiList, and that of Escherichia coli created by Monica Riley and her collaborators [ 18 ] (Additional file). In addition we created a field for the ontology describing underlying bioprocesses: explore, sense, shape, circulate, excrete, replicate, grow, respire, manage energy, store, scavenge, maintain, protect, control. They will be used in the future to color the arrows indicating genes in the picture of the region surrounding a gene of interest, allowing the user, at a glance, to have a rough idea of the processes encoded in the corresponding region. Queries using mining algorithms In addition to using keyword queries or sequence tags (such as molecular mass or isoelectric point of a protein) the database provides a versatile way to identify sequences from the biological knowledge viewpoint. In particular, as in many other databases, it allows the user to use Fasta, BlastP and BlastN to compare a sequence of interest to that of those present in the database. Furthermore, in contrast to most cases, it allows the user to extract information using motifs, that can be either continuous or discontinuous (e.g. finding all proteins with motif CXXCHX 12–25 C). This facility has already, in a quite unobtrusive but efficient way, permitted discovery of many unexpected functions. We have also provided means to explore the beginning and the end of protein sequences, as well as DNA regions upstream of putative operons, computed as strings of genes transcribed in the same orientation and separated by a maximum number of nucleotides (60 nt by default). Automatic updates Genome annotation is continuously updated by scientists all over the world, at a time when new genome sequences appear every three days or so. In order to cope with this enormous flux of information, a facility for browsing automatically new entries in major data libraries has been implemented. In the gene information panel, where each gene of interest is described after being identified as the result of a query (including resulting from a Blast or Pattern search), an "Automatic Blast" link provides a list of updated blast searches against the UniProt library (SWISSPROT+TREMBL). In addition, when the genome belongs to the 'Teraprot' Smith and Waterman Z-score family , the corresponding links (that are statistically much more significant than the results of Blast searches) are provided, allowing the user to look for remote kinships. To discuss the use of the databases we shall restrict our exploration to two databases from the package. LeptoList, that comprises two genomes (each one having two chromosomes) for Bacteria, and CunicuList, that describes the genome sequence and annotation of a small eukaryote. An example: LeptoList LeptoList is the reference database dedicated to the genome of Leptospira interrogans serovar Lai, the paradigm of leptospirosis causative agents [ 19 ]. It is presented together with a second sequence, that of L. interrogans serovar Copenhageni in order to allow easy comparison [ 20 ]. The WWW interface takes into account the fact that L. interrogans has two chromosomes (this feature was not yet displayed in GenoList databases). Using the regular comparison to the CDS to the non-redundant INSD protein database allowed us to suspect that a significant proportion of the short putative CDSs in the genome are likely ORFs and not authentic CDSs. This fits with the recent sequencing of the second Leptospira genome [ 20 , 21 ]. A couple of examples of its use are given here. We looked for counterparts of RRF, the ribosome release factor. In order to find the gene we used a known sequence, from B. subtilis ( frr gene product) and compared it using BLAST with the functionality implemented in LeptoList. This search led to a single gene, LA3295, located downstream of gene pyrH (as in most other bacterial genomes). This synteny is obviously highly significant. In the same way, the gene immediately upstream from pyrH (LA3297), as in other bacteria, is likely to be coding for elongation factor EFTs ( tsf ). When curating the database, we suggest to the curator that it would be of excellent policy to replace the gene numbers by the corresponding gene name. In another type of investigation, looking for patterns of the type TTGACA (1 ambiguity) – 17 nt – TATAAT# (1 ambiguity) (consensus sequence of the σ 70 -type promoter) in the 300 nt region upstream of genes revealed 70 sequences in chromosome I, many of which are likely to be promoters (at least they would be good guesses to start investigating promoters in L. interrogans ). In the same way, the putative DNA binding site located in the 300 nt nucleotide region upstream of genes, TGTGA (1 ambiguity) – 2 nt – KK – 2 nt – TCACA (1 ambiguity) (consensus sequence of the CAP/FNR family of transcriptional regulators), yielded 130 matches in chromosome I of serovar Lai and 72 matches in serovar Copenhageni and 2 in chromosome II of serovar Lai and 0 in serovar Copenhageni, allowing one to start investigating possible regulatory elements. This result is interesting as it suggests that chromosome I genes are submitted to a regulation recognizing that particular DNA-protein binding site. Furthermore, most genes found with the site in serovar Copenhageni are also found in serovar Lai, with sometimes several repeats in the latter, occuring upstream of some genes (such as fadH or prfC ), accounting for the higher total number of putative binding sites in that organism. It seems most interesting that genes involved in the control of respiration (cytochrome c biosynthesis), control of the TCA cycle (pyruvate dehydrogenase synthesis), control of the coupling between translation and transcription (stringent control) or translation itself (release factor 3 synthesis) are present in the list. While there are several putative adenylyl cyclase genes present in the organism, as well as several homologs of crp , it is plausible to propose that cAMP plays an important role in the life cycle of L. interrogans , perhaps suggesting ways to allow multiplication on plates of this elusive organism. LeptoList is accessible at the URL CunicuList: a database for a small eukaryote genome The GenoList structure has been initially constructed for organizing sequence data from prokaryotic genomes. However it may be extended to other organisms as well (the "genomic object" type must be extended accordingly). We have therefore tested the implementation of the structure for the genome of Encephalitozoon cuniculi , belonging to the Microsporidia taxon. Eleven chromosomes are present in this organism. Extraction of information is similar to that from other databases. For example we looked for counterparts of genes involved in tRNA modification (often essential genes). Using MesJ (TilS) [ 22 ] as well as TrmU [ 23 ] we found that gene Ecu03_1240 is most probably involved in driving the codon and amino acid specificity of a tRNA (possibly isoleucine or lysine tRNA). In the same way we could predict that gene Ecu07_1610 codes for synthesis of dihydrouridine in tRNA, a general feature of tRNA structure, because of its similarity with the yacF B. subtilis gene. Looking for counterparts of genes in the methionine salvage pathway [ 24 ], we failed to identify any gene that would code for the enzymes of the pathway, indicating that the parasite obtains all the metabolites derived from S-adenosylmethionine from its host. This is substantiated by the fact that the genes needed to synthesize queuosine [ 25 ] are apparently absent from the genome. Some organisms do not use this major tRNA modification, but this could be an interesting information for identification of drug targets against the parasite, since this suggests that those metabolites have to be transported into the cell by specific permeases. Database curation Several other similar bacterial databases are accessible at URL . Table 1 presents the list of microbial databases that are available at the Bioinfo server of the University of Hong Kong. Table 1 List of databases present at the Bioinfo server The GenoChore suite presented here manage bacterial genome data, except for CunicuList, which presents the sequence and annotation data of the small eukaryote Encephalitozoon cuniculi . AeruList Pseudomonas aeruginosa PA01 EMBL:AE004091 AnthraList Bacillus anthracis str. Ames EMBL:AE016879 CampyloList Campylobacter jejuni NCTC 11168 EMBL:AL111168 CereList Bacillus cereus ATCC 14579 EMBL:AE016877 CholeList Vibrio cholerae EMBL:AE003852 , EMBL:AE003853 CoeliList Streptomyces coelicolor A3(2) EMBL:AL645882 DiphteList Corynebacterium diphtheriae NCTC 13129 EMBL:BX248353 CunicuList Encephalitozoon cuniculi EMBL:AL391737 , EMBL:AL590442 , EMBL:AL590443 , EMBL:AL590444 , EMBL:AL590445 , EMBL:AL590446 , EMBL:AL590447 , EMBL:AL590448 , EMBL:AL590449 , EMBL:AL590450 , EMBL:AL590451 InfluList Haemophilus influenzae Rd KW20 EMBL:L42023 LeptoList Leptospira interrogans Lai str. 56601 EMBL:AE010300 , EMBL:AE010301 Leptospira interrogans Fiocruz L1-130 EMBL:AE016823 , EMBL:AE016824 MeningoList Neisseria meningitidis MC58 EMBL:AE002098 PutidaList Pseudomonas putida KT2440 EMBL:AE015451 SepiList Staphylococcus epidermidis ATCC 12228 EMBL:AE015929 SubtiList Bacillus subtilis str. 168 EMBL:AL009126 ThermaList Thermoanaerobacter tencongensis MB4 EMBL:AE008691 VulnifiList Vibrio vulnificus YJ016 EMBL:BA000037 , EMBL:BA000038 XylelList Xylella fastidiosa 9a5c EMBL:AE003849 Despite of – or because of – the large number of genome programs, once a sequence has been deposited at the INSD, its annotation is seldom updated. This is because the cost of curating annotations is extremely high, and usually not considered, despite its enormous importance. One of our aims was therefore to allow curation by selected teams by creating a curator page where such teams would input their annotations, that would then be propagated to the databases. The basic schema of the curator interface is shown in Figure 2 . In order to preserve the quality of the input data, potential users are asked to write to the database's webmaster to ask for account and passwords. We kept the interface of GenoList as published, except that a box providing access to protected curation of annotations is now provided, aiming initially at helping the first party (sequencing teams) annotators. If this works to our satisfaction this will be extended to selected third party annotators. Subsequently, on a yearly basis (or more frequently if needed) the collected re-annotation of the curators would be submitted as a new version of the same genome to the INSD. We hope that this service will be useful for the scientific community as a whole. Conclusions A set of 17 specialized end-user-oriented microbial databases (including one instance of Microsporidia) has been implemented in Hong Kong. They allow one to browse genome sequence and annotation data using the most frequent queries that end-users would like to ask. In addition they provide a weekly update of searches against the world-wide protein sequences data libraries, allowing one to monitor annotation on genes of interest. Finally, they allow users to search for patterns in DNA or protein sequences present in the databases. All comments, bug reports and suggestions for improvement are more than welcome: this work is meant to be useful for the community of microbiologists interested in genomics. Competing interests The author(s) declare that they have no competing interests. Authors' contributions GF wrote the parsers used to create the preliminary C/C++ MySQL package, and created important sections of the Perl package; CH created the procedure for renaming orthologs with reference to accepted names for model bacteria ( Bacillus subtilis and Escherichia coli ), created the link to Teraprot for identification of gene functions, and implemented parts of the PERL package; YQ implemented parts of the Perl package; CC and VC implemented most of the databases into the core structure; ZY wrote part of the new parsers, implemented the two chromosomes of LeptoList by changing the data structure in the database and set up with CC the first LeptoList database; FC set up and administered the Apache web server and MySQL database; IM over the years designed most of the GenoList data schema and user interface; AD was at the origin of the project, participated in the design and evolution of the data schema, was the systematic tester and end-user and wrote the core of the article. Supplementary Material Additional File 1 Functional categories. The genes' roles are listed into six major categories. The three first ones are directly linked to biological roles, while the remaining categories are created ad hoc : adaptation to atypical conditions correspond to miscellaneous roles, while the two last categories correspond to roles that have not yet been ascribed to genes because of lack of in vivo or in silico data Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549560.xml

528834

Phylogenetic inference in Rafflesiales: the influence of rate heterogeneity and horizontal gene transfer

Background The phylogenetic relationships among the holoparasites of Rafflesiales have remained enigmatic for over a century. Recent molecular phylogenetic studies using the mitochondrial matR gene placed Rafflesia , Rhizanthes and Sapria (Rafflesiaceae s. str.) in the angiosperm order Malpighiales and Mitrastema (Mitrastemonaceae) in Ericales. These phylogenetic studies did not, however, sample two additional groups traditionally classified within Rafflesiales (Apodantheaceae and Cytinaceae). Here we provide molecular phylogenetic evidence using DNA sequence data from mitochondrial and nuclear genes for representatives of all genera in Rafflesiales. Results Our analyses indicate that the phylogenetic affinities of the large-flowered clade and Mitrastema , ascertained using mitochondrial matR , are congruent with results from nuclear SSU rDNA when these data are analyzed using maximum likelihood and Bayesian methods. The relationship of Cytinaceae to Malvales was recovered in all analyses. Relationships between Apodanthaceae and photosynthetic angiosperms varied depending upon the data partition: Malvales (3-gene), Cucurbitales ( matR ) or Fabales ( atp1 ). The latter incongruencies suggest that horizontal gene transfer (HGT) may be affecting the mitochondrial gene topologies. The lack of association between Mitrastema and Ericales using atp1 is suggestive of HGT, but greater sampling within eudicots is needed to test this hypothesis further. Conclusions Rafflesiales are not monophyletic but composed of three or four independent lineages (families): Rafflesiaceae, Mitrastemonaceae, Apodanthaceae and Cytinaceae. Long-branch attraction appears to be misleading parsimony analyses of nuclear small-subunit rDNA data, but model-based methods (maximum likelihood and Bayesian analyses) recover a topology that is congruent with the mitochondrial matR gene tree, thus providing compelling evidence for organismal relationships. Horizontal gene transfer appears to be influencing only some taxa and some mitochondrial genes, thus indicating that the process is acting at the single gene (not whole genome) level.

Background Combining gene sequences from multiple subcellular compartments continues to provide increasingly well-resolved flowering plant phylogenies [ 1 ] and these have precipitated a new classification for angiosperms [ 2 ]. Whereas most groups have been placed at the ordinal level, seven of the 18 "taxa of uncertain position" are holoparasitic, nonphotosynthetic flowering plants. These parasites have been difficult to ally with green plants owing to extreme reduction and/or loss of morphological features [ 3 ]. Chloroplast genes commonly used to infer land plant phylogenetic relationships either show elevated substitution rates or are absent in these holoparasites [ 3 - 5 ]. Moreover, nuclear ribosomal genes also show greatly increased rates [ 6 ], thus analytical methods that accommodate such among-lineage rate heterogeneity must be used. Rafflesiales are a fascinating and enigmatic group of holoparasitic plants that includes Rafflesia , whose meter-wide flowers are the largest among all angiosperms, and Pilostyles , whose flowers are less than a centimeter in diameter. Such wide morphological variation has resulted in classifications that comprise four families: 1) the "small-flowered clade" (Apodanthaceae) with Apodanthes , Berlinianche , and Pilostyles , 2) the "large-flowered clade" (Rafflesiaceae s. str.) with Rafflesia , Rhizanthes , and Sapria , 3) the "inflorescence clade" (Cytinaceae) with Bdallophyton and Cytinus , and 4) the "hypogynous clade" (Mitrastemonaceae) with Mitrastema [ 7 , 8 ]. Recently, Barkman et al. [ 9 ] used DNA sequences of the mitochondrial gene matR to identify the closest photosynthetic relatives of two clades within Rafflesiales. Three genera, representing two of the four families in the order, were used in that study: Rafflesia and Rhizanthes (Rafflesiaceae s. str.) and Mitrastema (Mitrastemonaceae). Analyses of the matR data placed Rafflesiaceae s. str. within Malpighiales, an order that includes passionflowers ( Passiflora ), willow ( Salix ), and violet ( Viola ). Mitrastemonaceae was placed within Ericales, an order containing blueberries ( Vaccinium ), primroses ( Primula ), and tea ( Camellia ). The authors argued that these results were robust because they were congruent using different analytical methods (parsimony, neighbor-joining, Bayesian) and were not affected by long-branch attraction artifacts [ 10 ]. Moreover, because sequences from host plant lineages were included, and the parasites did not emerge as sister to these lineages, contamination and horizontal gene transfer (HGT) were discounted. In this study we expand upon the previous analysis [ 9 ] by including representatives of all Rafflesiales genera and families, thus allowing us to address the question of monophyly of the order. Moreover, parsimony, likelihood and Bayesian analyses were conducted on genes derived from all three subcellular compartments. These results were compared to assess the impact of artifacts such as long-branch attraction and HGT on various relationships. The data sets used were 1) mitochondrial matR , 2) mitochondrial atp1 and 3) a "3-gene" data set consisting of nuclear SSU rDNA plus two chloroplast genes: rbcL and atpB (the latter two only from nonparasites). Results Maximum likelihood (ML), maximum parsimony (MP) and Bayesian inference (BI) analyses of mitochondrial matR resulted in trees congruent with each other and with those previously generated [ 9 ] (Figure 1 and additional data file 1 ). As shown on the ML tree (Figure 1 ), Rafflesia , Rhizanthes , and Sapria were placed with strong support in Malpighiales. Mitrastema was placed in Ericales sister to Vaccinium . The Cytinus and Bdallophyton clade (Cytinaceae) was strongly supported and this clade was sister to one composed of four genera of Malvales, an order that contains cotton ( Gossypium ), rockrose ( Cistus ) and chocolate ( Theobroma ). For Apodanthaceae, Apodanthes and Pilostyles were sister taxa and derived from within Cucurbitales, an order that contains squash/pumpkin ( Cucurbita ) and Begonia . For Berlinianche , sequences homologous to matR could not be obtained using several primer combinations. Figure 1 ML strict consensus tree from mitochondrial matR . Strict consensus of two trees obtained from ML analysis of the 77-taxon mitochondrial matR matrix. Clades with Bayesian posterior probabilities between 0.9 and 1.0 are indicated by thick lines. Bootstrap percentages from MP analysis shown above lines. Rafflesiales taxa are shown in bold italics. Arrow represents a putative cases of horizontal gene transfer. The small phylogram is included to demonstrate branch length heterogeneity. All three analytical methods of the atp1 data produced trees that were generally congruent, thus the ML tree is illustrative (Figure 2 , additional data file 2 ). Clades among the monosulcates generally follow previously reported relationships, whereas the topology of the eudicot portion of the tree does not clearly reflect accepted clades, possibly owing to poor sampling within rosids and asterids (sequences for these taxa were not available from GenBank). Despite these shortcomings, this gene provides additional evidence useful in assessing the phylogeny and molecular evolution of Rafflesiales. With all three analytical methods, Mitrastema forms a clade with Beta (Caryophyllales), although this relationship does not receive strong support. This is remarkable given that 15 taxa from Ericales were included, yet a relationship with this order (as seen with matR ) was not obtained with atp1 . The large-flowered clade was strongly supported as monophyletic in all analyses, however, its position within the eudicots did not receive strong support. Parsimony analysis placed Pilostyles as sister to Pisum (Fabales) and this clade was sister to Berlinianche , but both with low bootstrap support. Apodanthes was strongly suported (90% bootstrap) as sister to Polemonium (Ericales) with MP but with ML this long-branch clade received lower support (Figure 2 ). The two genera of Cytinaceae, Cytinus and Bdallophyton , were sister to Malvales, with moderate (MP) to strong (BI) support. Figure 2 ML tree from mitochondrial atp1 . Phylogram obtained from ML analysis of the 71-taxon mitochondrial atp1 matrix. Clades with Bayesian posterior probabilities between 0.9 and 1.0 are indicated by thick lines. Rafflesiales taxa are shown in bold italics. Note that the clade with Apodanthes and Polemonium (asterisk) is poorly supported with a posterior probability of 0.54. Maximum parsimony analyses of the full-length (103 taxon) and reduced (77 taxon) 3-gene matrices were generally congruent and both resulted in all taxa of Rafflesiales being associated with Malvales (Figure 3 ), although with low bootstrap support for the monophyly of this clade. The two accessions of Pilostyles were sister to a clade composed of Pavonia and Gossypium , also with low bootstrap support. In constrast, BI analysis of the 3-gene matrix placed Mitrastema with Ericales and the large-flowered clade was a component of Malpighiales, the latter with strong support. The inflorescence clade ( Cytinus and Bdallophyton ) and the small-flowered clade ( Pilostyles ) were allied with Malvales (see additional data file 3 ), although posterior probablilities of this association were lower. Figure 3 Unconstrained MP tree from the 3-gene data matrix. Strict consensus of 12 trees obtained from an unconstrained maximum parsimony analysis of the 77-taxon "3-gene" matrix (nuclear SSU rDNA, rbcL , atpB ). Bootstrap support is shown above the lines. Rafflesiales taxa are shown in bold italics. Parsimony analysis of the nuclear SSU rDNA matrix, constrained to an accepted topology for nonparasites, showed the same pattern of relationships as the unconstrained 3-gene MP analysis, i.e., all Rafflesiales taxa were associated with Malvales (see additional data file 4 ). In contrast, the tree (Figure 4 ) resulting from ML analysis using the same constraint tree showed the same relationships as the BI tree for the 3-gene data set. Figure 4 Constrained ML tree from nuclear SSU rDNA. Tree resulting from the constrained ML analysis of the 77-taxon nuclear SSU rDNA matrix. Rafflesiales taxa are shown in bold italics. None of the consensus trees generated from MP analysis of the 100 nuclear SSU rDNA data sets simulated on 20-taxon trees matched the topology of the model tree. 58 of the 100 MP consensus trees showed a Mitrastema + Rafflesia/Rhizanthes/Sapria clade and 17 showed a Bdallophyton/Cytinus + Rafflesia/Rhizanthes/Sapria clade (Figure 5 ). Two other combinations, Bdallophyton/Cytinus + Pilostyles and Bdallophyton/Cytinus + Mitrastema + Rafflesia/Rhizanthes/Sapria accounted for 6% and 2% of the MP consensus trees, respectively. Thus, 83% of the MP trees contained incorrect clades, and most of these can be attributed to the long-branch Rafflesia clade. However, only two of the 100 MP trees showed all six long-branch taxa as monophyletic, a result seen on the original MP tree for the full 77-taxon data set. Results of parsimony analyses of data sets simulated on the full 77-taxon tree showed a similar pattern – 58 of the MP consensus trees showed a Mitrastema + Rafflesia/Rhizanthes/Sapria clade, 7 showed a Bdallophyton/Cytinus + Rafflesia/Rhizanthes/Sapria clade, and 14 showed a Bdallophyton/Cytinus + Pilostyles clade (Figure 5 ). In other words, MP returned an incorrect "long-branch" clade for 79% of the data sets simulated on the full 77-taxon model tree. In contrast, far fewer incorrect long-branch clades were recovered by ML for the 20-taxon simulations, and most (56%) ML trees matched the model tree in that the Rafflesia clade was sister to Passiflora , Mitrastema was sister to Helianthus/Nicotiana , and Pilostyles , Bdallophyton and Cytinus were associated with Gossypium . Figure 5 Rafflesiales long branches mislead MP. Proportion of simulated data sets (replicates) for which incorrect "long-branch" clades are recovered in maximum parsimony (black bars, 77 taxa), maximum parsimony (grey bars, 20 taxa), and maximum likelihood (open bars, 20 taxa) analyses. Inset is the model tree used to generate the simulated data sets. M = Mitrastema , B = Bdallophyton + Cytinus , R = Rafflesia + Rhizanthes + Sapria , P = Pilostyles . MP analyses of SSU data sets from which all but one parasite group had been removed resulted in phylogenetic placements that matched those found in the ML tree. MP analysis of a data set from which all Rafflesiales except Mitrastema had been removed resulted in trees that placed Mitrastema in Ericales. Removal of all parasites except Pilostyles or Bdallophyton + Cytinus individually placed both of these groups in Malvales. Finally, removal of all parasites except the large-flowered clade ( Rafflesia , Rhizanthes and Sapria ) placed this clade in Malpighiales. Thus, the positions of the parasite clades inferred in four separate MP analyses matched the positions found for these clades in the single ML tree. Discussion Rate heterogeneity and long-branch attraction artifacts Determining the photosynthetic relatives of Rafflesiales has long presented a challenge owing to the extreme reduction and/or modification of morphological structures that have accompanied the evolution of this lineage [ 3 , 11 ]. Molecular phylogenetic approaches, although providing great promise in resolving such questions, also come with their own set of challenges that includes losses of some genes, substitution rate increases in other genes, and horizontal gene transfer. Examples of the first process can be seen in chloroplast genes such as rbcL that are typically used to infer phylogenetic relationships among angiosperms but have not yet been amplified from any Rafflesiales and are presumed lost [ 5 ]. Increased substitution rates in the normally conservative plastid rDNA has been demonstrated in these holoparasites [ 4 , 12 ]. Similarly, accelerated rates in mitochondrial SSU rDNA, typically very conservative in many photosynthetic angiosperms, occur in Rafflesia and Cytinus [ 13 ]. Despite these complications, molecular phylogenetic analyses of some holoparasite lineages with comparatively lower rates have been tractable. For example, the mitochondrial genes atp1 and matR were used, in combination with nuclear rDNA and chloroplast genes, to reliably place Hydnoraceae with Aristolochiaceae [ 11 ]. Long-branch attraction, a bias in certain phylogenetic inference methods in which similarity due to convergent or parallel changes produces an erroneous phylogenetic grouping of taxa [ 10 ], is often implicated as the reason for anomalous phylogenetic groupings [ 14 ]. It has been suggested that some data sets with marked among-lineage rate heterogeneity cannot be applied to particular phylogenetic problems owing to hypothesized long-branch attraction artifacts [ 15 ]. In their unconstrained parsimony analysis of several angiosperm SSU rDNA sequences, Barkman et al. [ 9 ] found that the branch leading to Rafflesia was several times longer than any other branch, and that this branch was attracted to the second-longest branch in the tree – the one between gymnosperms and angiosperms. For these reasons, they argued that nuclear SSU rDNA sequences are of limited utility for assessing the phylogenetic position of Rafflesia . Barkman et al. [ 9 ] analyzed their SSU rDNA data using only parsimony, not model-based methods (e.g., ML or BI methods) that are less likely to be misled by long-branch attraction [ 16 ]. Our ML analysis of the SSU rDNA data recovers a topology that closely matches the matR topology presented by Barkman et al. [ 9 ] in which Rafflesia is closely related to Malpighiales and Mitrastema is a member of Ericales (Figure 4 ). These results highlight the requirement to analyze SSU rDNA data with methods less biased by long-branch attraction than parsimony, as well as the advantage gained by independent confirmation of results obtained from a single gene. Several authors have suggested that adding taxa can "break up" long branches and allow parsimony to recover the correct topology [ 17 - 19 ]. Our parsimony analysis of the 103- and 77-taxon SSU rDNA data sets, in which we included representatives of all genera of Rafflesiales (i.e., sequences that could potentially break the Rafflesia long branch), recovers a nearly monophyletic Rafflesiales containing all of the longest terminal branches in the tree (see additional data file 3 ). Based on our simulation study and MP analyses of data sets from which all but one parasite group was removed, we believe that this topology represents a case of long-branch attraction. These simulation results support the contention that the branches leading to the parasitic taxa are long enough to attract one another (Figure 5 ), a result in agreement with previous work [ 3 , 6 ]. Taxon sampling is not a cure-all for long-branch attraction problems [ 20 ]. Even for the data sets simulated on the full 77-taxon tree, MP returned incorrect long-branch clades nearly 80% of the time. MP did nearly as poorly with data sets simulated on a 77-taxon tree as it did on data sets simulated on a 20-taxon tree. Evaluation of the ML tree for the SSU data (Figure 4 ) shows that increasing the number of taxa from 20 to 77 did not improve the result because the long parasite branches were not broken. Instead, shorter (nonparasite) branches were broken which did not help MP recover the true topology for the simulated data sets. MP analyses of the full 77-taxon SSU data set that included all parasite clades resulted in a worse estimate of the phylogeny than MP analyses of smaller data sets in which only single parasite clades were included. Thus, the frequently stated view that increased taxon sampling can help MP avoid long-branch attraction problems may only be true if the added taxa are not distantly related long-branch clades themselves. Phylogenetic relationships of the four Rafflesiales clades Rafflesiaceae (the large-flowered clade) The results from analyses of Rafflesiales using independent data sets are summarized in Table 1 . For Rafflesiaceae s. str., placement in Malpighiales is supported by ML and BI analyses of the 3-gene and nuclear SSU rDNA data sets as well as mitochondrial matR . This placement in Malpighiales is also supported by a molecular phylogenetic study that used a single copy nuclear gene phytochrome C [ 21 ]. These authors proposed that Rafflesiaceae are most closely related to Ochnaceae or Clusiaceae which contrasts with presumed synapomorphies with Passiflora given by Barkman et al. [ 9 ]. Within Malpighiales, tremendous morphological diversity exists among the 27 families and 16,000 species. Moreover, relationships among the major clades are still poorly resolved [ 22 ]. Although the evidence for a malpighialean affinity of Rafflesiaceae appears strong, it is possible that the molecular data have only identified the stem group that represents the sister to the parasitic lineage. Table 1 Summary of phylogenetic analyses of Rafflesiales using different data partitions and methods of analysis. 3-Gene* 3-Gene nuSSU rDNA nuSSU rDNA matR matR atp1 atp1 Parsimony Bayesian Parsimony constrained Likelihood constrained Parsimony Likelihood & Bayesian Parsimony Likelihood & Bayesian Mitrastema Malvales Ericales Malvales Ericales Ericales Ericales Caryophyllales Caryophyllales Cytinus Malvales Malvales Malvales Malvales Malvales Malvales Malvales Malvales Bdallophyton Malvales Malvales Malvales Malvales Malvales Malvales Malvales Malvales Apodanthes N/A N/A N/A N/A Cucurbitales Cucurbitales Polemonium Polemonium Pilostyles Malvales Malvales Malvales Malvales Cucurbitales Cucurbitales Fabales Fabales Berlinianche N/A N/A N/A N/A N/A N/A Ericales/Fabales Ericales/Fabales Rafflesia Malvales Malpighiales Malvales Malpighiales Malpighiales Malpighiales Eudicots Eudicots Rhizanthes Malvales Malpighiales Malvales Malpighiales Malpighiales Malpighiales Eudicots Eudicots Sapria Malvales Malpighiales Malvales Malpighiales Malpighiales Malpighiales Eudicots Eudicots *Nuclear SSU rDNA plus chloroplast rbcL & atpB . Possible HGT events Long-branch artifact Barkman et al. [ 9 ] suggested that the floral similarities between Rafflesia and Passiflora , first noted by Robert Brown [ 23 ] represent morphological synapomorphies that support the results obtained from the matR gene tree. Arguments in favor of a number of other, equally credible relationships within eudicots could be made based on hypothetical evolutionary transformation series of morphological characters. Indeed Brown concluded that Rafflesia may have affinity with Passifloraceae (Malpighiales) but he also considered other groups such as Aristolochiaceae ("Asarinae", Piperales), Sterculiaceae (Malvales) and Cucurbitaceae (Cucurbitales). In general, different characters supported relationships with one or another group and therefore he left the subject as unresolved. Three proposed synapomorphies between Passifloraceae and Rafflesia were cited by Barkman et al. [ 9 ]: a hypanthium (perigone tube in Rafflesia ), an androgynophore (gynostemium or column in Rafflesia ), and an annular corona (diaphragm in Rafflesia ). Whether these structures are homologous is not clear and will likely require further morphological studies, possibly examining the floral development genes themselves. These hypotheses require scrutiny because the apparent similarities in structure are not clear when examined in detail. For example, the androgynophore of Passiflora is composed of a stalk that bears the androecium and gynoecium. In Rafflesia , the ovary is inferior (with no stalk), hence the central column must involve other gynoecial parts. The corona of Passiflora is very different in structure and function from the diaphragm of Rafflesia [ 24 ]. The observation of a physical union between Passiflora caerulea and Euonymus [ 25 ] was discussed by Barkman et al. [ 9 ] as a possible clue to the origin of parasitism in Rafflesia . Whether this association represents parasitism or not is a matter of semantics [ 26 ], for other similar associations exist such as Cissus and Opuntia growing on Yucca and Opuntia on Cercidium and Idria . In all of these cases, a true haustorium does not form and more likely these represent forms of grafting. It is difficult to state whether such rare occurrences have any bearing on the origin of parasitism in Rafflesiales or other parasitic flowering plants. Mitrastemonaceae (the hypogynous clade) Maximum likelihood and Bayesian analyses of the 3-gene and nuclear SSU rDNA data partitions placed Mitrastema in Ericales, a result congruent with that obtained using mitochondrial matR . As noted by Barkman et al. [ 9 ], this relationship within the asterids had not previously been proposed. Mitrastema has bisexual, protandrous flowers with a collar-shaped, four-merous perianth tube. The stamens are connate into a tube (androphore) crowned by a fertile zone of pollen-bearing locules. The staminal tube, open at the top by a small hole, circumscissally separates from the flower as it is pushed up by the growing gynoecium. The apical portion of the staminal tube is sterile, but below this is a series of vertical rings of ca. ten minute, pollen sacs each. The gynoecium is hypogynous, one-locular, with a thick, conical stigma. Placentation is parietal with 8–15 (-20) unequal placental lobes filling the locule. The numerous ovules are small (190 by 120 μm), anatropous, unitegmic (but with two cell layers), and tenuinucellar. Although some floral morphological features of Mitrastema are not in conflict with those seen in Ericales, such as extrorse anthers and cellular endosperm, features such as decussate leaves, circumscissile fruit dehiscence, and parietal placentation are too general to draw specific associations. Given that Mitrastema is an achlorophyllous holoparasite and that one clade of Ericaceae (Monotropoideae) contains achlorophyllous mycotrophs, it is intriguing to ask whether these groups share a common ancestor or evolved independently. The most specialized morphological feature found in Mitrastemonaceae, the athecal androecium, is not found in Ericales but in Malvaceae, the only angiosperm family that shows the entire gamut from taxa with normal stamens, to taxa with stamens deviating only slightly from the common pattern [ 27 , 28 ], to athecal androecia [ 29 ]. Cytinaceae (the inflorescence clade) The most consistent phylogenetic signal that is seen across all data sets and types of analyses is a relationship between Cytinaceae and Malvales (Table 1 ). Because the relationship between Cytinaceae and Malvales is the strongest among all four Rafflesiales clades, it is possible that this clade acts as an "attracter" for the other three Rafflesiales clades in some analyses. This is seen when using nuclear SSU rDNA sequences, either alone or with the topology of the tree stabilized through the addition of two chloroplast genes. In both cases, parsimony produces a monophyletic Rafflesiales within Malvales which contrasts with the result seen with the constrained ML SSU rDNA and the matR results. These results and those obtained from the simulation study indicate that the large-flowered clade and Mitrastema are artifactually attracted to Cytinaceae when parsimony is utilized. Unlike other Rafflesiales, members of Cytinaceae have multiple flowers arranged in an inflorescence. The floral structure called the diaphragm, seen in Rafflesia and Sapria (but not Rhizanthes ), is lacking in Cytinaceae. Bdallophyton is dioecious and Cytinus is either dioecious ( C. capensis , C. sanguineus ) or monoecious ( C. hypocistis ). The perianth is tubular, composed of four to nine imbricate organs. The androecium is connate, forming a compact synandrium with extrorse anthers and the pollen is 2-, 3-, or 4-porate. The female flower is epigynous with a columnar style terminated by a globose or capitate, viscous stigma with commissural lobes [ 30 ]. The ovary is unilocular with 8–14 deeply intrusive, discrete parietal placentae that bear numerous, orthotropous, tenuinucellate ovules. Apodanthaceae (the small-flowered clade) Maximum parsimony and likelihood analyses of the 3-gene data set and nuclear SSU rDNA sequences alone also place Pilostyles (the only Apodanthaceae for which SSU rDNA sequences are available) within Malvales, however, a sister relationship with Cytinaceae is not consistently obtained. A 3-gene alignment that included additional representatives of Malvales (16 taxa) gave similar results as shown in Figure 3 (i.e., Pilostyles on a clade separate from other Rafflesiales). These data, in conjunction with the results from the mitochondrial genes, support an evolution of Apodanthaceae independent from Rafflesiaceae s. str. The well-supported relationship between Pilostyles and Apodanthes using matR is expected given their very similar floral morphology [ 31 ], yet this clade is sister to two representatives of Cucurbitales ( Begonia and Cucurbita ). Contamination with host tissue is excluded because neither parasite is known to currently occur on a member of Cucurbitales. Apodanthaceae are grouped with Pisum (Fabales) and Polemonium (Ericales) on the atp1 tree, but no atp1 sequences from representatives of Cucurbitales were available from GenBank to test the matR result. The sister relationship between Apodanthes and Polemonium is strongly supported on the MP tree (bootstrap support value = 90%; additional data file 3 ), but this pairing must be viewed with caution given the low Bayesian posterior probability of the clade (0.54) and that both taxa are very long branches (Figure 2 ). Although ML is less susceptible to long-branch attraction artifacts than MP, it is not immune to it; thus, it remains unclear whether or not this relationship is artifactual. Moreover, the Polemonium sequence is separate from the clade containing 12 other members of this order, thus raising the possibility that the sequence results from contamination or HGT (see below). Additional sampling within the eudicots will be required to better understand the atp1 gene tree topology. Morphological features shared between Apodanthaceae and Cytinaceae are: unisexual flowers, a connate androecium, an inferior ovary, and a unilocular ovary with four parietal placentae bearing numerous, anatropous, tenuinucellate ovules [ 30 , 31 ]. Floral morphological features that might link Apodanthaceae and Cytinaceae with Malvales [ 31 ] include an androecial tube (e.g., Malvaceae), a trend toward synandria without anthers and thecae (e.g., Malvaceae) [ 29 ], tri- to hexamerous flowers (e.g., Thymelaeaceae), and parietal placentae (e.g., Cistaceae). The floral conditions of unisexuality and epigyny do occur in Malvales, albeit rarely. Unisexual flowers pose some difficulties for interpreting the morphological homologies of various floral organs. For Pilostyles and Apodanthes male flowers, a tubular synandrium surrounds and fuses with a central structure that could be interpreted as a sterile gynoecium. Support for the concept that such a central structure is a pistillode comes from Berlinianche where the upper portion of the synandrium is free from the central part. In female flowers of Apodanthaceae, there is no rudiment of an androecium, hence the central tissue is apparently entirely gynoecial. In contrast to the above discussion, the matR data indicate Apodanthaceae are related to Cucurbitales, an order with seven families, 129 genera and 2300 species. Hosts for Apodanthaceae are generally legumes, although Apodanthes occurs most frequently on Casearia (Salicaceae, Malpighiales). Thus, neither recent HGT nor contamination explains this result. Apodanthaceae shares some morphological features with members of Cucurbitaceae, subfamily Cucurbitoideae: unisexual, five-merous flowers ( Berlinianche ); carpellate flower with a unilocular, inferior ovary with parietal placentation; anatropous, bitegmic ovules; staminate flower with connate filaments (monadelphous) and a rudimentary gynoecium (pistillode) [ 32 ]. Conflicting characters also occur, such as a three-carpellate gynoecium in Cucurbitoideae (vs. four-carpellate in Apodanthaceae) and a valvate perianth (vs. imbricate). All of these characters, however, are less specialized than those shared between Apodanthaceae and Malvales. Background on horizontal gene transfer A requirement of the molecular phylogenetic approach to inferring evolutionary histories of organisms is vertical transmission of genetic material from parent to offspring. In contrast, horizontal gene transfer (HGT) describes the movement of genetic material between organisms of no direct ancestor-descendant relationship. Although the frequency of HGT is currently not well understood among prokaryotic and eukaryotic organisms, it is clear that HGT can compromise accurate inference of genealogical history. In plants, lateral movement of genetic material has been documented for mobile genetic elements such as introns [ 33 - 37 ] but only recently has convincing evidence emerged documenting HGT of mitochondrial genes [ 38 , 39 ]. Genes of the mitochondrion are extensively used to infer evolutionary relationships in plants [ 40 - 42 ], thus highlighting the importance of characterizing the frequency of HGT across genes and taxa. Incongruence among gene trees derived from different data sets can derive from a number of factors such as technical causes (insufficient data, gene choice, sequencing error, taxon sampling and identification), gene/genome-level processes, and organism-level processes (e.g., hybridization/introgression, lineage sorting, and HGT) [ 43 ]. HGT has only recently been recognized as a potentially important force in the evolution of plant mitochondrial genomes and detecting HGT is highly dependent upon the presence of multiple gene data sets with robust taxon sampling [ 38 , 39 ]. Evidence for horizontal gene transfer in parasitic plants We believe that incongruence between the the mitochondrial and the nuclear gene trees (Table 1 ) stem not just from long-branch attraction artifacts but also from cases of HGT. The placement of Apodanthes and Pilostyles on the atp1 tree as sister to Pisum (a legume, the family of hosts for Pilostyles ) represents a likely case of HGT. The atp1 data conflict with those from matR that associates Apodanthaceae with Cucurbitales. Moreover, we infer that the SSU rDNA tree better represents the organismal phylogeny because it seems less likely that nuclear genes would be influenced by HGT [ 44 , 45 ]. The main rationale for this is that nuclear rDNA cistrons are repeated hundreds to thousands of times in tandem arrays at nucleolar organizing regions of the chromosomes. Although it can be envisioned that concerted evolution could homogenize all rDNAs in the parasite with a form obtained via HGT, the probability of this happening is small given the vastly different number of starting copies. In their study of Rafflesiaceae s. str. and Mitrastemonaceae, Barkman et al. [ 9 ] discounted HGT as a possible explanation for their results because they state the phenomenon is rare and the overall topology of the matR tree closely matched results from other molecular phylogenetic investigations of angiosperms. The present study confirms that HGT is not implicated for the two lineages studied by Barkman as well as Cytinaceae, but this process could be invoked for Apodanthaceae. More recent work by these authors [ 46 ] identified several cases of HGT from host to parasite for atp1 . These included Dalea to Pilostyles , Tetrastigma to Rafflesia , and Lithocarpus to Mitrastema . In addition, HGT of another mitochondrial gene, nad1 , has been reported for Rafflesia and Sapria , both of which occur on the same clade as their hosts ( Tetrastigma ) on a gene tree [ 20 ]. These examples demonstrating the presence of host genes in parasitic plants provide the most compelling evidence for HGT. This form of transfer is intuitively logical given the intimate contact between cells of the two organisms via the endophytic haustorium. However, parasitic plants exist in complex ecosystems where they are in physical contact with many other organisms (bacteria, fungi, phytophagous and pollinating animals, etc.) that could potentially affect HGT. That such nonhost HGT may also be occurring is evidenced by the presence of an apparent cucurbitalean matR gene in Pilostyles and Apodanthes . Moreover, present-day hosts of parasitic angiosperms do not represent the only conduit for HGT if host choice has shifted through time as the parasite lineage evolves. For example, Barkman et al. [ 9 ] state that Mitrastema only parasitizes Fagales (e.g., Lithocarpus and Castanopsis , both Fagaceae) but this parasite has also been recorded from Aquifoliaceae, Asteraceae, Elaeocarpaceae, Juglandaceae, and Myrtaceae [ 47 ]. Host latitude for this species would be broader if rare hosts and hosts of parasite ancestors were fully known, thus expanding the taxonomic spectrum of potential HGT sources. Formidable contamination issues Contamination of parasite DNA with DNA from the host plant is an issue that must be given serious attention. Indeed, two sequences shown on the matR tree (Figure 1 ), Tetrastigma2 and Julbernardia are hosts for Rafflesia tuan-mudae and Berlinianche , respectively. These sequences were obtained by PCR amplification and sequencing from what was originally thought to be pure parasite genomic DNA. Sequences of the host (obtained from separate samples) were found to be identical to these "parasite" sequences, strongly suggesting contamination. In the case of Rafflesia , the DNA was obtained from a bud still attached to the host vine, both of which had been sectioned longitudinally. Disruption of these tissues likely resulted in transfer of host sap to the bud region where the tissue was sampled. Other samples of R. tuan-mudae from the same population, obtained as floral bracts with no host tissue, resulted in matR sequences that were similar to the other two Rafflesia species. For Berlinianche , whose flowers are much smaller than those of Rafflesia (5 mm in diameter), extreme care (using a stereo microscope) was exercised to remove floral parts devoid of any host tissue. Despite this, the matR sequence obtained from the first sample was that of the host, Julbernardia . Later, silica gel dried samples of other populations of the parasite were extracted, again using extreme care in avoiding host contamination. PCR products were obtained using several mitochondrial matR primers, but none were found to be homologous to this gene following BLAST searches. This result shows that host DNA was not present in this sample in sufficient amounts to amplify and that the parasite matR gene, if present, is highly divergent at the priming sites used. For all three Apodanthaceae genera, the conical style in female flowers is papillate and heavily secretory [ 31 ]. This sticky surface tends to capture a variety of environmental debris, likely including extraneous pollen, fungal spores, and host tissues that have been disrupted upon collecting. Obtaining a proper nuclear SSU rDNA sequence for Pilostyles was extremely difficult. Despite PCR products of the correct sizes using a variety of primer combinations, the sequences obtained from genomic DNA derived from flowers were deemed contaminants following BLAST searches that showed them to be most similar to monocots, fungi, etc. Only when sequences from two accessions of Pilostyles (Texas and California) both were most similar to Malvales was this considered good evidence for their true phylogenetic affiliation. Retrospectively, it is likely that the sticky flowers had accumulated wind-dispersed pollen (e.g., grasses) and that this DNA, despite being in low concentration, had less divergent priming sites than the parasite target DNA, allowing PCR to preferentially amplify the contaminant DNA. The mechanism of horizontal gene transfer: some considerations Given the accumulating molecular evidence for HGT from host to parasitic plant, it is worthwhile to consider potential mechanisms, along with their constraints, that may suggest further research. Relatively little information exists on the structure of the endophyte of Rafflesiales. Ultrastructural studies have been conducted on two species of Pilostyles : P. hamiltonii [ 48 ] and P. thurberi [ 49 ]. These authors conflict, however, as to whether there exists symplastic continuity between host and parasite via plasmodesmata; the former indicated that such connections are the major path of nutrient uptake by the parasite whereas the latter rejected this idea. Despite this controversy, heteroplastic plasmodesmatal connections have been documented in another parasitic plant, Cuscuta [ 50 ] and indeed such connections can even form in heterografts between distantly related plant taxa [ 51 ]. Given this, we assume that host-parasite plasmodesmatal connections exist in the endophytes of Rafflesiales. Transmission electron micrographs of Pilostyles suggest that intact, mature mitochondria are too large to pass through heteroplastic plasmodesmata, however, mitochondrial genomes or portions of the genome are certainly small enough for transmission. Once inside the parasite cell, there are various fates for the host gene. It could become incorporated into the parasite mitochondrial genome, and then either replace the parasite copy or exist as a duplicate, or the host gene could reside in the parasite nuclear genome. For the latter case, the gene would likely become a pseudogene given the requirement of mitochondrial-specific patterns of RNA editing. Two forms of atp1 are present in the primitive angiosperm Amborella trichopoda [ 38 ], one of which is derived from a HGT event from a eudicot. It is not known whether both forms of the gene exist in a single mitochondrial genome, in different mitochondrial genomes within the cell (i.e., heteroplasmy), or if one is nuclear and the other mitochondrial. Future work to address these questions would involve sequencing flanking regions of purported horizontally transferred genes to determine their subcellular location. Additionally, cDNA sequences obtained from matR mRNA would be useful to determine whether the gene is expressed and whether mitochondrial-specific RNA editing patterns are present. Conclusions In this study we have used data derived from nuclear, mitochondrial and chloroplast DNA and a variety of analytical approaches to address long-standing questions about the holoparasitic flowering plant order Rafflesiales. We show that Rafflesiales are not monophyletic but composed of at least three and possibly four independent lineages. Rafflesiaceae ( Rafflesia , Rhizanthes , and Sapria ) representing the large-flowered clade are monophyletic and are related to Malpighiales. The monogeneric family Mitrastemonaceae, the only member of the order with a superior ovary, is related to Ericales. The first of the remaining two families that have previously not been sampled is Cytinaceae ( Bdallophyton and Cytinus ) which is strongly supported as a member of Malvales. The last remaining unsampled family, Apodanthaceae ( Apodanthes , Berlinianche , and Pilostyles ) is either related to Malvales or Cucurbitales. Our simulation studies indicate that Mitrastema , Bdallophyton/Cytinus , and Rafflesia/Rhizanthes/Sapria have branches that are long enough to mislead parsimony. All of these relatively long branches appear to be attracted toward the Cytinaceae clade within Malvales. When nuclear SSU rDNA sequences are analyzed with ML, results fully congruent with those previously reported for two Rafflesiales clades using mitochondrial matR are obtained. If the phylogenetic affinityof Apodanthaceae are with Malvales, the results from the mitochondrial matR gene must represent a case of horizontal gene transfer (HGT) from Cucurbitales. If this proves to be the case, this provides an example of HGT from a nonhost plant to a parasitic angiosperm. To properly discern phylogenetic relationships in enigmatic parasitic taxa, our results demonstrate the need for robust taxon sampling, gene sequences from multiple subcellular compartments, and the use of analytical methods that accommodate rate heterogeneity and avoid the pitfalls of long-branch attraction. When the phylogenetic relationships among such holoparasitic taxa are poorly known, the strongest phylogenetic signal that can be obtained is congruence among results derived from independent sources (i.e., genes from different subcellular compartments). Comparisons among gene trees allows for the identification of HGT, a phenomenon that requires further investigation to determine its modes of action and frequency among taxa and through evolutionary time. Methods DNA extraction, PCR, sequencing DNA was extracted, amplified, cloned, and sequenced by using methods formerly reported [ 52 - 54 ]. The nuclear and mitochondrial sequences were PCR-amplified using primers reported elsewhere [ 6 , 55 , 56 ] and are also given on the first author's web site [ 57 ]. Sequencing was conducted using manual and automated methods (ABI Prism ® 377 automated DNA sequencer, Applied Biosystems) according to manufacturer's protocols. DNA alignments The initial matR alignment incorporated all of the Rafflesiales parasites and the nonparasite sequences previously published [ 9 ] as well as our newly generated sequences. The 106-taxon matrix represented over 40 orders and contained three gymnosperm outgroup taxa ( Ginkgo , Pinus , and Zamia ), 28 monosulcates, 63 nonparasitic eudicots, and 15 Rafflesiales. For two taxa ( Mitrastema and Rhizanthes ), our sequences, as well as those previously published, were from the same species but different accessions to test for consistency. Taking into account codon information, an alignment of 2177 sites was constructed manually using SeAl version 2.0 [ 58 ]. The full matrix was used for parsimony analyses whereas another, truncated to 77 taxa by removing all but three monosulcate taxa (Laurales used as outgroup), was constructed to facilitate likelihood analyses. This operation was justified because monosulcates were never implicated as relatives of Rafflesiales in any analyses. A 71-taxon, 1265-site atp1 alignment was similarly constructed and included the same gymnosperm outgroup genera as above, 24 monosulcates, 32 nonparasitic eudicots and 12 Rafflesiales. All of the monosulcate genera in the atp1 alignment were also represented in the matR data set, whereas eudicot sampling for atp1 was constrained by sequences available on GenBank (12 of the same genera as with matR or placeholders from same family). To test the position obtained for Rafflesiales taxa using mitochondrial genes with an independent data set derived from different compartments, a 4646-site "3-gene" matrix combining sequences from nuclear SSU rDNA and chloroplast rbcL and atpB was constructed that included 103 taxa (3 gymnosperms, 28 monosulcates, 58 nonparasitic eudicots, and 14 Rafflesiales). Sampling across angiosperm orders was very similar to the matR matrix, differing only by the presence of 11 placeholders and a second accession of Pilostyles . For the holoparasites, only nuclear SSU rDNA sequences were included; the chloroplast gene data for these taxa were coded as missing. The two chloroplast genes were included to add stability to the tree topology given that nuclear SSU has been shown to contain lower phylogenetic signal when used alone [ 15 ]. As with matR , the 103-taxon matrix was truncated to 77 taxa by removing all but five monosulcate taxa to facilitate likelihood analyses. All alignments reported in this paper have been deposited with TreeBASE [ 59 ]: study accession number S1177, matrix accession numbers = M2034–M2037 . Data analysis All three data sets were analyzed using maximum parsimony (MP) and maximum likelihood (ML) methods in PAUP* 4.0b10 [ 60 ] and Bayesian inference (BI) methods in MrBayes 3.0b4 [ 61 ]. Maximum parsimony All MP searches were performed using 100 random addition sequence replicates with tree-bisection-reconnection (TBR) branch-swapping, holding ten trees at each addition step, with all sites equally weighted. For the 77-taxon SSU data set, a series of four MP analyses were performed in which all but one parasite group ( Bdallophyton + Cytinus , Mitrastema , Pilostyles or the large-flowered clade comprising Rafflesia , Rhizanthes and Sapria ) was removed to determine the position of each parasite group in the absence of other long-branch parasite taxa in the analysis. This is a form of the test proposed by Siddall and Whiting [ 62 ]. Maximum likelihood For ML analyses, a MP tree was used in PAUP* to evaluate 56 nucleotide substitution models. ModelTest 3.06 [ 63 ] was used to select an appropriate model from the PAUP* output using hierarchical likelihood-ratio tests (hLRT's) and the Akaike Information Criterion (AIC). The general time-reversible (GTR) substitution model with among-site rate heterogeneity modeled with a "gamma + invariant sites distribution" (Γ + I) was chosen via the AIC as the best-fitting model for the atp1 data set. Investigation of the likelihood score output from PAUP* suggested that a simpler model not evaluated by ModelTest was not significantly worse than the GTR+Γ + I model (LRT; p = 0.520824). This submodel employed four (rather than six) relative rate parameters: one for A-C transversions and A-G transitions, one for A-T and C-G transversions, one for C-T transitions, and one for G-T transversions; the PAUP* LSET option used for analysis was "RCLASS = (a a b b c d)". Likewise, the models chosen by ModelTest for the matR data set were TVM+Γ (hLRT) and TIM+Γ (AIC), but a simpler statistically equal model (LRT; p = 0.583393) was used for analysis. This model employed three relative rate parameters: one for A-C, A-G, and G-T substitutions; one for A-T and C-G substitutions; and one for C-T substitutions; "RCLASS = (a a b b c a)", with among-site rate heterogeneity modeled with a gamma distribution. These simplified models were chosen to reduce computational time and to avoid estimation of unnecessary parameters, which can lead to greater variance in parameter estimates and higher topological uncertainty. A successive approximations approach was used for all ML analyses [ 19 , 64 ]. Substitution model parameters were estimated from the data on a MP tree. With parameter estimates fixed, starting trees for ML analyses were produced via random stepwise addition using five starting seeds, with each tree subjected to a round of tree bisection-reconnection (TBR) branch swapping. Substitution model parameters were then re-estimated on all resulting trees, followed by another round of random stepwise addition and TBR swapping. The tree with the highest likelihood was accepted as the ML tree. Nodal support Nodal support for all data sets was estimated using one or more of the following methods: equal-weights MP bootstrap analysis (100 pseudoreplicates, each consisting of a heuristic search using 100 random sequence addition replicates), ML bootstrap analysis (100 pseudoreplicates generated with SEQBOOT in PHYLIP and analyzed using successive approximations in PAUP*) [ 65 , 66 ], and Bayesian analysis (10 million generations, with the first one or two million discarded as burn-in and trees sampled every 500 generations for the matR and atp 1 data sets; 10 million generations, with the first 5 million discarded as burn-in and trees sampled every 500 generations for the 3-gene data set) [ 61 ]. The GTR+Γ + I submodels used in PAUP* are not available in MrBayes; a standard GTR+Γ + I model was used for the matR and atp1 data sets instead. A partitioned model was used for the 3-gene data set (see below). Two Bayesian runs were performed for all analyses in an attempt to determine if stationarity was reached, and plots of log likelihood and parameter convergence were also evaluated; log-likelihood plots alone are insufficient for monitoring chain mixing and convergence [ 67 , 68 ]. Partitioned analyses The 3-gene data set was also analyzed in MrBayes 3.0. A "fully partitioned" analysis was used in which the 3-gene data set was divided into seven partitions: nuclear SSU; atpB first, second and third codon positions; rbcL first, second and third codon positions. Appropriate substitution models for each data partition were chosen by computing likelihood scores for each partition on a MP tree for the 3-gene data set under 56 substitution models in PAUP* and comparing the scores in ModelTest. The GTR+Γ + I model was the best-fitting model for all partitions. The Bayesian analysis was performed with all model parameters (except branch lengths) unlinked across partitions. Constraints For the nuclear SSU rDNA data, constrained analyses were also performed. A constraint tree for 63 nonparasitic taxa was constructed using the MP topology of the "B series" tree from Soltis et al. [ 1 ] with relationships for poorly supported clades left unresolved. This tree was used as a backbone constraint for MP and ML analyses of 77 taxa including Rafflesiales. MP analyses were performed as described above. ML analyses followed a successive approximations approach similar to that described above. Simulations To investigate possible long-branch attraction in parsimony analyses of the SSU rDNA data set, two sets of simulations were performed. For the first set of simulations, a reduced data set of SSU rDNA sequences for 20 taxa (13 nonparasites and 7 Rafflesiales) was constructed and analyzed under ML (GTR+Γ + I model) in PAUP*. The tree resulting from this analysis, with its associated ML branch lengths and model parameters, was used as the model tree on which 100 data sets of length 1766 (the length of the original SSU rDNA data set) were simulated in Seq-Gen 1.2.7 [ 69 ]. For the second set of simulations, the ML tree for the full 77-taxon data set, with associated branch lengths and model parameters, was used as a model tree to simulate 100 data sets of length 1766 in Seq-Gen 1.2.7. Either MP and ML trees (20-taxon simulation) or just MP trees (77-taxon simulation) were estimated for all 100 simulated data sets. The trees (or strict consensus trees, if more than one MP or ML tree was recovered for a given simulated data set) were then inspected to determine the presence of "incorrect" clades (containing two or more "long-branch" Rafflesiales taxa) that were not present on the model tree. We do not expect to recover such clades at high frequencies unless long-branch attraction is biasing the analyses. List of abbreviations Γ + I – gamma + invariant sites distribution atp1 – ATP synthase alpha subunit atpB – ATP synthase beta subunit BI – Bayesian inference GTR – general time reversible model HGT – horizontal gene transfer matK – maturase K matR – maturase R ML – maximum likelihood MP – maximum parsimony rbcL – ribulose bisphosphate carboxylase/oxygenase, large subunit SSU – small subunit TBR – tree bisection-reconnection branch swapping Authors' contributions DLN coordinated all aspects of the study, obtained many of the genomic DNAs, generated all the nuclear SSU rDNA, conducted the sequence alignments, and drafted the manuscript. AB conducted the majority of the mitochondrial atp1 and matR sequencing and revised the text regarding morphological character comparisons. YQ provided primers, introduced AB to the field of molecular systematics, and supervised his Ph.D thesis. RVR conducted the PCR experiments showing host contamination of Rafflesia DNA and generated the matR sequences for several taxa. FEA performed the phylogenetic analyses. All authors read and approved the final manuscript. Supplementary Material Additional File 1 MP strict consensus tree from mitochondrial matR Strict consensus of 200,000+ trees obtained from maximum parsimony (unconstrained MP) analysis of the 77-taxon mitochondrial matR matrix. Bootstrap percentages are shown above the lines. Rafflesiales taxa are shown in bold italics. Click here for file Additional File 2 Strict consensus MP tree from mitochondrial atp1 Strict consensus of 328 trees resulting from a MP analysis of the 71-taxon mitochondrial atp1 matrix. Rafflesiales taxa are shown in bold italics. Bootstrap percentages are given above the branches. Click here for file Additional File 3 Majority rule consensus BI tree from 3-gene data set Majority rule consensus of 20,000 trees (10 million generations, 5 million burn-in) resulting from Bayesian analysis of the 77-taxon nuclear 3-gene matrix. Clades with Bayesian posterior probabilities are indicated above the clades. Rafflesiales taxa are shown in bold italics. Click here for file Additional File 4 Strict consensus constrained MP tree from nuclear SSU rDNA Strict consensus of 6 trees resulting from the constrained MP analysis of the 77-taxon nuclear SSU rDNA matrix. Rafflesiales taxa are shown in bold italics. Bootstrap percentages are given above selected nodes (Rafflesiales). Click here for file Additional File 5 Taxa used in this study MS Excel file giving taxon names and GenBank numbers for all genes used. Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC528834.xml

521179

Shut Down, Don't Stress Out

Among the many stresses faced by a cell, one of the most serious is exposure to oxidizing agents. An invading organism, for example, must defend itself against the oxidative assault mounted by a host's immune system. Since oxidation can rapidly destroy many types of molecules, cells have developed multiple means of protecting against it. Rapid mobilization of these defenses requires diversion of resources and temporary suspension of many normal cellular functions, including protein synthesis. In a new study, Elise Hondorp and Rowena Matthews show that when the Escherichia coli bacterium confronts oxidative stress, an enzyme that stands at a central point in the amino acid supply line for protein synthesis is rapidly and reversibly inactivated. Of the twenty amino acids that make up proteins, methionine plays a special role. It is the first amino acid added to every polypeptide chain, and without it, protein synthesis grinds quickly to a halt. Methionine is formed in E. coli through the action of the enzyme cobalamin-independent methionine synthase (MetE), which makes up between one and five percent of all protein in the cell. Thus, by turning off MetE in the face of oxidative stress, protein synthesis can be slowed or stopped, freeing cellular resources to be used elsewhere. Hondorp and Matthews show that in E. coli , MetE is acutely vulnerable to oxidation under a variety of conditions. These results are in accord with a companion study by Leichert and Jakob, also in PLoS Biology, showing that MetE is one of the proteins most sensitive to oxidative damage. When the active site of MetE is stressed by an oxidant, Hondorp and Matthews show, it is temporarily blocked by the attachment of a glutathione subunit. Glutathione is a small molecule that includes a reactive sulfur atom. During “glutathionylation” of MetE, a sulfur on an amino acid of the enzyme is oxidized and links up with a sulfur on glutathione. This study shows that glutathionylation occurs only on a specific amino acid (cysteine 645), which recent structural work indicates sits at the entrance to the active site. Attachment of the bulky glutathione subunit to this cysteine would be expected to block the active site, thus shutting down enzymatic activity. The results indicate that glutathionylation does indeed prevent activity of the enzyme, and furthermore, causes the enzyme to change its three-dimensional form. As the oxidative challenge abates, glutathionylation may be reversed, and the normal activity of the enzyme restored. Thus, glutathionylation of MetE may also serve to protect the active site from permanent oxidative damage. While glutathionylation is a common strategy in eukaryotes, MetE is so far one of the few proteins in bacteria known to be affected in this way. Shutting down MetE and limiting methionine production may play another important role, namely, communicating the bacterium's metabolic state to other nearby E. coli . Methionine is a precursor for the signaling molecule AI-2, which is released extracellularly and appears to serve as a key indicator of colony health and density. This information enables neighboring cells to better respond to changing and potentially hostile environments. Thus, the glutathionylation and inactivation of MetE may provide a simple mechanism by which a bacterium and its neighbors attempt to deal with oxidative stress.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC521179.xml

387283

The Elusive Baseline of Marine Disease: Are Diseases in Ocean Ecosystems Increasing?

Disease outbreaks alter the structure and function of marine ecosystems, directly affecting vertebrates (mammals, turtles, fish), invertebrates (corals, crustaceans, echinoderms), and plants (seagrasses). Previous studies suggest a recent increase in marine disease. However, lack of baseline data in most communities prevents a direct test of this hypothesis. We developed a proxy to evaluate a prediction of the increasing disease hypothesis: the proportion of scientific publications reporting disease increased in recent decades. This represents, to our knowledge, the first quantitative use of normalized trends in the literature to investigate an ecological hypothesis. We searched a literature database for reports of parasites and disease (hereafter “disease”) in nine marine taxonomic groups from 1970 to 2001. Reports, normalized for research effort, increased in turtles, corals, mammals, urchins, and molluscs. No significant trends were detected for seagrasses, decapods, or sharks/rays (though disease occurred in these groups). Counter to the prediction, disease reports decreased in fishes. Formulating effective resource management policy requires understanding the basis and timing of marine disease events. Why disease outbreaks increased in some groups but not in others should be a priority for future investigation. The increase in several groups lends urgency to understanding disease dynamics, particularly since few viable options currently exist to mitigate disease in the oceans.

Introduction Marine organisms serve as hosts for a diversity of parasites and pathogens. Mortalities affect not only the host population, but can cascade through ecosystems. Loss of biologically engineered habitats such as seagrass beds ( Lewis 1933 ; Taylor 1933 ) and cascading trophic effects due to removal of consumers ( Lessios 1988 ) can alter community structure. Understanding marine disease and the timing of outbreaks is increasingly important given escalating anthropogenic stressors affecting marine ecosystems. Humans directly affect community structure (e.g., overfishing [ Jackson et al. 2001 ; Myers and Worm 2003 ]) and facilitate introduction of terrestrial pathogens to marine organisms (e.g., canine distemper virus in Antarctic seals [ Bengtson and Boveng 1991 ]). Human-mediated climate change may also affect disease prevalence. A recent review predicts disease in both terrestrial and marine ecosystems could increase with future climate warming ( Harvell et al. 2002 ). Previous literature reviews suggesting a higher rate of disease outbreaks in the last three decades ( Epstein et al. 1998 ; Harvell et al. 1999 ), coupled with predictions of future increases due to climate change ( Harvell et al. 2002 ), lend new urgency to understanding causes of marine disease outbreaks. Evidence suggests the increase is real ( Harvell et al. 1999 ), yet lack of baseline data for most marine communities precludes a direct test of the hypothesis. We developed a proxy method to test a prediction of the increasing disease hypothesis: that reports of disease in the scientific literature, normalized to overall publication rates, increased since 1970. We searched an online literature database (ISI Web of Science) and quantified reports of disease in natural populations of marine organisms from 1970 to 2001. Nine marine taxonomic groups were searched: turtles, corals, mammals, urchins, molluscs, seagrasses, deca-pods, sharks/rays, and fishes. Previous analyses of ecological literature specifically assessed trends among scientists such as taxonomic bias ( Clark and May 2002 ) and taxonomic chauvinism ( Bonnet et al. 2002 ) in research. Our proxy method is to our knowledge the first quantitative use of normalized trends in the literature to investigate an ecological hypothesis. In the absence of baseline data, the literature proxy method detects important trends of disease in major groups of marine plants, invertebrates, and vertebrates. Results The largest confounding factor when using literature searches to correlate disease events with time is temporal change in the total number of publications (related to disease or not) on the taxonomic group. To control for changes in total publication, data were normalized using a yearly proportion of disease reports from natural populations relative to total literature inputs for each taxonomic group. Total disease reports, not normalized to literature inputs, increased in all groups ( Table 1 ). However, normalized results varied with taxonomic group. Normalized disease reports increased in turtles, corals, mammals, urchins, and molluscs. No significant trends were detected for seagrasses, decapods, and sharks/rays (though disease occurred in these groups). Counter to the hypothesis, disease reports decreased in fishes ( Figure 1 ). Figure 1 Percent of Literature Reporting Disease over Time in Each Taxonomic Group r s is Spearman's ρ. α is controlled for multiple comparisons with Holm's sequential Bonferroni adjustments. (A) Turtle. (B) Coral bleaching and disease (closed square); coral disease including infectious bleaching (open circle); coral bleaching (asterisk). (C) Mammal. (D) Urchin. (E) Mollusc. (F) Seagrass. (G) Decapod. (H) Shark/ray. (I) Fish. Table 1 Spearman's Rank Correlation Analysis The table shows total reports (not corrected for research effort), normalized reports, and normalized reports with most frequent author removed. r s is Spearman's ρ. α is controlled for multiple comparisons with Holm's sequential Bonferroni adjustments. Bold indicates significance The relevance of our approach hinges on the assumption that an actual change in disease over time is accompanied by a corresponding change in publication frequency by scientists. We evaluated this assumption by testing the protocols with a case in which the baseline was known (raccoon rabies [ Rupprecht and Smith 1994 ]). Normalized reports of raccoon rabies increased since 1970 (see Table 1 ) just as the disease increased from an index case in Virginia in 1977 to an epizootic affecting eight mid-Atlantic states and the District of Columbia by 1992 ( Rupprecht and Smith 1994 ). Despite improvements in search protocols, use of a literature proxy is limited by the inability to distinguish between an event that did not occur and an event that was not reported. We tested whether particular authors contributed disproportionate primary literature inputs that could bias results. Papers by the most prolific author in each taxonomic group were removed to determine whether there was an “author effect,” and none was observed in any taxonomic group (see Table 1 ). Multiple reports of a single disease event could also bias the data. Multiple reports were removed from the turtle, coral, urchin, mammal, shark/ray, and seagrass literature. Removal of the reports did not alter the significance of the results; thus, multiple reports in the mollusc, decapod, and fish literature were not removed, owing to the large volume of literature in these groups. Discussion We address an ecological hypothesis, that disease of marine organisms increased since 1970, using a quantitative literature proxy method. Although total reports of marine disease increased over time ( Epstein et al. 1998 ; see Table 1 ), a parallel increase in publication rates confounds interpretation of this pattern. Our approach normalizes data to overall publication within each group to circumvent this problem. While an increase in disease reports was detected in many taxa, our finding that disease did not increase in all taxa has two important implications. First, the increases were not exclusively the result of increased study of disease by marine biologists. Second, factors such as global change may have complex effects on disease. Although some aspects of global change, such as warming and pollution, are predicted to make hosts more susceptible to infection ( Scott 1988 ; Holmes 1996 ), some stressors may impact parasites more than their hosts ( Lafferty 1997 ). Signs of infection with coldwater disease in salmonids, for example, occur between 4°C and 10°C and disappear as water temperature increases ( Holt et al. 1989 ). In addition, stressors that depress host population density may reduce density-dependent transmission of host-specific infectious disease by reducing contact rates between infected and uninfected individuals ( Lafferty and Holt 2003 ). New or increasing stressors, such as global warming, could increase disease if stressed hosts are more susceptible to infection. Elevated sea surface temperature due to El Niño events is a common explanation for coral bleaching ( Williams and Bunkley-Williams 1990 ; Hoegh-Guldberg 1999 ) and may increase coral susceptibility to disease ( Harvell et al. 2001 ). Increases in turtle and mollusc disease also appear temperature-related. Green turtle fibropapilloma tumors are hypothesized to grow rapidly in summer and reach a debilitating size by winter, when cold water temperatures further stress turtles, resulting in winter strandings ( Herbst 1994 ). The geographic range of the oyster parasite Perkinsus marinus extended 500 km north owing to an increase in average winter low temperatures ( Ford 1996 ). Pollution is another ubiquitous and increasing stressor. Bioaccumulation of lipophillic toxins in marine mammals affects the immune system and increases susceptibility to disease ( Lafferty and Gerber 2002 ). Disease could also increase if transmission increases with host density. Some sea urchins experienced increased populations due to overfishing of their predators, and these high-density populations are associated with bacterial disease ( Lafferty and Kushner 2000 ). Regulations such as the United States Marine Mammal Protection Act of 1972 fully protect pinniped populations, and several species have increased in abundance to levels where transmission efficiency would be expected to increase. The decline in infectious diseases of wild fishes over time corresponds to documented reductions in fish populations through intense fishing ( Jackson et al. 2001 ; Myers and Worm 2003 ). Fisheries that reduce the abundance of a fished species should also reduce infectious disease transmission ( Dobson and May 1987 ). This has been documented in experiments ( Amundsen and Kristoffersen 1990 ) and in observations of parasite declines associated with overfishing ( Sanders and Lester 1981 ). Grouping diseases within taxa could obscure important patterns. For example, the trend for increasing coral disease was driven by coral bleaching ( r s = 0.87, p < 0.0001), while infectious coral diseases, including infectious bleaching, did not increase over time ( r s = 0.13, p = 0.4934; see Figure 1 B). The infectious bleaching literature includes several papers since 1996. To ensure the lack of a significant coral disease trend was not due to multiple papers published on this topic at the end of the time range surveyed, an additional analysis was conducted with all infectious bleaching papers excluded; r s and p values did not change ( Table 2 ). Table 2 Normalized Coral Disease Reports Original data include papers on infectious bleaching. r s and p values are the same for both analyses. Italics indicate changes in proportions after removal of infectious bleaching literature While we did not detect an increase in normalized coral disease reports over time, impacts of disease can be high. The recent shift of dominant corals ( Acropora to Agaricia ) on reefs due to white band disease was unprecedented in the last 3,000 y ( Aronson et al. 2002 ). Future research should take a finer-scale look at disease, particularly disease impacts, within each taxonomic group. Further investigation is also warranted to determine why some groups showed no temporal pattern in disease reports. We examined temporal trends in disease reports since 1970 to identify groups experiencing increased outbreaks. The strong pattern of increased reports in groups such as turtles, mammals, and urchins reflects perceived changes noted by scientists ( Harvell et al. 1999 ). Trends in other groups, such as seagrasses and fishes, suggest that an increase in disease did not occur across all taxa. Although this proxy approach does not directly test hypotheses of temporal changes in disease, a strong signal likely reflects an underlying pattern in nature. In the absence of baseline data, this is a useful approach for detecting quantitative trends in disease occurrence. Understanding disease dynamics, including trends in disease occurrence, is fundamental to conserve ecosystems faced with rising anthropogenic stresses. Materials and Methods We searched the Science Citation Index Expanded (5,900 journals, ISI Web of Science versions 1.1 and 1.2) for papers published from 1970 to 2001 with titles containing specific host taxonomic strings alone and in combination with a disease string ( Table 3 ). We excluded articles clearly about disease in nonnatural settings, such as hatcheries, aquaculture, and mariculture, or about experimental or laboratory infections. Searches for corals were performed twice to quantify reports of bleaching separately from infectious bleaching (e.g., Vibrio shiloi [ Israely et al. 2001 ]) and disease. Only titles were searched, as online abstracts are not available for many articles prior to 1990. Searching the complete citation would bias results after 1990 because more text of each publication would be searched. Table 3 Taxonomic Groups and Search Strings Abstracts (or entire manuscripts, when necessary) were obtained for articles within the turtle, coral, urchin, mammal, shark/ray, and seagrass literature that appeared to report the same disease event (e.g., multiple reports of the Caribbean Diadema urchin mortality). If more than one paper reported an event, only the earliest published report was included in the analysis. Because significance of results was not altered, multiple reports of disease were not removed from mollusc, decapod, and fish literature owing to the large number of publications returned for each group. Often, returned titles contained part of the search string, but were not relevant (e.g. “crab nebula” when searching “crab”). Modifications to search strings excluded most irrelevant articles, and titles were read to determine relevance. If more than 50 titles were returned, titles were randomly sorted and the greater of 20% (maximum of 200) or 50 returned titles were read. Total relevant articles were calculated as the proportion of relevant articles read times the total number returned. Protocols were tested using raccoon rabies, a disease for which baseline data are available ( Rupprecht and Smith 1994 ). Potential biases were considered and tested. Extensive descriptive or taxonomic work early in the study of a group could bias results against a large number of disease reports. If such a bias existed, one would expect both a large number of disease reports and a large number of nondisease publications in the beginning of the literature survey period. Neither prediction is true—the number of both disease reports and nondisease publications either remains relatively constant or increases through time in all groups. Frequent publishing by one author could bias results. Papers by the most published author in a taxonomic group were removed from the analysis to determine their effect. Papers on a particular “hot” topic could also bias results, particularly if that topic is disease and inflates normalized disease reports late in the survey period. For example, a recent mortality event could increase scientists' awareness of disease, resulting in increased publishing without a concomitant increase in the phenomenon. This likely does not affect our results because (a) disease is not the only “hot” topic experiencing increased publication rates and (b) while multiple papers on disease may be published, not all are reports of disease in natural populations. A 3-y running mean was used to reveal trends obscured by clustered reporting (e.g., a symposium volume on a topic) and time lags between observation and publication (approximately 3 y, determined by comparing event and publication dates). Data were analyzed with Spearman's rank correlation (JMP version 5.0) with α controlled for multiple comparisons using Holm's sequential Bonferroni adjustments.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC387283.xml

544939

The birth of Emerging Themes in Epidemiology: a tale of Valerie, causality and epidemiology

Emerging Themes in Epidemiology (ETE) is a new, online, Open Access peer-reviewed journal. The Journal is unique in that it was conceived and is managed by research degree students in epidemiology and related public health fields. The Journal's management is overseen by its Editor-in-Chief and Associate Faculty Editors, all of whom are senior members of faculty. ETE aims to encourage debate and discussion on the theoretical, methodological and practical aspects of epidemiologic research and practice. In addition, ETE is dedicated to the promotion of Open Access publication and the training of research students in the scientific publishing process. This editorial, to coincide with the launch of ETE , sets out the Journal's philosophy and aims. Epidemiology is a rich and innovative science that has much to gain from broader discussion of the causal frameworks that underpin it. ETE aims to be a major forum for such discussion.

" Causality. There is no escape from it, we are forever slaves to it. Our only hope, our only peace is to understand it, to understand the 'why' [ 1 ]." - The Merovingian As far as we know, the Wachowski brothers' Merovingian was not an epidemiologist, but his sentiments should resonate well with those in the field. Probably more than those in any other profession, epidemiologists are slaves to causality. Our professional lives are dedicated to the pursuit of pumphandles, spiderwebs and causal pies [ 2 , 3 ]. Adding to these colourful metaphors are ones proposing frameworks for epidemiologic research that involve Chinese boxes, computer-generated fractals and prison breaks [ 2 , 4 , 5 ]. But what is the student of epidemiology to make of all these curious abstractions? Lost in a sea of metaphors, they might very likely throw their arms up in the air and, in thorough confusion, decide to take a long and much-needed coffee break [ 6 - 8 ]. The epidemiologic literature on causality certainly makes for stimulating reading, but it would be interesting to know how many of us have causal pies and fractals on our minds as we reach for that red folder labelled "Logistic regression 101". Discursive articles on the usefulness of such metaphors are widely regarded as philosophical flights of fancy that we might eventually get around to reading after clearing that backlog of papers waiting to be written in the next two months. Yet what are our alternatives? The newly-qualified epidemiologist leaves their degree with a solid grasp of error, bias, confounding and statistical methodology, but with perhaps a single lecture on Koch-Henlé postulates and Bradford-Hill 'criteria' as the extent of their training into causal thinking. It is interesting to note that Last's Dictionary of Epidemiology does not include the term 'cause', opting instead to give a definition of 'causality' that involves a brief discussion of necessity and sufficiency [ 9 ]. Any reasoned assessment will quickly lead to the conclusion that guidelines for determining causal pathways as commonly taught in epidemiology courses are woefully inadequate, regardless of whether one decides to take an inductionist, refutationist, or hypothetico-deductivist view [ 10 , 11 ], or admits to not having a clue what any of these terms mean. The challenge for the modern epidemiologist is to put those newly-learned methods to use from within a causal limbo, with Robert and Austin as their guides and John Snow as spiritual counsellor. The anatomy of a cause James Wong's 2000 teen horror movie Final Destination [ 12 ] is unlikely to go down in history as a cinematic classic, but it is memorable for its clever, if rather gruesome, depictions of causal processes leading invariably to the death of a number of its unfortunate characters. The movie's motto is that "you can't cheat death's plan". Having narrowly avoided a fatal plane accident thanks to the protagonist's chilling premonition, a French teacher and five of her students are destined to fall victims to the Grim Reaper one by one- in the order they would have died had they been on the ill-fated plane with the rest of their class- unless they can find a way to break the cycle and cheat death. In the most elaborate death scene, Valerie, the French teacher, alone in her house and still visibly shaken by the loss of her colleague and students, becomes unnerved by noises outside. With John Denver's Rocky Mountain High ironically playing in the background, she tries to calm herself by making some tea. Moving to the kitchen sink, she fills the kettle with water, spilling some down the side. She wipes the kettle, turns towards the gas stove and tosses the towel carelessly behind her, which catches onto a knife block. With the kettle whistling, she picks up a school coffee mug, drops two tea bags inside and fills it with boiling water. Picking up the mug, she suddenly recognizes the school logo and, in shock, reflexively throws the mug's contents into the sink. Opting now for something stronger to calm her nerves, she takes some ice cubes from the freezer, drops them into the still-warm mug and re-fills it, this time with vodka. Oblivious to the crack that has appeared in the mug, she walks towards the living room, leaving a trail of vodka behind. As she stands by her computer monitor, vodka drips into the circuitry. An electrical surge creates a spark that ignites the alcohol, causing the monitor to explode and sending out flying shards of glass that slash her throat. Shocked and bleeding, she stumbles towards the kitchen sink, chased by a trail of burning alcohol. Reaching the stove, the trail of flames ignites the gas burners, lighting up her clothes and hair. Falling to the ground and still bleeding, she rolls around violently trying to put out the flames. In an act of desperation, she reaches up and grabs the dangling towel, tilting the knife block and sending a half dozen blades cascading into her stomach while flames catch the curtains and set the house on fire. Suppose you were the investigator arriving on the scene. There is a half burnt-down house, a blown-up computer, a broken mug and a corpse with third-degree burns, stab wounds and a cut throat, but no signs of struggle. What would you determine was the cause of this tragedy? The severe burns, the protruding knives and the neck wound would be pretty obvious choices. But perhaps it is more complicated than that. Perhaps there were extenuating circumstances without which this tragic outcome might not have occurred. The exploding computer maybe, or what about the vodka that burned leaving no trace, or the cracked mug? No, maybe it was the towel, complicitly catching onto the knife block. Or maybe we should blame the teacher's drinking habit. The point of this rather unsavoury story is that without the benefit of such extrinsic observation, detailed reconstructions of causal processes are unattainable. A reasoned observer might conclude that all of the above factors were in some way responsible. They all contributed to the process in their own small way, and had any one of them not been involved things may not have turned out the way they did. They were all what one might call 'component causes'. But is this enough to convince us of what the real cause was? Clearly not. Suppose now that you were an audience member and were somehow able to communicate directly with the characters in the movie. You might wish to warn Valerie of her impending ill fate. At which point in the whole sequence would you alert her so that her death could be prevented? Clearly you would not deny her a mug of tea and you would most likely have no way of knowing that the towel would land on the kitchen block with dire consequences just a few moments later. You might, of course, realize this at the last moment and warn her against reaching up and grabbing the towel, but she might still have died from her neck wound. You might have shouted for her to grab a fire extinguisher and put out the trail of flames, but it is unlikely she would have listened to you as she tried hopelessly to stop the bleeding from her neck. You might, with better foresight, have pointed out to her that vodka was dripping from her mug. Or perhaps with hindsight, you might have recognized that the best thing would have been to provide some moral support and consolation in her time of grief, with which the whole sorry incident might have been avoided altogether. One thing is now clear. There are steps within causal processes on which we can act to try and alter their course- the trail of flames might have been extinguished, and the consequences of the dripping vodka might have been avoided. There are other steps on which we can have no influence, eg. the tossed towel landing on the knife block. Another important thing is also apparent: causal processes have hierarchies. Depending on what happens at one stage, a number of alternative events may result at the next. Had Valerie been able to stop the trail of flames, she might have reached the kitchen sink, realized she could not stop the bleeding and called an ambulance. Or she might have run out of the house shouting for help and her neighbour, trained in first aid, might have saved her life. Saving victims of horror movies, however, is not an easy job. Knowing when best to step in is not necessarily that simple, as we have seen with Valerie. In some cases we may be given a number of opportunities and intervening at any of them might lead to a positive outcome. But in other cases, once certain factors are in place there will be an air of inevitability in everything that follows, and all our attempts to intervene thereafter may prove to be little more than an exercise in futility. Perhaps it is now time to admit that I have extended this fanciful analogy far beyond what is appropriate. I make no apology if in so doing I have in some way managed to convey the idea that epidemiology thrives on causal processes, on elucidating their complexities and identifying the most effective points for intervention no matter at what level of their elaborate hierarchies. If this is a worthwhile venture, then the discussion of how we conceptualize and study these causal processes surely is so too. Emerging Themes in Epidemiology (ETE) was born out of this ideal- that contrary to common belief, epidemiology is not merely a collection of standardized tools to be applied at will to any health-related situation, but that it is a rich and innovative science that aims to describe reality in all its complexity, spanning the molecular to the global, with the ultimate goal of improving the health of individuals and populations. And that in order for this to be achieved, we need to improve our understanding of how factors, at any level of biological or social organization, eventually lead to ill health. ETE is founded on three core principles: • That epidemiology and epidemiologists have much to gain from a broader and more fundamental discussion of the concepts and theoretical frameworks that underpin the practice of epidemiologic research • That Open Access publication has a crucial role to play in reducing the current inequities in access to scientific information, which should be a universal and freely-available resource for the benefit of the whole of society • That students of epidemiology and related fields can make substantial contributions to the introduction of new concepts and ideas into mainstream epidemiologic research, not only through writing, but also through having a direct influence over what is published In keeping with this philosophy, we recognize that epidemiology has much to gain from developments in other fields and we welcome contributions from all public health professionals. We will consider articles that comment critically on current epidemiologic theory and practice, either generally or within a specific specialty, including articles from other fields that have implications for the conduct of epidemiologic research. ETE will not generally publish research reports, although exceptions may be made in cases where the results can be placed within a broader public health context to present a new concept or theoretical framework. By making all of this material freely available online under the auspices of the Open Access publisher, BioMed Central, we aim to make Emerging Themes in Epidemiology a global forum for the discussion of new developments in epidemiologic thinking and practice that will benefit the global public health community. In doing so, we recognize that there is much that the scientific community can do to support Open Access publication. The health consequences of inequitable access to scientific information remain largely ignored, yet for years we have adhered to a system of publication that is restrictive and largely subsidized by institutions and libraries at great expense. Open Access is an important step towards making the publication process, and its associated costs, more transparent. We thus call on individuals, institutions, funding bodies and governments to engage in the Open Access movement by promoting and supporting Open Access publication. This will involve a major shift not just in publishing costs, but also in thinking. Our current measures for assessing the impact of academic research are inherently intertwined with our inequitable publication tradition. Developing new ways of assessing the quality of scientific research that are independent of journals' perceived 'impact' are imperative for the wider recognition of Open Access. In promoting the role of students in the publication process, we intend for ETE to be a training ground for postgraduate students, providing them with an opportunity to be involved at every stage of the publication process, including commissioning, reviewing and writing articles. Our editorial board is formed principally of research students, who operate the Journal with support from an international group of associate faculty. Our collaborations extend to a growing number of students from diverse institutions serving as article referees. We welcome suggestions for extending these collaborations in the future. The third millennium has brought with it exciting challenges for epidemiologists. An explosion in the availability of genetic and molecular information, the development of bioinformatics, the increasing application of sophisticated statistical analyses to model complex systems and the gradual incorporation of sociological approaches to understanding health inequalities have arrived together with sobering statistics on the state of global public health. Knowing when and how to apply these and other developments at a time of rapid political, social and biological change while maintaining clear sight of our public health goals will be the major challenge for current and future generations. It is our hope that Emerging Themes in Epidemiology will be a tool for epidemiologists confronting that challenge.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC544939.xml

514883

Monomethyl Branched-Chain Fatty Acids Play an Essential Role in Caenorhabditis elegans Development

Monomethyl branched-chain fatty acids (mmBCFAs) are commonly found in many organisms from bacteria to mammals. In humans, they have been detected in skin, brain, blood, and cancer cells. Despite a broad distribution, mmBCFAs remain exotic in eukaryotes, where their origin and physiological roles are not understood. Here we report our study of the function and regulation of mmBCFAs in Caenorhabditis elegans, combining genetics, gas chromatography, and DNA microarray analysis. We show that C. elegans synthesizes mmBCFAs de novo and utilizes the long-chain fatty acid elongation enzymes ELO-5 and ELO-6 to produce two mmBCFAs, C15ISO and C17ISO. These mmBCFAs are essential for C. elegans growth and development, as suppression of their biosynthesis results in a growth arrest at the first larval stage. The arrest is reversible and can be overcome by feeding the arrested animals with mmBCFA supplements. We show not only that the levels of C15ISO and C17ISO affect the expression of several genes, but also that the activities of some of these genes affect biosynthesis of mmBCFAs, suggesting a potential feedback regulation. One of the genes, lpd-1, encodes a homolog of a mammalian sterol regulatory element-binding protein (SREBP 1c). We present results suggesting that elo-5 and elo-6 may be transcriptional targets of LPD-1. This study exposes unexpected and crucial physiological functions of C15ISO and C17ISO in C. elegans and suggests a potentially important role for mmBCFAs in other eukaryotes.

Introduction Fatty acids (FAs) belong to a physiologically important class of molecules involved in energy storage, membrane structure, and various signaling pathways. Different FAs have different physical properties that determine their unique functions. Among the most abundant in animal cells as well as the most studied are those of long-chain even-numbered saturated and unsaturated FAs. C15ISO and C17ISO are saturated tetradecanoic and hexadecanoic FAs with a single methyl group appended on the carbon next to the terminal carbon ( Figure 1 ). Monomethyl branched-chain FAs (mmBCFAs) in ISO configuration as well as in anteISO configuration (methyl group appended on the second to the terminal carbon) also seem to be ubiquitous in nature. They are present in particularly large quantities in various bacterial genera, including cold-tolerating and thermophilic species ( Merkel and Perry 1977 ; Annous et al. 1997 ; Ferreira et al. 1997 ; Batrakov et al. 2000 ; Jahnke et al. 2001 ; Groth et al. 2002 ; Nichols et al. 2002 ). There, mmBCFAs contribute to the membrane function, regulating fluidity ( Rilfors et al. 1978 ; Suutari and Laakso 1994 ; Cropp et al. 2000 ; Jones et al. 2002 ) and proton permeability ( van de Vossenberg et al. 1999 ). Figure 1 Structure of mmBCFAs of 15 and 17 Carbons C15ISO, 13-methyl myristic acid; C17ISO, 15-methyl hexadecanoic acid; C17anteISO, 14-methyl hexadecanoic acid. Other mmBCFAs mentioned in the text are the following: C13ISO, 11-methyl lauric acid; C15anteISO, 12-methyl tetradecanoic acid. C15ISO and C17ISO are readily detectable in C. elegans. Although comprehensive reports on mmBCFAs in eukaryotes are lacking, sporadic data indicate that they are present in the fungi, plant, and animal kingdoms ( Garton 1985 ; Seyama et al. 1996 ; Martinez et al. 1997 ; Cropp et al. 2000 ; Wolff et al. 2001 ; Destaillats et al. 2002 ). In mammals, mmBCFAs have been detected in several tissues, including skin ( Aungst 1989 ), Vernix caseosa ( Nicolaides and Apon 1976 ), harderian and sebaceous glands ( Nordstrom et al. 1986 ), hair ( Jones and Rivett 1997 ), brain ( Ramsey et al. 1977 ), blood ( Holman et al. 1995 ), and cancer cells ( Hradec and Dufek 1994 ). The fact that mmBCFAs are present in a wide variety of organisms implies a conservation of the related metabolic enzymes and consequently important and perhaps unique functions for these molecules ( Jones and Rivett 1997 ). Nevertheless, their physiological roles and metabolic regulations have not been systematically studied and thus remain fragmentary. It was found that C21anteISO is the major covalently bound FA in mammalian hair fibers. A removal of this FA from its protein counterparts results in a loss of hydrophobicity ( Jones and Rivett 1997 ). Other studies indicated that C17anteISO esterified to cholesterol binds to and activates enzymes of protein biosynthesis ( Tuhackova and Hradec 1985 ; Hradec and Dufek 1994 ). A potential significance of mmBCFAs for human health is associated with a long-observed correlation between amounts of these FAs and disease conditions such as brain deficiency ( Ramsey et al. 1977 ) and cancer ( Hradec and Dufek 1994 ). More recent studies have revealed a role of another mmBCFA, C15ISO, as a growth inhibitor of human cancer where it selectively induces apoptosis ( Yang et al. 2000 ). Given how important these FA molecules may be and how little is known about their biosynthesis and functions in eukaryotes, it is an opportune problem to study. De novo synthesis of long-chain mmBCFAs described for bacteria is quite different from the biosynthesis of straight-chain FAs ( Smith and Kaneda 1980 ; Oku and Kaneda 1988 ; Toal et al. 1995 ). While the latter uses acetyl-coenzyme A (acetyl-CoA) as a primer condensing with a malonyl-CoA extender, branched-chain FA synthesis starts with the branched-chain CoA primers derived from the branched-chain amino acids leucine, isoleucine, and valine. To synthesize branched-chain FAs, organisms must have a system for supplying branched-chain primers along with the enzymes utilizing them ( Smith and Kaneda 1980 ). No such enzymes have been previously characterized in vivo for any eukaryotic organisms. Here we describe our approach to characterize the biosynthesis and function of mmBCFAs using the free-living nematode Caenorhabditis elegans. Combining genetic, molecular, and biochemical analyses, we show that the worm is not only able to synthesize mmBCFAs de novo but is also absolutely dependent on these FA species for its growth and development. Results/Discussion C. elegans Synthesizes Branched-Chain FAs De Novo and Uses Two FA Elongation Enzymes to Produce C15ISO and C17ISO In characterizing FA elongation in C. elegans, we identified eight sequences homologous to the yeast long-chain FA elongation enzymes ( Kniazeva et al. 2003 ). To test for their possible functions in vivo, we applied RNAi to the corresponding genes, followed by an analysis of FA composition in whole animals using gas chromatography (GC). RNAi treatment of four genes— elo-3 (D2024.3), elo-4 (C40H1.4), elo-7 (F56H11.3), and elo-8 (Y47D3A.30)—did not produce any notable phenotypes, whereas suppression of elo-1 (F56H11.4) and elo-2 (F11E6.5) affected the elongation of straight long-chain saturated and polyunsaturated FAs ( Beaudoin et al. 2000 ; Kniazeva et al. 2003 ). Surprisingly, the RNAi treatment of the two remaining genes, elo-5 (F41H10.7) and elo-6 (F41H10.8), affected the levels of branched-chain FA. Transcriptional reporter constructs ( elo-5Prom :: GFP and elo-6Prom :: GFP ) indicated that both genes are expressed in the gut ( Figure 2 ). In addition, elo-5 was expressed in unidentified head cells and elo-6 was expressed in neurons, pharynx, and vulva muscles. Figure 2 The Expression of elo-5Prom :: GFP and elo-6Prom :: GFP Constructs in Wild-Type Worms (A, C, E, and G) DIC images; (B, D, F and H) fluorescence images. (A–D) Strong expression of the elo-5Prom :: GFP construct in the gut and in the head is shown. (E–H) The expression of the elo-6Prom :: GFP construct in the gut, vulvae (white arrows), and nerve ring is shown. Scale bars, 100 μm. The RNAi of elo-6 significantly reduced the amount of only C17ISO, while the RNAi of elo-5 dramatically reduced quantities of both C15ISO and C17ISO ( Figure 3 ). These results indicate that ELO-5 might be involved in the biosynthesis of C15ISO and possibly also C17ISO, whereas ELO-6 may function in elongating C15ISO to C17ISO ( Figure 3 C and 3 D). To our best knowledge, these are the first enzymes that have been shown to be involved in long-chain mmBCFA biosynthesis in a nonbacterial in vivo system and the first enzymes of the long-chain FA elongation family related to mmBCFA production. Figure 3 RNAi Treatment of elo-5 and elo-6 Significantly Alters the FA Composition (A and B) GC profiles showing the FA composition in the wild-type strain (Bristol N2) containing the RNAi feeding control vector and in the elo-5(RNAi) feeding strain. Arrowheads point to the peaks corresponding to C15ISO and C17ISO. (C) Comparison of FA composition in three strains: wild type, elo-5(RNAi), and elo-6(RNAi). C17ISO is decreased in both RNAi strains, while C15ISO is only decreased in elo-5(RNAi). (D) Suggested elongation reactions catalyzed by ELO-5 and ELO-6 in C15ISO and C17ISO biosynthesis. FAs are elongated by an addition of two carbon groups at a time. Combined data presented in this figure and in the text suggest that ELO-6 acts at the elongation step from C15 to C17, whereas ELO-5 may be involved in the production of both C15ISO and C17ISO. In bacteria, mmBCFA biosynthesis utilizes branched-chain α-keto-acids of leucine, isoleucine, and valine to produce mmBCFA acyl-CoA primers that substitute for acetyl-CoAs in conventional FA biosynthesis ( Oku and Kaneda 1988 ). Key enzymes engaged in synthesizing the mmBCFA acyl-CoA primers are branched-chain aminotransferase (BCAT) and the branched-chain α-keto-acid dehydrogenase (BCKAD) complex ( Figure 4 A). The elongation of the mmBCFA backbone is then carried out by fatty acid synthetase (FAS). Figure 4 The C. elegans BCKAD Homolog Is Involved in mmBCFA Biosynthesis (A) Early steps of mmBCFA biosynthesis in bacteria, based on Smith and Kaneda (1980) , Oku and Kaneda (1988) , and Toal et al. (1995) . IVD, isovaleryl-CoA dehydrogenase. Predicted corresponding C. elegans genes encoding predicted orthologs were identified (shown in italicized names of reading frames). (B) GC profiles reveal differences in the FA composition in the wild-type animals and animals treated with RNAi of E1 alpha subunit of BCKAD encoded by Y39E4A.3. Black arrowheads point to C15ISO and C17ISO. (C) A summary of several independent preparations shows a significant decrease in both mmBCFAs in the Y39E4A.3 dsRNA-treated animals ( p = 0.001 and 0.008 for C15ISO and C17ISO, respectively). The ability of C. elegans to grow on the chemically defined axenic medium CbMM ( Lu and Goetsch 1993 ), which lacks the potential mmBCFA precursors, has suggested that the animals can synthesize mmBCFA de novo. If so, a disruption of the BCKAD complex could affect mmBCFA levels. We identified a predicted C. elegans protein, Y39E4A.3, with significant sequence homology to the E1 alpha subunit of BCKAD (Y39E4A.3 scores expect value 8e-50 on 57% of the length with the Bacillus subtilis BCKAD and 1.4e-134 on 88.4% of the length with the Homo sapiens BCKADs). RNAi of Y39E4A.3 led to a significant decrease in C15ISO and C17ISO production ( Figure 4 B and 4 C). RNAi suppression of another predicted component of the BCKAD complex, pyruvate dehydrogenase (T05H10.6), resulted in a similar decrease in C15ISO and C17ISO (unpublished data), indicating a role for the C. elegans BCKAD protein in long-chain mmBCFA biosynthesis. Thus, C. elegans appears to use the same initial reactions to produce mmBCFAs as bacterial cells. In addition, the worm uses enzymes of the FA elongation family, ELO-5 and ELO-6, to complete the pathway. A connection between BCKAD functions and mmBCFA quantities has been previously reported in humans ( Jones et al. 1996 ). Normally hair fibers are densely covered with C21anteISO, which contributes about 38.2% to the total hair FAs ( Jones and Rivett 1997 ). It was observed that patients with maple syrup urine disease, which is caused by an inherited mutation in the BCKAD gene, had a drastically reduced level of mmBCFAs in their hair. Together, these data suggest that long-chain mmBCFA biosynthesis could be similar in bacteria, C. elegans, and human. Blocking ELO-5 Function Causes Growth and Developmental Defects While the suppression of elo-6 activity by feeding double-stranded RNA (dsRNA) to wild-type animals did not cause obvious morphological or growth defects, the suppression of elo-5 resulted in more pronounced phenotypes ( Figure 5 ). Worms originating from wild-type eggs laid on the elo-5(RNAi) plates displayed no obvious growth or morphological abnormality until the second day of adulthood, when they developed an egg-laying defect ( Figure 5 B). Eggs of the next generation hatched on time but the progeny arrested at the first of the four larval stages (L1). The small larvae maintained morphological integrity and could survive on a plate for up to 3–4 d. The arrest was only observed in progeny of parents exposed to elo-5 RNAi at the L1 stage. Figure 5 RNAi Treatment of elo-5 Causes L1 Arrest and Other Physiological Defects (A–C) Nomarski images of worms grown from eggs placed on RNAi plates. Scale bars, 100 μm. (A) Young adults had normal morphology and growth rates. (B) On the second day of adulthood, these animals displayed an egg-laying defect; eggs hatched inside the worms. Arrows point to the late embryos and hatched larvae inside a worm. (C) F1 generation arrested uniformly at the first larval stage (L1), and larvae arrested for 4–5 d died. (D–F) Images of worms derived from late larvae (L2–L4) placed on the RNAi plates. (D) The F1 progeny of worms developed from the treated larvae had smaller size and a scrawny morphology compared to the wild type shown in (A). Scale bar, 100 μm. (E) These animals produced very few oocytes, some of which gave rise to embryos and L1 worms. White arrows indicate embryos. Some oocytes remained unfertilized (black arrow). Scale bar, 10 μm. (F) The proximal part of the gonads undergoes deterioration resulting in sterility. The white arrow indicates spermatica, the black arrow shows an abnormally amorphous oocyte, and the two-way arrow points to the clumsy gonad arm that is finely ordered in wild-type animals. Scale bar, 10 μm. When parental animals were subjected to elo-5 RNAi at later larval stages (L2–L4), their progeny did not arrest in L1 but continued to develop into adulthood. These animals had no obvious defects in locomotion, pharyngeal pumping, intestinal contractions, chemotaxis response, touch sensitivity, or general anatomy (unpublished data). However, the growing worms became progressively sick ( Figure 5 D– 5 F). The gonads appeared normal at the L4 and early adult stages, but after fertilization of one to ten oocytes, oogenesis became impaired. Gonad degeneration began with a pronounced vacuolization in the midsection of the gonad followed by the appearance of disorganized clumps of nuclei in the proximal part. An egg-laying defect became apparent and only a few progeny arose from these worms, which then arrested at L1. The development of the elo-5 RNAi phenotypes is likely due to a gradual elimination of the ELO-5–associated functions. Our data suggest that these functions are crucial for larval growth and development. We also obtained a likely null mutant of the elo-5 gene, elo-5(gk208), which has a 245-bp deletion eliminating the predicted first exon (Genome Science Center, BC Cancer Research Center, Vancouver, British Columbia, Canada). This allele phenocopies the L1 arrest phenotype of the elo-5(RNAi) animals. A Deficiency of C15ISO and C17ISO FAs Is Solely Responsible for the Defects Caused by elo-5(RNAi) We reasoned that if the defects observed in the elo-5(RNAi) animals resulted directly from the deficiency of C15ISO and C17ISO, then feeding these worms with C15ISO and C17ISO should mask a shortage of endogenous C15ISO and C17ISO and permit the animals to grow normally. As predicted, the C17ISO and C17anteISO supplements rescued the elo-5 RNAi defects (in 52 of 60 and 58 of 60 plates, respectively). A partial rescue was observed on the plates supplemented with C15ISO and C15anteISO (23 of 38 and 20 of 28 plates, respectively). Corroborating results were obtained when homozygous elo-5(gk208) animals supplied with C17ISO grew normally. In sharp contrast, neither saturated or mono- or polyunsaturated FA molecules (C16:0, C16:1 n7, C17:0, C18:3 n6), mmBCFAs with shorter or longer backbones (C13ISO, C18ISO, C19ISO), nor polymethyl branched phytanic acid were able to rescue or reduce defects (0 of 30 plates in each experiment). Therefore, we have determined that only dietary 17-carbon mmBCFAs are competent to bypass the biochemical defect caused by loss of ELO-5 function. GC analysis of FA composition in elo-5(RNAi) worms grown on supplemented plates revealed that only C17ISO and C17anteISO are significantly incorporated into lipids ( Figure 6 A– 6 C). Because the addition of C15ISO did not result in elongation to C17ISO ( Figure 6 A), we wanted to determine whether ELO-6 was capable of extending an FA backbone in the absence of ELO-5, or whether the supplied free FA molecules could enter a different metabolic pathway, for instance, a degradation pathway. To distinguish between these two possibilities, we added mmBCFA-producing bacteria on top of the elo-5(RNAi) feeding Escherichia coli strain (HT115), which lacks mmBCFAs. This mmBCFA-producing strain was identified by chance; we noticed that in the presence of a certain bacterial contaminant the animals could overcome the elo-5(RNAi) effects. Using a rapid bacterial identification method ( Lane et al. 1985 ), we determined the contaminant to be Stenotrophomonas maltophilia. GC analysis revealed that this bacterial strain produced a high quantity of C15ISO and C15anteISO but not 17-carbon mmBCFAs ( Figure 6 D). GC analysis of elo-5(RNAi) animals fed with S. maltophilia indicated that they not only accumulated bacterial C15ISO and C15anteISO but also efficiently elongated these FA species to C17ISO and C17anteISO, which are absent in S. maltophilia ( Figure 6 D and 6 E). This suggested that elongation from C15ISO to C17ISO mmBCFA was not impaired in the elo-5(RNAi) animals. Therefore, ELO-6 function remains intact in elo-5(RNAi). Apparently, C15ISO added to the plates could not be utilized by ELO-6 whereas C15ISO-CoA and/or C15anteISO-CoA originating from the bacterial food could, suggesting that free and esterified mmBCFAs were likely to enter alternative pathways. Figure 6 The FA Composition in Worms Maintained on elo-5 RNAi Plates Supplemented with FA or with S. maltophilia Enriched with C15ISO and C15anteISO FA Black arrowheads indicate positions of mmBCFAs. (A) Animals grown with C15ISO supplements were partially rescued to the wild-type phenotype; however, no accumulation of C15ISO or its elongation to C17ISO was detectable. (B and C) Animals grown with the (B) C17ISO or (C) C17anteISO supplements were fully rescued. Peaks corresponding to C17ISO and C17anteISO are prominent. (D) The FA composition in S. maltophilia. Arrowheads point to the major FAs, C15ISO and C15anteISO. (E) The elo-5(RNAi) animals are able to elongate dietary C15ISO and C15anteISO into C17ISO and C17anteISO. Arrowheads indicate mmBCFAs. The horizontal arrow illustrates the elongation from C15 to C17 mmBCFA. The essential roles of C15ISO and C17ISO were also supported through an examination of the elo-5(gk208) deletion mutant. The homozygous mutants grew without any obvious morphological defects when maintained on the plates supplemented with C17ISO or seeded with S. maltophilia. However, removal of the FA supplements or S. maltophilia by bleaching resulted in the same L1 arrest phenotype seen for the elo-5(RNAi) worms. L1 Arrest of the elo-5(RNAi) Animals Is Reversible and Related to the Variations in Levels of C17ISO during Development We then asked if elo-5(RNAi) animals arrested at the L1 stage could be recovered by adding the 17-carbon mmBCFA supplements. Indeed, C17ISO and C17anteISO could effectively release L1 larvae from the developmental arrest; about 50% of 2-d-arrested and 1% of 4-d-old L1 were rescued to full growth and proliferation. Since C17anteISO could not be detected in the laboratory animals under normal conditions of culturing, C17ISO appeared to be the principal molecule conveying the ELO-5 function. Therefore, the L1 arrest of the C17ISO-depleted worms is both completely penetrant and reversible, indicating that C17ISO plays a critical role in growth and development at the L1 stage. The analysis of the FA levels of staged worms revealed that the C17ISO level increases gradually from a relatively low level at L1 to its peak in gravid adults containing eggs ( Figure 7 A). Based on the analysis of green fluorescent protein (GFP) reporter constructs (unpublished data) and in situ hybridization data (results from NextDB by Y. Kohara, Tokyo, Japan), neither elo-5 nor elo-6 is significantly expressed in eggs or L1. Therefore, C17ISO likely accumulates in embryos during oogenesis. It may be directly transported from gut to gonads, since both ELO-5 and ELO-6 were expressed mainly in the gut and since feeding C17ISO rescued the elo-5 mutant phenotypes. When RNAi-mediated disruption of elo-5 occurs at the L1 stage of a parent and consequently blocks C17ISO synthesis from that stage on, the eggs and L1 animals of the next generation are expected to contain a critically low concentration of C17ISO, halting further development. Because the arrested L1 can be rescued by a dietary supply of the mmBCFA, the deficiency is not likely to cause critical defects during the embryonic and early postembryonic periods. Figure 7 A Fluctuation of the C17ISO Amounts in Development (A) Relative amounts of C15ISO and C17ISO in the worm samples collected at different developmental stages. The amount of the mmBCFA molecule is presented as the percentage of total FA in each sample. Grey bars, C15ISO; black bars, C17ISO. (B) Proposed relationship between the levels of mmBCFA during development and the RNAi effects. Depending on the time of RNAi onset, the amount of C17ISO in F1 eggs varies. If elo-5 is suppressed in parental animals after they have begun to synthesize mmBCFA, then their eggs will have a reduced C17ISO level that is still above the critical low level, which permits these animals to grow but causes them to display gonadal defects. These worms produce a small number of progeny that is then arrested in L1. If parental animals are treated with elo-5(RNAi) right after hatching, they are unable to initiate mmBCFA biosynthesis and the levels of C15ISO and C17ISO in their eggs are reduced to below the critical low level, resulting in L1 arrest of their progeny. If elo-5 RNAi is applied to the parent worms at or after the L2 larval stage, when the amount of C17ISO has already been elevated and/or the RNAi effect is less penetrant, the progeny may receive sufficient C17ISO to pass the L1 arrest stage. The resulting animals, however, become visibly unhealthy at the L4 and adult stages as mentioned earlier, suggesting that C17ISO also plays a role in late developmental stages. Based on these results, we propose a relationship between the amounts of C17ISO and developmental stages ( Figure 7 B). In this model, the level of C17ISO is monitored at the first larval stage and the decision is made whether to proceed or pause in development. The analysis of GC data from staged animals has also indicated that the variation of the C17ISO level is correlated with only two other FA species, suggesting a potential compensatory and coregulatory mechanism. The C17ISO Level Correlates with the Levels of Two Other FAs during Development FA homeostasis implies that relative amounts of various FA species are coordinated and balanced for optimal performance. To obtain information that may help us understand why and how numerous FAs and their specific metabolic enzymes are maintained in nature, we carried out analysis to determine a possible correlation between changes in the levels of C17ISO and other FAs detected in worms. We have analyzed a large amount of GC data ( n = 50) obtained from mixed populations of wild-type animals where the fractions of eggs, larvae, and adults randomly varied. We also included GC data separately obtained from staged worms: eggs, L1, L2, L3, L4, and gravid adults. We found that the amounts of C17ISO significantly correlated with only two other FA molecules: linoleic acid (C18:2 n6) and vaccenic acid (C18:1 n7) ( Figure 8 ). A potential physiological significance of these correlations is intriguing. Figure 8 Correlation between the Level of C17ISO and the Levels of Linoleic and Vaccenic Acids during Development Graphical illustrations of the correlation between the levels of C17ISO and (A) vaccenic acid and (B) linoleic acid. Data were obtained by GC analysis of synchronized populations of worms. Combined with the GC measurements generated from 50 additional samples (see Materials and Methods), these data were used to calculate correlation coefficients: CORREL C17ISO/C18:2 n6 = 0.82772, T-TEST = 6.54814 × 10 −7 , and CORREL C17ISO/C18:1 n7 = −0.85162, T-TEST = 4.74094 × 10 −5 . Black bars, C17ISO; white bars, vaccenic acid; grey bars, linoleic acid. The observed negative correlation between the levels of C17ISO and C18:1 n7 throughout development may indicate a compensatory adjustment important for physiological functions, such as retention of the cell membrane physical properties. mmBCFAs and monounsaturated straight-chain FAs have been previously implicated in regulating membrane fluidity, which depends on the ratio of saturated FA to monounsaturated and branched-chain FA content in bacterial cells ( Rilfors et al. 1978 ; Suutari and Laakso 1994 ; Cropp et al. 2000 ). An elevation in monounsaturated FA amounts in response to the decrease of branched-chain FAs, but not vice versa, was observed in Streptomyces avermitilis ( Cropp et al. 2000 ), suggesting that monounsaturated FAs may sense a state of membrane fluidity. In the elo-5(RNAi) -treated worms, a substantial loss of C15ISO and C17ISO is also accompanied by a change in the FA composition, most noticeably by the elevation in C18:1 n7 (see Figure 3 C), a result consistent with the above observation. To estimate the effect of the C15ISO and C17ISO deficiency on the membrane saturation, the saturation index (SI = [saturated FA]/[mmBCFA + monounsaturated FA]) was calculated. No significant differences were detected in elo-5(RNAi) worm compared to wild type (SI = 0.325 ± 0.011 [ n = 6] and SI = 0.320 ± 0.032 [ n = 5], respectively). Therefore, elo-5(RNAi) may not cause massive cell membrane dysfunction. A positive correlation between the amounts of C17ISO and C18:2 n6 may suggest a potential common function during development. In addition to the importance of linoleic acid as a substrate for polyunsaturated FA biosynthesis, its hydroxylated fatty acid derivative (HODEs) is known as a signaling molecule affecting chemotaxis ( Kang and Vanderhoek 1998 ), cell proliferation ( Eling and Glasgow 1994 ), and modulation of several enzymatic pathways ( Hsi et al. 2002 ). A correlation between C17ISO and linoleic acid may also suggest a similar regulation of biosynthesis of the two molecules. The changes in the FA composition associated with a decrease in C15ISO and C17ISO indicate that the metabolism of straight-chain FA species is responsive to the mmBCFA levels and suggest a cross regulation. Interestingly, in the elo-5(RNAi) animals fed with C15ISO or C15anteISO containing bacterial supplement (S. maltophilia), the FA composition was significantly altered (see Figure 6 E). It appears that mmBCFAs become principal components in a range of 16–18-carbon FAs. This suggests that large quantities of mmBCFAs are not toxic. In contrast, because these worms grow and proliferate well, mmBCFAs seem to be efficient substitutes for saturated and monounsaturated straight-chain FAs. The Worm SREBP Homolog Controls Production of Branched-Chain FAs In mammals, straight-chain FA biosynthesis depends on the 1c isoform of sterol regulatory element binding protein (SREBP-1c), which promotes the expression of FA metabolic enzymes ( Edwards et al. 2000 ; Horton 2002 ; Matsuzaka et al. 2002 ). There is only one protein in C. elegans that is homologous to mammalian SREBPs, Y47D3B.7 (the gene has been named lpd-1, for “lipid depleted 1”) ( McKay et al. 2003 ). McKay and coauthors have shown that worms treated with lpd-1 RNAi display a lipid-depleted phenotype. They have also shown that lpd-1 regulates the expression of several lipogenic enzymes, acetyl-CoA carboxilase (ACC), FAS, and glycerol 3-phosphate acyltransferase (G3PA) ( McKay et al. 2003 ). Thus, similar to its mammalian homolog, lpd-1 is involved in straight-chain FA biosynthesis. We wanted to see if lpd-1 also plays a role in mmBCFA metabolism. We first applied RNAi to lpd-1 and determined the FA composition of the mutant worms. As expected, the FA content of treated animals was significantly changed, but surprisingly the most reduced were the levels of C15ISO and C17ISO ( Figure 9 ). Also significantly reduced was the amount of C18:2 n6. In contrast, the C16:0 level was elevated. These data indicate that, in addition to regulating the first steps of global FA biosynthesis through the activation of the ACC and FAS transcription, the worm SREBP homolog regulates mmBCFA elongation as well as desaturation of straight-chain FAs. Figure 9 RNAi of the C. elegans SREBP Homolog Alters the FA Composition (A and B) The GC profiles of (A) wild-type and (B) lpd-1(RNAi) -treated worms. (C) A summary of several independent GC runs. Bars represent the percentages of total FAs. The levels of C15ISO, C17ISO, and C16:0 are significantly altered by the RNAi treatment. Black arrowheads point to differences in the C15ISO and C17ISO amounts. Grey arrowheads indicate the changes in palmitic acid, C16:0. As reported previously, disruption of lpd-1 through a mutation or RNAi injection caused early larval arrest ( McKay et al. 2003 ). The effect of lpd-1 RNAi feeding in our experiments was apparently less severe. The RNAi-treated animals displayed slow growth, morphological abnormalities, and egg-laying defects but no larval arrest. Supplementing C17ISO to the plates did not significantly rescue these defects. LPD-1 and LPD-2 Diverge in Functions LPD-2 (C48E7.3) is another C. elegans homolog of a mammalian lipogenic transcription factor, CCAAT/enhancer-binding protein (C/EBP). McKay and coauthors have shown that the lpd-2(RNAi) and lpd-1(RNAi) phenotypes are quite similar; affected worms are defective in growth, pale and scrawny in appearance, and lacking in fat content ( McKay et al. 2003 ). They have also shown that LPD-1 and LPD-2 control the expression of the same lipogenic enzymes: ACC, FAS, ATP-citrate lyase, and G3PA. We tested to see if LPD-1 and LPD-2 function similarly in the regulation of mmBCFA biosynthesis. In contrast to the result from lpd-1 (RNAi), the FA composition in lpd-2(RNAi) worms was not significantly different from that of wild-type animals even though these animals had a noticeably sick appearance (unpublished data). This result suggested that, in addition to having some common targets, LPD-1 and LPD-2 have distinct functions. LPD-1 is important for production of mmBCFAs as well as other very-long-chain FAs, whereas LPD-2 has no specificity for any particular type of FA. elo-5 and elo-6 Are Likely Targets of LPD-1 The changes in FA composition observed in lpd-1(RNAi) would be consistent with downregulation of elo-5, elo-6 (decrease in mmBCFA), elo-2 (increase in C16:0) ( Kniazeva et al. 2003 ), and Δ 9- and/or Δ 12 -desaturase genes (decrease in C18:2 n6). The genes encoding mammalian orthologs of the C. elegans elo-2 and Δ 9 -desaturase genes are known targets of SREBP-1c ( Edwards et al. 2000 ; Horton 2002 ; Horton et al. 2002 ). To examine if elo-5 and elo-6 are targets of lpd-1, we analyzed the expression of elo-5, elo-6, and lpd-1. Evaluation of the expression from an lpd-1Prom :: GFP fusion construct (a gift of J. Graff) in transgenic animals revealed that, in addition to the previously reported expression in intestinal cells ( McKay et al. 2003 ), the construct is strongly expressed in a subset of head neurons ( Figure 10 A— Figure 10 D). Using a lipophilic dye, DiI, which highlights chemosensory ciliated neurons, we identified these neurons as amphids ( Murphy et al. 2003 ). In the strains carrying elo-5Prom :: GFP and elo-6Prom :: GFP reporter constructs, GFP fluorescence was also detected in the gut and several head neurons, including amphid neurons ( Figure 10 E– 10 H and Figure 2 ). Figure 10 The Expressions of elo-5 and lpd-1 Reporter Constructs Are Spatially Similar (A and B) Nomarski and GFP-filtered images of an adult animal containing the lpd-1Prom :: GFP construct, showing strong expression in two symmetrical head neurons, each of which has processes to the nose and around a nerve ring. Scale bars, 10 μm. (C) DiI staining of amphid neurons in lpd-1Prom :: GFP (dsRed filter). Arrows indicate neuronal nuclei shown in (D). Scale bar, 10 μm. (D) GFP expression in the animal shown in (C). Scale bar, 10 μm. (E and F) Nomarski and GFP-filtered images of an animal containing elo-5Prom :: GFP, revealing fluorescence in the similar amphid neuron. Scale bar, 7.5 μm. (G and H) The intestinal and intestinal-muscle GFP expression in (G) lpd-1Prom :: GFP and (H) elo-5Prom :: GFP constructs. Scale bar, 7.5 μm. If LPD-1 promotes elo-5 and elo-6 expression, then RNAi of lpd-1 should alter GFP intensity in elo-5Prom :: GFP and elo-6Prom :: GFP reporter strains. The level of GFP expression driven by elo-5 and elo-6 promoters is high in conventionally cultured animals. In the worms maintained on the lpd-1(RNAi) plates, the expression was noticeably weakened, suggesting a downregulation of the promoter activities ( Figure 11 A– 11 D). No significant changes in GFP expression were detected in a control strain containing a kqt-1Prom :: GFP construct that also expresses GFP in head neurons and the gut (unpublished data). Figure 11 The Expression of GFP Fusion Constructs Suggests the Involvement of lpd-1, acs-1, and pnk-1 in mmBCFA Biosynthesis (A–D) elo-5Prom :: GFP expression is downregulated in the lpd-1(RNAi) background. Scale bars, 100 μm. (A and C) GFP-filtered images of (A) elo-5Prom :: GFP and (C) elo-6Prom :: GFP in wild-type animals, showing the characteristic bright intestinal fluorescence. (B and D) GFP-filtered images of (B) elo-5Prom :: GFP and (D) elo-5Prom :: GFP in lpd-1(RNAi) animals, revealing diminished fluorescence in the gut. (E–H) lpd-1 expression is upregulated in neurons of the elo-5(RNAi) animals deficient for C15ISO and C17ISO. Scale bars, 15 μm. (E and F) Nomarski and GFP images of wild-type L1 larvae containing lpd-1Prom :: GFP. (G and H) Nomarski and GFP images of elo-5(RNAi) -treated animals (L1 arrested) containing lpd-1Prom :: GFP, showing a visibly brighter fluorescence than that seen in (E) and (F). Circles are centered on the pharyngeal back bulb. (I–K) acs-1Prom :: GFP expression is upregulated in the elo-5(RNAi) animals deficient for C15ISO and C17ISO. Panels show GFP images of acs-1Prom :: GFP animals grown on the (I) control, (J) elo-5(RNAi), and (K) lpd-1(RNAi) plates. The fluorescence from acs-1Prom :: GFP in (J) is significantly stronger than that in (I). Scale bars, 100 μm. (L–N) pnk-1Prom :: GFP expression is upregulated by elo-5(RNAi) but downregulated by lpd-1(RNAi). Panels show GFP images of pnk-1Prom :: GFP animals grown on the (L) control, (M) elo-5(RNAi), and (N) lpd-1(RNAi) plates. The fluorescence of the fusion construct is stronger in (M) but weaker in (N) than that in the control (L). Scale bars, 100 μm. To test if the disruption of FAS, a target of LPD-1 ( McKay et al. 2003 ), could contribute to the observed decrease of C15ISO and C17ISO in lpd-1(RNAi), we analyzed FA composition in FAS(RNAi) strains. There is one predicted FAS gene, F32H2.5, and its shorter homolog, F32H2.6, in the C. elegans genome. The latter can only encode the N-terminal portion of the protein. These genes share extended nucleotide identity, and RNAi of one could thus possibly affect the other. Consistent with a critical role for FAS in the first steps of FA biosynthesis, the RNAi-mediated disruption of F32H2.5 and F32H2.6 resulted in multiple defects and a lethal growth arrest (unpublished data). The FA composition (the content and relative amounts of various FA species) of the affected animals remained, however, unchanged. This suggests that disruption of FAS does not selectively alter FA biosynthesis and that neither FAS protein is specific for mmBCFA. Therefore, downregulation of FAS by loss of lpd-1 cannot account for the severe deficiency of mmBCFA in lpd-1(RNAi). Thus, we have shown that disruption of lpd-1 affects C15ISO and C17ISO biosynthesis. The fact that lpd-1, elo-5, and elo-6 are expressed in the same cells concurrently and that the GFP reporter analysis indicated that elo-5 and elo-6 transcription is downregulated in the absence of lpd-1 suggests that elo-5 and elo-6 are likely to be the targets of lpd-1. Since ACC and FAS catalyze the first steps in the biosynthesis of straight-chain FAs while ELO-5 and ELO-6 extend mmBCFA molecules, LPD-1 appears to integrate conventional and “unusual” FA biosyntheses. It seems reasonable to predict that in order to differentiate between these metabolic pathways and mediate compensatory or adaptive changes in FA composition, LPD-1 must interact with other factors such as nuclear receptors activated by specific FA ligands. It is thus important to screen for such interactions to better understand FA homeostasis in C. elegans. A Reciprocal Correlation between lpd-1 Expression and mmBCFA Levels Because mammalian SREBP-1c regulates polyunsaturated FA biosynthesis and is feedback inhibited by polyunsaturated FAs ( Jump 2002 ), we asked if lpd-1 could be regulated by mmBCFAs at the transcriptional level. Our microarray data (discussed below) indicated a 1.68-fold upregulation of lpd-1 in the elo-5(RNAi) animals, while no changes were detected in its levels between samples from wild-type animals at different developmental stages (see Materials and Methods ). To examine the influence of the mmBCFA deficiency on lpd-1 expression, we grew the lpd-1Prom :: GFP -containing strain on the elo-5(RNAi) and control plates to compare GFP fluorescence. No obvious difference in the GFP expression driven by the lpd-1 promoter in intestinal cells was detected on the elo-5(RNAi) plates versus the control plates. A modest change in the transcription level (1.68-fold) could be masked by a variability of the expression between individual animals and even between individual cells (unpublished data). In contrast to the observation for the intestinal cells, a strong induction of GFP was detected in amphid neurons of lpd-1Prom :: GFP ; elo-5(RNAi) animals ( Figure 11 E– 11 H). This suggests that a chronic deficiency of mmBCFA in elo-5(RNAi) animals may transcriptionally stimulate LPD-1 production, at least in neuronal cells. Collectively, our results suggest that the relationship between lpd-1 and C15ISO/C17ISO is reciprocal; while downregulation of lpd-1 transcription results in the C17ISO deficiency, the C15ISO and C17ISO deficiency upregulates lpd-1 transcription at least in a subset of cells. Therefore, the worm SREBP homolog, LPD-1, may play an important role in mmBCFA homeostasis. Screening for Additional Genes Involved in mmBCFA Homeostasis Because C15ISO and C17ISO play critical roles in animal development and growth, we suspected mechanisms might exist to respond to and regulate their levels. Regulation of mmBCFA homeostasis may involve transcription factors, metabolic enzymes, and transport and binding proteins. It is reasonable to suggest that a deficiency of mmBCFA triggers a compensatory alteration in the expression of these genes. It is also possible that a comparative analysis of global gene expressions between wild-type and mmBCFA-deficient animals may reveal these potential changes and the changes underlying developmental and growth functions of mmBCFA. We used DNA microarray analysis to compare the total gene expression in elo-5(RNAi) and wild-type animals. To select candidate genes, we applied restrictive criteria and excluded genes of which the expression was also changed in the spt-1(RNAi) strain ( Protocol S1 and Table S1 ). The spt-1 (C23H3.4) gene encodes a predicted C. elegans homolog of serine-palmitoyl transferase subunit 1. RNAi of spt-1 strongly affects the FA composition without reducing the C15ISO or C17ISO levels (unpublished data). The F1 generation of spt-1(RNAi) animals developed gonadal and egg-laying defects that were similar to the phenotype of F1 animals from parents treated with elo-5(RNAi) at a late larval stage (described earlier; see Figure 5 B and 5 F). We thought that by deselecting genes that have altered expressions in spt-1(RNAi), we would be able to eliminate variations in gene expressions unrelated to the mmBCFA deficiency. Such variations might emerge from altered straight-chain FA metabolism and from general sickness. Here, we discuss the analysis of the first set of candidate genes that are differentially expressed in elo-5(RNAi) and may relate to the C15ISO and C17ISO homeostasis. Twenty-five genes were selected in the screen ( Table 1 ) and each was functionally tested by RNAi and GC analysis for its role in C15ISO and C17ISO metabolism. RNAi of four of these genes ( pnk-1 [C10G11.5], nhr-49 [K10C3.6], acs-1 [F46E10.1], and C27H6.2) significantly affected the FA composition ( Figure 12 ). All four genes encoded products structurally homologous to the known proteins (PNK-1, human pantothenate kinase; NHR-49 , nuclear hormone receptor; ACS-1, very-long-chain FA CoA ligase; and C27H6.2, RuvB-like DNA binding protein). Figure 12 RNAi of Four Candidate Genes with Altered Expression in elo-5(RNAi) Worms Affects the FA Composition (A) GC profile of the wild type. (B–E) GC profiles of the RNAi-treated worms. (B–D) RNAi of the three genes resulted in a decrease of the C17ISO or both C15ISO and C17ISO levels, indicated by black arrowheads. In addition, a significant elevation in straight-chain saturated FA levels, indicated by grey arrowheads, is observed in K10C3.6(RNAi). (E) C27H6.2(RNAi) does not cause significant changes in mmBCFA but results in an elevation of straight-chain monounsaturated FA and C18:1 n7, indicated by white arrowheads. Statistical analysis of several GC runs on each of the samples was also carried out (unpublished data). Table 1 Candidate Genes and Their Encoded Proteins Selected from Microarray Data for Functional Tests (RNAi and GC Analysis) a Open reading frames (ORFs) predicted by the C. elegans Genome Project (WormBase.org) b Data from comparing arrays from the experimental sample with that from a baseline control sample (Sample I; see Protocol S1) Analysis of the Candidate Genes Circumstantial evidence suggests that these four candidate genes may be involved in feedback regulation of mmBCFA biosynthesis. First, the expression of these genes is not variable in nature, as judged by a comparison of the microarray data obtained from developmentally different populations of N2 ( Protocol S1 ) as well as for vulval development pathway mutants (data obtained for an unrelated project, J. Chen, personal communication). Second, the direction of the changes for three of the genes is in concordance with the proposed feedback regulation: pnk-1, nhr-49, and acs-1 were upregulated in C17ISO-deficient elo-5(RNAi) . Lastly, a functional analysis shows that these three candidate genes are required for the normal level of mmBCFA production (RNAi of the genes affects mmBCFA production). The fourth candidate gene, C27H6.2, affects the level of vaccenic acid (C18:1 n7), which is related to the levels of mmBCFA (see Figure 8 ), suggesting cross-talk. To detect potential feedback regulation involving acs-1 and pnk-1, we made reporter strains with GFP expression driven by acs-1 and pnk-1 promoters, acs-1Prom :: GFP and pnk-1Prom :: GFP, respectively. These two genes showed a higher degree of upregulation than the other candidates according to the microarray data. In addition, RNAi of these two genes resulted in a significant loss in the mmBCFA fraction. The GFP fluorescence from acs-1Prom :: GFP and pnk-1Prom :: GFP was readily detectable in the gut. Expression of acs-1Prom :: GFP was also detected in the canal-associated neurons in the head neurons and vulval cells. A comparison of synchronized animals grown on the control and elo-5(RNAi) plates indicated a significantly brighter fluorescence in the RNAi worms (see Figure 11 I, 11 J, 11 L, and 11 M), suggesting upregulation of acs-1 and pnk-1 under C15ISO or C17ISO deficiency. These results were in concordance with the microarray data. Moreover, pnk-1, but not acs-1, seemed to be regulated by LPD-1 because pnk-1Prom :: GFP expression was significantly reduced on lpd-1(RNAi) (see Figure 11 L and 11 N). It was interesting to note that the pnk-1 and acs-1 genes were previously selected in two different screens as potential targets of the daf-2/daf-16 (Y55D5A.5 and R13H8.1, respectively) pathway. pnk-1 had been identified in a screen for genes affecting the life span and metabolism of C. elegans through analysis of promoter regions, and it was confirmed as a direct target of DAF-16, a forkhead transcriptional factor ( Lee et al. 2003 ). acs-1 had been identified in a microarray screen for DAF-16 targets that influence life span ( Murphy et al. 2003 ). A third gene, nhr-49, had been previously selected in a screen for fat regulatory genes ( Ashrafi et al. 2003 ). It was shown that RNAi of this gene leads to an increase in fat accumulation in affected animals. Our analysis of nhr-49(RNAi) animals showed that reduction of nhr-49 activity results in upregulation of saturated FA biosynthesis that may contribute to fat accumulation. Although the regulatory path for this process remains unknown, the involvement of daf-2 has not been ruled out. A potential link between the candidate genes and DAF-2/insulin signaling is very intriguing. The C. elegans insulin-signaling pathway is involved in sensing nutritional state and metabolic conditions as well as controlling growth and diapause ( Kimura et al. 1997 ; Ailion et al. 1999 ). As we report in this paper, a mmBCFA deficiency causes transient L1 arrest. This phenotype strikingly resembles L1 arrest of worms hatched in the absence of food (a method commonly used to obtain synchronized animals). An investigation of possible roles for mmBCFA in food sensation and insulin signaling pathways is underway. Downregulation of the fourth candidate gene, C27H6.2, may result in a significant increase in monounsaturated FA levels ( Figure 12 ). This is consistent with the enlarged fraction of monounsaturated FAs observed in the elo-5(RNAi) animals (see Figure 3 C). Downregulation of C27H6.2 may have an adaptive effect to compensate for the loss of mmBCFAs in cell membranes. If so, C27H6.2 may be part of a mechanism that senses and tunes physical properties of membranes. C27H6.2 is homologous to RuvB/TIP49a/Pontin52, an evolutionarily conserved protein essential for growth and proliferation ( Kanemaki et al. 1997 ; Bauer et al. 1998 ; Qiu et al. 1998 ). Its mammalian ortholog acts as a transcriptional cofactor that binds to β-catenin, TATA-box binding protein, and likely to a number of other diverse transcription factors ( Bauer et al. 1998 ). Concluding Remarks Two mmBCFAs are normally detected in C. elegans: C15ISO and C17ISO. A deficiency of these FAs is lethal and cannot be compensated by any other FA present, indicating their crucial importance for growth and development. There are two sources of C15ISO and C17ISO available for worms. First, they possess a system for mmBCFA biosynthesis that includes two FA elongation enzymes, ELO-5 and ELO-6, which are regulated at least in part by the nematode homolog of SREBP-1c (lpd-1). Second, worms may obtain mmBCFAs from their diet (bacteria). Therefore, C. elegans is able to produce, activate, transport, and utilize mmBCFAs and is vitally dependent on this system. The level of C15ISO and C17ISO in eggs appears to be critical for growth and development, as animals depleted of C15ISO or C17ISO completely arrest at the L1 stage. The uniformity and reversibility of the arrest would be consistent with a regulatory role in growth and development for these mmBCFAs or for more complex lipid molecules containing them. However, it cannot be ruled out that the arrest is due to the failure of a metabolic or structural function that is essential for growth and development at the first larval stage. In addition, C15ISO and C17ISO may directly or indirectly regulate genes involved in FA homeostasis. Consistent with this, their deficiency triggers a large alteration in gene expression that may reflect a complex feedback mechanism. Among the potentially responsive genes are transcription factors and metabolic genes. Ubiquitous and unattended mmBCFAs come forth as physiologically important molecules that regulate essential functions in eukaryotes. Many interesting questions regarding mmBCFAs remain to be addressed. What are the other components of the mmBCFA biosynthetic machinery? What are the components of their transport system? Does an organism have a mechanism by which the mmBCFA level is measured? What are the signaling pathways involved in the mmBCFA responses? How do mmBCFAs exert their physiological function? Do mmBCFAs act alone or as parts of more complex lipids? How are mmBCFAs synthesized in mammals? Lastly, what are the specific physiological functions of mmBCFAs in mammals? Both genetic and biochemical approaches will be taken to address these questions. Materials and Methods RNA interference by feeding The RNAi feeding vectors were either made in our laboratory using Taq PCR and cloning genomic fragments into a double T7 vector, pPD129.36 (gift of A. Fire), or obtained from the C. elegans whole genome RNAi feeding library (J. Ahringer, MRC Geneservice, Cambridge, United Kingdom). The RNAi feeding strain was E. coli HT115 transformed with either empty pPD129.36 vector (controls) or with dsRNA-producing constructs. Plates were prepared as described in Kamath et al. (2001) . Unless stated differently, wild-type N2 Bristol animals were plated as synchronized adults. To obtain synchronized worms of various stages, a large quantity of N2 gravid adults were collected, bleached, and grown to the required stage on HT115 that had been transformed with pPD129.36 (control). GC analysis A mixed population of well-fed worms were washed off the plates with water, rinsed 3–4 times, and, after aspirating away water, frozen at −80 °C. FA methyl esters and lipid extraction were performed as described in Miquel and Browse (1992) . GC was performed on the HP6890N (Agilent, Palo Alto, California, United States) equipped with a DB-23 column (30 m × 250 μm × 0.25 μm) ( Kniazeva et al. 2003 ). Each experiment was repeated at least five times. Average values and standard deviations were then calculated for each of the compounds in the experiments. Staging worms to test for FA composition After bleaching gravid adults, an aliquot of the eggs was set apart, and the rest was incubated overnight in M9 at room temperature. On the next day, an aliquot of L1 was frozen for GC analysis. The rest of L1 was plated on agar plates. Subsequently, L2, L3, L4, young adults, and adults along with hatched L1 were collected as the separate samples. Mixed populations of worms starved for 24–100 h were included in the experiment to monitor a possible effect of the starvation. Phenotype rescue using FA supplements. Ninety microliters of the 4 mM solution of FA (Sigma, St. Louis, Missouri, United States) in 1% NP40 was dropped on the side of the bacterial lawn that contained either elo-5 dsRNA-producing plasmid or the control HT115 vector. Two synchronized young adults were plated and their progeny was scored 4 and 5 d later. Each experiment was performed in at least 30 replicates. For recovering elo-5(RNAi) worms from L1 arrest, wild-type adults were placed on the elo-5(RNAi) plates. Four days later, their progeny was removed and eggs of the next generation were left on the plates. Hatched L1 were kept for 2 or 4 d before transferring as agar chunks to new elo-5(RNAi) plates. FA supplements were added to spots next to the chunks. Ten plates were prepared for each FA supplement. Control plates contained no supplements. To verify that an addition of supplements did not affect RNA interference per se, we used let-418(RNAi) animals, which have a sterile phenotype, as a control. Neither C15 nor C17 mmBCFA added to let-418(RNAi) plates modified the expected phenotype. Designing of GFP reporter constructs. To prepare the GFP fusion constructs, genomic fragments were PCR amplified and cloned in frame into one of the GFP fusion vectors (gift of A. Fire). The locations of the genomic fragments and PCR primers used are listed below: (1) elo-5Prom :: GFP, starting at 3.894 kb genomic upstream of the first codon and ending 4 bp into the first exon; primers, F-BamHI-TTTAGGTCATTTTTTGAGTCGCCA and R-BamHI-TAGTCTGGAATTTTGAAATTGAACGG; vector, pPD95.69; (2) elo-6Prom :: GFP, a 4.764-kb fragment covering 3,104 bp upstream and 1,660 bp downstream of the predicted start codon and ending 14 bp into the third exon; primers, F-Sph1-GCCCTTGGAAACCATCTACGACGAATC and R-Sma1-TCCGAACAGAACGACATAAGAGATTTCC; vector, pPD95.77; (3) acs-1Prom :: GFP, a 3.142-kb genomic fragment containing 3,048 kb upstream of the first predicted ATG and ending 24 bp into the second predicted exon; primers, F-SphI-CATAATTACTATTGCGTCACATG and R-SphI-CTCTTCCAAACTGGCGATGTCGA; vector, pPD95.69; (4) pnk-1Prom :: GFP, a 1.14-kb fragment that includes 937 bp upstream of the first predicted codon of the C10G11.5 and 203 bp downstream, ending 24 bp into the second exon; primers, F-SphI-TCGTACGATCGGACCATAGGCTAA and R-SphI-CTGATCCTCTGTAGCAGCGGCCCT; vector, pPD95.69. These constructs were injected into C. elegans at 10–50 ng/μl to form extrachromosomal arrays. In the case of acs-1, the extrachromosomal array had been integrated into the C. elegans genome. Staining chemosensory ciliated neuron with DiI Worms were soaked in a 5-μg/ml solution of DiI (Molecular Probes, Eugene, Oregon, United States) in M9 buffer for 1 h. They were then rinsed three times with M9 and visualized by fluorescence using the Texas Red filter. Correlation analysis The FA quantities obtained by GC were expressed as a percentage of the total. t test (two-tailed distribution) and correlation analysis were performed using the Microsoft Excel program. Visualization and scoring of the GFP expression in promoter::GFP lines Synchronized adults were placed on control (HT115 bacterial strain transformed with empty vector, pPD129.36) and RNAi (HT115 bacterial strain transformed with dsRNA construct) plates. Several worms of the next generation were picked from the control and RNAi plates and mounted on the same microscopic slide. GFP images were obtained with the fixed settings and exposure. Microarray analysis One young adult of the N2 Bristol strain was plated on each control and RNAi feeding plate. Control plates contained the E. coli HT115 strain transformed with empty pPD129.36 vector. Experimental RNAi plates contained E. coli HT115 transformed with corresponding dsRNA constructs. The growth conditions, RNA preparations, and data analyses are described in detail in Protocol S1 . Expression data are presented in Dataset S1 . Supporting Information Dataset S1 Microarray Expression Data (1.8 MB TXT). Click here for additional data file. Figure S1 Expression of Collagen Genes as an Indicator of Developmental Differences in Mixed Populations of Worms Samples I, II, and III represent mixed populations of wild-type animals started simultaneously from one young adult. Each was harvested at three time points, when mostly adults represented the F1 generation and the embryos and larvae in different proportions represented the F2 generation (see Protocol S1 ). Sample III corresponds to the most diverse mixture of worms. Numbers of collagen genes that were differentially expressed between pairs of samples are shown above or bellow the arrow brackets. Sample I and Sample III, which originated from the most distal mixed populations, have the largest number of differentially expressed collagens. Sample I and an experimental sample corresponding to the elo-5(RNAi) phenotype have a lower number of the changed collagen genes, suggesting that populations on these experimental and control plates are similar. (24 KB PPT). Click here for additional data file. Protocol S1 Microarray Data Analysis (37 KB DOC). Click here for additional data file. Table S1 Filtering Candidate Genes by Comparing Different Mutant and Wild-Type Samples (28 KB DOC). Click here for additional data file.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC514883.xml

539044

Picturing AIDS: Using Images to Raise Community Awareness

In Botswana, explicit color photos of people with AIDS have been used to spread knowledge, with the aim of saving lives

Southern Africa poses special problems for AIDS educators and health care workers. Because there is a strong tradition of oral communication in the region, written educational materials often do not have as much impact as the spoken word. We have found that using colour images of HIV/AIDS in a workshop setting to provoke discussion can be a useful alternative to more conventional, written materials. In this article, we discuss our experience of using such images to raise community awareness about the AIDS epidemic in Botswana. Who We Are Teaching-aids At Low Cost (TALC) is a nongovernmental organisation that supplies cheap teaching aids and books to raise standards of health care and standards of living—especially in poverty-stricken areas—worldwide ( http://www.talcuk.org/ ). The organisation has traditionally focused on developing countries, particularly sub-Saharan Africa and Asia. In recent years, TALC has become more global; it now distributes materials to more than 200 countries and sends educational materials on CD-ROM at no cost to health workers in developing nations. In 1964, TALC was founded at the London School of Hygiene and Tropical Medicine as a way of providing low-cost colour transparencies to help students from resource-poor countries to teach after they returned home. By the early 1980s, nearly half a million transparencies were being sold at cost each year. Those who used them came to appreciate how important colour images could be, particularly amongst people who have grown up in societies where knowledge is spread primarily through oral communication and less use has been made of the written word. Early on in the HIV/AIDS epidemic, we decided that our experience of distributing visual teaching materials could be used to spread information about this new pandemic, which was hitting African societies particularly hard. We produced four sets of 24 colour transparencies on HIV/AIDS, with a detailed accompanying text. Edwin Mapara: The Botswana Experience Today, there are an estimated 260,000 people in Botswana living with HIV. This—in a country with a total population of 1.6 million—gives Botswana a prevalence rate of 36.5%, the second highest in the world after Swaziland [1] . As a medical student in Zambia in 1985, I studied patients with Kaposi's sarcoma. The consultant in charge appreciated that this was due to HIV infection, but when she started to acknowledge this publicly, she was strongly censored by the existing authorities and was almost forced to leave the country. I realised that if this kind of denial persisted, the epidemic would spread more widely and would become an even greater disaster. I wanted to try to bring home to both the authorities and the African people the truth about the spread of the disease and the need for fundamental changes in sexual behaviour. In 1990, on qualifying, I took up a post at the Athlone Hospital, a 175-bed district hospital in the Lobatse region of Southern Botswana. I joined other health workers who shared my concerns. We started the Athlone Anti-AIDS Project to address HIV prevention and care both in the hospital and in the wider community. We began to have organised discussions with local people about HIV/AIDS. The response we heard was often, “You talk about this terrible disease, which may affect us, but show us a patient”. This is how we came to use a set of slides from TALC, in a teaching programme that the Ministry of Health in Botswana called “radical and insensitive”. We emphasised the essential messages about AIDS prevention by using coloured pictures of black Africans. These pictures included explicit images of ulcers on a penis and a vagina. The slides included clinical manifestations of HIV/AIDS (such as herpes zoster and Kaposi's sarcoma) and other sexually transmitted diseases, images that explained the basic virology and transmission of HIV ( Figure 1 ), and images about HIV prevention (such as condoms) and care (such as caring for orphans infected by HIV/ AIDS). Figure 1 Don't Worry—Only a Few Sticks This slide is used in workshops to show that while we only see a handful of patients with symptomatic HIV, many more of us are HIV positive and are infecting others; we do not know our HIV status, since we have not been tested and we have no symptoms. (Illustration: TALC) Showing these pictures to local people was hugely controversial. For example, some elderly participants walked out when they saw the explicit pictures. Some community members approached local political counsellors to voice their concerns about a “decay of culture” and a “lack of respect”. Some parents did not want their children to see the images at all, because “they would corrupt their morals and young minds” and would encourage children to “experiment with sex”. As the team leader, I was fined chickens on several occasions by local chiefs and elders for the “crime” of showing these explicit TALC slides. The government gave our teaching project very little support in the early days; we were even cautioned by the highest authorities at the Ministry of Health. The Church, too, wanted nothing to do with our programme of “loose morals”. Despite these obstacles, over the next ten years we held over a hundred workshops, which eventually involved all government departments and levels of society in Botswana. Today, we are still using the same pictures. In 2000, the United Nations Development Programme declared Athlone Hospital's initiative as one of the “best practices” in Botswana [2] , and it is being replicated nationwide. TALC slides have been shown from the pulpits of churches, and community members will ask for the colour pictures specifically when the team is invited to lead a workshop. Given the terrible impact that AIDS has had on the community, the same community members who once resisted our teaching project ask us angrily why doctors were not sufficiently aggressive in using pictures in the early days of AIDS. One telling statement made in a workshop was: “you doctors are to blame for what has happened to Africa, and particularly to our children. You should have done this ten years ago before one quarter of the population became infected. The blood of our children, who have died, rests on your heads”. Making the Best Use of Pictures In Botswana, I used a slide projector and occasionally a mobile electrical generator, but such equipment is not widely available in most African countries. As an alternative to using colour slides, TALC has developed a folded A4 (210 mm × 279 mm) sheet with 12 colour images as a way of presenting the important messages about HIV/AIDS. This leaflet is available on request; E-mail: info@talcuk.org (or mail TALC, P. O. Box 49, St. Albans, AL1 5TX, United Kingdom). In our experience, the slides or leaflet work best if you can get the participants to sit in small groups for discussion. Each group should have at least one set of pictures. In your introduction, mention that to talk about sex or death is not taboo in a world of AIDS. Encourage active participation by all. Show one picture at a time—”let the picture talk”—and do not initially look at the accompanying text. Ask the participants to describe what they see in English or in the vernacular. Encourage participants to work out for themselves what the message is in the picture. Discuss all the possible answers. Then look at the text that accompanies the picture. Provide an answer built from the participants' words. If appropriate, ask people about their own relevant experiences. Finally, ask the participants to pin the pictures to the wall, making sure that each picture is put up by a different participant. Revise the lessons learned at the end of the session. Revise again, weeks later, if possible. At the end of the day, the participants should be able to say, “we did it by ourselves”. The pictures show examples of how HIV/AIDS can affect people. They must not be thought of, or used, as a way to diagnose HIV/AIDS in participants or their relatives or friends. Emphasise to all participants that if they have any reason to suspect that they (or anyone they know) have HIV/ AIDS, they should attend a clinic where trained health workers can help them. Conclusion For people in Botswana, “seeing is believing”. Written descriptions are often not enough; showing pictures of herpes zoster, syphilis ulcers, or tuberculosis lymphadenopathy can be a powerful teaching tool. Once the initial shock is overcome, these colour pictures offer a straightforward way to demonstrate the realities of the disease far and wide.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539044.xml

522868

Computational identification of developmental enhancers: conservation and function of transcription factor binding-site clusters in Drosophila melanogaster and Drosophila pseudoobscura

27 predicted gene-regulatory regions in the Drosophila melanogaster genome were analyzed in vivo , confirming 15 active enhancer regions. A comparison with Drosophila pseudoobscura sequences revealed that conservation of binding-site clusters accurately discriminates functional regions from non-functional ones.

Background The transcription of protein-coding genes in distinct temporal and spatial patterns plays a central role in the differentiation and development of animal embryos. Decoding how the unique expression pattern of every transcript is encoded in DNA is essential to understanding how genome sequences specify organismal form and function. Understanding gene regulation requires discovering the cis -acting sequences that control transcription, identifying which trans -acting factors act on each regulatory sequence, and determining how these interactions affect the timing and organization of transcription. The first step in this process is by no means straightforward. Regulatory regions are often large and complex. Functional cis -acting sequences are found 5' and 3' of transcripts and in introns, and can act over short or long distances. Most of the described animal regulatory sequences were identified by experimental dissection of a locus, and astonishingly few of these are well characterized. Despite the paucity of good examples, as multiple regulatory sequences from different organisms were identified and characterized, some common features became apparent [ 1 , 2 ]. Most animal regulatory sequences act as compact modular units, with regions of roughly a kilobase (kb) in size controlling specific aspects of a gene's transcription. These regulatory units - referred to here as cis -regulatory modules (CRMs) - tend to contain functional binding sites for several different transcription factors, often with multiple sites for each factor. As the first animal genome sequences were completed [ 3 - 6 ], researchers began to tackle the challenge of identifying regulatory sequences on a genomic scale. We and several other groups began to ask whether common characteristics of regulatory sequences - modularity and high binding-site density - might be distinguishing characteristics that would permit the computational identification of new regulatory sequences. A number of in silico methods to identify regulatory sequences on the basis of binding-site clustering have been developed and applied to animal genomes [ 7 - 10 ]. Some of the predictions have the expected in vivo regulatory activity [ 11 - 17 ], yet few of these predictions have been systematically evaluated. The transcriptional regulatory network governing early Drosophila development is perhaps the best system in which to apply and evaluate these methods. Development of the Drosophila embryo is arguably better understood than that of any other animal. Sophisticated genetic screens [ 18 , 19 ] have identified most of the key regulators of early development, and the molecular biology and biochemistry of these factors and their target sequences have received a great deal of attention. The spatial and temporal embryonic expression patterns of a large number of genes are known from microarray [ 20 ] and in situ expression studies [ 21 ]. Transcriptional regulation plays a uniquely important role in pre-gastrula patterning, as most of the key events occur in the absence of cell membranes and the cell-cell signaling systems that play a crucial role later in fly development and throughout the development of most other animals. In a previous study [ 11 ], we identified 37 regions of the Drosophila melanogaster genome with unusually high densities of predicted binding sites for the early-acting transcription factors Bicoid (BCD), Hunchback (HB), Krüppel (KR), Knirps (KNI) and Caudal (CAD). As nine of these regions overlapped previously known CRMs, we proposed the remaining 28 as predicted CRMs (pCRMs). We tested one of the previously untested pCRMs for enhancer activity in a standard reporter gene assay [ 22 , 23 ] and showed that it is responsible for directing a portion of the embryonic expression pattern of the gap transcription factor gene giant ( gt ) in a posterior stripe. Here, we report the systematic testing of the remaining 27 untested pCRMs for enhancer activity, resulting in collections of both bona fide positive and false-positive predictions, allowing us to develop and evaluate methods to improve the accuracy of methods for identifying functional cis -regulatory sequences. We were particularly interested in methods based on the comparison of genome sequences of related species. The genome sequence of D. pseudoobscura (which diverged from D. melanogaster approximately 46 million years ago [ 24 ]) was recently completed by the Baylor Human Genome Sequencing Center, and several other Drosophila species are currently being sequenced. The morphological and molecular events in early embryonic development are highly conserved among drosophilids, and we expect the activity of the transcriptional regulators and the architecture of regulatory networks to be highly conserved as well. Most D. melanogaster regulatory sequences should have functional orthologs in other Drosophila species [ 25 , 26 ], and a major rationale for sequencing other Drosophila species is the expectation that regulatory sequences have characteristic patterns of evolution that can be used to identify them and to better understand their function. Most methods used to identify regulatory sequences from interspecies sequence comparison are fairly simple. They identify 'conserved' non-coding sequences (CNSs), operationally defined as islands of non-coding sequence with relatively high conservation flanked by regions of low conservation, and assume that this conservation reflects regulatory function. Although crude, these methods have been remarkably effective in identifying mammalian regulatory sequences [ 27 , 28 ], and preliminary studies in Drosophila suggest that similar methods will be valuable in insects as well [ 29 ]. However, despite such successes, the extent of the efficacy of comparative sequence analysis in regulatory sequence discovery remains unclear. A systematic comparison of human-mouse sequence conservation in known regulatory regions and ancestral repeats (which provide a model for neutral evolution) suggests that regulatory regions cannot generally be distinguished on the basis of simple sequence conservation measures alone [ 30 , 31 ]. Similarly, a recent analysis of D. melanogaster and D. pseudoobscura showed that known regulatory regions are only slightly more conserved than the rest of the non-coding genome [ 32 ], highlighting the need for further study and the development of comparative methods that go beyond measures of sequence identity. Results Expression patterns of pCRM containing transgenes The 37 pCRMs are shown in Table 1 . Each has been assigned an identifier (of the form PCEXXXX). The first nine overlap previously known enhancers of runt ( run ), even-skipped ( eve ), hairy ( h ), knirps ( kni ) and hunchback ( hb ). To determine whether any of the remaining 28 pCRMs also function as enhancers, we generated P-element constructs containing the pCRM sequence with minimal flanking sequence on both sides fused to the eve basal promoter and a lacZ reporter gene (see Materials and methods). As the margins of the tested sequences do not precisely correspond to the margins of the clusters, we assigned a unique identifier (of the form CEXXXX) to each tested fragment (identical CE and PCE numbers correspond to the same pCRM). We successfully generated multiple independent transgenic fly lines for 27 of the 28 pCRMs. We repeatedly failed to generate transgenes containing CE8007. This sequence contains five copies of an approximately 358 base-pair (bp) degenerate repeat. One additional pCRM (CE8002) also contains tandem repeats. While we were able to generate transgenes for CE8002 and assay its expression, these two tandem repeat-containing pCRMs (CE8007 and CE8002) were excluded from subsequent analyses. We examined the expression of these constructs by in situ RNA hybridization to the lacZ transcript in embryos at different stages in at least three independent transformant lines. Nine of the 27 transgenes showed mRNA expression during embryogenesis (Figure 1 ), while the remaining 18 assayed transgenes showed no detectable expression at any stage during embryogenesis. To identify the genes regulated by the nine pCRMs with embryonic expression, we examined the expression patterns of genes containing the pCRM in an intron and genes with promoters within 20 kb of the CRM (see Figure 1 ). We used the embryonic microrarray and whole-mount in situ expression data available in the Berkeley Gene Expression Database [ 21 ], supplemented with additional whole-mount in situ experiments where necessary (data not shown; these new in situ 's will be included in the public expression database [ 33 ] at its next release). Six of the active pCRMs drive lacZ expression in patterns that recapitulate portions of the expression of a gene adjacent to or containing the pCRM. Four of these new enhancers act in the blastoderm and two during germ-band elongation. CE8001 is 5' of the gene for the gap transcription factor giant and recapitulates the posterior domain (65-85% egg length measuring from the anterior end of the embryo) of gt expression in the blastoderm as previously described [ 11 ]. CE8011 is 5' of the gene for the POU-homeobox transcription factor nubbin ( nub ). The CRM recapitulates the endogenous blastoderm expression pattern of nub , first detected as a broad band extending from 50 to 75% egg length. Although nub expression continues in later embryonic stages, CE8011 expression is limited to the blastoderm stage. CE8010 is 5' of the pair-rule gene odd-skipped ( odd ) and drives expression of two of its seven stripes: stripe 3 at 55% and stripe 6 at 75% egg length. This CRM also has the ability to drive later, more complex, patterns of expression. During stages 6 and 7, expression is detected in the procephalic ectoderm anlage and in the primordium of the posterior midgut. By stage 13, expression is also detected in the anterior cells of the midgut which will give rise to the proventriulus, the first midgut constriction, the posterior midgut and microtubule primordial as well as cells in the hindgut, all similar to portions of the pattern of wildtype odd protein expression previously described [ 34 ]. CE8024 is 3' of the pair-rule gene fushi-tarazu ( ftz ) and drives expression of two of its stripes: stripe 1 at 35% and stripe 5 at 65% egg length. Using a similar CRM reporter assay, this pattern of expression was also detected by [ 35 ]. CE8012 is in the third intron of POU domain protein 2 ( pdm2 ) and appears to completely recapitulate its stage-12 expression pattern, which is limited to a subset of the developing neuroblasts and ganglion mother cells of the developing central nervous system. A similar pattern of expression was previously described for the protein product of pdm2 [ 36 ]. It is worth noting that we do not detect expression of CE8012 in the blastoderm stage, whereas the endogenous gene exhibits a blastoderm expression pattern similar to nub . CE8027 is 3' of the gene for the Zn-finger transcription factor squeeze ( sqz ) and recapitulates the wild-type expression pattern of sqz RNA in a subset of cells in the neuroectoderm at stage 12. The wild-type sqz expression pattern was previously described [ 37 ]. The remaining three active pCRMs cannot be easily associated with a specific gene. CE8005 drives expression in the ventral region of the embryo. It is 3' of a gene encoding a ubiquitously expressed Zn-finger containing protein ( CG9650 ) that is maternally expressed and deposited in the embryo. This strong maternal expression potentially obscures a zygotic expression pattern. Two additional adjacent genes, CG32725 and CG1958 , showed no expression in whole-mount in situ hybridization of embryos. CE8016 drives a seven-stripe expression pattern in the blastoderm. It is in the first intron of CG14502 which shows very low level expression by microarrays in the blastoderm, and has no obvious detectable pattern of expression in whole-mount in situ hybridization of embryos. This pCRM is approximately 2 kb 5' of scribbler (sbb) , which is expressed maternally, possibly obscuring an early zygotic expression pattern (a few in situ images show a hint of striping). sbb is also expressed later in development in the ventral nervous system. An additional potential target, Otefin ( Ote ), is also expressed maternally and relatively ubiquitously through germ-band extension. All other nearby genes displayed in Figure 1 showed no embryonic expression in whole-mount in situ hybridization or by microarray. CE8020 drives an atypical four-stripe pattern in the blastoderm - two stripes at 7% and 26% that are anterior to the first ftz stripe and two stripes at 39% and 87%. It is in the first intron of ome ( CG32145 ), which is not expressed maternally and has no blastoderm expression, but is expressed late in salivary gland, trachea, hindgut and a subset of the epidermis. All other nearby genes displayed in Figure 1 showed no embryonic expression in whole-mount in situ hybridization or by microarray. With these results, and the nine previously known enhancers, at least 15 of the 37 highest density clusters of the five transcription factors used in our initial screen have early-embryonic enhancer activity. The remainder of this paper examines 35 of the original 37 clusters, with the two tandem repeat-containing clusters excluded. We divide these 35 into three categories - 15 positives (the nine overlapping previously known enhancers plus the six new enhancers identified here), three ambiguous (the three positives without a clear regulated gene), and 17 negatives (see Table 2 ). We largely focus on differences between the positives and negatives. Distinguishing active and inactive clusters All 15 positives are within 20 kb of the transcription start site (or, where the transcription start site is unknown, the start of the gene annotation) of transcripts expressed in spatiotemporal patterns consistent with regulation by the maternal and gap transcription factors used in our screen (that is, in anterior-posterior patterns in the blastoderm or in the developing neuroblasts of the central nervous system). Only one of the 17 negatives was located within 20 kb of a plausible target (PCE8021 is 7 kb upstream of reaper ), so out of 16 pCRMs located within 20 kb of a gene with appropriate expression, 15 (94%) are active enhancers. The positives are, on average, larger than the negatives (average cluster size of positive = 900 bp, while average cluster size of negatives was 711 bp), a difference that is significant by the Komogorov-Smirnov (KS) test ( p = 0.017). The positives have a slightly higher density of binding sites, but this difference was not significant. The binding site composition of the positives and negatives are similar (the positives contain more KR, and fewer BCD binding sites, but again these differences are not highly significant). Although others have reported that some factors have characteristic spacings with respect to themselves and other factors [ 38 ], we could not find evidence for such spacing or identify other differences that could distinguish positive pCRMs from negative (Figure 2 ). Use of D. pseudoobscura We assembled the D. pseudoobscura genome from traces deposited in the NCBI's TraceDB using the Celera assembler [ 39 , 40 ]. These assemblies were used to examine the conservation of our pCRMs and to assess whether conservation could be used instead of or in addition to binding site clustering as a way to identify CRMs. We first assessed whether positive pCRMs could be distinguished from their flanking sequences based on degree of conservation. In vertebrate comparative genomics, relatively simple methods (such as VISTA [ 41 ]) are commonly used to identify CNSs that are a surprisingly rich source of new cis -regulatory sequences. We evaluated the potential of using such methods with D. melanogaster and D. pseudoobscura in two ways. First, we constructed percent-identity plots for the regions containing all of the 37 pCRMs (Figure 3 ; similar plots for all pCRMs are available in the online supplement at [ 42 ]) with the location of pCRMs and other known regulatory sequences clearly indicated. Although it appears that some CRMs (that is, eve stripe 3/7) would have been successfully identified by such simple comparative methods, positive pCRMs do not collectively appear distinguishable from flanking sequence on the basis of conservation alone. Although positive pCRMs are almost all in highly conserved blocks, there is a surprisingly high amount of non-coding sequence conservation throughout these regions, and most negative pCRMs are also contained in highly conserved blocks. It remains to be seen whether this difference in the conservation landscape of Drosophila non-coding sequences compared to vertebrates reflects a significant difference in the functional organization of non-coding sequences, or simply indicates that there is too little divergence between D. melanogaster and D. pseudoobscura to detect useful differences in the rates of evolution (see Discussion). We next assessed whether positive pCRMs can be distinguished from negative pCRMs on the basis of their degree of similarity between D. melanogaster and D. pseudoobscura . For each pCRM-containing region, we identified orthologous contigs from the D. pseudoobscura assembly and aligned them using the alignment program LAGAN [ 43 ]. We were able to find orthologous regions for 32 pCRMs (see Table 2 ). Using the simple measure of percent identity, we find that positive pCRMs are, on average, more highly conserved than negative pCRMs (see Table 2 ). Although this difference is significant ( p = 0.002 by KS test), the distribution of conservation scores for positive and negative pCRMs overlap considerably, and thus conservation alone is not a useful way of distinguishing positive and negative pCRMs (see Figure 4b ). To get a genome-wide perspective on the degree of conservation in positive pCRMs, we analyzed the conservation of CRM-sized (1 kb) regions in randomly chosen sections of the genome (Figure 4b ). Positive pCRMs are, generally, more conserved than average CRM-sized sequences, and some positive pCRMs are among the most highly conserved non-coding sequences in the genome. However, a conservation cut-off necessary to select the majority of positive pCRMs would select roughly one third of the non-coding regions of the genome, and thus is not a practical method for prioritizing sequences for functional analysis. Conservation of binding sites and conservation of clustering We expect that most genes will have similar expression patterns in D. melanogaster and D. pseudoobscura , and that most D. melanogaster enhancers should have functional orthologs in D. pseudoobscura . For those enhancers we seek to identify here - namely those where binding site clustering reflects their function - we expect clustering to be found in both D. melanogaster and D. pseudoobscura . Conversely, clusters that simply occur by chance in either genome but do not reflect the function of the sequence (as, we believe, is the case for many of our false-positive predictions) should not be conserved. Thus, looking for conservation of binding-site clustering should provide a valuable way of distinguishing functional and non-functional binding-site clusters in the D. melanogaster genome. We used the alignments described above to examine the conservation of individual predicted binding sites in all of the pCRMs (Table 2 ). We refer to a predicted D. melanogaster binding site that overlaps a predicted D. pseudoobscura binding site for the same factor in an alignment as an 'aligned' site. We require overlap and not perfect alignment to compensate for alignment ambiguity; the overwhelming majority (85%) of aligned sites are perfectly aligned. Although there is only a subtle difference in the binding-site density in the positive and negative pCRMs in D. melanogaster (22.7 sites/kb compared to 22.2), the density of aligned binding sites in positive pCRMs (13.8 sites/kb) is nearly twice that in negative pCRMs (6.8 sites/kb). This is a highly significant difference ( p < 0.001 by KS test) and aligned site density better discriminates positive and negative pCRMs than sequence conservation (compare Figure 4c and 4b ). Sixty-one percent of the predicted binding sites in positive pCRMs are aligned, while only 30% of the sites in negative pCRMs are aligned. Across the genome, 22.3% of predicted binding sites are aligned meaning that there is a roughly fourfold increase over background in the probability that a binding site in a positive pCRM is conserved in place compared to a binding site in a negative pCRM. Sixty-one percent is almost certainly an underestimate of the fraction of pCRM sites that are functionally conserved. The D. melanogaster - D. pseudoobscura alignments were not always unambiguous (using simulations we have assessed the role of alignment algorithms in identifying conserved transcription factor binding sites, see [ 44 ]), and some orthologous binding sites may not have been properly aligned. More important, studies of the evolution of various Drosophila enhancers suggest that the positions of binding sites within an enhancer are somewhat plastic, and the functional conservation of a binding site does not necessarily require positional conservation [ 25 , 26 ]. To characterize the extent of binding site conservation independent of positional conservation, we computed a second measure of binding-site conservation. We consider an unaligned binding site in D. melanogaster to be 'preserved' if it can be matched to a corresponding site in the D. pseudoobscura pCRM (allowing each D. pseudoobscura site to match only one D. melanogaste r site). If we consider both aligned and preserved sites to be conserved, then roughly 80% of the sites in positive pCRMs are conserved compared with 40% in negative pCRMs. The density of preserved but not aligned sites in positive pCRMs (4.3/kb) is considerably higher than in negative pCRMs (2.2/kb) or random sequences (1.8/kb). Thus, in the D. pseudoobscura orthologs of active D. melanogaster CRMs we observe an increase in binding-site density that cannot be explained by the positional conservation of sites found in D. melanogaster or the random occurrence of sites in the genome. Several of the 15 positive CRMs have high densities of these preserved but unaligned sites, but two in particular, runt stripe 3 and hairy stripe 6, stand out from the rest. These two have almost as many preserved sites as strictly aligned sites. Aligned plus preserved (conserved) site density (Figure 4d ) almost perfectly separates positive from negative pCRMs. Only one of the positive pCRMs (PCE8012) has a conserved site density below 14 sites/kb, while only one of the negative pCRMs (PCE8021) has a conserved site density above 14 sites/kb. eCIS-ANALYST: a comparative enhancer finder As the conservation of binding sites and binding-site clusters between D. melanogaster and D. pseudoobscura successfully distinguishes positive and negative predictions made using the D. melanogaster sequence alone, we incorporated comparative sequence data into our enhancer-prediction algorithm CIS-ANALYST [ 11 ]. Instead of searching for clusters of predicted binding sites in a single genome, eCIS-ANALYST (the 'e' is for evolutionary) searches for conserved clusters of sites between the two genomes (see Materials and methods). eCIS-ANALYST is available at [ 45 ]. Using 17 negative pCRMs and an expanded set of 25 positive pCRMs (which included the 15 positive predictions discussed above and 10 functional enhancers known to respond to the five factors; these 10 additional enhancers were discussed and analyzed in [ 11 ] but had binding-site densities below the threshold used there), we compared the ability of CIS-ANALYST and eCIS-ANALYST to identify positive pCRMs and to distinguish positive and negative pCRMs at different binding-site density cutoffs (Figure 5 ). The incorporation of the conservation criteria greatly improves the algorithm's apparent performance. The expected fraction of false positives is markedly reduced, and it is possible to lower the binding site threshold to recover six of the ten previously missed positive enhancers without increasing the number of expected false-positive predictions. New predictions As eCIS-ANALYST has markedly better specificity than CIS-ANALYST, we sought to identify BCD, HB, KR, KNI and CAD targets that were missed with the relatively stringent criteria used in our previous analysis. Rather than use a stringent cutoff (15 binding sites per 700 bp) as we did in [ 11 ], we performed three separate runs with lower cutoffs (for example, 10 sites per 700 bp in one run) and applied a conservation threshold (see Materials and methods and Additional data file 3) to select 929 conserved binding-site clusters. There were 842 new pCRMs within 20 kb or in an intron of an annotated transcript (Additional data file 7) and 87 more than 20 kb (Additional data file 8). We ranked these new pCRMs by a simple scoring scheme that measures both the density and the total number of sites conserved (we evaluated several different scoring schemes, and selected one that optimally identified regions near genes with blastoderm expression patterns; see Materials and methods). The 75 highest-scoring pCRMs within 20 kb of an annotated transcript are shown in Table 3 . Thirteen of the 15 positive pCRMs described above are in the top 75 ( ftz stripe 1/5 is number 107 and the pdm2 neurogenic enhancer is number 418) as are five other known enhancers. One of our negative pCRMs, CE8021, is ranked number 12. To focus our search for new enhancers on genes likely to be regulated by BCD, HB, KR, KNI and/or CAD, we searched FlyBase [ 46 ] and a database of Drosophila embryonic expression patterns [ 21 ] and identified 278 genes with anterior-posterior patterns in the blastoderm (AP genes; Figure 6 and see also Additional data files 2 and 9). Thirty-one of the 75 highest-scoring new predictions are adjacent to or within 20 kb of one or more of these genes, including 11 pCRMs that do not overlap previously described enhancers. The 75 highest-scoring predictions within 20 kb of an AP gene but not in Table 3 , are shown in Table 4 . In Tables 3 and 4 together, there are 106 high-scoring conserved binding-site clusters near AP genes, 90 of which do not overlap known enhancers. Discussion We performed a large and comprehensive evaluation of the efficacy of computational methods for the identification of functional cis -regulatory modules in Drosophila . Analysis of the in vivo activity of 36 high-density clusters of predicted BCD, HB, KR, KNI and CAD binding sites identified in our previous study [ 11 ] offers compelling support for the use of transcription factor binding-site clustering as a method to identify regulatory sequences, as at least 15 of these sequences function as early developmental enhancers in vivo . An evolutionary analysis of these sequences - based on comparisons of the D. melanogaster and D. pseudoobscura genomes - shows that sequence conservation alone can not reliably discriminate cluster-containing regions that function in vivo from those that do not. However, a new method that combines binding-site clustering and comparative sequence analysis to search for binding-site clusters that are present in multiple species does reliably discriminate active and inactive clusters. Using this method, we make several hundred predictions of new CRMs, a large number of which are located near likely target genes. Binding-site clustering The success of relatively simple binding-site clustering methods here and in other work is remarkable given the crudeness of these methods. As our negative predictions demonstrate, the mere presence of a cluster of binding sites is not sufficient to make an active embryonically expressed CRM. Although these 17 sequences have binding-site densities and compositions indistinguishable from their functional cousins, they do not function as enhancers in a simple transgene assay. It is possible that some of these negative pCRMs may be functional enhancers that respond to the factors used in our screen, perhaps requiring a different promoter or other flanking sequences not used in the transgene. While further experiments could address this possibility, we felt these were a low priority, as few of the D. pseudoobscura orthologs of these negative pCRMs have binding-site clusters, and few are near genes with appropriate expression patterns. Thus it is unlikely that many function in their endogenous locations in vivo . Both the general activity and, more important, the specific regulatory output of a CRM are a complex, and still poorly understood, function of the specific architecture of its sites. The emerging picture of the ordered multiprotein complexes that mediate enhancer activity suggests constraints on enhancer composition and architecture [ 1 , 2 , 47 ] whose elucidation will form a critical part of the future dissection of the function of cis -regulatory sequences. It is intriguing that three of the clusters we tested direct expression patterns that bear no obvious relationship to the expression of a neighboring gene despite our extensive efforts to identify such genes. We cannot yet exclude the possibility that these pCRMs have an in vivo function related to their observed expression patterns. However, the poor conservation of these elements in D. pseudoobscura suggest that they do not have a regulatory function, and raises the possibility that some 'random' clusters of binding sites (that occur by chance or perhaps through selection on some functionally unrelated sequence feature) have the necessary characteristics to be active enhancers in the proper genomic environment (that is, near a promoter and not silenced by trans -acting chromatin mechanisms). That any such sequences exist suggests that the compositional and architectural constraints on binding sites in enhancers may be fairly weak. Whatever the nature of these constraints, it is clear that binding-site density is not the sole defining characteristic of functional enhancers. However, it is a surprisingly effective distinguishing one, and the usefulness of this and related methods [ 48 ] suggests that the broader application of such methods to different collections of transcription factors will be extremely valuable in annotating the regulatory content of animal genomes. New enhancers We identified double-stripe enhancers for ftz and odd . ftz and odd are generally classified as 'secondary' pair-rule genes whose expression is governed by other pair-rule genes rather than by the maternal and gap transcription factors that govern the so-called 'primary' pair-rule genes ( eve , h and runt ) ([ 49 ]; also reviewed in [ 50 ]). However, the ftz and odd enhancers described here were identified on the basis of binding sites for maternal and gap transcription factors, and function like the enhancers of primary pair-rule genes in directing expression in specific stripes. It has been suggested that the ftz enhancer is an evolutionary relic of the homeotic role played by ftz in primitive insects [ 51 ], a view supported by the apparently normal expression and activity of ftz when this element is missing. However, given our observation that non-functional binding sites clusters are not conserved, even over the relatively short evolutionary distance separating D. melanogaster and D. pseudoobscura , it seems unlikely that this element is purely vestigial. In fact, Yu and Pick [ 52 ] examined the expression pattern of the endogenous ftz gene and show that stripes 1 and 5 appear before other ftz stripes and they postulate the existence of stripe-specific regulatory elements that may exist outside of the characterized zebra and upstream elements such as the one identified and characterized in this study. The conservation of binding sites in both the ftz and odd enhancers suggest that they play an important role in development, and further call into question the distinction between primary and secondary pair-rule genes. Two of the new enhancers (CE8011 and CE8012) are adjacent to and apparently regulate two linked genes with very similar patterns of embryonic expression. Both nub (also known as pdm1 ) and pdm2 are expressed in the anterior and posterior midgut primordium and in neuroblasts. CE8011, found immediately upstream of nub , regulates its early expression, and not its later neuroblast expression. In contrast, CE8012, found in an intron of pdm2 regulates its expression only in neuroblasts and not earlier. While we did not detect a neuroblast enhancer for nub or a blastoderm enhancer for pdm2 in our single-species binding-site cluster search, a number of interesting pdm2 regions were discovered in our eCIS-ANALYST search (two are listed in Table 4 ). Regulatory models and improving the accuracy of CRM prediction The accuracy of our enhancer predictions would almost certainly be improved if we restricted our search space to genomic regions adjacent to genes known to be regulated by particular transcription factors. Drosophila enhancers have been known to work at distances of up to 100 kb, but most are within 10 kb of their target gene. All of our true-positive predictions were within 10 kb of the known or predicted transcription start site of a gene with a pattern that was known, or plausibly could have been, regulated by the five regulators used in our screen (anterior-posterior patterns in the blastoderm; expression in neuroblasts). In contrast, only one of the negative predictions was this close to such a gene - an additional four were within 50 kb. As the comprehensive atlas of embryonic expression patterns is completed [ 21 , 53 ] it will be possible to restrict searches for CRMs to regions of the genome near genes with expression patterns that could arise from the regulators being considered, or to prioritize the results of whole-genome screens on the basis of whether they are near plausible targets. Comprehensive methods for inferring regulatory interactions where they are not already known will be critical for the widespread application of binding-site clustering methods. In addition to allowing less stringent focused screens, they will also help overcome the combinatorial challenge raised by the existence of up to 700 sequence-specific transcription factors in Drosophila . Even assuming the availability of binding data for all of these factors, it will not be possible to search for targets of all combinations of these factors - there are too many possibilities. This is not just a practical problem - it is a fundamental statistical problem. While the false-positive rate for a single combination of factors is low, if we tried even all pairs of factors, it is likely that every region of the genome would have a high binding-site density for some collection of factors. Sequence data from other Drosophila species may allow us to determine which of these collections are conserved and therefore likely to be functional, but it is unlikely that all aspects of regulation can be inferred from comparative analyses and therefore it is essential that we continue to dissect the regulatory network by traditional means. A greater current limitation in the widespread application of binding-site clustering methods is the absence of high-quality binding data for most Drosophila transcription factors. The initial success of methods that use in vitro binding data to predict regulatory targets has prompted the characterization of binding specificities for many additional factors. However, the heterogeneity of approaches used makes it difficult to combine these data in an optimal manner. In addition, most of the available transcription factor binding data consists of a few to several dozen high-affinity sites. While these data are very useful, they do not fully represent the binding capacity of a factor and thus do not permit the identification of intermediate or low-affinity sites which are known to be important in some regulatory systems [ 54 ]. We have begun to apply high-throughput methods [ 55 ] to characterize a broad spectrum of target sites for all of the transcription factors involved in early embryogenesis. The results will ultimately allow us to estimate the binding affinity of each factor for any target sequence. Comparative genomics in CRM predictions The extent of non-coding sequence conservation between D. melanogaster and D. pseudoobscura was surprising. A major motivation for the National Human Genome Research Institute (NHGRI) support of the D. pseudoobscura genome sequencing was the identification of conserved regions that would guide the annotation of functional sequences in D. melanogaster . D. pseudoobscura was chosen as the second member of this genus to be sequenced in part because it was felt that it had separated from D. melanogaster sufficiently long ago that non-functional sequences would exhibit substantial divergence. However, despite an evolutionary separation that is greater than human and mouse (an average synonymous substitution rate of 1.8-2.6 substitutions/site [ 29 ] compared to 0.6 substitutions/site [ 30 ]), and despite some variation in conservation in non-coding sequences, we were not able to use standard measures of sequence conservation to differentiate active pCRMs from their flanking sequence or from inactive pCRMs, reinforcing other recent observations [ 32 ]. One reason for the limited efficacy of these methods is that they do not recognize the specific patterns of conservation characteristic of different classes of functional sequences. For example, coding sequences can be easily recognized from the characteristic triplet pattern in evolutionary rates where the third (and often synonymous) position of codons tends to evolve at a greater rate than the first two positions [ 56 , 57 ]. Similarly, RNAs that form conserved secondary structures can be recognized by patterns of co-substitution ([ 58 ] and references cited within). The early developmental enhancers we are studying here are made up of large collections of transcription factor-binding sites, and it is expected that both individual functional binding sites and the overall composition of functional CRMs will be conserved [ 25 , 26 ]. Conservation of binding-site clustering is a specific evolutionary signature of this class of functional regulatory sequences, and, like the evolutionary signatures of protein-coding and RNA genes, can be used to specifically identify these sequences from comparative sequence data. Contrast PCE8010 (the odd stripe enhancer) and PCE8015 (Figure 3 ). Both have the same overall amount of sequence conservation, indicating that they are under some functional constraint. However, 80% of the predicted binding sites in PCE8001 are conserved, compared to 20% for PCE8015. The conservation of binding sites (both number and location) in PCE8001 makes it highly unlikely that the cluster was found by chance in D. melanogaster , and suggests (correctly) that this sequence is actively responding to the presence of these binding sites. The poor conservation of binding sites in PCE8015 (no greater than is found in random regions of genome) suggests either that the BCD, HB, KR, KNI and CAD sites in this region are not functional or that the region is undergoing rapid functional diversification. Of course the absence of binding site conservation does not suggest that the sequence is non-functional, merely that these sequences are unlikely to have the particular function we are studying here. From the data shown in Figure 4 , we expect the incorporation of binding-site conservation into the CRM search process to greatly reduce the number of false-positive predictions. We anticipate that a significant number of the new predictions from our genome-wide screen and screen targeted at genes with early anterior-posterior patterns to be active CRMs, and we have begun testing these predictions. The pattern of binding-site conservation in positive pCRMs sheds additional light on the processes that govern CRM evolution. We find that predicted binding sites in positive D. melanogaster pCRMs are roughly three times more likely to be aligned to predicted sites in the D. pseudoobscura compared to predicted binding sites in negative pCRMs, in the sequences flanking pCRMs, or in random regions of the genome. The demonstration that this strictest form of binding-site conservation is strengthened in functional CRMs contrasts with an earlier study that concluded that binding sites in functional CRMs had only a slightly elevated probability of falling in conserved sequence [ 32 ]. Their methodology differed from ours in that they used randomly shuffled binding-site positions within functional CRMs as the background, while we used actual predicted binding-site positions in randomly picked regions of the genome. In addition to this colinear conservation, we also observe that there is an overall enrichment for binding sites in positive pCRMs independent of the conservation of individual sites. Specifically, the presence of a binding site for a factor in a positive D. melanogaster pCRM increases (relative to negative pCRMs and random genomic fragments) the probability of finding a site for the same factor in the orthologous region of D. pseudoobscura , even if the site is not in the same (aligned) position. Thus, in this set of positive pCRMs, there appears to be selection to maintain binding site composition, but not always the specific order and orientation of sites. This is consistent with models of enhancer plasticity that have been proposed and discussed elsewhere [ 25 , 59 - 61 ]. The relative importance of binding-site architecture and binding-site composition to maintaining the function of an enhancer over evolutionary time remains unclear. Over relatively short evolutionary distances (as between D. melanogaster and D. pseudoobscura ) most binding sites are conserved and found in the same place. Over longer evolutionary distances, individual binding sites are often poorly conserved even as the overall composition and function of a CRM is conserved. From a practical perspective, this requires adjusting how conservation is incorporated into searches for clusters of binding sites that are likely to be CRMs. For relatively short evolutionary distances, searches for clusters of aligned sites will be less sensitive to noise and will focus on functional binding sites. For longer distances, where binding site turnover will likely preclude searching for clusters of conserved sites, searches for conserved binding site clusters should still work well. In fact, this latter method can work - with some modification - among species whose sequences can no longer be aligned. Anopheles gambiae diverged from its common ancestor with D. melanogaster roughly 220 million years ago, and there is little or no detectable non-coding sequence similarity between these two species. Nonetheless, we find clusters of HB, KR and KNI binding sites in the vicinity of gap and pair-rule genes and suggest that many of these are functional orthologs of D. melanogaster CRMs. Despite strong selection to maintain function, enough binding-site turnover has occurred in these CRM during their 220 million years of independent evolution to eliminate detectable sequence similarity. But they remain functionally similar and we can detect this functional similarity through its evolutionary signature. With methods like the one we have presented here, aided by new and better binding data on Drosophila transcription factors and an impending wealth of comparative sequence data, we anticipate rapid progress on the identification and functional characterization of regulatory sequences. We will then be able to turn our attention to the next great challenge - understanding the precise relationship between the binding-site composition and architecture of regulatory sequences and the expression patterns they specify. Materials and methods Collection of CRMs The collection of CRM sequences was previously described [ 11 ] Transgenics DNA fragments identified as candidate CRMs were amplified from either bacterial artifical chromosome (BAC) or y; cn bw sp fly genomic DNA by PCR using two primers containing unique sequence and synthetic Asc I and Not I restriction sites (Additional data file 5). The PCR product was digested with Asc I and Not I, and inserted in its native orientation into the Asc I- Not I site of a modified CaSpeR-AUG-bgal transformation vector [ 62 ] containing the eve basal promoter, starting at -42 bp and continuing through codon 22 fused in-frame with lacZ [ 63 ]. The P-element transformation vectors were injected into w 1118 embryos, as described previously [ 63 , 64 ]. Transgenic fly lines containing CRMs CE8005 (7A), CE8016 (55C) and CE8020 (70EF) were verified by generating genomic DNA [ 65 ] from each line for PCR. PCR products were amplified using primers designed from the CaSpeR-AUG-bgal vector - forward primer 5' CGCTTGGAGCTTCGTCAC and reverse primer 5' GAGTAACAACCCGTCGGATTC and 35 cycles (Gene Amp 9700, Perkin-Elmer). The resulting PCR products were sequenced using standard conditions with BigDye version 3.0 and electrophoresed on a 3730 capillary sequencer (ABI). Whole-mount in situ hybridizations Embryonic whole-mount in situ RNA hybridizations were performed as previously described [ 21 ]. RNA probes were generated using cDNA clones RE29225 ( gt ), RE14252 ( odd ), RE34782 ( nub ), RE49429 ( pdm2 ), and RE47384 ( sqz ). Exon 1 of the ftz gene was amplified from genomic DNA using forward primer 5' GCGTTGCGTGCACATC and reverse primer 5' ATTCTTCAGCTTCTGCGTCTG. The PCR product was cloned into the TA vector (Invitrogen) and used to generate ftz RNA probe. Double-labeling RNA probes, using cDNAs or genomic DNA as templates, were labeled with fluorescein-12-UTP while lacZ RNA probes were labeled with digoxigenin-11-UTP (Roche). Hybridizations were performed as described above with the following modifications: (1) 2 μl of each probe were added to give a final concentration of 1:50; (2) sequential alkaline phosphatase staining was performed first with Sigma Fast red to detect endogenous transcripts, stopped by washing for 30 min in 0.1 M glycine-HCl pH 2.2, 0.1% Tween-20 at room temperature, and then continued as described to detect lacZ expression. Assembly The input to the genome assembly was the set of whole-genome shotgun reads from the Baylor Genome Sequencing Center retrieved from the National Center for Biotechnology Information (NCBI) Trace Archive, consisting of 2,607,525 total sequences. After trimming the sequences to remove vector and low-quality regions, the average read length was 607 bp. Approximately 75% of the reads were from short insert (approximately 2.5-3.0 kb) libraries, with another 25% from longer (6-7 kb) libraries. Another 46,040 reads came from the ends of 40-kb fosmids. We ran the Celera Assembler several times, and found that by adjusting one parameter in particular we could produce considerably better assemblies. In particular, the assembler has an arrival rate statistic j , which measures the probability that a contig is repetitive on the basis of its depth of coverage. The default setting is very conservative: if a contig has more than 50% likelihood of being repetitive, it is marked as such and is set aside during most of the assembly process. For large highly repetitive mammalian genomes this setting may be appropriate, but for D. pseudoobscura we found that setting it to 90% or higher produced considerably better contigs, while apparently causing few if any misassemblies. The overall assembly contained 10,089 scaffolds and 10,329 contigs, containing 165,864,212 bp. The estimated span of the scaffolds, using the gap sizes estimated from clone insert sizes, is 172,362,884. The largest scaffold was 3.05 million base-pairs (Mbp) and the scaffold N50 size was 418,046. (The N50 size is the size of the smallest scaffold such that the total length of all scaffolds greater than this size is at least one half the total genome size, where genome size here is 172 Mbp.) There are 308 scaffolds larger than 100,000 bp, whose total span is 129.5 Mbp. The N50 contig size, using 166 Mbp as the genome size (not counting gaps), was 43,555. Another measure of assembly quality is the number of large contigs: if we define 'large' as 10 kbp, then the assembly contains 3177 large contigs whose total length is 131,067,828 bp. (For reference, the assembly produced by the Baylor Human Genome Sequencing Center contains 129.4 Mbp in all contigs, including small ones, and the span of all scaffolds is 139.3 Mbp.) All of our contigs and scaffolds are freely available by anonymous ftp at [ 66 ]. Alignment and conservation of pCRMs The extent and pattern of conservation between D. melanogaster and D. pseudoobscura in regions containing pCRMs were determined as follows. The D. melanogaster genomic sequence of the region of interest (with known repetitive elements masked) was extracted from a BioPerl genome database [ 67 ] containing Release 3.1 sequence and annotations from the Berkeley Drosophila Genome Project [ 68 ]. Potentially orthologous D. pseudoobscura contigs/scaffolds were identified using WU-BLAST 2.0 [ 69 ] using default parameters except for (-span1 -spsepqmax = 5000 -hspsepsmax = 5000 -gapsepmax = 5000 -gapsepsmax = 5000). High-scoring pairs (HSPs) with E-values less than 1e-20 were flagged as potential homologous regions. HSPs located more than 5,000 bp from each other in the D. melanogaster sequence were treated as separate hits. After examining dot-plots of the hits, we noticed a large number of small, local inversions that were found in both our assembly and the assemblies released by the Baylor Human Genome Sequencing Center. We used BLASTZ [ 70 ]) to automatically identify inversions, and when necessary inverted the corresponding D. pseudoobscura sequence. Each D. pseudoobscura sequence was aligned to the D. melanogaster corresponding sequence using LAGAN 1.2 [ 43 ] with default settings. A total of 31 genomic loci of approximately 50 kb were examined; these regions contain 36 pCRMs (the eve and h loci contain three pCRMs each, and PCE8003 and PCE8004 are within 20 kb of each other). Twenty-eight regions had aligned D. pseudoobscura sequence that spanned all or most of the region. For three regions (PCE8002, PCE8003/8004 and PCE8009) we were not able to identify large regions of orthologous sequence; these were excluded from subsequent comparative analyses. Dot-plots of the alignments from all 30 regions are available at [ 42 ]. Scoring gross conservation of pCRMs The conservation of a specific genomic segment was scored as the fraction of D. melanogaster bases aligned to the identical base in aligned regions (percent identity). Scoring binding-site conservation of pCRMs We used two definitions of binding-site conservation. A binding site was considered 'aligned' if it overlaps a predicted D. pseudoobscura binding site for the same factor in the LAGAN alignment. Only overlap, and not strict alignment, was required to compensate for small errors in the alignment. A non-aligned binding site was considered 'preserved' if it could be matched to a D. pseudoobscura site for the same factor within the bounds of the pCRM, allowing each D. pseudoobscura site to be the match for only a single D. melanogaster site. The number of aligned plus preserved sites for each factor in a region is thus equal to the minimum number of sites for that factor in the two species. Generating an orthology map for genome searches To develop an orthology map for genome-wide searches, we used NUCmer [ 71 ] to align the Release 3 D. melanogaster genome (with annotated repetitive elements and transposable elements masked) and the D. pseudoobscura scaffolds described above. NUCmer was run with the command line parameters (-c 36 -g 10 --mum -d 0.3 -l 9). NUCmer generated a collection of short, highly conserved regions of homology ('anchors') spaced on average every 1 kb throughout the D. melanogaster genome. Anchors flanking either side of a D. melanogaster region of interest were used to pull out the corresponding D. pseudoobscura region, and additional flanking anchors were examined to ensure that the region was unambiguously orthologous. The region identified was re-aligned to the melanogaster region with LAGAN 1.2 using default settings. Random sampling of non-coding genome To characterize properties of non-coding sequences across the genome, we picked 4,000 1-kb segments of the D. melanogaster genome, sampled uniformly from all non-coding sequence. For 3,300 of these, we could find orthologous regions in D. pseudoobscura , and these were used to calculate the properties of random non-coding sequence shown in Figure 4 and discussed in the text. Properties determined using this data are considered properties of only the portion of the genome that is detectably orthologous under our conditions. The regions themselves are available as supplemental material at [ 42 ]. eCIS-ANALYST genome searches Binding-site clusters in the D. melanogaster genome were determined as described in [ 11 ], where the minimum number of sites (min_sites) and the window size (wind_size) are variable. Release 3 genomic sequence with exons masked was searched with PATSER [ 72 ] using the following command line options: -c -d2 -l4. An 'alphabet' file (specified with the command line parameter '-a') was used to provide the following background frequencies: A/T = 0.297, G/C = 0.203. Position weight matrix (PWM) models were identical to those used in [ 11 ]. In the online version of eCIS-ANALYST, the minimum PWM match threshold site_p is also variable, but in the current study it was held constant at 0.0003 for all factors. Tests using alternate values for this variable did not lead to significant improvement in prediction efficacy. For each potential D. melanogaster cluster, we identified the corresponding D. pseudoobscura region using the homology anchors described above. A pairwise alignment was made using LAGAN 1.2 (default parameters), and the number of aligned and preserved binding sites were determined as described above. The 2-kb flanking either side of the pCRM was included in the alignment to avoid edge effects, and was subsequently removed when calculating pCRM properties. We examined our functional (positive) and non-functional (negative) pCRMs and noticed that in the positives, the lower bound for the number of conserved sites as a function of D. melanogaster sites followed an approximately logarithmic curve (Additional data file 3). From this observation, we classified a D. melanogaster binding site cluster as conserved if: where NS m is the number of binding sites in the D. melanogaster pCRM and NS c is the number of conserved binding sites. Different values of the logarithmic base b give different behavior. The data shown in Additional data file 3 support values of b between 1.15 and 1.4. We defined a more intuitive parameter, CF (conservation factor), which can range from 0 to 1 where 0 is the least stringent threshold ( b = 1.4) and 1 is the most stringent ( b = 1.15) b = 1.4 - ( CF * (1.4 - 1.15))     (2) We performed genome searches with CF values of 0.25, 0.5, 0.55 and 0.75 and manually inspected the results with respect to false-negative and false-positive rates based on our 15 positive and 17 negative pCRMs (Additional data file 3). While we did not strictly optimize a single metric, we picked the values that gave a reasonable balance between false positives and false negatives, b = 0.25 for aligned sites alone, and b = 0.55 for aligned plus preserved sits. Genome-wide predictions eCIS-ANALYST genome searches were run with the following parameters: min_sites = 10, wind_size = 700 (run #1), and min_sites = 13, wind_size = 1,100 (run #2). All conserved clusters (with conservation defined as described in Equations 1 and 2 above) were combined. In order to capture weaker clusters, we performed an additional run (run number 3) using min_sites = 9, wind_size = 700. For this low stringency run, we used a non-standard conservation threshold different from the one described above, accepting all clusters with at least four aligned plus preserved sites, independent of the number of sites in D. melanogaster . We merged overlapping clusters from runs 1-3, yielding 929 non-overlapping clusters as described in Results. Four metrics were then used to rank these 929 pCRMs: the number of aligned binding sites; the density of aligned binding sites; the number of aligned plus preserved binding sites; and the density of aligned plus preserved binding sites. All values were normalized according to background distribution of random non-coding sequences. The four normalized values were then summed to compute an overall score, which was then renormalized to arrive at a final z-score used to rank pCRMs in Tables 3 and 4 and Additional data files 7, 8, 10, and 11. Additional data files The following additional data files are available with the online version of this article. Additional data file 1 shows the binding site densities (column 1), aligned site densities (column 2), and aligned plus preserved site densities (column 3) for individual transcription factors. The top portion of each panel contains a histogram of the values for randomly chosen 1,000 bp regions of the D. melanogaster genome. The blue line plots the cumulative distribution. The colored asterisks show the average values for each class of pCRM. The panel below the histogram shows the values for each pCRM (each dot represents one pCRM, with positives in blue, negatives in red, ambiguous in green). Additional data file 2 shows expression patterns of 65 genes adjacent to 122 pCRMs identified by eCIS-ANALYST. The images were obtained from the BDGP Embryonic Expression Pattern Database [ 33 ], and include all pCRMs from Additional data files 7,8,10,11 for which an adjacent gene had an early segmentation pattern. Additional data file 3 shows discrimination of positive and negative pCRMs. Comparisons of the number of predicted binding sites in D. melanogaster pCRMs to the number of aligned sites (top panel) and aligned plus preserved sites (bottom panel). Blue dots represent the 15 positive pCRMs from the text; green dots the ten known CRMs that were below the threshold used in [ 11 ]; red dots negative pCRMs; pink dots ambiguous pCRMs. Gray boxes represent the distribution of values for random 1,000 bp non-coding regions. The blue line shows the discrimination function (see Materials and methods). Additional data file 4 shows new pCRMs. Three 30 kb regions were chosen to illustrate new predictions: (A) the argos locus, (B) the CG4702 locus (note that CG31361 is not expressed in blastoderm embryos and PCE8494 is a low-scoring pCRM), and (C) the SoxN locus. Exons are shows as blue boxes, introns are represented with horizontal lines, and the direction of transcription is indicated by the arrow. New pCRMs are shown as gray ovals. The green graphs show average (in 300 bp windows) percent identity and fraction of bases in conserved blocks. Below the percent identity plots are shown insertions (gray boxes) and deletions (orange boxes) in the D. melanogaster sequence relative to their D. pseudoobscura ortholog. The location of binding sites in D. melanogaster , binding sites in D. pseudoobscura and aligned binding sites along with the density of sites averaged over 700 bp are shown in the bottom three panels for each region. Additional data file 5 gives the primers used to amplify pCRMs for transgenics. Additional data file 6 gives additional information from Table 2 . Additional data file 7 gives all new pCRMs from genome-wide eCIS-ANALYST located within 20 kb of annotated transcript. Additional data file 8 gives all new pCRMs from genome-wide eCIS-ANALYST located more than 20 kb from annotated transcript. Additional data file 9 lists genes with anterior-posterior patterns and the source of the information. Additional data file 10 gives all new pCRMs from genome-wide eCIS-ANALYST located within 20 kb of gene with anterior-posterior pattern. And, finally, Additional data file 11 gives all new pCRMs from genome-wide eCIS-ANALYST located between 20 kb and 50 kb from gene with anterior-posterior pattern. Supplementary Material Additional data file 1 The binding site densities (column 1), aligned site densities (column 2), and aligned plus preserved site densities (column 3) for individual transcription factors Click here for additional data file Additional data file 2 Expression patterns of 65 genes adjacent to 122 pCRMs identified by eCIS-ANALYST Click here for additional data file Additional data file 3 Discrimination of positive and negative pCRMs. Comparisons of the number of predicted binding sites in D. melanogaster pCRMs to the number of aligned sites (top panel) and aligned plus preserved sites (bottom panel) Click here for additional data file Additional data file 4 New pCRMs Click here for additional data file Additional data file 5 The primers used to amplify pCRMs for transgenics Click here for additional data file Additional data file 6 Additional information from Table 2 Click here for additional data file Additional data file 7 All new pCRMs from genome-wide eCIS-ANALYST located within 20 kb of annotated transcript Click here for additional data file Additional data file 8 All new pCRMs from genome-wide eCIS-ANALYST located more than 20 kb from annotated transcript Click here for additional data file Additional data file 9 Genes with anterior-posterior patterns and the source of the information Click here for additional data file Additional data file 10 All new pCRMs from genome-wide eCIS-ANALYST located within 20 kb of gene with anterior-posterior pattern Click here for additional data file Additional data file 11 All new pCRMs from genome-wide eCIS-ANALYST located between 20 kb and 50 kb from gene with anterior-posterior pattern Click here for additional data file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC522868.xml

539050

Randomized, Controlled Trial of Therapy Interruption in Chronic HIV-1 Infection

Background Approaches to limiting exposure to antiretroviral therapy (ART) drugs are an active area of HIV therapy research. Here we present longitudinal follow-up of a randomized, open-label, single-center study of the immune, viral, and safety outcomes of structured therapy interruptions (TIs) in patients with chronically suppressed HIV-1 infection as compared to equal follow-up of patients on continuous therapy and including a final therapy interruption in both arms. Methods and Findings Forty-two chronically HIV-infected patients on suppressive ART with CD4 counts higher than 400 were randomized 1:1 to either (1) three successive fixed TIs of 2, 4, and 6 wk, with intervening resumption of therapy with resuppression for 4 wk before subsequent interruption, or (2) 40 wk of continuous therapy, with a final open-ended TI in both treatment groups. Main outcome was analysis of the time to viral rebound (>5,000 copies/ml) during the open-ended TI. Secondary outcomes included study-defined safety criteria, viral resistance, therapy failure, and retention of immune reconstitution. There was no difference between the groups in time to viral rebound during the open-ended TI (continuous therapy/single TI, median [interquartile range] = 4 [ 1 – 8 ] wk, n = 21; repeated TI, median [interquartile range] = 5 [ 4 – 8 ] wk, n = 21; p = 0.36). No differences in study-related adverse events, viral set point at 12 or 20 wk of open-ended interruption, viral resistance or therapy failure, retention of CD4 T cell numbers on ART, or retention of lymphoproliferative recall antigen responses were noted between groups. Importantly, resistance detected shortly after initial viremia following the open-ended TI did not result in a lack of resuppression to less than 50 copies/ml after reinitiation of the same drug regimen. Conclusion Cycles of 2- to 6-wk time-fixed TIs in patients with suppressed HIV infection failed to confer a clinically significant benefit with regard to viral suppression off ART. Also, secondary analysis showed no difference between the two strategies in terms of safety, retention of immune reconstitution, and clinical therapy failure. Based on these findings, we suggest that further clinical research on the long-term consequences of TI strategies to decrease drug exposure is warranted.

Introduction Antiretroviral therapy (ART) has been a milestone in the treatment of HIV infection. Current treatment guidelines for HIV-1 infection in the United States recommend the initiation of ART in patients with CD4 T cell counts of less than 350 cells/μl [ 1 ]. In implementing these guidelines, health-care providers face the ongoing challenge of developing treatment strategies that minimize drug-related toxicity and adverse effects while retaining effective control of viral replication. Furthermore, treatment-associated costs (particularly in resource-poor areas), difficulty in maintaining long-term optimal adherence [ 2 ], and the emergence of viral resistance [ 3 , 4 , 5 ] have limited the feasibility of life-long ART-mediated viral suppression, increasing the need for alternative treatment strategies. Intermittent therapy strategies, consisting of alternating cycles on and off ART, have increasingly emerged as a potential intervention to address limitations of continuous ART [ 6 , 7 , 8 , 9 ]. Therapy interruption (TI) studies in ART-treated patients with suppressed HIV infection [ 10 ] have addressed the general questions as to whether such strategies can achieve greater viral control through increased antiviral responses (autoimmunization hypothesis) or simply serve as a strategy to reduce cost of long-term therapy and drug-associated toxicity. While pilot studies and uncontrolled (or incomplete) trials in patients with chronic HIV infection have addressed viral and immune outcomes of fixed-length TI and fixed on-drug cycles [ 11 , 12 , 13 , 14 , 15 , 16 ], no completed randomized, controlled trial has yet addressed by intent-to-treat analysis the outcome during an open-ended TI of sequential TIs versus continuous treatment in patients with confirmed suppression. The largest study to date in this area is the prospective single-arm Swiss–Spanish Intermittent Trial (SSITT) conducted in 133 recruited patients undergoing sequential 2-wk TIs and showing a lack of impact of this strategy on achieving sustained viral loads of less than 5,000 copies/ml off therapy in those that completed the study [ 11 ]. However, the lack of a control arm in this study has left unanswered questions about the impact of multiple TIs on time to rebound, immune reconstitution, therapy failure, and viral resistance when analyzed against a randomized control arm of continuous treatment followed for equal time before a single open-ended interruption. We completed a randomized, controlled trial on the outcome of repeated 2- to 6-wk TIs in patients with chronic infection in which the comparator group maintained continuous therapy and then an open-ended interruption period was applied in both treatment groups. The study addressed the potential for repeated interruptions of therapy to delay time to viral rebound as a primary outcome and analyzed secondary outcomes regarding study-defined safety criteria, viral suppression and resistance, and retention of immune reconstitution. Methods Participants Between August 2000 and December 2003, we enrolled 42 patients infected with HIV who were older than 18 y and on ART; eligibility criteria included CD4 counts of greater than 400 cells/μl on ART with a nadir of no less than 100 cells/μl, ART-mediated suppression (< 500 copies/ml) for more than 6 mo and less than 50 copies/ml at recruitment on any antiretroviral regimen. Approval of the study protocol was obtained from the institutional review board (IRB) of the Philadelphia Field Initiating Group for HIV Trials (Philadelphia, Pennsylvania, United States). Written informed consent was obtained from all patients. Human experimentation guidelines of the United States Department of Health and Human Services and of the authors' institutions were followed. The study protocol, including the patient consent form, the CONSORT form, and the IRB approval, can be found in Protocols S1–S4 . Randomization and Study Design Forty-two eligible patients from the Jonathan Lax Immune Disorder Clinic in Philadelphia, Pennsylvania, were randomized via sealed envelopes in a 1:1 fashion to a first phase (phase I) of either (1) three successive TIs of 2, 4, and 6 wk, respectively, or (2) maintenance of ART for 40 wk before a final interruption of therapy in both arms (phase II) subject to therapy reinitiation criteria as described below. Phase II consisted of an open-ended interruption to allow for virological and immunological comparisons between the groups off therapy. Study visits were every 2 wk for the repeated interruptions group and every 4 wk for the continuous ART group during phase I. Both groups were followed every 2 wk during phase II. We followed a study design with step-wise increases in the length of TI cycles to address potential safety concerns (resuppression was confirmed after shorter TIs before longer interruptions were initiated) and the hypothesis that sequential viral replication intervals would stimulate viral control and a delay in time to viral rebound. Phase I procedures for the repeated interruptions group included the following. (1) Interruption of therapy was individually timed to occur after two HIV RNA measurements of less than 50 copies/ml without any viral load measurements greater than 400 copies/ml in between; these interruptions increased from 2 to 4 to 6 wk sequentially. (2) If a 0.5-log or greater reduction in viral load did not occur by 6 wk of reinitiated therapy or less than 50 copies/ml was not achieved within 20 wk of reinitiated therapy, patients were withdrawn as therapy failures and a resistance test was performed. (3) Patients were also withdrawn as therapy failures if (a) the CD4 cell number declined by more than 45% of the baseline CD4 count, (b) participants developed an opportunistic infection, even if retaining required CD4 count levels, or (c) a viral load of greater than 500,000 copies/ml occurred once, with or without development of acute retroviral syndrome as defined by fever, skin lesions, and pharyngitis. Phase I procedures for the continuous therapy arm included the following: (1) patient monitoring if detected viremia was between 50 and 999 copies/ml, with the patient withdrawn if their viral load did not return to less than 50 copies/ml immediately prior to phase II, and (2) patient study withdrawal as therapy failure if during the 40-wk ART period viral load rebounded to more than 1,000 copies/ml at two consecutive time points. Phase II procedures for both arms included the following: (1) monitoring for patient study withdrawal criteria as described in phase I, (2) determining time to primary end point of a viral load greater than 5,000 copies/ml, (3) monitoring until the time of therapy reinitiation at a viral load greater than 30,000 copies/ml for three consecutive time points, and (4) after reinitiation of therapy, follow-up on therapy to confirm resuppression to less than 50 copies/ml at 6, 10, and 14 wk on therapy. Clinical and laboratory parameters (CD4 count and viral load) were monitored at each visit, and venous blood was collected for additional secondary outcomes during selected study visits. In both phase I and II, participants taking non-nucleoside reverse-transcriptase inhibitors (NNRTIs) were instructed to stop them a day earlier than the remaining drugs in the regimen. Primary and Secondary Outcomes The primary outcome was time to confirmed virological rebound during phase II. Rebound was defined as first time point with greater than 5,000 copies/ml. Viral replication magnitude as defined by mean HIV-1 plasma RNA area under the curve (AUC HIV RNA ) was measured as a secondary outcome at weeks 12 and 20 of phase II based on reinitiation-of-therapy criteria outlined above. Additional secondary outcomes included (1) safety outcomes (serious adverse events [SAEs] and patient withdrawal based on criteria defined above), (2) retention of ART-mediated immune reconstitution, and (3) detection of viral resistance. Retention of immune reconstitution was analyzed by (1) same-day whole blood flow-cytometry-based analysis of CD4 and CD8 T cells, including total and naïve (CD62 l/CD45RA) and memory (CD45RO) subsets as described [ 17 ], and (2) same-day recall response analysis of peripheral blood mononuclear cell lymphoproliferative responses to Candida albicans as described [ 17 ]. Viral resistance mutations were retrospectively analyzed on cryopreserved plasma samples by genotyping of first available sample with viral load greater than 100 copies/ml following each interruption using the TruGene Assay (Visible Genetics, Toronto, Canada) at the Gladstone Institute of Virology and Immunology (San Francisco, California, United States) as previously described [ 18 , 19 ]. Sample Size The sample size required was calculated using PS [ 20 ] software, and based on a type I error of 0.05, with 90% power, to detect a difference of 4 wk or more in time to viral rebound between arms. Eighteen patients per group resulted in sufficient power (18 for 90%, 13 for 80%) to determine a difference of 4 wk or greater between groups in time to rebound of virus during the open-ended interruption. Assuming a loss to follow-up of 15%, we targeted 21 patients per group, or 42 total. Statistical Analysis The primary analysis was an intent-to-treat analysis in which dropouts were assigned a week 0 rebound time (e.g., maximum failure to delay rebound). In secondary analyses, these dropouts were excluded. The log-rank test was used to test the null hypothesis of no difference between arms in the number of weeks from initiation of the open-ended TI to reaching viral rebound as defined. Patients not reaching end point at 26 wk after the beginning of the open-ended TI were censored. Wilcoxon rank sum tests were used to compare baseline and week 0 of the open-ended interruption between groups. Wilcoxon signed rank tests were used to test for no change from baseline to week 0 of phase II. Finally, Wilcoxon rank sum tests were employed to test between groups for equality of the mean AUC HIV RNA up to 12 and 20 wk. In all cases, a two-sided alpha level of 0.05 was used to define statistical significance. Unless otherwise stated, results are presented as median (interquartile range) in text and tables. Results Patient Flow and Discontinuations Trial patient flow is summarized in Figure 1 . Between August 2000 and December 2003, 42 patients at the Jonathan Lax Immune Disorder Clinic at the Philadelphia Field Initiating Group for HIV Trials were enrolled, randomized, and followed as shown in Figure 2 . In the continuous therapy/single interruption arm, 16 of 21 patients reached the open-ended interruption. Reasons for study discontinuation in this arm were loss to follow-up ( n = 1; patient moved away) and virological failure during continuous therapy ( n = 4; further discussed below). In the repeated interruptions arm, 18 of 21 patients reached the open-ended interruption following three TIs of 2, 4, and 6 wk duration, with median peak rises in viral loads of 136 (50–2,590), 13,651 (180–222,589), and 18,887 (3,893–96,101) copies/ml, respectively. Median time to less than 50 copies/ml after resumption of therapy was 2 (0–4), 3 (1.8–12), and 9.5 (2–12) wk, respectively, with 9, 18, and 20 wk as the maximum time needed to achieve suppression in 100% of patients before reaching the open-ended interruption. Study discontinuation in the repeated interruptions arm was due to protocol violation ( n = 1; patient restarted therapy during interruption out of protocol), loss to follow-up ( n = 1; patient imprisoned), and virological failure during on-therapy period ( n = 1; further discussed below). Figure 1 Study Flow Figure 2 Study Design (Phases I and II) Baseline Criteria and Follow-Up The demographic and clinical characteristics of the two groups at baseline are summarized in Table 1 . Seventy-five percent of participants were on their second to fourth regimen while 25% were in their first regimen . No significant difference was found in baseline parameters between arms, with 33%–47% of patients on protease-inhibitor-containing and 61%–71% on NNRTI-containing regimens. Owing to the high participation of patients on NNRTI-based regimens and concerns about TI and safety in general, patient outcomes and treatment failure were reviewed monthly by the IRB of this study during the first 8 mo of study, quarterly for the following 4 mo, and semi-annually thereafter. Figure 2 shows study design for both arms, with a median follow-up of 41 (41–42) wk during phase I for the continuous therapy/single interruption arm and 42 (30–51) wk for the repeated interruptions arm. Follow-up during phase II had a median duration of 27 wk in both arms (continuous therapy/single interruption arm, 27 [8.75–47]; repeated interruptions arm, 27 [16.5–35]). Following reinitiation of therapy after phase II, patients suppressed viral replication to less than 50 copies/ml by a median time of 10 (6–12) wk in both arms, excluding for two patients in the continuous therapy/single interruption arm who elected to stay off ART indefinitely and one patient from the repeated interruptions arm who reported nonadherence following regimen reinitiation yet reached 52 copies/ml before withdrawing from additional follow-up. Table 1 Baseline Demographic and Clinical Characteristics per Study Arm a Numbers include cases of PI/NNRTI combined use at study entry AA, African American; C, Caucasian; H, Hispanic, IV, intravenous drug usage; PI, protease inhibitor; S, sexual transmission Primary Outcome An intent-to-treat analysis of the time to viral rebound (>5,000 copies/ml) in the open-ended interruption showed no difference between groups (continuous therapy/single TI, median = 4 [ 1 – 8 ] wk, n = 21; repeated TI, median = 5 [ 4 – 8 ] wk, n = 21; p = 0.36). Figure 3 (top panel) shows the probability of plasma HIV-1 RNA remaining less than 5,000 copies/ml for the two groups ( n = 21 per group). Exclusion of drop-outs in an as-treated analysis did not alter conclusions (single TI, median = 5 [ 4 – 9 ] wk, n = 18; repeated TI, median = 6 [ 4 – 8 ] wk, n = 16; p > 0.05). Additional secondary analysis of the magnitude of viral load as shown in Figure 3 (second panel) showed similar viral replication as determined by mean AUC HIV RNA analysis at week 12 (single TI, median = 124,621 [ 23 ,326–262,348] AUC HIV RNA ; repeated TI, median = 100,400 [47,221–365,731] AUC HIV RNA ; p > 0.05) or week 20 (single TI, median = 114,550 [ 31 ,829–362,628] AUC HIV RNA ; repeated TI, median = 153,097 [67,427–515,421] AUC HIV RNA ; p > 0.05). Figure 3 Lack of a Difference between Groups in Plasma HIV-1 RNA during Phase II Top panel shows Kaplan-Meyer plot summarizing time to a viral load of more than 5,000 copies/ml in both arms. Second panel shows viral load (mean ± standard error) per arm during 27 wk of TI (median time of phase II). Bottom table shows number of patients at time points shown for viral load in the second panel; the decrease in viral load over time is due to the reinitiation of therapy in patients with higher viral loads. Secondary Outcomes SAEs and patient discontinuation No patient discontinuation in either group was due to study-defined changes in CD4 cell count (reviewed further below) or due to study-associated SAEs (disease progression or acute retroviral syndrome). However, four non-study-related SAEs occurred: two patients from the continuous therapy/single interruption arm were hospitalized, one for a cholecystectomy and one for acute rectal bleeding, during the 40-wk ART period; a patient from the repeated interruptions arm died of liver cancer during week 26 of the open-ended interruption after previously reaching a viral load greater than 5,000 copies/ml yet electing to stay off ART; and a patient from the repeated interruptions arm developed a transient ileitis. Immune reconstitution No significant difference was observed between groups in CD4 T cell counts at the start of phase II, as illustrated in Figure 4 . In addition, no difference in the percentage of naïve CD4 cells or decrease of recall response to C. albicans was observed, confirming the absence of significant differences in the retention of baseline immune reconstitution correlates between arms. However, a significant decrease in the abundance of CD4 cells relative to other T cell types as summarized in CD4% (but not in absolute CD4 count ) was present in the repeated TI arm, corresponding to a significant increase in CD8 T cell count. In spite of fluctuations in CD4 T cell count levels between the start and end of each monitored TI, a recovery of CD4 count levels was achieved upon resuppression following each TI in conjunction with a retention of lymphoproliferative responses against C. albicans before, during, and after each TI, as illustrated in Figure 5 . Figure 4 T Cell Subsets and Recall Lymphoproliferative Response at the End of Phase I End of phase I values for each arm are summarized (median and first and third quartiles) in the stacked figures showing from top to bottom: CD4 T cells/μl, CD4%, CD4 − CD45RA + CD62L + % (naïve phenotype), CD8 T cells/μl, CD8%, and C. albicans lymphoproliferative response (shown as stimulation Index, SI). Unpaired p values for each variable are shown above corresponding bracket. Figure 5 CD4 T Cells/μl and T Cell Recall Lymphoproliferative Response during Sequential TIs in Phase I Shown are data from the repeated interruptions arm. Panels show the TI initiation visit and TI end visit of each sequential TI inclusive of the initiation visit for phase II (open-ended TI). Viral resistance mutations and therapy failure An intent-to-treat analysis of the combined number of patients per arm with detected resistance mutations irrespective of therapy failure in phase I and during the final TI in phase II showed no significant difference between arms (continuous therapy/single TI, 7/21; repeated TI, 10/21; p > 0.05). Study-defined criteria for therapy failure of a previously suppressive regimen were met by 4/21 patients in the continuous therapy/single interruption arm (patients S37, S47, S52, and S59) in association with self-reported nonadherence to therapy and detection of resistance mutations in phase I, as listed in Table 2 . One patient in the repeated interruptions arm (1/21; patient S56) failed therapy after 20 wk following the third TI by maintaining a viral load between 50 and 999 copies/ml in the presence of previously undetected resistance mutations. Table 2 Therapy Failures with Plasma HIV-1 Protease and Reverse Transcriptase Inhibitor–Associated Resistance Patterns during on Therapy Periods (Study Phase I) Bold identifies drugs for which mutations were detected in plasma 3TC, lamivudine; ABV, abacavir; ddI, didanosine; d4T, stavudine; EFZ, efavirenz; NVP, nevirapine; RT, reverse transcriptase a Mutations associated with patient's regimen In patients who reached phase II in the absence of therapy failure, a total of 12 patients were identified to have resistance mutations at the first viremic time point (continuous therapy/single TI, 3/16; repeated TI, 9/18; p = 0.06). A greater number of resistance mutations was detected in the repeated interruption arm, as summarized in Table 3 . In ten out of these 12 patients, a change in resistance patterns was observed when comparing the first viremic time point to the last. All 11 of 12 patients in Table 3 who reinitiated therapy retained suppressive ability of their respective regimens, as did all other patients who did not show resistance mutations in phase II. In the repeated interruptions arm, analysis of newly detected resistance mutations in phase II, as defined by a lack of detection during viremic time points in phase I, identified 3/18 patients (patients S4, S22, and S43) with this pattern (see notations in Table 3 ). Table 3 Non-Therapy Failures with Resistance Detected off ART at First and Last Viremic Time Point in Comparator Open-Ended TI (Phase II) Bold identifies drugs against which mutations were detected a Mutations associated with patient's regimen b Patient/physician changed regimen after open-ended interruption for reasons not related to suppression activity on previous regimen: patient S7 changed to 3TC, TNV, EFZ, NVP; patient S40 changed to LOP, RTV, ddI, TNV; and patient S35 changed to LOP, RTV, ABV, TNV c Mutations not detected at the first plasma HIV-1 RNA tested during prior TIs d Resistance shown for patient S35 is last available, at week 2 of the third TI e Patient S45 was lost to follow-up after the end of the third TI. Resistance shown is last available, at week 6 of the third TI. Resuppression noted after completion of the third TI 3TC, Lamivudine; ABV, Abacavir; d4T, Stavudine; ddI, Didanosine; EFZ, Efavirenz; LOP, Lopinavir; NLF, Nelfinavir; NVP, Nevirapine; TNV, Tenofovir; ZDV, Zidovudine Discussion Earlier reports on TI strategies in patients with chronic HIV infection include multiple pilot or single-arm study designs centered on the effects on viral control by comparison with pre-therapy periods, detection of resistance mutations without parallel follow-up of a continuously treated arm, and inclusion of variable criteria regarding viral resuppression before proceeding with repeated TIs [ 11 , 12 , 14 , 16 ]. In contrast, our strategy mandated resuppression of viral replication to less than 50 copies/ml before each TI and presents the first comparison of viral replication during a final open-ended interruption of therapy between patients randomized to complete three sequential TIs or stay under continuous therapy. Our data, based on intent-to-treat analysis, did not show that repeated TIs resulted in a clinically significant virological benefit as measured by the time to viral rebound to more than 5,000 copies/ml (see Figure 3 ). Secondary as-treated analysis on viral replication magnitude also indicated a lack of difference between arms. Consistent with the findings of SSITT [ 11 ], analysis of our data by the categorical classification of a “responder” as a patient with viral load less than 5,000 copies/ml at week 12 off therapy showed no significant difference in this frequency between arms (single TI, 5/18; repeated TI, 5/16), suggesting the presence of “responders” irrespective of previous protocol-mandated TIs. Based on secondary outcome measures, the incidence of adverse events (SAEs, therapy failure, and patient discontinuation) or clinical disease progression (as indicated by CD4 count on therapy or opportunistic infections) was not observed to be different between arms. Prospective safety outcomes in our study are in accordance with reports from a retrospective analysis of 1,290 patients who interrupted treatment at least once (< 3 mo) without an increased risk of HIV-associated morbidity or mortality (with the exception of patients in Center for Disease Control and Prevention stage C during first interruption only) [ 21 ]. In regards to immunological outcomes, a concern associated with interruption of suppressive therapy is the potential for irreversible, viral-mediated CD4 T cell loss leading to disease progression [ 6 , 22 ]. We did not observe a decrease in CD4 cell numbers or lymphoproliferative responses against C. albicans when measured between arms before the open-ended TI (see Figure 4 ), nor following resuppression after monitored TI reinitiation cycles in the repeated interruptions arm (see Figure 5 ). The latter is consistent with observations by others and does not support an immediate immunological “cost” to short-term TIs [ 12 , 14 , 15 , 16 , 23 ]. However, we do show that monitoring CD4 cell numbers by percentage could lead to misinterpreting a significant loss of CD4 cells as a result of a significant increase in CD8 count following TIs, even though absolute CD4 count numbers remained unchanged (see Figure 4 ). Interestingly, the increase in CD8 T cell number also corresponded with an increase in HIV-specific responses as measured by interferon-gamma expression (data not shown), which in light of an absence of effect on viral load between arms further supports that TI strategies alone may not significantly alter the pre-existing balance between viral replication and host antiviral responses [ 14 , 16 , 23 , 24 ]. Importantly, no evidence for an increase of viral resistance in association with therapy failure was present in the repeated interruptions arm (See Table 2 ). We did not observe a greater clinical failure of NNRTI-based regimens in the repeated interruption arm due to “single drug” periods as predicted by recently redefined drug half-life estimates and the presence of viral replication during each interruption [ 25 , 26 , 27 ]. However, the percentage of patients with resistance mutations detected in this study in the repeated interruption arm (47%) is higher than the 17% observed in the SSITT cohort [ 11 ], in which patients with prior treatment failures were excluded [ 28 ]. We interpret this difference to mean that the resistance detected off drug in both our and their cohorts is likely associated with the greater number of drug-experienced patients in our cohort (75%) and the detection of prior archived resistance mutations as supported by Metzner et al. [ 29 ], who documented in 14/25 (56%) SSITT patients the presence of minor populations of M184V occurring at least once off drug during interruption of therapy. In spite of the lack of difference in the total number of patients with resistant mutations detected on therapy during phase I and off therapy in phase II (7/21 [33%] versus 10/21 [47%], respectively) in both arms, we do report in similarity to others a greater detection of resistance mutations in the TI arm when restricting analysis to the last off-drug period only [ 29 , 30 ] as three of 16 (18%) had mutations detected off drug in the continuous therapy/single interruption arm compared to nine of 18 (50%) in the repeated interruption arm. However, based on the lack of association between viral resistance detected off-drug shortly after TI and resuppression by the same regimen in all patients, it remains undetermined to what extent TIs favor the detection of archived mutations in chronically suppressed patients and to what extent these mutations are a signal for a future therapy failure. The latter is best exemplified by the data we collected on patients on NNRTI-based regimens in the repeated interruptions arm where two patients (S19 and S43) showed K103N detection (only during the off-drug periods) in the absence of therapy failure while maintaining the same regimen after each TI, including post-study follow-up ( Table S1 ). On the other hand, virological failure in the continued presence of an NNRTI-based regimen in phase I was associated with detection of K103N, as observed in one patient (S56) in the repeated interruption arm and three patients (S37, S52, and S59) in the continuous therapy arm with self-reported non-adherence. Drug resistance that occurs during virological drug failure predicts virological responses to salvage treatment [ 31 , 32 , 33 ]. In contrast, the clinical implications of drug resistance mutations that appear shortly after TI in chronically suppressed patients are not clear. Case reports in this cohort of patients have demonstrated that drug-resistant variants that appeared during TIs may not persist in subsequent time points even after repeated use of the same antiretroviral regimen [ 19 , 34 ]. We now observe that drug resistance appearing during TIs can be transient since 50% and 33% of patients listed in Table 3 showed complete and partial reversion to wild type, respectively, when comparing to resistance at the last available viremic time point in phase II (See Table 3 ). Further, we observed durable resuppression of plasma viral RNA level in many patients who had drug-resistance mutations off therapy that would otherwise be expected to affect part of their treatment regimen when reinitiated (see Table S1 ). Virus populations that expand shortly after TI may lack all of the adaptations required to achieve high levels of plasma viremia in the presence of drug during continuous treatment. These adaptations may include the resistance-associated mutations, which were detected, as well as secondary mutations that may increase the viral replication capacity [ 35 , 36 ] or envelope adaptations required to escape concurrent humoral immune responses [ 37 , 38 ]. It is of interest to note that despite the large amount of research activity on TIs in patients with suppressed chronic infection and the hundreds of monitored interruptions studied to date, only limited cases of development of clinical resistance (as evidenced by a lack of viral resuppression following therapy reinitiation) have emerged, in contrast to the multiple reports of detection of viral sequences off ART associated with resistance as shown in this study and others [ 11 , 19 , 29 , 30 , 39 , 40 ]. Taken together, while our data show no clinically significant benefit for repeated TIs of less than 1.5 mo in patients with CD4 counts greater than 400 on therapy with regard to viral control as defined by time to rebound, secondary outcomes document no significant difference in levels of retention of immune reconstitution between arms and no increased incidence of virological failure as a consequence of TIs. While our data indicate that this TI strategy should not be pursued outside of a clinical trial setting, we argue that it will be important to collect additional data on the potential benefits of drug-sparing regimens (such as reduced long-term toxicity and reduced cost) and to define long-term outcomes in comparison with continuous therapy. Supporting Information Registration of randomized trial at clinicaltrials.gov under identifier NCT00051818. Protocol S1 Protocol Text: Effects of Sequential TI (614 KB DOC). Click here for additional data file. Protocol S2 Study IRB Approval Current IRB approval for study at clinical site. (179 KB PDF). Click here for additional data file. Protocol S3 Wistar IRB Approval IRB approval to receive study biological material at the Wistar Institute for research. (201 KB PDF). Click here for additional data file. Protocol S4 CONSORT Checklist (50 KB DOC). Click here for additional data file. Table S1 Patients with Detected Resistance during Phase II: Regimen at Initiation of Phase II and Subsequent Post-Study Follow-Up to August 2004 (36 KB DOC). Click here for additional data file. Patient Summary Why Was This Study Done? Highly active antiretroviral therapy has revolutionized HIV treatment for patients who have access to the medications. But the drugs are expensive, have side effects, and can become ineffective when the virus develops resistance. Structured treatment interruptions (STIs), also known as “drug holidays” (because patients take a holiday from their drugs), have been suggested as possible alternatives to continuous therapy. Initially, there was fear that patients who went back on therapy after an interruption would not be able to control the virus again, but there was also hope that STIs might actually strengthen the immune system. In addition, STIs might alleviate some side effects, and they would certainly reduce costs. This study uses a particular design to examine the risks and benefits of STIs. What Did the Researchers Do? The researchers studied 42 patients who received either continuous therapy for 40 weeks or three successive treatment interruptions of two, four, and six weeks, followed by a final open-ended interruption for both groups. The researchers then recorded how long patients were able to control the virus before their viral load reached a certain threshold and they had to restart therapy. They also examined CD4 counts and therapy failure, and looked for resistant viruses on and off therapy. What Did They Find? In terms of being able to control the virus, it made no difference whether patients were on continuous therapy or had three STIs. In other words, when both groups stopped treatment at 40 weeks, the length of time that the patients could control the virus was the same in both groups. Eventually, all patients (except two who elected to stay off antiretroviral therapy) re-initiated therapy because of a rising viral load, and the patients once on therapy all regained control over the virus. Resistant viruses were found in patients from both groups, but during the final interruption they were more common in the group that had received the three STIs. What Does This Mean? The study confirms that STIs do not help with viral control, consistent with other studies that found that STIs had no clinical benefit. On the other hand, no short-term adverse events were present, as all patients were able to regain control over the virus after they went back on treatment (without a drop in CD4 count), even after several rounds of interruptions and tests to detect of resistant viruses. There remains concern about whether recurrent cycles of viral replication and suppression might in themselves be harmful, and whether the presence of resistant virus is a signal for future treatment failure. Given these unanswered questions, STIs should only be undertaken within clinical trials. What Next? Possible risks and benefits of STIs in the management of therapy remain an active area of research. Evidence so far has not shown clinical benefits. Ongoing studies need to clarify whether there are long-term risks (and what they are), so that we can weigh these against the benefits of reducing costs and side effects. Additional Online Information The Body information Web page on STIs: http://www.thebody.com/treat/sti.html Information on “continuing antriretroviral treatment” from AVERT, an international HIV and AIDS charity based in the United Kingdom: http://www.avert.org/conttrt.htm Information on STIs from NAM, a United Kingdom registered charity: http://www.aidsmap.com/en/docs/7980314C-97B5–412F-93B1-AD8B64F51F73.asp Factsheet on HIV treatments from the United States National Institute for Allergy and Infectious Diseases: http://www.niaid.nih.gov/factsheets/treat-hiv.htm Search results from Clinicaltrials.gov when searching for “HIV” and “treatment interruption” combined terms: http://www.clinicaltrials.gov/search/term=%22Treatment+Interruption%22%5BCONDITION%5D+AND+HIV+%5BCONDITION%5D

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC539050.xml

549548

The evolution of the Sin1 gene product, a little known protein implicated in stress responses and type I interferon signaling in vertebrates

Background In yeast, birds and mammals, the SAPK-interacting protein 1 (Sin1) gene product has been implicated as a component of the stress-activated protein kinase (SAPK) signal transduction pathway. Recently, Sin1 has also been shown to interact with the carboxyl terminal end of the cytoplasmic domain of the ovine type I interferon receptor subunit 2 (IFNAR2). However, the function of Sin1 remains unknown. Since SAPK pathways are ancient and the IFN system is confined to vertebrates, the organization of the Sin1 gene and the sequences of the Sin1 protein have been compared across a wide taxonomic range of species. Results Sin1 is represented, apparently as a single gene, in all metazoan species and fungi but is not detectable in protozoa, prokaryotes, or plants. Sin1 is highly conserved in vertebrates (79–99% identity at amino acid level), which possess an interferon system, suggesting that it has been subjected to powerful evolutionary constraint that has limited its diversification. Sin1 possesses at least two unique sequences in its IFNAR2-interacting region that are not represented in insects and other invertebrates. Sequence alignment between vertebrates and insects revealed five Sin1 strongly conserved domains (SCDs I-V), but an analysis of any of these domains failed to identify known functional protein motifs. SCD III, which is approximately 129 amino acids in length, is particularly highly conserved and is present in all the species examined, suggesting a conserved function from fungi to mammals. The coding region of the vertebrate Sin1 gene encompasses 11 exon and 10 introns, while in C. elegans the gene consists of 10 exons and 9 introns organized distinctly from those of vertebrates. In yeast and insects, Sin1 is intronless. Conclusions The study reveals the phylogeny of a little studied gene which has recently been implicated in two important signal transduction pathways, one ancient (stress response), one relatively new (interferon signaling).

Background Sin1 was originally described as a human protein that modulated Ras function in Saccharomyces cerevisiae [ 1 ]. Strains of yeast that expressed the constitutively activated RAS2 Val 19 mutation had elevated levels of cyclic AMP, impaired growth control and were acutely sensitive to heat shock. This phenotype was reversed when the yeast strain was transfected with a cDNA (clone JC310) that encoded a then unknown protein. Although the authors suggested that the inferred interaction between the JC310 product and RAS might be fortuitous, they favored the possibility that that the unknown protein either was a true inhibitor of RAS or that it was a RAS target protein, which when over-expressed, had a protective action. A S. cerevisiae protein encoded by the AVO1 gene showed distant similarity the human JC310 product [ 2 , 3 ]. Approximately eight years after the identification of JC310 it was again identified, on this occasion in a yeast two-hybrid screen of a Schizosaccharomyces pombe cDNA library as a 665 amino acid protein that bound via polypeptide sequences in its C-terminal 244 amino acids to the Sty1/Spc1, stress activated MAP kinase (SAPK) [ 2 ]. A fission yeast strain lacking the Sin1 gene was sterile, sensitive to multiple types of stress, including heat shock, and had delayed cell cycles compared to a parental strain. Sin1 acted downstream of activated Sty1/Spc1 and appeared to be necessary for normal function of the transcription factor Atf1, a homolog of human ATF2. Wilkinson et al. [ 2 ] found that an apparent full length homolog of Sin1 from chicken allowed the heat sensitive strain of S. pombe to grow at 37°C, albeit very poorly. Moreover, fusion of the first 486 amino acids of yeast Sin1 (which does not restore growth) with the C-terminal 182 amino acids of the chicken Sin1 sequence protected against heat shock. Together, these data showed that Sin1 functions as a component of the stress-activated Sty1/Spc1 MAP kinase pathway in S. pombe and that a functional homolog of Sin1 exists in vertebrates. No further information concerning Sin1 appeared since the paper of Wilkinson et al. [ 2 ] until our discovery that the ovine (ov) Sin1 associated via its C-terminus to the cytoplasmic domain of IFNAR2, a subunit of the type I IFN receptor [ 4 ], and Schroder et al. [ 5 ] described transcripts for Sin1 in human tissue and provided an analysis of the human gene. The latter study confirmed that Sin1 was relatively well conserved across Metazoa and fungi (Ascomyctes and Basidiomycetes) and was also represented in amoebae, but not in other protozoan species. Ovine Sin1, which is 88% identical in sequence to chicken Sin1, can be co-immunoprecipated with the IFN receptor subunit IFNAR2 and shows a similar subcellular distribution to the receptor protein when co-expressed in mammalian cells [ 4 ]. Although ovSin1 was identified from a cDNA present in ovine endometrium and was initially considered to have a role in reproduction associated with the action of IFN-τ on the uterus during early pregnancy in the sheep, it became clear that the Sin1 gene was expressed in tissues other than endometrium and might have a general role in the action of type 1 IFN. In particular, it seemed possible that Sin1 might link the action of IFN to the stress activated SAPK signal transduction pathways. Such a linkage has been inferred from earlier studies in which early activation of p38 MAPK had been noted following exposure of a variety of cell lines to IFN-α, -β, or -τ [ 6 - 12 ]. Although the SAPK pathway is itself ancient and is found in all the species in which the Sin1 gene exists, the IFN system of receptors and ligands is restricted to vertebrates. We reasoned, therefore, that an analysis of Sin1 gene sequences might not only provide insight into the function of Sin1, but indicate how the protein evolved to interact with IFNAR2. The fact that the Sin1 gene appears to be expressed ubiquitously, that it is highly conserved across a wide range of taxa, and that it is a likely participant in several important signaling pathways, makes it an intriguing candidate for a functional/evolutionary analysis. Results Conservation of the Sin1 gene from yeast to mammals A combination of searching methods was employed to locate Sin1 genes in available cDNA and genome data bases (Table 1 ). Sin1 sequences were found in two yeast species ( Schizosaccharomyces pombe and Saccharomyces cerevisiae ), the red bread mold ( Neurospora crassa ) and a number of other fungal species (not shown here), Caenorhabditis elegans , a mosquito species ( Anopheles gambiae ), fruit fly ( D. melanogaster ), frog ( Xenopus laevis ), two fish species ( Fugu rubripes and Danio rerio ), chicken ( Gallus gallus ), mouse ( Mus musculus ), rat ( Rattus norvegicus ), human ( Homo sapiens ), sheep ( Ovis aries ), cattle ( Bos taurus ), and pig ( Sus scrofa ) (Table 1 ). No apparent ortholog could be detected in the plant Arabidopsis thaliana . Nor could sequences corresponding to Sin1 be found in protozoa other than amoebae and prokaryotic species. Table 1 Sin1 genes and their GenBank accession numbers Organism GenBank Accession No. Comments Yeast ( Saccharomyces cerevisiae NP_014563 Blastp the yeast protein database with fission yeast Sin1 protein. Yeast ( Schizosaccharomyces pombe NP_594703 Wilkinson et al. 1999. Red bread mold ( Neurospora crassa XP_322410 Blastp protein databases with budding yeast Sin1 protein. Worm ( Caenorhabditis elegans ) NM_064195 Blastp the worm protein database with ovSin1 protein. Fly ( Drosophila melanogaster ) AE003814 Blastp the fly protein database with ovSin1 protein. Mosquito ( Anopheles gambiae ) XM_319576 Blastp the mosquito protein database with ovSin1 protein. Fish ( Fugu rubripes ) N.A. Blastp the fugu protein database with ovSin1 protein. Frog ( Xenopus lavis ) BC043789 Search EST databases with chicken Sin1 cDNA. Chicken ( Gallus gallus ) AF153127 Wilkinson et al. 1999. Mouse ( Mus musculus ) BQ713136, BF781677, BU152256 Search mouse EST and genome databases with sheep Sin1 cDNA Rat ( Rattus norvegicus ) CK476507, BE127132, BF553331, BU759329, AW141364 Search EST and genome databases with sheep Sin1 cDNA. Pig ( Sus scrofa ) CF791532, CF178115, BP459453, CF177341 Search EST databases with sheep Sin1 cDNA. Cattle ( Bos taurus ) BF230134, AV603930, CB433957, BM480500 Search EST databases with sheep Sin1 cDNA. Sheep ( Ovis aries ) AY547378 Wang oberts, 2004 Human ( Homo sapiens ) NM_024117, BC002326 Search human EST and genome database with sheep Sin1 cDNA "Comments" briefly describe the methods used to obtain the sequences. The marked dissimilarity in inferred amino acid sequence between Sin1 from vertebrates and C. elegans (25% identity, Table 2 ), between the two yeast species (29% identity, Table 2 ; see Additional file: 1 ) and between S. pombe and N. crassa (28% identity, Table 2 , see Additional file: 2 ) in the approximately 500 aa of overlap suggests that even if homologs existed in plants and prokaryotes they would likely be overlooked by the search methods employed. Table 2 Pairwise comparisons of Sin1 cDNA and amino acid sequences from various species S. pomb e (665 aa) N. crassa (798 aa) C. elegans (684 aa) D. melanogaster (569 aa) A. gambiae (548 aa) F. Rubripes (530 aa) X. lavis (520 aa) G. gallus (522 aa) M. musculus (522 aa) R. norvegicus (522 aa) O. aries (522 aa) B. Taurus (522 aa) S. scrofa (522 aa) H. sapiens (522 aa) S. pombe (665 aa) - 28.2 34.1* 28.8* 35.6* 21.7 28.2 21.1 24.6 26.1 25.5 25.4 24.8 25.3 N. crassa (798 aa) NA - 32.6* 28.3* 30.7* 22.6 31.9* 20.3 36.2* 20.5 20.9 20.9 20.9 20.5 C. elegans (684 aa) NA NA - 27.5* 29.4* 22.3 22.6 21.8 23.0 23.0 23.5 23.2 22.8 23.2 D. melanogaster (569 aa) NA NA NA - 46.0 31.8 34.5 32.9 35.2 35.3 31.1 31.7 32.0 31.9 A. gambiae (548 aa) NA NA NA NA - 34.9 33.4 35.3 33.8 34.0 33.2 33.0 33.2 33.4 F. rubripes (530 aa) NA NA NA NA NA - 80.0 82.9 78.7 79.2 79.8 80.2 79.8 80.4 X. lavis (520 aa) NA NA NA NA NA 71.8 - 88.5 84.4 85.0 84.6 85.2 85.2 85.6 G. gallus (522 aa) NA NA NA NA NA 74.7 78.0 - 88.0 88.3 88.3 89.3 89.3 90.0 M. musculus (522 aa) NA NA NA NA NA 73.6 76.1 80.9 - 99.2 91.7 96.9 97.3 96.9 R. norvegicus (522 aa) NA NA NA NA NA 73.6 76.3 80.5 96.2 - 91.3 96.9 96.9 96.9 O. aries (522 aa) NA NA NA NA NA 73.3 75.3 81.7 96.0 96.0 - 98.7 97.9 98.1 B. taurus (522 aa) NA NA NA NA NA 73.4 75.6 81.7 91.9 91.8 98.2 - 98.9 99.0 S. scrofa (522 aa) NA NA NA NA NA 73.1 75.8 82.3 92.5 92.2 94.7 95.7 - 98.7 H. sapiens (522 aa) NA NA NA NA NA 73.9 75.9 82.8 92.4 92.5 94.1 94.9 95.7 - Notes: 1. Numbers in the upper-right half above the diagonal are identity percentages for amino acid sequences. 2. Numbers in the lower-left half below the diagonal are identity percentage for DNA sequences. 3. Numbers below the species names are the lengths of the Sin1 protein. 4. NA, Not applicable, i.e. no significant similarity was found. 5. Astericks, significant similarity occurs only in one region of the protein. For details, see the notes below: S. pombe-C. elegans: significant similarity occurs in one region (170 aa: 252–391). S. pombe- D. melanogaster : significant similarity occurs in one region (120 aa: 278–407). S. pombe- A. gambiae: significant similarity occurs in one region (94 aa: 282–375). N.crassa-C. elegans: significant similarity occurs in one region (90 aa: 376–465). N.crassa-D. melanogaster: significant similarity occurs in one region (58 aa: 407–464). N.crassa-A. gambiae: significant similarity occurs in one region (207 aa: 250–456). N.crassa-Xenopus: significant similarity occurs in one region (127 aa: 338–464). N.crassa-M. musculus: significant similarity occurs in one region (127 aa: 338–464). C. elegans-D. melanogaster: significant similarity occurs in one region (169 aa: 198–366). C. elegans-A. gambiae: significant similarity occurs in one region (178aa: 198–375). 6. Sequences and their GenBank accession numbers are: O. aries (AY547378), B. taurus (BF230134, AV603930, CB433957, BM480500), H. sapiens (NM_024117, BC002326), S. scrofa (CF791532, CF178115, BP459453, CF177341), M. muscus (BQ713136, BF781677, BU152256), R. norvegicus (CK476507, BE127132, BF553331, BU759329, AW141364), G. gallus (AF153127), X. laevis (BC043789), F. rubripes (Sequence accessible at ), D. melanogaster (AE003814), A. gambiae (XM_319576); S. pombe (AL136521, NP_594703, CAB66311); N. crassa (XP_322410). Figure 1 Alignment of Sin1 proteins from the fission yeast and sheep. The GAP program was used to align the two sequences. Black shading shows identical residues. Abbreviations: S. pombe, Schizosaccharomyces pombe (fission yeast. GenBank accession No. AL136521). O. aries, Ovis aries (sheep. GenBank accession No. AY547378). Figure 2 A phylogenetic tree for Sin1 primary sequences from various species. Sin1 polypeptide sequences were aligned by the program ClustalW, and the alignment output used by the program MEGA to generate a neighbor joining phylogenetic tree for the regions of alignment. GeneBank accession numbers for Sin1 sequences are listed in Table 1. Numbers beside branch points indicate the confidence levels for the relationship of the paired sequences as determined by bootstrap statistical analysis (1000 replicates). The lengths of the arms represent the extent of amino acid differences between the paired sequences, with the scale bar equivalent to 50 residues. Sin1 from the yeast species, S. cerevisiae and S. pombe which consist of 1172 aa and 665 aa, respectively, and also from the red bread mold, N. crassa (798 aa) are much longer than Sin1 from vertebrate and insect species, which are ~520 aa long. The regions of similarity among these three fungal proteins are confined entirely to the carboxyl termini of these molecules, although several gaps have to be introduced to align them. No similarities are detectable in the amino terminal extensions, which, in the case of S. cerevisiae , is 370 aa long. It is the carboxyl regions of the fungal proteins that can also be aligned with the Sin1 sequences from C. elegans , insects, and vertebrate species, including Ovis aries , the sheep (Fig. 1 ). A phylogenetic tree reconstructed from an alignment of amino acid sequences of Sin1 is shown in Fig. 2 . As anticipated, the sequences from the three fungi, C. elegans , the two insect species, and vertebrate species fell into distinct branches of the tree. The sequences for the mammalian species were tightly clustered, with identities ranging from 99% (humans and cattle) to 91.3% (sheep and rat) (Table 2 ). All the vertebrate cDNA encoded polypeptides of 522 aa (Table 2 ). There is considerable conservation of Sin1 from mammals to birds (~90%), amphibians (~85%), and fish (~80%) (Table 2 ). The insect sequences are rather longer than the ones from vertebrates, and several gaps have to be introduced to provide alignments (Fig. 3 , 4 , 5 ). Nevertheless, the insect amino acid sequences are approximately 33% identical to those of the mammals (Table 2 ). Five blocks of sequence (SCD I-V) are significantly more conserved than others when two insects, a fish, an amphibian, a bird and several mammals are compared (Figs. 3 , 4 , 5 & 6 ). Three of these regions are located towards the N-terminus and two additional regions towards the C-terminus. The most diverse region is located centrally. Figure 3 The alignment of Sin1 polypeptide sequences from insects and vertebrates. The ClustalW program was used to align all the protein sequences. Symbols (*, :, and .) show residues that are either identical(*), strongly similar (:), or weakly similar (.), respectively. Five Sin1 conserved domains (SCD) are highlighted as SCD I-V. The GenBank accession numbers for the sequences are: O. aries (AY547378), B. taurus (BF230134, AV603930, CB433957, BM480500), H. sapiens (NM_024117, BC002326), S. scrofa (CF791532, CF178115, BP459453, CF177341), M. musculus (BQ713136, BF781677, BU152256), R. norvegicus (CK476507, BE127132, BF553331, BU759329, AW141364), G. gallus (AF153127), X. laevis (BC043789), F. rubripes , D. melanogaster (AE003814), A. gambiae (XM_319576). Figure 4 The alignment of Sin1 polypeptide sequences from insects and vertebrates. The ClustalW program was used to align all the protein sequences. Symbols (*, :, and .) show residues that are either identical(*), strongly similar (:), or weakly similar (.), respectively. Five Sin1 conserved domains (SCD) are highlighted as SCD I-V. The GenBank accession numbers for the sequences are: O. aries (AY547378), B. taurus (BF230134, AV603930, CB433957, BM480500), H. sapiens (NM_024117, BC002326), S. scrofa (CF791532, CF178115, BP459453, CF177341), M. musculus (BQ713136, BF781677, BU152256), R. norvegicus (CK476507, BE127132, BF553331, BU759329, AW141364), G. gallus (AF153127), X. laevis (BC043789), F. rubripes , D. melanogaster (AE003814), A. gambiae (XM_319576). Figure 5 The alignment of Sin1 polypeptide sequences from insects and vertebrates. The ClustalW program was used to align all the protein sequences. Symbols (*, :, and .) show residues that are either identical(*), strongly similar (:), or weakly similar (.), respectively. Five Sin1 conserved domains (SCD) are highlighted as SCD I-V. The GenBank accession numbers for the sequences are: O. aries (AY547378), B. taurus (BF230134, AV603930, CB433957, BM480500), H. sapiens (NM_024117, BC002326), S. scrofa (CF791532, CF178115, BP459453, CF177341), M. musculus (BQ713136, BF781677, BU152256), R. norvegicus (CK476507, BE127132, BF553331, BU759329, AW141364), G. gallus (AF153127), X. laevis (BC043789), F. rubripes , D. melanogaster (AE003814), A. gambiae (XM_319576). Figure 6 The five highly conserved domains of Sin1 proteins. Sin1 primary sequences from various species were aligned by using the ClustalW program, and the five most conserved domains identified from the alignment in Fig. 3 and the sequences listed in Table 1. Conserved domains are shown as boxes with remaining regions as solid lines. SCD, S in1 c onserved d omain. Numbers beneath the species names are the lengths of the Sin1 proteins. Values in the boxes are the number of amino acid residues within a conserved domain. Numbers on the lines reflect the lengths of that region. The region of the greatest identity between these divergent insect and vertebrate sequences is an acidic region placed in conserved SCD III (Fig. 3 , 4 , 5 ). In mammals, this region is completely conserved and corresponds to residues L232-K267 (LHIAEDDGEVDTDFPPLDSNEPIHKFGFSTLALVEK; Figs. 3 , 4 , 5 ; Fig. 7 ). However, an analysis of this sequence reveals no known functional motifs and no strong similarity to sequences represented in other known proteins. Schroder et al. [ 5 ] have also noted this conserved sequence in their analyses of Sin1 sequences and have named it CRIM for c onserved r egion i n the m iddle. Figure 7 Alignment of Sin1 conserved domain III from various species. Sequences have been aligned by using the GCG PILEUP and GeneDoc programs. Degree of conservation is illustrated by intensity of shading (black, complete identity; light gray with black letters, complete identity across some but not all species; dark gray with white letters, high conservation but with conservative differences). The GenBank accession numbers for the sequences are: mm, M. musculus (BQ713136, BF781677, BU152256); rn, R. norvegicus (CK476507, BE127132, BF553331, BU759329, AW141364); bt, B. taurus (BF230134, AV603930, CB433957, BM480500); oa, O. aries (AY547378); ss, S. scrofa (CF791532, CF178115, BP459453, CF177341); hs, H. sapiens (NM_024117, BC002326); gg, G. gallus (AF153127); xl, X. laevis (BC043789); fr, F. rubripes ); dm, D. melanogaster (AE003814); ag, A. gambiae (XM_319576); ce, Caenorhabditis elegans (NM_064195); sp, Schizosaccharomyces pombe (AL136521, NP_594703, CAB66311); nc, Neurospora crassa (XP_322410). Sin1 from C. elegans retains the highly conserved 36 amino acid SCD I and the 127 amino acid Domain III (Fig. 6 & 7 ). SCD III is also retained in the fission yeast and the red bread mold. Vertebrates possess several unique sequences not present in insects and yeast, and, therefore, potentially implicated in the IFN signal transduction pathway including a carboxyl terminal region (KLSRRTSFSFQKDKK) immediately following the end of SCD V. Functional motifs in the Sin1 primary sequence When the ovine Sin1 sequence is scanned for functional motifs [ 13 ], the structure appears unusually barren. Two weak bipartite nuclear localization signals (NLS) [ 14 ] can be detected. One (residues 82–98, RRSNTAQRLERLRKERQ) is present in the SCDII domain, and the other (residues 503–519, RKLNRRTSFSFQKEKKS) is almost at the C-terminus within conserved domain V (Fig. 3 ). Nevertheless, data from the subcellular localization experiment showed that Sin1 is excluded from the nucleus when transfected in COS1 or L929 cells [ 4 ], suggesting these NLS are probably not functional. There are numerous motifs that are recognized as potential but weak sites for phosphorylation by either casein kinase II (CK2), protein kinase C, or protein kinase A (data not shown). None of the 17 CK2 sites, the 12 protein kinase C, or the 5 protein kinase A sites present in the ovine Sin1 primary sequence are conserved from mammals to fission yeast, although many are retained across all vertebrates. A weak site for myristylation (ovine residues 170–175, GTTATK; Figs. 1 & 3 , 4 , 5 ), and hence for membrane association, is retained in all the vertebrate species examined, but is absent in insects and yeast. In absence of any data on the functional significance of these sites, they will not be discussed further. Gene structure of Sin1 from various species The genomic sequence encompassing the transcribed region of the gene could be retrieved from the genome data bases for S. pombe , S. cerevisiae , C. elegans , D. melanogaster , A. gambiae , F. rubripes , R. norvegicus , M. Musculus , H. sapiens [ 15 , 16 ]. Sin1 exists as a single copy gene in all these species. For example, the human Sin1 gene is located on chromosome 9 (9q34.11-9q34.12) (data not shown) with the transcribed region composed of 11 exons and 10 introns and spanning a region of about 240 kb (Fig. 8 ). Exon 7 is spliced out of the shorter form of Sin1 [ 4 , 5 ]. The lack of exon 7 does not cause a frame shift because the intron phases of the two introns on both sides of exon 7 are identical (data not shown). Schroder et al. [ 5 ] have also demonstrated or predicted other minor splice variants for Sin1 in the human. The 11 exons account for only 0.9% of the gene sequence. It is, of course, unclear how many additional exons and introns are associated with the 5' UTR beyond the transcription start site(s), whose location has not been determined. Figure 8 A comparison of the Sin1 gene structure across species. The gene structure for all species was retrieved from the genome database of the species by using the BLASTn program to analyze the open reading frame of each Sin1 cDNA sequence. Only the regions of the gene containing the open reading frame are shown in the diagram. All sequences begin with start codons and end with stop codons. The numbers under species names are the protein length. Currently, the sheep and bovine genome sequences are not available, but it is likely that the Sin1 gene organization will be similar to that in the human. The current comparative synteny maps between human, sheep and cattle [ 17 - 19 ] predict that the Sin1 gene is located on sheep chromosome 3 (3p1.7-3p2.6) and bovine chromosome 11 (11q2.3-11q2.8), respectively. A comparative map for all the genes is shown in Fig. 8 . In fission yeast and insects, the Sin1 gene consists of a single exon. In worm, fish, rat, mice, and human, Sin1 has multiple exons. The exon/intron pattern, consisting of 11 exons, is observed in all vertebrates, including the two fish species (Fig. 8 ). It is noteworthy that although the genomic sequences of sheep and cattle are not available, the exon/intron pattern of their Sin1 genes is similar to that of other vertebrates based on the comparison between sheep or cattle Sin1 cDNA and human genomic sequence of Sin1 (data not shown). The lengths of these 11 exons are also remarkably conserved and fall within the normal range (50–200bp for most internal exons) (International Human Genome Sequencing Consortium, 2001). As expected, the sizes of the introns differ across species, and some are extremely long. Intron sizes generally decrease in the order human > mouse > rat > fish (Fig. 8 ). As expected, intron sizes were quite similar between rodents and human. The Sin1 gene from C. elegans is organized quite differently from that in mammals. It consists of 10 exons interrupted by nine relatively short introns. The region of the C. elegans gene that contains regions of similarity with the mammalian protein sequences consists of exon 1 (SCD I) and exon 5 (SCD III). As noted above and in Figure 8 , the Sin1 gene from insects and S. pombe is comprised of only a single exon. Discussion Sin1 is a little studied gene product of unclear function found in species ranging from mammals to fungi. Although the S. pombe gene product is longer than that of mammals, with an extension at its N-terminus, human Sin1 can rescue the stress sensitivity noted in the phenotype of a S. pombe strain that expressed a constitutively active form of RAS, indicating that function, as well as structure, has been conserved over hundreds of millions of years. Two facts should be considered when attempting to infer a role for Sin1 in vertebrates. The first, as discussed in the Background, is the known ability of type 1 IFN to activate MAPK/SAPK in mammalian cells. The second is the proven involvement of Sin1 in the yeast SAPK (Sty1/Spc1) pathway and its involvement in controlling transcription of stress-activated genes [ 2 ]. The present analysis was conducted in an attempt to gain more detailed information about Sin1 function from a phylogenetic analysis and comparison of Sin1 genes and gene products in different taxonomic groups. The Sin1 gene is remarkably divergent in both length and sequence identity within the fungi S. pombe , S. cerevisiae , and N. crassa , emphasizing the evolutionary distance between these three species. The regions of similarity are confined to the ~600 amino acid C-terminal regions of the three sequences (data not shown), and it is this region that is also conserved in insects and vertebrates (see Additional file: 1 & 2 ). This diversity in structure within the fungi is probably reflected in divergence of function. AVO1, the apparent Sin1 ortholog of S. cerevisiae , forms a membrane-associated complex with TOR2 and other protein components (AVO2, AVO3 and LST8), which control cell growth in response to nutrients [ 3 , 32 ]. Cells with deletion of AVO1 are unable to organize their actin cytoskeleton [ 3 ]. In contrast, the Sin1 ortholog of S. pombe is involved in a stress response signaling pathway by interacting with Sty1 [ 2 ]. A cross-species comparison of all the Sin1 sequences available, indicates five regions of greatest conservation, only one of which, a ~127 amino acid central region (SCD III), was easily defined in all taxa (Figs. 3 , 4 , 5 & 6 ). Even this region is poorly conserved in the budding yeast, S. cerevisiae , although certain landmark amino acids are retained (data not shown). Interestingly, Sin1 from insects and vertebrates, despite having only about 35% identity, are of similar length and possess the five regions of high identity. Conceivably, the SCD III domain is functionally essential in all the species, while SCDs I, II, IV, and V have evolved conserved function within the Metazoa. A not unreasonable assumption is that that Sin1 plays an evolutionarily conserved role in SAPK signaling across a broad range of taxa, including all metazoan and fungal species [ 5 ] but has assumed an additional function in vertebrates in mediating crosstalk with the IFN-signal transduction pathway. In vertebrates Sin1 falls into a class of highly conserved gene products. Its conservation is lower than that of two structural proteins, histone H3 and β-actin, but is comparable to that of CDK1 (Table 3 ). However, while CDK1 in yeast and insects retains considerable sequence identity with the vertebrate orthologs, much of the conservation of Sin1 is lost. It is tempting to speculate that Sin1 has been subjected to powerful evolutionary constraint that has limited its amino acid sequence divergence within vertebrates. It should be noted that our analyses cannot exclude the possibility that conservation of Sin1 among vertebrates reflects recent divergence of the sampled vertebrates relative to the other taxa examined. Once data become available, it will be instructive to compare Sin1 gene sequences from the invertebrate chordates (Tunicata and Cephalochordata) with those of the other metazoan taxa. Table 3 Comparison across species of the amino acid sequence conservation of Sin1 with some other conserved genes Yeast Drosophila Frog Chicken Mouse Cattle Human Histone H3 91.2% 98.5% 94.9% 95.6% 97.8% 98.5% 100% β-actin 90.4% 97.9% 99.5% 100% 100% 98.1% 100% CDK1 64.9% 71.7% 88.5% 93.3% 97.0% 98.7% 100% Sin1 25.3% 31.9% 85.6% 90.0% 96.9% 99.0% 100% Values for percentage identities were obtained by aligning amino acid sequences from various species with their human counterparts. CDK, cyclin-dependent kinase. The GenBank accession numbers for sequences are as follows. Human ( Homo sapiens ): histone H3 (AAH66884), β-actin (NP_001092), CDK1 (P06493), Sin1 (NM_024117, BC002326). Cattle ( Bos taurus ): histone H3 (P16105), β-actin (AAM98378), CDK1 (P48734), Sin1 (BF230134, AV603930, CB433957, BM480500). Mouse ( Mus musculus ): histone H3 (NP_062342), β-actin (NP_031419), CDK1 (NP_031685), Sin1 (BQ713136, BF781677, BU152256). Chicken ( Gallus gallus ): histone H3 (I50245), β-actin (NP_990849), CDK1 (P13863); Sin1 (AF153127). Frog ( Xenopus laevis ): histone H3 (P02302), β-actin (AAC27796), CDK1 (P35567, Sin1 (BC043789). Fly ( Drosophila melanogaster ): histone H3 (NP_724345), β-actin (NP_511052), CDK1 (NP_476797), Sin1 (AE003814). Yeast ( Schizosaccharomyces pombe ): histone H3 (NP_595567), β-actin (NP_595618), CDK1 (NP_595629), Sin1 (NP_014563). Sin1 was shown to be associated with the cytoplasmic domain of IFNAR2, a subunit of the type I IFN receptor [ 4 ]. Since insects appear to lack genes for type I IFN and their receptors (R. M. Roberts, unpublished observations), whereas vertebrates utilize this system primarily as an anti-viral response [ 20 - 22 ], it should be theoretically possible to define a sequence in silico unique to vertebrates but clearly absent in both D. melanogaster and A. gambiae that might account for the association of Sin1 with IFNAR2. Sin1 binds to the carboxyl end of the cytoplasmic domain of IFNAR2 via its own carboxyl 114 amino acids [ 4 ]. At least two candidate sequences exist in that part of Sin1. One is the rather basic carboxyl terminus (aa 510–522), another a HDYKHLYFESDA (aa 458–469) sequence, both of which are absent in the insect proteins (Figs. 3 , 4 , 5 ). Whether these sequences are participants in the interaction of Sin1 with IFNAR2 in vertebrates has not been examined experimentally. Of course, it is quite possible that insect Sin1 can bind vertebrate IFNAR2 or that amino acid substitutions elsewhere in the carboxyl end of the vertebrate sequence have evolved to promote the interaction. These possibilities have also not been tested. In this regard, IFNAR2, with which Sin1 interacts, has evolved much more rapidly than Sin1 itself. The sequence of human IFNAR2, for example, shows only about 58% and 29% identity to those of ovine and chicken IFNAR2, respectively [ 21 , 23 ], while orthologs have yet to be defined for IFNAR2 in frogs and fish, even though these animals are believed to have a functional IFN system, which includes the production of Type I IFN and downstream components in response to double stranded RNA [ 20 , 22 ]. Interestingly, the only highly conserved continuous sequence of chick and mammalian IFNAR2 within the Sin1 binding region is an acidic region (aa 493–515; human IFNAR2 numbering) at the very carboxyl terminus of the molecule ([ 23 ]; R.M. Roberts, unpublished observations). It seems possible that this conserved sequence provides the scaffold for Sin1 binding. As also observed by Schroder et al. [ 5 ], Sin1 is represented by a single gene in all species where it exists. In both insects and the two yeast species, the gene is intronless, while in C. elegans and in vertebrate species introns are present (Fig. 8 ). In budding yeast, only a small number (3.8%) of genes have introns [ 24 ], whereas in most other eukaryotes, including Drosophila , intronic sequences are a feature of the majority of genes and must be excised to produce a functional mRNA [ 25 ]. For D. melanogaster , for example, there is an average of 3 introns per gene [ 26 ]. These introns are short, averaging 240 bp in Drosophila [ 27 ]. Why the Sin1 genes are intronless in these species is unclear, but there is considerable evidence that retrotransposition occurs in yeast, Drosophila [ 28 ] and mammals [ 29 ]. In this process, reverse transcription of mRNA from a parental gene creates an intronless copy of the parental gene at a new position in the genome. If this mechanism created the Sin1 gene, a remnant or evolved version of the parental gene might be anticipated to exist, particularly if the transposition event occurred in recent evolutionary time [ 28 ]. It is unclear whether the intronless Sin1 gene in Drosophila resulted from such a retrotransposition event since there is not a detectable intronic copy elsewhere in the genome. The Sin1 gene from C. elegans has introns, but is organized very differently from that of vertebrates, where the intron/exon organization is highly conserved (Fig. 8 ). Unfortunately, the function of Sin1 is unknown. Its structural conservation from vertebrates to yeast [ 30 ] and its expression in most, if not all tissues of mammals [ 4 ] suggest a central, if elusive, role in life processes. Conclusions SAPK-interacting protein 1 (Sin1), a little-studied but widely expressed gene product, is encoded by a single gene in fungi, nematodes, insects, and all vertebrates analyzed and shows modest conservation of amino acid sequence that is consistent with some degree of conserved function in stress-activated signal transduction pathways. Sin1 is highly conserved in vertebrates where it has been implicated in linking interferon responses to the SAPK pathway. Methods Databases Sin1 genomic sequences from human, mouse, rat, fruit fly, mosquito, C. elegans , S. pombe , and S. cerevisiae , were retrieved from at NCBI Genome databases [ 18 ]. Sin1 cDNA sequences from human, mouse, rat, cattle and pig, and other Sin1 ESTs were retrieved from GenBank EST database after BLASTn analysis at NCBI [ 18 ]. For fish Sin1 genomic sequences, the incomplete puffer fish ( Fugu rubripes ) and zebrafish ( Danio rerio ) genome databases at the Ensembl site [ 16 ] were used. The budding yeast ( Saccharomyces cerevisiae ) ORF (open reading frame) database [ 33 ] was used to retrieve budding yeast Sin1. Software programs used to analyze sequences Pairwise global sequence alignment was performed by using either the BESTFIT or the GAP program from GCG (Madison, WI). Multiple global sequence alignment was performed by using either the PILEUP program (GCG, Madison, WI) and GeneDoc [ 34 ] or ClustalW program [ 35 ]. The phylogenetic tree for Sin1 was generated by using the ClustalW program and the MEGA program [ 36 ]. Motif search was performed by using the ScanProsite program [ 13 ]. Methods for obtaining Sin1 sequences from various species Fission yeast ( Schizosaccharomyces pombe ) and chicken ( Gallus gallus ): The two Sin1 sequences were published by Wilkinson et al. [ 2 ]. Budding yeast ( Saccharomyces cerevisiae ): The BLASTp program was used to search the budding yeast ORF database for any protein sequence that had significant similarity to the fission yeast Sin1 protein. The obtained budding yeast Sin1 protein sequence had a GenBank link where its cDNA was available. The cDNA sequence was used to analyze its genomic structure at the NCBI yeast genome site. Red bread mold ( Neurospora crassa ): Sin1 protein was retrieved from the Neurospora crossa protein data base by searching (BLASTp) with the budding yeast Sin1 protein. Worm ( Caenorhabditis elegans ): The Sin1 protein sequence was obtained from the C. elegans protein database by searching with ovine Sin1 protein. The cDNA sequence was then obtained from the GenBank link and used to determine the structure of the Sin1 gene. Fly ( Drosophila melanogaster ): The fruit fly Sin1 protein sequence was retrieved from the D. melanogaster protein database as above. The cDNA sequence was obtained from the GenBank link. Unexpectedly, querying the Drosophila genomic sequence with the C. elegans Sin1 sequence and vice-versa failed to yield a match in either case. Mosquito ( Anopheles gambiae ): The mosquito Sin1 protein sequence was retrieved from the Anopheles gambiae str. PEST protein database as above. The cDNA sequence was then obtained from the GenBank link. Puffer fish ( Fugu rubripes ) and Zebrafish ( Danio rerio ): Both Fugu rubripes and Danio rerio genome databases, which are accessible at two websites, NCBI and ENSEMBL, were queried with Sin1 cDNA sequences from sheep, chicken, and frog. For both species, only the Ensembl site provided the complete genomic sequence. Although the Fugu rubripes genome sequence is incomplete, the exons of Sin1 cDNA can be retrieved and successfully assembled into the full length structure by alignment with other Sin1 cDNA and gene sequences. No GenBank entry was available for the Fugu rubripes Sin1 gene. When a similar method was used to retrieve the Zebrafish Sin1 cDNA sequence, the full length sequence could not be obtained because the region (~20 kb) covering one exon (exon 4) was incomplete. Therefore, the fish Sin1 protein sequence used here is from Fugu rubripes . Frog ( Xenopus lavis ): The full-length cDNA sequence of Sin1 reported here was from African clawed frog, and was obtained by blasting the Xenopus EST database [ 37 ] with the chicken Sin1 sequence. The protein sequence was deduced from this cDNA sequence. Mouse ( Mus musculus ): The mouse Sin1 cDNA sequence was obtained by editing several ESTs, after performing a BLASTn search of the Mus Musculus EST database with the ovine Sin1 cDNA sequence. Searching the mouse genome database with the mouse Sin1 cDNA coding region then allowed the gene, down stream of its transcription start site to be located and its structure to be inferred. Rat ( Rattus norvegicus ): The rat Sin1 cDNA sequence was retrieved from several overlapping ESTs, which were obtained by searching the Rattus norvegicus EST database with the ovine Sin1 cDNA sequence. The coding region of the rat Sin1 cDNA was then used to search the rat genome database at the NCBI website for the genomic structure of the gene. Cattle ( Bos taurus ): The full length bovine Sin1 cDNA sequence was obtained from overlapping ESTs, which were obtained by searching the NCBI EST database with the ovine Sin1 cDNA sequence. Pig ( Sus scrofa ): The swine Sin1 cDNA sequence was obtained as above by searching the NCBI EST database with the ovine Sin1 cDNA sequence. Human ( Homo sapiens ): The sequence published by Colicelli et al. [ 31 ] was confirmed by performing a BLASTn search on human EST data bases with the ovine Sin1 cDNA sequence. Since the previously published sequence was not full-length, an additional human Sin1 EST (GenBank Acc. No. BC002326) was used to assembly the full length Sin1 cDNA sequence. The location of the gene and its structure downstream of its transcription start site were determined by searching the full human genome database with the Sin1 open reading frame. Sheep ( Ovis aries ): The sheep Sin1 cDNA sequence was cloned from a sheep endometrial cDNA library in a yeast two-hybrid screen [ 4 ]. GenBank accession numbers are summarized in Table 1 . Authors' contributions SW carried out the majority of the computational analyses under the direction of RMR, and wrote the first draft of the manuscript. RMR conceived of the study and participated in its design and coordination. All authors read and approved the final manuscript. Supplementary Material Additional file 1 Alignment of Sin1 proteins from the fission yeast and the budding yeast. The Bestfit program was used to align the two sequences. Black shading shows identical residues. A conserved region (SCD III; see Fig. 4) is highlighted by a line above the sequence, and appears not so well conserved in the budding yeast as in other species. Abbreviations: S. pombe, Schizosaccharomyces pombe (fission yeast. GenBank accession No. AL136521). S. cerevisae, Saccharomyces cerevisae (budding yeast. GenBank accession No. NP_014563). Click here for file Additional file 2 Alignment of Sin1 proteins from the fission yeast and the red bread mold. The Bestfit program was used to align the two sequences. Black shading shows identical residues. Abbreviations: S. pombe, Schizosaccharomyces pombe (fission yeast. GenBank accession No. AL136521). N. crassa, Neurospora crassa (red bread mold. GenBank accession No. XP_322410). Click here for file

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549548.xml

549206

Comparative analysis of protein coding sequences from human, mouse and the domesticated pig

Background The availability of abundant sequence data from key model organisms has made large scale studies of molecular evolution an exciting possibility. Here we use full length cDNA alignments comprising more than 700,000 nucleotides from human, mouse, pig and the Japanese pufferfish Fugu rubrices in order to investigate 1) the relationships between three major lineages of mammals: rodents, artiodactyls and primates, and 2) the rate of evolution and the occurrence of positive Darwinian selection using codon based models of sequence evolution. Results We provide evidence that the evolutionary splits among primates, rodents and artiodactyls happened shortly after each other, with most gene trees favouring a topology with rodents as outgroup to primates and artiodactyls. Using an unrooted topology of the three mammalian species we show that since their diversification, the pig and mouse lineages have on average experienced 1.44 and 2.86 times as many synonymous substitutions as humans, respectively, whereas the rates of non-synonymous substitutions are more similar. The analysis shows the highest average dN/dS ratio in the human lineage, followed by the pig and then the mouse lineages. Using codon based models we detect signals of positive Darwinian selection in approximately 5.3%, 4.9% and 6.0% of the genes on the human, pig and mouse lineages respectively. Approximately 16.8% of all the genes studied here are not currently annotated as functional genes in humans. Our analyses indicate that a large fraction of these genes may have lost their function quite recently or may still be functional genes in some or all of the three mammalian species. Conclusions We present a comparative analysis of protein coding genes from three major mammalian lineages. Our study demonstrates the usefulness of codon-based likelihood models in detecting selection and it illustrates the value of sequencing organisms at different phylogenetic distances for comparative studies.

Background Large scale sequencing projects of many different species allow us to investigate phylogenetic issues in much more detail and to identify whether certain genes have had an extraordinary evolution in one or more species and thus gain insight into the actions of natural selection. Despite the sequencing of an increasing number of mammalian genomes and the implementation of more sophisticated evolutionary models using maximum likelihood and Bayesian methodology, the branching order within the mammalian phylum is still not completely resolved. The main reason for this uncertainty is that the diversification of these orders occurred over a short period of time, making the inference of branching order a difficult problem. One of the highly debated issues concerns the relative order of branching among primates, artiodactyls and rodents [ 1 - 9 ]. Here, the Japanese pufferfish Fugu rubrices is used as an outgroup to estimate the branching order of the three species relative to each other. Codon based models [ 10 , 11 ] allow for powerful analysis of protein coding nucleotide sequences. Evolutionary hypotheses may be tested using likelihood ratio tests between nested models. For an introduction to the practical use of these models see [ 12 ], for a more thorough review of the methodology see [ 13 ]. The parameter of primary interest is the ratio of nonsynonymous to synonymous substitutions (ω), also known as the dN/dS ratio. The dN/dS ratio measures the relative importance of evolutionary forces that have shaped a particular protein. A dN/dS ratio significantly larger than one strongly suggests that positive Darwinian selection has acted on the protein. Different extensions to the basic codon model exist, and these can be divided into three main categories: (1) Lineage-specific models that average ω over sites but differentiate between lineages [ 14 ]; (2) site-specific models that average ω over lineages but differentiate over sites [ 15 ]; (3) branch-site specific models that combine the two previous extensions by allowing ω to vary over sites in all background lineages, but allow for a different value of ω in one or more pre-specified lineages [ 16 ]. The models we use here and their relationships are shown in Table 1 . Numerous studies have shown the ability of the site-specific and the branch-site specific models to detect positive selection in cases where the branch-specific models did not, indicating that averaging over sites is generally a more serious problem than averaging over lineages and that in many cases using a branch-site specific model increases the power to detect positive selection [ 17 - 22 ]. In a recent study of cDNA trios of human, mouse and chimpanzee a codon based branch-site specific model was used to search for human genes that have undergone positive selection since our divergence from other primates [ 23 ]. Here, a similar search is done on a different phylogenetic level using a collection of porcine genes. While the study by Clark and colleagues concentrates on the divergence between humans and chimpanzees (branch a in Figure 1 ) our study searches for genes that have undergone positive selection since the divergence of primates, artiodactyls and rodents. Several recent studies have shown that some of the branch-site specific models under certain conditions might have a high false positive rate when used to detect positively selected sites [ 24 , 25 ]. This problem has recently been addressed by Yang and colleagues with the implementation of a new Bayes empirical Bayes (BEB) method for predicting positively selected sites. This new method is much better at avoiding false positives while still retaining a high sensitivity (Z. Yang, pers. comm.). Here we use the new and improved BEB version of the branch-site specific model originally presented in [ 23 ] to detect genes that may have been influenced by positive selection. Results The distribution of sequence lengths of the 1120 three-species alignments is shown in Figure 2 . Since the full length cDNAs were assembled from random ESTs, there is a bias towards assembling relatively short genes. Therefore the subset of genes used in this analysis is not a random sample from the pig genome. This decreases the power of our evolutionary tests, since short alignments have less power when testing for positive selection, but we do not anticipate any other systematic bias in our results. Mammalian phylogeny The relative branching order of the three mammalian species was investigated with the individual genes as well as with a concatenated super gene. Using the empirical amino acid substitution model of Whelan and Goldman [ 26 ] we maximized the likelihood under the three conflicting topologies shown in Figure 3a–c . In 123 of the 988 alignments all amino acids are identical in the three mammalian species giving us no information to discriminate between the three topologies. Of the remaining 865 alignments 245 favour topology A, while 440 and 180 favour topology B and topology C respectively. A concatenated super gene of all 988 alignments clearly favoured topology B over topology A, which again has a higher likelihood than topology C, consistent with the results from the individual gene comparisons (Table 2 .). We used the baseml program of PAML to compare the three topologies in a nucleotide based framework. Different nucleotide based substitution models were used to maximize the likelihood on the three topologies for each of the three codon positions separately. The results of using different models of nucleotide evolution were highly similar so here we only discuss the results obtained with the HKY85 model [ 27 ]. The results based on the third codon position shows that Fugu is too distantly related to the three mammals to be informative in placement of the root of the mammals (results not shown). The first and second codon positions do not show such saturation and should therefore be useful in comparing the three topologies. Consistent with the results based on the amino acid substitution model we see that topology B is favoured in most genes, followed by topology A and topology C, respectively. The actual numbers from the second codon position are 215, 386 and 179 in favour of topology A, topology B and topology C respectively and 208 alignments are uninformative. The corresponding numbers for the first codon position are 215, 545, 175 and 53 (Table 2 .). The internal branch is rather short in all cases. Therefore in the remaining analyses we treat the mouse, human, pig split as a trifurcation. Depending on which topology is actually the right one, the only bias introduced by treating the topology as a star tree, as shown in Figure 3d , is a minor overestimation of the branch length of the species that actually roots the other two. The rates of evolution The three-species alignments were used to estimate the synonymous and nonsynonymous substitution rates of the three branches under the free ratio model, see Table 3 . Figure 4a–f shows the distribution of the synonymous and nonsynonymous branch lengths for each gene in all three species. The synonymous rates are significantly different between the three species. The average synonymous substitution rate, estimated using the concatenated super gene, is approximately 2.86 times larger in mouse compared to pig, and approximately 1.44 times larger in pig than in human. The nonsynonymous rates are more similar among the three species. The corresponding values for the nonsynonymous rates are 2.08 and 1.17 respectively. Table 3 shows the mean, median and variance of both the synonymous and nonsynonymous rate distributions as well as the values obtained from the concatenated super gene. The average values from the individual genes are highly similar to the results obtained from the concatenated super gene. Positive Darwinian selection The dN/dS ratios on the three different lineages were estimated under the free ratio model (Figure 4g–i ). Most genes in all three species have an average dN/dS ratio very close to zero with the average dN/dS ratio higher in human than in pig, which again is higher than in the mouse lineage. The one ratio model averages over sites and lineages, which makes this an extremely conservative method of detecting positive selection. Only four of the 1120 three-species alignments have an average dN/dS ratio larger than one, see Table 4 , and of those only one is significantly larger than one (XM_165930). The free ratio model allows each lineage to have its own dN/dS ratio. This model has slightly more power than the one ratio model due to its ability to find lineage specific signals. The likelihood ratio test (LRT) of these two models should not be considered as a stringent test for positive selection, but more as a test for different selective forces among lineages. The LRT shows that 154 genes have significantly different dN/dS ratios among lineages at the 5% significance level, 73 at 1% and 41 at the 0.1% level of significance. Table 5 shows the 24 genes that have a dN/dS ratio larger than one in one or more lineages as well as the result from each gene of a LRT that tests whether the estimated value of ω is significantly larger than one. As with the one ratio model only one gene shows a result significantly larger than one. The gene is the same one as reported with the one ratio model (XM_165930) and the lineage with a dN/dS ratio significantly larger than one is the lineage leading to pig. Several studies have shown that averaging over sites is more conservative when searching for positive selection than is averaging over lineages. The branch-site specific model A and model B [ 16 ] were originally designed to search for genes where only a small fraction of codons in a specific foreground lineage has evolved under positive selection. Several studies have shown that the original models are prone to predicting false positives under certain conditions, and one should therefore be very careful drawing conclusions from studies based on those models. Here we use a new and improved version of a branch-site model developed for the analyses of human, chimpanzee and mouse gene trios [ 23 ]. The new model we use here is implemented in PAML v. 3.14 and uses the new and improved Bayes empirical Bayes approach to predict which sites have evolved under positive selection in the foreground lineage. Likelihood ratio tests were done separately with human, pig and mouse as the predefined foreground lineage. The LRT when contrasting the neutral model with the branch-site model has two degrees of freedom. By using the human lineage as foreground lineage we find 288 genes that show signals of positive selection (dN/dS in the foreground lineage is larger than one). In 58 of those genes the branch-site model fits the data significantly better than the neutral model at the 5% significance level. We find 34 and 15 genes at the 0.01 and 0.001 levels of significance respectively. The corresponding numbers of genes using pig as foreground lineage are 314, 55(0.05), 23(0.01) and 5(0.001). Using mouse as foreground lineage results in 352, 67(0.05), 25(0.01) and 4(0.001). The genes found to be under positive selection in any of the three species with a LRT significance level of 0.001 are shown in Table 6 . The molecular function of the genes predicted to be under positive selection was determined using the Panther server [ 28 ] and the NCBI server using the newest build of the human genome. Both annotation servers are updated on a regular basis when new information becomes available. During the course of this study the annotation of several genes changed. Of our 1120 alignments 188 are currently not annotated as functional genes indicating that they might possibly be pseudogenes in human; see the Discussion for more details on this subject. The proportion of genes that we report to have undergone positive selection in the human lineage at the 5% level of significance can therefore be viewed as either 58/1120 ~5.2% or 43/931 ~4.6%, indicating that possible pseudogenes are only slightly overrepresented in the genes predicted to have undergone adaptive evolution. The genes predicted to have been under positive selection in the pig and mouse lineage show a similar trend. Several different models have been developed that allow for heterogeneity of ω over sites in an alignment. We used the M4 model [ 15 ] which allows each codon to fall into one of 5 categories corresponding to ω equal to 0, 1/3, 2/3, 1 and 3. The first category represents the fraction of codons that have evolved under strong purifying selection allowing no nonsynonymous changes to occur. The next two categories represent different intensities of purifying selection. The category with ω = 1 represents neutrally evolving sites, while the last category with ω = 3 represents codons that have evolved under positive selection. The results of this analysis on the concatenated super gene can be seen in Table 7 . Only 1.6 % of all codons appear to have evolved under positive selection, and approximately 69 % have been under strong functional constraints. Codon usage bias The concatenated super gene was also used to investigate the patterns of codon usage in the three species; the results of this investigation are summarized in Table 8 . A test for equal codon distributions is rejected in all three pair wise comparisons (P < 0.0001, 60 d.f.). Using nucleotide frequencies to estimate the codon equilibrium frequencies fits the data poorly, so does the equal frequency model (Table 9 ). For a description of the codon equilibrium frequency models, see the Methods. The F3 × 4 model was extended with one extra parameter that accounts for CpG avoidance at the second and third codon position. Since all changes in the second position of a codon are nonsynonymous, the frequency of NCG codons is expected to be lower than under the F3 × 4 model. The extra parameter introduced improves the log likelihood by approximately 1236 units (~44%). This can be compared to the approximately 321 units per extra parameter introduced when going from the F3 × 4 model to the codon table model. When analysing the super gene it is still better to use the actual codon frequencies, but with individual genes the number of codons can sometimes be so small that the use of actual codon counts can be problematic. We also implemented a similar model that incorporated the avoidance of CG in first and second position by introducing an additional parameter but this does not improve the fit of the model significantly (results not shown). This is probably caused by the fact that all four codons with CG in the first and second position code for the same amino acid, Arginine. Arginine has six different codons and the two codons without a CG pair (AGA and AGG) are generally favoured over the other four (Table 8 ), but this tendency is apparently accounted for when modelling nucleotide frequencies at the three codon positions, so here we only present the model that accounts for CpG avoidance at the second and third codon position. Table 9 shows that the choice of codon equilibrium frequency model has detectable effects on the parameter estimates. Most striking is the apparent overestimation of the transition/transversion ratio and the dN/dS ratio when the model is less parameter-rich. Discussion The phylogeny of the early mammalian radiation has been extensively debated over the last two decades. The classical view based on fossil evidence states that all major orders of placental mammals first appear right after the Cretaceous-Tertiary (KT) boundary approximately 65 million years ago [ 29 ]. This sudden appearance of all major placental orders is known as the mammalian radiation. With the use of molecular data this late radiation has been challenged and it is now widely accepted that the radiation of the placental orders probably occurred many million years before the KT boundary [ 29 - 31 ]. Molecular data have also been used to investigate the relative branching orders of many of the larger clades of placental mammals [ 1 - 7 , 9 , 30 ]. One of the issues that have been debated extensively is the placement of Rodentia in the placental tree. Some studies favour a basal placement of the rodents [ 1 , 3 - 5 , 32 , 33 ] while other studies favour a sister relationship between primates and rodents [ 6 - 8 ]. Recently strong evidence based on insertions, deletions and ancient transposable elements in favour of a sister relationship of primates and rodents has been reported [ 2 , 34 ]. The incongruence of single gene phylogenies was investigated in a recent study of eight yeast species [ 35 ]. The phylogeny commonly believed to be correct is completely resolved when concatenating 20 or more randomly chosen genes to form a super gene. A concatenated multi gene approach was also shown to resolve single gene incongruences in a recent study on green algae [ 36 ]. Here we use 988 full cDNA alignments comprising 672,918 nucleotides to investigate the branching order of the three mammalian species. We present results based on both single gene phylogenies and a concatenated super gene. All genes including the concatenated super gene were analysed with both nucleotide and amino acid based substitution models. All methods favour a primate-artiodactyls clade with rodents as an outgroup but with a relatively short internal mammalian branch, indicating that the mammalian radiation happened within a short period of time. The different methods used in this study have very different assumptions but they all show the same general results. The HKY85 model takes into account differences in nucleotide frequencies and transition/transversion biases and allows for differences in substitution rates among the lineages. However, it is still possible that complexities unaccounted for such as non-stationarity and irreversibility of the substitution process have created biases that lead to long-branch attraction of Fugu and Mouse and an erroneous conclusion. Furthermore, the incongruence between our analysis and many recent studies is also affected by the following. (1) The choice of outgroup; bony fishes are believed to have diverged approximately 450 million years ago [ 31 ], making saturation effects in synonymous sites a real problem. We are therefore forced to only consider nonsynonymous sites or amino acid replacements in the phylogenetic analyses. The recently completed genome sequence of the chicken ( Gallus gallus ) shows that the average value of dS between human and chicken genes is approximately 1.66 [ 37 ], which indicates that many genes may still be too distantly related for synonymous sites to avoid problems with saturation. A marsupial species would provide a much better outgroup when available [ 3 , 32 ]. (2) Taxon sampling; by only using three species the variance of the parameter estimates can be quite high and the power to discriminate between two conflicting topologies quite low. The sequencing of more species will lessen this problem. (3) Overly simplistic evolutionary models; here we use only nucleotide and amino acid based models. If a more closely related outgroup was available the use of more complex codon based models could be beneficial in resolving the apparent conflict. Several extensions have been made to the codon models during the past few years. One obvious extension to the codon models is a model that incorporates CG avoidance within and over codon boundaries. This will clearly improve the fit of the data to the model and therefore give more accurate parameter estimates. Including context dependencies over codon boundaries and information about protein structure have also been shown to increase the fit of the models to protein coding data and therefore should result in better parameter estimates [ 38 , 39 ]. (4) Gene trees and species trees can be different; the split between the three groups probably occurred within a very short period of time, allowing for the possibility that different genes actually have different phylogenies due to ancient polymorphisms at the time of the speciation. Using even larger number of genes and a sufficiently sophisticated model should lessen this problem [ 35 , 36 ]. The rate of synonymous substitution was estimated to be almost three times higher in rodents than in other mammals, in agreement with previous investigations that also showed an elevated rate in rodents [ 40 - 42 ]. This has historically often been explained by a generation time effect. Species that have short generation times experience more generations in the time span we consider and consequently they will experience more neutral substitutions over time. The fact that the pig, which has a generation time intermediate between mouse and humans, has an intermediate rate of synonymous substitutions, seems to agree with this theory. For a more thorough discussion of the generation time hypothesis in mammals see [ 43 ]. The nearly neutral theory of molecular evolution predicts that the generation time effect should be smaller for non-synonymous substitutions [ 42 , 44 , 45 ]. The simple argument is that animals with short generation times such as rodents often have a very large effective population size. In a population with a large effective population size slightly deleterious mutations will be removed from the gene pool more effectively than in a population with a small effective population size, where genetic drift will reduce the efficiency of natural selection. Figure 4g–h shows the distribution of the dN/dS ratio in the three lineages. The average dN/dS ratio is highest in humans suggesting a small effective population size, while it is smallest in mouse suggesting a larger effective population size. Previous studies of the occurrence of positive selection based on pair wise comparisons have revealed a very low occurrence of positive selection. In a study of 3595 alignments only 17 genes showed evidence of positive selection [ 46 ]. The branch specific models used here only find one gene where the dN/dS ratio is significantly larger than one. The gene reported is XM_165930. XM_165930 was originally annotated as being similar to cold shock domain protein A, but it has recently been removed from Genbank as a result of standard genome annotation processes. Codon based branch-site models similar to the ones used here were used in a paper based on a three way comparison among chimpanzees, humans and mice [ 23 ]. They report that approximately 1.6 % of all the genes studied have been undergoing positive selection in the lineage leading to modern humans. Using a similar criterion our study indicates that approximately 3.0 % of the genes studied have been undergoing positive selection on the lineage leading to humans; the corresponding numbers for pig and mouse are 2.0 % and 2.2 % respectively. When comparing these two studies it is important to consider the following three things: (1) the relatively short average length of the genes studied here decreases the power of the models to detect positive selection; (2) the use of the new BEB method for detecting positively selected sites should reduce the number of false positives, making our estimates more conservative and more accurate; (3) our study deals with a completely different phylogenetic level, covering a much longer time span than the study by Clark and colleagues. The multiple testing and the small number of taxa used in a study like this imply that the results presented should not be taken as conclusive evidence for positive selection, but more as an approach to searching among the thousands of genes to look for genes that may have evolved in a biologically interesting manner. Comparative approaches such as the one we use here can only be a first step towards showing that positive Darwinian selection may be a key part in the evolution of many different gene families. Further experimental and computational analyses must then be used to investigate the suggested candidates more thoroughly. During the course of our investigation a large fraction of the genes were re-annotated as putative pseudogenes: 188/1120 ~16.8%. However, all these genes have uninterrupted reading frames in all three species; only a tiny fraction of all codons seems to have evolved in a neutral-like fashion (ω~1), and the distributions of the synonymous as well as the nonsynonymous rates of these putative pseudogenes are almost identical to the distributions of the remaining genes (results not shown). The only difference is a slight increase in the dN/dS ratio in the human lineage, which is actually due to a few genes that experience an unusually high dN/dS ratio. Omitting these genes from the analysis removes the observed differences completely. Thus, if all these genes are indeed pseudogenes in human, the loss of function must have occurred quite recently and they may not be pseudogenes in pig and mouse. Conclusions The collection of a large set of pig cDNA sequences has enabled us to study long term evolutionary trends in mammalian genes. Our results indicate that the codon models are able to detect evolutionary signals indicating adaptive evolution in several genes. Our phylogenetic investigation of the primate, rodent, artiodactyl split disagree with most recent findings in favouring a primate, artiodactyl clade with rodents as an outgroup. Our study indicates that several genes that are not classified as genes in the most recent human annotation might after all be real genes; or at least they have become pseudogenes very recently, and the orthologous genes in mouse and pig might still be functional. This shows the potential of comparative methods in identifying functional regions of the genome. Methods cDNA alignment Complete cDNA from the domesticated pig Sus scrofa was assembled at the Danish Institute of Agricultural Sciences (DIAS) from cDNA libraries from 100 different tissues constructed at DIAS and the Royal Veterinary and Agricultural University in the following way. Total RNA was purified from selected tissues using Rneasy (Qiagen) or Tri ReagentR and poly(A+) mRNA was selected using Oligotex (Qiagene) or PolyATract (Promega). Directional cloneable cDNA was synthezised from Poly(A+) mRNA using the cDNA Synthesis Kit (Stratagene) and was ligated into Eco RI/Xho I digested pTrueBlue (GenomicsOne) or pBluescript (Stratagene) followed by electrotransformation into E. coli XL1-Blue MRF' (Stratagene). 5'-EST sequencing was performed using standard protocols (Applied Biosystem). The sequences were trimmed to the longest open reading frame and the termination codons were removed. Homologues sequences from human, mouse and the Japanese pufferfish Fugu rubrices were obtained with the blastall program with default parameters; the E-score was set to 10 -8 . We constructed two different datasets, one with and one without Fugu rubrices . Individual alignments were made using ClustalW version 1.83 with default parameters [ 47 ]. We kept the pig reading frame intact in the alignments by removing any columns where the alignment gave rise to gaps in the pig sequence. Alignments that resulted in premature stop codons, or were shorter than 30 codons, were removed. We used the one ratio model to estimate the total branch length of the tree as well as the synonymous branch lengths. These distributions were used to detect peculiar genes where one or more sequences might not be a true orthologue, and all outliers were thereafter removed from the dataset. This analysis gave 1120 alignments of mouse, human and pig, and of these 988 also included Fugu. The 1120 original cDNAs from Sus scrofa have been deposited in Genbank with the following accession numbers: AY609387-AY610506. Phylogeny and rates of evolution Nine hundred and eighty-eight four-species alignments were concatenated into a super gene. The three topologies were compared using the super gene as well as each individual gene. Both nonsynonymous nucleotide substitutions and amino acid substitutions were investigated with PAML v. 3.14 [ 48 ]. The nonsynonymous substitutions were represented by the first and second codon positions of all codons, and the three different topologies were investigated with baseml using the HKY85[ 27 ] model (model = 4) of nucleotide substitutions. The likelihood was then maximized under the three different topologies using all the individual genes as well the concatenated super gene. The codeml program with the codons translated to amino acids (seqtype = 3) were also used to investigate the three topologies. We used different models of amino acid evolution to maximize the likelihood under the three topologies and since the results were highly similar we only present the results from the empirical method of Whelan and Goldman (model = 2, aaratefile = wag.dat)[ 26 ]. Using the 1120 three species alignments, the synonymous and nonsynonymous rates of evolution were estimated with the codeml program (seqtype = 1) using the free ratio model (model = 2) with the transition/transversion ratio estimated from the data (fix_kappa = 0). Investigation of selection The different tests for positive Darwinian selection are all based on extensions of the basic codon based likelihood model [ 11 ]. Likelihood ratio tests (LRTs) were used to compare nested models where one allows for positive selection and the other does not. The probability that a codon i substitutes into another codon j during the time interval t is determined by the rate matrix Q = ( q ij ) with entries for i ≠ j , with corresponding substitution probability matrix given by exp(Qt). Here π j is the equilibrium codon frequency of codon j , κ is the transition/transversion ratio and ω is the dN/dS ratio. All parameters are estimated independently for each gene. The star topology of the three species is used to estimate the branch lengths (τ human , τ pig , τ mouse ) for synonymous and non-synonymous substitutions. Positive selection was tested in two different ways. Test 1 averages over sites but differentiates among lineages. The LRT compares the free ratio model where all three lineages have a different value of ω estimated from the data with the one ratio model where all three lineages share a common value of ω [ 14 ]. We note that this test is more a test of variable dN/dS ratios among lineages than a test for positive selection. The free ratio model has three parameters for ω and the one ratio model only one. The LRT statistic is calculated as 2 times the differences in maximum log likelihood and is asymptotically distributed as a χ 2 distribution with 2 degrees of freedom. The genes found in one or more lineages evolving with a dN/dS ratio > 1 are compared to a nested model where the dN/dS ratio is fixed at 1 in the lineages shown to have a dN/dS ratio larger than one to see whether the result can be attributed to natural selection or just relaxation of selective pressures. Test 2 is based on a new and improved version of the branch-site method presented in [ 23 ]. We will refer to this model as model A. The LRT is based on a comparison of the neutral model and model A. The neutral model assumes two categories of sites, a proportion p 1 of sites where ω 1 are estimated from the data and is forced to lie between 0 and 1, and a proportion p 2 of neutrally evolving sites where ω 1 = 1 ( p 1 + p 2 = 1). Model A furthermore allows a pre-specified branch to have a proportion of sites that evolve with a different value of ω estimated from the data. This value cannot be smaller than 1. The LRT follows a χ 2 distribution with 2 degrees of freedom. If the value of ω in the foreground lineage is estimated to be equal to one the model collapses to the neutral model. PAML v. 3.14 [ 48 ] was used to estimate likelihood and parameters under each model. Codon equilibrium frequencies can be estimated from data using either simple proportions in the full data set (the CT model with 60 parameters), assuming equal frequencies (Fequal model), multiplying overall counts of nucleotide frequencies (F1 × 4 model, 3 parameters) or counts of nucleotide frequencies for each codon position (F3 × 4 model, 9 parameters). The codon table (CT) was used for analysis of the concatenated super gene and the F3 × 4 model was used on the individual genes. CpG Extension of the codon models A simple extension of the F3 × 4 codon equilibrium frequency model can incorporate CpG avoidance by adding an extra parameter that penalizes a C followed by a G in the second and third codon position. The new model is parameterised as follows Here π i 1 1 represents the frequency of nucleotide i 1 , at codon position 1, and ψ(0 < ψ < 1) is a CpG penalizing parameter. The scaling factor c ψ ensures that the codon frequencies sum to one. Authors' contributions FGJ carried out the analyses and was the primary writer of the text. AH and FGJ together implemented some of the analysis tools. FGJ, AH and MHS together developed the ideas and discussed the interpretation of the results. MF and CB gathered the EST data used. HHJ assembled the cDNAs and carried out Blast searches. All authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC549206.xml

548695

DNA repeat arrays in chicken and human genomes and the adaptive evolution of avian genome size

Background Birds have smaller average genome sizes than other tetrapod classes, and it has been proposed that a relatively low frequency of repeating DNA is one factor in reduction of avian genome sizes. Results DNA repeat arrays in the sequenced portion of the chicken ( Gallus gallus ) autosomes were quantified and compared with those in human autosomes. In the chicken 10.3% of the genome was occupied by DNA repeats, in contrast to 44.9% in human. In the chicken, the percentage of a chromosome occupied by repeats was positively correlated with chromosome length, but even the largest chicken chromosomes had repeat densities much lower than those in human, indicating that avoidance of repeats in the chicken is not confined to minichromosomes. When 294 simple sequence repeat types shared between chicken and human genomes were compared, mean repeat array length and maximum repeat array length were significantly lower in the chicken than in human. Conclusions The fact that the chicken simple sequence repeat arrays were consistently smaller than arrays of the same type in human is evidence that the reduction in repeat array length in the chicken has involved numerous independent evolutionary events. This implies that reduction of DNA repeats in birds is the result of adaptive evolution. Reduction of DNA repeats on minichromosomes may be an adaptation to permit chiasma formation and alignment of small chromosomes. However, the fact that repeat array lengths are consistently reduced on the largest chicken chromosomes supports the hypothesis that other selective factors are at work, presumably related to the reduction of cell size and consequent advantages for the energetic demands of flight.

Background Genomes sizes (as measured by the DNA mass per diploid nucleus) are smaller on average in birds than in other tetrapod classes, and genome sizes within the class Aves show less variation than those of other tetrapod classes [ 1 , 2 ]. It has been proposed that reduced genome size in birds represents an adaptation to the high rate of oxidative metabolism in birds, which results primarily from the demands of flight [ 1 - 4 ]. Cell size and nuclear genome mass are correlated in vertebrates, and cell sizes of birds are smaller than those of mammals [ 1 ]. Smaller cells are advantageous in an animal with a high rate of oxidative metabolism because a smaller cell has a greater surface area per volume of cytoplasm, thus facilitating gas exchange. An alternative to the hypothesis that the reduced genome size is adaptive is the hypothesis that it resulted from an event of genomic DNA loss that was fixed in the ancestor of all birds due to genetic drift. The fixation of even a deleterious mutation is possible if the population undergoes an extreme bottleneck [ 5 ]. Some authors have argued that such a bottleneck may have occurred in the ancestor of birds at the end of the Cretaceous period [ 6 ], although this conclusion is not consistent with recent molecular evidence placing the radiation of the avian orders well prior to that time [ 7 ]. In order to decide whether genome reduction in birds was adaptive or due to a random event, Hughes and Hughes [ 8 ] compared the lengths of corresponding introns of orthologous chicken ( Gallus gallus ) and human ( Homo sapiens ) genes. They found that corresponding introns were significantly shorter in chickens, indicating that numerous independent deletions have occurred in the introns of birds. These results support the hypothesis that genome size reduction in birds is adaptive, since it is unlikely that such a large number of independent deletion events were due to chance alone. Additional evidence in support of the adaptive hypothesis is provided by the observation that a secondary increase in genome size has occurred in avian lineages which have become flightless or have reduced flying ability [ 9 ]. It has been suggested that an important factor in genome size reduction in birds has been that birds have lower levels of repetitive DNA than other vertebrates [ 10 ]. Genomes of mammals and reptiles are estimated to consist of about 30–50% repeats, while those of birds have been estimated to consist of only 15–20% repeats [ 10 - 12 ]. In birds chromosomes are of two types: a minority of macrochomosomes (3–6 μm in length) and a larger number of microchromosomes (0.5–2.5 μm in length). In the chicken, there are six pairs of macrochromosomes, and thirty-three pairs of microchromosomes [ 13 ]. There is a high rate of chiasma formation on avian microchromosomes, and this may be an adaptation that ensures correct pairing of these chromosomes during meiosis and mitosis [ 14 ]. Burt [ 10 ] proposed that the avoidance of repeats in the avian genome may in turn be an adaptation that enhances the probability of chiasma formation between homologous microchromosomes. This hypothesis and the hypothesis that genome size reduction represents an adaptation to flight are not mutually exclusive, since both factors may be at work simultaneously. Consistent with Burt's hypothesis, Wicker et al. [ 15 ] reported that in the chicken genome the ratio of repeats to protein-coding genes is higher on macrochromosomes than on minochromosomes. The sequencing of a substantial portion of the chicken genome has made it possible to examine quantitatively the distribution of repeating sequences on different chromosomes in the genome. Here we compare the distribution of repeats on 28 sequenced autosomes of chicken with that on the 22 human autosomes in order to test the hypothesis that reduction in repeat density in the avian genome has occurred as a result of adaptive evolution. Results The characteristics of repeat arrays on the 28 sequenced chicken chromosomes are summarized in Table 1 . The number of repeat arrays varied from 319 on chromosome 16 to 283,761 on chromosome 1; and the percent of the chromosome occupied by repeats varied from 4.1% on chromosome 32 to 14.9% on chromosome 1. In spite of the considerable variation among chicken chromosomes with respect to the percent of the chromosome occupied by repeats, the overall percentage of the chicken genome occupied by repeats (10.3%) was less than one quarter the percentage of the human genome occupied by repeats (44.9%) (Table 2 ). Even the most repeat-rich chicken chromosome, chromosome 1, had a repeat density less than one third that of the human genome (Tables 1 and 2 ). The range of repeat array lengths was much narrower Table 1 DNA sequence repeats on the assembled portion of the chicken genome. Chromosome Chromosome length (bp) No. repeat arrays Total repeat length (bp) (% of sequence) 1 188,239,860 283,761 27,978,835 (14.9%) 2 147,590,765 214,512 19,430,497 (13.2%) 3 108,638,738 151,571 12,198,434 (11.2%) 4 90,634,903 121,663 8,905,732 (9.8%) 5 56,310,377 69,048 4,638,645 (8.2%) 6 33,893,787 38,873 2,468,824 (7.3%) 7 37,338,262 41,189 2,397,200 (6.4%) 8 30,024,636 33,974 2,086,343 (6.6%) 9 23,409,228 24,255 1,384,475 (5.9%) 10 20,909,726 19,914 1,075,555 (5.1%) 11 19,020054 20,514 1,095,858 (5.8%) 12 19,821,895 19,419 1,116,593 (5.6%) 13 17,279,963 16,894 1,015,160 (5.9%) 14 20,603,938 21,588 1,417,684 (6.9%) 15 12,438,626 11,830 640,595 (5.2%) 16 239,457 319 18,614 (7.8%) 17 10,632,206 9,508 554,602 (5.2%) 18 8,919,268 8,312 574,276 (6.4%) 19 9,463,882 8,635 491,763 (5.2%) 20 13,506,680 12,826 766,482 (5.7%) 21 6,202,554 6,001 359,040 (5.8%) 22 2,228,820 2,636 183,334 (8.2%) 23 5,666,127 4,932 234,823 (5.7%) 24 5,910,111 5,435 356,373 (6.0%) 26 4,255,270 3,385 188,003 (4.4%) 27 2,668,888 2,833 252,335 (9.5%) 28 4,731,479 5,183 446,256 (9.4%) 32 1,018,878 806 42,242 (4.1%) Table 2 Features of DNA sequence repeats on human and chicken autosomes. Human Chicken No. chromosomes analyzed 22 28 Total sequence length (bp) 2,864,255,932 901,598,378 No. repeat arrays 4,698,717 1,160,319 Total repeat length (bp) (% of sequence) 1,287,381,310 (44.9%) 92,440,122 (10.3%) Mean repeat array length (bp) [median] (range) 274.0 [188.0] (7–160,603) 79.7 [25.0] (6–7,096) Mean no. repeat arrays per chromosome [median] (range) 213,578 [219,247] (57,109–409,783) 41,440 [14,860] (319–283,761) In the chicken genome, there was a significant positive correlation (r = 0.847; P < 0.001) between chromosome length and the percentage of the chromosome occupied by repeats (% repeats)(Figure 1 ). The four largest chicken chromosomes (chromosomes 1–4) contributed strongly to this positive correlation. In the case of the four largest chromosomes, there was a clear linear relationship between chromosome length and % repeats (Figure 1 ). In the human genome, there was also a positive, but non-significant correlation (r = 0.412; n.s.) between chromosome length and % repeats (Figure 1 ). Figure 1 The percentage of the chromosome occupied by repeats (% repeats) as a function of chromosome length in human ( blue dots ) and chicken ( red dots ). As illustrated in Figure 1 , % repeats values for human chromosomes were considerably higher than those for chicken chromosomes, even when the chromosome length were similar. An analysis of covariance was used to compare % repeats between the two species, with chromosome length as a covariate. There was a significant difference between species (P < 0.001) and a significant effect of chromosome length (P < 0.001), but there was not a significant interaction between species and chromosome length. These results imply that there is a similar slope to the linear relationship between chromosome length and % repeats in the two species, but that the % repeats values for human are significantly higher than those for chicken. Comparison of summary statistics human and chicken genomes showed that both mean and median repeat array lengths were considerably greater in the former species than in the latter (Table 2 ). In order to provide a statistical test of this difference that was not biased by the presence of different array types in the two genomes, we conducted paired tests on the 294 simple sequence repeat types shared by the two genomes (Table 3 ). Both the mean array length and the maximum array length were significantly greater in human than in chicken (Table 3 ). By contrast, the minimum array length did not differ significantly between chicken and human (Table 3 ). In order to test whether these differences between the two species were due mainly to the influence of the smaller chicken chromosomes, we repeated the analysis using only the four largest chicken chromosomes (chromosomes 1–4). In this case also, both the mean array length and the maximum array length were again significantly greater in human than in chicken, while the minimum array length was not significantly different between species (Table 3 ). Table 3 Mean (± S.E.) of variables describing simple sequence repeat types shared between human and chicken. Human Chicken P (paired-sample t-test) All chicken chromosomes (294 repeat types): Mean array length (bp) 83.6 ± 2.8 58.1 ± 1.5 < 0.001 Minimum array length (bp) 22.1 ± 1.6 22.1 ± 0.7 n.s. Maximum array length (bp) 457.9 ± 25.0 193.5 ± 8.1 < 0.001 Chicken chromosomes 1–4 (286 repeat types): Mean array length (bp) 83.8 ± 2.9 56.4 ± 1.7 < 0.001 Minimum array length (bp) 21.7 ± 1.6 24.8 ± 1.1 n.s. Maximum array length (bp) 466.7 ± 25.5 153.3 ± 6.9 < 0.001 Discussion Tabulation of DNA repeat arrays in the assembled portion of the chicken autosomes showed the overall percentage of repeats to be 10.3%. This value is similar to, but slightly lower than, previously published estimates (about 15%) based on reassociation kinetics [ 11 , 12 ]. By contrast, a similar tabulation in the human autosomes showed the overall percentage of repeats to be 44.9%. Because the value for chicken is substantially lower than the mammalian value, the results support the hypothesis that a relative scarcity of repeating DNA is a major factor in causing the relatively compact size of the avian genome [ 15 ]. Moreover, when simple sequence repeat array types shared between chicken and human genomes were compared, mean repeat array length and maximum repeat array length were significantly lower in the chicken than in human. The fact that these differences occurred consistently in nearly 300 distinct array types is evidence that the reduction in repeat arrays in the chicken has involved numerous independent evolutionary events. Mutational changes to simple sequence repeat arrays typically involve slippage events that either decrease or increase the number of repeat units [ 16 ]. The fact that simple sequence repeat arrays are shorter in the chicken than in the human implies that mutational events increasing array length have been eliminated by selection in the chicken to a greater extent than in human. Such long arrays might have included some that were inherited from the ancestors of birds and others that arose due to mutational events within Aves. In either case, the evidence for numerous, independent events of elimination of long arrays implies that reduction of DNA repeat length and thus of overall genome size in birds has occurred as a result of adaptive evolution. There were substantial differences among chicken chromosomes with respect to the percentage of the chromosome occupied by repeats, and % repeats increased significantly as a function of chromosome length. This trend implies that the avian genome is characterized by an especially pronounced avoidance of longer repeats on the smaller chromosomes. This finding is consistent with the hypothesis of Burt [ 10 ] that the reduction of repeating DNA in avian genomes is adaptive in permitting chiasma formation and alignment of microchromosomes. However, even the largest chicken chromosomes had repeat densities much lower than human chromosomes of similar length (Figure 1 ). This implies that avoidance of repeats on microchromosomes cannot be the sole factor at work in repeat avoidance in avian genomes. This interpretation is further supported by the fact that mean repeat array length and maximum repeat array length of repeat types shared between chicken and human genomes were significantly lower on the largest four chicken chromosomes than in human. Thus, the largest chicken chromosomes, like the rest of the chicken genome, showed a pattern indicating adaptive reduction of repeat array length. Our results imply that some other selective factor besides the need for alignment of minichromosomes contributes to genome size reduction in birds. Together with previous evidence [ 9 ], the results are thus consistent with the hypothesis that genome size reduction in birds is adaptive in that it leads to reduction of cell size and thus is advantageous in view of the energetic demands of flight. Methods The chicken ( Gallus gallus ) genome assembly (May 2004 freeze, release 25.1b.1) was downloaded from Ensembl web site at . Only autosomes were used in the analyses; data were available for chromosomes 1 through 24, 26, 27, 28 and 32. We extracted Ensembl annotations of the features of repeat arrays (including repeat name, start and end positions on the chromosome, and orientation). The human genome assembly (May 2004 freeze, build 35 (hg17)) was downloaded via the UCSC Genome Browser . Repeat information based on the RepeatMasker annotations (repeat name, start and end positions on the chromosome and orientation) was extracted from the UCSC genome annotation database. Only autosomes (chromosomes 1 through 22) were used. For both chicken and human, repeats tallied included simple sequence repeats, class I elements, class II elements, low-complexity regions, and satellite regions. In addition, we compared between genomes a set of 294 simple sequence repeats which were present in both genomes; i.e., repeats of the same short nucleotide motif were present in both genomes. For these 294 repeat types, mean, minimum and maximum length of repeat arrays were compared in pairwise fashion between human and chicken. Authors' contributions HP gathered and summarized the data. ALH conducted statistical analyses and wrote the manuscript. Both authors read and approved the final manuscript.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC548695.xml

554096

Theories of schizophrenia: a genetic-inflammatory-vascular synthesis

Background Schizophrenia, a relatively common psychiatric syndrome, affects virtually all brain functions yet has eluded explanation for more than 100 years. Whether by developmental and/or degenerative processes, abnormalities of neurons and their synaptic connections have been the recent focus of attention. However, our inability to fathom the pathophysiology of schizophrenia forces us to challenge our theoretical models and beliefs. A search for a more satisfying model to explain aspects of schizophrenia uncovers clues pointing to genetically mediated CNS microvascular inflammatory disease. Discussion A vascular component to a theory of schizophrenia posits that the physiologic abnormalities leading to illness involve disruption of the exquisitely precise regulation of the delivery of energy and oxygen required for normal brain function. The theory further proposes that abnormalities of CNS metabolism arise because genetically modulated inflammatory reactions damage the microvascular system of the brain in reaction to environmental agents, including infections, hypoxia, and physical trauma. Damage may accumulate with repeated exposure to triggering agents resulting in exacerbation and deterioration, or healing with their removal. There are clear examples of genetic polymorphisms in inflammatory regulators leading to exaggerated inflammatory responses. There is also ample evidence that inflammatory vascular disease of the brain can lead to psychosis, often waxing and waning, and exhibiting a fluctuating course, as seen in schizophrenia. Disturbances of CNS blood flow have repeatedly been observed in people with schizophrenia using old and new technologies. To account for the myriad of behavioral and other curious findings in schizophrenia such as minor physical anomalies, or reported decreased rates of rheumatoid arthritis and highly visible nail fold capillaries, we would have to evoke a process that is systemic such as the vascular and immune/inflammatory systems. Summary A vascular-inflammatory theory of schizophrenia brings together environmental and genetic factors in a way that can explain the diversity of symptoms and outcomes observed. If these ideas are confirmed, they would lead in new directions for treatments or preventions by avoiding inducers of inflammation or by way of inflammatory modulating agents, thus preventing exaggerated inflammation and consequent triggering of a psychotic episode in genetically predisposed persons.

Background When the solution to a clinical or scientific puzzle eludes us for more than a century, as with schizophrenia (formerly dementia praecox), we need new ways of thinking about the problem [ 1 , 2 ]. Efforts to understand schizophrenia have focused on neurons and, especially, the role of presumed excess dopamine neurotransmission. We believe that genetic, environmental, and stochastic factors combine with epigenetic factors to create episodes of the illness [ 3 - 5 ]. Thus, the syndrome of schizophrenia is viewed as an endpoint in a dynamic process variously conceptualized as degenerative or developmental or alternating at different points in the process [ 6 - 10 ]. Degenerative models imply that after a period of normal development, the organism, or one of its parts, takes a wrongful turn in its trajectory and begins to malfunction. This describes the eventual outcome for all life forms and is a biological restatement of the second law of thermodynamics. Since degeneration is universal, stating that an illness is degenerative is not particularly helpful. What would be helpful is to determine when in the life course the degeneration begins and how the degeneration is initiated and proceeds. Answers to the "when?" and "how?" questions would then describe the degenerative process in developmental terms. Developmental models of schizophrenia implicate abnormalities of early brain development predisposing to future schizophrenia. The proponents of the model further argue that the perturbations of development are limited to the early times of development and are discontinuous. Without this qualifier, developmental models are indistinguishable from degenerative models where the degeneration commences early in the life span. The early abnormalities are not necessarily the cause of schizophrenia, but, instead, create a state of risk for a future episode of schizophrenia. That is, a diathesis or predisposition is not a disease. Consequently, there must be factors later in life that convert the vulnerability to an illness. These additional factors are presumed to damage development in such a way that a predisposition becomes actualized. To gain a complete understanding of the syndrome, we must again return to the question of " what happens?" Following this line of reasoning, the distinction between degenerative and developmental models blurs. In fact, a medical-behavioral condition can be both developmental and degenerative as exemplified by Down syndrome [ 11 - 13 ]. Individuals born with trisomy 21 exhibit a number of developmental anomalies including cardiac malformations, abnormal dermatoglyphics, skeletal changes, and muscular hypotonia, to name a few. As trisomy 21 infants mature, most exhibit degrees of mental retardation. By about age 50, these individuals invariably develop Alzheimer-like CNS degenerative changes that can be seen at autopsy [ 13 ]. Schizophrenia involves both developmental and degenerative features. From the time of Bleuler [ 14 ] and Kraepelin[ 15 ], "It is certain that many a schizophrenia can be traced back into the early years of the patient's lives..." [ 14 ] p. 252. The 'follow back' studies of schizophrenia support these views [ 16 ]. Likewise, prospective studies of children at high risk for schizophrenia report developmental anomalies in motor skills, cognition, and attention long before the onset of overt illness [ 17 - 19 ]. Overt psychotic symptoms for some individuals usually start in the late teenage years or early twenties, but the illness can start as early as middle childhood [ 20 ] and may, more rarely, start in old age [ 21 ] p 73]. The evidence suggesting early developmental perturbations in schizophrenia is compelling. At the same time, there certainly are examples of deterioration reminiscent of Kraepelin's suggestion for some people with schizophrenia. However, deterioration in clinical course may not indicate CNS deterioration. Instead, the decline could be a secondary consequence of an illness that disrupts education, economic achievement, and social functioning leading to a downward spiral in all aspects of adult life. Consistent with an early degenerative process, there are reports of declining cognitive function preceding onset of psychosis [ 22 ]. Proponents of neurodevelopmental models suggest that the premorbid cognitive abnormalities are developmental risk factors for future schizophrenia (c.f [ 23 ]) and argue that such abnormalities show little evidence of decline after onset [ 6 , 24 ]. Whether developmental or degenerative, the premorbid cognitive deficits seen in schizophrenia are also seen in other disorders [ 25 ] and lack specificity and sensitivity thus detracting from the concept that the cognitive abnormalities seen in schizophrenia are useful endophenotypes [ 26 ]. The strongest evidence for a neurodegenerative phenomenon comes from imaging studies showing progressive loss of brain volumes [ 27 - 29 ]. Neuropathological studies fail to find widespread classic signs of neurodegeneration such as gliosis though there are exceptions to this generalization [ 30 ]. Observations of abnormal dendritic arborization [ 31 , 32 ] are consistent with the neuroimaging evidence suggesting abnormal connectivity between brain regions [ 29 ]. As a cautionary note, most of the neuroimaging and neuropathology results are subject to confounds from the effects of medications and various other treatments, post-mortem intervals, possible effects of diet, smoking habits, as well as a myriad of other potential confounds associated with glucocorticoid mediated stress following chronic illness and associated life's limitations [ 33 , 34 ]. The symptoms of schizophrenia are highly variable. Within families (and thus presuming relative homogeneity of genetic and environmental factors) symptoms can vary widely over time, as illustrated by identical quadruplets concordant for schizophrenia [ 35 ]. Even within affected individuals, symptoms will wax and wane and may even remit [ 36 ] suggesting a life long process. The major behavioral symptoms of schizophrenia include alterations in cognition, memory, perception, thought (inferred from language), motor functions, and affect. People with schizophrenia may show abnormal dermatoglyphics and other minor physical anomalies [ 37 - 42 ]. Other oddities to be incorporated in a comprehensive explanation of schizophrenia include highly visible nail fold capillaries [ 43 , 44 ] and the rarity of rheumatoid arthritis among schizophrenic persons [ 45 ]. These physical characteristics suggest the need to look beyond the nervous system per se to have a comprehensive view of the illness. The fact that the schizophrenia syndrome, as currently defined, is relatively common provides important information about the frequency of causal factors. About 1% of the population will experience schizophrenia during the lifespan. Except for a few rare exceptions, this 1% risk is remarkably constant around the globe regardless of culture, geography, or ethnicity. Men and women are affected equally. These facts mean that the risk factors for schizophrenia must also be common and ubiquitous. Given that the concordance rate for schizophrenia in identical twins [ 46 ] is only about 50%, there must be at least two global risk-increasing categories for schizophrenia, i.e., something(s) genetic and something(s) environmental. Assuming these risk factors are independent of each other, the joint probability of acquiring both risk factors is the product of their population frequencies that, for schizophrenia, equals about .01. To make a simplifying assumption to allow easy calculations, let us say that the two risk factors are present with about equal frequency in the population. With this simplification, straightforward mathematics indicates that the individual frequencies of these factors are close to the square root of the population frequency of 1%. That would mean that about 10% of the population would encounter at least one risk factor. The math indicates that the greater the number of independent risk factors, the more common they are. [See [ 47 ] for further elaboration]. Our challenge is to develop a theory of schizophrenia that can plausibly explain an illness that affects all domains of behavior (thought, affect, motor performance, etc), that has elements of developmental perturbations early in life leaving clues such as minor physical abnormalities, and also has elements of degenerative changes. At the same time, the defect is so subtle that we can't find the cause(s) with our best modern technology. Furthermore, in spite of brain-wide dysfunctions, many individuals with schizophrenia remain sufficiently intact that, with good treatment and a bit of luck, can maintain jobs and function usefully in society. Thus, we need to find frequent and ubiquitous factors that can affect virtually all brain functions as well as creating somatic signs, but they operate in ways that leave these functions only slightly "off kilter" as compared to the complete disruption seen in strokes, or classical degenerative disorders such as Alzheimer, or as seen in Down syndrome where the behavioral pathology is apparent from earliest stages. As we try to explain schizophrenia, we must account for most all of the developmental and degenerative features of schizophrenia. To account for the panoply of signs and symptoms seen in schizophrenia, any complete theory of schizophrenia must include organism wide systems. In addition to the nervous system, the immune system and the vascular system are defensible candidates. Both are invoked in the following theory: Some schizophrenia psychoses are the result of damage to the micro-vascular system in the brain initiated by genetically influenced abnormal inflammatory processes acting in response to ubiquitous environmental factors that trigger inflammatory responses, including infection, trauma, or hypoxia. It is the relative infrequency of the vulnerable genotypes in the population [ 48 ] that results in only a small proportion developing overt psychosis. We wish to emphasize that our hypothesis specifically identifies the microvascular system as the critical site of inflammation. We postulate that the inflamed micro-vessels lose their coupling with astrocytes, leading to disrupted regulation of cerebral blood flow and damage to the blood brain barrier. These disruptions in homeostatic mechanisms then lead to abnormal signal processing. Our focus on inflammation of the vessels differentiates our hypothesis from models of widespread parenchymal inflammation such as seen in psychotic syndromes following, for example, encephalitis lethargica, or paraneoplastic syndromes. Many acute inflammatory disorders of the brain involve inflammation of both the parenchyma and the vasculature. By contrast, we are proposing a chronic, smoldering, inflammation of the vessels alone. And, finally, we distinguish our hypothesis from the theories of schizophrenia implicating direct parenchymal infection of the brain (c.f. [ 49 ]) and also differentiates our hypothesis from speculations about schizophrenia that invoke infectious agents altering DNA [ 50 ]. Many prior debates about inflammation in the brains of people with schizophrenia have focused on the presence of absence of gliosis (see [ 51 ] for review). The consensus opinion is that gliosis, though present in some cases, is not a consistent feature of the neuropathology of schizophrenia. However, as Harrison [ 51 ] points out, evaluating gliosis is fraught with a multitude of problems and is not a definitive indicator of degenerative/inflammatory changes in the brain. More recent efforts have demonstrated activation of microglia in the brains of some individuals with schizophrenia implying an ongoing immunopathological process in addition to what ever happened early in development [ 52 ]. Ongoing neurodegenerative processes are suggested by increased levels of S100B, a small calcium binding astrocytic protein that is involved in inducing apoptosis and modulating proinflammatory cytokines [ 53 - 55 ]. It is likely that the current clinical syndrome of schizophrenia is etiologically heterogeneous. We do not pretend to explain all (DSM or ICD) cases of syndromal schizophrenia. Instead, we put forward our hypothesis as an attempt to define a psychiatric syndrome in terms of a particular pathophysiology. Following this course may then help refine our nosology (see also section on 'specificity' below) and cause us to recalculate basics 'facts' such as prevalence rates. Discussion A primer on CNS blood supply Neurons derive their energy from oxygen and glucose delivered by the vascular system, plus lactate and glycogen derived from astroglia [ 56 ]. The combination of neurons, astroglia, and micro-vessels form a metabolic trio [ 56 ] whereby the glia extend processes interacting with neurons on the one hand and, on the other, form endplates interdigitated into capillary walls. Rather than being passive conduits, the CNS vascular system is the most precisely managed and the most complex fluid dynamic system known. Regulation of cerebral blood flow (CBF) is managed primarily by a coupling between astrocytic glial cells [ 56 - 59 ] and capillary endothelium [ 60 - 65 ]. Astrocytes sense local neuronal metabolic activity and adjust blood flow as needed. Cerebral vessels change caliber in response to vasoactive substances released by astrocytes activated by glutamate receptors [ 56 , 66 , 67 ]. Serotonin [ 68 ], acetylcholine [ 69 ] and dopamine [ 66 , 70 , 71 ] transmission between astrocytes and micro vessels also play roles. When the neuronal activation of discrete areas is sustained over longer periods, vasoactive substances stimulate angiogenesis resulting in increased capillary density [ 67 ] thus enhancing local neuronal circuitry. Conversely, decrease in capillary density is likely to reduce the functional capacity of brain areas so affected [ 67 ]. Consequently, capillary beds in the cortex are not distributed in uniform fashion [ 72 ]. There are close relationships among local neuronal activity, density of capillary bed, and the distribution of valve-like flow control structures [ 73 ]. Developmentally, the CNS vascular system originates from capillary endothelial cells that migrate into developing neuro-ectoderm under the influence of trophic factors such as vascular endothelial growth factor (VEGF) [ 74 ] and erythropoietin [ 75 ] both produced by astroglia. The developing micro-vasculature, although comprising only 0.1% of the entire brain, and operating under the influence of genetic directives, has a key role in the development, maintenance and repair of the brain [ 76 ]. In turn, VEGF has trophic effects on neurons and glial cells, and the activity of VEGF influenced angiogenesis is directly proportional to the high metabolic activity of neocortical development [ 77 ]. Thus, angiogenesis and neurogenesis occur simultaneously and synergistically [ 78 - 80 ]. In addition to formation of capillaries themselves, intricate anastomoses between micro-vessels further 'fine tune' the metabolic support of developing glia and neurons [ 81 ] The genetics of infectious & inflammatory diseases When infectious agents give rise to inflammatory vascular disease, the nature of the infectious agent may be less important that an individual's genetically influenced inflammatory response. The concept that infectious disease may have a genetic component is, of course, not new. Many agricultural geneticists make their livings by breeding disease resistance into both plants and animals [ 82 , 83 ]. One of the founders of behavioral genetics, Franz Kallmann [ 84 ], showed genetic factors influenced acquiring tuberculosis (DZ concordance = 26%, MZ concordance = 87%), an observation that was confirmed in modern times [ 85 , 86 ]. Many other infectious diseases appear to have genetic factors influencing susceptibility or resistance to the infection [ 87 - 97 ]. Mechanisms for genetically mediated responses to infection occur through genetic variations in immune mediators such as cytokines[ 96 ] and HLA factors [ 98 , 99 ]. Familial Mediterranean Fever (FMF) [ 100 , 101 ] provides a heuristic Mendelian example. The gene for FMF is located on the short arm of chromosome 16 and produces pyrin (marenostrin) that functions in a negative feed back loop to suppress inflammation. Absence of pyrin leads to exaggerated inflammatory responses. Vasculitis is one of the consequences [ 102 ]. Additionally, very high rates of rheumatic fever (RF) or rheumatic heart disease (RHD) are found in relatives of patients with FMF[ 103 ]. Having even one mutant gene appears to lead to immune hyperactivity to streptococcal antigens. We also know that antibody [ 104 ] production and cytokine activity [ 105 ] in RF patients is more marked than non-rheumatics. It is clear that genes influence the host's response to infection. A similar line of reasoning applies to other inducers of inflammation such as traumatic injury [ 106 ] or hypoxia [ 107 , 108 ]. Just as the CNS blood supply is highly regulated, the inflammatory systems in the brain require 'fine tuning.' Given the limited ability for adult brain to regenerate, and assuming there is little tissue to spare, it would make sense that the brain should be protected from overabundant inflammatory reactions [ 109 ]. Astrocytes play a key role in the expression of inflammatory cytokines, chemokines, and growth factors involving the modulation of gene expression for these factors [ 109 - 111 ]. Let us suppose that schizophrenia develops following an infection (or trauma or anoxia – the environmental contributors) but the host's response is determined by genetic factors regulating the nature and degree of inflammation. That infectious agents may be operative in schizophrenia is supported by several of lines of evidence. Summaries can be found in numerous sources [ 49 , 50 , 112 - 116 ]. The same concept applies to trauma [ 106 ] or anoxia [ 79 , 107 ] that may also stimulate inflammatory processes. Vascular disease and psychopathology The syndrome of schizophrenia is likely to be etiologically heterogeneous and a multitude of CNS disorders can give rise to schizophrenic-like psychoses [ 117 ]. The idea that CNS micro-vascular diseases, in particular, are factors in psychotic disorders is also an old idea [ 118 , 119 ] that deserves a second look in light of new perspectives offered by developments in the genetics of inflammatory diseases. There are many examples of psychoses resulting from micro-vascular CNS disease including lupus and Sjögren syndrome [ 120 ]. Neuroimaging and neurocognitive deficits in these disorders are similar to those seen in schizophrenia [ 121 ]. Psychoses associated with substance abuse are also associated with CNS vasculitis [ 122 ]. Furthermore, infectious agents such as syphilis [ 123 ] and rheumatic fever (RF – see below), lead to micro-vascular disorders of the CNS that are associated with psychiatric symptoms including psychoses. Thomas, et. al. [ 124 ] also demonstrated small vessel abnormalities in the depressed elderly. At the same time, there is growing interest in cytokines and other inflammatory agents in psychoses[ 125 ] as well as growing awareness that inflammatory reactions are modulated by neuropeptides [ 126 ]. Inflammatory processes often damage the precise regulation of cerebral blood flow. The wide spectrum of clinical conditions thought to be created, in part, by inflammatory CNS micro-vessel disease include Alzheimer disease where it is thought that inflammatory processed damage the micro-vascular endothelium causing insufficient blood flow leading to oxidative stress, a build up of amyloid, and eventual cell death [ 127 - 135 ]. Cerebral palsy is also conceptualized as an infectious-inflammatory-vascular disorder where the vascular lesion is complete thrombosis [ 136 ]. Neurotoxic effects of methamphetamine and cocaine appear to be due to induction of inflammatory genes in small vessel endothelial cells [ 122 , 137 ], thus explaining the vascular damage seen in amphetamine and cocaine abuse that was previously attributed to contaminants of injected drugs [ 122 , 138 - 140 ]. Returning to the early stages of life, we have seen that the development of the neurons and glia are intimately associated with, and dependent on, the parallel development of the CNS vasculature. If the stated theory is correct, and given the developmental perspective of schizophrenia ---early developmental perturbations of the CNS set the stage for later schizophrenia--- we would expect to find support for the idea that inflammatory events early in life affect CNS vascular function. Such is the case. Whether the early insults are traumatic, infectious, or hypoxic; inflammatory process are involved in the attempts to protect and repair by modulating angiogenesis [ 141 - 148 ]. Thus, the reports implicating pregnancy and birth complications (anoxia, trauma or maternal infections) in the development of some cases of schizophrenia [ 149 , 150 ] could all be mediated by the common pathway of inflammatory-vascular mechanisms. Individuals who's genes created perturbations in inflammatory-vascular regulation would continue to experience abnormalities of protection and repair in response to subsequent CNS insults. Over time, the accumulation of 'hits' could lead to brain dysfunction to the extent seen in psychoses. The greater the number and duration of 'hits,' the greater the risk for a deteriorating /degenerative course. That neuroleptics may alter the permeability of the blood brain barrier and modify immunoregulation in the CNS [ 151 ] strengthens the argument for early treatment as a strategy to prevent deterioration. Alterations of cerebral blood flow in schizophrenia Since the time of Seymour Kety's pioneering efforts [ 152 , 153 ], there has been interest in altered cerebral blood flow in people with schizophrenia. An in-depth review of this large literature is beyond the scope of this paper. The interested reader is referred to discussions of reduced anterior cerebral perfusion leading to the concept of 'hypofrontality' in schizophrenia [ 154 , 155 ] and to more recent reviews [ 156 - 158 ]. Bachneff's [ 159 ] review and theory about defects in regulation of CNS microvascular systems is particularly relevant. These reviews summarize a consistent body of evidence showing reduced cerebral blood flow in brains of people with schizophrenia especially to anterior regions. Flow deficits are seen in medication-naive new onset cases [ 160 , 161 ] and more established cases free of neuroleptics [ 162 ] suggesting that flow perturbations are neither the consequence of duration of illness nor treatment. Neuroleptics can alter cerebral blood flow [ 163 , 164 ] although the effects may be regionally and drug specific [ 165 , 166 ]. Decreased frontal flow is often associated with negative symptoms [ 167 , 168 ]. In addition to the frontal cortex, flow abnormalities in people with schizophrenia have been noted in the cingulate cortex [ 169 , 170 ], thalamus [ 171 ], basal ganglia [ 172 ], parietal cortex [ 167 , 170 ] and cerebellum [ 171 ]. Furthermore, in some instances, flow rates are increased [ 160 , 170 ]. Rather than a simple hypothesis of hypofrontality in schizophrenia, theorizing is evolving toward a concept of "dysfunctional circuits"[ 160 ] or "inefficient dynamic modulation" [ 173 ] of cerebral metabolism which is supported by other examples of abnormal modulation of cerebral blood flow in response to activation tasks [ 171 , 174 ]. Disturbances of blood flow in schizophrenia are well documented but are not limited to schizophrenia. Disturbed cerebral blood flow is also reported in obsessive compulsive disorder [ 175 ] and depression [ 176 , 177 ] as well as in Alzheimer disease (cited earlier). The usual interpretation is that alterations of blood flow arise as a consequence of abnormal neuronal metabolism. The theory proposed by this paper turns the causal arrow around to suggest that abnormalities of blood flow lead to altered neuronal-glial function that, in turn, leads to psychopathology. There has been scant direct visualization of the vascular system in schizophrenia, but at least one laboratory has found evidence of atypically simplified angioarchitecture and failure of normal arborization of small vessels [ 32 ]. Post- streptococcal behavioral syndromes as a model Post-streptococcal neuropsychiatric syndromes include Syndenham chorea, the PANDAS/obsessive compulsive syndrome, tics including Tourette syndrome, and possibly, ADHD [ 178 - 184 ]. Psychotic disorders are also implicated [ 183 , 185 ] and see citations below. Sydenham chorea is the best-known neuropsychiatric complication following streptococcal pharyngitis. The association of psychoses and Sydenham chorea as well as with RF even in the absence of chorea, was discussed in the 17 th and 18 th centuries starting with Sydenham himself (see [ 186 ]). The interest in psychoses associated with RF continued throughout the 1900's [ 187 - 197 ]. People with a history of Sydenham chorea and/or rheumatic fever are at high risk for developing psychopathology later in life [ 198 , 199 ] with a relative risk for schizophrenia as high as 8.9 in a 10 year follow-up of 29 Sydenham patients [ 200 ]. There is a suggestion that the family members of Sydenham patients are also at higher risk for psychosis [ 201 ]. During the 1940's-1960's when RF was still quite prevalent, people with psychoses appeared to have higher than expected rates of histories of RHD or RF)[ 195 , 202 , 203 ] or rheumatic chorea [ 204 ]. Psychotic patients with RHD more often had early (<age 19) onset, movement disorders, progressively insidious courses and poor long-term outcomes [ 203 ]. Preliminary data from a Minnesota study also finds increased rates of RHD in psychotic patients, a pattern of increased psychiatric hospitalization following an epidemic of RF, and a clinical course for "rheumatic psychoses" that disproportionately led to a severe and continuous decline in function [ 205 ]. Although schizophrenia-like psychoses were the most common psychopathology related to rheumatic syndromes, manic-depressive, involutional, and senile psychoses were also observed [ 183 , 197 ]. An inflammatory reaction of the CNS vascular endothelium (vasculitis) is a common denominator in the both acute and chronic cerebral consequences of rheumatic fever. [ 186 , 187 , 190 , 195 , 197 , 206 - 209 ]. The microvascular lesions suggest both an obliterating process likely due to micro-emboli from rheumatic cardiac valves and an inflammatory process involving irregular proliferative changes in the vascular endothelium, dilatation of the lymphatic spaces surrounding the capillaries suggesting increased permeability of the capillary endothelium, and inflammatory cell infiltrates. Disruption of the blood brain barrier suggested by the evidence of increased permeability of the small vessels could compromise the immunological protection of the brain leading to the formation of the anti-neuronal antibodies seen in post-streptococcal CNS syndromes. In parallel fashion, people with schizophrenia show evidence of altered blood brain barrier and consequent alterations in immunological markers [ 210 ] The post-strep psychopathologies provide a precedent for the hypothesis of this paper by demonstrating that an infectious process can trigger a series of inflammatory reactions that lead to a variety of somatic and psychiatric syndromes, including psychoses where vascular pathology is implicated. The pathogenicity of a strep infection is a function of the strain (genotype) of the bacterium and the genetically mediated inflammatory mechanisms of the host [ 211 ] and illustrates how a ubiquitous and often relatively benign environmental factor can create more serious sequelae in a limited number of genetically predisposed individuals-true genotype by environment interaction. Summary The ideas here are not completely new. Eugen Bleuler [ 14 ] remarked: "The fragility of the blood vessels which appears in many schizophrenics, both acute and chronic, seems to indicate a real vascular pathology (p.167)." We bring old ideas forward into the light of new understandings offered by molecular genetics and inflammatory diseases. Since the late 1800's there has been evidence of inflammatory neuro-vascular abnormalities in psychiatric illness that were initiated by infectious agents. CNS lues (syphilis) is the best-known example. This paper expands the concept to suggest that a variety of environmental insults (infection, trauma, anoxia) may follow a common final pathway to psychopathology by stimulating inflammatory processes that damage the capillary-glial-neuron triad as illustrated in Figure 1 . Abnormal behaviors develop as a result of disruptions in astroglial mediated coupling of cerebral blood flow to neuronal metabolic needs. These subtle disruptions are hard to find, as the microvasculature comprises only about 0.1% of the brain and are of a scale more appropriate for electron microscopy. None-the-less, the hemodynamic perturbations have sufficient impact to cause subtle but widespread disruption of the normally harmonious coordination of CNS function leading to a condition variously conceived as a "neurointegrative defect"[ 212 ], "synaptic slippage" [ 213 ], "abnormal signal transduction" [ 4 ], "inefficient dynamic modulation" [ 173 ] or "synaptic destabilization" [ 214 ]. The ultimate impact would lead to psychopathology including psychoses as the vascular-glial-neuron triad is progressively damaged over time after repeated inflammatory episodes. The resultant failure to regulate the delivery of oxygen and energy adequately would lead to oxidative stress [ 215 - 217 ]. Oxidative stress, in turn, can further damage the microvasculature and the blood brain barrier [ 218 - 220 ]. The astroglial-capillary partnership that protects the integrity of the blood brain barrier would be compromised, thus exposing neural tissue to damage from immunological attack [ 221 ]. Known precedents of such processes are found in the behavioral changes seen in CNS vascular inflammatory diseases such as lupus and the post-strep syndromes described above. This theory could explain how developmental events such as prenatal infections [ 150 , 222 ], and other birth and pregnancy complications [ 149 ] including anoxia [ 223 ] are linked to later schizophrenia – infection, trauma, or anoxia all stimulate inflammatory processes [ 224 ]. The data suggesting an (statistical) influence of season of birth [ 116 ] is also consistent with the hypothesis as infectious epidemics often follow seasonal patterns. Some of the minor physical anomalies such as unusual scalp hair patterns and dermatoglyphic changes are explained because the development of these phenomena are linked to each other [ 225 ], to the development of the central nervous system [ 226 ], and are developmentally modulated by the pleiotropic effects of the same substances that modulate brain vascular development (e.g., vascular endothelial growth factor/vascular permeability factor [ 227 ] and epidermal growth factor [ 228 ]). The waxing and waning of symptoms would correspond to waxing and waning of inflammations as individuals are exposed, recover, and then re-exposed in conjunction with other physiological and hormonal influences, as seen in lupus [ 229 ]. The nature and severity of symptoms would depend on where in the brain the inflammation takes place and this may be stochastic. As the micro- vascular system is everywhere in the brain, lesions could produce the variety of symptoms seen in schizophrenia including dysfunctions of thought, emotion, memory, motor skills and autonomic regulation. The developmental age of the individual will also make a difference. Inflammatory processes that alter angiogenesis during prenatal development will likely have more dire consequences than inflammatory reactions that start after CNS maturation although even the adult brain remains susceptible [ 230 ]. We have attempted to schematically illustrate this dynamic process in Figure 2 . This theory also captures many of the little oddities observed in schizophrenia. For example, the reported abnormalities of the nail fold capillary beds seen in some people with schizophrenia [ 44 ] are also seen in people with inflammatory disorders such as FMF [ 231 ] and rheumatoid arthritis [ 232 ]. Another oddity is the negative association between schizophrenia and rheumatoid arthritis [ 45 ]. There are parallels in the post-streptococcal syndromes where RF and acute post-streptococcal glomerulonephritis very rarely occur in the same patient [ 233 ]. Some strains of group-A-streptococci identified by their M-protein serotypes are rheumatogenic while others are nephritogenic [ 233 , 234 ]. Phage or phage-like elements inserted into the streptococcal DNA are a major source of variation between streptococcal strains and these elements determine pathogenicity [ 235 ]. Additionally, host variation in humoral and cellular immune response shape the outcome of infection[ 211 ] By analogy, individuals with vascular/CNS involvement following, for example, streptococcal infections may be systematically spared from joint involvement as a function of both the invading strain and the individuals susceptibilities. Alternatively, as postulated for Alzheimer disease (cited earlier) that is also less common in people treated for arthritis, the anti-inflammatory treatments for arthritis might reduce the risk of inflammatory brain disease. Another line of evidence compatible with this theory is the observation that genetic linkages for schizophrenia coincide with sites for glial growth factor cell regulators [ 214 ] and, as we have seen, the glia are key intermediaries of CNS inflammation and vascular regulation. More specifically, emerging data demonstrate associations between schizophrenia and genetic polymorphisms in regulators of inflammation such as tumor necrosis factor alpha genes [ 236 , 237 ] and interleukin-1 genes [ 238 ]. Another piece that fits into the puzzle is the fact that neuroleptics have inflammatory modulating properties [ 239 - 244 ] and neuroleptic treatment may be synergized by addition of anti-inflammatory drugs [ 245 ]. It may well be that the environmental components of psychiatric illness such as schizophrenia are relatively minor, ubiquitous, or chance events [ 246 , 247 ] that have the potential to stimulate the inflammatory systems. However, the nature of the insults may be less important than individuals' genetically influenced and idiosyncratic responses to the insults, similar to individuals with FMF who have an exaggerated inflammatory response. Thus, the genetic components of the inherited predisposition to mental illness may lie "upstream" in the immune system rather than in the CNS per se. The possibility that the environmental agents may be nearly universal (e.g. who has not had a strep throat or viral syndrome?), will mean that the prevalence of the etiological factor will be similar in control and experimental groups thus making it too easy to dismiss key environmental factors in null hypothesis designs [ 47 , 248 ]. Rather than focus on the environmental contributors that could be non-specific and ubiquitous, it will be more productive to look for genotypes that respond abnormally to triggers of inflammation and microvascular dysfunction (cf[ 48 ]). These individuals would be the ones who are at high risk for psychiatric illness. However, the inflammatory processes involve a cascade of steps involving many genes. But this, too, fits with the polygenic features of schizophrenia [ 249 ]. Identification of high-risk individuals, combined with such tools as immunizations or anti-inflammatory agents may promote prevention of much psychiatric morbidity. Already, the cytokine regulator and vascular growth factor erythropoietin is suggested as a possible neuroprotective factor in schizophrenia [ 250 ] Future directions The speculations about psychoses developing from vascular/inflammatory processes provide direction for future research across many domains. In addition to pursuing direct evidence of altered activities in inflammatory/immune systems in people with psychoses, the inflammatory/vascular theory has implications for epidemiology, genetics, neuroimaging and neuropathology. For the epidemiologist, the challenge will be to detect relatively small signals against a very noisy background. We hypothesize that the triggers for inflammation can be many and varied and are common factors in the environment. Imagine starting with the clinical syndrome of Sydenham chorea and comparing the rates of strep throat in those affected vs. comparison sample of people without Sydenham chorea. Null hypothesis testing with small sample sizes and nearly ubiquitous etiological agents are clearly not adequate. A second epidemiological challenge is to cast a broad enough net to capture the wide variety of possible contributing factors. Rather than taking a one by one approach to exploring the etiological contributions of, say, virus titers, anoxia, physical trauma, the epidemiologist should look for any and all. It would be predicted that individuals with multiple "hits" (e.g. in utero exposure to virus and low Apgar scores and childhood head trauma) would be at greater risk than those exposed to just one event. If in utero inflammatory processes are active in the genesis of schizophrenia we would also predict an increased rate of fetal deaths in families of schizophrenic probands. A third epidemiological opportunity lies in the search for non-psychiatric inflammatory-related disease or traits in people with psychosis. If something is askew in the inflammatory process in schizophrenia, the effects will show up in other parts of the body. Though requiring replication, the association of psychosis with hemolytic anemia in lupus [ 251 ] provides an illustrative example. In addition to rheumatoid arthritis, the associations of diabetes and cancer have been explored in schizophrenia; one of is exploring rheumatic heart disease [ 205 ]. Population-based health registries should be used in a search for co-morbid physical illness. For geneticists, the proposed theory obviously points to linkage/association studies using inflammation genes; a few examples were cited previously [ 236 - 238 ]. A simple step with extant data might start with a meta analysis defining chromosomal "hot spots" for linkage with schizophrenia and search the gnome maps for immune regulators at these sites as Moises, et al [ 214 ] have done for glial growth regulators. Family, twin, and adoption methodologies can all be applied to the issue of co-morbid or co-segregating physical conditions. The inflammatory/vascular theory has much to suggest to neuroimaging research especially in the realm of reinterpreting regional perturbations in metabolic activity as primary disturbances of flow regulation rather than intrinsic neuronal metabolic abnormalities. It would be interesting to assess the impact of vasoactive compounds and inflammatory modulators on neuroimaging studies of regional blood flow. Likewise, further pursuit of neuroimaging evidence of disrupted blood brain barrier, as initiated by Dysken, et al [ 252 ], and with manipulation of inflammatory systems as suggested by Mueller and Ackenheil [ 253 ] would test our hypothesis. The neuropathology of schizophrenia, focused mostly on the neurons, is notable for inconsistencies in findings (see [ 51 , 254 ] for reviews). Such inconsistency is exactly what would be predicted by an inflammatory/vascular theory where the lesions are truly functional in the sense that the function of the brain alters in relation to perturbations in blood flow regulation. Only the more prolonged and serious inflammation will leave visible traces of neuronal damage and such damage may be patchy and inconsistent from one patient to another. However, over the early years of CNS development, alterations in cellular organization or migration may result from disrupted angiogensis that must go hand in hand with neuronal and glial development. The location and extent of CNS change will be a function of severity of inflammation and timing during development. Such consequences will be hard to demonstrate in human post-mortem tissues and animal or in vitro models may be more fruitful areas for study the effects of inflammation on neurogenesis and blood flow regulation. To our knowledge, human post mortem studies have not utilized vascular cast methodology and this should be considered, perhaps casting one half of a specimen brain while subjecting the other half to cellular analysis. Specificity Because of our interests and expertise, we have focused our attention on schizophrenia as the behavioral phenotype resulting from inflammatory-vascular pathology but the theory presented here is likely to be more general. Indeed, our use of examples of psychoses associated with known inflammatory- vascular pathologies (e.g. autoimmune CNS vascular disease or infectious CNS vascular disease as seen in syphilis) makes it clear that a vascular-inflammatory theory may apply to a wide range of psychotic conditions that may also include psychoses associated with mood disorders. Whereas, the classical genetic studies support the separateness of schizophrenia and mood disorders [ 255 ], there are modern molecular signs that schizophrenia and mood disorders share genetic elements in common [ 256 , 257 ]. Furthermore, mood disorders, like schizophrenia, show evidence of frontal lobe pathology, enlarged ventricles, abnormal cerebral blood flow [ 33 , 258 ] and vascular abnormalities [ 124 ]. To what extent all of these changes are epiphenomena of being psychotic (treatment effects or stress, etc) remain debatable [ 259 ]. However, finding similar brain changes in a variety of psychotic conditions does not necessarily mean these changes are epiphenomena. Examples from neuropsychiatry teach us that the underlying pathology does not necessarily define the behavioral symptoms. Thus, psychoses with Huntington disease may be affective-like or schizophreniform. Similar pathophysiological mechanisms may underlie a variety of psychotic phenotypes. The evolution of behavioral symptoms for any given pathophysiology may depend on a variety of moderating variables such as an individual's developmental age when the disease process begins, gender, hormones, genetic 'landscape' upon which the disease process unfolds, along with the nature, frequency, and intensity of successive triggers of inflammatory response. Reprise A broad spectrum of observations leads to a working hypothesis that schizophrenia and, possibly, other psychiatric syndromes are the result of genetically mediated inflammatory reactions that damage the neuron-glial-capillary triad with resultant loss of ability to fine tune regional brain metabolism. This hypothesis incorporates genetic, epigenetic [ 260 ], and environmental factors. Furthermore, an inflammatory/vascular theory can explain the variety of behavioral symptoms seen in schizophrenia, the variable course of the illness, and the numerous other puzzling observations such as an excess of minor physical anomalies. Should this theory prove heuristic, it would point to the use of inflammatory modulators in treating the illness. Perhaps more importantly, identifying individuals who were at high risk for the disorder in high genetic risk families as well as the general population, because of abnormalities of their inflammatory systems, holds hope for prevention through early intervention using inflammatory modulators. List of abbreviations ADHD attention deficit hyperactivity disorder BDNF brain derived neurotropic factor CBF cerebral blood flow CNS central nervous system DZ dizygotic FMF familial Mediterranean fever MZ monozygotic NGF nerve growth factor NO nitric oxide PANDAS pediatric autoimmune neurological disorder associated with strep. RF rheumatic fever RHD rheumatic heart disease VEGF vascular endothelial growth factor Competing interests The author(s) declare that they have no competing interests. Authors' contributions This article was the joint effort of both authors with input as noted below. Pre-publication history The pre-publication history for this paper can be accessed here:

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC554096.xml

387268

Neural Activity When People Solve Verbal Problems with Insight

People sometimes solve problems with a unique process called insight, accompanied by an “Aha!” experience. It has long been unclear whether different cognitive and neural processes lead to insight versus noninsight solutions, or if solutions differ only in subsequent subjective feeling. Recent behavioral studies indicate distinct patterns of performance and suggest differential hemispheric involvement for insight and noninsight solutions. Subjects solved verbal problems, and after each correct solution indicated whether they solved with or without insight. We observed two objective neural correlates of insight. Functional magnetic resonance imaging ( Experiment 1 ) revealed increased activity in the right hemisphere anterior superior temporal gyrus for insight relative to noninsight solutions. The same region was active during initial solving efforts. Scalp electroencephalogram recordings ( Experiment 2 ) revealed a sudden burst of high-frequency (gamma-band) neural activity in the same area beginning 0.3 s prior to insight solutions. This right anterior temporal area is associated with making connections across distantly related information during comprehension. Although all problem solving relies on a largely shared cortical network, the sudden flash of insight occurs when solvers engage distinct neural and cognitive processes that allow them to see connections that previously eluded them.

Introduction According to legend, Archimedes shouted “Eureka!” (“I have found it!”) when he suddenly discovered that water displacement could be used to calculate density. Since then, “Eureka!,” or “Aha!,” has often been used to express the feeling one gets when solving a problem with insight . Insight is pervasive in human (and possibly animal [ Epstein et al. 1984 ]) cognition, occurring in perception, memory retrieval, language comprehension, problem solving, and various forms of practical, artistic, and scientific creativity ( Sternberg and Davidson 1995 ). The Archimedes legend has persisted over two millennia in part because it illustrates some of the key ways in which insight solutions differ from solutions achieved through more straightforward problem solving. We examine the neural bases of these different problem-solving methods. Although many processes are shared by most types of problem solving, insight solutions appear to differ from noninsight solutions in several important ways. The clearest defining characteristic of insight problem solving is the subjective “Aha!” or “Eureka!” experience that follows insight solutions ( Schooler et al. 1993 ). This subjective experience can lead to a strong emotional response—according to legend, Archimedes ran home from the baths shouting “Eureka!” without donning his clothes first. In addition, problem solving with insight is characterized by the following features. (1) Solvers first come to an impasse, no longer progressing toward a solution ( Duncker 1945 ). Archimedes, for example, was stymied by King Hiero's challenge to determine whether his new crown was pure gold without damaging the crown. (2) Solvers usually cannot report the processing that enables them to reinterpret the problem and overcome the impasse ( Maier 1931 ). Insight often occurs when people are not even aware they are thinking of the problem, as reportedly happened to Archimedes while in the baths. (3) Solvers experience their solutions as arising suddenly ( Metcalfe and Wiebe 1987 ; Smith and Kounios 1996 ) and immediately recognize the correctness of the solution (or solution path). (4) Performance on insight problems is associated with creative thinking and other cognitive abilities different from those associated with performance on noninsight problems ( Schooler and Melcher 1997 ). Some researchers have argued that all these characteristics of insight solutions are essentially epiphenomenal, that insight and noninsight solutions vary only in emotional intensity, and that they are attained with precisely the same cognitive (hence neural) mechanisms ( Weisberg and Alba 1981 ; Weisberg 1986 ; Perkins 2000 ). Persistent questions about insight concern whether unconscious processing precedes reinterpretation and solution, whether distinct cognitive and neural mechanisms beyond a common problem-solving network are involved in insight, and whether the apparent suddenness of insight solutions reflects truly sudden changes in cognitive processing and neural activity. Recent work suggests that people are thinking—at an unconscious level—about the solution prior to solving problems with insight. Specifically, while working on a verbal problem they have yet to solve, people presented with a potential solution word read the actual solution word faster than they read an unrelated word ( Bowden and Beeman 1998 ). This “solution priming” effect is greater—and in fact people make solution decisions about presented words more quickly—when words are presented to the left visual hemifield, which projects directly to the right hemisphere (RH), than when words are presented to the right visual hemifield, which projects to the left hemisphere (LH). This suggests that RH semantic processing is more likely than LH semantic processing to produce lexical or semantic information that leads to the solution. These RH advantages occur only when solvers experience insight—the “Aha!” or “Eureka!” feeling that comes with insight solutions ( Bowden and Jung-Beeman 2003a ). Moreover, when subjects try to solve classic insight problems, they benefit more from hints presented to the left visual field (i.e., the RH) than from hints presented to the right visual field (i.e., the LH) ( Fiore and Schooler 1998 ). Problem solving is a complex behavior that requires a network of cortical areas for all types of solving strategies and solutions, so solving problems with and without insight likely invokes many shared cognitive processes and neural mechanisms. One critical cognitive process distinguishing insight solutions from noninsight solutions is that solving with insight requires solvers to recognize distant or novel semantic (or associative) relations; hence, insight-specific neural activity should reflect that process. The most likely area to contribute to this component of insight problem solving is the anterior superior temporal gyrus (aSTG) of the RH. Language comprehension studies demonstrate that the RH is particularly important for recognizing distant semantic relations ( Chiarello et al. 1990 ; Beeman 1998 ), and bilateral aSTG is involved in semantic integration. For example, sentences and complex discourse increase neural activity in aSTG bilaterally ( Mazoyer et al. 1993 ; Stowe et al. 1999 ), and discourse that places particular demands on recognizing or computing distant semantic relations specifically increases neural activity in RH temporal areas ( St. George et al. 1999 ; Mason and Just 2004 ), especially aSTG ( Meyer et al. 2000 ; Kircher et al. 2001 ). If this prediction of RH aSTG involvement is confirmed, it will help constrain neurocognitive theories of insight. Other cortical areas, such as prefrontal cortex and the anterior cingulate (AC) may also be differentially involved in producing insight and noninsight solutions. We used functional magnetic resonance imaging (FMRI) in Experiment 1 and electroencephalogram (EEG) measurement in Experiment 2 to test the empirically and theoretically derived hypothesis that solving problems with insight requires engagement of (or increased emphasis on) distinct neural mechanisms, particularly in the RH anterior temporal lobe. Event-related experimental designs compared neural activity when people solved verbal problems with insight to neural activity when they solved problems (from the same problem set) without insight. As in earlier behavioral work, we used a set of compound remote associate problems ( Bowden and Jung-Beeman 2003b ) adapted from a test of creative cognition ( Mednick 1962 ). Figure 1 illustrates the sequence for each trial. Subjects saw three problem words (pine, crab, sauce) and attempted to produce a single solution word (apple) that can form a familiar compound word or phrase with each of the three problem words (pineapple, crab apple, applesauce) . We relied on solvers' reports to sort solutions into insight solutions and noninsight solutions, avoiding the complication that presumed insight problems can sometimes be solved without insight ( Davidson 1995 ) and circumventing the use of different types of problems requiring different cognitive operations. Thus, we made use of the most important defining characteristic of insight problems: the subjective conscious experience—the “Aha!” A similar technique revealed distinct behavioral characteristics when people recognized solutions with insight ( Bowden and Jung-Beeman 2003a ). Note that this is a very “tight” comparison. In both conditions problems are solved using a network of processes common to both insight and noninsight solutions. If insight ratings reflect some distinct cognitive processes, this contrast will reveal the distinct underlying brain activity. In other words, within the cortical network for problem solving, different components will be engaged or emphasized for insight versus noninsight solutions. FMRI ( Experiment 1 ) should reveal neuroanatomical locations of processes that are unique to insight solutions, and EEG ( Experiment 2 ) should reveal the time course (e.g., whether insight really is sudden) and frequency characteristics of neurophysiological differences. Figure 1 Sequence of Events for Each Trial (A) The “Compound” prompt was presented for 0.5 s, then persisted for a variable amount of additional time (0–2 s) until a cue from the scanner indicated the beginning of a new whole brain acquisition. (B) A three-word problem appeared in the center of the screen and persisted until subjects indicated with a bimanual button press that they had solved the problem, or until the 30-s time limit elapsed. Thus, event timing and condition were completely dependent on subjects' responses. (C) Following the button press or time limit, subjects were prompted to verbalize the solution (or press the buttons and say “Don't know” if the time limit expired prior to solution) then (D) prompted to indicate (with a bimanual button press) whether they felt insight, as described prior to the experiment. (E) Next, subjects performed 9 s of an unrelated filler task (three line-matching trials, 3 s each), allowing BOLD signal to return to baseline (in areas not involved in line matching). Results Experiment 1 Subjects solved 59% of the problems presented, and pressed buttons indicating “insight” for 56% (s.d. = 18.2) of their solutions, “no insight” for 41% (s.d. = 18.9) of their solutions, and “other” for 2% of their solutions. We marked a point about 2 s (rounded to the nearest whole second) prior to each solution button press as the solution event, and examined a time window 4–9 s after this event (i.e., 2–7 s after the button press) to isolate the corresponding hemodynamic response. Solving problems and responding to them required a strict sequence of events (reading of words, solving effort, solving, button press, verbalizing the solution, insight decision), but this sequence was identical whether subjects indicated solving with or without insight, so differences in FMRI signal resulted from the degree to which distinct cognitive processes and neural systems led to insight or noninsight solutions. Figure 2 illustrates the most robust insight effect: as predicted, insight solutions were associated with greater neural activity in the RH aSTG than noninsight solutions. The active area was slightly anterior to primary auditory cortex, posterior to temporal pole, and along the medial aspect of the aSTG, extending down the lateral edge of the descending ramus of the Sylvian fissure to midway through the middle temporal gyrus (MTG). (This site is also close to the superior temporal sulcus, which has been implicated in language). Across all 13 subjects, the peak signal difference at a single voxel within the RH aSTG was 0.25% across the 6-s window, and 0.30% at a single time to repetition (TR), i.e., the time needed to repeat the image of the whole brain. Overall signal in this region was robust, reaching 96.8% of the brainwide average (after removing voxels in other brain areas with signal below a standard criterion). Within the cluster of voxels identified across the group, 12 subjects showed from 0.03% to 0.35% greater signal for insight than for noninsight solutions; one subject showed 0.02% greater signal for the noninsight solutions. It is not likely that RH aSTG is involved only in output or in emotional response following insight solutions, because neural activity in this area also increased when subjects first encountered each problem ( Figure 3 ). Thus, RH aSTG is involved in processing the problem words both initially and at solution. (Of course, event-related FMRI signal occurred in many other cortical regions at problem onset, especially visual cortex). There was no insight effect in response windows immediately preceding or following the defined response window. All indications point to a striking transient event in the RH aSTG near the time when subjects solve problems with insight. Figure 2 FMRI Insight Effect in RH aSTG (A) Voxels showing greater FMRI signal for insight than noninsight solutions, overlaid on the averaged normalized structural image of all subjects. The active area has a volume of 531 mm 3 (peak t = 4.89 at 44, −9, −9 in Talairach space). (B) and (C) Group average signal change following the solution event, for insight (red line) and noninsight (blue line) solutions (yellow arrow indicates button press): (B) over entire LH aSTG region; (C) over entire RH aSTG region. (D) Insight solution signal change minus noninsight solution signal change, in RH aSTG (error bars show the standard error of the mean of the difference at each timepoint). Figure 3 FMRI Signal in RH aSTG during Initial Solving Efforts (A) Voxels in right temporal lobe showing baseline-to-peak event-related FMRI signal when subjects first encounter problems, overlaid on the averaged normalized structural image of all subjects. The cluster is in RH aSTG, with a volume of 469 mm 3 , with peak t value of 4.37 at 41, −6, −12 in Talairach space, clearly overlapping with the cluster showing an insight effect at solution. (B) Group average signal change following problem onset (time = 0), for the cluster defined by signal at the problem onset (green line) and the cluster (illustrated in Figure 2 A) showing the insight effect at solution (white line). Error bars show the standard error of the mean of the difference at each time point. The involvement of the RH rather than the LH for this verbal task is not due to greater difficulty in producing insight solutions: subjects produced insight solutions at least as quickly (mean solution time = 10.25 s, s.d. = 3.58 s) as they produced noninsight solutions (mean = 11.28 s, s.d. = 4.13 s) ( t < 1.0, p > 0.3). More importantly, the hemodynamic responses to both insight and noninsight solutions in the homologous area of the LH are about equivalent to the response to noninsight solutions in the RH aSTG—it is the strong response to insight solutions in the RH aSTG that stands out. There is no insight effect anywhere within temporal cortex of the LH. At statistical thresholds below significant levels ( p < 0.1 uncorrected), there are as many voxels in LH temporal cortex showing a noninsight effect as showing an insight effect. Several other cortical areas showing insight effects that did not meet significance criteria are listed in Table 1 (see also Figure S1 ). Some of these effects were in frontal cortex, which is notable because various frontal areas have been implicated in problem solving and reasoning. Patients with prefrontal damage have particular difficulty integrating relations in reasoning tasks ( Waltz et al. 1999 ), and when healthy subjects perform the same task, neural activity increases in rostrolateral prefrontal cortext ( Christoff et al. 2001 ). Some problem solving increases activity in dorsolateral prefrontal cortex ( Prabhakaran et al. 1997 ), perhaps because of working memory demands. Solving of poorly structured problems seems particularly impaired following damage to the prefrontal cortex of the RH ( Goel and Grafman 2000 ). Moreover, the inferior frontal gyrus (IFG) is highly active when people engage in directed semantic retrieval ( Wagner et al. 2001 ) or when they select particular semantic concepts over competing ones ( Thompson-Schill et al. 1997 ), e.g., to generate a response ( Frith et al. 1991 ). Usually in these circumstances the IFG activity is stronger in the LH, even when people are reasoning about spatial problems ( Goel et al. 1998 ), but the IFG responds particularly strongly in the RH when subjects select more distant semantic relations because of task demands ( Seger et al. 2000 ) or comprehension goals ( Robertson et al. 2000 ). Because of its putative importance for problem solving, semantic retrieval, and semantic selection, IFG was an a priori region of interest. One question we had hoped to answer was whether the semantic selection of insight solutions would preferentially evoke activity in RH or LH IFG, but the insight effects in both areas were too small (in area and in reliability) to test this question. When a more lenient statistical threshold was adopted, small clusters of signal were observed in both RH and LH IFG ( Table 1 ; Figure S1 A). Indeed, within the small region surpassing this weak statistical threshold, signal change in the RH IFG region was moderately strong (peak = 0.21% across the whole window). However, as is often the case, FMRI signal in this region was low (about 72% of the brainwide average) and variability was high, decreasing our confidence in the effect. Table 1 Full FMRI Results of Insight Effect All areas showing an “insight effect”—stronger signal for insight solutions than noninsight solutions (ordered by mean percent signal change). All cluster sizes represent active voxels at t (12) = 3.43, p < 0.005, except bilateral inferior frontal gyrus areas (*), shown at 2.83, p < 0.015, because it was an a priori region of interest. Location of cluster centers and peak t values are shown in Talairach coordinates After RH aSTG, the second largest area showing an insight effect in FMRI signal was the medial frontal gyrus in the LH ( Table 1 ; Figure S1 B). Although this area was 85% as large (453 mm 3 at p < 0.005 threshold) as RH aSTG, the event-related signal within it was weak and the insight–noninsight difference (peak difference = 0.15%) was relatively small. (The insight effect may be attributable as much to a negative response for noninsight solutions as to a positive response for insight solutions.) There also was an insight effect in small clusters in or near bilateral amygdala or parahippocampal gyrus. Again, regional signal was low (83% of the brainwide average), and the signal difference was small (peak = 0.16%). However, an amygdalar response may be expected, given the emotional sensation of the insight experience (Parsons and Osherson, 2001). Hippocampal or parahippocampal involvement is also plausible, if memory interacts with insight solutions differently from how it interacts with noninsight solutions. For instance, insight problems may encourage distinct memory encoding ( Wills et al. 2000 ) or may require distinct retrieval. Finally, a small cluster in the LH posterior cingulate (PC) also showed an insight effect. There was strong, sustained FMRI signal for both solution types in this region; on the fringe of this responding region, FMRI signal began earlier following insight than noninsight solutions. The lateness of the FMRI signal across LH PC suggests that this effect began later in the response sequence, rather than during solution generation. Finally, as in most FMRI studies, signal was relatively weak in temporal pole and orbitofrontal areas due to magnetic susceptibility artifact, so we cannot rule out undetected effects in those areas. Several cortical areas showed strong solution-related FMRI signal, but approximately equally for insight and noninsight solutions. Some of these areas (e.g., motor cortex) relate to the response sequence rather than solution processes; other areas probably reflect component processes of a problem-solving network common to both insight and noninsight solving, such as retrieving potential solutions. Two areas that may be of interest for future studies are AC and posterior middle/superior temporal gyrus. Both these areas, in the RH only, showed strong, negative solution-related signal, approximately equal in the two solution types. AC is an area that might be predicted to be involved in reorienting attention as solvers overcome impasses, given its role in performance monitoring and cognitive control ( MacDonald et al. 2000 ). RH posterior MTG is active when subjects “get” jokes ( Goel and Dolan 2001 ) and when they attempt to solve problems with deductive reasoning ( Parsons and Osherson 2001 ). However, in our experiment, only the RH aSTG showed a robust insight effect. Experiment 2 A separate group of subjects participated in fundamentally the same paradigm while we continuously recorded EEGs from the scalp. We then compared time-frequency analyses of the EEGs associated with insight solutions versus noninsight solutions. EEG provides temporal resolution greatly superior to that of FMRI and thus can better elucidate the time course and suddenness of the insight effect. Furthermore, complex EEG oscillations can be parsed into constituent frequency components, some of which have been linked to particular types of neural and cognitive processes ( Ward 2003 ). The high temporal resolution of EEG allows us to address one of the fundamental questions raised earlier: does insight really occur suddenly, as subjective experience suggests? For problems typically solved without insight, solvers report gradually increasing closeness to solution. In contrast, for problems typically solved with insight, solvers report little or no progress until shortly before they actually solve the problem ( Metcalfe 1986 ; Metcalfe and Wiebe 1987 ). Similarly, quantitative analyses of the distributions of response times and accuracies during anagram solving (a task frequently eliciting the experience of insight) reveal that a solution becomes available in a discrete transition from a state of little or no information about the correct response directly to the final state of high accuracy. This contrasts with various language and memory tasks not associated with insight, which yield partial outputs before processing has been completed ( Kounios and Smith 1995 ; Smith and Kounios 1996 ). We predicted that a sudden change in neural activity associated with insight solutions would produce an EEG correlate. Specifically, we predicted that high-frequency EEG oscillations in the gamma band (i.e., greater than 30 Hz) would reflect this sudden activity, because prior research has associated gamma-band activity with the activation of perceptual, lexical, and semantic representations ( Tallon-Baudry and Bertrand 1999 ; Pulvermüller 2001 ). Gamma-band electrical activity correlates with the blood oxygenation level–dependent (BOLD) response apparent in FMRI signal; lower-frequency EEG components do not seem to have direct correlates in FMRI signal ( Foucher et al. 2003 ; Laufs et al. 2003 ). Consequently, based on the language literature discussed earlier and on our FMRI results, we predicted a discrete insight-related increase in gamma-band activity at electrodes over the anterior temporal lobe of the RH. Participants solved 46% (s.d. = 8.2) of the problems correctly within the time limit. Of correctly solved problems, subjects reported more insight solutions (56%, s.d. = 8.4) than noninsight solutions (42%, s.d. = 9.0), ( t [18] = 3.47, p =0.003); there was no difference in mean response times (insight solutions = 9.94 s, s.d. = 2.60; noninsight solutions=9.25 s, s.d. = 3.06; t < 1.0). There was a burst of gamma-band activity associated with correct insight solutions (but not noninsight solutions) beginning approximately 0.3 s before the button-press solution response at anterior right temporal electrodes ( Figure 4 ), with no significant difference between insight and noninsight solutions over homologous LH sites. A repeated-measures analysis of variance (ANOVA) performed on log-transformed gamma-band (39 Hz) EEG power at left and right temporal electrode sites (T7 and T8, respectively) for insight and noninsight trials using two time windows (−1.52 to −0.36 s and −0.30 to −0.02 s, measured with respect to the solution response) yielded significant insight × time window ( F [1,18] = 6.68, p = 0.019) and insight × time window × Hemisphere ( F [1,18] = 8.11, p = 0.011) interactions. The overall interaction occurred because there was an insight × hemisphere interaction from −0.30 to −0.02 s ( F [1,18] = 4.61, p = 0.046) but no effect in the −1.52 to −0.36 s time window. Within the −0.30 to −0.02 s interval for these two electrodes, there was a significant insight effect at the right temporal (T8) site ( t [18] = 3.48, p = 0.003), but not at the homologous left temporal (T7) site or any other LH temporal electrode. Laplacian mapping of this effect ( Figure 4 B) is remarkably consistent with the FMRI signal in RH aSTG observed in Experiment 1 . (EEG does not have the spatial resolution of FMRI. However, we used the Laplacian transform [i.e., second spatial derivative] to localize observed activity. The Laplacian derivation acts as a high-pass spatial filter that reduces the contribution from activity in distant areas of the brain to the signal at a given electrode, and therefore reflects relatively focal and proximal brain activity. Given our FMRI results and the demonstrated correspondence between high-frequency EEG activity and FMRI signal [ Foucher et al. 2003 ; Laufs et al. 2003 ], we are confident in the localization of this effect.) Figure 4 Gamma-Band Power for Insight and Noninsight Solutions (A) Grand average time course of EEG power (in v 2 ) at 39 Hz estimated with the Morlet wavelet transform at right superior temporal electrode T8. The x -axis represents time (in seconds) with the yellow arrow and R marking the point in time of the solution button-press response (i.e., 0.0 s). The green horizontal bars above the x -axis represent the time intervals used in the statistical analyses and topographic maps. Note that gamma-band power for insight trials (red line) starts to increase above power on noninsight trials (blue line) by approximately 0.3 s before the button press. (B) Time-frequency plots of the insight minus noninsight difference shown in (A). The y -axis represents frequency (in Hz); the x -axis represents time (in seconds, with respect to the button press, exactly as shown in [A]). Red areas of the plot reflect times and frequencies at which insight EEG power is greater than noninsight EEG power; blue areas reflect times and frequencies at which noninsight EEG power is greater than insight EEG power. Note the sudden emergence of increased gamma power for insight solutions approximately 0.3 s before the button press. (C) Insight minus noninsight gamma-band differences plotted as topographic maps (LH and RH) of scalp current density (in v 2 /m 2 ) estimated by a spline-based Laplacian transform computed with a realistic FMRI-derived head model. The Laplacian transform acts as a high-pass spatial filter that minimizes the contribution of activity distant from each electrode, thereby manifesting discrete, relatively superficial sources. The maps are thresholded to show foci of current density at the upper and lower 20% of the scale. Note the prominent effect of insight (effect for insight greater than effect for noninsight, in red) at the right superior temporal electrode (T8) and surrounding electrodes present from −0.30 to −0.02 s (measured with respect to the solution response) that is not present in the earlier epoch (−1.52 to −0.36 s). The blue area over left inferior parietal cortex (electrode P7) indicates that noninsight gamma power is nonsignificantly greater than insight power ( F [1,19] < 1) over this region. The gamma burst in the right temporal area cannot be attributed to motor processes involved in making the response because (A) motor activity associated with the bimanual button press would have caused a bilateral gamma burst, not a unilateral one; (B) the location of the gamma burst as determined by Laplacian mapping ( Figure 4 B) is not consistent with hand-related motor cortex activity; and (C) both insight and noninsight solutions required button presses. Other planned statistical tests (ANOVAs) examined possible insight-related frontal theta (5–8 Hz), posterior alpha (8–13 Hz), and fronto-central beta (13–20 Hz) activity. There were no statistically significant theta or beta effects. (Visual inspection and post hoc statistical tests suggested insight-related frontal 4-Hz activity, but this effect cannot be reliably distinguished from possible artifacts due to small vertical eye movements.) There was a significant posterior alpha effect, which is discussed below. Discussion Complex problem solving requires a complex cortical network to encode the problem information, search memory for relevant information, evaluate this information, apply operators, and so forth. The FMRI and EEG results reported here conclusively demonstrate that solving verbal problems with insight requires at least one additional component to this cortical network, involving RH aSTG, that is less important to solving without insight. The insight effect in RH aSTG accords with the literature on integrating distant or novel semantic relations during language comprehension. When people comprehend (read or listen to) sentences or stories, neural activity increases in aSTG or temporal pole bilaterally more than when comprehending single words ( Mazoyer et al. 1993 ; Bottini et al. 1994 ; Stowe et al. 1999 ; Humphries et al. 2001 ; Meyer et al. 2000 ). Neural activity increases in predominantly RH aSTG during tasks that emphasize integration across sentences to extract themes ( St. George et al. 1999 ) or to form more coherent memories for stories ( Mason and Just 2004 ). RH aSTG is also selectively active when subjects must generate the best ending to a sentence ( Kircher et al. 2001 ) or mentally repair grammatically incorrect sentences ( Meyer et al. 2000 ), both of which likely require intense semantic integration. Like the results in language processing, the current results are predicted by the theory that the RH performs relatively coarse semantic coding ( Beeman 1998 ; similarly, Chiarello et al. 1990 ). This theory contends that when people encounter words, semantic processing in several LH areas engages in relatively fine semantic coding which produces small semantic fields—i.e., this processing strongly focuses on a few concepts closely related to the input word in the given context. This is very effective for most straightforward language processing. In contrast, the homologous RH areas engage in relatively coarse semantic coding, which produces large and weak semantic fields—i.e., this processing includes many concepts, even concepts distantly related to the input words and context. This process is ineffective for rapid interpretation or selection but increases semantic overlap among multiple semantic fields ( Beeman et al. 1994), which is useful when drawing together parts of a story or conversation that are only distantly related ( Beeman 1993 ; Beeman et al. 2000 ). In this view, the coarseness of semantic coding is largely influenced by slight asymmetries in neural microcircuitry that produce more discrete, less redundant input fields in pyramidal neurons of the LH language cortex, and more overlapping input fields in corresponding neurons in the RH (for reviews see Beeman 1998 ; Hutsler and Galuske 2003 ). We suggest that semantic integration, generally, is important for connecting various problem elements together and connecting the problem to the solution, and that coarsely coded semantic integration, computed in RH aSTG, is especially critical to insight solutions, at least for verbal problems (or problems that can be solved with verbal or semantic information). People come to an impasse on insight problems because their retrieval efforts are misdirected by ambiguous information in the problem or by their usual method for solving similar problems. Large semantic fields allowing for more overlap among distantly related concepts (or distantly associated lexical items) may help overcome this impasse. Because this semantic processing is weak, it may remain unconscious, perhaps overshadowed by stronger processing of the misdirected information ( Schooler et al. 1993 ; Smith 1995 ), and solvers remain stuck at impasse. Eventually, solution-related information bursts into awareness “in a sudden flash.” This can happen after misdirected processing decays or is suppressed, after solution-related processing grows, or after environmental cues occur—such as the water overflowing the bathtub when Archimedes got in. Archimedes had semantic and verbal knowledge about how to compute density from weight and volume, but struggled with measuring the volume of an irregularly shaped crown without harming the crown (e.g., melting it). His observation of water displacement allowed him to connect known concepts in new ways. This is the nature of many insights, the recognition of new connections across existing knowledge. A persistent question has been whether the cognitive and neural events that lead to insight are as sudden as the subjective experience. The timing and frequency characteristics of the EEG results shed light on this question. We propose that the gamma-band insight effect in Experiment 2 reflects the sudden transition of solution-related cognitive processing from an unconscious to a conscious state. Recent research associates gamma-band oscillations with the ignition of neural cell assemblies supporting the transient feature binding necessary to activate a representation ( Tallon-Baudry and Bertrand 1999 ; Pulvermüller 2001 )—in this case, a phonological, lexical, or semantic representation corresponding to the solution word and its associations to the problem words. According to this hypothesis, greater synchronous gamma-band activity for insight than for noninsight solutions could reflect a more integrated or unitized solution representation. Furthermore, synchronous gamma-band activity has been hypothesized to play a critical role in the accessibility to consciousness of such representations ( Engel and Singer 2001 ). The timing (with respect to the solution button press) of the insight gamma-band effect closely approximates estimates derived from cognitive behavioral studies of the amount of time required to access an available solution and generate a two-alternative, forced-choice button-press response (e.g., Kounios et al. 1987 ; Meyer et al. 1988 ; Smith and Kounios 1996 ). The present experiments had no response choice (i.e., always the same bimanual button press for solutions), so subjects could easily have responded 0.3 s after solving the problems. Thus, we infer that the observed gamma burst reflects the sudden conscious availability of a solution word resulting from an insight. Suddenly recognizing new connections between problem elements is a hallmark of insight, but it is only one component of a large cortical network necessary for solving problems with insight, and recognizing new connections likely contributes to other tasks, such as understanding metaphors ( Bottini et al. 1994 ) and deriving a story theme ( St. George et al. 1999 ). Similar tasks may depend on related cortical networks. For example, appreciating semantic jokes ( Goel and Dolan 2001 ) and engaging in deductive reasoning that sometimes involves insight ( Parsons and Osherson 2001 ) both increase activity in RH posterior MTG. It is striking that the insight effect observed in the RH in our experiments occurred when people solved verbal problems, which traditional views suggest should involve mostly LH processing with little or no contribution from the RH. It is possible that insight solutions to nonverbal problems would require different cortical networks. However, the observed effect cannot be due simply to verbal retrieval, which must occur for both insight and noninsight solutions; it could be due to a type of verbal retrieval specific to insight solutions, but not involved in noninsight solutions. We turn now to another result from the EEG time-frequency analysis, which was not predicted but nevertheless suggests a provocative interpretation. The gamma burst thought to reflect the transition of the insight solution from an unconscious to a conscious state was preceded by insight-specific activity in the alpha band (8–13 Hz). Specifically, there was a burst of alpha power (estimated at 9.8 Hz) associated with insight solutions detected over right posterior parietal cortex from approximately 1.4 s until approximately 0.4 s before the solution response, at which point insight alpha power decreased to the level of noninsight alpha power, or below ( Figure 5 ). An ANOVA was performed on log-transformed alpha-band (9.8 Hz) EEG power at left and right parietal-occipital electrode sites (PO7 and PO8, respectively) for insight and noninsight trials using three time windows: −2.06 to −1.56 s, −1.31 to −0.56 s, and −0.31 to 0.06 s (measured from the solution button press). This analysis yielded a significant insight × time window interaction ( F [2,36] = 4.13, p = 0.027, with the Huynh-Feldt correction). Follow-up t -tests in each time window yielded significant effects of insight in the first time window at both electrode sites (PO7: t [18] = 2.32, p = 0.033; PO8: t [18] = 2.42, p = 0.026) and in the second time window only at the RH site (PO8: t [18] = 2.17, p = 0.043), with a reversal of the direction of the effect. The third time window yielded no significant effects. Figure 5 Alpha-Band Power for Insight and Noninsight Solutions (Same conventions as in Figure 4 ). (A) Time course of EEG power at 9.8 Hz (in v 2 ) at right parietal-occipital electrode (PO8). The x -axis represents time (in seconds), with the green horizontal bars above the x -axis representing the time intervals used in the statistical analyses and topographic maps. The yellow arrow and R (at 0.0 s) signify the time of the button-press response. (B) Time-frequency plots of the insight minus noninsight difference shown in (A). (C) Insight minus noninsight alpha-band differences plotted as topographic maps of scalp current density (in v 2 /m 2 ). Note that alpha-band power is significantly greater for insight solutions than noninsight solutions during the −1.31 to −0.56 s interval, but not during the preceding (−2.06 to −1.56 s) or subsequent (−0.31 to +0.06 s) intervals. This alpha burst was embedded in a slow decrease in alpha (see [A]), probably reflecting a general increase in cortical activity as effort increases during the course of problem solving. Alpha rhythms are understood to reflect idling or inhibition of cortical areas ( Pfurtscheller et al. 1996 ). Increased alpha power measured over parietal-occipital cortex indicates idling or inhibition of visual cortex. This has been attributed to gating of visual information flowing into the perceptual system in order to protect fragile or resource-intensive processes from interference from bottom-up stimulation ( Ray and Cole 1985 ; Worden et al. 2001 ; Jensen et al. 2002 ; Cooper et al. 2003 ; Ward 2003 ). This interpretation assumes that brain areas are normally highly interactive, and that allowing one process to proceed relatively independently requires active attenuation of this interaction. For instance, when subjects attend to visual space in the hemifield projecting to one hemisphere, posterior alpha increases over the other hemisphere, which receives inputs from the unattended hemifield ( Worden et al. 2001 ). Analogously, the present results suggest selective gating of visual inputs to the RH during the interval preceding the insight-related right temporal gamma burst ( Figure 6 ). Hypothetically, this allows weaker processing about more distant associations between the problem words and potential solutions to gain strength, by attenuating bottom-up activation or other neural activity not related to solution that would decrease the signal-to-noise ratio for the actual solution. Figure 6 The Time Course of the Insight Effect Alpha power (9.8 Hz at right parietal-occipital electrode PO8) and gamma power (39 Hz at right temporal electrode T8) for the insight effect (i.e., correct insight solutions minus correct noninsight solutions, in v 2 ). The left y -axis shows the magnitude of the alpha insight effect (purple line); the right y -axis applies to the gamma insight effect (green line). The x -axis represents time (in seconds). The yellow arrow and R (at 0.0 s) signify the time of the button-press response. Note the transient enhancement of alpha on insight trials (relative to noninsight trials) prior to the gamma burst. This interpretation of the early insight-specific alpha effect is consistent with previous behavioral research suggesting that, prior to an insight, the solution to a verbal problem can be weakly activated ( Bowers et al. 1990 ), especially in the RH (Bowden and Beeman 1998 ; Bowden and Jung-Beeman 2003a ). Thus insight solutions may be associated with early unconscious solution-related processing, followed by a sudden transition to full awareness of the solution. We suggest that, in Experiment 2 , the early posterior alpha insight effect is an indirect correlate of the former, and the right temporal gamma effect is a direct correlate of the latter. In sum, when people solve problems with insight, leading to an “Aha!” experience, their solutions are accompanied by a striking increase in neural activity in RH aSTG. Thus, within the network of cortical areas required for problem solving, different components are engaged or emphasized when solving with versus without insight. We propose that the RH aSTG facilitates integration of information across distant lexical or semantic relations, allowing solvers to see connections that had previously eluded them. In the two millennia since Archimedes shouted “Eureka!,” it has seemed common knowledge that people sometimes solve problems—whether great scientific questions or trivial puzzles—by a seemingly distinct mechanism called insight. This mechanism involves suddenly seeing a problem in a new light, often without awareness of how that new light was switched on. We have demonstrated that insight solutions are indeed associated with a discrete, distinct pattern of neural activity, supporting unique cognitive processes. Materials and Methods Subjects Ten men and eight women were paid to participate in Experiment 1 ; 19 new subjects (nine men, ten women) were paid to participate in Experiment 2 . All were young (18–29) neurologically intact, right-handed, native English speakers; Experiment 1 participants met safety criteria for FMRI scanning. After hearing about all methods and risks and performing practice trials, they consented to participate. In Experiment 1 , data from four men and one woman were excluded due to poor FMRI signal or because subjects provided fewer than ten insight or noninsight responses. This research was approved by the University of Pennsylvania Institutional Review Board. Behavioral paradigm Following practice, subjects attempted 124 compound remote associate problems during FMRI scanning. These problems ( Bowden and Jung-Beeman 2003 b) can be solved quickly and evoke an “Aha!” experience, producing a distinct behavioral signature ( Bowden and Jung-Beeman 2003 a), roughly half the time they are solved. Figure 1 illustrates the sequence of events for each trial. Each trial began with the task label “Compound” presented on liquid crystal diode goggles for 0.5 to 2.5 s. A gating signal from the scanner triggered the central presentation of three problem words, which persisted until subjects solved the problem or 30 s elapsed. If subjects solved the problem, they made a bimanual button press, after which the word “Solution?” prompted them to verbalize their solution. After 2 s the word “Insight?” prompted subjects to press buttons indicating whether they solved the problem with insight. Prior to the experiment subjects were told the following: “A feeling of insight is a kind of ‘Aha!' characterized by suddenness and obviousness. You may not be sure how you came up with the answer, but are relatively confident that it is correct without having to mentally check it. It is as though the answer came into mind all at once—when you first thought of the word, you simply knew it was the answer. This feeling does not have to be overwhelming, but should resemble what was just described.” The experimenter interacted with subjects until this description was clear. This subjective rating could be used differently across subjects (or even across trials), blurring condition boundaries; yet the distinct neural correlates of insight observed across the group demonstrate that there was some consistency. If subjects failed to solve problems within 30 s, the “Solution?” prompt appeared, and subjects pressed the “no” buttons and verbalized “Don't Know.” Then the “Insight?” prompt appeared, and subjects pressed the “no” buttons again. After the insight rating, subjects performed three line-matching trials (3 s each) to distract them from thinking about the problems, allowing the critical BOLD signal to return to baseline ( Binder et al. 1999 ). The total time from the end of one problem to the onset of the next was 14.5–16.5 s. The condition (e.g., insight or noninsight solution) and time of events was determined by subjects' responses. Image acquisition Imaging was performed at the Hospital of the University of Pennsylvania, on a 1.5 Tesla GE SIGNA scanner with a fast gradient system for echo-planar imaging and a standard head coil. Head motion was restricted with plastic braces and foam padding. Anatomical high-resolution T1-weighted axial and sagittal images were acquired while subjects performed practice trials. Functional images (21 slices, 5 mm thick; 3.75-mm × 3.75-mm in-plane resolution; TR = 2000 ms for 21 slices; time to echo = 40 ms) were acquired in the same axial plane as the anatomical images using gradient-echo echo-planar sequences sensitive to BOLD signal ( Kwong et al. 1992 ; Ogawa et al. 1992 ). Each functional run was preceded by a 20-s saturation period. Subjects participated in four 15-min runs and a fifth run of varying length, depending on the number of remaining problems. Image analysis Images were coregistered through time with a three-dimensional registration algorithm ( Cox 1996 ). Echo planar imaging volumes were spatially smoothed using a 7.5-mm full-width half-maximum Gaussian kernel. Within each run, voxels were eliminated if the signal magnitude changed more than 10% across successive TRs, or if the mean signal level was below a noise threshold. Functional data were transformed ( Collins et al. 1994 ) to a standard stereotaxic atlas ( Talairach and Tournoux 1988 ) with a voxel size of 2.5 mm 3 . Data were analyzed using general linear model analysis that extracted average responses to each trial type, correcting for linear drift and removing signal changes correlated with head motion. Each TR was divided into two 1-s images to improve time locking of the solving event and the functional image data (time-course data were temporally smoothed in Figures 2 and 3 ). Solution-related responses were calculated using the average signal change within the window 4–9 s (to account for hemodynamic delay) after the solving event (beginning about 2 s prior to the button press). Differences between insight and noninsight solution events were estimated for each participant, then combined in a second-stage random effects analysis to identify differences consistent across all subjects. A cluster threshold was set at regions at least 500 mm 3 in volume (32 normalized voxels, or 7.1 original-sized voxels) in which each voxel was reliably different across subjects, ( t [12] > 3.43, p < 0.005 uncorrected). Monte Carlo simulations with similar datasets reveal low false positive rates with these criteria. RH aSTG was the only cluster to exceed these criteria, and converging evidence and the a priori prediction about RH aSTG strengthen confidence in this result. Experiment 2 Behavioral procedures were similar to those of Experiment 1 , except that (A) problem words were presented at smaller visual angles to discourage eye movements, (B) there were 2-s delays between each event in the response sequence, and (C) subjects triggered a new problem directly after responding to the previous problem (i.e., no line task occurred between problems). EEG methods Continuous high-density EEGs were recorded at 250 Hz (bandpass: 0.2–100 Hz) from 128 tin electrodes embedded in an elastic cap (linked mastoid reference with forehead ground) placed according to the extended International 10–20 System. Prior to data analysis, EEG channels with excessive noise were replaced with interpolated data from neighboring channels. Eyeblink artifacts were removed from the EEG with an adaptive filter separately constructed for each subject using EMSE 5.0 (Source Signal Imaging Inc., San Diego, California, United States). Induced oscillations were analyzed by segmenting each subject's continuous EEG into 4-s segments beginning 3 s before each solution response. (An analysis epoch beginning at an earlier point in time would have resulted in the loss of trials associated with response times of less than 3 s.) Time-frequency transforms (performed with EMSE 5.0) were obtained by the application of complex-valued Grossmann-Morlet wavelets, which are Gaussian in both time and frequency. Following Torrence and Campo (1998 ), the mother wavelet, ω 0 , in the time domain has the form where ω 0 is a nondimensional frequency. In this case, ω 0 is chosen to be 5.336, so that ∫ϕ 0 ( t ) ≅ 0. The constant π−¼ is a normalization factor such that ∫(ϕ 0 ( t )) 2 = 1. For the discrete time case, a family of wavelets may be obtained as where δ t is the sample period (in seconds), s is the scale (in seconds), and n is an integer that counts the number of samples from the starting time. The Fourier wavelength λ is given by In the frequency domain, the (continuous) Fourier transform of Equation 2 is where One reasonable way to measure the “resolution” of the wavelet transform is to consider the dispersion of the wavelets in both time and frequency. Since the wavelets are Gaussian in both domains, the e -folding time and frequency may serve as quantitative measures of dispersion. Note that these dispersions are a function of the scale, s . For a selected frequency, 𝒻 c = 1/λ, or from Equation 3 Then substituting into Equation 2 , we find that the e -folding time is for frequency 𝒻 c . From Equation 2 , the e -folding frequency is . To make this concrete, we find that for a 10-Hz (alpha-band) center frequency, the e -folding time is 0.12 s and the e -folding frequency is 2.6 Hz. For a 40-Hz ( gamma-band) center frequency, the e -folding time is 0.03 s and the e -folding frequency is 10.5 Hz. Note that these e -folding parameters imply that wavelet scaling preserves the joint time-frequency resolution (equal areas in time-frequency space), with higher temporal resolution but broader frequency resolution as the wavelet scale decreases. Segments corresponding to trials for which individual subjects produced the correct response were isolated and averaged separately according to whether or not the subject reported the experience of insight. Planned statistical tests (repeated-measure ANOVAs) were performed in order to detect insight-related effects on frontal midline theta (5–8 Hz), posterior alpha (8–13 Hz), fronto-central beta (13–20 Hz), and left and right temporal gamma (20–50 Hz). Response-locked event-related potentials (ERPs) were also computed using the same analysis epoch. Standard ERP analyses yielded no evidence of statistically significant effects, likely because ERPs reflect phase-locked activity rather than the induced (i.e., nonphase-locked) activity examined in the wavelet analyses; due to the long response times evident in this experiment, phase locking resulting from problem presentation would not be expected. EEG effects were topographically mapped by employing spline-based Laplacian mapping with an FMRI-derived realistic head model and digitized electrode positions. Localization of EEG/ERP signals is a form of probabilistic modelling rather than direct neuroimaging. In contrast to other techniques, source estimation by Laplacian mapping indicates the presence of superficial foci of neuroelectric activity with minimal assumptions. Supporting Information Figure S1 Cortical Regions Showing “Insight Effects” Below Cluster Size Threshold The far left lane shows for each region a single slice best depicting the cluster activated above threshold; middle lane shows time course of signal following insight (red line) and noninsight (blue line) solutions, across the entire active cluster; right panel shows the “insight effect” (insight signal minus noninsight signal, error bars show the standard error of the mean of the difference at each timepoint). (A) depicts bilateral IFG with lowered threshold ( t [12] = 2.83, p < 0.015); (B–D) depict clusters of FMRI signal at the same t -threshold used in the main paper ( t [12] = 3.43, p < 0.005), but the clusters are too small to surpass cluster criterion. (B) LH medial frontal gyrus; (C) LH PC gyrus; (D) LH amygdala (there was also a small cluster near RH amygdala). Spatial coordinates and other are details listed in Table 1 . (914 KB PDF). Click here for additional data file.

/Users/keerthanasridhar/biomedlm/data/PMC000xxxxxx/PMC387268.xml