DNA pooling analysis of ADHD and genes regulating vesicle release of neurotransmitters.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 139B:33 –37 (2005) DNA Pooling Analysis of ADHD and Genes Regulating Vesicle Release of Neurotransmitters K.J. Brookes, J. Knight, X. Xu, and P. Asherson* MRC Social, Genetic Developmental Psychiatry Centre, Institute of Psychiatry, United Kingdom ADHD is one of the most prevalent, and heritable behavioural disorders in childhood. Genetic associations have been reported with polymorphic variants within or near to dopamine pathway genes. Recently snap-25 has also shown association with ADHD in several datasets. We therefore investigated other genes that produce proteins that interact with SNAP-25 in the mechanism of vesicular release of neurotransmitters at the synapse. A total of 106 SNPs were screened for minor allele frequency greater than 5% and 61 SNPs selected for analysis in DNA pools made up from an ADHD clinical sample of DSM-IV combined type probands (n ¼ 180) and a control sample of 90 males and 90 females. Initial screening identified several SNPs that showed allele frequency differences of 5% or more. One SNP in the synaptophysin gene showed suggestive evidence of association following case-control and TDT analysis and warrants further investigation. ß 2005 Wiley-Liss, Inc. KEY WORDS: DNA pooling; ADHD synaptic components; association INTRODUCTION Despite the relative success of candidate gene association mapping in ADHD [Asherson, 2004] there have been limitations to the rate of progress. The rate limiting-step for most candidate gene studies has been the cost-effective generation of individual genotypes. Genotyping individual samples, especially where large sample sizes are required to obtain adequate statistical power remains expensive and has no guarantee that an association will be found. For these reasons many studies have restricted themselves to the analysis of one or only a few markers for each gene investigated. There are only a few examples of more systematic analyses that aim to test for association with markers spanning intragenic regions and, Abbreviations: ADHD, attention deficit hyperactivity disorder; SNP, single nucleotide polymorphism; TDT, transmission disequilibrium test; snap-25, synaptosomal associated protein– 25 kD; stx1a, syntaxin 1A; vamp2, vesicle associated membrane protein 2; syt1, *Synaptotagmin 1; syp, synaptophysin; vmat2, vesicular monoamine transporter 2; SNARES, soluble N-ethylmaleimide-sensitive factor-attachment protein receptors; MRM, meta-regression method. Grant sponsor: Wellcome Trust project (to PA). *Correspondence to: P. Asherson, MRC Social, Genetic Developmental Psychiatry Centre, Institute of Psychiatry, United Kingdom. E-mail: firstname.lastname@example.org Received 24 March 2005; Accepted 2 June 2005 DOI 10.1002/ajmg.b.30216 ß 2005 Wiley-Liss, Inc. as a consequence, negative reports fail to exclude association to a particular gene, even within the power constraints set by sample size. One approach to overcome these problems is the use of DNA pooling to screen dense sets of markers. We have previously argued that at the present time a DNA pooling approach using all available single nucleotide polymorphisms (SNPs) spanning a candidate gene is an efficient method for initial screening of genes for association [Jawaid et al., 2004]. The approach of using haplotype-tagging SNPs identified from the HAPMAP database is a suitable alternative for selection of SNPs for DNA pooling, particularly in regions of strong marker-marker linkage disequilibrium, but may not pick up associations with SNPs that are not strongly correlated with available HAPMAP markers [Sham et al., 2002]. DNA pooling reduces the amount of work involved when investigating multiple genetic polymorphisms by combining DNA samples together into case and control pools and estimating allele frequency differences from the pooled allele image patterns [Sham et al., 2002]. Positive associations suggested by the DNA pooling screening method can then be confirmed and investigated further by conventional use of individual genotypes. Here were have applied a DNA pooling strategy to search for associations between ADHD and single nucleotide polymorphisms (SNPs) spanning six genes involved in the vesicular release of neurotransmitters at the synapse; syntaxin 1A (stx1a), vesicle associated membrane protein 2 (vamp2), synaptotagmin (syt1), synaptophysin (syp), and vesicular monoamine transporter 2 (vmat2). Interest in vesicular synaptic proteins follows from reports of association between ADHD and polymorphic variants within the synaptosomal associated protein (snap-25) gene [Barr et al., 2000; Brophy et al., 2002; Mill et al., 2002; Kustanovich et al., 2003; Mill et al., 2004]. SNAP-25 forms part of a stable membrane bound ternary complex to permit neurotransmitter release from vesicles into the synaptic cleft. Proteins involved in these complexes are known as SNARES (soluble N-ethylmaleimide-sensitive factor-attachment protein receptors) and are essential for intracellular trafficking. The release of neurotransmitters relies on the fusion of the vesicle with the plasma membrane and three main proteins have been identified that are essential in this process; SNAP-25, STX1A, and VAMP2. In the neurone, the SNARE proteins form a bridge between the synaptic vesicle and the plasma membrane, driving the membrane fusion required for neurotransmitter release. SNARES form a coiled bundle of 4 alpha helices that run parallel, aligning the transmembrane domains of VAMP2 and STX1A at the same end of the bundle. There are two SNARE groups that complex during exocytosis. (1) The vesicle membrane SNARE (v-SNARE) is made up of VAMP2, and contributes one helix to the complex. The availability of VAMP2 is limited, and is therefore a rate-limiting step in the control of exocytosis. (2) The target membrane SNARE (tSNARE) is made up of a complex of SNAP-25 and STX1A that contributes three helices to the fusion process. It is believed that the proteins STX1A and VAMP2 are the minimal 34 Brookes et al. machinery needed for membrane fusion [Weber et al., 1998]. It was found that these t- and v- SNAREs could form a stable complex in the absence of other proteins, although they appear to be under kinetic control via regulatory proteins in vivo [Rothman and Sollner, 1997]. Therefore it would seem that any alteration in their function, might impair the release of dopamine into the synapse. Other proteins that function in conjunction with SNARES are synaptotagmin (syt1) and synaptophysin (syp). Synaptotagmin encodes a protein that is integral to the synaptic vesicle membrane and binds calcium required for the release of neurotransmitter at the synapse [Fernandez-Chacon et al., 2001]. Synaptophysin also encodes a synaptic vesicle integral membrane protein although is precise function is not known [McMahon et al., 1996]. The vesicular monoamine transporter 2 (vmat2) acts to accumulate monoamines from the cytoplasm into synaptic vesicles. It is essential for the correct functioning of monoamine systems, and has previously been implicated in human neuropsychiatric disorders [Peter et al., 1993]. METHODS Study Design For the initial screening step, equal quantities of DNAs were combined together into case and control pools from which allele frequency differences were estimated using a primer extension assay (Snapshot, ABI Foster City). Several investigators have reported that the SNapShot reaction provides an accurate method for determining allele frequency differences between case and control pools [e.g., Norton et al., 2002]. Significant associations suggested by the DNA pooling step were confirmed by individual genotyping of case and control samples and within family analysis using parental genotypes to guard against potential stratification effects. Samples The clinical ADHD samples were ascertained from Child and Adolescent clinics in London and the South East of England. Cases were referred for assessment if they were thought by experienced clinicians to have a diagnosis of the combined subtype of ADHD under DSM-IV criteria, with no significant Axis I co-morbidity apart from oppositional defiant disorder (ODD) and conduct disorder (CD). Parents of referred cases were interviewed with a modified version of the Child and Adolescent Psychiatric Assessment (CAPA) [Angold et al., 1995]. Information on ADHD symptoms at school was obtained using the long form of the Conners questionnaire [Conners, 1995]. Following assessments HYPESCHEME data sheets were completed using data gathered from the research interview, questionnaire and where necessary review of case notes. HYPESCHEME is an operational criteria checklist for ADHD and hyperkinetic disorders that summarises and applies DSM- IV and ICD-10 operational criteria [Curran et al., 2000]. HYPESCHEME diagnoses were checked against researcher applied DSM-IV criteria and discrepancies reviewed by two researchers (PA and SR). Where consensus could not be reached, cases were brought to case conference and final consensus agreement made with a senior clinical researcher (ET). DNA was obtained using cheek swabs as described in Freeman et al. . Samples were all of English speaking Caucasians of European origin. All probands had a research diagnosis of ADHD combined subtype following the DSM-IV criteria [APA, 1994]. Exclusion criteria included IQ less than 70, neurological disorders or brain damage and autism. The sample analyzed here consisted of 180 probands. DNA was available from both parents for 121 probands and from only the mother for 64 probands. Ninety-six percent of the probands were male. A control sample was generated from ethnically matched individuals taking part in a population twin sample [Trouton et al., 2002]. This comprised 90 unrelated females and 90 unrelated males selected for low ADHD symptoms scores (lower 20%) using a composite index of ADHD symptoms. The composite index was derived from the average of maternally rated strength and difficulty questionnaire [SDQ; Goodman, 1997] scores at ages 2,3, and 4 [Price et al., 2001]. DNA Pooling Screen SNPs were identified from dbSNP (http://www.ncbi.nlm. nih.gov). We aimed to include all SNPs within the region of each gene that had minor allele frequencies of 0.05 or above. In addition we included coding region SNPs and additional SNPs with unknown minor allele frequencies in regions where those with known heterozygosity were not listed. In the first stage of the analysis, SNP markers were screened for association using a DNA pooling approach. DNA pools were constructed by mixing equal quantities of DNA quantified to a final concentration of 5 ng/ml (0.5 ng) prior to mixing. The concentration of each DNA sample was measured using the PicoGreen dsDNA quantitation reagent (Cambridge Biosciences) in a Fluorimeter (Thermo Life Sciences). DNA pools constructed consisted of two case pools (n ¼ 90, n ¼ 90) and two control pools (n ¼ 90, n ¼ 90). Each genotype assay was analyzed in triplicate on each pool using the SNaPshotTM method (ABI, Foster City). Allele frequencies (AF) were estimated from the DNA pool images by averaging across each set of triplicate data with no adjustment for unequal peak heights observed in heterozygote samples: AF(allele 1) ¼ PH(allele 1)/ (PH(allele 1) þ PH(allele 2)). In order to account for both technical error and sampling error in estimating an appropriate significance value for observed allele frequency differences between case and control pools, we adopted a meta-regression method (MRM) for the analysis of multiple pools [Xu et al., 2005]. For each pool we TABLE I. SNP Markers That Were Selected for Individual Genotyping Marker names Gene SNP STX1A SYP SYT1 VAMP2 VMAT2 rs1569061 rs2293945 rs1245769 rs316987 rs363225 DNA pooling Individual genotyping Estimated allele frequency difference MRM P-value Case allele frequency Control allele frequency Case-control P-value T NT TDT P-value 0.09 0.09 0.06 0.05 0.07 0.75 0.03 0.09 0.29 0.49 0.94 0.25 0.68 0.905 0.44 0.92 0.32 0.72 0.904 0.39 0.24 0.09 0.23 0.91 0.31 15 16 41 26 49 6 4 37 19 41 0.05 0.005 0.73 0.30 0.40 DNA pooling, significance values were estimated from DNA pooling data using the meta-regression method (MRM); Case-control, allele counts and frequencies from case and control pools; TDT, Transmitted (T) and non-transmission (NT) of the putative risk allele from heterozygote parents. DNA Pooling Analysis of Synaptic Component derived an estimate of the effect size, the variance of the estimate and an independent variable relating to the phenotype of the individuals in each pool. Effect size of each pool is taken to be the average allele frequency over the measurements from the replicate pools. The measurement variance for each marker (j), within a pool, is calculated using data from the replicate pools: n P j Þ2 ðpij p i¼1 2 sj ¼ n1 TABLE II. A Summary of the Findings in This Study Gene stx1a N ¼ 27 with n the number of replicates; pj the average allele freq across the replicates; pij the allele freq for each replicate. An average measurement of experimental variance for a marker across the different pools is calculated. The total variance for each pool is then calculated: Variance ¼ j Þ s j ð1 p p ^2 þ n 2y where pj is average allele frequency; y is number of individuals; n is number of pools. SNPs that obtained an MRM significance level <0.1 were individually genotyped in the case and control samples, plus parents and siblings of the probands. TDT analysis was performed using unphased [Dudbridge, 2003]. RESULTS 106 SNPs were screened in the test pool of 50 non-ADHD subjects to evaluate whether each SNP was sufficiently polymorphic to be used to screen for associations in the experimental pools. A marker was dropped if no second allele was detectable or if the minor peak was less than 5% of the total combined peak heights. Although in a few cases this might lead to discarding a few markers with true MAF greater than 5% (because we did not take into account peak height ratios from heterozygote samples), we consider that peak heights less than 5% of the total give inaccurate estimates of allele frequency differences between pools and cannot confidently be discriminated from background. Of the 106 database SNPs analyzed, 61 SNPs (58%) were found to be sufficiently polymorphic to be tested in the experimental pools. A summary of the data from the total set of SNP markers screened for heterozygosity is available online (Table II). As expected, the majority of SNPs analyzed showed no estimated allele frequency differences in either of the DNA pool comparisons. Out of the total of 61 comparisons made there were two comparisons that were significant at the P < 0.1 significance level. These two SNPs (SYP- rs2293945; SYT1– rs1245769) were individually genotyped. An additional three markers were also selected for individual genotyping that had estimated allele frequency differences greater than 7% between case and control pools (STX1-rs1569061, VAMP2rs1061032, VMAT2-rs363225). A summary of the data from SNPs selected for individual genotyping is listed in Table I. The three SNPs selected on the basis of allele frequency differences alone were not significant following case-control of the individually genotyped data. Unexpectedly, TDT analysis of one of the markers, rs1569061 in STX1a, showed nominal significance (P ¼ 0.04). This finding was however nonsignificant when adjusted for the number of tests performed and is therefore likely to represent a type I error. Of the two markers selected on the basis of an estimated MRM P-value <0.1, only one remained significant following case-control and TDT analysis of the individually genotyped data. SNP rs2293945 in SYP showed an estimated 8% difference in allele frequency between the proband and control pools with MRM P-value ¼ 0.03. Individual genotyping confirmed the proband control association with a trend for significance 35 vamp2 N ¼ 17 syt1 N ¼ 13 syp N ¼ 19 SNP reference rs2293485 rs2091449 rs1569061 rs1049738 rs941299 rs941298 rs875342 rs867500 rs3830638 rs3793243 rs2030921 rs2030922 rs4504524 rs4582436 rs4717806 rs4363087 rs7807316 rs7793506 rs7793508 rs4717102 rs5884917 rs7776771 rs6460051 rs6951030 rs6956879 rs7790069 rs7794407 rs3208049 rs316987/1061032 rs1061089 rs930497 rs2278637 rs959303 rs1051498 rs1150 rs3178041 rs1051667 rs1527593 rs2304909 rs1051780 rs8079444 rs8066511 rs8066513 rs8067606 rs2037743 rs1918188 rs1918191 rs1465054 rs1465053 rs1245815 rs1245819 rs941133 rs1526958 rs1245769 rs7972950 rs1245816 rs1245820 rs1049749 rs7889267 rs2737731 rs2856747 rs2737732 rs3841666 rs6417804 rs7055023 rs3817678 MAF NP NP 0.055 NP 0.405 NP NP NP NP NP NP MAF according to pooling 0.49 — 0.17 — 0.25 0.32 0.13 — 0.19 0.44 0.26 — 0.11 — 0.2 0.43 0.02 — — — — 0.39 0.36 0.28 0.49 0.03 NP 0.03 0.20 0.01 NP 0.465 0.466 NP NP NP 0.45 0.27 0.17 — 0.02 NP NP NP NP NP NP 0.32 0.393 0.388 0.259 0.03 NP 0.298 0.05 NP 0.495 NP 0.354 NP 0.15 0.42 0.32 0.21 0.1 0.3 0.24 0.1 0.03 0.35 0.43 NP NP 0.04 0.27 0.5 NP NP Casecontrol ns ns ns — ns ns ns — ns ns ns — ns — ns ns ns — — — — ns ns ns ns ns — ns ns ns — ns ns — ns ns — ns — — — — — — ns ns ns ns ns — ns ns — 0.09 ns — ns — ns — — ns ns ns — — (Continued ) 36 Brookes et al. TABLE II. (Continued ) Gene vmat2 N ¼ 17 SNP reference rs2293945 rs7060540 rs6417805 rs6520404 rs7061766 rs5906752 rs5905721 rs5905722 rs5905723 rs2737733 rs363390 rs363333 rs363397 rs363342 rs363343 rs929493 rs2283140 rs363225 rs363272 rs363231 rs363236 rs2619094 rs1042543 rs3026051 rs363387 rs363388 rs363411 MAF MAF according to pooling 0.32 NP 0.46 0.34 NP 0.18 0.35 0.31 NP NP 0.12 0.07 NP 0.23 0.473 0.021 NP NP NP 0.08 NP 0.144 0.39 0.15 0.24 0.48 0.16 0.07 0.35 0.12 Casecontrol 0.034 — ns ns — ns ns ns — — ns ns — BA ns ns ns 0.208 ns ns ns — — — 0.539 — BA MAF, reported (database) minor allele frequency; MAF-pool, estimated minor allele frequency from test pools; NP, non-polymorphic markers; ns, non-significant MTR P-value; BA, bad assay. (P-value ¼ 0.09). TDT analysis provided increased evidence for the association with a P-value of <0.01. DISCUSSION In summary we used a DNA pooling approach to screen 61 SNPs from a selection of 106 database SNPs for association between ADHD and six genes involved in the mechanisms of vesicular release of neurotransmitters. In the first stage of the analysis we discarded 45 SNPs where the allele peaks was less than 5% of the combined peak heights. From the remaining 61 SNPs that were used in the DNA pooling comparisons of proband versus control we identified only two markers that generated estimated significance values <0.1. Only SNP rs2293945 in SYP showed overall evidence for association by passing each stage of the analysis. The level of significance for this marker following individual genotyping was however only 0.09 that would be expected to occur by chance alone more than four times in this study. On the other hand, a greater level of significance was obtained following TDT analysis of 0.01 that would occur less than once by chance alone with the number of SNPs analyzed. The syp gene is located on the X chromosome, with the SNP rs2293945 located within the 50 UTR, a region known to possess regulatory properties. The fact this is an X chromosome gene might be of interest since ADHD occurs more commonly in males than females [Ford et al., 2003]. The reasons for this are unknown although quantitative genetic data suggests that the same genetic risks apply equally to males and females; however the mean for ADHD symptoms in the population is lower for females than males suggesting a possible twothreshold model. X chromosome genes may play a role since females who are heterozygote for a risk allele will on average have a lower risk for ADHD. The second X chromosome may also protect females from the effects of an associated polymorphism since genes can escape or show only partial Xinactivation. Males only have one X chromosome therefore would not be as protected. We did not have sufficient females in this sample to explore the strength of this association in females. Limitations of this study include small sample size with 80% power to detect odds ratios of around 1.5, depending on minor allele frequency. There was a tendency for initial estimated significance values to be inflated compared to individual genotyping of markers. This is most likely due minor differences in quantification of individual DNA samples prior to mixing into pooled samples and demonstrates the critical importance of the initial quantification step in the first phase of the experiment. The use of TDT alongside casecontrol analysis guards against the possibility that observed differences are due solely to population stratification effects and can in some cases identify SNP associations that are significant when controlled for in this way. Our data show that the meta-regression method of analysis can correctly identify significant allele associations in case-control pooling screens. ACKNOWLEDGMENTS We thank all families involved in this research for their participation. REFERENCES American Psychiatric Association. 1994. Diagnostic and statistical manual of mental disorders. 4th edition. Angold A, Prendergast M, Cox A, Harrington R, Simonoff E, Rutter M. 1995. The child and adolescent psychiatric assessment (CAPA). Psychol Med 25(4):739–753. Asherson P. 2004. Attention-deficit hyperactivity disorder in the postgenomic era. Eur Child Adolesc Psychiatry 13(Suppl):150–170. Barr CL, Feng Y, Wigg K, Bloom S, Roberts W, Malone M, Schachar R, Tannock R, Kennedy JL. 2000. Identification of DNA variants in the SNAP-25 gene and linkage study of these polymorphisms and attentiondeficit hyperactivity disorder. Mol Psychiatry 5(4):405–409. Brophy K, Hawl Z, Kirley A, Fitgerald M, Gill M. 2002. Synaptosomalassociated protein 25 (SNAP-25) and attention deficit hyperactivity disorder (ADHD): Evidence of linkage and association in the Irish population. Mol Psychiatry 7(8):913–917. Conners CK. 1995. The Conners rating scale: Instruments for the assessments of childhood psychopathology. Duke University. Curran S, Newman S, Taylor E, Asherson P. 2000. Hypescheme: An operational criteria checklist and minimum data set for molecular genetic studies of attention deficit and hyperactivity disorders. Am J Med Genet 96(3):244–250. Dudbridge F. 2003. Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol 25(2):115–121. Fernandez-Chacon R, Konlgstorfer A, Gerber SH, Garcia J, Matos MF, Stevens CF, Brose N, Rizo J, Roscnmund C, Sudhof TC. 2001. Synaptotagmin I functions as a calcium regulator of release probability. Nature 410(6824):41–49. Ford T, Goodman R, Maltzer H. 2003. The British child and adolescent mental health survey 1999: The prevalence of DSM-IV disorders. J Am Acad Child Adolesc Psychiatry 42(10):1203–1211. Freeman B, Smith N, Curtis C, Huckett L, Mill J, Craig IW. 2003. DNA from buccal swabs recruited by mail: Evaluation of storage effects on longterm stability and suitability for multiplex polymerase chain reaction genotyping. Behav Genet 33(1):67–72. Goodman R. 1997. The strengths and difficulties questionnaire: A research note. J Child Psychol Psychiatry 38(5):581–586. Jawaid A, Sham PC, Makoff AJ, Asherson J. 2004. Is haplotype tagging the panacea to association mapping studies? Eur J Hum Genet 12(4):259– 262. Kustanovich V, Mcrriman B, McGough J, McCracken JT, Smalley SL, Nelson SF. 2003. Biased paternal transmission of SNAP-25 risk alleles in attention-deficit hyperactivity disorder. Mol Psychiatry 8(3):309– 315. DNA Pooling Analysis of Synaptic Component McMahon HT, Bolshakov VY, Janz R, Hammer RE, Siegelbaum SA, Sudhof TC. 1996. Synaptophysin—A major synaptic vesicle protein—Is not essential for neurotransmitter release. Proc Natl Acad Sci USA 93(10):4760–4764. Mill J, Curran S, Kent L, Gould A, Huckett L, Richards S, Taylor E, Asherson P. 2002. Association study of a SNAP-25 microsatellite and attention deficit hyperactivity disorder. Am J Med Genet 114(3):269–271. 37 Price TS, Simonoff E, Waldman I, Asherson P, Plomin R. 2001. Hyperactivity in preschool children is highly heritable. J Am Acad Child Adolesc Psychiatry 40(12):1362–1364. Rothman JE, Sollner TH. 1997. Throttles and dampers: Controlling the engine of membrane fusion. Science 276(5316):1212–1213. Sham P, Bader JS, Craig I, O’Donovan M, Owen M. 2002. DNA pooling: A tool for large scale association studies. Nat Rev Genet 3:862. Mill J, Richards S, Knight J, Curran S, Taylor E, Asherson P. 2004. Haplotype analysis of SNAP-25 suggests a role in the aetiology of ADHD. Mol Psychiatry 9(8):801–810. Trouton A, Spinath FM, Plomin R. 2002. Twins early development study (TEDS): A multivariate, longitudinal genetic investigation of language, cognition, and behavior problems in childhood. Twin Res 5(5):444–448. Norton N, Williams NM, Williams HJ, Spurlock G, Kirov G, Morris DW, Hoogendoorn B, Owen MJ, O’Donovan MC. 2002. Universal, robust, highly quantitative SNP allele frequency measurement in DNA pools. Hum Genet 110(5):471–478. Weber T, Zernelman BV, McNew JA, Westermann B, Gmachl M, Parlati F, Sollner TH, Rothman JE. 1998. SNAREpins: Minimal machinery for membrane fusion. Cell 92(6):759–772. Peter D, Finn JP, Klisak I, Liu Y, Kojis T, Heinzmann C, Roghani A, Sparkes RS, Edwards RH. 1993. Chromosomal localization of the human vesicular amine transporter genes. Genomics 18(3):720–723. Xu X, Knight J, Brookes K, Mill J, Sham P, Craig I, Taylor E, Asherson P. 2005. DNA pooling analysis of 21 norepinephrine transporter Gene SNPs with attention deficit hyperactivity disorder: No evidence for association. Am J Med Genet B Neuropsychiatr Genet 134(1):115–118.