American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:914 –917 (2008) Association of DAO and G72(DAOA)/G30 Genes With Bipolar Affective Disorder Diana Prata,1,2 Gerome Breen,1,2 Sarah Osborne,2 Janet Munro,2 David St. Clair,3 and David Collier1,2* 1 Social Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, London, UK Division of Psychological Medicine, Institute of Psychiatry, London, UK 3 Department of Mental Health, University of Aberdeen, Aberdeen, UK 2 There is growing evidence of partial aetiological overlap between schizophrenia and bipolar disorder (BP) from linkage analysis, genetic epidemiology and molecular genetics studies. In the present study we investigated whether individual polymorphisms or haplotypes of the DAO and G72(DAOA)/G30 genes, which have been previously implicated in schizophrenia, are also associated with bipolar disorder. For each gene, we genotyped 213 cases and 197 controls for SNPs previously associated with schizophrenia: rs2111902 (MDAAO-4), rs3918346 (MDAAO-5), rs3741775 (MDAAO-6) and rs3918347 (MDAAO-7) in DAO and rs746187 (M7), rs3916966 (M13), rs2391191 (M15) and rs3916972 (M25) in G72. Although none of the individual SNPs in these genes reached statistical significance, we found haplotype wise associations with bipolar disorder for both genes. These included a two-SNP haplotype in DAO (rs2111902-A and rs3918346-T; global P ¼ 0.003, individual P ¼ 0.002, Z ¼ 3.1) and a twoSNP haplotype for G72(DAOA)/G30 (rs746187-G and rs3916972-G; global P ¼ 0.05; individual P ¼ 0.005, Z ¼ 2.81). However, we found no evidence for an epistatic interaction between the SNPs and/or haplotypes of the two genes. In summary, our findings provide some support for the individual involvement of DAO and G72(DAOA)/G30 in the etiology of bipolar disorder. ß 2007 Wiley-Liss, Inc. KEY WORDS: bipolar disorder; DAO; G72; SNPs; haplotypes Please cite this article as follows: Prata D, Breen G, Osborne S, Munro J, St. Clair D, Collier D. 2008. Association of DAO and G72(DAOA)/G30 Genes With Bipolar Affective Disorder. Am J Med Genet Part B 147B:914–917. INTRODUCTION Twin studies indicate a substantial role for genetic factors in bipolar affective disorder with heritability of 80–90% [Craddock and Forty, 2006]. This is a similar level of heritability to schizophrenia and schizoaffective disorder *Correspondence to: David Collier, Section of Molecular Genetics, Division of Psychological Medicine, Institute of Psychiatry, London SE5 8AF, UK. E-mail: firstname.lastname@example.org Received 21 June 2007; Accepted 30 October 2007 DOI 10.1002/ajmg.b.30682 ß 2007 Wiley-Liss, Inc. [Cardno et al., 1999]. Over the last decade family-based linkage analysis has provided evidence for a locus for schizophrenia on chromosome 13q, over a broad region of about 70 cM [Detera-Wadleigh and McMahon, 2006]. Suggestive linkage was first reported by Lin et al. . Since there have been a series of both positive [e.g., Blouin et al., 1998; Brzustowicz et al., 1999] and negative linkage studies, which culminated in two important meta analyses, one positive, using the multiple scan probability method [MSP; Badner and Gershon, 2002], and another which was negative, using genome scan meta-analysis (GSMA), a rank-order method [Lewis et al., 2003]. Linkage evidence for chromosome 13q is also suggestive in BP: one of three meta-analyses showed significant association [Badner and Gershon, 2002]. Overall, it seems probable but not certain that there is a locus for schizophrenia and for bipolar affective disorder on chromosome 13q [Detera-Wadleigh and McMahon, 2006]. A number of candidate genes have been examined on chromosome 13q, but only one systematic study has examined the locus [Chumakov et al., 2002], which used homozygosity mapping and case–control association in a French-Canadian isolate to identify two distinct clusters in 13q32–33 of schizophrenia-associated SNPs. These mapped to novel transcripts, G72 and G30, which are plausible candidate genes. Most subsequent effort has focused on G72, a human-specific gene expressed in the caudate and amygdala [Chumakov et al., 2002] with excess expression in the dorso-lateral pre-frontal cortex of schizophrenics [Korostishevsky et al., 2004]. Interactive cloning using yeast two-hybrid analysis, revealed that G72’s protein product binds to and activates D-amino acid oxidase (DAO), which is also expressed in the human brain [Chumakov et al., 2002]. Hence, G72 has since been referred to as D-amino-acid oxidase activator (DAOA). DAO maps to 12q24, a region with some evidence of linkage to schizophrenia and bipolar disorder [Sklar, 2002]. Subsequently association between bipolar affective disorder and the G72(DAOA)/G30 locus has also been reported with in two North American family samples [Hattori et al., 2003] and a further North American family sample [Chen et al., 2004], a German case–control sample [Schumacher et al., 2004] and a UK case–control sample [Williams et al., 2006]. In all studies, evidence for association came from individual SNPs as well as multilocus haplotypes, although there is no consensus about the specific risk alleles or haplotypes across studies. Further studies of schizophrenia have also shown association with G72(DAOA)/G30 [Hall et al., 2004; Korostishevsky et al., 2004; Wang et al., 2004]. Recently, a meta-analysis [DeteraWadleigh and McMahon, 2006] of 18 individual SNPs from 10 association studies found evidence for association of G72 (DAOA)/G30 with schizophrenia and, to a lesser extent with bipolar disorder. A second meta-analysis by Li and He  of 16 polymorphisms in schizophrenia also found evidence for association, albeit weak. The largest study to date was of 2831 individuals, of whom 709 had DSMIV schizophrenia, 706 had bipolar-I disorder, and 1416 were controls [Williams et al., 2006]. This study suggested that G72(DAOA)/G30 does Association of DAO and G72(DAOA)/G30 Genes not contribute to the susceptibility to psychosis per se, but to episodes of mood disorder across the bipolar and schizophrenia spectrum. DAO has been examined in one German case–control study of bipolar disorder [Schumacher et al., 2004] which found no association. However, Fallin et al.  reported a significant association of this gene with bipolar affective disorder in an Ashkenazi Jewish case-parent trios design, although not with schizophrenia. Given the evidence for the involvement of its interacting partner, G G72(DAOA)/G3072, in psychosis and mood disorders, DAO warrants further investigation in bipolar disorder. Thus we set out to investigate its effect in a Scottish sample of cases with bipolar I disorder and controls. We also tried to detect a gene-by-gene interaction as suggested by Chumakov et al. . MATERIALS AND METHODS Subjects Our sample consisted of 213 bipolar disorder I patients, and 197 healthy controls. Subjects were recruited through psychiatric hospitals in Scotland, met DSM-IV criteria for bipolar I disorder, and had been receiving lithium therapy for at least 3 years. All subjects were Caucasian with at least three of their four grandparents born in Scotland. Consensus diagnosis was determined by trained psychologists and psychiatrists based on case-note review and clinical interview using semistructured diagnostic questionnaires. The Operational Criteria Checklist for Psychotic Illness program was used to define diagnoses. Patients with schizoaffective disorder were excluded. Final decisions on any diagnostic disagreements ([N ¼ 31] for reasons of missing data or discrepancies between DSM-IV and International Classification of Diseases-10 diagnoses [N ¼ 31]) were made by one of the authors (D.St.C.) and corroborated by another (R.S.). The sample has 97% power to detect a heterozygote odds ratio of 2 at the 0.05 level for a risk allele of frequency 0.19. Power analysis was performed using the genetic power calculator (http://statgen.iop.kcl.ac.uk/gpc/). Genotyping We genotyped four SNPs in the DAO gene used in the original study of Chumakov et al. : rs2111902 (alleles G or T; alias MDAAO-4), rs3918346 (C/T; MDAAO-5), rs3741775 (G/T; MDAAO-6), rs3918347 (A/G; MDAAO-7). rs2111902 occurs in an untranslated mRNA region and the remaining three SNPS in introns. In the G72(DAOA)/G30 gene, we genotyped the four SNPs originally analyzed by Chumakov et al. : rs746187 (C/T; alias M7), rs3916966 (A/C; M13), rs2391191 (A/G; M15); and by Hall et al. , rs3916972 (G/ T; M25). rs2391191 is in an untranslated mRNA region and the remaining three in regions of uncertain function. Genotyping was performed under contract by KBioscience (http://www.kbioscience.co.uk) using a competitive allele specific PCR system (CASP). Quality control criteria were that genotypes form three distinct clusters, water controls were negative, the number of callable genotypes was higher than 90% and minor allele frequency was greater than 0.02. In addition interplate and intraplate duplicate testing of known DNAs was performed. All genotyping was performed sequentially by the same method and blind to status. For the genotype and allele-wise statistical analysis we used both w2 and Fisher’s Exact Test to analyze the individual polymorphisms. GENECOUNTING software (version 1.3, March 2003; Zhao et al., 2002] was used to estimate the haplotype frequencies from the genotype data (with 1000 permutations/simulations) and Cramer’s V and Absolute D0 to assess linkage disequilibrium (LD) between the SNPs within each gene. To test for interaction between the two genes, we used GAIA (Genetic Association Interaction Analysis) software [Macgregor and Khan, 2006] as well as WHAP version 2.06 [Purcell et al., 2007]. RESULTS All SNPs were in Hardy–Weinberg equilibrium. Pairwise LD analysis between the markers in each gene, with Cramer’s V and Absolute D0 tests showed the SNPs within each gene were in strong LD (Table I). Analysis of each SNP individually showed no significant association (data not shown). However, one SNP within the G72(DAOA)/G30 gene, rs746187, showed a trend for a higher frequency of the G allele in the cases group (P ¼ 0.07; OR ¼ 1.31). Haplotype analysis revealed a two-SNP haplotype of rs2111902 and rs3918346 of DAO to be globally significant at P ¼ 0.003, the haplotypes A–T being the most probable risk haplotype at an individual P ¼ 0.002 (Z-score 3.1). Adding allele T of rs3741775 or the A allele of rs3918347 to this haplotype still gave a significant association of global P ¼ 0.01 and P ¼ 0.02, respectively, and a specific P-value of 0.002 (both Z-score 3.1) for each individual haplotype (Table II). For G72(DAOA)/G30, we found borderline significance for a twoSNP haplotype composed of rs746187 and rs3916972 (global P ¼ 0.05) but an individual P ¼ 0.005 (Z score 2.81) for the G–G haplotype (Table II). We then tested for interaction between DAO and G72(DAOA)/G30. Using GAIA software, we investigated the presence of additive and multiplicative interaction between any of the SNPs of each gene. With WHAP, we attempted to TABLE I. Pairwise LD Analysis (Cramer’s V and Absolute D0 ) Between the Markers in Each Gene Absolute D0 DAO 1 rs2111902 2 rs3918346 3 rs3741775 4 rs3918347 Cramer’s V G72(DAOA)/G30 1 rs746187 2 rs3916966 3 rs2391191 4 rs3916972 a ns. 1 rs2111902 2 rs3918346 3 rs3741775 4 rs3918347 0.80 0.40 0.98 0.62 0.96 0.57 0.69 0.22 0.59 0.46 0.87 0.30 1 rs746187 2 rs3916966 3 rs2391191 4 rs3916972 0.02 0.01 1.00 0.004 0.06 0.06 0.02a 0.01a 0.002a 915 0.99 0.04a 0.04a 916 Prata et al. TABLE II. Haplotype Association Between DAO Bipolar Affective Disorder Global P-value Risk haplotypes 14.94, df ¼ 3 16.76, df ¼ 7 15.22, df ¼ 6 0.003 0.01 0.02 2–2: A–T 2–2–1: A–T–T 2–2–1: A–T–A 0.047/0.011 0.048/0.011 0.047/0.011 3.07 3.05 3.08 0.002 0.002 0.002 6.64, df ¼ 3 0.05 1–1: G–G 0.182/0.110 2.81 0.005 w2 SNPs DAO 1, 2 1, 2, 3 1, 2, 4 G72(DAOA)/G30 1, 4 Cases/controls Z-score Specific P-value 2 Presented is the w statistic text and the correspondent global P-value as well as the frequency and P-values for the specific risk haplotypes. detect an interactive effect by analyzing the significant haplotype of DAO while covarying for risk genotypes of G72(DAOA)/G30, assuming an additive model. We found no indication of epistatic effects. DISCUSSION Our results support a contribution of DAO to the development of bipolar disorder, which is in line with previous linkage findings [Sklar, 2002]. We found significant association of twoand three-SNP risk haplotypes with the disorder in our case– control design. The original study of Chumakov et al.  found schizophrenia to be associated with all the four SNPs investigated here. So far, two published association studies have investigated this gene in bipolar disorder [Schumacher et al., 2004; Fallin et al., 2005]. Schumacher et al. used three of the four SNPs (rs2111902, rs3918346, and rs3741775) analyzed in the present study. They found a significant association for each of the individual SNPs and the three-SNP haplotype with schizophrenia, but not with bipolar disorder. The family trio study of Fallin et al.  found this gene to be associated with bipolar disorder in single-SNP and haplotype-based transmission disequilibrium tests, using other SNPs as part of two LD blocks spanning roughly half of the gene. We detected a significant association for that three SNP haplotype (A–T–T) and a stronger association for the two-SNP rs2111902– rs3918346 haplotype (A–T) with bipolar disorder. We found the rs2111902 risk allele (A), to be the same as the schizophrenia risk allele in Schumacher et al. , but not the rs3918346 or rs3741775 alleles which were C and G respectively in that study. The associated three-marker haplotype in Chumakov et al.  was C–T–T. In contrast to DAO, a body of evidence provides compelling evidence of association between G72(DAOA)/G30 with bipolar disorder [Detera-Wadleigh and McMahon, 2006]. Although there may be difficulties with the haplotypic association approach, our association replicates previous positive findings of association with G72(DAOA)/G30 in the form of a two-SNP haplotype of rs746187-G and rs3916972-G. Across studies of both bipolar disorder and schizophrenia, the specific associated alleles of nearly every marker have varied, as shown in DeteraWadleigh and McMahon , which could reflect the possibility that multiple, possibly rare disease alleles exist across different populations. From our set of four SNPs in G72(DAOA)/G30, we did not find a significant individual SNP association with bipolar disorder, even thought SNP rs2391191 (M15) had shown strong significance of association (P ¼ 0.0006) with schizophrenia in the meta-analysis of Detera-Wadleigh and McMahon . In terms of haplotype association, these four SNPs might mark a significantly undertransmitted (possibly protective) haplotype (M7–M12–M12– M13–M14–M15–M23-–M25) in schizophrenia as found by Hall et al.  in South African and USA samples. Within it, a core haplotype flanked by M12 and M15 has also been found to be over-transmitted in schizophrenia with complimentary alleles, in the French Canadian sample of Chumakov et al. . In our Scottish bipolar sample, we found a significant over-representation of M7–M25 haplotype, for the G–G allelic version, which is under-transmitted in schizophrenic in the USA schizophrenia sample of Hall et al. . It remains unclear why different (although closely related) haplotypes show association across populations, although in our case, this may represent a difference in the genetic etiology of each disorder. The fact that our sample did not show any gene by gene interaction is possibly due to power limitations that did not allow us to detect small rare allelic effects or to lack of strong association with the putative risk polymorphisms in the gene. CONCLUSION Our study provides additional evidence for an involvement of G72(DAOA)/G30 and DAO in the etiology of bipolar disorder, but does not support epistatic interaction. Given that, to date, more studies have investigated G72(DAOA)/G30 and DAO in schizophrenia than in bipolar disorder, a larger pool of data, sufficient for meta-analysis, will be required for the association of these genes with bipolar disorder to be definitive. ACKNOWLEDGMENTS We thank FCT—Fundacao para a Ciencia e Tecnologia (Portugal) for sponsoring the genotyping, and the SIM foundation, the University of Aberdeen (UK) and the patient and control volunteers for the case–control sample. REFERENCES Badner JA, Gershon ES. 2002. Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Mol Psychiatry 7:405–411. Blouin JL, Dombroski BA, Nath SK, Lasseter VK, Wolyniec PS, Nestadt G, et al. 1998. Schizophrenia susceptibility loci on chromosomes 13q32 and 8p21. Nat Genet 20:70–73. Brzustowicz LM, Honer WG, Chow EW, Little D, Hogan J, Hodgkinson K, et al. 1999. Linkage of familial schizophrenia to chromosome 13q32. Am J Hum Genet 65:1096–1103. Cardno AG, Marshall EJ, Coid B, Macdonald AM, Ribchester TR, Davies NJ, Venturi P, Jones LA, Lewis SW, Sham PC, Gottesman II, Farmer AE, McGuffin P, Reveley AM, Murray RM. 1999. Heritability estimates for psychotic disorders. Arch Gen Psychiatry 56:162–168. Chen YS, Akula N, Detera-Wadleigh SD, Schulze TG, Thomas J, Potash JB, DePaulo JR, McInnis MG, Cox NJ, McMahon FJ. 2004. Findings in an independent sample support an association between bipolar affective disorder and the G72G30 locus on chromosome 13q33. Mol Psychiatry 91:87–92. Chumakov I, Blumenfeld M, Guerassimenko O, Cavarec L, Palicio M, Abderrahim H, Bougueleret L, Barry C, Tanaka H, La Rosa P, Puech A, Tahri N, Cohen-Akenine A, Delabrosse S, Lissarrague S, Picard FP, Maurice K, Essioux L, Millasseau P, Grel P, Debailleul V, Simon AM, Caterina D, Dufaure I, Malekzadeh K, Belova M, Luan JJ, Bouillot M, Sambucy JL, Primas G, Saumier M, Boubkiri N, Martin-Saumier S, Nasroune M, Peixoto H, Delaye A, Pinchot V, Bastucci M, Guillou S, Association of DAO and G72(DAOA)/G30 Genes Chevillon M, Sainz-Fuertes R, Meguenni S, Aurich-Costa J, Cherif D, Gimalac A, Van Duijn C, Gauvreau D, Ouellette G, Fortier I, Raelson J, Sherbatich T, Riazanskaia N, Rogaev E, Raeymaekers P, Aerssens J, Konings F, Luyten W, Macciardi F, Sham PC, Straub RE, Weinberger DR, Cohen N, Cohen D. 2002. Genetic and physiological data implicating the new human gene G72 and the gene for D-amino acid oxidase in schizophrenia. Proc Natl Acad Sci USA 99:13675–13680. Craddock N, Forty L. 2006. Genetics of affective (mood) disorders. Eur J Hum Genet 14:660–668. Detera-Wadleigh SD, McMahon FJ. 2006. G72G30 in schizophrenia and bipolar disorder: Review and meta-analysis. Biol Psychiatry 60:106– 114. Fallin MD, Lasseter VK, Avramopoulos D, Nicodemus KK, Wolyniec PS, McGrath JA, Steel G, Nestadt G, Liang KY, Huganir RL, Valle D, Pulver AE. 2005. Bipolar I disorder and schizophrenia: A 440-single-nucleotide polymorphism screen of 64 candidate genes among Ashkenazi Jewish case-parent trios. Am J Hum Genet 77(6):918–936. Hall D, Gogos JA, Karayiorgou M. 2004. The contribution of three strong candidate schizophrenia susceptibility genes in demographically distinct populations. Genes Brain Behav 3:240–248. Hattori E, Liu C, Badner JA, Bonner TI, Christian SL, Maheshwari M, Detera-Wadleigh SD, Gibbs RA, Gershon ES. 2003. Polymorphisms at the G72G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. Am J Hum Genet 725:1131–1140. Korostishevsky M, Kaganovich M, Cholostoy A, Ashkenazi M, Ratner Y, Dahary D, Bernstein J, Bening-Abu-Shach U, Ben-Asher E, Lancet D, Ritsner M, Navon R. 2004. Is the G72G30 locus associated with schizophrenia? Single nucleotide polymorphisms, haplotypes, and gene expression analysis. Biol Psychiatry 56:3169–3176. Lewis CM, Levinson DF, Wise LH, DeLisi LE, Straub RE, Hovatta I, Williams NM, Schwab SG, Pulver AE, Faraone SV, Brzustowicz LM, Kaufmann CA, Garver DL, Gurling HM, Lindholm E, Coon H, Moises HW, Byerley W, Shaw SH, Mesen A, Sherrington R, O’Neill FA, Walsh D, Kendler KS, Ekelund J, Paunio T, Lonnqvist J, Peltonen L, O’Donovan MC, Owen MJ, Wildenauer DB, Maier W, Nestadt G, Blouin JL, 917 Antonarakis SE, Mowry BJ, Silverman JM, Crowe RR, Cloninger CR, Tsuang MT, Malaspina D, Harkavy-Friedman JM, Svrakic DM, Bassett AS, Holcomb J, Kalsi G, McQuillin A, Brynjolfson J, Sigmundsson T, Petursson H, Jazin E, Zoega T, Helgason T. 2003. Genome scan metaanalysis of schizophrenia and bipolar disorder. Part II. Schizophrenia. Am J Hum Genet 73:34–48. Li D, He L. 2007. G72/G30 genes and schizophrenia: A systematic metaanalysis of association studies. Genetics 175:917–922. Lin MW, Sham P, Hwu HG, Collier D, Murray R, Powell JF. 1997. Suggestive evidence for linkage of schizophrenia to markers on chromosome 13 in Caucasian but not Oriental populations. Hum Genet 99:417–420. Macgregor S, Khan I. 2006. GAIA: An easy-to-use web-based application for interaction analysis of case-control data. BMC Med Genet 7:34. Purcell S, Daly MJ, Sham PC. 2007. WHAP: Haplotype-based association analysis. Bioinformatics 23:255–256. Schumacher J, Jamra RA, Freudenberg J, Becker T, Ohlraun S, Otte AC, Tullius M, Kovalenko S, Bogaert AV, Maier W, Rietschel M, Propping P, Nothen MM, Cichon S. 2004. Examination of G72 and D-amino-acid oxidase as genetic risk factors for schizophrenia and bipolar affective disorder. Mol Psychiatry 92:203–237. Sklar P. 2002. Linkage analysis in psychiatric disorders: The emerging picture. Annu Rev Genomics Hum Genet 3:371–413. Review Wang Z, Wei J, Zhang X, Guo Y, Xu Q, Liu S, Shi J, Yu Y, Ju G, Li Y, Shen Y. 2004. Association of G72G30 with schizophrenia in the Chinese population. Biochem Biophys Res Commun 319:1281–1286. Williams NM, Green EK, Macgregor S, Dwyer S, Norton N, Williams H, Raybould R, Grozeva D, Hamshere M, Zammit S, Jones L, Cardno A, Kirov G, Jones I, O’Donovan MC, Owen MJ, Craddock N. 2006. Variation at the DAOAG30 locus influences susceptibility to major mood episodes but not psychosis in schizophrenia and bipolar disorder. Arch Gen Psychiatry 634:366–373. Zhao JH, Lissarrague S, Essioux L, Sham PC. 2002. GENECOUNTING: Haplotype analysis with missing genotypes. Bioinformatics 18:1694– 1695.