SIR—Recently, Hamilton et al1 performed a genome-wide scan in 60 multiplex pedigrees with a ‘panic disorder syndrome’ and found the most convincing linkage evidence on 13q32–q33. This chromosomal region has received strong attention in the study of schizophrenia and bipolar affective disorder (BPAD), since genetic variants within or near to the overlapping genes D-amino-acid oxidase activator (DAOA) and G30 located on 13q33 were associated with both schizophrenia and BPAD.2, 3, 4 In this study, we hypothesized that the DAOA/G30 locus might also confer risk for the development of panic disorder (PD). Several lines of evidence support this hypothesis: firstly, DAOA/G30 is located within the linkage peak for panic disorder syndrome observed by Hamilton et al.1 Secondly, results of several clinical and epidemiological studies suggest a shared etiologic background between PD and schizophrenia, for example, Bayle et al,5 as well as between PD and BPAD, for example, MacKinnon et al.6 Thirdly, recent findings suggest the involvement of DAOA in the glutamatergic signaling pathway,2 a pathway that has been implicated in anxiogenesis.
We analyzed SNPs at the DAOA/G30 locus in a sample consisting of 152 patients with PD and 208 controls of German descent. Patients were diagnosed according to DSM-IIIR and DSM-IV on the basis of structured clinical interviews (SADS-LA7 and CIDI8) and/or a review of medical records. The control individuals were anonymous blood donors, who had been recruited independently from the controls used in our previous association study of schizophrenia and BPAD.4 We focused on SNPs rs2391191 (M15), rs3918342 (M23), and rs421292 (M24), as these had shown the strongest association in our previous work.4 Additionally, we typed two flanking SNPs, rs778293 (M22) and rs3916972 (M25), covering this chromosomal region. Genotyping was carried out using the MassARRAY system (Sequenom, San Diego, USA) according to the genotyping method described by Ding and Cantor,9 with minor modifications (see Supplementary table). We performed both single-marker and haplotype analyses with the program COCAPHASE 2.40.10
Of the five variants at the DAOA/G30 locus typed, three SNPs (rs778293, rs3918342, and rs1421292) proved to be in significant LD with PD (Table 1), with the strongest evidence being observed for rs1421292 (P=0.031, OR 1.39). At the genotypic level, significant association was found between PD patients and rs1421292 under a dominant model (P=0.024, OR 2.12, Table 1). Given that a Bonferroni correction to adjust for multiple testing of loci would be overly conservative, and given the high levels of LD of these closely related markers, we used a permutation procedure instead. Our best P-value in the allelewise analysis was 0.031. In our simulation, a P-value of 0.031 or smaller would be seen by chance alone in 1129 times out of 10 000 replicates, yielding an empirical P-value of 0.113. In a subsequent step, we performed haplotype analysis (data not shown) comprising SNPs rs778293, rs3918342, and rs1421292, which were associated with disease in the single-locus analysis. The three-marker haplotype T–T–A was significantly less frequent in patients with PD (40 vs 48% in controls, P=0.030), and the haplotype C–C–T was more frequent in PD patients (42 vs 35% in controls, P=0.072). Although far from being significant (P-values >0.1), it is interesting to note that the allelic and haplotypic associations seem to be stronger in PD than in schizophrenia and BPAD. For example, the same alleles of rs3918342 and rs1421292 previously shown to be associated with schizophrenia and BPAD4 were even more frequent in patients with PD (rs3918342 allele C: 53% in schizophrenia and BPAD,4 57% in PD; rs1421292 allele T: 56 and 55% in schizophrenia and BPAD,4 respectively, 60% in PD).
Despite the fact that the observed single-marker associations do not withstand a stringent and robust adjustment for multiple testing at a nominal level of significance, our results suggest that variability at the DAOA/G30 locus may be involved in the etiology of PD. Since we have previously observed the same alleles and haplotypes to be associated with disease in our study on BPAD and schizophrenia, we believe that our findings are of potential interest for the psychiatric genetic community and may give a new perspective on conventional diagnostic categories.
Hamilton SP et al. Further genetic evidence for a panic disorder syndrome mapping to chromosome 13q. Proc Natl Acad Sci USA 2003; 100: 2550–2555.
Chumakov I et al. 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 2002; 99: 13675–13680.
Hattori E et al. Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. Am J Hum Genet 2003; 72: 1131–1140.
Schumacher J et al. Examination of G72 and D-amino-acid oxidase as genetic risk factors for schizophrenia and bipolar affective disorder. Mol Psychiatry 2004; 9: 203–207.
Bayle FJ et al. Clinical features of panic attacks in schizophrenia. Eur Psychiatry 2001; 16: 349–353.
MacKinnon DF et al. Association of rapid mood switching with panic disorder and familial panic risk in familial bipolar disorder. Am J Psychiatry 2003; 160: 1696–1698.
Mannuzza S et al. Schedule for affective disorders and schizophrenia—lifetime version modified for the study of anxiety disorders (SADS-LA): rationale and conceptual development. J Psychiatr Res 1986; 20: 317–325.
Robins LN et al. The composite international diagnostic interview. Arch Gen Psychiat 1988; 45: 1069–1077.
Ding C, Cantor CR . Direct molecular haplotyping of long-range genomic DNA with M1-PCR. Proc Natl Acad Sci USA 2003; 13: 7449–7455.
Dudbridge F . Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol 2003; 25: 115–121.
About this article
AIMS Genetics (2018)
Psychiatry Research (2017)
Molecular Psychiatry (2016)
Psychiatric Genetics (2016)
European Archives of Psychiatry and Clinical Neuroscience (2015)