Susceptibility genes for bipolar disorder, with focus on pleiotropy and amygdala activity

Abstract

Bipolar disorder (BD) is a common and highly heritable disorder, but few susceptibility genes have been identified and the underlying biological mechanisms remain poorly understood. Epidemiological and molecular genetic studies have provided evidence for genetic overlap between BD and other psychiatric disorders, including schizophrenia (SZ). However, we lack knowledge on which genetic variants and pathophysiological processes are involved in this overlap, and which are confined to one diagnostic category or mechanism. Recent large genome-wide association (GWA) studies have detected several novel candidate risk variants for BD. In order to confirm these findings, they should be replicated in independent samples, and their diagnostic specificity as well as neurobiological effects must be determined. Among these recently identified genetic variants are single nucleotide polymorphisms (SNPs) in the genes DGKH, PALB2, ANK3 and CACNA1C. We genotyped 37 DGKH SNPs, one PALB2 SNP, three ANK3 SNPs and one CACNA1C SNP in Nordic BD and SZ casecontrol samples. SZ cases were included for the purpose of testing for genetic overlap, and ten SNPs in the gene BRCA2 were also genotyped, as this gene is functionally related to PALB2. To test the hypothesis of increased amygdala activity as a potential genetically conditioned underlying mechanism for BD, we measured amygdala activity in a subsample of Norwegian individuals genotyped for a CACNA1C risk variant with a functional magnetic resonance imaging (fMRI) negative faces paradigm. We confirmed the association between the previously identified PALB2 SNP and BD in a meta-analysis, including our Nordic samples and international replication samples, and identified one new candidate risk SNP for BD in BRCA2. There was no significant association between these SNPs and SZ. We also replicated the association between two ANK3 SNPs and BD, but found no evidence for genetic overlap with SZ. There was no significant association between the DGKH SNPs, including one previously identified variant, and BD in our Danish and Norwegian samples. Carriers of the CACNA1C risk allele were found to have increased activity in the left amygdala, with indications of a more pronounced effect in BD cases than in SZ cases and healthy controls. Taken together, these findings further support that PALB2 and ANK3 are BD risk genes, and indicate that BRCA2 might be of interest for further investigation. As both PALB2 and BRCA2 are involved in DNA repair, this mechanism could potentially be related to the development of BD. Although we found no evidence for DGKH, further studies are needed to finally determine the role of DGKH in BD susceptibility. Our findings also support the hypothesis that increased amygdala activity is a mechanism underpinning the clinical phenotype of BD, and that this mechanism might be conditioned by the CACNA1C risk variant. Our findings also implicate ion channelopathy as a putative pathophysiological process in BD, taken into consideration that both ANK3 and CACNA1C encode for proteins related to ion channel functioning. Furthermore, the current findings indicate a partial genetic overlap between BD and SZ, as some of the variants investigated in this study were found to be more specific for BD than SZ, while other were associated with the same biological mechanism, although more prominent in BD than SZ.