Identification of genetic variants for suicidal behaviors in patients with bipolar or major depressive disorders

[[abstract]]Background: Suicide is the second leading cause of death in young adults, and is accounted for 1.4 percent of all deaths worldwide. More than 90% of suicide attempters or victims had been diagnosed with psychiatric disorders, particularly Major Depressive Disorder (MDD) and Bipolar Disorder (BD). Suicidal behaviour has substantial heritability, however, the underlying etiology of patients diagnosed with different mood disorders may not identical. In the current study, we adopted the Genome-Wide Association Study design (GWAS) to identify genetic variants associated with suicidal behaviors in patients with BD or MDD. Our goal is to identify possible common and unique genetic variants that might be associated to suicidal behavior between the two patient groups. We further test the predictive ability for suicide in the MDD group by using the calculated Polygenic Risk Score (PRS) for suicide in BD patients and vice versa. Methods: We recruited 662 BD and 428 MDD patients in Taiwan, who were referred by psychiatrists based on DSM-IV criteria. BD patients were interviewed to collect information of demographic, clinical characteristics, and suicidal behaviors. In addition, the suicide item in Hamilton depression rating scale was used to obtain suicide information for MDD patients. Genotyping was conducted using Affymetrix CHB1 chip for BD, and Illumina Human Omni Express Exome bead chip for MDD with imputation for association analyses. Genetic association and PRS analyses were employed by Plink v1.07. The signi fi cance level was set at 5E-08 for adjustment of multiple comparisons, and a p-value less than 1E-06 was considered suggestive association. Results: In our samples, 45% of BD and 30% of MDD subjects were male. The onset age for BP and MDD were 18.24 and 38.04 years respectively. The prevalence of suicide attempt was 35% in BD group and 43% in MDD group. There were no loci that reached the genome-wide signi fi cance level for suicidal behavior in BP patients. The most signi fi cant marker was rs34945506 (p = 4.13E-06) located in RCSD1 gene on chromosome 1. In MDD group, marker rs76015177 (p = 9.94E-07) showed suggestive association with suicidal behavior. This marker is located on chromosome 9 and is not mapped to any coding gene. There were no common SNPs associated with suicidal behavior in the MDD and BP groups. Discussion: Genetic variants that showed suggestive signals with suicidal behavior in the present study were not reported in previous studies. However, due to limited sample size, the power of this study was only moderate, and large-scale replication studies are needed for further validation. One mapped gene, RCSD1, has been reported to regulate the ability of F-actin-capping protein to remodel actin fi lament assembly via stress-induced phosphorylation. Actin was found to be related to some neurodegeneration diseases, such as Alzheimer's disease. For MDD, two suggestive markers (rs78434258 and rs79070953) on chromosome 9 are found near to the most signi fi cant marker (rs76015177) were located on LOC105375976 gene, which is a non-coding RNA. However, the function of this gene is mostly unknown. Lastly, the non-overlapping genetic results for suicidal behaviors in the MDD and BP patients may indicate different pathogenesis and biological pathways for suicide by different diagnosis groups