(M1230-10-56) A Targeted Approach to Studying DNA Methylation in Individuals with Bipolar Disorder and Suicidal Behavior

Purpose: Genome-wide association studies (GWAS) aim to understand the genetic component of complex disease phenotypes, such as suicidal behavior, which do not obey the standard Mendelian inheritance rules. However, the risk factors for suicidal behavior cannot be explained only by the genetic variants identified by GWAS studies. Both genetic variants and environmental factors play a role in developing complex diseases. The interaction between genetic variants (SNPs) and the environment can be translated as epigenetic modifications, leading to gene expression changes without changing the underlying DNA sequence. DNA methylation is a common epigenetic modification that can change the phenotypic outcome of SNPs. Thus, to fully understand the functional role of genetic variants that are associated with suicide, it is essential to study the change in DNA methylation of the CpG sites adjacent to SNPs associated with suicide. In this project, we aim to study DNA methylation close to SNPs associated with suicidal behavior in individuals with bipolar disorder to develop a DNA methylation-based biomarker.

Methods: We performed extensive research of the literature to identify SNPs (N=24) that showed a significant association with suicidal behavior in individuals with bipolar disorder. After identifying these variants, we generated a customized panel that covered our chosen genetic variants and adjacent CpG sites. We profiled DNA methylation patterns of blood samples collected from bipolar disorder participants with suicidal behavior (BDSB; n = 55) and without suicidal behavior (BDNSB; n= 51) using bisulfite amplicon sequencing. We conducted differential methylation analysis at the level of one CpG and region level using a logistic regression model to fit the count of methylated and unmethylated reads as the outcome and sample groups as the predictor considering age, sex, race, and smoking history as covariates. Additionally, we tested whether suicide-related genotypes explained the DNA methylation phenotype by conducting a methylation-eQTL analysis. We have genotype data for the following SNPs for 112 subjects: rs300774, rs75981117, rs6265, rs11130703, rs2766533, rs30079, rs669838, and rs924134. We used a linear model that evaluated whether the number of alleles (coded 0, 1, and 2) predicted DNA methylation (log (percentage percent +1) at each site, considering covariates for age, sex, smoking history, and the first two principal components from the genotype data to account for ethnicity differences. We also tested the correlation between the lethality and number of lifetime suicide attempts with the DNA methylation percentage of 88 sites for BDSB subjects (n=49).

Results: Significant DNA methylation changes in BD individuals with a history of suicide attempts - There are a total of 7 CpGs (Table 1) and 5 regions (Table 2) that only significantly differed in DNA methylation levels in the BDSB group (corrected p-value < 0.05). These sites and regions are located in genes involved in modulating stress response and the immune system. Significant methyl-eQTLs - In total, we have 30 cis methyl-eQTL pairs, and 2 CpG-SNP pairs showed a significant result (p-value < 0.05). The top cis SNP is rs2766533 which showed the strongest association to the methylation level of genomic location chr6.35717737. We found that higher DNA methylation at chr6.35717737 was associated with increasing the copies of the minor allele (A). rs2766533 is located downstream of chr6.35717737 in intron 2 of the FKBP5 gene. This finding provides additional support for the role of FKBP5 variants in modulating the gene function through changing DNA methylation patterns. CpG methylation correlated with the lethality and the number of suicide attempts - We also found that the lethality and the number of suicide attempts are associated with the methylation of CpG sites close to SNPs linked to suicide (corrected p-value < 0.05).

Conclusion: Genetic risk variants associated with suicidal behavior in BD groups are also associated with changes in DNA methylation. In BD individuals, the methylation of CpG sites near SNPs associated with suicidal behavior is correlated with both the lethality and the frequency of suicide attempts. This study provides evidence of the relationship between genetic and epigenetic variation in the context of suicidal behavior in bipolar disorder.

Acknowledgments: My advisor: Dr. Marie Gaine
The Bipolar Disorder Research Program of Excellence, UIOWA.
Pharmaceutical Sciences and Experimental Therapeutics (PSET) Department, UIOWA.
Study Participants

List of the distinct CpG sites that showed significant difference in methylation only in BD with history of suicide attempt vs the BD with no history of suicide attempt



List of the distinct regions that showed significant difference in methylation only in BD with history of suicide attempt vs the BD with no history of suicide attempt.