5 - AN INTEGRATED GENETIC-EPIGENETIC ANALYSIS OF SCHIZOPHRENIA: EVIDENCE FOR CO-LOCALIZATION OF GENETIC ASSOCIATIONS AND DIFFERENTIAL DNA METHYLATION FROM A LARGE META-ANALYSIS OF WHOLE BLOOD DNA

Abstract

Background Schizophrenia is a severe, highly heritable, neuropsychiatric disorder characterized by episodic psychosis and altered cognitive function. Despite recent successes in identifying genetic variants robustly associated with susceptibility, there remains uncertainty about the causal genes involved in disease pathogenesis and how their function is regulated. There is growing interest in the role of developmentally regulated epigenetic variation in the molecular etiology of schizophrenia, with studies of disease-discordant monozygotic twins, clinical sample cohorts and post-mortem brain tissue identifying methylomic variation associated with disease. Leveraging on the considerable investment in genome-wide association studies (GWAS) we are examining genome-wide patterns of DNA methylation across multiple cohorts with the aim of undertaking an integrated genetic-epigenetic approach to schizophrenia. Methods DNA methylation profiled in whole blood samples from five schizophrenia case-control cohorts using the Illumina 450 K HumanMethylation array (total n = ~3,200 samples). Each sample was also genotyped and polygenic scores for schizophrenia calculated. We performed two parallel epigenome-wide association studies (EWAS) to identify methylomic variation associated with 1) schizophrenia diagnosis and 2) polygenic burden calculated from the results of latest PGC schizophrenia GWAS, combing the results from each individual cohort by meta-analysis. Results Differentially methylated positions (DMPs) associated with schizophrenia identified in each cohort showed consistent direction of effects across other datasets. Combing the results across the five cohorts, we identified 3,911 schizophrenia-associated DMPs annotated to 2,307 genes. This included further replication of 305/365 DMPs identified in our previous schizophrenia EWAS and an additional 2,434 novel DMPs. Furthermore, we identify overlap in regions characterized by differential DNA methylation and loci nominated in the largest GWAS of schizophrenia conducted by the PGC. Finally, we show how DNA methylation quantitative trait loci in combination with Bayesian co-localization analyses can be used to annotate extended genomic regions nominated by studies of schizophrenia, and to identify potential regulatory variation causally involved in disease. Discussion This study represents the largest integrated analysis of genetic and epigenetic variation in schizophrenia. Combining data from multiple cohorts has enabled us to identify novel DMPs associated with both diagnosis and elevated polygenic burden for schizophrenia. We demonstrate the utility of using a polygenic risk score to identify molecular variation associated with etiological variation, and of using DNA methylation quantitative trait loci to refine the functional and regulatory variation associated with schizophrenia risk variants. Finally, we present strong evidence for the co-localization of genetic associations for schizophrenia and differential DNA methylation.