Impact of genetic risk loci for multiple sclerosis on expression of proximal genes in patients

Abstract

Despite advancements in genetic studies, it is difficult to understand and characterize the functional relevance of diseaseassociated genetic variants, especially in the context of a complex multifactorial disease such asmultiple sclerosis (MS). As a large proportion of expression quantitative trait loci (eQTLs) are context-specific, we performed RNA-Seq in peripheral bloodmononuclear cells from MS patients (n=145) to identify eQTLs in regions centered on 109MS risk single nucleotide polymorphisms and 7 associated human leukocyte antigen variants. We identified 77 statistically significant eQTL associations, including pseudogenes and non-coding RNAs. Thirty-eight out of 40 testable eQTL effects were colocalized with the disease association signal. As many eQTLs are tissue specific, we aimed to detail their significance in different cell types. Approximately 70% of the eQTLs were replicated and characterized in at least onemajor peripheral blood mononuclear cell-derived cell type. Furthermore, 40% of eQTLs were found to bemore pronounced inMS patients comparedwith non-inflammatory neurological diseases patients. In addition, we found two single nucleotide polymorphisms to be significantly associated with the proportions of three different cell types. Mapping to enhancer histone marks and predicted transcription factor binding sites added additional functional evidence for eight eQTL regions. As an example, we found that rs71624119, shared with three other autoimmune diseases and located in a primed enhancer (H3K4me1) with potential binding for STAT transcription factors, significantly associates with ANKRD55 expression. This study providesmany novel and validated targets for future functional characterization of MS and other diseases.