Epigenome-Wide Analyses Identify Two Novel Associations With Recurrent Stroke in the Vitamin Intervention for Stroke Prevention Clinical Trial.

DNA methylation, a well-characterized epigenetic modification that is influenced by both environment and genetic variation, has previously been implicated in a number of complex diseases, including cardiovascular disease and stroke. The goal of this study was to evaluate epigenome-wide associations with recurrent stroke and the folate one-carbon metabolism-related trait, plasma homocysteine (hcy). Differential methylation analyses were performed on 473,864 autosomal CpG loci, using Illumina HumanMethylation 450K array data in 180 ischemic stroke cases from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial. Linear regression was used to assess associations between number of strokes prior to VISP enrollment and measures of hcy with degree of methylation (β-values), while logistic regression was used to evaluate recurrent stroke status and incident recurrent stroke associations. All regression analyses were stratified by race. Two differentially methylated CpG sites exceeded epigenome-wide significance ( ≤ 1.055 × 10) for prior number of strokes (PNS) in European Americans. The top locus, cg22812874, was located in the ankyrin repeat and SOCS box containing 10 gene (; = 3.4 × 10; β = -0.0308; 95% CI = -0.040, -0.002). Methylation locus cg00340919, located in an intron of the tetratricopeptide repeat domain 37 gene, was also statistically significant (; = 8.74 × 10; β = -0.0517; 95% CI = -0.069, -0.034). An additional 138 CpG sites met our threshold for suggestive significance ( ≤ 5 × 10). We evaluated DNA methylation associated with recurrent stroke and hcy phenotypes across the epigenome. Hypermethylation at two CpG sites located in and was associated with fewer strokes prior to VISP enrollment. Our findings present a foundation for additional epigenome-wide studies, as well as mechanistic studies into epigenetic marks that influence recurrent stroke risk.