Epigenetic patterns in blood associated with lipid traits predict incident coronary heart disease events and are enriched for results from genome-wide association studies

摘要

Background Genome-wide association studies (GWAS) have identified loci influencing circulating lipid concentrations in humans; further information on novel contributing genes, pathways and biology may be gained through studies of epigenetic modifications. Methods and Results To identify epigenetic changes associated with lipid concentrations, we assayed genome-wide DNA methylation at cytosine-guanine dinucleotides (CpGs) in whole blood from 2,306 individuals from two population-based cohorts, with replication of findings in 2,025 additional individuals. We identified 193 CpGs associated with lipid levels in the discovery stage (P 1.08E-07) and replicated 33 (at Bonferroni-corrected P0.05), including 25 novel CpGs not previously associated with lipids. Genes at lipid-associated CpGs were enriched in lipid and amino acid metabolism processes. A differentially methylated locus associated with TG and HDL-C (cg27243685; P= 8.1E-26 and 9.3E-19) was associated with cis-expression of a reverse-cholesterol transporter (ABCG1; P=7.2E-28), and incident cardiovascular disease events (HR per SD increment = 1.38, 95% CI, 1.15-1.66, P=0.0007). We found significant cis methylation quantitative trait loci (cis-meQTLs) at 64% of the 193 CpGs with an enrichment of signals from GWAS of lipid levels (PTC =0.004, PHDL-C=0.008 and PTG=0.00003) and coronary heart disease (P=0.0007). For example, genome-wide significant variants associated with LDL-C and coronary heart disease at APOB were cis-meQTLs for a LDL-C-related differentially methylated locus. Conclusions We report novel associations of DNA methylation with lipid levels, describe epigenetic mechanisms related to previous GWAS discoveries and provide evidence implicating epigenetic regulation of reverse cholesterol transport in blood in relation to occurrence of cardiovascular disease events. Keywords: lipids, cardiovascular diseases, epigenetics, gene expression/regulation, genome-wide analysis