Ligthart · Genome biology 2016 · epigenome-wide association study meta-analysis · n=12,974

DNA methylation signatures of chronic low-grade inflammation are associated with complex diseases.

Cited 345 times in the scientific literature.

Level 3 - non-randomized controlled study

Meta-analysis of observational cohort studies (CEBM level 3).

PubMed 27955697 · doi:10.1186/s13059-016-1119-5 · record verified 2026-08-31

What was done

The authors conducted a meta-analysis of epigenome-wide association studies (EWAS) evaluating associations between peripheral blood DNA methylation and serum C-reactive protein (CRP), a marker of chronic low-grade inflammation. The discovery cohort included 8,863 individuals of European ancestry, followed by trans-ethnic replication in 4,111 African American individuals. Replicated CpG sites were further tested for associations with whole-blood gene expression in cis, genetic sequence variants, and cardiometabolic outcomes.

What was found

In the European discovery panel, differential methylation at 218 CpG sites associated with CRP (P < 1.15 x 10^-7). Fifty-eight CpG sites across 45 unique loci replicated in African Americans (P < 2.29 x 10^-4). Among replicated sites, 9 (16%) associated with cis-gene expression in whole blood (P < 8.47 x 10^-5), 10 (17%) associated with nearby genetic variants (P < 2.50 x 10^-3), and 51 (88%) associated with at least one cardiometabolic outcome (P < 9.58 x 10^-5). An additive weighted methylation score accounted for up to 6% (R2) of age- and sex-adjusted CRP variation, independent of known CRP-related genetic variants.

Why it matters

This study maps robust, trans-ethnic epigenetic loci linked to systemic inflammation and cardiometabolic disease, identifying candidate molecular pathways and targets for potential therapeutic intervention.

Limits

The associations are observational, preventing causal inference regarding whether DNA methylation alters or responds to chronic inflammation. Analysis was limited to whole blood rather than target metabolic tissues, and detailed clinical phenotypes and potential residual confounding factors beyond age and sex are not described in the abstract.

Cited by