Epigenome-wide association study of adiposity and future risk of obesity-related diseases

Int J Obes (Lond). 2018 Dec;42(12):2022-2035. doi: 10.1038/s41366-018-0064-7. Epub 2018 May 1.


Background: Obesity is an established risk factor for several common chronic diseases such as breast and colorectal cancer, metabolic and cardiovascular diseases; however, the biological basis for these relationships is not fully understood. To explore the association of obesity with these conditions, we investigated peripheral blood leucocyte (PBL) DNA methylation markers for adiposity and their contribution to risk of incident breast and colorectal cancer and myocardial infarction.

Methods: DNA methylation profiles (Illumina Infinium® HumanMethylation450 BeadChip) from 1941 individuals from four population-based European cohorts were analysed in relation to body mass index, waist circumference, waist-hip and waist-height ratio within a meta-analytical framework. In a subset of these individuals, data on genome-wide gene expression level, biomarkers of glucose and lipid metabolism were also available. Validation of methylation markers associated with all adiposity measures was performed in 358 individuals. Finally, we investigated the association of obesity-related methylation marks with breast, colorectal cancer and myocardial infarction within relevant subsets of the discovery population.

Results: We identified 40 CpG loci with methylation levels associated with at least one adiposity measure. Of these, one CpG locus (cg06500161) in ABCG1 was associated with all four adiposity measures (P = 9.07×10-8 to 3.27×10-18) and lower transcriptional activity of the full-length isoform of ABCG1 (P = 6.00×10-7), higher triglyceride levels (P = 5.37×10-9) and higher triglycerides-to-HDL cholesterol ratio (P = 1.03×10-10). Of the 40 informative and obesity-related CpG loci, two (in IL2RB and FGF18) were significantly associated with colorectal cancer (inversely, P < 1.6×10-3) and one intergenic locus on chromosome 1 was inversely associated with myocardial infarction (P < 1.25×10-3), independently of obesity and established risk factors.

Conclusion: Our results suggest that epigenetic changes, in particular altered DNA methylation patterns, may be an intermediate biomarker at the intersection of obesity and obesity-related diseases, and could offer clues as to underlying biological mechanisms.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adiposity / genetics*
  • DNA Methylation / genetics*
  • Epigenomics / methods*
  • Genetic Markers / genetics
  • Genome-Wide Association Study
  • Humans
  • Leukocytes, Mononuclear / chemistry
  • Myocardial Infarction* / epidemiology
  • Myocardial Infarction* / genetics
  • Neoplasms* / epidemiology
  • Neoplasms* / genetics
  • Obesity* / epidemiology
  • Obesity* / genetics


  • Genetic Markers