Blood lipids influence DNA methylation in circulating cells.
Dekkers KF., van Iterson M., Slieker RC., Moed MH., Bonder MJ., van Galen M., Mei H., Zhernakova DV., van den Berg LH., Deelen J., van Dongen J., van Heemst D., Hofman A., Hottenga JJ., van der Kallen CJH., Schalkwijk CG., Stehouwer CDA., Tigchelaar EF., Uitterlinden AG., Willemsen G., Zhernakova A., Franke L., 't Hoen PAC., Jansen R., van Meurs J., Boomsma DI., van Duijn CM., van Greevenbroek MMJ., Veldink JH., Wijmenga C., BIOS Consortium None., van Zwet EW., Slagboom PE., Jukema JW., Heijmans BT.
BACKGROUND: Cells can be primed by external stimuli to obtain a long-term epigenetic memory. We hypothesize that long-term exposure to elevated blood lipids can prime circulating immune cells through changes in DNA methylation, a process that may contribute to the development of atherosclerosis. To interrogate the causal relationship between triglyceride, low-density lipoprotein (LDL) cholesterol, and high-density lipoprotein (HDL) cholesterol levels and genome-wide DNA methylation while excluding confounding and pleiotropy, we perform a stepwise Mendelian randomization analysis in whole blood of 3296 individuals. RESULTS: This analysis shows that differential methylation is the consequence of inter-individual variation in blood lipid levels and not vice versa. Specifically, we observe an effect of triglycerides on DNA methylation at three CpGs, of LDL cholesterol at one CpG, and of HDL cholesterol at two CpGs using multivariable Mendelian randomization. Using RNA-seq data available for a large subset of individuals (N = 2044), DNA methylation of these six CpGs is associated with the expression of CPT1A and SREBF1 (for triglycerides), DHCR24 (for LDL cholesterol) and ABCG1 (for HDL cholesterol), which are all key regulators of lipid metabolism. CONCLUSIONS: Our analysis suggests a role for epigenetic priming in end-product feedback control of lipid metabolism and highlights Mendelian randomization as an effective tool to infer causal relationships in integrative genomics data.