Abstract 945: A Novel Role of Cardiac DNA Methylation as a Regulator of Fibrosis in Human Diabetic Heart Failure

Abstract

Human heart failure (HF) is accompanied by changes in cardiac gene expression; however, the impact of diabetes mellitus on this transcriptional regulation remains unclear. Our laboratory studies the role of epigenetics in HF and focuses on DNA methylation, commonly accepted as a negative regulator of gene expression when present in a gene’s promoter. In the current study, we hypothesized that diabetes alters cardiac DNA methylation and the transcriptome. We performed an integrated analysis of RNA-sequencing and DNA methylation using human left ventricle biopsies from HF patients to determine the effect of diabetes on this reprogramming. Our analysis identified global changes in cardiac DNA methylation as well as mRNA expression sufficient to distinguish diabetic from non-diabetic patients. Specifically, we identified significant regulation of 1,133 genes (|1.5| fold-change, P<0.05) and 9,302 methylation sites (|5%| methylation, P<0.05). Within the 391 co-regulated genes, only 174 genes had altered promoter-associated methylation. Consistent with the accepted dogma that promoter methylation inversely regulates gene expression, we found that 119 of those 174 genes displayed this inverse relationship; of these, an overwhelming majority (102) were hypomethylated in diabetic relative to non-diabetic hearts. These findings contrast the cardiac hypermethylation we have reported for ischemic HF. Gene set enrichment of the 119 inversely-regulated genes identified numerous pathways involved in fibrosis, including extracellular matrix organization (FDR<0.05, 3.9% enriched) and collagen biosynthesis (FDR<0.05, 6.3% enriched), which suggest that DNA methylation contributes to the adverse cardiac remodeling associated with diabetic heart failure. To identify putative regulators of these transcriptional and epigenetic differences, a candidate gene approach was used to reveal induction (2.5-fold, Q<0.05) and demethylation (20.5%, P<0.05) of EGR2, a known regulator of fibrosis. Furthermore, we found induction of GADD45beta (2.6-fold, Q < 0.05) a known regulator of DNA demethylation. Taken together, these observations suggest that epigenetic mechanisms underlie an etiology-specific reprogramming of cardiac fibrosis in diabetic HF.