High-resolution DNA methylation changes reveal biomarkers of heart failure with preserved ejection fraction versus reduced ejection fraction

Abstract

Novel biomarkers are needed to better identify-and distinguish-heart failure with preserved ejection fraction (HFpEF) from other clinical phenotypes. The goal of our study was to identify epigenetic-sensitive biomarkers useful to a more accurate diagnosis of HFpEF. We performed a network-oriented genome-wide DNA methylation study of circulating CD4⁺ T lymphocytes isolated from peripheral blood using reduced representation bisulfite sequencing (RRBS) in two cohorts (i.e., discovery/validation) each of both male and female patients with HFpEF (n = 12/10), HF with reduced EF (HFrEF; n = 7/5), and volunteers lacking clinical evidence of HF (CON; n = 7/5). RRBS is the gold-standard platform for measuring genome-wide DNA methylation changes at single-cytosine resolution in hypothesis-generating studies. We identified three hypomethylated HFpEF-specific differentially methylated positions (DMPs) associated with FOXB1, ELMOD1, and DGKH genes wherein ROC curve analysis revealed that increased expression levels had a reasonable diagnostic performance in predicting HFpEF (AUC ≥ 0.8, p < 0.05). Network analysis identified additional three genes including JUNB (p = 0.037), SETD7 (p = 0.003), and MEF2D (p = 0.0001) which were significantly higher in HFpEF vs. HFrEF patients. ROC curve analysis showed that integrating the functional H₂FPEF classification with the expression levels of the FOXB1, ELMOD1, and DGKH as well as the JUNB, SETD7, and MEF2D genes improved diagnostic accuracy, with AUC = 0.8 (p < 0.0001) as compared to H₂FPEF score alone (p > 0.05). Besides, increased expression levels of SETD7-RELA-IL6 axis significantly discriminated overweight/obese HFpEF vs. HFrEF patients (AUC = 1; p = 0.001, p = 0.006, p = 0.006, respectively). We support an emerging dogma that indirect epigenetic testing via high-resolution RRBS methylomics represents a non-invasive tool that may enable easier access to both diagnostic and mechanistic insights of HFpEF. An epigenetic-oriented dysregulation of network-derived SETD7-RELA-IL6 axis in circulating CD4⁺ T lymphocytes may drive pro-inflammatory responses which, in turn, may lead to cardiac remodeling in overweight/obese HFpEF.

Keywords: Circulating CD4+ T lymphocytes; DNA methylation; Diagnostic biomarkers; HFpEF; Liquid biopsy; Network analysis.