Background: Atrial fibrillation (AF) is defined as a supraventricular tachyarrhythmia with uncoordinated electrical pulses, impairing atrial function. It is the most common arrhythmia leading to stroke and heart failure. It is already known that ectopic pacemaker usually near the pulmonary veins stimulate atrial arrhythmia. Molecular and epigenetic regulations in the right atria of patients with permanent AF are unknown compared to patients with sinus rhythm (SR).

Methods: Human samples were collected from patients with permanent AF during surgery or patients with SR after non-cardiovascular death. Samples from the right atrial appendage (RAA) were isolated by fluorescence-activated nuclei sorting (FANS) based on phospholamban (PLN) staining prior further processing of cardiomyocytes (CMs). Next generation sequencing approaches including bulk and single nuclei RNA sequencing (RNA-seq) as well as whole genome bisulfite sequencing (WGBS) were used to define transcriptomic and DNA methylation changes in CMs of AF patients. Genomic regulatory regions and gene expression data were also linked to genome wide association studies (GWAS).

Results: In bulk RNA-seq CMs from patients with permanent AF compared to SR samples showed changes in gene expression, including ion channels, sarcomere genes, and transcription factors (TFs) (AF vs SR, 583 DEGs, p-adjusted < 0.05, n=16/15). AF-specific low methylated regions (AF-LMRs, n=1237), representative for regulatory sites in the genome, were enriched for motifs of the ETS-related family (e.g. ETV4, p-val. = 2.66e-31) and could be associated with SNPs found in AF (AF-LMRS/random, log2FC = 0.74, -log10(Chi-square p-val) = 8 ). The heart is composed of many different cell types and non-CM cells influencing the arrhythmic behaviour of the atria. Transcriptome analysis on single nuclei level revealed the intercellular communication and importance of endothelial cells in AF patients. Gene ontology analysis of the individual cell types showed enrichment of biological processes. AF induced downregulation of genes associated with calcium regulation and axon guidance in CMs, MAPK-, ERBB- and TGF-beta signalling in endothelial cells (ECs) and AP-1 signalling in endocard and fibroblasts, as well as immune response pathways including interleukin signalling in monocytes/ macrophages. Enriched GO terms of upregulated genes, e.g. in CMs were dilated cardiomyopathy and muscle contraction and extra cellular matrix interactions in ECs and fibroblasts. Transcriptome analysis on single nuclei level revealed the extensive changes in intercellular communication due to AF. The total number of interactions of ECs and immune cells to other cell types increase in AF patients.  Ligand-receptor pairing in EC subtypes were strongly altered, especially semaphorins and neuregulins, whereas CMs showed an upregulation in apelin signalling and downregulation in BMP7 expression.

Conclusion: In this study, we assessed differences in gene expression and epigenetic regulation in AF. Here we combined the genetic predisposition for AF resembled in GWAS features and the epigenetic regulations including gene expression of TFs, DNA methylation changes and cell-cell interactions landscape. This study brought us one step closer to a better understanding of the complex pathology of AF.