Abstract
BACKGROUND: Atrial fibrillation (AF) is the most common sustained arrhythmia worldwide and imposes a substantial clinical burden. Epicardial adipose tissue (EAT), the cardiac visceral fat depot, has been implicated in AF pathogenesis, but the cellular, molecular, and immune features underlying EAT-mediated AF remain elusive.</p>
METHODS: We combined population-based analysis, Mendelian randomization, single-nucleus RNA sequencing, and biomarker validation. The visceral adipose tissue-AF association was examined in the National Health and Nutrition Examination Survey, genetic evidence was evaluated using Mendelian randomization based on UKB (UK Biobank), EAT was profiled in patients with coronary artery disease (6 with AF, 3 without), and key circulating signals were evaluated in the Fuwai Hospital cohort and UKB.</p>
RESULTS: Higher visceral adipose tissue was associated with increased AF risk in National Health and Nutrition Examination Survey, and Mendelian randomization provided genetic evidence consistent with a potential causal contribution of visceral adipose tissue to AF. Single-nucleus RNA sequencing showed extensive EAT remodeling in patients with coronary artery disease and AF, including impaired neuroprotective and myelin-maintenance programs, metabolic imbalance, profibrotic activation, and a chronic inflammatory milieu with immunoglobulin G-related humoral immune activation. Drug2cell prioritized candidate drug-cell associations across specific EAT cell populations, and summary data-based Mendelian randomization prioritized PTPRN2 as a putative AF susceptibility gene. Higher plasma interleukin-17 levels were associated with postoperative AF in the Fuwai Hospital cohort, whereas circulating interleukin-17C and interleukin-17D were independently associated with incident AF in the UKB.</p>
CONCLUSIONS: We constructed the first single-nucleus transcriptomic atlas of human AF-related EAT and identified potential mechanisms linking EAT remodeling to arrhythmogenesis. Elevated plasma interleukin-17 may serve as a predictive biomarker for AF.</p>