Telomere shortening in heart failure due to excess neurohormonal activation is associated with loss of nuclear TRF-2 and increased telomeric DNA damage and predicts cardiac recovery

Abstract

Abstract The mechanisms and clinical significance of telomere shortening in heart failure remain elusive. Mammalian cardiomyocyte (CM) regeneration is limited and CM cell division cannot account for telomere shortening. Whether telomere shortening in turn affects cardiac recovery remains unexplored. We induced heart failure by excess neurohormonal activation (NHA), a universal dysregulation in the failing heart. B6 mice were subjected to AngII-infusion, uninephrectomy and high-salt (AngII++) to induce excess NHA for 5 weeks. Cardiac function was assessed by ultrasound. Human iPSC-derived CMs, mouse CMs and rat ventricular CMs (H9C2) were stimulated with AngII. Human cardiac biopsies from patients with heart failure with reduced ejection fraction (HFrEF) were grouped depending on cardiac recovery. Telomere length was quantified by Q-FISH. ROS-induced DNA/RNA-damage was evaluated after staining for 8-oxo-7,8-dihydro-2'-deoxyguanosine, -guanine and – guanosine-residues (8-oxo-dG). H9C2-cells were co-stained for telomeres, telomere repeat binding factor 2 (TRF-2) and 8-oxo-dG, telomere-specific DNA-damage was quantified by automated colocalization analysis. Superoxide (O2-) was quantified by 2-hydroxyethidium (2-HE) using HPLC analysis. Fibrosis was quantified by Sirius red staining. AngII++-mice exhibited left ventricular (LV) hypertrophy (p<0.0001 vs B6) along with reduced LV ejection fraction (p=0.03 vs B6), reduced global longitudinal strain (p<0.001 vs B6) along with significant telomere shortening in CMs isolated from AngII++-mice (p=0.0082 vs B6). Telomere length showed a significant correlation with LV systolic function (r2=0.64, p=0.006). While 8-oxo-dG-staining revealed significantly increased ROS-induced DNA/RNA damage in AngII++-myocardial sections overall (p<0.0001 vs B6), DNA/RNA-damage was highest in areas of fibrotic adverse remodeling congruent with significant telomere shortening (p<0.0122 fibrotic vs non-fibrotic AngII++-myocardium, p<0.0001 vs B6). In all cells tested, NHA resulted in significant telomere shortening. As a potential explanation, CMs stimulated with AngII revealed significantly increased, dose-dependent O2–production (AngII 10 nM p=0.03, AngII 50 nm p=0.0004 vs H9C2) along with significant loss of nucelar TRF2 (AngII nM p=0.0049, Ang II 50 nM p<0.0001 vs H9C2), while colocalization analysis revealed a significantly increased ROS-induced damage to the telomeric DNA (AngII 10/50 nm p<0.05 vs H9C2). HFrEF-patients that had recovered cardiac function (ΔLVEF +5 – 30%) showed significantly longer myocardial telomeres than patients with no recovery/worsening (ΔLVEF −10 – 0%) (p=0.008), overall the telomere signal correlated significantly with ΔLVEF (r2=0.41, p=0.03). Our data provides insides into how excess NHA via increased ROS, loss of nuclear TRF-2 and increased telomeric DNA-damage causes myocardial telomere shortening in heart failure and further that the extent of telomere shortening determines cardiac recovery. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): BMBF 01EO1003