Background: Respiratory heart rate variability (RespHRV) is a natural beat-to-beat variation in heart rate linked to respiration that is lost in heart failure. We have previously shown that restoring RespHRV using a novel biofeedback pacemaker improves cardiac function in an ovine model of heart failure with reduced ejection fraction (HFrEF). Recent proteomic and metabolomic analyses suggested improved mitochondrial function as a potential mechanism underlying this response. We therefore investigated whether RespHRV pacing restores mitochondrial structure in the failing heart.
Methods: Heart failure was induced in female Romney sheep by sequential coronary microembolisation. Animals were assigned to control (n=5), heart failure (HF; n=5), or heart failure treated with RespHRV pacing for two weeks (HF+R; n=5). Heart failure was induced in sheep by sequential coronary microembolization. Three months following heart failure induction, animals underwent instrumentation surgery and received two weeks of pacing therapy. Sheep were then euthanized, and cardiac tissue was collected for subsequent analyses. Surgical procedures, anaesthesia, and pharmacological protocols were performed as previously described by Shanks et al., (2022). Left ventricular tissue was analysed using stimulated emission depletion (STED) microscopy of TOMM20-labelled mitochondria and transmission electron microscopy (TEM) of mitochondrial ultrastructure. Proteomic and metabolomic analyses were used to provide mechanistic context. STED and TEM images were analysed using a linear mixed-effects model with group (C, HF, HF+R) as a fixed effect and animal included as a random intercept to account for clustering of myocytes within animals. Planned comparisons performed between groups used Least Significant Difference (LSD) tests. Data was expressed as mean and SEM.
Results: STED microscopy demonstrated a highly organised columnar arrangement of mitochondria in control myocardium that was disrupted in HF. Directionality analysis revealed loss of mitochondrial alignment in HF that was restored following RespHRV pacing (p=0.051). Quantification of TOMM20 labelling demonstrated a significant increase in TOMM20-positive area in HF+R compared with HF (p=0.005). TEM revealed marked disruption of mitochondrial cristae architecture in HF hearts, characterised by a significant reduction in cristae density compared with controls (p<0.001). Following two weeks of RespHRV pacing, cristae density was significantly increased compared with HF (p<0.001) and mitochondrial ultrastructure restored towards the control phenotype despite no change in total mitochondrial area fraction. Complementary proteomic analyses demonstrated restoration of mitochondrial proteins associated with oxidative metabolism and fatty acid β-oxidation, while paired plasma metabolomic analyses showed reductions in circulating long-chain acylcarnitines following pacing, consistent with improved mitochondrial oxidative function.
Conclusions: RespHRV pacing induces reverse remodelling of mitochondrial architecture in the failing heart. STED microscopy and TEM demonstrate restoration of mitochondrial organisation and cristae density, providing a structural basis for improved mitochondrial function. These findings identify mitochondrial recovery as a key component of the therapeutic response to RespHRV pacing and highlight the value of advanced imaging approaches for understanding mechanisms of cardiac repair.