HFrEF and HFpEF: are mitochondria at the heart of the matter?

Heart failure (HF) is the inability of the heart to pump enough blood to meet the body's requirements, particularly during exercise. Traditional therapies for HF, including neurohormonal modulators, are effective at improving patient outcomes (Pandey et al., 2023). HF is categorised into two main subtypes according to the left ventricular (LV) ejection fraction (EF): HF with reduced EF (HFrEF; EF < 40%) is characterised by weakened LV contraction, whereas HF with preserved EF (HFpEF; EF > 40%) presents with diastolic dysfunction. Although the two HF phenotypes have shared clinical manifestations, they have distinct pathophysiological processes, and only HFrEF patients respond positively to existing treatments. Therefore, novel diagnostic strategy and phenotype-targeted therapies are needed to improve clinical outcomes for all HF subtypes. A recent study published in The Journal of Physiology by Méndez-Fernández et al. (2024) adds to the current understanding of the common and diverging mechanisms that contribute to HFrEF and HFpEF. Méndez-Fernández et al. (2024) make use of two mouse models of HFpEF and HFrEF to compare their pathophysiology. The HFrEF model was re-created by exposing mice to angiotensin II (0.7 mg/kg/day) and a high salt diet. Similarly, a mouse model of HFpEF was re-created in age-matched mice (∼8–12 weeks) of the same strain (C57BL/6), but they were fed a high fat diet (HFD) to create metabolic disturbances. Both HF mice received nitro-l-arginine methyl ester (l-NAME) in their drinking water to promote haemodynamic stress, albeit at different concentrations, and exposure times for each model. Age-matched control mice were fed a normal chow diet only. Both HF models displayed similarly elevated systolic and diastolic blood pressure demonstrating hemodynamic stress caused by the high dose l-NAME (0.5 g/l; HFpEF) or combined low dose l-NAME (0.1 g/l) with angiotensin II and high salt (HFrEF). The HFpEF mice displayed severe adiposity and dyslipidaemia indicative of metabolic disturbance, yet insulin resistance and glucose intolerance were not measured, which may be a contributing mechanism in this model of HFpEF. Although the authors present age-matched mice, the HFpEF model required an 8-week HFD intervention compared to the HFrEF model where mice were exposed to only 4 weeks of angiotensin II and 5 weeks of high salt and l-NAME before developing HF. The study utilised an appropriate sample size in the end-point body weight data (n = 19–31/group). Following the 8-week study timeline both models had lung congestion (n = 5–9), indicating cardiac insufficiency. However, it is unclear if there was significant attrition in mice due to humane end-point euthanasia occurring before the full 8-weeks. It would be useful to know what overt signs of HF, if any, were displayed by the mice, such as dyspnoea, poor grooming, and rapid weight loss, to constitute a humane end-point euthanasia and potentially a more severe disease model. LV function was measured invasively with a pressure-volume catheter under anaesthesia to derive steady-state stroke volume (SV) and LV EF. In addition, the inferior vena cava was transiently occluded to obtain preload-dependent LV end-systolic pressure-volume relationship (PVR) slope and end-diastolic PVR stiffness index, to assess LV systolic and diastolic function respectively. The HFrEF model displayed decreased SV, EF and end-systolic PVR slope recapitulating systolic dysfunction. The HFpEF model had a normal EF and increased end-diastolic stiffness comparable to human HFpEF (Westermann et al., 2008). Notably, at elevated heart rates resembling exercise, HFpEF patients typically show reduced end-diastolic volume and SV with a left-shifted PVR. Méndez-Fernández et al. (2024) have presented limited PV-derived parameters to characterise their models and did not examine the model's cardiac function under different conditions (i.e. exercise), or time points. While PV loop measurements are the gold standard in evaluating cardiac function, they are not routinely conducted clinically. Conventional echocardiographic and Doppler imaging, which are non-invasive methods to diagnose cardiac function, would have allowed for continuous monitoring of cardiac function and HF progression across the study timeline. Méndez-Fernández et al. considered differences in intracellular Ca2+ cycling by utilising confocal imaging of isolated ventricular cardiomyocytes loaded with Fluo-4. They reported smaller Ca2+ transients in HFrEF cardiomyocytes, consistent with previous models of HFrEF (Kilfoil et al., 2020) and their findings of systolic dysfunction in vivo. Impaired Ca2+ release in HFrEF is well established in the literature and is often accompanied by decreased transverse-tubule density and integrity, resulting in asynchronous Ca2+-induced-Ca2+ release. Transverse-tubules were not examined in this study but there was asynchronous Ca2+ release of the Ca2+ transient leading edge which may be the related to changes in transverse-tubule remodelling, like that reported by Kilfoil et al. (2020) in a rat model of ischaemia-induced HFrEF. In contrast, the Ca2+ transients in HFpEF cardiomyocytes were of greater amplitude, but these changes were not mirrored by an increased systolic function in vivo as SV and EF were not different to control animals. Given that the afterload (systolic blood pressure) was greater in these animals, it is plausible that to maintain EF cardiomyocytes would have to produce more force. HFpEF cardiomyocytes also had asynchronous Ca2+ release on the leading edge of these Ca2+ transients. The authors propose that the asynchronous release was caused by impaired sensitivity of the ryanodine receptor (RyR), based on previous reports that transverse-tubule structure is maintained in a salt-sensitive hypertensive rat model of HFpEF (Kilfoil et al., 2020). However, the same report describes an increase in the influx of Ca2+ into HFpEF cardiomyocytes, without any change in the proportion of this influx relative to the Ca2+ transient amplitude, and enhanced Ca2+ release synchrony, which may be interpreted as unchanged RyR sensitivity. Méndez-Fernández et al. reported diastolic dysfunction in both HFrEF and HFpEF, based on an end-diastolic PVR stiffness index, which was mirrored in an increased time to 50% Ca2+ transient decay. In the case of their HFrEF model, this change appeared to be accompanied by cardiomyocyte hypertrophy and fibrosis of the ventricular wall, which could contribute to the increased wall stiffness. However, this does not explain why the decay phase of the intracellular Ca2+ transient would be prolonged in either case. It may have been useful for the authors to have reported changes in other periods of the decay phase, as these can be attributed to different Ca2+ reuptake and extrusion pathways. The time to 50% decay alone can be confounding in room temperature experiments such as these, where a 'bump' in the decay phase is observed, which appeared to be more prominent in the authors' representative transient from HFpEF cardiomyocytes. While most results in the study point to distinct cellular mechanisms between HFrEF and HFpEF, the mitochondria were found to be similarly depolarised in both groups compared to control. Protein hyperacetylation in both HF models could explain these findings; however, without identifying acetylation of specific mitochondrial proteins, there is limited evidence that it is the cause for mitochondrial depolarisation. The authors use tetramethylrhodamine ethyl ester (TMRE; 300 nM), which accumulates within mitochondria in inverse proportion to mitochondrial membrane potential. However, interpretation of TMRE can be difficult as changes in dye intensity are dependent on whether the dye is being used in quenching mode (typically > 50 nm) or unquenching mode (typically < 30 nm). The higher concentrations used in the quenching mode lead to greater mitochondrial accumulation and dye aggregation, which quenches fluorescence emission (Perry et al., 2011). The authors assume the dye to be in the unquenching mode and conclude there is a similar energetic impairment in both HF types. Given this is the only shared mechanism of the two models, further investigation of mitochondrial dysfunction, such as respirometry measurements and mitochondrial Ca2+ handling will strengthen this finding. Ca2+ within the mitochondrial matrix stimulates Kreb's cycle dehydrogenases to fuel the electron transport system and is a means of matching energy supply to demand within cardiomyocytes. Matrix Ca2+ could be limited in the HFrEF cardiomyocytes due to the smaller Ca2+ transient amplitudes. However, too much Ca2+ uptake can also lead to transient or permanent opening of the mitochondrial permeability transition pore, which could contribute to the depolarised state in the HFpEF cardiomyocytes. In the pursuit for a 'silver bullet' in HF therapeutics, perhaps we ought to turn our attention to the correction of mitochondrial dysfunction. Interestingly, the benefits of sodium-glucose cotransporter 2 inhibitors for treating both HFrEF and HFpEF may be acting through this shared mechanism and improving cardiomyocyte energetics (Pandey et al., 2023). Further understanding of the mitochondrial mechanisms behind HF will be a crucial step in understanding this shared feature of the HF subtypes. which is not a focus of current pharmacological treatments. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. No competing interests declared. All authors contributed to conception or design of the work and drafting the work or revising it critically for important intellectual content. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. Health Research Council: Amelia Sally Power, 24/747; Auckland Medical Research Foundation (AMRF): Amelia Sally Power, 1124006.