Distinguishing HF with reduced and preserved ejection fraction at the level of individual cardiomyocytes: implications for therapeutic development

Abstract

Heart failure is a significant public health issue, representing the leading cause of death and hospitalization in developed countries (Borlaug & Redfield, 2011). The term ‘heart failure’ (HF) broadly encompasses a clinical condition in which the pump function of the heart is impaired, accompanied by symptoms including dyspnoea, fatigue and oedema (Borlaug & Redfield, 2011). HF can be divided into two subtypes, depending on whether the cardiac ejection fraction (EF) is impaired. HF with reduced EF (HFrEF) is characterized by an EF < 45–50% (Borlaug, 2014), often accompanied by chamber dilatation and impaired systolic function (Borlaug & Redfield, 2011). The pathophysiology of HFrEF is relatively well understood, with many effective treatment options available. In contrast, HF with preserved EF (HFpEF) is characterized by an EF > 50% (Borlaug, 2014), often accompanied by impaired ventricular relaxation and increased stiffness (Borlaug & Redfield, 2011). While HFpEF accounts for roughly 50% of total HF cases, this subtype remains poorly understood, owing to a lack of available human tissue or representative models for preclinical study (Borlaug, 2014). Furthermore, therapies with proven efficacy in HFrEF patients often show limited benefit in HFpEF patients (Borlaug & Redfield, 2011), suggesting that there are key molecular and physiological differences that distinguish the two conditions. The development of effective therapies for HFpEF will therefore require an improved understanding of how HFpEF manifests at the cellular and molecular level, and how these mechanisms differ from HFrEF. A recent study by Kilfoil et al. (2020), published in this issue of The Journal of Physiolgoy, addressed this question at the level of individual cardiomyocytes, focusing on how Ca2+ regulation is affected in HFrEF vs. HFpEF at baseline and in response to β-adrenergic stimulation (Fig. 1). These findings help to elucidate the distinct cardiomyocyte-specific pathological features of HFrEF and HFpEF, with implications for the development of improved treatment and management strategies for HF patients. HFrEF typically arises from conditions where cardiomyocytes are directly and acutely affected, including genetic cardiomyopathies and ischaemic injury (Borlaug & Redfield, 2011). As such, the HFrEF phenotype is often generated in rodents by surgical induction of myocardial infarction, or by introducing genetic mutations associated with human cardiomyopathies (Noll et al. 2020). To model HFrEF, Kilfoil et al. used surgical ligation of the left anterior descending coronary artery to induce myocardial infarction in Sprague Dawley (SD) rats. Echocardiographic measurement of the ejection fraction and E/A ratio demonstrated that following surgery, the rats developed systolic and diastolic dysfunction, respectively. As with other surgical models, the coronary ligation model accurately recapitulates the cardiac remodelling that occurs following myocardial infarction; however, the systemic factors that would contribute to the onset of myocardial infarction in human patients, such as vascular disease are not represented (Noll et al. 2020). Consequently, whether this model is a true representation of the complete HFrEF phenotype remains an open question. In contrast to HFrEF, HFpEF develops secondary to a variety of systemic conditions, including hypertension, diabetes and obesity (Borlaug & Redfield, 2011). Typically, HFpEF is modelled in rodents by experimentally inducing one of these co-morbidities, using environmental factors or genetic manipulation (Noll et al. 2020). In the study by Kilfoil et al., an HFpEF model was created by maintaining Dahl Salt-Sensitive (DS) rats on a high-salt diet, thereby inducing hypertension. An additional unique feature of this model is the development of insulin resistance; therefore, two relevant HFpEF co-morbidities are represented simultaneously. After 11 weeks on a high-salt diet, rats displayed echocardiographic indicators of diastolic dysfunction such as reduced E/A and increased E/e′ ratio, while the ejection fraction remained >65%, indicative of preserved systolic function. In addition to the absence of additional co-morbidities, such as obesity or age, a potential specific limitation of the DS rat model is that high salt intake is the primary factor contributing to the development of hypertension, while hypertension in human patients is caused by a number of environmental and genetic factors (Noll et al. 2020). Therefore, this model accurately represents the haemodynamic factors contributing to the onset of HFpEF, but does not account for the various risk factors associated with hypertension in humans. Alterations in excitation-contraction (EC) coupling have been demonstrated to contribute to the contractile dysfunction observed in HFrEF; however, the role of EC coupling in HFpEF is poorly characterized (Kilfoil et al. 2020). Using the models described above, Kilfoil et al. characterized EC coupling in single cardiomyocytes isolated from HFrEF and HFpEF rats. EC coupling was assayed by imaging Ca2+ transients using the Ca2+ indicator Fura-2. In agreement with echocardiographic measurements, the transient amplitude was decreased relative to healthy controls in HFrEF but not HFpEF cardiomyocytes, indicative of reduced contractile force production. Additionally, ultra-rapid line scanning confocal microscopy was used to assess EC coupling at the level of individual couplons, revealing reduced synchronicity and increased latency of Ca2+ spark production in HFrEF cardiomyocytes relative to controls. Interestingly, EC coupling efficiency was generally enhanced in HFpEF cardiomyocytes relative to controls, which Kilfoil et al. attributed to an adaptive enhancement of contractile function in order to overcome the increased stiffness associated with HFpEF. Kilfoil et al. also noted that this phenomenon has been observed in other models, in which HFpEF develops as a result of cardiomyocyte hypertrophy or ageing; however, it is important to note that all of these experimental models encompass only a subset of the factors that contribute to the development of HFpEF in human patients. Given that deficits in systolic function have been observed in HFpEF patients using other measures such as longitudinal tissue shortening or speckle tracing echocardiography (Borlaug, 2014), further studies should aim to confirm that EC coupling is preserved in cardiomyocytes derived from human HFpEF patients. In contrast to systolic dysfunction, diastolic dysfunction is present in both HFrEF and HFpEF. Diastolic dysfunction results from insufficient ventricular relaxation and filling during diastole (Borlaug, 2014). In addition to non-cardiomyocyte factors such as fibrosis, diastolic dysfunction may also be associated with cardiomyocyte-intrinsic factors, such as delayed myofilament relaxation kinetics and increased myofilament stiffness (Eisner et al. 2020). Elevated diastolic Ca2+ levels have been associated with the diastolic dysfunction present in both HFrEF and HFpEF (Eisner et al. 2020); therefore, Kilfoil et al. compared diastolic Ca2+ levels in HFrEF and HFpEF using the ratiometric Ca2+ indicator Fura-2. In agreement with echocardiographic data, diastolic Ca2+ concentrations in both HFrEF and HFpEF cardiomyocytes were elevated relative to controls. Elevated diastolic Ca2+ levels in HFrEF cardiomyocytes are commonly attributed to reduced Ca2+ uptake into the sarcoplasmic reticulum (SR), due to reduced SERCA expression and PLN hypophosphorylation (Eisner et al. 2020). In contrast, the underlying mechanism of elevated diastolic Ca2+ levels in HFpEF is poorly understood. Quantification of SR-associated protein expression levels and phosphorylation states revealed that the SR Ca2+ uptake system in HFpEF cardiomyocytes is generally similar to controls, apart from an increase in PLN expression. To investigate potential alternative mechanisms for elevated diastolic Ca2+ levels, Kilfoil et al. demonstrated that Ca2+ current, Cav1.2 expression and phosphorylated RyR were increased in HFpEF relative to controls. Given that no compensatory increases in Ca2+ uptake or efflux were observed, these observations suggest that the increase in diastolic Ca2+ concentration may result from increased Ca2+ influx via L-type Ca2+ channels, or leakage of SR Ca2+ into the cytosol via RyR channels (Fig. 1; Eisner et al. 2020; Kilfoil et al. 2020). Future experiments aiming to elucidate the precise mechanism by which diastolic dysfunction occurs in HFpEF cardiomyocytes may be of clinical interest, as these studies would have important implications in the development of therapeutic strategies to manage and prevent the onset of HFpEF in human patients. Exercise intolerance is a key feature of HFpEF, and is often associated with impaired cardiac responses to β-adrenergic stimulation. During exercise, activation of the β-adrenergic signalling cascade leads to PKA-dependent phosphorylation of various components of the Ca2+ handling apparatus, thereby increasing cardiomyocyte contractility, or inotropy, and accelerating relaxation kinetics, or lusitropy (Kilfoil et al. 2020). To assess inotropic and chronotropic reserve, Kilfoil et al. recorded Ca2+ transients in HFpEF cardiomyocytes stimulated with the β-adrenergic agonist isoprenaline (isoproterenol). In agreement with clinical observations, the isoprenaline-induced increases in transient amplitude and decay rate were blunted in HFpEF cardiomyocytes relative to controls. This observed lack of sensitivity to β-adrenergic activation could be related to elevated diastolic Ca2+ levels at baseline; however, this interpretation could be challenged based on the observation that many components of the Ca2+ handling apparatus are not dysregulated in unstimulated HFpEF cardiomyocytes. An alternative explanation could instead involve defects in the function of the β-adrenergic pathway in HFpEF cardiomyocytes. While Kilfoil et al. noted that downregulation of the β-adrenergic receptor has not been observed in the DS rat model, the function of downstream pathway components has not been thoroughly investigated in this model. Future experiments focused on the function of downstream components such as cAMP and PKA may provide additional insight as to whether defects in the β-adrenergic pathway contribute to the pathogenesis of HFpEF. Further elucidation of the cellular deficits leading to reduced inotropic and lusitropic reserve could have important implications for the development of treatments for HFpEF, particularly given the general failure of β-blockers in HFpEF patients (Borlaug, 2014). None. Sole author. None.