Mitochondrial heterogeneity across cardiac cell types in heart failure with preserved ejection fraction

Abstract

Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous and increasingly prevalent syndrome for which effective disease-modifying therapies remain limited, in part because its multicellular pathophysiology remains incompletely understood. Mitochondrial dysfunction has emerged as an important contributor to HFpEF, yet most studies have examined mitochondrial abnormalities within individual cell types rather than as part of an integrated cardiac microenvironment. This review discusses evidence that cell-type-specific mitochondrial alterations in cardiomyocytes, fibroblasts, and endothelial cells may interact through redox, metabolic, and paracrine signaling to promote diastolic dysfunction, myocardial stiffening, and impaired energetic reserve. In cardiomyocytes, disrupted mitochondrial calcium (_mCa²⁺) handling and redox imbalance uncouple ATP production from energetic demand, contributing to Ca²⁺ overload, mitochondrial reactive oxygen species (mtROS) generation, and impaired relaxation. In fibroblasts, mitochondrial and metabolic reprogramming support profibrotic activation and extracellular matrix (ECM) deposition. In endothelial cells, mitochondrial dysfunction reduces nitric oxide bioavailability, enhances oxidative stress, and contributes to microvascular dysfunction. Together, these cell-type-specific mitochondrial abnormalities may converge on a redox–calcium–energetics axis that helps explain the dissociation between preserved systolic function and impaired diastolic performance in HFpEF. We further highlight mitochondrial heterogeneity and intercellular crosstalk as potential determinants of disease progression and discuss emerging therapeutic strategies targeting mitochondrial pathways as a rationale for precision approaches in HFpEF.

Graphical abstract

Graphical Abstract. Mitochondrial heterogeneity and intercellular crosstalk drive HFpEF pathogenesis. Mitochondrial dysfunction manifests in a cell-type-specific manner across the HFpEF myocardium. In cardiomyocytes, disrupted _mCa²⁺ handling and redox imbalance promote Ca²⁺ overload, excessive mtROS production, impaired ATP generation, and energetic uncoupling. In fibroblasts, mitochondrial and metabolic reprogramming enhance profibrotic signaling, ECM synthesis, and myocardial stiffening. In endothelial cells, mitochondrial dysfunction increases mtROS, promotes eNOS uncoupling, reduces nitric oxide (NO) bioavailability, and contributes to microvascular dysfunction. These abnormalities interact through ROS-dependent signaling, paracrine mediators, and metabolic crosstalk, converging on a pathogenic redox–calcium–energetics axis. Emerging therapeutic strategies targeting _mCa²⁺ flux, oxidative stress, MQC, bioenergetics, and endothelial function may support cell-type-specific precision therapies in HFpEF.

1. Introduction

Heart failure with preserved ejection fraction (HFpEF) represents a growing clinical burden and is characterized by clinical symptoms of HF despite preserved left ventricular ejection fraction (LVEF ≥50%) [1,2]. Unlike HF with reduced ejection fraction (HFrEF), HFpEF is a highly heterogeneous syndrome encompassing cardiometabolic, hypertensive, inflammatory, and aging-associated phenotypes. Thus, mechanisms identified in one HFpEF model or endotype may not generalize across all human HFpEF phenotypes. This biological complexity has contributed to the persistent therapeutic gap, as treatments effective in HFrEF have shown limited benefits in HFpEF [2]. Consequently, current management remains largely focused on symptom control rather than disease-modifying therapies. The burden of HFpEF continues to rise with population aging and increasing cardiometabolic disease. HFpEF now accounts for more than half of all heart failure cases and has become the predominant heart failure phenotype, underscoring the urgent need for deeper mechanistic understanding and targeted therapies [3,4].

Mitochondrial dysfunction has emerged as an important contributor to HFpEF pathophysiology, linking metabolic stress to impaired cardiac function. Beyond ATP production, mitochondria integrate redox balance, calcium (Ca²⁺⁾ signaling, and metabolic adaptation. In HFpEF, disrupted mitochondrial calcium (_mCa²⁺) handling, increased reactive oxygen species (ROS) generation, and reduced metabolic flexibility contribute to energetic failure, oxidative stress, and diastolic dysfunction [[5], [6], [7], [8]]. Notably, mitochondrial alterations in HFpEF are not confined to cardiomyocytes. Cardiac fibroblasts, endothelial cells (ECs), and immune cells exhibit distinct mitochondrial adaptations that promote fibrosis, microvascular dysfunction, and inflammation. However, these cell-type-specific alterations are frequently studied in isolation, limiting a comprehensive understanding of disease mechanisms [9,10]. A major barrier to therapeutic development is the lack of an integrated view of how mitochondrial dysfunction across cardiac cell populations contributes to HFpEF. Here, we synthesize evidence that cell-type-specific mitochondrial alterations in cardiomyocytes, fibroblasts, and endothelial cells may interact through redox, metabolic, and paracrine signaling to promote disease progression. This perspective provides a mechanistic basis for developing cell-type-specific, mitochondria-targeted therapeutic strategies and supports the advancement of precision approaches in HFpEF. In this review, we integrate _mCa²⁺ handling, redox regulation, bioenergetic remodeling, and intercellular mitochondrial crosstalk across cardiomyocytes, fibroblasts, and endothelial cells into a multicellular framework for HFpEF pathogenesis.

2. Mitochondria in normal cardiac physiology: integration of energetics, calcium, and redox signaling

The heart operates under exceptionally high energetic demand, requiring continuous ATP production at rest with the ability to rapidly scale output during increased workload. To meet this requirement, mitochondria occupy approximately 30–35% of cardiomyocyte volume and serve as the primary source of myocardial ATP through oxidative phosphorylation (OXPHOS) [11,12]. Efficient ATP production depends on maintenance of a strongly negative mitochondrial membrane potential (ΔΨm), which drives proton re-entry through ATP synthase and sustains energy generation [13,14]. Beyond ATP synthesis, mitochondria act as dynamic signaling hubs that integrate Ca²⁺, metabolic flux, and redox balance into a coordinated regulatory system [14]. Central to this function is _mCa²⁺ handling. Ca²⁺ enters the mitochondrial matrix via the mitochondrial calcium uniporter (MCU) and is extruded primarily through the sodium–calcium exchanger (NCLX), enabling tight control of matrix Ca²⁺ concentration in response to beat-to-beat changes in workload [[15], [16], [17]].

_mCa²⁺ links excitation–contraction coupling to mitochondrial metabolism by activating Ca²⁺-sensitive tricarboxylic acid (TCA) cycle dehydrogenases including pyruvate dehydrogenase, isocitrate dehydrogenase, and α-ketoglutarate dehydrogenase, thereby increasing NADH and FADH₂ production, ETC flux, and ATP synthesis [18]. Through this mechanism, cytosolic Ca²⁺ signals are matched to mitochondrial energy production. Cardiac mitochondria also display considerable metabolic flexibility, utilizing fatty acids, glucose, lactate, ketones, and amino acids [18,19]. This adaptability is supported by the phosphocreatine shuttle, which buffers ATP flux and facilitates rapid energy transfer between sites of production and utilization [20].

The spatial organization of mitochondria further optimizes these functions. Interfibrillar mitochondria, located between myofibrils, supply ATP for contraction; subsarcolemmal mitochondria support ion transport and membrane-associated signaling; and perinuclear mitochondria contribute to transcriptional regulation [21]. Physical interactions between mitochondria and the sarcoplasmic reticulum (SR), often referred to as mitochondria-associated membranes (MAMs), create specialized microdomains that enable efficient Ca²⁺ transfer and localized metabolic regulation [22,23]. Mitochondria also modulate intracellular Ca²⁺ dynamics essential for excitation–contraction coupling. During systole, L-type Ca²⁺ influx triggers SR Ca²⁺ release, activating the contractile apparatus. Relaxation requires rapid cytosolic Ca²⁺ clearance, largely mediated by SERCA-dependent reuptake into the SR, a process that is highly ATP-dependent [24]. Importantly, mitochondrial ROS (mtROS), generated as byproducts of electron transport, function as signaling molecules under physiological conditions. The primary mtROS species is superoxide, which is generated predominantly at complexes I and III of the Electron transport chain (ETC) when electron transfer becomes partially uncoupled from OXPHOS [25]. Superoxide is rapidly converted to hydrogen peroxide (H₂O₂) by manganese superoxide dismutase (SOD2), allowing ROS signals to participate in reversible redox signaling [26]. Under physiological conditions, mitochondrial antioxidant systems including SOD2, glutathione peroxidases, peroxiredoxins, thioredoxin, and NADPH-dependent reducing pathways maintain redox balance and prevent oxidative damage [27]. Thus, mitochondrial function depends not only on ATP production but also on tight regulation of mtROS generation and antioxidant capacity. Low levels of ROS regulate redox-sensitive pathways involved in metabolism, Ca²⁺ handling, and gene expression [28]. However, these effects require tight control to avoid oxidative damage, highlighting the importance of balanced redox homeostasis.

Collectively, these processes form an integrated redox–calcium–energetics axis, in which _mCa²⁺ regulates metabolic flux, ATP production supports Ca²⁺ cycling, and ROS modulates both signaling and stress responses (Fig. 1). This axis enables the myocardium to dynamically match energy supply with contractile demand while maintaining redox balance and cellular homeostasis. Disruption of any component of this tightly coordinated system can initiate maladaptive feedback loops. Excess _mCa²⁺ and oxidative stress promote opening of the mitochondrial permeability transition pore, leading to collapse of ΔΨm, impaired ATP synthesis, reduced Ca²⁺ buffering capacity, and further ROS generation [27,29].

Fig. 1.

The mitochondrial redox–calcium–energetics axis in normal cardiac physiology. Mitochondria are central regulators of cardiac function through an integrated redox–calcium–energetics axis, which coordinates _mCa²⁺ signaling, ROS generation, and ATP production to match energy supply with cardiac demand. In cardiomyocytes, _mCa²⁺ uptake through the mtCU activates TCA-cycle dehydrogenases, increasing NADH/FADH₂ production, ETC flux, and ATP synthesis. Controlled ROS generation supports adaptive signaling, whereas excessive ROS disrupts Ca²⁺ homeostasis and energetic efficiency. Distinct mitochondrial subpopulations and SR–mitochondrial contact sites further coordinate contraction, relaxation, and metabolic adaptation.

Although the redox–calcium–energetics axis represents a conserved framework of mitochondrial function, its manifestations differ substantially across cardiac cell populations. Cardiomyocytes rely on this network to support excitation–contraction coupling and energetic demand, whereas fibroblasts and endothelial cells utilize mitochondria primarily as signaling platforms that regulate cellular differentiation, redox homeostasis, inflammatory responses, and tissue remodeling. Consequently, disruption of mitochondrial signaling in HFpEF produces distinct cell-type–specific phenotypes that extend beyond individual cellular dysfunction and set the stage for the downstream discussion of intercellular metabolic, redox, and paracrine crosstalk.

3. Disruption of the redox–calcium–energetics axis in HFpEF: cell-type-specific mitochondrial derangements

Building on this physiological framework, HFpEF can be examined through cell-type-specific disruption of mitochondrial Ca²⁺ handling, redox signaling, and energetic adaptation. Although cardiomyocytes account for most myocardial mass (70–85%), non-myocyte populations, including fibroblasts, ECs, and immune cells, contribute importantly to disease progression through metabolic, paracrine, and extracellular matrix (ECM)–mediated signaling. Together, these observations support a multicellular perspective on mitochondrial dysfunction and provide the rationale for examining how cell-type–specific mitochondrial abnormalities converge through intercellular signaling pathways [[30], [31], [32]].

3.1. Cardiomyocytes

Cardiomyocytes are the principal determinants of myocardial contraction and relaxation and therefore provide the most direct cellular context in which disruption of the redox–calcium–energetics axis manifests as diastolic dysfunction. In healthy myocardium, mitochondrial ATP production, SR Ca²⁺ cycling, _mCa²⁺ uptake, and redox homeostasis are tightly coordinated to match energy supply with contractile demand. In HFpEF, this integration becomes progressively destabilized. Altered SR Ca²⁺ handling, dysregulated _mCa²⁺ uptake and efflux, impaired OXPHOS, and excess mtROS generation reinforce one another, producing a feedforward network of energetic stress and impaired relaxation.

3.1.1. Distinct remodeling of mitochondrial calcium signaling in HFpEF versus HFrEF

Cytosolic–SR Ca²⁺ cycling is a central determinant of cardiomyocyte function and is differentially remodeled in HFpEF versus HFrEF. Efficient excitation–contraction coupling depends on the close structural and functional association between the SR network and t-tubule membranes [33]. In HF, ventricular cardiomyocytes typically exhibit reduced t-tubule density, while atrial myocytes can demonstrate near-complete t-tubule loss [34]. In contrast, HFpEF is characterized by relatively preserved, and in some cases increased, t-tubule density, which supports maintenance of systolic Ca²⁺ transients despite increased myocardial stiffness [35]. This likely reflects a compensatory adaptation that preserves contractile function while masking underlying defects in relaxation.

At the molecular level, Ryanodine receptor 2 (RyR2) regulation differs between HFrEF and HFpEF. In HFrEF, pathological phosphorylation of RyR2 by protein kinase A (PKA) and Ca²⁺/calmodulin-dependent kinase II (CaMKII) promotes diastolic Ca²⁺ leak, partly through destabilization of the RyR2–FKBP12.6 complex, resulting in reduced SR Ca²⁺ content and impaired contractile performance [36,37]. In contrast, HFpEF exhibits preserved or enhanced systolic RyR2-mediated Ca²⁺ release, which helps maintain systolic function[35]. In this setting, diastolic Ca²⁺ leak appears more selective and is primarily associated with PKA-dependent signaling rather than widespread CaMKII activation or oxidative destabilization [35,38]. Together, these findings indicate that HFpEF is characterized by preserved systolic Ca²⁺ release but impaired diastolic Ca²⁺ clearance, generating a Ca²⁺-handling phenotype distinct from that of HFrEF.

SR Ca²⁺ reuptake is also differentially affected. In HFrEF, reduced Sarco/endoplasmic reticulum Ca²⁺-ATPase 2a (SERCA2a) expression and activity limit Ca²⁺ sequestration into the SR [39], while phospholamban further inhibits SERCA2a function, contributing to elevated cytosolic Ca²⁺ during diastole [40]. In HFpEF, SERCA2a abnormalities are more variable, with some models demonstrating preserved expression but altered regulation, including reduced phospholamban phosphorylation [7,35]. These findings suggest that impaired Ca²⁺ clearance in HFpEF cannot be fully explained by SR dysfunction alone and may involve additional buffering mechanisms.

Consistent with this, _mCa²⁺ handling is increasingly recognized as a critical contributor to Ca²⁺ homeostasis. In HFpEF, reduced SERCA2a activity and insufficient phospholamban phosphorylation delay SR Ca²⁺ reuptake, while RyR2 mediated Ca²⁺ leak further elevates cytosolic Ca²⁺ levels [41,42]. In parallel, dysfunction of the mitochondrial calcium uniporter (mtCU) complex impairs _mCa²⁺ buffering, prolonging cytosolic Ca²⁺ transients and delaying diastolic relaxation [23,43]. Disrupted _mCa²⁺ signaling also reduces activation of Ca²⁺-dependent dehydrogenases, further impairing ATP production and reinforcing a feedforward loop between energetic deficiency and Ca²⁺ dysregulation [44].

Oxidative stress further amplifies these defects. ROS, generated by mitochondrial dysfunction and enzymes such as NADPH oxidase, xanthine oxidase, and nitric oxide synthase (NOS), promote oxidative modification of RyR2, increasing diastolic Ca²⁺ leak [27,45]. This enhances reliance on sodium–calcium exchange mechanisms for Ca²⁺ extrusion [46], yet diastolic Ca²⁺ remains elevated due to persistent SR leak and impaired reuptake [47].

_mCa²⁺ cycling differs fundamentally between HFpEF and HFrEF. In HFrEF, reduced _mCa²⁺ uptake leads to diminished activation of TCA-cycle enzymes, impaired ATP production, and increased ROS generation. Several cardiometabolic HFpEF models exhibit increased or dysregulated _mCa²⁺ accumulation, although this finding has not been consistently observed across all HFpEF models. Although elevated _mCa²⁺ can transiently enhance OXPHOS, sustained overload promotes mitochondrial dysfunction, including permeability transition pore opening and ROS amplification [27,29,44,48]. Notably, some models (e.g., ZSF1) report reduced _mCa²⁺ content despite mitochondrial swelling, suggesting that _mCa²⁺ dynamics in HFpEF may vary depending on disease stage or model [49]. Thus, model-dependent findings should be distinguished from mechanisms validated in human HFpEF. Nevertheless, impaired mitochondrial respiration, particularly reduced complex activity, is a consistent finding [7,50]. These differences reflect distinct regulation of _mCa²⁺ transporters. In HFrEF, increased NCLX activity favors Ca²⁺ efflux, reducing _mCa²⁺ load [51,52], whereas MCU upregulation restores _mCa²⁺ levels and improves cardiac function [53,54].

In HFpEF, _mCa²⁺ dysregulation appears to result from altered _mCa²⁺ flux rather than consistent changes in MCU expression. Although elevated mitochondrial Ca²⁺ has been reported in several HFpEF models, the relative contributions of MCU-mediated uptake and NCLX-mediated efflux remain incompletely defined. Emerging evidence suggests that impaired NCLX-dependent Ca²⁺ extrusion may contribute to _mCa²⁺ accumulation and mitochondrial dysfunction, but this mechanism requires validation in human HFpEF [7,55].

Alterations in MAMs further influence Ca²⁺ dynamics. In HFrEF, disruption of MAM structure and downregulation of proteins such as IP3R2 and PACS2 reduce SR-to-mitochondria Ca²⁺ transfer [56]. In contrast, some metabolic HFpEF models exhibit disturbed mitochondrial-cytosolic Ca²⁺ handling and mitochondrial dysfunction, raising the possibility that altered SR-mitochondrial communication and MAM signaling contribute to disease progression [43,57]. Although these findings may not fully generalize to all HFpEF phenotypes [58], they suggest that enhanced SR–mitochondrial coupling may contribute to _mCa²⁺ accumulation and mitochondrial dysfunction in selected HFpEF settings.

Collectively, these findings highlight that although both HFpEF and HFrEF involve mitochondrial and Ca²⁺ handling abnormalities, their dominant mechanisms differ substantially (Table 1). Many experimental HFpEF models are characterized by impaired diastolic relaxation, _mCa²⁺ dysregulation, endothelial dysfunction, and metabolic remodeling. In contrast, HFrEF is more strongly associated with systolic failure, impaired mitochondrial energetics, and adverse ventricular remodeling [59]. These distinctions support disease-specific therapeutic strategies targeting distinct mitochondrial and cellular pathways in each HF subtype.

Table 1.

Comparative mitochondrial and cellular mechanisms distinguishing HFpEF and HFrEF. This table summarizes major differences in _mCa²⁺ handling, mitochondrial signaling, energetics, oxidative stress, endothelial dysfunction, fibrosis, and inflammatory remodeling between HFpEF and HFrEF. HFpEF is characterized predominantly by impaired diastolic relaxation, _mCa²⁺ overload, microvascular dysfunction, and multicellular metabolic remodeling. In contrast, HFrEF is more strongly associated with systolic failure, impaired mitochondrial energetics, and adverse ventricular remodeling. Together, these distinct mitochondrial and cellular phenotypes highlight divergent pathophysiological mechanisms that may require disease-specific therapeutic strategies. Features listed under HFpEF are derived predominantly from experimental HFpEF models, particularly cardiometabolic and metabolic-risk models, and should not be interpreted as universal characteristics of all HFpEF phenotypes. Human validation remains limited for several mitochondrial pathways.

Comparative mitochondrial and cellular mechanisms in HFpEF vs. HFrEF
Molecular pathways	HFpEF	HFrEF
Primary functional abnormality	Preserved ejection fraction with impaired relaxation and increased ventricular stiffness [60]	Reduced ejection fraction with impaired systolic contractility [59]
Cardiac remodeling phenotype	Predominantly concentric remodeling and myocardial stiffening [8,60]	Predominantly eccentric remodeling and ventricular dilation [35,59]
T-tubule remodeling	Relatively preserved or mildly altered t-tubule organization [35]	Marked t-tubule disruption and loss [35]
RyR2-mediated SR Ca2+ leak	Preserved systolic Ca2+ release with impaired diastolic Ca2+ homeostasis and selective SR Ca2+ leak [35,41]	Pronounced RyR2 destabilization with increased CaMKII-mediated SR Ca2+ leak [36,61]
SERCA2a regulation	Variable; often preserved SERCA2a expression but impaired regulatory control [7,35]	Reduced SERCA2a expression and impaired SR Ca2+ reuptake [40,59]
mCa2+ handling	mCa2+ overload or dysregulated accumulation contributing to ROS generation and permeability transition pore opening [7,29,48]	Reduced mCa2+ uptake associated with impaired energetic coupling [48]
MCU/NCLX balance	Contributions of MCU-mediated uptake versus NCLX-mediated efflux remain incompletely defined and appear model-dependent [7,55]	Increased NCLX activity favors mCa2+ efflux [51]
mtROS production	Increased mtROS linked to metabolic stress, inflammation, and endothelial dysfunction [8,60,62]	Increased mtROS generation associated with ETC dysfunction, impaired oxidative phosphorylation, and energetic deficiency [59,63,64]
Energetic phenotype	Impaired metabolic flexibility with reduced oxidative reserve and mitochondrial efficiency despite preserved systolic function [8,35,65]	Reduced OXPHOS and impaired ATP production [20,59]
MAMs	Disturbed mitochondrial-cytosolic Ca2+ communication and putative dysregulation of SR-mitochondrial/MAM signaling [7,57,58]	Disrupted SR–mitochondrial communication and reduced Ca2+ transfer [56]
NO signaling	Reduced NO–sGC–cGMP signaling central to coronary microvascular dysfunction [60,62]	NO signaling impaired but less central to disease progression [8,60,62]

3.1.2. Oxidative stress and redox imbalance in HFpEF cardiomyocytes

ROS act as both signaling molecules and mediators of injury, and their role in HFpEF cardiomyocytes is central to the disruption of the redox–Ca²⁺–energetics axis. In HF, excessive ROS production contributes to oxidative modification of contractile and cytoskeletal proteins, including actin and myosin, impairing both myocardial contraction and relaxation [66,67]. ROS-mediated modification of RyR2 increases diastolic Ca²⁺ leak, leading to sustained cytosolic Ca²⁺ elevation and delayed relaxation [41,66]. Mitochondria act as both a source and a target of ROS [63]. Increased _mCa²⁺ loading enhances ETC activity but also increases electron leakage from complexes I and III, promoting mitochondrial superoxide generation. Although transient mtROS production contributes to adaptive signaling, sustained _mCa²⁺ overload produces excessive superoxide and H₂O₂ generation that overwhelms endogenous antioxidant defenses, leading to oxidative modification of Ca²⁺-handling proteins, respiratory complexes, and metabolic enzymes [63,67].

Over time, oxidative damage to respiratory complexes reduces ATP production and weakens _mCa²⁺ buffering capacity. This establishes a feedforward loop in which redox imbalance exacerbates Ca²⁺ dysregulation and energetic deficiency, ultimately contributing to impaired diastolic function in HFpEF. Together, impaired _mCa²⁺ buffering, excessive mtROS generation, and reduced energetic reserve create a feedforward network that selectively compromises diastolic relaxation in HFpEF.

3.1.3. Therapeutic implications in cardiomyocytes

Targeting Ca²⁺ handling remains a central therapeutic strategy in HF. Approaches that enhance SR Ca²⁺ uptake or reduce SR Ca²⁺ leaks, including SERCA2a gene therapy, phospholamban inhibition, and RyR2 stabilization have demonstrated beneficial effects in preclinical and clinical studies [39,41]. However, the distinct mitochondrial phenotype of HFpEF suggests that therapies targeting _mCa²⁺ handling and redox balance may be particularly important [41,63]. Increasing _mCa²⁺ content improves outcomes in HFrEF but has limited benefits in HFpEF, highlighting fundamental differences in disease biology [35,68]. Modulation of mitochondria-associated membranes and mitochondrial dynamics, including regulation of Drp1-dependent fission, represents an emerging therapeutic avenue [69,70].

These distinctions suggest that HFpEF therapies may need to prioritize restoration of _mCa²⁺ handling, redox balance, and diastolic energetic reserve rather than approaches developed primarily for systolic failure. Cardiomyocyte mitochondrial dysfunction may also influence endothelial and fibroblast responses through redox, energetic, and Ca²⁺-dependent signaling.

3.2. Cardiac fibroblasts

Activated cardiac fibroblasts acquire a mitochondrial phenotype distinct from cardiomyocytes, characterized by altered morphology, metabolic rewiring, and impaired mitochondrial quality control (MQC). These adaptations support myofibroblast differentiation, ECM biosynthesis, and persistence of the profibrotic state.

3.2.1. Mitochondrial remodeling and metabolic adaptation in cardiac fibroblasts

Mitochondrial dynamics are increasingly recognized as important regulators of fibroblast fate. In healthy cells, mitochondrial fusion and fission maintain organellar quality, distribute metabolites, and facilitate adaptation to energetic demands [71,72]. Activated fibroblasts display altered mitochondrial dynamics characterized by increased mitochondrial fragmentation. Excessive Drp1-mediated mitochondrial fission promotes mitochondrial dysfunction and facilitates fibrotic remodeling, whereas preservation of mitochondrial integrity through balanced fusion and fission limits fibroblast activation [71,73]. Consistent with this concept, pharmacological inhibition of mitochondrial fission attenuates cardiac fibrosis in experimental models, supporting a functional role for mitochondrial morphology in regulating fibroblast behavior [73,74].

A characteristic feature of activated fibroblasts is metabolic rewiring toward enhanced glutamine utilization [75]. In addition to supporting ATP production, glutaminolysis replenishes TCA-cycle intermediates required for biosynthetic pathways and collagen production. Experimental inhibition of glutamine metabolism reduces myofibroblast persistence and reverses established cardiac fibrosis, highlighting the importance of mitochondrial metabolic pathways in sustaining fibroblast activation [75]. Furthermore, mitochondrial metabolites such as α-ketoglutarate and acetyl-CoA can influence chromatin remodeling and transcriptional programs, linking mitochondrial metabolism directly to fibroblast cell-fate decisions [76,77]. In parallel, MQC mechanisms are frequently impaired during fibroblast activation. Defective mitophagy allows accumulation of dysfunctional mitochondria with reduced respiratory efficiency and increased oxidative burden, thereby sustaining profibrotic signaling. Restoration of MQC pathways has been shown to suppress fibroblast activation and limit fibrosis in multiple experimental settings, suggesting that maintenance of mitochondrial homeostasis is essential for preventing chronic fibroblast persistence [[78], [79], [80]]. These cell-autonomous adaptations provide the basis for the intercellular mechanisms discussed in Section 4.

3.2.2. Redox signaling and fibrotic activation

Activated fibroblasts and myofibroblasts are the principal mediators of myocardial fibrosis [78,81]. Upon activation, fibroblasts increase synthesis of ECM proteins, including collagen and fibronectin, and can differentiate into myofibroblasts characterized by expression of α-smooth muscle actin, conferring contractile properties. This transition is regulated by both mechanical stress and biochemical signaling pathways. Transforming growth factor-β (TGFβ) is a key profibrotic mediator in HF and is produced by fibroblasts, infiltrating immune cells, and macrophages. Acting in both autocrine and paracrine manners, TGFβ drives transcriptional programs that promote ECM deposition and myofibroblast differentiation. ROS have been shown to mediate TGFβ-induced fibroblast activation and differentiation [82,83], establishing a feedforward signaling axis.

TGFβ signaling further induces expression of NOX4, an intracellular membrane-associated NADPH oxidase, increasing both cytosolic ROS and mitochondria-linked ROS amplification [84,85]. This creates a positive feedback loop in which ROS reinforces TGFβ signaling, thereby promoting fibroblast activation and ECM production. Elevated ROS levels also regulate ECM remodeling through multiple mechanisms, including transcriptional activation of ECM genes, post-translational modification of collagen, and regulation of matrix metalloproteinases (MMPs) [86,87]. Because MMPs are typically secreted in inactive forms and activated by oxidative modification, ROS play a dual role in both ECM deposition and degradation, ultimately controlling matrix composition and stiffness. Impaired MQC pathways may further amplify this process by increasing _mtROS accumulation, thereby reinforcing the feedforward interaction between mitochondrial dysfunction, TGFβ signaling, and myofibroblast persistence [78,82].

3.2.3. Ca²⁺-dependent signaling and mitochondrial integration in fibroblasts

Mechanical stress is transduced into intracellular signaling in fibroblasts through plasma membrane mechanoreceptors, including transient receptor potential (TRP) channels. These Ca²⁺-permeable channels facilitate extracellular Ca²⁺ influx and are key regulators of fibroblast activation and differentiation [88]. Activation of TRP channels promotes downstream signaling through calcineurin-mediated nuclear factor of activated T cells (NFAT) pathways, which drive transcriptional programs associated with fibrosis. TRPV4, a mechanosensitive TRP channel, has been specifically implicated in cardiac fibroblast activation. TRPV4-mediated increases in cytosolic Ca²⁺ promote myofibroblast differentiation through nuclear translocation of myocardin-related transcription factor A (MRTF-A), which regulates expression of α-smooth muscle actin and other profibrotic genes. Sustained Ca²⁺ elevation further activates Ca²⁺-dependent proteases such as calpains and transcriptional regulators, including NFAT, enhancing fibroblast proliferation, collagen synthesis, and ECM deposition [89,90]. Mitochondria contribute to these processes by buffering cytosolic Ca²⁺ and integrating Ca²⁺ signals with metabolic activity. _mCa²⁺ uptake links Ca²⁺ signaling to mitochondrial metabolic adaptation in activated fibroblasts [63,72]. In this way, fibroblast activation is supported by an integrated network of mechanotransduction, Ca²⁺ signaling, mitochondrial metabolism, and redox amplification.

3.2.4. Therapeutic implications for fibroblasts

Cardiac fibrosis represents a double-edged sword in cardiac pathology. While ECM deposition is essential for maintaining structural integrity following injury such as preventing ventricular rupture after myocardial infarction, excessive fibrosis in HFpEF contributes primarily to pathological stiffening without conferring additional functional benefit [78,81]. Unlike HFrEF, HFpEF is not typically associated with myocardial thinning or systolic failure, and thus fibroblast activation and myofibroblast differentiation provide limited compensatory advantage while significantly impairing diastolic function.

Accordingly, targeting fibroblast activation and ECM remodeling represents a promising therapeutic strategy in HFpEF. Modulation of TGFβ signaling, redox balance, and Ca²⁺-dependent pathways may attenuate fibroblast activation and limit fibrosis. Interventions aimed at reducing _mtROS production or restoring mitochondrial function may disrupt the feedforward cycle of oxidative stress and profibrotic signaling [73,82,91]. These approaches highlight the importance of fibroblast-specific mitochondrial pathways as potential targets for mitigating myocardial stiffening and improving diastolic function in HFpEF.

3.3. Cardiac endothelial cells

Cardiac endothelial cells regulate vascular tone, barrier function, microvascular perfusion, inflammatory signaling, and redox balance. In HFpEF, endothelial mitochondrial dysfunction links systemic stress to microvascular dysfunction and impaired myocardial reserve.

3.3.1. Mitochondrial regulation of endothelial homeostasis

Although ECs rely largely on glycolysis for ATP production, endothelial mitochondria are essential regulators of redox balance, metabolic sensing, stress adaptation, and vascular homeostasis. This specialization distinguishes endothelial mitochondria from the highly bioenergetic mitochondrial network of cardiomyocytes. In HFpEF, endothelial mitochondrial remodeling includes altered mitochondrial dynamics, impaired MQC pathways, and disrupted NAD⁺-dependent signaling [8,92].

Maintenance of mitochondrial quality is particularly important in ECs because dysfunctional mitochondria can promote oxidative stress and inflammatory activation. Under physiological conditions, balanced fusion, fission, and mitophagy preserve mitochondrial network integrity and limit the accumulation of damaged organelles. In contrast, impaired mitophagy and excessive mitochondrial fission increase endothelial stress, whereas restoration of mitochondrial turnover improves endothelial resilience and vascular integrity [93,94].

Endothelial mitochondrial function is also linked to NAD⁺ homeostasis and sirtuin signaling. NAD⁺ depletion impairs MQC and enhances inflammatory signaling, while NAD⁺ restoration promotes mitochondrial biogenesis, antioxidant defense, and stress resistance [95]. Thus, endothelial mitochondria support vascular function through MQC, redox regulation, and metabolic signaling, reinforcing the concept of cell-type–specific mitochondrial heterogeneity in HFpEF. Because ECs line the coronary microvasculature, mitochondrial stress in these cells may also influence neighboring fibroblasts, immune cells, and cardiomyocytes through paracrine signaling.

3.3.2. Endothelial mitochondria as regulators of redox and metabolic signaling

Endothelial _mtROS are generated predominantly through ETC dysfunction and are further amplified by impaired mitophagy, linking MQC directly to vascular dysfunction [96]. In cardiometabolic forms of HFpEF, chronic low-grade inflammation associated with obesity, insulin resistance, and metabolic dysfunction has been proposed as a major driver of endothelial mitochondrial stress. Increased _mtROS directly reduces nitric oxide (NO) bioavailability through scavenging and promotes endothelial NOS (eNOS) uncoupling via oxidation of tetrahydrobiopterin. This shifts ECs from a vasodilatory, anti-inflammatory phenotype toward a vasoconstrictive, pro-inflammatory state characterized by impaired NO-soluble guanylate cyclase (sGC)–cGMP signaling [8,62]. Oxidative disruption of mitochondrial redox homeostasis further impairs endothelial responsiveness to metabolic and hemodynamic demand [60,62,96]. Loss of NO signaling is central to the HFpEF microvascular paradigm and contributes to impaired vascular relaxation, increased myocardial stiffness, and downstream fibrotic remodeling.

3.3.3. Ca²⁺ signaling at ER–mitochondria interfaces

Endothelial mitochondrial function is closely linked to Ca²⁺ signaling at endoplasmic reticulum (ER)–mitochondria contact sites. In ECs, Ca²⁺ signals generated at ER microdomains are transmitted to adjacent mitochondria, where _mCa²⁺ uptake regulates local cytosolic Ca²⁺ buffering, metabolic enzyme activity, and ROS production. These ER–mitochondria contacts couple Ca²⁺ dynamics to redox signaling [97,98]. Disruption of this coupling may impair endothelial adaptation to metabolic and inflammatory stress. In HFpEF, inflammatory and cardiometabolic stressors may remodel these Ca²⁺–redox microdomains, promoting pathological Ca²⁺ oscillations, NFAT and NF-κB activation, and further oxidative stress [60,92,99,100].

3.3.4. Endothelial mitochondria, perfusion, and exercise intolerance

Endothelial mitochondrial dysfunction may contribute to HFpEF symptoms by impairing microvascular perfusion and oxygen delivery during stress. Exertional dyspnea and reduced peak oxygen consumption in HFpEF correlate closely with impaired coronary microvascular function and reduced coronary flow reserve [8].

Mitochondrial dysfunction impairs NO-dependent vasomotor responses and perfusion matching during increased metabolic demand, promoting subendocardial ischemia and reducing diastolic reserve during exercise [60]. In parallel, endothelial mitochondrial stress may promote endothelial senescence, impaired angiogenic capacity, capillary rarefaction, and reduced oxygen diffusion. Together, these changes link endothelial mitochondrial dysfunction to impaired perfusion reserve and exercise intolerance in HFpEF.

3.3.5. Therapeutic implications for endothelium

Therapeutic strategies that preserve endothelial mitochondrial function may improve NO signaling, reduce oxidative stress, and support microvascular integrity in HFpEF. Potential approaches include mitochondria-targeted antioxidants, NAD⁺ restoration, and enhancement of MQC pathways. Mitochondria-targeted antioxidants such as MitoQ and SS-31 improve endothelial function and restore NO bioavailability in preclinical models [95,101,102]. NAD⁺ precursors such as nicotinamide riboside enhance mitochondrial function, promote sirtuin-mediated MQC, and reduce inflammatory signaling [95]. Sodium–glucose cotransporter 2 (SGLT2) inhibitors also improve endothelial function and reduce mitochondrial oxidative stress in cardiometabolic disease contexts, suggesting that part of their benefit in HFpEF may involve endothelial mitochondrial pathways [103]. Mitoprotective agents targeting cardiolipin stabilization have similarly improved microvascular integrity in large-animal models. However, because HFpEF is heterogeneous, future studies should define which HFpEF endotypes are most likely to benefit from endothelial mitochondria-targeted therapies.

Collectively, endothelial mitochondrial alterations may contribute to HFpEF through mtROS accumulation, impaired NO signaling, microvascular dysfunction, and paracrine communication. These cell-type–specific endothelial abnormalities provide an important foundation for the intercellular crosstalk mechanisms discussed in Section 4.

4. Intercellular crosstalk

In HFpEF, mitochondrial dysfunction may be conceptualized as a multicellular, network-level phenotype, rather than an isolated defect within cardiomyocytes. ECs, fibroblasts, immune cells, and cardiomyocytes exhibit distinct forms of mitochondrial remodeling that may interact through metabolic, redox, inflammatory, and paracrine pathways. Together, these processes may contribute to myocardial stiffening and diastolic dysfunction in selected HFpEF phenotypes [8,60,104]. In ECs, altered mitochondrial metabolism and increased mtROS reduce NO bioavailability, promoting inflammatory signaling and impaired microvascular–myocyte coupling [60,93]. In parallel, fibroblasts undergo metabolic reprogramming characterized by increased glycolysis, altered glutamine metabolism, and defective mitophagy, supporting myofibroblast activation and ECM deposition. Mitochondrial dysfunction also promotes inflammatory signaling via NLRP3 activation and cytokine release (IL-1β, IL-18), further amplified by IL-6, TNF-α, and TGF-β–mediated ROS generation [99,105].

These mitochondrial disturbances propagate across the myocardium through paracrine signaling, redox-dependent modifications, extracellular vesicles (EVs), and ECM remodeling. EVs containing _mtDNA, oxidized lipids, metabolic enzymes, and small RNAs transmit mitochondrial stress between ECs, fibroblasts, and cardiomyocytes [106,107]. Concurrently, progressive matrix stiffening alters mechanotransduction pathways, feeding back onto _mCa²⁺ handling, substrate utilization, and redox homeostasis across cell types. Impaired mitophagy further amplifies this network by allowing accumulation of ROS-generating mitochondria, reinforcing intercellular dysfunction [108]. Collectively, these processes synchronize metabolic inflexibility across the myocardium, resulting in reduced oxidative reserve, impaired substrate switching, and sustained redox imbalance (Fig. 2).

Fig. 2.

Intercellular mitochondrial crosstalk driving HFpEF pathogenesis. Systemic cardiometabolic stress disrupts _mCa²⁺ handling, redox balance, and energetic adaptation across cardiomyocytes, endothelial cells, fibroblasts, and immune cells. Endothelial mitochondrial dysfunction reduces NO signaling and promotes inflammation; fibroblast metabolic remodeling drives ECM deposition and stiffness; cardiomyocyte mitochondrial dysfunction impairs relaxation; and immune activation amplifies inflammatory signaling. These cell-type-specific abnormalities interact through mtROS, _mCa²⁺ signaling, EVs, paracrine mediators, and mechanotransduction, establishing feedback loops that promote microvascular dysfunction, fibrosis, diastolic dysfunction, and exercise intolerance.

Collectively, these observations support a model in which mitochondrial dysfunction across interacting cardiac cell populations may contribute to myocardial remodeling and disease progression in HFpEF. We propose that coordinated disturbances in mitochondrial signaling, metabolism, and redox homeostasis represent a useful mechanistic framework for understanding intercellular contributions to HFpEF pathogenesis. However, given the substantial heterogeneity of HFpEF and the limited concordance among experimental models and human studies, this framework should currently be viewed as a hypothesis-generating paradigm rather than a defining feature of the syndrome.

4.1. Endothelial–fibroblast signaling in fibrosis

Endothelial dysfunction links systemic comorbidities to myocardial remodeling in HFpEF. Chronic inflammation and oxidative stress impair the coronary microvasculature, reducing NO signaling and creating a pro-fibrotic environment [60]. Endothelial mitochondrial dysfunction amplifies this process through excess mtROS generation and disruption of redox-sensitive signaling pathways, accelerating NO depletion and suppressing NO–sGC–cGMP–PKG activity. Loss of this anti-fibrotic signaling promotes fibroblast activation, proliferation, and collagen synthesis [109]. In addition, injured ECs secrete profibrotic mediators, including TGF-associated factors and matricellular proteins such as CTGF/CCN2, which further stimulate fibroblasts and ECM accumulation [110].

A particularly important endothelial contribution is endothelial-to-mesenchymal transition (EndMT). Under inflammatory and oxidative conditions (and in response to TGF-β family signaling), subsets of ECs downshift endothelial identity and acquire mesenchymal/myofibroblast-like features, increasing the pool of matrix-producing cells and enhancing pro-fibrotic paracrine output. EndMT has been demonstrated as a contributor to cardiac fibrosis in foundational experimental work and has also been discussed specifically in HFpEF-related contexts (including cardiorenal/HFpEF frameworks) [111]. Once activated, fibroblasts can further worsen endothelial and whole-tissue mitochondrial stress through ECM overproduction and architectural remodeling. Collagen accumulation, crosslinking, and expansion of the interstitial matrix increase myocardial stiffness and distort microvascular interstitial geometry, impairing oxygen and metabolite diffusion and raising mechanical load on resident cells, conditions that favor _mtROS generation and maladaptive stress signaling [112].

This sets up a classic mechanotransduction loop: fibroblasts sense the stiffened matrix and translate it into sustained transcriptional programs (notably through mechanosensitive pathways such as YAP/TAZ and related nodes), which further stabilize myofibroblast identity and collagen synthesis. Accumulating evidence demonstrates that ECM stiffness modulates the activation state of cardiac fibroblasts, whereas inhibition of YAP-mediated mechanosignaling reduces profibrotic remodeling and cardiac fibrosis in vivo [113,114]. The net result may be a reinforced HFpEF mechanical phenotype: progressive ECM deposition and mechanosensitive reinforcement drive a myocardium with high passive stiffness and impaired relaxation, while ongoing endothelial dysfunction sustains the pro-fibrotic drive. This is why HFpEF fibrosis often behaves less like a transient scar response and more like a self-maintaining remodeling program organized by endothelial-fibroblast crosstalk [60,112].

4.2. Disrupted fibroblast–cardiomyocyte metabolic coupling

In the healthy myocardium, cardiac fibroblasts function not only as matrix-producing cells; they are metabolic partners that help stabilize the cardiomyocyte environment. One well-described example is metabolite shuttling: fibroblasts can increase glycolytic flux and export lactate, while cardiomyocytes import lactate via monocarboxylate transporters and oxidize it as a fuel, conceptually similar to the astrocyte-neuron lactate shuttle. This has been supported by proteomic and co-culture evidence for a fibroblast-to-cardiomyocyte lactate shuttle, and by mechanistic studies showing that interrupting lactate transport (e.g., MCT1-related steps) can blunt adverse remodeling in stress settings such as hypertension [115,116]. Classic studies have demonstrated metabolic coupling of glutathione between cardiac cells, highlighting the capacity of the myocardium to engage in intercellular antioxidant cooperation during oxidative stress [117].

In HFpEF (and in pressure overload/aging states that model key HFpEF features), this supportive coupling shifts because fibroblasts undergo mitochondrial and metabolic rewiring during activation. A prominent and reproducible feature of fibroblast activation is a rise in aerobic glycolysis and lactate production (Warburg-like remodeling), which supports proliferation, ECM synthesis, and myofibroblast contractile programs [78]. HFpEF progression is associated with broader myocardial metabolic remodeling, including impaired oxidative metabolism and altered amino acid/lipid utilization, which contribute to reduced metabolic flexibility, and increased reliance on adaptive redox-buffering mechanisms [65].

Activated fibroblasts undergo metabolic remodeling that alters their interaction with neighboring cardiomyocytes [75,115]. Multi-omics studies of cardiac fibrosis indicate that metabolic reprogramming, including impaired oxidative metabolism and altered substrate utilization, promotes fibroblast activation programs that reinforce extracellular matrix deposition, tissue stiffening, and adverse cardiac remodeling [109].

Together, these shifts bias the tissue toward an energetically expensive remodeling state: fibroblasts prioritize biomass production (collagen, ECM) and stress adaptation, while cardiomyocytes face reduced metabolic flexibility and higher energetic burden during diastole, when ATP demand for relaxation and Ca²⁺ reuptake is substantial. The net effect is that what is normally metabolic “support” becomes maladaptive coupling that worsens diastolic energetic insufficiency and helps lock in HFpEF physiology.

4.3. _mCa²⁺ and ROS act as intercellular signaling currencies

Beyond their intracellular roles, _mCa²⁺ and ROS can function as intercellular signaling mediators. In HFpEF, dysregulated Ca²⁺–ROS coupling may propagate stress signals between cardiomyocytes, endothelial cells, fibroblasts, and immune cells, thereby linking cell-type-specific mitochondrial dysfunction to tissue-level remodeling [8,63,118].

Importantly, _mtROS do not remain confined within the originating cell. Low to moderate ROS levels act as diffusible second messengers, modulating redox sensitive kinases, phosphatases, and transcription factors in neighboring cells [63]. Thus, _mtROS serve as paracrine amplifiers of myocardial remodeling. Beyond simple diffusion, stressed mitochondria contribute to intercellular communication through EVs and mitochondrial component release. Cardiomyocytes, ECs, and fibroblasts can package mtDNA, oxidized lipids, and metabolic enzymes into EVs, which are taken up by neighboring cells and activate innate immune or stress pathways [63,106,107,119]. mtDNA-containing vesicles can activate pattern-recognition receptors such as TLR9 and cGAS-STING, linking mitochondrial oxidative stress to sterile inflammation, a hallmark of HFpEF [120].

Together, _mCa²⁺ and ROS act as transferable stress signals that connect endothelial dysfunction, fibroblast activation, and cardiomyocyte relaxation defects. Endothelial _mtROS, fibroblast-derived redox signals, matrix stiffening, and cardiomyocyte _mCa²⁺ overload may reinforce one another through a multicellular feedback network involving ROS output and inflammatory EV release. Because systolic ATP generation may remain relatively preserved in early HFpEF, these redox-mediated disturbances are likely to preferentially impair diastolic energetics and relaxation rather than produce immediate contractile failure [60,63,119]. Thus, dysregulated _mCa²⁺ handling and oxidative stress may shift intercellular mitochondrial signaling from adaptive communication toward maladaptive, tissue-level dysfunction.

4.4. HFpEF heterogeneity and translational considerations

Cardiometabolic HFpEF models dominate the available mechanistic literature and have helped define abnormalities in _mCa²⁺ handling, redox signaling, endothelial dysfunction, and metabolic remodeling. However, mitochondrial adaptation remains less defined in aging-associated, hypertensive, renal-associated, or frailty-dominant HFpEF. Thus, mechanisms identified in one HFpEF endotype should not be assumed to generalize across the broader HFpEF spectrum. Integrating human myocardial datasets with comparative model analyses will be essential to distinguish conserved mitochondrial pathways from phenotype-specific adaptations and guide precision therapeutics [8,121,122].

Not all proposed mitochondrial abnormalities are consistent across HFpEF studies. For example, recent human cardiometabolic HFpEF analyses reported preserved fatty acid oxidation and minimal NAD⁺/NADH changes, contrasting with several experimental models and some human cohorts. These discrepancies underscore the need for caution when extrapolating model-derived mitochondrial mechanisms to the broader HFpEF population [118,123,124].

5. Therapeutics targeting the mitochondrial redox–calcium–energetics axis in HFpEF

Currently, no curative therapy exists for HFpEF, and management remains largely focused on symptom relief and reducing hospitalization risk. However, increasing evidence implicates mitochondrial dysfunction as an important contributor to disease progression, prompting the development of therapeutic strategies that target mitochondrial pathways across multiple cardiac cell populations. Importantly, mitochondrial abnormalities in HFpEF are cell-type-specific, with cardiomyocytes exhibiting impaired calcium handling and bioenergetics, fibroblasts undergoing profibrotic metabolic reprogramming, and endothelial cells developing redox imbalance and NO deficiency. Consequently, emerging therapies act through distinct but overlapping mechanisms in these cell types, including reduction of mitochondrial oxidative stress, restoration of mitochondrial energetics and metabolic flexibility, modulation of _mCa²⁺ signaling, preservation of mitochondrial structure and MQC, attenuation of fibroblast activation, and improvement of endothelial function. Table 2 summarizes the mitochondrial effects of these therapeutic approaches across cardiomyocytes, fibroblasts, and ECs, highlighting how modulation of shared mitochondrial pathways may influence multiple contributors to HFpEF pathogenesis.

Table 2.

Mitochondria-targeted therapeutic strategies in HFpEF and their cell-type-specific effects across the cardiac microenvironment. The table summarizes representative therapies that target mitochondrial dysfunction in cardiomyocytes, cardiac fibroblasts, and ECs. Highlighted mechanisms include modulation of _mCa²⁺ signaling, redox balance, bioenergetics, MQC, fibrotic remodeling, and endothelial NO signaling. These therapies demonstrate how restoration of mitochondrial function across diverse cardiac cell populations converges on the redox–calcium–energetics axis to mitigate key pathological features of HFpEF, including diastolic dysfunction, myocardial fibrosis, microvascular impairment, and reduced energetic reserve. Evidence is derived from experimental and translational studies in HFpEF and related cardiovascular disease models.

Drug	Mitochondrial effects of therapeutic agents across cardiac cell types
	Cardiomyocytes	Fibroblasts	ECs
Elamipretide (SS-31)	Restores ATP production, reduces electron leakage and mtROS, stabilizes cardiolipin [125,126]	Reverses mitochondrial fragmentation, reduces mtROS [82,127]	Restores mitochondrial density and cardiolipin; improves microvascular integrity; anti-inflammatory [102,128]
MitoQ	Reduces hypertrophy; decreases oxidative stress [82,101]	Inhibits TGF-β1 signaling and profibrotic activation via Nrf2 preservation [82]	Improves NO bioavailability [101]; prevents mitochondrial oxidative damage and apoptosis [129]
Coenzyme Q10	Enhances ETC function, preserves ΔΨm, reduces ROS and apoptosis; improves I/R injury outcomes [130]	Attenuates adverse remodeling and fibrosis-associated signaling [131]	Improves endothelial function and reduces oxidative stress through mitochondrial protective signaling pathways [132]
Vitamin E	Reduces lipid peroxidation and oxidative stress while preserving mitochondrial integrity during ischemia-reperfusion injury [133]	Reduces oxidative stress and apoptosis while improving fibroblast survival under stress conditions [134]	Mitochondria-targeted antioxidant forms reduce oxidative damage and apoptosis in endothelial cells [129]
Semaglutide (GLP-1RA)	Improves mitochondrial energetics, mitochondrial quality control, and mitochondrial structural integrity [135]	Reduces profibrotic remodeling and extracellular matrix accumulation through metabolic reprogramming [136]	Reduces prothrombotic and inflammatory signaling; improves vascular homeostasis and endothelial phenotype [103,136]
Tirzepatide (GIP/GLP-1RA)	Preserves cardiac systolic and diastolic function, improves mitochondrial bioenergetics, preserves mitochondrial structure, and reduces oxidative stress [137,138]	Potential antifibrotic and anti-inflammatory effects through metabolic remodeling pathways [139]	Improves endothelial function and endothelial metabolic homeostasis [140]
Nicotinamide mononucleotide (NMN)	Restores NAD+ pools, improves mitochondrial homeostasis, and attenuates heart failure progression [141]	Potential antifibrotic effects through restoration of NAD+-dependent mitochondrial signaling [142]	Enhances NAD+-dependent mitochondrial signaling and improves resistance to oxidative stress [95,143]
Berberine	Modulates mitochondrial Ca2+ uptake via MCU inhibition, limits mCa2+ overload, and attenuates mitochondrial dysfunction and apoptosis [144,145]	Attenuates profibrotic remodeling and fibrosis-associated signaling pathways [146]	Improves endothelial function and attenuates vascular dysfunction in experimental HFpEF models [147,148]
Icariin	Reduces oxidative stress and supports mitochondrial function [149]	Attenuates fibrosis-associated oxidative stress and profibrotic remodeling [149]	Improves endothelial function and promotes eNOS/NO signaling [149]
Astragaloside IV	Improves myocardial metabolism and mitochondrial function while reducing inflammation in HFpEF models [150]	Potentially reduces oxidative stress and supports mitochondrial biogenesis, mechanisms that may contribute to antifibrotic effects [151]	Restores mitochondrial homeostasis and NO signaling via AMPK/SIRT1-dependent pathways [152]
Mdivi-1 (Drp1 inhibitor)	Inhibits Drp1-mediated mitochondrial fission and reduces cardiomyocyte injury and cell death [153]	Reduces fibrosis-associated remodeling by inhibiting Drp1-mediated mitochondrial fission and oxidative stress [154]	Improves endothelial function by inhibiting Drp1-mediated mitochondrial fission and reducing oxidative stress [94]

5.1. Targeting mitochondrial ROS and redox imbalance

Excess _mtROS contributes to oxidative damage, calcium dysregulation, endothelial dysfunction, and energetic impairment in HFpEF. Coenzyme Q10 (CoQ10), MitoQ, and vitamin E have been reported to improve endothelial function and restore NO bioavailability in preclinical models. CoQ10 enhances electron transport chain activity, preserves ΔΨm, improves endothelial respiration, and reduces inflammation and thrombosis [101,132,133].

MitoQ reduces oxidative stress, preserves Nrf2 signaling and endothelial function, and NO bioavailability, and protects against mitochondrial damage and apoptosis. Vitamin E limits lipid peroxidation, ROS accumulation, preventing oxidative injury. Collectively, these agents improve mitochondrial function and cellular resilience across cardiomyocytes, fibroblasts, and endothelial cells [82,101,129,133,134].

5.2. Targeting mitochondrial bioenergetics and NAD⁺ metabolism

Impaired substrate utilization and NAD⁺ depletion contribute to mitochondrial dysfunction in HFpEF. Restoration of intracellular NAD⁺ pools through precursors such as nicotinamide mononucleotide (NMN), activates sirtuin-dependent pathways, improves mitochondrial respiration, enhances bioenergetics, and suppresses oxidative stress and adverse cardiac remodeling [95,141,142].

Similarly, semaglutide and tirzepatide may improve mitochondrial metabolism, respiration, calcium handling, mitophagy, and cellular energetics while reducing inflammation, fibrosis, and endothelial dysfunction [135,136,140,155]. Together, these therapies restore mitochondrial energy homeostasis and metabolic adaptability in HFpEF.

5.3. Targeting mitochondrial structure and dynamics

Maintenance of mitochondrial architecture is essential for efficient ATP production and calcium homeostasis. Elamipretide stabilizes cardiolipin, restores ATP production, reduces electron leak and mtROS generation, and improves mitochondrial density and microvascular integrity [102,127,128]. Mdivi-1 inhibits Drp1-mediated mitochondrial fission, limiting mPTP opening, oxidative stress, and cell death [94,153,154]. Berberine modulates _mCa²⁺ homeostasis by inhibiting MCU, thereby limiting _mCa²⁺ overload and protecting against mitochondrial dysfunction and cardiac injury. Additional studies suggest that berberine improves MQC, attenuates oxidative stress, and exerts antifibrotic effects in cardiovascular disease models [[144], [145], [146]]. Collectively, these agents preserve mitochondrial structure, MQC, and functional integrity in HFpEF.

5.4. Emerging natural compounds modulating mitochondrial signaling

Natural compounds with mitochondrial actions have emerged as potential adjunctive therapies in HFpEF. Icariin enhances mitochondrial function through antioxidant and Nrf-dependent mechanisms, increases eNOS-derived NO production, and attenuates fibrosis and oxidative stress [149]. Astragaloside IV restores ATP production, NAD⁺ levels, complex I activity, mitochondrial biogenesis, and mitochondrial homeostasis while activating AMPK/SIRT1-dependent pathways and improving NO signaling [[150], [151], [152]]. Together, these compounds strengthen MQC, redox balance, and endothelial function while limiting fibrotic remodeling.

6. Conclusion and future perspective

HFpEF is a complex and heterogeneous syndrome in which mitochondrial dysfunction may represent an important mechanistic link among cardiometabolic stress, inflammation, microvascular dysfunction, fibrosis, and impaired diastolic reserve. Beyond their canonical role in ATP production, mitochondria serve as dynamic regulators of Ca²⁺ signaling, redox homeostasis, metabolic flexibility, and intercellular communication across cardiomyocytes, fibroblasts, and ECs. Disruption of this integrated redox–Ca²⁺–energetics axis may impair relaxation, promote myocardial stiffening, and reduce energetic reserve despite preserved systolic function. A major theme emerging from this review is that mitochondrial dysfunction in HFpEF is not restricted to cardiomyocytes. Instead, cell-type-specific mitochondrial abnormalities may interact through redox signaling, paracrine mediators, EVs, immune activation, and ECM remodeling. These multicellular interactions provide a useful framework for understanding how mitochondrial stress in one cardiac cell population may influence neighboring cells and contribute to progressive myocardial dysfunction.

Therapeutically, mitochondria represent promising but complex targets in HFpEF. Approaches aimed at reducing mtROS, restoring NAD⁺ metabolism, improving MQC, modulating _mCa²⁺ handling, and enhancing endothelial or fibroblast mitochondrial function may hold translational potential. However, because HFpEF encompasses diverse phenotypes, broad mitochondria-targeted strategies may not be equally effective across all patients. Future therapies will likely need to be tailored to specific HFpEF endotypes and to the dominant cell-type-specific mitochondrial abnormalities present in each disease context.

A key remaining challenge is determining which mitochondrial abnormalities are conserved across HFpEF endotypes and which are model- or phenotype-specific [112]. Future studies integrating human myocardial datasets with comparative experimental models, single-cell and spatial multi-omics, advanced metabolic imaging, mitochondrial functional profiling, and computational patient stratification will be essential for defining clinically relevant mitochondrial targets. Such approaches may enable biomarker-guided precision medicine and identify patients most likely to benefit from mitochondria-targeted therapies. Ultimately, coordinated mitochondrial dysfunction should be viewed as a promising mechanistic framework rather than an established unifying mechanism for all HFpEF. Continued validation in human studies will be necessary to determine whether targeting mitochondrial pathways across interacting cardiac cell populations can provide meaningful disease modification in HFpEF.

CRediT authorship contribution statement

Anupriya Sinha: Writing – original draft, Visualization, Investigation. Devin Kolmetzky: Writing – original draft. Dhanendra Tomar: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization. Pooja Jadiya: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this manuscript, the authors used Microsoft Copilot exclusively for language refinement and readability enhancement. All generated content was carefully reviewed and edited by the authors, who assume full responsibility for the accuracy and integrity of the final manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data was used for the research described in the article.