Mitochondrial Calcium Regulation of Cardiac Metabolism in Health and Disease

Abstract

Oxidative phosphorylation is regulated by mitochondrial calcium (Ca²⁺) in health and disease. In physiological states, Ca²⁺ enters via the mitochondrial Ca²⁺ uniporter and rapidly enhances NADH and ATP production. However, maintaining Ca²⁺ homeostasis is critical: insufficient Ca²⁺ impairs stress adaptation, and Ca²⁺ overload can trigger cell death. In this review, we delve into recent insights further defining the relationship between mitochondrial Ca²⁺ dynamics and oxidative phosphorylation. Our focus is on how such regulation affects cardiac function in health and disease, including heart failure, ischemia-reperfusion, arrhythmias, catecholaminergic polymorphic ventricular tachycardia, mitochondrial cardiomyopathies, Barth syndrome, and Friedreich’s ataxia. Several themes emerge from recent data. First, mitochondrial Ca²⁺ regulation is critical for fuel substrate selection, metabolite import, and matching of ATP supply to demand. Second, mitochondrial Ca²⁺ regulates both the production and response to reactive oxygen species (ROS), and the balance between its pro- and antioxidant effects is key to how it contributes to physiological and pathological states. Third, Ca²⁺ exerts localized effects on the electron transport chain (ETC), not through traditional allosteric mechanisms but rather indirectly. These effects hinge on specific transporters, such as the uniporter or the Na⁺/Ca²⁺ exchanger, and may not be noticeable acutely, contributing differently to phenotypes depending on whether Ca²⁺ transporters are acutely or chronically modified. Perturbations in these novel relationships during disease states may either serve as compensatory mechanisms or exacerbate impairments in oxidative phosphorylation. Consequently, targeting mitochondrial Ca²⁺ holds promise as a therapeutic strategy for a variety of cardiac diseases characterized by contractile failure or arrhythmias.

Introduction

Calcium ions (Ca²⁺) are fundamental for a multitude of cellular processes, allowing the rapid transmission and interconversion between electrical and biochemical signals during electrical conduction, muscular contraction, and metabolic signaling (1). The endoplasmic reticulum (ER) is the primary Ca²⁺ store in the cell (200–500 µM free Ca²⁺), whereas levels within the mitochondrial matrix are lower (<100 µM free Ca²⁺, though the resting levels have been the topic of some controversy) (2–5). Thus, when released from the ER, Ca²⁺ levels rapidly increase in nearby mitochondria. Within mitochondria, Ca²⁺ signaling has long been recognized to increase NADH, double ATP synthesis rates, and regulate reactive oxygen species (ROS) production (6–8). In this review, we focus on more recent investigations conducted in murine models of normal or diseased cardiac function that have expanded our understanding of mitochondrial Ca²⁺ signaling in relation to oxidative phosphorylation (OXPHOS) and oxidative stress. Beyond direct binding and allosteric effects on mitochondrial enzymes, these studies have revealed a multitude of indirect mechanisms that fine-tune Ca²⁺ regulation of OXPHOS. Notably, these novel forms of regulation often depend on local interactions between Ca²⁺ transporters and mitochondrial complexes, emphasizing the localized nature of Ca²⁺ signaling (FIGURE 1). Over the course of the review, we also highlight critical unanswered questions and controversies regarding the mechanisms by which Ca²⁺ regulates OXPHOS.

FIGURE 1.

The localized nature of mitochondrial Ca²⁺ signaling In the heart, Ca²⁺ transfer from sarcoplasmic reticulum (SR) to mitochondria occurs at sites of close contact between the SR and mitochondria, allowing rapid transport through RyR2 (cardiac ryanodine receptor), voltage-dependent anion channel (VDAC), and MCU [mitochondrial Ca²⁺ uniporter (pore-forming subunit)] into the mitochondrial matrix. Ca²⁺ export back to the SR occurs in a more distal location, with Ca²⁺ exchange for Na⁺ via NCLX (Na⁺/Ca²⁺ exchanger) and finally uptake back into the SR via sarco(endo)plasmic reticulum Ca²⁺ ATPase (SERCA) pumps. IMM, inner mitochondrial membrane; IMS, intermembrane space; MICU1/2, mitochondrial Ca²⁺ uptake (gating subunit); OMM, outer mitochondrial membrane.

Mitochondrial Ca²⁺ Transporters and Their Spatial Distribution

An overview of the molecules involved in mitochondrial Ca²⁺ transport is included in Table 1.

Table 1.

Key components of Ca²⁺ transporting proteins

Gene	Protein	Tissue Expression	Localization	Comments	Reviews/ References
VDAC1	Voltage-Dependent Anion Channel 1	Ubiquitous, abundant in heart, liver, kidney, and brain	OMM	Responsible for Ca2+ transport across the OMM. Regulated by cholesterol, fatty acids, and NADH	(9, 10)
VDAC2	Voltage-Dependent Anion Channel 2	Heart, skeletal muscle, brain, lung, pancreas	OMM	Responsible for Ca2+ transport across the OMM. Targeted by cyclophilin D in the mitochondrial permeability transition pore opening	(9, 10)
VDAC3	Voltage-Dependent Anion Channel 3	Testis, spleen, ovary, brain	OMM	Responsible for Ca2+ transport across the OMM. Regulated by ceramide, calcium, and pH response to oxidative stress, altered mitochondrial respiration	(9, 10)
MCU	Mitochondrial Calcium Uniporter	Ubiquitous, high levels in heart, muscle, and brain	IMM	Pore-forming subunit of the mitochondrial Ca2+ uniporter, allowing Ca2+ influx across the IMM. Inhibited by Ru360, ruthenium red, mitoxantrone, and others	(11, 12)
MCUB	Mitochondrial Calcium Uniporter dominant negative subunit beta	Broad expression. Low levels in heart but upregulated during injury.	IMM	Homologous to MCU but inhibits Ca2+ uptake when incorporated into the channel pore in a dominant-negative fashion	(11–14)
MICU1	Mitochondrial Calcium Uptake 1	Ubiquitous, highest in heart, muscle, and brain	IMS	Component of the uniporter. Acts as a gatekeeper, modulating MCU activity in response to Ca²+ levels	(11, 12)
MICU2	Mitochondrial Calcium Uptake 2	Ubiquitous	IMS	Component of the uniporter. Collaborates with MICU1 in regulating MCU activity	(11, 12)
MICU3	Mitochondrial Calcium Uptake 3	Ubiquitous	IMS	Component of the uniporter. Collaborates with MICU1 in regulating MCU activity	(11, 12)
SMDT1 (EMRE)	Essential MCU regulator	Ubiquitous	IMM	Transmembrane protein that surrounds MCU, opening the channel and helping bridge the interaction with MICU1	(11, 12)
MCUR1	Mitochondrial calcium uniporter regulator 1	Ubiquitous	IMM	Positive regulator of mitochondrial Ca2+ uptake through the uniporter	(11, 12)
SLC8B1 (NCLX)	Mitochondrial Na+/Ca2+ exchanger	Ubiquitous, high expression in heart, brain, and muscle	IMM	Mediates the efflux of Ca2+ from the matrix via exchange with Na+. Helps prevent mitochondrial Ca2+ overload	(15, 16)
LETM1	Leucine zipper and EF-hand containing transmembrane protein 1	Ubiquitous	IMM	Mediates the efflux of Ca2+ from the matrix via exchange with H+. May also be involved in K+/H+ exchange	(17, 18)
TMEM65	Transmembrane protein 65	Widely expressed, with higher levels in brain, heart, and skeletal muscle	IMM	New potential component of Na+/Ca2+ exchanger	(19–21)
TMBIM5	Transmembrane BAX inhibitor motif-containing protein 5	Ubiquitous expression, higher levels in brain, heart, and muscle tissues	IMM	New potential component of the Ca2+/H+ exchanger	(22–24)

IMM, inner mitochondrial membrane; IMS, intermembrane space; OMM, outer mitochondrial membrane.

Mitochondrial Outer Membrane

Voltage-dependent anion channel.

The outer membrane of the mitochondria is not felt to be a significant barrier to calcium diffusion because of the presence of voltage-dependent anion channel (VDAC) channels. These channels have an array of antiparallel beta strands that create a large pore (9). Even when these channels constrict to block diffusion of larger molecules, small ions such as Ca²⁺ can diffuse through them freely. Although the channels themselves do not form critical barriers to Ca²⁺ diffusion, by altering their localization across the outer membrane they can facilitate Ca²⁺ transfer from sites of release in the ER or sarcoplasmic reticulum (SR) to the inner membrane (10). In fact, cardiac-specific loss of VDAC isoforms can lead to dilated cardiomyopathy due to insufficient Ca²⁺ transfer to mitochondria (25). Thus, processes that alter the clustering of VDAC channels can also alter the rates of Ca²⁺ influx into the mitochondria.

Mitochondrial Inner Membrane Ca²⁺ Import

Mitochondrial Ca²⁺ uniporter.

High-capacity Ca²⁺ influx, ascribed to a uniporter driven by mitochondrial membrane potential (ΔΨ), occurs through a highly selective Ca²⁺ channel (26–28). Whereas initial efforts with limited control over ΔΨ and Ca²⁺ buffering had suggested that this pathway was low affinity (29), more precise recent measurements have shown the ability of the channel to conduct Ca²⁺ at submicromolar levels, especially in the heart (30–33). Efforts in the past decade culminated with the discovery of the genes encoding this multisubunit channel, known as the mitochondrial Ca²⁺ uniporter (11, 34–40). From here onward, we refer to the entire channel as the “uniporter,” whereas individual subunits {e.g., MCU [mitochondrial Ca²⁺ uniporter (pore-forming subunit)]} are specified by their protein or gene abbreviation. The pore-forming subunit is encoded by the MCU gene. Recent structural studies revealed a tetrameric organization of MCU, centered around a narrow pore lined with concentric rings of glutamate residues that create a highly negatively charged selectivity filter with nanomolar affinity for Ca²⁺ (12, 41–46). A highly homologous subunit, MCUB, appears to act in a dominant-negative manner, inhibiting Ca²⁺ transport in channels into which it incorporates (39). In metazoans, the transmembrane protein EMRE (SMDT1 gene) surrounds the MCU channel, helping to open the pore where it exits into the matrix and bridging interactions between MCU and the accessory MICU1 subunits (45, 47, 48). The MICU1–3 subunits are located in the intermembrane space, contain Ca²⁺-binding EF-hands, gate the channel by blocking the pore, and help the channel dimerize (43, 49). Micromolar Ca²⁺ binds to the MICU EF-hands, releasing them from the pore, allowing Ca²⁺ passage and turning the uniporter into a Ca²⁺-gated Ca²⁺ channel. These core components of the uniporter can also bind to other mitochondrial proteins to regulate Ca²⁺ import. Although compelling evidence described below indicates that genetic deletion of MCU appears to largely prevent Ca²⁺ import, basal mitochondrial Ca²⁺ levels are often only mildly affected. It may be that other transporters, either the known Ca²⁺ exchangers described below or others, may compensate (50, 51).

As with VDAC channels on the outer membrane, the spatial localization of the uniporter helps regulate mitochondrial metabolism. MICU1 subunits appear restricted to the boundary membrane, the portion of the inner mitochondrial membrane closest to the outer mitochondrial membrane; MCU itself is more homogeneously distributed and is found in cristae as well (52). During Ca²⁺ signals, the distribution of these two shows much greater overlap. Superresolution microscopy has revealed that the uniporter tends to cluster at the edges of cardiomyocyte mitochondria in the vicinity of T tubules and the sarcoplasmic reticulum (SR), where cardiac Ca²⁺ signals originate (53). Thus, within the heart, the uniporter appears to be clustered at sites of active Ca²⁺ signaling, localizing close to the outer membrane VDAC and SR-embedded ryanodine receptors (RyRs) and guaranteeing seamless Ca²⁺ transfer from SR to mitochondrial matrix (FIGURE 1).

Outstanding questions: 1) From recent structural studies, it is clear that uniporter channel dimerizes via both the MCU NH₂-terminal domain and MICU1–3 subunits. This may allow cooperative activation, with Ca²⁺ binding at only one set of MICU subunits, allowing opening of the adjacent channel as well, though this has not been demonstrated functionally and any impact on OXPHOS remains unexplored. 2) MCUB inhibits Ca²⁺ transport when incorporated into uniporter channels, but the mechanism of inhibition remains unclear. When incorporated into lipid bilayers, it may form Na⁺ channels, but how it inhibits the native uniporter complex remains unanswered. 3) Given the restricted localization of MICU subunits relative to the homogeneous MCU distribution in the inner membrane, what is the mechanism for MCU-MICU redistribution during Ca²⁺ signals, and how does this affect the time course of uptake during Ca²⁺ signaling?

Mitochondrial Inner Membrane Ca²⁺ Export

Na⁺/Ca²⁺ exchange.

Ca²⁺ export from the mitochondrial matrix has not been studied to the degree that Ca²⁺ import has. However, there are two well-characterized pathways mediating such export. One pathway involves the influx of Na⁺ ions in exchange for Ca²⁺ efflux in a possibly electroneutral ratio (54, 55). The exchanger is encoded by the NCLX protein (SLC8B1 gene) (15). Loss of this transmembrane protein is lethal within cardiomyocytes, likely because of overwhelming Ca²⁺ overload when this efflux pathway is blocked (16). The spatial distribution of NCLX may also contribute to its effects on Ca²⁺ regulation. Within cardiomyocytes NCLX seems to be distributed further away from sources of Ca²⁺, and closer to the sarco(endo)plasmic Ca²⁺-ATPase (SERCA), which takes up Ca²⁺ back into the SR (56) (FIGURE 1). In recently posted preprints, TMEM65 also appears to contribute to Na⁺/Ca²⁺ exchange, though it is unclear whether this protein interacts with NCLX or encodes an entirely separate pathway (19–21).

Ca²⁺/H⁺ exchange.

The second pathway for Ca²⁺ export from the mitochondrial matrix involves the electroneutral exchange of Ca²⁺ for H⁺ via the Ca²⁺/H⁺ exchanger. One protein mediating Ca²⁺/H⁺ exchange has been identified as LETM1, though it may also be involved in the exchange of K⁺ and H⁺ (17, 18, 57–60). Within heart cells, despite expression of the protein and closely related homologs, this activity is thought to be less important for the export of Ca²⁺ compared with NCLX, since it cannot compensate for the loss of NCLX in knockout (KO) mice (16). Whether LETM1 homologs (LETMD1 and LETM2) are also involved in Ca²⁺/H⁺ exchange has also not been well established. Although loss of LETM1 is lethal, the precise consequences of altered Ca²⁺/H⁺ exchange on mitochondrial mechanisms are not well understood, nor do we know whether there is heterogeneity in spatial distribution along the mitochondrial inner membrane. Finally, very recent reports have also identified TMBIM5 as critical for Ca²⁺/H⁺ exchange, potentially via direct interaction with LETM1 (22–24), though the precise mechanisms by which it contributes to Ca²⁺/H⁺ exchange remain to be established.

Mitochondrial Ca²⁺ Buffering

Some of the most poorly understood aspects of mitochondrial Ca²⁺ physiology are the identities, mechanisms, and localization of matrix Ca²⁺ buffers. The critical difficulty appears to be that, unlike the ER, there are no well-characterized high-capacity Ca²⁺ binding proteins in the mitochondrial matrix that would allow genetic or pharmacological modification. Nonetheless, the mitochondrial matrix has remarkable capacity to buffer Ca²⁺ influx, with total matrix Ca²⁺ thought to be 10³- to 10⁵-fold greater than free Ca²⁺ concentrations (61–63). This substantial buffering power comes from the presence of amorphous calcium phosphate, a substance strongly regulated by pH (64–66). Within the matrix, phosphate can oligomerize into complex polymers with high-energy bonds and high capacity for binding Ca²⁺. Cleavage of these polymers in cultured cells can reduce ΔΨ, increase NADH, and produce a shift away from OXPHOS toward glycolysis, though whether these effects depend on altering Ca²⁺ levels is unclear (67, 68). In addition, Ca²⁺ phosphate complexes can form insoluble granules within the matrix (62). How these localized structures regulate Ca²⁺ physiology remains difficult to study experimentally, though in modeling studies they may interfere with Complex I-dependent respiration (69).

Mitochondrial Ca²⁺ buffering can potentially alter the timing and strength of cytoplasmic Ca²⁺ signals. These altered cytoplasmic Ca²⁺ levels can engage a variety of signaling pathways, such as 5′ AMP-activated protein kinase or the cAMP response element-binding protein, in a variety of tissues (70, 71). Such effects have been reviewed elsewhere (72–74). In the heart, Ca²⁺ extrusion from the cytoplasm is dominated by transporters at the plasma membrane and SR, so there is limited direct effect on the action potential from mitochondrial Ca²⁺ buffering (75, 76). However, chronic alteration of mitochondrial Ca²⁺ uptake can alter action potential dynamics, contributing to arrhythmias in both the atrium and the ventricle, as described below in Catecholaminergic Polymorphic Ventricular Tachycardia (77).

Ca²⁺ in Mitochondrial Metabolism

Allosteric Ca²⁺ Binding Enhances the Activity of Metabolite Import

Ca²⁺-activated mitochondrial carriers (CaMCs) are mitochondrial inner membrane transporters of metabolites, which are regulated by Ca²⁺ via intermembrane space-facing EF-hands, similar to MICU1 (78) (FIGURE 2). When Ca²⁺ levels rise within the cytoplasm and intermembrane space, Ca²⁺ binding to these carriers enhances metabolite transport rates. The aspartate/glutamate exchangers (AGCs, aralar and citrin) possess longer NH₂-terminal extensions with multiple EF-hand copies. Both AGCs are expressed in the heart, are involved in the aspartate-malate shuttle, and, although active at baseline, can be further enhanced by relatively modest increases in Ca²⁺ (<500 nM) (79). In contrast, the shorter CaMCs are ATP-Mg²⁺/Pi exchangers that need higher concentrations of Ca²⁺ to become active. These CaMCs also have NH₂-terminal extensions containing EF-hands that more closely resemble those in calmodulin (78). In recent studies, Ca²⁺-dependent enhancement of substrate transport via these carriers was a primary mechanism by which cytoplasmic Ca²⁺ could accelerate delivery of substrates for OXPHOS during cardiac mitochondrial metabolism (80).

FIGURE 2.

Overview of established and novel forms of mitochondrial metabolism regulation by Ca²⁺ MCU [mitochondrial Ca²⁺ uniporter (pore-forming subunit)] imports Ca²⁺ into the matrix, whereas export mechanisms include Ca²⁺/H⁺ exchange via LETM1/TMBIM5 and Na⁺/Ca²⁺ exchange via NCLX/TMEM65. Well-established mechanisms of Ca²⁺ regulation are shown with solid red arrows and include allosteric regulation of aspartate-glutamate carriers (AGC), glycerol-3-phosphate dehydrogenase 2 (GPD2), isocitrate dehydrogenase (IDH), α-ketoglutarate dehydrogenase (αKDH), pyruvate dehydrogenase (PDH) phosphatase, and short Ca²⁺-binding mitochondrial carriers (SCaMC). Newer indirect mechanisms are shown as dashed red arrows and involve regulation of Complex I, coenzyme Q (CoQ), Complex IV, and ATP synthase. Cyt c, cytochrome c; IMM, inner mitochondrial membrane; IMS, intermembrane space; MCU, mitochondrial Ca²⁺ uniporter (pore-forming subunit); MPC, mitochondrial pyruvate carrier; OMM, outer mitochondrial membrane.

The glycerol-3-phosphate shuttle can also be subject to Ca²⁺ regulation (81). This pathway provides another route for electrons to enter the ETC. In this pathway, dihydroxyacetone conversion into glycerol-3-phosphate is catalyzed by cytoplasmic glycerol-3-phosphate dehydrogenase (GPD). Glycerol-3-phosphate can then diffuse into the intermembrane space, where it is oxidized by GPD2, transferring electrons to coenzyme Q (CoQ) upstream of Complex III. Modest increases in Ca²⁺ (100–250 nM) in the intermembrane space can bind to the COOH-terminal EF-hands within GPD2 to enhance its activity (81–84). This pathway appears to be minimally active in the healthy heart but may be upregulated during ischemic disease, particularly when Complex I of the ETC is dysfunctional (81, 85–88).

Ca²⁺ Regulation of Pyruvate Dehydrogenase Alters Fuel Substrate Choice

Much of the work on the uniporter over the past decade has shown that it is a critical determinant of mitochondrial fuel substrate preference, which we have reviewed recently and summarize briefly here (89–92). The key nexus for this regulation is the pyruvate dehydrogenase (PDH) complex, a multisubunit enzyme that transforms pyruvate into acetyl CoA. PDH activity is inhibited when PDH kinase phosphorylates it and enhanced when PDH phosphatases remove these modifications. Ca²⁺ stimulates PDH indirectly, by allosterically binding (K_0.5 ∼1 µM) and promoting the activity of PDH phosphatase (93, 94) (FIGURE 2).

In recent studies of cardiac and skeletal muscle-specific knockouts of MCU, flux through PDH was reduced (89–91). Because pyruvate contributes to the tricarboxylic acid (TCA) cycle downstream of glycolysis, deletion of MCU produced a shift of substrate utilization away from glucose toward fatty acids. Moreover, in reciprocal fashion, inhibition of the import of pyruvate into mitochondria was associated with inhibition of mitochondrial Ca²⁺ uptake (95). This mechanism may prevent mitochondrial Ca²⁺ overload when nutrient depletion causes bioenergetic stress.

Allosteric Ca²⁺ Binding Stimulates Multiple TCA Cycle Dehydrogenases

Since the discovery of Ca²⁺ import into the mitochondria, the best-established mechanism of metabolic regulation has been through binding and allosteric enhancement of several key dehydrogenases of the TCA cycle, which have been reviewed extensively (6, 93, 96) (FIGURE 2). Ca²⁺ activates isocitrate dehydrogenase (IDH) (97, 98) [K_0.5 at Ca²⁺ 5–50 µM (99, 100)] and α-ketoglutarate dehydrogenase (AKGDH) (101, 102) [K_M at Ca²⁺ 0.2 µM with ADP or ∼2 µM with ATP (99)]. Surprisingly, despite substantial evidence supporting this traditional model, recent investigations using knockouts of the mitochondrial calcium uniporter, especially in the heart, found relatively preserved respiration and ATP production (103–108). This suggests either that the well-established effects of Ca²⁺ on TCA cycle dehydrogenases may be less important compared with its upstream effects on metabolite transport and fuel substrate choice or that the absence of Ca²⁺-induced TCA cycle enhancement may be compensated by other mechanisms, such as product inhibition. Importantly, although uniporter deficiency minimally affects basal cardiac function, it is clear that the uniporter is required to boost ATP synthesis rapidly during periods of stress (see Ischemia-Reperfusion Injury and Systolic Heart Failure) (105, 107, 109).

Ca²⁺ Alters Oxidative Phosphorylation via Indirect Mechanisms

Ca²⁺ regulation of the electron transport chain (ETC) and F₁-F_O-ATP synthase is an unsettled question. Direct Ca²⁺ regulation of the ETC via allosteric binding has never been established at physiological Ca²⁺ concentrations, though a recent preprint suggests that Ca²⁺ may bind multiple subunits of ETC Complexes I–IV and the ATP synthase (110). Regardless, various studies have shown Ca²⁺-dependent increases in activity of the ETC and ATP synthase in isolated mitochondria or purified enzyme systems (6). The key benefit of Ca²⁺ regulation of OXPHOS is its very rapid onset, occurring within ∼250 ms of Ca²⁺ entry (111). In two recent studies of OXPHOS in cardiac and skeletal mitochondria, Ca²⁺-induced increases in ATP production were explained by its effects on substrate import, TCA cycle flux, and increased ΔΨ rather than direct allosteric effects on the ETC or ATP synthase (33, 80). However, a different study examining skeletal muscle mitochondria revealed more than twofold increases in OXPHOS rates induced by Ca²⁺ at ETC Complexes I, III, and IV, even after controlling for the effects of Ca²⁺ on substrate import, ΔΨ, and NADH production (112). The use of different assays in these studies, including respirometry, cytochrome redox absorbance, and fluorescent reporters, may account for some of the discrepant findings. Furthermore, the use of the creatine kinase clamp to mimic resistance to ATP production under physiological conditions may have affected the results (80, 112), as the effects of Ca²⁺ on respiration are minimized when saturating levels of ADP are used (113). Despite conflicting data on direct Ca²⁺ allosteric regulation of OXPHOS, there is a growing list of recent studies described below that show that mitochondrial Ca²⁺ is a critical indirect regulator of OXPHOS (FIGURE 2).

Complex I (NADH:ubiquinone oxidoreductase).

Complex I is the largest component of the ETC, consisting of an L-shaped structure with multiple transmembrane subunits responsible for H⁺ pumping and a matrix-facing soluble arm that is responsible for transferring electrons from NADH to coenzyme Q (CoQ). Until a very recent preprint (see future directions and conclusions), there were no data to suggest that Ca²⁺ binds directly to Complex I. Nevertheless, our group has recently discovered that Complex I interacts with the uniporter and regulates its stability, via a mechanism we term Complex I-induced protein turnover (CLIPT) (114) (FIGURE 3). Under normal physiology, during NADH oxidation, some electrons leak away from Complex I at a low rate (<1%) (115). Such electrons rapidly react with molecular oxygen to produce reactive oxygen species (ROS). These species are highly reactive and primarily disrupt Complex I locally (116). We identified the NH₂-terminal domain of MCU as an interactor with Complex I, precisely near the electron transfer sites of its soluble arm. Under normal conditions, the electron leak from Complex I oxidizes MCU and leads to its degradation by quality-control proteases (114). Therefore, MCU is constantly turning over and potentially protecting Complex I from oxidation caused by this harmful electron leak. When MCU levels decline, loss of this protective function may produce long-term decreases in Complex I levels, supported by recent studies that show loss of Complex I when the uniporter is disrupted in cardiac tissue (89, 109).

FIGURE 3.

New mechanisms involving mitochondrial Ca²⁺ A: Complex I interacts with MCU [mitochondrial Ca²⁺ uniporter (pore-forming subunit)] and increases its turnover because of reactive oxygen species (ROS) leak (CLIPT), which helps protect Complex I from oxidation. B: matrix acidification during ischemic injury solubilizes Ca²⁺ from Ca²⁺ phosphate precipitates. The increased matrix Ca²⁺ via NCLX boosts matrix Na⁺, thereby decreasing inner mitochondrial membrane (IM) fluidity and slowing coenzyme Q (CoQ) diffusion and the activity of Complexes II + III. C: excess Ca²⁺ may trigger rearrangement of the F1 domain of the ATP synthase to produce a channel in the membrane. IMS, intermembrane space; NTD, NH₂-terminal domain of MCU.

Complex I is the ETC complex most affected by congenital mutations. Our group found that cardiac mitochondrial Ca²⁺ increases when Complex I is impaired, as a potential compensatory mechanism needed for survival. Impairment in Complex I led to a loss of the interaction with MCU. Absent this interaction, MCU became stabilized, with a consequent increase in channel levels (114, 117). This effect was extremely well conserved, evident in Drosophila, mice, and induced pluripotent stem cells (IPSCs) derived from an infant with Complex I deficiency. Moreover, we found that the uniporter activity was essential for survival. In Drosophila with Complex I deficiency, suppressing uniporter expression led to severe developmental mortality, whereas overexpressing the uniporter rescued both survival and function. Such results suggest that altering mitochondrial Ca²⁺ may be a potential therapeutic target in the setting of Complex I dysfunction.

Complex II (succinate dehydrogenase).

Complex II is a component of the TCA cycle, producing FADH₂, which is subsequently oxidized to transfer electrons to coenzyme Q in the ETC. There has been no evidence of direct or indirect Ca²⁺ regulation of this complex under normal physiology, though loss of Complex II may impair the ability to take up mitochondrial Ca²⁺ in cell lines (118).

Complex III (CoQH₂-cytochrome c reductase).

Complex III is a dimeric structure that transfers electrons from reduced CoQ (CoQH₂) to cytochrome c, through a loop known as the Q cycle. In this process, a net two protons are pumped into the intermembrane space. As with Complex II, physiological levels of Ca²⁺ do not appear to directly affect Complex III activity, although there is at least one publication that suggests that Ca²⁺ at very low submicromolar concentrations may alter flux through an unclear mechanism (119).

Despite the absence of direct effects of Ca²⁺ on Complex III during normal physiology, a recent publication details a novel mechanism by which Ca²⁺ affects the flux of CoQ from Complex II to Complex III (120) (FIGURE 3). Ischemia-reperfusion injury leads to increases in cytoplasmic Na⁺ levels, due to limited activity of the Na⁺-K⁺-ATPase, and acidification of the mitochondrial matrix. This acidification releases Ca²⁺ from calcium phosphate granules located in the matrix. Then, increases in Ca²⁺ in the matrix as well as Na⁺ in the cytoplasm cause overactivation of NCLX, contributing to Na⁺ overload within the mitochondrial matrix. These increases in Na⁺ within the mitochondria are thought to decrease the fluidity of the inner mitochondrial membrane and slow the diffusion of CoQH₂ from Complex II to Complex III, decreasing flux through the ETC. Moreover, the Complex III semiquinone is prevented from completing the Q cycle, and its excess electrons can leak out as ROS, promoting oxidative damage. Here, rather than compensating for deficits in the ETC, as during Complex I impairment, increases in mitochondrial Ca²⁺ may contribute to oxidative damage.

Complex IV (cytochrome-c oxidase).

Complex IV transfers electrons from cytochrome c to oxygen. This redox reaction is coupled to the translocation of four protons into the intermembrane space, contributing to setting ΔΨ. Notably, the intermembrane space side of complex IV has a binding site for Na⁺ and Ca²⁺ (121). Under resting Ca²⁺ levels, the site is normally bound to Na⁺. However, during Ca²⁺ signals (∼1 µM, though prior reports suggested >20 µM), the Na⁺ is displaced, inhibiting activity of the complex with an IC₅₀ of 0.5 µM. It appears that this inhibition may only be evident when fuel substrates are limited. In addition, Complex IV may also be indirectly regulated by the uniporter. MCU regulator 1 (MCUR1) appears to be an assembly factor for Complex IV as well as a scaffold factor for the uniporter (122–125). MCUR1 inhibition slows mitochondrial Ca²⁺ uptake and may destabilize Complex IV. Moreover, during periods of mitochondrial Ca²⁺ overload, MCUR1 appears to be a critical sensor of the Ca²⁺ threshold triggering the permeability transition (126), a process described in more detail below. The mechanistic basis for these effects remains unexplored.

Supercomplexes.

Complex I binds to two subunits of Complex III and a single subunit of Complex IV to assemble a supercomplex known as the respirasome (127, 128). Although these respirasomes are seen repeatedly in biochemical assays, whether they exert functional effects on energetic flux or reflect optimal packing density within mitochondrial cristae is an ongoing topic of research (129–131). In fact, recent studies suggest that disrupting supercomplex formation has limited effects on cardiac function (132). Nevertheless, multiple interventions that alter mitochondrial Ca²⁺ uptake also affect supercomplex formation, though the association between Ca²⁺ uptake and supercomplex formation is not consistent across studies (133–135). For example, in one study sprint interval training exercise led to increases in mitochondrial Ca²⁺ and supercomplex formation (136). In contrast, in another report decreased levels of mitochondrial Ca²⁺ in female cardiomyocytes, relative to males, were associated with increased supercomplex formation through the COX7RP interfacial subunit, though respiration was unaffected (137). This effect was associated with lower ROS levels and thought to partly explain the lower propensity for arrhythmias in premenopausal females.

F₁-F_O-ATP synthase.

The F₁-F_O-ATP synthase contains a circular membrane-embedded F_O module and a rotary, soluble F₁ module, connected by central and peripheral stalks (138, 139). This allows the catalytic F₁ portion to synthesize ATP from ADP and phosphate, driven by H⁺ flux through an oligomeric ring in F_O. No evidence exists for direct allosteric regulation by Ca²⁺ during normal physiology (33, 140). Intriguingly, in trypanosomal parasites the uniporter may directly interact with the ATP synthase c subunit, the highly conserved protein that forms the oligomeric ring of F_O (141).

In pathophysiological states, however, it is possible that Ca²⁺ exerts a profound effect on ATP synthase structure and function. It is well established that excessive Ca²⁺ uptake opens a cyclosporine-sensitive, large-conductance channel known as the mitochondrial permeability transition (MPT) pore (142–145). When this channel opens for prolonged periods it leads to mitochondrial depolarization, swelling, and membrane disruption and can ultimately lead to cell death. One current model for the MPT suggests that it occurs via abnormal changes in ATP synthase conformation (142) (FIGURE 3). Mg²⁺ typically binds at the interface between the α- and β-subunits of the soluble F₁ module, helping to coordinate ATP or ADP within the nucleotide-binding pocket. In the ATP synthase model of the MPT, at high concentrations Ca²⁺ displaces Mg²⁺ at this site, leading to the transmission of abnormal strain to the transmembrane F_O module and opening of a large-pore channel within the ring formed by c subunits. This model has received some support in a recent preprint showing Ca²⁺ binding to the F₁ module β-subunit (110).

Outstanding question: Because opening of the MPT appears to be one of the earliest irreversible events in the pathophysiology of injury during heart attacks, strokes, and other forms of ischemic injury, deciphering its mechanism remains one of the most significant unanswered questions in mitochondrial Ca²⁺ signaling. How Ca²⁺ produces MPT through the ADP-ATP translocase, the other major complex involved, has not been resolved (146, 147).

Mitochondrial Ca²⁺ Transport Regulation of OXPHOS, Oxidative Stress, and Mitochondrial Dysfunction in Heart Disease

Ischemia-Reperfusion Injury and Systolic Heart Failure

Acute cardiac injury occurs most frequently during acute coronary syndromes, leading to ischemia-reperfusion injury. Both this type of insult as well as damage caused by other processes such as hypertension (excessive neurohormonal signals), drugs, or genetic disorders can lead to chronic systolic heart failure, characterized by impaired cardiac contractile function. In rodent models, these pathologies have been investigated by either injuring the heart through nongenetic manipulations or studying transgenic animals with disease-causing mutations. In studies of nongenetic heart disease, three rodent models have been used repeatedly: 1) acute ischemia-reperfusion, 2) chronic pressure-overload-induced heart failure, and 3) chronic β-adrenergic stress-induced heart failure. The phenotypes associated with uniporter deletion in these models of heart disease have been reviewed recently (148–151). Here, we focus on four interrelated mechanistic concepts that can be generalized from investigating mitochondrial Ca²⁺ uptake in these heart injury disease models (FIGURE 4).

FIGURE 4.

Mitochondrial Ca²⁺ pathways involved in health and disease A: by regulating multiple enzymes involved in metabolite transport, the tricarboxylic acid (TCA) cycle, and the electron transport chain (ETC), Ca²⁺ allows rapid matching of ATP supply to demand. B: Ca²⁺ overload produces permeability transition, in which the ATP synthase, ATP/ADP translocase (ANT), and mitochondrial Ca²⁺ uniporter (MCU) regulator 1 (MCUR1) are implicated. C: by boosting flux through the TCA cycle, Ca²⁺ can produce both prooxidant effects [via NADH oxidation and reactive oxygen species (ROS) leak through Complex I] and antioxidant effects [enhancing the supply of NADPH by indirectly increasing substrates to malic enzyme 3 (ME3), IDH2 (isocitrate dehydrogenase 2), and nicotinamide nucleotide transhydrogenase (NNT)]. The balance of these effects determines the ultimate consequence of Ca²⁺ on oxidative stress. D: metabolism can be altered differently depending on the chronicity of uniporter inhibition. Acute inhibition abrogates supply-demand matching and Ca²⁺ overload during ischemia-reperfusion injury. However, chronic inhibition can potentially activate other cell death pathways, cause compensation of Ca²⁺ influx by reversing the direction of Ca²⁺/H⁺ and Na⁺/Ca²⁺ exchangers, or cause the loss of protective uniporter functions such as its interaction with Complex I. All of these chronic mechanisms can offset the benefit on Ca²⁺ overload of uniporter inhibition and may underlie the benefits of boosting uniporter activity in chronic cardiac disease. GSSH, reduced glutathione.

Concept 1: Ca²⁺-dependent enhancement of OXPHOS in heart injury.

First, mitochondrial Ca²⁺ uptake during cardiac function appears to be important for matching ATP supply and demand, via the effects of Ca²⁺ on OXPHOS and fuel choice described in ca²⁺ in mitochondrial metabolism. Such contributions to energetic homeostasis are most evident when cardiac workload increases, since mice lacking uniporter components are unable to rapidly increase heart rate or cardiac contractility during pharmacological or physical stress or exercise (105, 107, 108). In addition, overexpression of the inhibitory uniporter subunit MCUB produces a transient cardiomyopathy, suggesting that acute loss of Ca²⁺ uptake prevents supply-demand matching (13).

As heart failure progresses, energy supply fails to meet demand, and individuals with more severe supply-demand mismatch have reduced survival (152). Therefore, overexpression of MCU has been suggested as a potential therapy for the failing heart. Of note, in several of the studies listed below cardiac energetics was only partly investigated. However, global changes in ATP are a late manifestation of energetic dysfunction, whereas one of the earliest markers of cardiac energetic failure is diminished activity of SERCA, the protein responsible for reloading the SR with Ca²⁺ (153). Therefore, finding decreases in SR Ca²⁺ leak and rescue of SR Ca²⁺ uptake are useful surrogates for improved cardiac function during heart failure. In a guinea pig pressure-overload model of heart failure, acute viral transduction of MCU increased uniporter levels by ∼50% and improved the contractile response. Although the consequence to energetics was not directly measured, it is notable that oxidative stress was decreased and Ca²⁺ leak from the SR was diminished (154). Furthermore, in experiments where heart failure was induced in mice by chronic isoproterenol infusion, which leads to excessive β-adrenergic stimulation, overexpressing MCU fourfold with an inducible transgene approach also ameliorated the decline in cardiac function (155). Critically, the increase in Ca²⁺ uptake was associated with improved Ca²⁺ uptake into the SR, suggesting that MCU overexpression improved cellular energy homeostasis in this heart failure model. Taken together, these studies suggest that modest increases in mitochondrial Ca²⁺ uptake can help boost cardiac energy synthesis and preserve function during heart failure. What remains to be established in these scenarios is to what degree the changes in metabolism occur at the level of ATP synthesis/OXPHOS, altered fuel choice via changes to pyruvate dehydrogenase activity, or the alteration of other matrix targets of Ca²⁺.

Concepts 2 and 3: Ca²⁺-dependent injury via the regulation of reactive oxygen species and the mitochondrial permeability transition.

In contrast to the potential beneficial effects of Ca²⁺ on energetics in heart failure, in acute cardiac injury the greater concern is mitochondrial Ca²⁺ overload, via two toxic downstream mechanisms involving the OXPHOS machinery. These are the excessive production of ROS and the potential activation of the MPT. Ca²⁺-induced activation of the MPT has been discussed above (in F₁-F_O-ATP synthase). Ca²⁺-induced boosting of ROS can occur via its effects on NADH production. During ischemic injury, hypoxia leads to stalling of forward flow of electrons down the ETC, and these can leak out as ROS, particularly during reperfusion, when mitochondrial dysfunction leads to reverse electron flow through Complex I (156). Concurrently, cytoplasmic Ca²⁺ overload occurs because of the failure of Ca²⁺ recycling into the SR and increased permeability of the plasma membrane (153, 157, 158) and subsequently enters the matrix via the uniporter to produce mitochondrial Ca²⁺ overload (159). Thus the combined effect of increased Ca²⁺-mediated NADH production and dysfunctional ETC activity is to increase pathological ROS production. Because of the combined effects on pathological ROS and MPT, mice with whole body or cardiac-specific knockout of MCU, pharmacological inhibition of MCU, or cardiac deletion of EMRE all had reduced cardiomyocyte necrosis during experimental ischemia-reperfusion injury (105, 107, 109, 159, 160). The benefit of preventing Ca²⁺ overload could also be replicated by increasing NCLX expression approximately threefold, which increased mitochondrial Ca²⁺ efflux and improved survival (16). Inhibition of MCU and EMRE and enhancement of NCLX have all been associated with reduced susceptibility to the MPT in these cardiac models, and less ROS generation has been noted in MCU KO mouse embryonic fibroblasts subject to ischemia-reperfusion injury (16, 105, 107). Moreover, ischemic cardiac injury leads to endogenous downregulation of mitochondrial Ca²⁺ uptake, as multiple studies have shown that the inhibitory uniporter subunit MCUB, usually poorly expressed in the heart, is substantially induced in the postischemic period (13, 14). Consistent with this effect, transgenic overexpression of MCUB reduces the degree of injury following ischemia-reperfusion, whereas deletion of MCUB increases both injury and susceptibility to MPT (13, 14).

Nevertheless, the effects of Ca²⁺ on ROS are not entirely straightforward, as Ca²⁺ can also indirectly contribute to the antioxidant response (8). For example, increasing matrix Ca²⁺ via inhibition of NCLX appears to prevent the production of pathological ROS in failing cardiomyocytes (161). The mitochondrial glutathione and thioredoxin-dependent enzymes involved in the antioxidant response rely on the NADPH/NADP⁺ redox potential (149). Mitochondrial Ca²⁺ can indirectly contribute to boosting this potential by its effects on the TCA cycle. The TCA cycle metabolites malate and isocitrate are substrates for NADP⁺-dependent malic enzyme (ME3) and isocitrate dehydrogenase 2 (IDH2) (149, 162), both of which have robust cardiac expression. Though these enzymes are not known to bind Ca²⁺ directly, increases in their substrates via Ca²⁺-induced boosting of TCA cycle flux can lead to increased NADPH and subsequent antioxidant pathways. Similarly, enhanced Ca²⁺-induced NADH production can be interconverted to NADPH via the nicotinamide nucleotide transhydrogenase (NNT), a ΔΨ-dependent transmembrane protein. In pressure-overload heart failure models, it appears that reversed NNT activity depletes NADPH to maintain the NADH pool (163), and thus Ca²⁺ boosting of NADH production may also contribute to maintaining NADPH levels via this mechanism. The loss of this protective mechanism may offset the benefit of preventing Ca²⁺ overload in pressure-overload models and explain why deletion of the uniporter has no effect on survival in this form of myocardial injury (103, 105). In contrast, as discussed above, modest MCU overexpression may in fact prove beneficial in pressure-overload models, implying that the benefit conferred to energetics and redox homeostasis may outweigh concerns about Ca²⁺ overload (154).

Outstanding questions: Since mitochondrial Ca²⁺ can indirectly promote both an antioxidant response via the contribution to NADPH as well as a prooxidant response via NADH oxidation at Complex I, what factors contribute to the balance of these effects in health and disease? Furthermore, there has been only limited investigation into whether Ca²⁺ can directly regulate the enzymes involved in the mitochondrial antioxidant response (164).

Concept 4: Chronic uniporter inhibition may affect metabolism beyond mitochondrial Ca²⁺ uptake.

The final mechanistic concept arising from studies investigating therapeutic depletion of the uniporter is perhaps the most intriguing: the effects vary greatly depending on the timing of the uniporter inhibition relative to the onset of injury. In the studies listed above, acute deletion of MCU or EMRE in the weeks before ischemia-reperfusion injury led to improved outcomes in uniporter-inhibited animals. In contrast, germline and whole body deletion of uniporter components have consistently shown minimal effect on outcomes, despite loss of mitochondrial Ca²⁺ uptake, in the same ischemia-reperfusion models (103, 106, 108, 109, 148). In a recent study, these differential effects were shown to be dependent on the chronicity of uniporter loss relative to the injury. The authors induced cardiac-specific loss of the EMRE subunit with tamoxifen and studied the effects of heart injury either 3 wk or 3 mo after the deletion. Whereas loss of EMRE was protective when injury followed shortly thereafter, the effects were absent when chronic loss of EMRE preceded the injury.

Outstanding question: The mechanisms for the adaptation to uniporter loss remain to be established. It appears that either 1) other mechanisms compensate for the loss of uniporter-mediated mitochondrial Ca²⁺ uptake or 2) the benefits of preventing mitochondrial Ca²⁺ overload are offset in the long term by detrimental effects of absent Ca²⁺ influx on energetics. Evidence for both these possibilities exists, as alterations in the activity of NCLX are seen in in vitro models of ischemia-reperfusion injury (165), suggesting it may compensate as an alternative Ca²⁺ flux pathway (though see Ref. 104), whereas chronic loss of the uniporter leads to reductions in Complex I levels and an increase in mitochondrial oxidative modifications, suggesting potential harm from chronic uniporter loss (89, 109, 165).

Arrhythmias Induced by Heart Injury

Beyond assessment of overall cardiac function, investigations have also focused on the susceptibility to arrhythmias, a separate mechanism for cardiac death. In several heart failure models, regulation of NCLX function appears critical for arrhythmia generation (FIGURE 5A). Both acute and chronic pharmacological inhibition of NCLX by transfusion of the inhibitor CGP-37157 reduced ventricular tachyarrhythmias in failing hearts. In one study, acute inhibition of NCLX in a pig coronary ischemia-reperfusion model resulted in suppression of ventricular tachyarrhythmias, though no effect was seen on cardiac infarct size (166). Similarly, in a study of guinea pigs subjected to combined aortic constriction and chronic β-adrenergic stress, control animals developed heart failure and fatal tachyarrhythmias, whereas chronic infusion of CGP-37157 led to a suppression of both functional decline as well as arrhythmias (161). The mechanism was presumed to be the prevention of excess Ca²⁺ depletion, leading to beneficial effects on ATP production via classical allosteric mechanisms, though these were not tested directly. Notably, in the guinea pig study, prolonged cardiac stress leads to cytoplasmic Na⁺ overload, which is then imported into mitochondria via NCLX in exchange for Ca²⁺, leading to Ca²⁺ depletion. In this model, cardiac deficits result from insufficient Ca²⁺ to support NADH production through the TCA cycle. The increased NAD⁺-to-NADH ratio impairs electron flow both for ATP production (via the electron transport chain) as well as for interconversion to NADPH, which is critical for the redox response, as discussed above (Concepts 2 and 3: Ca²⁺-dependent injury via the regulation of reactive oxygen species and the mitochondrial permeability transition). Thus, inhibiting NCLX preserves Ca²⁺ for both energetics and combating oxidative stress. However, the therapeutic window for this intervention may well be limited, as loss of NCLX is lethal (16).

FIGURE 5.

Mitochondrial Ca²⁺ alterations involved in cardiac disease A and B: contributions to arrhythmogenesis after cardiac injury. A: depletion of mitochondrial Ca²⁺ by excessive Na⁺/Ca²⁺ exchange can lead to energetic impairment via insufficient NADH production and promote arrhythmias. B: Ca²⁺ overload in states of overnutrition promotes the production of mitochondrial reactive oxygen species (mROS), promoting oxidative injury to ryanodine receptor (RyR)2, diastolic Ca²⁺ leak, action potential (AP) lengthening, and consequent arrhythmias. C: the balance between the effects described in A and B may also determine whether mitochondrial Ca²⁺ promotes or inhibits arrhythmias in CPVT. D: in mitochondrial cardiomyopathies, injury to Complex I prevents interaction and oxidation of MCU (mitochondrial Ca²⁺ uniporter), leading to buildup in uniporter levels and subsequent energetic compensation. E: cardiolipin (blue) can bind within a fenestration between the MCU transmembrane domains. Loss of cardiolipin in Barth syndrome may alter MCU structure to inhibit interaction with accessory subunits and prevent Ca²⁺ influx. IM, inner mitochondrial membrane; IMS, intermembrane space.

Arrhythmias induced by diabetes or overnutrition represent another area where mitochondrial Ca²⁺ may be contributing by exerting prooxidant effects (FIGURE 5B). In a mouse model of diabetes induced by high-fat diet, excess mitochondrial ROS was noted and led to oxidation of RyR2 receptors and diastolic Ca²⁺ leak, which can prolong the cardiac action potential and trigger arrhythmias (167). In a prediabetic model induced by chronic fructose feeding, pharmacological uniporter enhancement exacerbated arrhythmias, whereas MCU ablation reduced them (168). In another study, control and MCU KO cardiomyocytes were exposed to palmitic acid as a model for metabolic stress. Whereas the control cardiomyocytes had increases in mitochondrial Ca²⁺, ROS, and depolarization of mitochondrial membranes, these changes were abrogated in MCU KO cells. In animals exposed to a high-fat diet for 4 wk, cardiomyocytes also had prolonged action potentials and easily induced ventricular tachyarrhythmias, whereas cardiac-specific deletion of MCU abrogated these changes (169). However, not all investigations have been consistent with mitochondrial Ca²⁺ promoting these pathologies. In particular, in a study of diabetes induced by overexpression of human amylin in rat pancreas, mitochondrial Ca²⁺ levels were low and pharmacological inhibition of NCLX prevented arrhythmias (170). Nevertheless, the bulk of studies suggest that increases in mitochondrial Ca²⁺ may have proarrhythmic effects in diabetes/overnutrition.

The effects of mitochondrial Ca²⁺ on cardiac arrhythmias in heart failure versus diabetes/obesity models are strikingly discordant. In the heart failure models, mitochondrial Ca²⁺ appears to decreasing ROS by its effects on NAD(P)H, whereas in the overfeeding or diabetic models, mitochondrial Ca²⁺ appears to boost ROS through Complex I.

Outstanding question: As discussed above, this raises the question of what mechanisms alter the balance between the pro- and antioxidant effects of mitochondrial Ca²⁺ and whether Ca²⁺ contributes to these beyond the established mechanisms.

Catecholaminergic Polymorphic Ventricular Tachycardia

Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a rare inherited arrhythmia syndrome that can cause sudden cardiac death in young people. It is characterized by exercise- or emotional stress-induced bidirectional or polymorphic ventricular tachycardia in the setting of a structurally normal heart and a normal ECG. CPVT is caused by mutations affecting proteins that regulate cardiomyocyte SR Ca²⁺ levels, most commonly RyR2. The resulting intracellular Ca²⁺ overload leads to delayed afterdepolarizations and triggered activity, which can induce ventricular tachycardia and fibrillation (171, 172).

Given the centrality of Ca²⁺ transfer between RyRs and mitochondria for energy supply-demand matching, several studies have investigated whether aberrant Ca²⁺ release from CPVT RyR2 mutants affects mitochondrial function (FIGURE 5C). In one study (173), RyR Ca²⁺ leak induced by caffeine led to increased mitochondrial ROS production, setting up a feedback loop where mitochondria-derived ROS promoted further Ca²⁺ leak via RyR oxidation. Attenuating the mitochondrial production of ROS and inhibiting mitochondrial Ca²⁺ uptake by viral expression of dominant-negative MCU both decreased arrhythmogenesis parameters in isolated cardiomyocytes, whereas cardiomyocytes from CPVT mice had elevated expression of the inhibitory MCUB subunit. How such interventions may affect mitochondrial Ca²⁺-dependent energetics was not explored but remains an intriguing question, because triggered activity does not always lead to clinical arrhythmias. That such a disconnect may exist between ROS promotion of triggered activity versus altered energetics promotion of an arrhythmogenic substrate comes from contrasting data on drugs that promote, rather than inhibit, mitochondrial Ca²⁺ uptake. In two studies in zebrafish models of CPVT, enhancement of SR-to-mitochondrial Ca²⁺ transfer, via either the VDAC2 agonist efsevin or the MCU agonists ezetimibe and disulfiram, restored normal cardiac rhythms by more than threefold (174, 175). Strategies that promote SR-to-mitochondria Ca²⁺ transfer were also successful in preventing arrhythmias in live RyR2^R4496C/WT mice. When efsevin, the VDAC2 agonist, or kaempferol, a weak MCU agonist, was infused via osmotic minipumps in this model, there was a ∼40% reduction in the number of mice with ventricular tachyarrhythmias, an effect also replicated in cardiomyocytes derived from induced pluripotent stem cells (176). Surprisingly, boosting MCU Ca²⁺ uptake with kaempferol actually reduced ROS in the calsequestrin KO model of CPVT, an effect in part attributed to global increases in the ability of mitochondria to buffer Ca²⁺ (168). Conversely, inhibiting MCU with a tamoxifen-inducible knockout drastically worsened arrhythmias, leading to rapid lethality (168). The evidence for Ca²⁺ overload as a mediator of injury in CPVT is also somewhat controversial, as calsequestrin KOs that were crossed with mice with inhibited MPT demonstrated increased ROS production and an increased propensity to arrhythmias (177).

Outstanding question: How to reconcile these studies with surprisingly opposite outcomes? One possibility is the use of pharmacological versus genetic manipulations, though it should be noted that even genetic manipulations led to apparently opposite outcomes [decreased arrhythmogenesis with a dominant-negative MCU in rat ventricular myocytes vs increased lethality with MCU KO (168, 173)]. An intriguing possibility is the disconnect between in vitro phenotypes, which have variably shown both worsened and improved arrhythmogenesis and susceptibility to Ca²⁺ overload, versus whole animal outcomes, which have consistently shown improvements in outcomes with Ca²⁺ uniporter enhancement. In vitro assays typically measure parameters in isolated cardiomyocytes without energetic or contractile resistance, as occurs in vivo. Therefore, it may be that the Ca²⁺ contribution to cardiac energetic homeostasis, poorly evaluated in vitro, may be a key determinant to long-term benefit during dysregulated SR Ca²⁺ release in CPVT.

Mitochondrial Cardiomyopathies

Energetic failure can be a primary cause of cardiomyopathy in mitochondrial diseases. Such diseases involve deficient oxidative phosphorylation (OXPHOS) and arise from mutations in mitochondrial proteins encoded by either the nuclear or the mitochondrial genome (mtDNA) (178, 179). These diseases typically occur in infants and children, with a prevalence of 1 in 5,000 (180). Because cardiac contraction requires high energy supply, cardiac dysfunction occurs in 20–40% of children with mitochondrial diseases and increases their mortality nearly threefold (181). In these disorders, the reduction in OXPHOS produces heart failure due to the energy supply-demand mismatch.

Investigators have previously studied OXPHOS deficiency caused by mutations (118, 182–186), deletion of Tfam (the transcription factor for mtDNA) (187), or chemical mtDNA depletion (188). The typical finding is reduced or unchanged Ca²⁺ uptake in the presence of diminished ΔΨ, though increased Ca²⁺ was seen in skeletal muscle (187). Diminished ΔΨ correlated with disease severity. This presents a technical problem in assessing mitochondrial Ca²⁺ transport pathways. Ca²⁺ influx through the uniporter is driven by ΔΨ, so a change to either ΔΨ or uniporter activity can alter the magnitude of Ca²⁺ influx. Imaging assays are the most commonly used method to measure mitochondrial Ca²⁺ influx, but they cannot distinguish whether changes in flux are due to altered ΔΨ or uniporter activity. Analysis becomes especially difficult when these two factors change in opposite directions. In OXPHOS-deficient mitochondria, precisely such diminished ΔΨ may mask any compensatory increases in uniporter activity in typical in vitro imaging assays. We used complementary imaging and electrophysiological assays to measure uniporter activity directly in mice with cardiac-specific deletion of Tfam (Tfam cKO) (117), a rodent model that best mimics pediatric mitochondrial cardiomyopathies (189, 190). These animals had substantial reductions in cardiac mtDNA-encoded OXPHOS subunits and ETC activity. In these mice, we noted a doubling in the rate of Ca²⁺ uptake and a three- to fivefold increase in the size of mitochondrial Ca²⁺ currents, associated with a doubling of protein, but not transcript, levels of multiple uniporter subunits. We found that these changes in Ca²⁺ influx boosted mitochondrial respiration and ATP synthesis. In the absence of Ca²⁺, oxygen consumption rates in Tfam cKO mitochondria were half of wild-type values, but incubation in Ca²⁺ restored these rates to near control levels. The mechanism for such energetic compensation was CLIPT, the regulation of protein lifetime by Complex I described above in Complex I (NADH:ubiquinone oxidoreductase) (FIGURE 5D). These studies reinforce the concept that mitochondrial Ca²⁺ can be a critical compensatory contributor to energetic homeostasis during ETC injury.

Outstanding question: Given the benefits afforded by mitochondrial Ca²⁺ in these diseases, are there therapeutic approaches that can maximize the stability of uniporter channels without overactivating oxidant responses or triggering MPT? In general, such an approach may be applicable not only to these diseases but to a range of the other cardiac disorders described here.

Barth Syndrome

Barth syndrome is a rare X-linked genetic disorder characterized by cardiomyopathy, skeletal myopathy, growth retardation, neutropenia, and elevated urinary excretion of 3-methylglutaconic acid (3-MGA) (191). The disease results from mutations in the tafazzin gene (TAZ). Tafazzin plays a crucial role in the remodeling of cardiolipin, a unique phospholipid predominantly found in the inner mitochondrial membrane. Mutations in tafazzin lead to reductions in cardiolipin levels compared with precursors. Because cardiolipin binds to multiple mitochondrial membrane proteins, including components of the ETC, this imbalance disrupts mitochondrial bioenergetics and membrane integrity (191, 192). In a mouse model of Barth syndrome generated by RNA interference against Taz, a systolic cardiomyopathy develops over 10–20 wk, leading to reductions in OXPHOS (193). In hearts from these mice, as well as human cardiac tissue and inducible pluripotent stem cell-derived cardiomyocytes (CM-IPSCs) obtained from patients with Barth syndrome, there was >50% reduction in MCU expression, along with a modest ∼10% increase in MCUB levels (193). In separate studies, loss of cardiolipin also affected the stability of MICU1 posttranscriptionally in several model systems (194, 195). The mechanism for loss of MCU appears to be direct, with structures of the uniporter revealing cardiolipin molecules buried in the fenestration between MCU transmembrane domains (49) (FIGURE 5E). Notably, the lack of mitochondrial Ca²⁺ uptake produced pronounced effects during increases in cardiac workload, leading to diminished respiratory capacity, excess oxidation of NAD(P)H, and insufficient ATP generation (193). In contrast to other examples of dysregulated cardiac mitochondrial Ca²⁺ homeostasis, there were no obvious differences in ROS measurements, likely because of an upregulation of endogenous ROS metabolizers. [It is worth noting that in a separate study using cardiac-specific Taz knockout an increase in mitochondrial ROS, mediated by excessive RyR leak and elevated diastolic Ca²⁺, was found (196).] To restore mitochondrial Ca²⁺ levels, investigators used in vivo pharmacological inhibition of Na⁺/Ca²⁺ exchange. Although this reversed arrhythmic parameters, the inotropic response remained blunted. This modest recovery engendered by mitochondrial Ca²⁺ boosting is expected, since cardiolipin binding regulates not just MCU but also multiple components of the ETC (133, 134). However, given the known differential localization of the uniporter relative to NCLX, this result also may further support the tenet that mitochondrial Ca²⁺ regulation of OXPHOS may be a highly local phenomenon, not adequately reflected in global assessments of mitochondrial Ca²⁺ concentrations.

Friedreich’s Ataxia

Friedreich’s ataxia is a peripheral and central neurodegenerative disorder that frequently affects the heart, producing a hypertrophic cardiomyopathy. In this disorder, triplet expansion repeats within the frataxin (FXN) gene cause mitochondrial dysfunction, primarily affecting iron-sulfur cluster biogenesis and heme production. Deficiencies in these central processes disrupt cellular iron homeostasis, impair ATP synthesis, and cause substantial oxidative stress (197, 198). Initial studies in cultured SH-SY5Y (neuroblastoma-like) cells revealed that mitochondrial depolarization and the primary energetic deficiency prevented efficient mitochondrial Ca²⁺ uptake, though whether this contributed to energetic deficiency was not explored (199). Such results were replicated in a separate study, directly revealing diminished Ca²⁺ uptake into isolated FTX-deficient SH-SY5Y cells (200). Here, the decreased accumulation of Ca²⁺ was attributed to dysregulation in ER-mitochondrial contacts, perhaps directly by a variant form of frataxin. Artificially restoring these contacts rescued mitochondrial Ca²⁺ uptake. In a further study, neonatal rat cardiomyocytes acutely depleted of frataxin developed increased sensitivity to Ca²⁺ overload, with swollen mitochondria (201). As in SH-SY5Y cells (200), MCU levels were unchanged. However, in the cardiomyocytes, NCLX levels were decreased, implying that mitochondrial Ca²⁺ overload may lead to toxicity via the MPT. In fact, global chelation of Ca²⁺ or treatment with cyclosporine A, a MPT inhibitor, reduced mitochondrial swelling in these cardiomyocytes (201).

Outstanding questions: What remains unknown is whether mitochondrial Ca²⁺ levels are increased in FXN-deficient cardiomyocytes and whether these produce compensatory boosts to energetics during disease. It is notable that treatment with Ca²⁺ chelation had a less robust phenotype compared to inhibition of the MPT, which may suggest that modest increases in Ca²⁺ may produce salutary effects on mitochondrial energetics before the onset of overload and MPT. In fact, when Ca²⁺ uptake in neurons was enhanced in fruit flies deficient in FTX via overexpression of MCU, locomotor function and survival increased substantially (200). Another interesting point is that interventions to protect mitochondria, such as maintaining organisms under chronic hypoxia, reverse oxidative stress signaling and neurological deficits but fail to affect cardiomyopathy (202), suggesting that the energetic deficits are a key factor for this organ and compensatory Ca²⁺ signaling may offer an independent means of promoting survival.

Future Directions and Conclusions

Because most of the studies conducted on Ca²⁺ regulation of OXPHOS have been conducted in the heart and skeletal muscle, investigators need to establish whether these mechanisms are more generally conserved across organs or if novel tissue-specific Ca²⁺-dependent pathways exist. Moreover, although knockout studies have provided valuable insights, the lack of highly specific compounds for acute manipulation of mitochondrial Ca²⁺ levels hinders investigations during physiological activities. If these can be combined with novel sensors for Ca²⁺, ATP, NAD⁺, and other metabolites, multimodal assessments will reveal how physiological outcomes correlate with changes in critical mitochondrial parameters.

Additionally, emerging technologies are poised to deepen our understanding of mitochondrial Ca²⁺ regulation. The use of isotope-labeled tracers will hopefully offer more precise assessments of how Ca²⁺ influences metabolic fluxes and enzymatic activity, beyond global measurement of respiration. In addition, a recent preprint appears poised to greatly expand the repertoire of targets of mitochondrial Ca²⁺ regulation (110). In this study, assessment of proteomic thermostability in the presence or absence of Ca²⁺ identified >200 potentially Ca²⁺-sensitive mitochondrial proteins, including 11 Complex I, 4 Complex III, 1 Complex IV, and 8 F₁-F_O-ATP synthase subunits, along with 2 Complex II assembly factors. Moreover, Ca²⁺ binding to 2,4-Dienoyl-CoA Reductase 1, a rate-limiting enzyme in mitochondrial polyunsaturated fatty acid oxidation, suggests that Ca²⁺ effects on fuel substrate choice may extend to lipids as well. Such discoveries are overturning the dogma of limited effects of mitochondrial Ca²⁺ on metabolism.

In summary, mitochondrial Ca²⁺ is essential for the cardiac response to physiological and pathological stress. Although it enhances ATP production and shifts the balance in mitochondrial fuel preference, its indirect effects on OXPHOS are also emerging. Recent discoveries of binding partners between ETC complexes and Ca²⁺ regulators provide insights into cellular management of mitochondrial Ca²⁺. As with cytoplasmic Ca²⁺ regulation, a central principle of mitochondrial Ca²⁺ signaling is that these effects are highly localized. Ultimately, understanding how Ca²⁺ governs OXPHOS may have therapeutic implications, since mitochondrial damage is a frequent pathophysiological event across organs and diseases.