Mitochondrial ion channels in cardiac function

Abstract

Mitochondria have been recognized as key organelles in cardiac physiology and are potential targets for clinical interventions to improve cardiac function. Mitochondrial dysfunction has been accepted as a major contributor to the development of heart failure. The main function of mitochondria is to meet the high energy demands of the heart by oxidative metabolism. Ionic homeostasis in mitochondria directly regulates oxidative metabolism, and any disruption in ionic homeostasis causes mitochondrial dysfunction and eventually contractile failure. The mitochondrial ionic homeostasis is closely coupled with inner mitochondrial membrane potential. To regulate and maintain ionic homeostasis, mitochondrial membranes are equipped with ion transporting proteins. Ion transport mechanisms involving several different ion channels and transporters are highly efficient and dynamic, thus helping to maintain the ionic homeostasis of ions as well as their salts present in the mitochondrial matrix. In recent years, several novel proteins have been identified on the mitochondrial membranes and these proteins are actively being pursued in research for roles in the organ as well as organelle physiology. In this article, the role of mitochondrial ion channels in cardiac function is reviewed. In recent times, the major focus of the mitochondrial ion channel field is to establish molecular identities as well as assigning specific functions to them. Given the diversity of mitochondrial ion channels and their unique roles in cardiac function, they present novel and viable therapeutic targets for cardiac diseases.

INTRODUCTION

Heart diseases are among the top causes of morbidity and mortality (1), but the underlying mechanisms that lead to heart failure are not fully understood. There is an urgency to understand and decipher the precise molecular mechanism(s) involved in cardiac diseases. Ion channels and transporters located in different cardiac cells are often targeted in therapeutic interventions for over 60 years (2–7). Amiodarone and lidocaine were the first groups of compounds targeted for sarcolemmal ion fluxes to treat arrhythmias (8). An intricate analysis of the molecular basis of cardiac arrhythmias has been the driving force behind establishing the molecular identity of the ion channels that generate the cardiac action potential (9). As of today, the genes encoding all the major cardiac membrane ion channels have been sequenced and identified (9, 10). As expected, several studies using pharmacological and genetic approaches have revealed a greater complexity than originally predicted (9, 11). Several of these ion channels function as macromolecular complexes. They either physically interact with each other or are functionally coupled at specific sites within the plasma membrane (9). Molecular and technical advancements that regulate the action potential hold the promise that these genes could be manipulated to treat arrhythmias (10, 11).

Even though technical advances have facilitated the establishment of proof of principle to treat cardiac diseases, the precise mechanism for cardiac diseases such as arrhythmias is still not completely deciphered. There are several gaps in our understanding of the molecular basis of cardiac function and dysfunction. It is becoming clear that there are subcellular pathways responsible for aberrant cardiac dysfunction. One of the key organelles responsible for meeting the crucial energy demand of the heart is the mitochondria. Heart dysfunction is often accompanied by mitochondrial dysfunction, providing a crucial link between bioenergetics and cardiac mechanics (12–14). In a typical mammalian heart, mitochondria account for 30% of the myocardium by volume and produce over 90% of cardiac adenosine triphosphate (ATP) (15, 16). Mitochondria oxidize dietary hydrogen (from food fed through the pyruvate cycle) with O₂ to generate heat and ATP. Ideally, the inner membrane of mitochondria is maintained at around −180 to −240 mV. The depolarization of the inner mitochondrial membrane is often linked to disruption of ATP production, generation of reactive oxygen species (ROS), and triggering apoptosis. Mitochondria are one of the most negatively charged organelles in the cell and also maintain ionic homeostasis inside the matrix, which is different from cytosol and extra cellular environment (Table 1). Ionic homeostasis in mitochondria is regulated by ion channels and transporters present in mitochondrial membranes.

Table 1.

Ion concentrations in a different compartment of the cell

Ion	Cytosol	Mitochondrial Matrix	Extracellular (Serum)	References
H+	62 nM	10–20 mM	0.1 mM	(18–20)
K+	150 mM	215 mM	5 mM	(6, 21–23)
Na+	10 mM	5 mM	150 mM	(24–28)
Ca2+	40–100 nM	0.1–2 µM	2.5 mM	(18, 29–34)
Mg2+	17–20 mM	0.35–2.4 mM	0.85–1.10 mM	(28, 31, 35, 36)
Cl−	5–60 mM	0.9–22.2 mM	96–105 mM	(37–39)
PO4−	∼6 mM	∼15 mM	0.83–1.5 mM	(40, 41)

Mitochondria have two membranes, outer and inner that create two isolated compartments (Fig. 1): the intermembrane space and the mitochondrial matrix (17). Though the outer membrane of mitochondria possesses ion channels (42), it is less selective and acts as a molecular sieve. In contrast, the inner membrane is highly ion-selective (43–47). The key function of the inner mitochondrial membrane is the generation of proton motive force, utilizing the energy produced by burning fats and sugars (48–50). The proton motive force is then used by ATP synthase in mitochondria to synthesize ATP from ADP and inorganic phosphate (Pi). Membrane potential across the inner membrane is very high, which causes the mitochondrial matrix to accumulate Ca²⁺ ions and form complexes with phosphates and carbonates (51). The majority of the ion channels characterized so far are present in the inner mitochondrial membranes (14, 51). Surprisingly, there are continuous dynamic processes involving the transportation of ions across the inner membrane resulting in the accumulation of salt precipitates in the mitochondrial matrix (24). These salt precipitates (24) are vital during pathological stressors such as ischemic or hypoxic insults where they act as pH buffers and contribute to the continuous production of ATP. Mitochondrial ion channels are extensively studied in terms of protecting the heart or cardiac cells from ischemia-reperfusion injury. The molecular identity of several of these cardioprotective mitochondrial ion channels has been established, facilitating the more focused approach to target them for therapeutics. However, whether mitochondrial ion channels play any role in the physiology of the heart is still being elucidated (Fig. 1).

Figure 1.

Ion channels present in outer, inner, or both membranes regulate ion flux across mitochondrial membranes. Mitochondrial ionic homeostasis regulates the membrane potential of the mitochondrion. Electron transport chain (ETC) coupling present in the inner membrane is controlled by mitochondrial membrane potential which in turn modulates ATP and reactive oxygen species (ROS) production. Mitochondrial ions also directly regulate ROS production by mitochondrial ETC complexes. Mitochondrial ions and ROS in turn regulate mitochondrial permeability transition pore (mPTP). Ionic homeostasis is coupled with mitochondrial ROS either directly or indirectly via regulating ion channels in mitochondria or other organelles. All the factors, mitochondrial ions, ROS, and ATP, modulate the cardiac function and also participate in cardioprotection against ischemia-reperfusion injury.

This review aims to highlight the role of mitochondrial ion channels in cardiac function, emphasizing that mitochondrial ion channels, similar to the cardiac plasma membrane channels, also exist as macromolecules and are functionally coupled with each other either in specific sites or by physical interactions. The majority of the focus for this review will concentrate on ion channels present in cardiac mitochondria.

MITOCHONDRIA AS THERAPEUTIC TARGETS

Due to the extensive role of mitochondria in different physiological and pathophysiological conditions, a new term, “mitochondrial medicine,” has been coined to focus on mitochondrial diseases as well as mitochondria as a therapeutic target (52, 53). During oxidative phosphorylation (OXPHOS), electrons are collected from the oxidation process to allow the generation of reducing equivalents nicotinamide adenine dinucleotide (NAD) + hydrogen (H) and flavin adenine dinucleotide (FADH2). Electrons flow through the complexes of the electron transport chain (ETC) (Fig. 1), H⁺ are pumped out into the intermembrane space from the matrix. This creates a strong negative mitochondrial membrane potential designated as Δψ_m, which can be utilized to target drugs to the mitochondria. The movement of H⁺ down the proton gradient across the inner membrane drives the final complex of the ETC, ATP synthase, which converts ADP to ATP. The proton gradient is coupled with other mitochondrial ions to maintain appropriate ionic homeostasis. Any disruption in mitochondrial ion homeostasis due to activation or deactivation or abnormal behavior of mitochondrial channels could lead to mitochondrial dysfunction. The major readouts from mitochondria such as ATP, ROS, and Ca²⁺ that are regulated by mitochondrial ion channels and transporters play a significant role in cardiac function and cardioprotection (Fig. 1). Therefore, it is imperative to establish the molecular identity of mitochondrial ion channels and transporters. Once the molecular identity is established, it will facilitate elucidation of the specific role and function of cardiac mitochondrial ion channels. The next key question is how these mitochondrial ion channels target specifically to outer/inner or both membranes of mitochondria. Some of the channels possess a classical mitochondrial targeting sequence (54, 55) but several of these channels lack the mitochondrial targeting sequence. Another important mitochondrial targeting factor is splice variation (56, 57), which facilitates the interaction of ion channels with chaperones (58) to target them to mitochondria (59–62).

MITOCHONDRIAL CALCIUM CHANNELS

Mitochondrial Ca²⁺ plays an important role in the regulation of mitochondrial and cellular function. Mitochondria take up Ca²⁺ from the cytosol in response to cellular stimuli which in turn modulate the TCA cycle, energy production, and cellular metabolism. To uptake Ca²⁺, mitochondria are often located in close proximity to a Ca²⁺ source [e.g., near the plasma membrane, endoplasmic reticulum (ER), or sarcoplasmic reticulum (SR)]. Ca²⁺ transport into mitochondria is tightly regulated as excessive Ca²⁺ influx will activate the formation and opening of mitochondrial permeability transition pore (mPTP). When the oxidative phosphorylation chain is active, the membrane potential on the inner mitochondrial membrane is highly negative (∼−180 mV), which causes the influx of calcium to the matrix through a high capacity Ca²⁺-selective ion channel called mitochondrial calcium uniporter (MCU) channel (63–65). The channel consists of four MCU subunits. The MCU super complex consists of EMRE, MICU1/2, MCUb, MCUR1, and SLC25A23 (66). Ca²⁺ flux through the MCU channel (67) super complexes regulates Ca²⁺-dependent cell functions such as apoptosis, autophagy, transcription, and cell survival (68).

In cardiac muscle, Ca²⁺ plays a significant role in contraction and relaxation. Mitochondrial Ca²⁺ in cardiac tissue plays a critical role in meeting the energy demands. Though pharmacological and mitochondrial Ca²⁺ studies implicated MCU in heart function, the major breakthrough came after the discovery of MCU and its components as well as genetic mice models. Global and cell-specific null-mutant mice as well as gain-of-function indicated that MCU is involved in the onset and transmission of Ca²⁺ transients during cardiac contractions. At the resting stage, null-mutant mice for MCU present normal cardiac function (69) but during stress such as fight-or-flight responses MCU null-mutant mice showed an impaired heart rate acceleration (70). During ischemia and reperfusion, Ca²⁺ flux increases in mitochondria, resulting in the formation and opening of mitochondrial permeability transition pore (mPTP). The absence of MCU prevents Ca²⁺ overload of mitochondria, hence offers cardioprotection against ischemia-reperfusion injury (71, 72). Given the importance of Ca²⁺ in mitochondria function, it is expected to possess several other Ca²⁺ channels and transporters such as mitochondrial Na⁺/Li/Ca²⁺ exchanger (NCLX) (73). In agreement, the lack of MCU significantly reduced mitochondrial matrix Ca²⁺ (to around 25% of wild type) but does not completely eliminate it (74), indicating the presence of an independent Ca²⁺ uptake mechanism. Therefore, to decipher the role of mitochondrial Ca²⁺ channels in cardiac function, other possible Ca²⁺ channels are to be identified. On these lines, mitochondrial NCLX encoded by Slc8b1 was shown to be a vital mechanism involved in mitochondrial Ca²⁺ flux. Overexpression of NCLX offers cardioprotection to mice subjected to the permanent occlusion of the left coronary artery or pressure overload (73). In contrast, ablation of NCLX caused sudden death, and surviving mice presented left ventricular dilation and compromised left ventricular function (73). Mice lacking NCLX presented cardiac fibrosis and hypertrophy (73) indicating that Ca²⁺ is important for the maintenance of cardiac structure and function.

Studies carried out on mitochondrial ion fluxes and mechanisms involved demonstrate that mitochondrial Ca²⁺ homeostasis is regulated by MCU, and NCLX is indispensable for cardiac function at the basal as well as during stress.

MITOCHONDRIAL SODIUM CHANNELS

Analysis of Na⁺ ions in mitochondria (mitoNa⁺) versus cytosol (cytoNa⁺) indicate that mitoNa⁺ concentrations are either equal or lower than cytoNa⁺ concentrations (27, 28, 75). Using different types of assays, mitoNa concentration is estimated to be around 5 mM in cardiac cells (28, 75), and 50–113 mM in Madin-Darby Canine Kidney (MDCK) cells (26). When compared with other ions, mitoNa⁺ was nearly equivalent to the H⁺ gradient. However, the uptake of Pi by respiring mitochondria decreases matrix pH and increases matrix [Na⁺]. The concentration of mitoNa⁺ is maintained as a result of its coupling with an endogenous Na⁺-H⁺ antiporter. In ventricular myocytes, inhibition of both aerobic metabolism and glycolysis produces a gradual increase in mitoNa. An increase in mitochondrial Na⁺ concentration is coupled with a decrease in systolic contraction and ATP levels (28, 75). The existence of Na⁺/H⁺ exchanger (NHE) was proposed by Mitchell and Moyle in 1969 (76), and later on, NHE was proposed by Garlid in 1991 (77). NHE was shown to be the major cation transporter in mitochondria.

In myocardial infarction (MI), mitoNa⁺ levels show an early increase which could be due to decreased activity of NHE and hence restricted efflux of Na⁺. During MI, NCX could still be active causing accumulation of cytosolic Na⁺. In due course, mitochondrial NCX (NCLX) is also reversed which will cause the second phase decrease in mitoNa⁺ levels. The second phase decrease was completely abolished in the presence of CGP-37157, a specific inhibitor of the mitochondrial NCLX (36). These results indicate that levels of mitoCa²⁺, mitoNa⁺, and mitoH⁺ and the mitochondrial membrane potential influence mitochondrial Na⁺ handling.

One of the key roles attributed to Na⁺ ions in mitochondria is to regulate Ca²⁺ homeostasis. Earlier experiments using nigericin (acts as K⁺, Na⁺, H⁺, and Pb²⁺ ionophore, and usually used as antiporter for K⁺ and H⁺ across biological membranes) showed that high rates of Ca²⁺ efflux can be maintained by mitoNa⁺ levels. Na⁺ gradient is usually maintained by the Na⁺-H⁺ antiporter and provides a strong electromotive force to drive Ca²⁺ into mitochondria. Though there is clear evidence for the existence of an efficient Na⁺ exchange mechanism, there are no specific ion channels present to transport Na²⁺. Inhibition of Na⁺ uniporter usually by Mg²⁺ directly impacts mitochondrial function (28). In contrast, mitochondrial ROS and NAD⁺ levels are known to regulate cardiac voltage-gated sodium channel (SCN5A) currents (I_Na). Using SCN5 null-mutant mice (28) it was shown that an increased cytosolic NADH results in decreased I_Na, which can be reversed by NAD⁺, suggesting a link between metabolism and I_Na.

MITOCHONDRIAL POTASSIUM CHANNELS

Mitochondrial matrix volume is regulated by the net balance of K⁺ uptake and release (78). As the concentration of K⁺ is higher in the cytosol compared with mitochondria, they can be assumed as “potassium sinks” within the cell. Mitochondrial membrane potential across the inner membrane drives the electrophoretic influx of K⁺ and is balanced by the electroneutral K⁺/H⁺ antiporter (79, 80). Dysregulation of the K⁺ influx could result in mitochondrial depolarization and matrix swelling. Therefore, K⁺ flux in and out of mitochondria is required to be tightly regulated by modulating and fine-tuning ion channels and transporters. Opening of ion channels permeable to K⁺ is associated with a mild depolarization of mitochondrial membrane potential which could result in cytoprotective effects (80, 81). Complete depolarization of mitochondrial membrane potential could result in mitophagy, which is a highly energy-consuming process and cells will need to invest their resources in producing new mitochondria to replace degraded mitochondria.

Over the last two decades, several K⁺ channels have been described in the inner membrane of mitochondria (Table 2; 56, 82, 83, 87, 89, 92–94, 98–100, 102, 103). The majority of the focus is on ATP-sensitive K⁺ channel (mitoK_ATP, inhibited by physiological levels of ATP) (4, 86, 104, 105) and large-conductance Ca²⁺ and voltage-gated K⁺ channel (mitoBK_Ca) (2, 6, 56, 90, 106–109). Recently, additional K⁺ channels were identified in the inner membrane of mitochondria (86, 110). There are several outstanding questions on the existence of multiple potassium channels in mitochondria. It is possible that the presence of mitochondrial potassium channels is cellular or organ-specific. One or more potassium channels co-exist in the inner membrane of mitochondria which are functionally coupled with each other as well as with K⁺/H⁺ antiporter and other transporters (81). Mitochondrial potassium channels form a dedicated potassium ion transport mechanism that can either function in tandem or compensate for the loss of one or the other channel. At the organelle level, these potassium channels are implicated in regulating mitochondrial function, such as energy-production, ROS generation, ionic homeostasis, Ca²⁺ retention capacity (CRC) handling, and membrane potential (2, 111). At cellular and organ levels, the opening of mitochondrial K⁺ channels is associated with protecting cells and organs from hypoxia/ischemia-reperfusion injury (2, 80, 112). There is a well-established cross-coupling between sarcolemmal K⁺ channels and fluctuation in Ψ_mito. Specifically, the opening of sarcolemmal K_ATP channels causes hyperpolarization of the membrane, which creates a conduction block and reduces mitochondrial energy production which is termed as a “metabolic sink” (113).

Table 2.

Potassium channels in mitochondrial membranes

Potassium Channels	Gene	References
mitoKATP channel	KCNJ1, CCDC51	(82–86)
mitoSKCa channel	KCNN1, KCNN2, and KCNN3	(87, 88)
mitoIKCa channel	KCNN4	(89)
mitoBKCa channel	KCNMA1	(56, 90, 91)
mitoKv1.3 channel	KCNA3	(92)
mitoKv1.5 channel	KCNA5	(93)
mitoKv7.4 channel	KCNQ4	(94)
mitoTASK3 channel	KCNK9	(95–97)
mitoSLO2	KCNT2	(98, 99)
mitoHCN	HCN1, HCN2, HCN3, and HCN4	(100, 101)

BK_Ca, large conductance calcium-activated potassium channels; HCN, hyperpolarization-activated cyclic nucleotide-gated; IK_Ca, intermediate-conductance Ca²⁺-gated K⁺ channel; K_ATP, ATP-gated potassium channel; K_v, voltage-gated potassium channel; SK_Ca, small-conductance Ca²⁺-gated K⁺ channel.

Though the role of mitochondrial potassium channels in cardioprotection is well established and reviewed by several independent groups, here, I will focus on the role of mitochondrial potassium channels in cardiac function. The major challenge in defining roles for mitochondrial potassium channels is their presence in other membranes such as plasma membrane, and nuclear or endoplasmic reticulum membranes. Unless otherwise specified, these functions could also originate from nonmitochondrial structures. A significant increase in ventricular funny currents (I_f) in a variety of cardiac diseases due to hypertension, and ischemic and dilated cardiomyopathy has been reported (114). I_f rises from hyperpolarization-activated cyclic nucleotide-gated (HCN) channels, and their expression also varies in the myocardium and is highly dependent on the development and the presence of pathological state of the heart. Na⁺-activated potassium channels (Slick, K_Na, SLO2) are activated by Ca²⁺ and Cl⁻ and present high conductance and prominent subconductance states (115). These channels were shown to be present in cardiomyocyte plasma and mitochondrial membranes. Activating K_Na results in cardioprotection from ischemia-reperfusion injury. Rate pressure product (RPP), a measurement of the internal workload of the heart, indicates that the expression of Slo2 is vital for maintaining heart rate and systolic blood pressure (99). A member of the KCNK gene family, TASK3 channels are leaking conductances that result in background currents and are usually present in most cell membranes. They are also localized to the mitochondria where they are predicted to regulate cristae morphology, membrane potential, apoptosis, and aldosterone biosynthesis (95–97). Aldosterone receptors are present in the heart and are known to modulate heart rate (116).

Among voltage-gated potassium channels, Kv_1.3, Kv_1.5, and Kv_7.4 have been reported in inner mitochondrial membranes (92–94). Specifically, retigabine, a Kv_7.4 opener in the rat heart, improved cardiac function such as RPP and change in pressure over time (dP/dt) in a Langendorff model albeit at a higher concentration (94). Though Kv_1.5 is known to play a vital role in repolarizing K⁺ currents, the role of mitoKv_1.5 is not yet elucidated. Kv_1.3 is a key player in metabolism as the null mutant mice present lower body weight and decreased adiposity, increased total energy expenditure, increased locomotor activity, and resistance to either diet- or genetic-induced obesity. Both Kv_1.3 and Kv_1.5 are shown to play a crucial role in coupling myocardial blood flow to cardiac metabolism.

mitoK_ATP is one of the most extensively studied mitochondrial K⁺ channels. Opening of K_ATP protects the heart from IR injury but under physiological conditions, the channel is expected to be dormant. The molecular identity and precise pharmacological tools for mitoK_ATP are still being characterized. Recently, CCDC51 together with a member of the ABC transporter family (ABCB8) was shown to form a mitoKATP-like channel. The renal-outer-medullary-potassium (ROMK, KCNJ1) channel and MITOK (CCDC51) are the forefront candidates for the mitoK_ATP channel (85, 86). Pharmacological opening of K_ATP by diazoxide improves cardiac function including left ventricular ejection fraction (LVEF) and fractional shortening (LVFS). ROMK null-mutant mice have reduced blood pressure at baseline (117), a decreased afterload, and altered arterial elastance, but increased preload-independent indexes indicating a direct effect on cardiac contractility (85). ROMK null-mutant mice do not develop bradycardia or ECG abnormalities at the neonatal stage but adult mice present significant abnormalities in ECG, which are predicted to be caused by electrolyte imbalances associated with hypo-hyperkalemia (85). The decrease in left ventricle dimension presented in ROMK null-mutant mice (KCNJ1^−/−) is also caused by systemic defects. However, the ROMK cardiomyocyte-specific knockout mouse (KCNJ1^fl/fl × α-MHC-Cre mouse) presented no changes in the cardiac structure (85). The majority of the pharmacological studies focused on K_ATP involve diazoxide or the substituted fatty acid 5-hydroxydecanoate (5-HD) (118–120). In the literature, there is evidence of nonspecific actions of diazoxide and 5-HD as well as the existence of ATP-sensitive K⁺ currents that are not sensitive to known “mitoK_ATP” activators or blockers (68, 118–121). However, pharmacological blocking of mitoK_ATP does prevent the beneficial antiarrhythmic effects seen with ischemic preconditioning. However, activation of mitoK_ATP channels presents contrasting results in preventing arrhythmia before the onset of ischemia but not after ischemic insult (122, 123). Genetic models do provide an insight into the possible role of ATP-sensitive channels in cardiac function or at least with respect to cardioprotection against ischemia-reperfusion injury (86).

Our group and others have established that BK_Ca is present in cardiac mitochondria (2, 56, 79, 90, 91, 106, 124–130), and is a C-terminus splice variant encoded by the Kcnma1 gene (56). In adult cardiomyocytes, BK_Ca is exclusively present in the inner membrane of mitochondria and not in the plasma membrane (56, 131). BK_Ca has been shown to play a role in cardiac function. Inhibition of BK_Ca channels mediated bradycardia is the direct result of sinoatrial (SA) nodal cell inhibition and reduction of SA nodal action potential generation (132, 133). Using pharmacological tools, it was shown that BK_Ca does play a role in cardiac function (88, 134). Paxilline (cell-permeable blocker, possibly targeting mitoBK_Ca) altered the heart rate, LVEF, and global longitudinal strain within a range of 1–15 min, but iberiotoxin [IBTX, cell impermeable blocker (135)] did not present any change up to 40 µg/mL. IBTX did reduce the coronary artery mean velocity indicating the involvement of plasma membrane BK_Ca in coronary arteries (134). In the guinea pig, blocking BK with Paxilline or SK with NS8593 or both antagonists at the same time increased myocardial infarction and worsened the left ventricular pressure (88). In agreement with pharmacological data (134), implicating mitoBK_Ca in cardiac function, a cardiomyocyte-specific BK_Ca knockout mouse also showed a reduction in global longitudinal strain, LVEF, and an increase in fibrosis (90). These studies exclusively performed in adult animal models show that mitoBK_Ca participate in cardiac function as well as cardioprotection. The expression of BK changes with age and specifically in coronary myocytes that affect its contractility (136). In the future, it will be of interest to study how changes in the expression of cardiac BK with age (neonates, young, and old) affect cardioprotection from IR injury and cardiac function.

Surprisingly, most of the mitoK channels are opened by cAMP (137–139) or cGMP (90, 140–143) or both (144), which provides a common theme for their physiological activation in mitochondria in addition of voltage and Ca²⁺ and other factors. Initial data, from studies on ROMK and BK_Ca suggest that regardless of the mechanism involved, mitochondrial K⁺ channels can be tested as possible targets on a short-term or acute basis to modulate cardiac inotropy. The key finding is that these channel-mediated impacts are completely reversible which should be taken into account for use during short-term treatment and emergencies.

MITOCHONDRIAL ANION CHANNELS

Chloride currents are implicated in the diastolic potential, action potential duration, and membrane conductance of cardiac tissues. In fact, voltage-clamp experiments indicated that a transient outward current is carried by Cl⁻ ions (145, 146). Though there was some initial evidence of the physiological roles of Cl⁻ in cardiac function, they were largely ignored once potassium was shown to be a major charge carrier for the transient outward current in sheep Purkinje fibers (147). However, persistent efforts by several groups indicated that a distinct Cl⁻ current exists in cardiac cells (148) and in mitochondria (149).

Anion channels, originally termed as leak channels, are being increasingly recognized as key players in regulating mitochondrial ionic homeostasis. Prominent anion channels shown by experiments are presented in Table 3. Though anion channels are often ignored, the very first ion channel to be recognized in mitochondrial was the voltage-dependent anion channel (VDAC). VDAC is predominantly present in the outer mitochondrial membrane and also allows Ca²⁺ to pass through. There are three VDAC isoforms identified of which VDAC1 is the most abundant protein. One of the most prominent roles associated with VDAC apart from the transport of anions, cations, and metabolites is apoptosis. The role of VDAC in cardiac IR injury or function is not yet deciphered but inhibiting VDAC opening significantly attenuates myocardial IR injury (177, 178). Recent studies (179) show that the expression of VDAC1 increases in the human heart after MI, as well as in the left ventricle of patients with chronic cardiac dilation, hypertrophy, and dysfunction. Similar overexpression of VDAC1 was observed in the left atrium and left ventricle of rodents after MI or excessive aldosterone levels. These studies implicate VDAC1 in the myocardium under different pathological conditions, and blocking VDAC1 could provide beneficial effects. However, it is not clear whether the flux of anions or Ca²⁺ is responsible for VDAC1-related myocardial dysfunction. A recent study implicated VDAC2 in influencing excitation-contraction coupling (ECC) and cardiac disease progress progression (180). So far, both VDAC1 and VDAC2 are projected as therapeutic targets for heart failure.

Table 3.

Anion channels in mitochondrial membranes

Anion Channels	Gene	References
VDAC1	Vdac1	(150–153)
VDAC2	Vdac2	(154–158)
VDAC3	Vdac3	(54, 153, 159–162)
CLIC4	Clic4	(163–168)
CLIC5	Clic5	(149, 169)
IMAC	Not identified	(170–172)
Ca2+-activated chloride channel, anoctamin-1	TMEM16A	(173,174)
Ceramide channels	Lipid molecules	(175,176)

CLIC, chloride intracellular channel proteins; IMAC, inner membrane anion channel; VDAC, voltage-dependent anion channel.

Although VDAC was the first mitochondrial ion channel characterized, it was the only anion channel localized to the outer mitochondrial membranes. In the inner mitochondrial membranes, the existence of an Mg²⁺ and pH-dependent anion transport mechanism was described in the early 1960s (181), which was reiterated by a study that indicated anion transportation decreases in low pH (182). Inner mitochondrial membrane presents a varying degree of permeability to anions like Cl⁻, Br⁻, I⁻, SCN⁻, NO³⁻, PO₄⁻, HCO³⁻, and SO₄²⁻ indicating a presence of inner membrane anion channel (IMAC) (183). The single-channel conductance of IMAC is around ∼107 pS (184). Though the molecular identity of IMAC is still not deciphered, it is involved in volume homeostasis. In the heart, IMAC has been implicated in electrical and contractile dysfunction in the heart after ischemic injury (51, 185, 186). In cardiac cells, IMAC inhibition prevents reverse oscillatory mitochondrial depolarizations induced by substrate deprivation and oxidative stress (51, 113, 186, 187). Along with porins, channels made by lipid molecules, Ceramide [C2, C16, C18,-ceramide (N-palmitoyl-d-erythro-sphingosine)] have been proposed to exist in the outer mitochondrial membrane forming a large channel with a diameter of 10 nm (175). These channels are implicated in regulation of apoptosis and could play a role in cardiac protection, but the area remains largely unexplored (175, 176).

Another anion channel, anoctamin 1 (ANO1), a Ca²⁺-activated Cl⁻ channel (CaCC) encoded by TMEM16A, usually found in gastrointestinal epithelia, skeletal, and smooth muscle cells were recently identified in the inner membrane of mitochondria of rat lung microvascular endothelial cells (173). In mitochondria, ANO1 interacts with OPA1 and is implicated in the regulation of mitochondrial bioenergetics and cristae formation. The role of ANO1 in cardiac function is not known but after ischemic injury the expression of ANO1 increases in the cardiac tissue. The channel is associated with accelerated early phase 1 repolarization of action potentials (APs) and a deeper “spike and dome” after ischemia. The role of ANO1 in ischemia-induced arrhythmias is attributed to the plasma membrane variant as information on the mitochondrial variant is not yet complete in cardiac cells.

A novel class of anion channel (37, 188, 189), chloride intracellular channel proteins 4 (CLIC4) and 5 (CLIC5) were recently characterized in mitochondrial membranes (3, 149, 169). CLIC4, originally known as p64H1 was localized to mitochondrial membranes and is also called mtCLIC (163,164, 190,191). Our recent study revealed that CLIC4 is localized to the outer membranes of cardiac mitochondria. On the other hand, CLIC5 was localized to the inner mitochondrial membranes. There are no specific roles assigned to CLIC4 or CLIC5 in cardiac function. The single-channel conductance of CLIC4 is ∼15 pS but CLIC5 can present up to ∼110 pS conductance. The channel activity of CLIC4 and CLIC5 is blocked by IAA-94. IAA-94 was originally used to purify p64 protein (192) and which was later named CLIC. IAA-94 increases myocardial infarction after ischemia-reperfusion injury and cardiac cell death implicating CLIC proteins in cardioprotection (193–196). IAA-94 also reverses and even stopped high ventricular fibrillation frequencies caused by hypo-osmotic stress, indicating that IAA-94-sensitive Cl⁻ channels in the maintenance of ventricular fibrillation (197). All the studies carried out so far implicate IAA-94-sensitive Cl channels, such as CLICs in cardioprotection and arrhythmias. In the future, null-mutant mice could provide information on mitochondrial anion channels in cardiac function.

Mitochondrial anion channels are increasingly being implicated in cell death, pH and volume regulation, cardioprotection against ischemic injury as well as arrhythmias. A combination of genetic and specific pharmacological tools is required to dissect the specific roles of these channels in cardiac function.

MITOCHONDRIAL GAP JUNCTIONS

The key mechanism of communications between cardiac cells is the gap junction. Gap junctions, also termed connexins, are also responsible for the transport of regulatory molecules between cells (198). There are four different members of the connexin (Cx) family of gap junctions in the heart: Cx43, Cx45, Cx40, and Cx37. So far, Cx43 is localized to the cardiac mitochondria with its carboxy-terminus directed toward the intermembrane space (199, 200). Cx43 forms a channel with six subunits allowing a low-resistant current propagation and passage of small molecules. In cardiac mitochondria, it regulates oxygen consumption and potassium fluxes. Inhibition of Cx43 by pharmacological and genetic approaches reduces complex I-mediated oxygen consumption and ATP production (200). Cx43 also regulates mitochondrial ROS and calcium levels (201, 202) in heart cells, implicating them in cardioprotection as well as cardiac function. The expression of mitochondrial Cx43 declines with age (203), hypertension, diabetes, and hypercholesterolemia, and heart failure (199). In the myocardial infarction model, an increase in ROS generation has been linked to the downregulation of Cx43 (204). Decreasing mitochondrial superoxide production rescues mitochondrial function and restores Cx43 expression, which increases conduction velocity and reduces arrhythmias inducibility (122, 205, 206). Though the role of Cx43 in heart function is well established, it is vital to dissect the role of mitochondrialCx43 from the plasma membrane from Cx43.

OTHER ION TRANSPORT MECHANISMS

Another important class of ion transport mechanism in the mitochondrion is SLC25 family proteins. The SLC25 is also known as mitochondrial carrier family (MCF) and has a large diversity in terms of solutes being transported through them. The transport facilitated by these MCFs could be electroneutral or electrogenic. They are divided into citrate (CIC), phosphate (PIC), ADP and ATP (AAC), dicarboxylates (DIC), aspartate, glutamate (AGC), and sulfate (UCPs). MCFs transport ions, amino acids, nucleotides, dinucleotides, carboxylates, keto acids, and substrates transported through them and play a significant role in regulating mitochondrial ionic homeostasis and function. To keep the review focused on ion homeostasis, I have not included nonionic transports through MCFs or some ion channels which are capable of transporting small molecules (150, 151, 157, 207–212).

The ion transport mechanism responsible for regulating mitochondrial proton concentrations is uncoupling proteins (UCPs). There are five UCP homologs; UCP1 (SLC25A7), UCP2 (SLC25A8), UCP3 (SLC25A9), UCP4 (SLC25A27), and UCP5 (SLC25A14) in mammals, and they are located in the inner membrane of the mitochondrion. UCP1 is chiefly localized to the brown adipose tissue, UCP2 is present in the heart, and UCP-3 is present in skeletal muscles. In the heart, loss of UCP2 from fibroblasts results in loss of right ventricular function upon pressure overload (213), however, in heart failure, UCP-2 inhibition in cardiomyocytes is beneficial as it improves the efficiency of glucose oxidation (214). In failing rat hearts, levels of UCP2 increase in pressure overload (215) and after myocardial infarction (216). These studies emphasize the role of UCPs in the transition from cardiac hypertrophy to heart failure. In terms of mechanism, there are two possible mechanisms proposed: 1) regulation of ROS by UCPs and 2) regulation of metabolic pathways. As proton is a major player in regulating ionic homeostasis, the key mechanism involved could be directly associated with proton concentration inside the mitochondria modulated by UCPs, and disruption will result in loss of mitochondrial membrane potential. The loss of mitochondrial membrane potential will increase ROS generation and decoupling of ETC, hence causing disruption of metabolic pathways.

Phosphates are the major anion present in the mitochondrial matrix. Mitochondrial phosphate carrier, SLC25A3, is the major mechanism in PO₄⁻ transport, and energy production. Two MCFs, PiCA and PiCB are responsible for PO₄⁻ transport for ADP phosphorylation but no specific ion channel has been identified for PO₄⁻ transport. Abnormal activity or expression of MCFs is known to cause several diseases and disorders including cardiomyopathy, ventriculomegaly, and congenital heart defects. Mice lacking PiC develop cardiac phenotype reminiscent of SLC25A3 deficiency in the long-term, such as hypertrophy with ventricular dilation and compromised cardiac function. However, the deletion of PiC attenuated ischemia-reperfusion injury and partially protected cells from apoptosis (217). In patients with coronary artery disease, changes in PO₄⁻ transport result in mitochondrial DNA deletions and damage to ETC complexes (218). Taken together these studies indicate that PO₄⁻ transportation is tightly coupled with ETC function and metabolism.

Another key ion involved in several important enzymes as a cofactor is Mg²⁺. Mg²⁺ is the fourth most abundant mineral in the mammalian body, and the second most abundant cation. It is also essential for the energy production and function of several kinases, ATPases, and nucleic acid metabolism. The mitochondrion is capable of take-up and extrudes Mg²⁺ by respiration-dependent reactions. Mg²⁺ is directly involved in the regulation of Ca²⁺ transport in mitochondria. Mitochondrial Mg²⁺ transport was first described around 60 years ago (219, 220). The transporters of Mg²⁺ recently identified in mitochondria are Ymr166c/Mme1 (221) and SLC41A3 (222). Mg²⁺ levels are directly involved in cardiovascular diseases, type 2 diabetes mellitus, heart failure, arrhythmias, and its serum levels also influence the outcome after heart failure in patients (223–230). Mg²⁺ supplementation restores mitochondrial membrane potential, increased ATP production, decreased ROS and Ca²⁺ overload, and cardiac diastolic function in diabetes mellitus mice (230). In patients, Mg²⁺ supplementation is used to acutely treat arrhythmias, decrease mortality after MI, treat hypokalemia, reduce stroke, and mortality associated with diabetes mellitus (231). Although promising in some cardiovascular diseases and associated mortality, Mg²⁺ supplementation does not have the same outcome for re-entrant rhythms, atrial fibrillation, and ischemic heart diseases (231). Similar to other ions, it is vital to dissect the role of mitochondrial versus extramitochondrial Mg²⁺ impact on cardiac function.

PERSPECTIVE VIEW

The field of mitochondrial ion channels has been increasingly recognized for their roles during stress or in apoptosis. A growing number of ion channels have been recognized and reported in mitochondrial membranes. The lack of information on the identity of mitochondrial ion channels is the biggest challenge in the field. So far a handful of cation and anion channels have been identified at the molecular level and in the future, we anticipate the unmasking of additional mitochondrial ion channel proteins. Another challenge in the field is to specifically target mitochondrial ion channels and associated post-translational modifications. All of the ion channels or their complex components are encoded by nuclear DNA (with an exception of ceramides which are made by lipid molecules). Some of the mitochondrial ion channels share biophysical and pharmacological properties with plasma membrane ion channel counterparts (232) making it challenging to specifically target mitochondrial ion channels (233).

At the pharmaceutical and therapeutical end, it is important to unravel methods to specifically target these channels. The high throughput screening is one of the most promising pathways to identify channel-specific modulators. The next significant challenge will be to specifically deliver these compounds to mitochondria and explicitly to channels present in the inner membrane. Recent developments in mitochondrial medicine by using lipophilic cation moieties [e.g., triphenylphosphonium (TPP) and Dequalinium (DQA)], cardiolipin targeting peptides (also known as Szeto-Schiller peptides), and mitochondrial targeting signal peptides (20–40 amino acids N-terminal targeting sequences) have shown promise to target mitochondrial ion channels. In clinical studies, TPP-based compounds MitoQ and SkQ1 are used for trials for diastolic dysfunction (NCT03586414), peripheral arterial disease (NCT03506633), heart failure (NCT02966665), and hypogonadism (NCT02758431) (234). In all the clinical trials, patients are being recruited and it will be interesting to assess the outcomes of these studies in the context of mitochondrial medicine.

In summary, mitochondrial ion channels are potential therapeutic targets for transplant medicine and cardiovascular and neurogenerative diseases. These channels, once identified and characterized, can be targeted by incorporating novel and specific mechanisms to treat diseases and disorders.

GRANTS

The research in the laboratory is supported by NIH Grant HL133050, American Heart Association (AHA) Grants 16GRNT29430000 and 11SDG230059, and W. W. Smith Charitable Trust.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author.

AUTHOR CONTRIBUTIONS

H.S. analyzed data; prepared figures; drafted manuscript; edited and revised manuscript; approved final version of manuscript.