Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. 1999 Sep 1; 519(Pt 2): 347–360.
PMCID: PMC2269505

ATP-sensitive K+ channel openers prevent Ca2+ overload in rat cardiac mitochondria


  1. Mitochondrial dysfunction, secondary to excessive accumulation of Ca2+, has been implicated in cardiac injury. We here examined the action of potassium channel openers on mitochondrial Ca2+ homeostasis, as these cardioprotective ion channel modulators have recently been shown to target a mitochondrial ATP-sensitive K+ channel.
  2. In isolated cardiac mitochondria, diazoxide and pinacidil decreased the rate and magnitude of Ca2+ uptake into the mitochondrial matrix with an IC50 of 65 and 128 μm, respectively. At all stages of Ca2+ uptake, the potassium channel openers depolarized the mitochondrial membrane thereby reducing Ca2+ influx through the potential-dependent mitochondrial uniporter.
  3. Diazoxide and pinacidil, in a concentration-dependent manner, also activated release of Ca2+ from mitochondria. This was prevented by cyclosporin A, an inhibitor of Ca2+ release through the mitochondrial permeability transition pore.
  4. Replacement of extramitochondrial K+ with mannitol abolished the effects of diazoxide and pinacidil on mitochondrial Ca2+, while the K+ ionophore valinomycin mimicked the effects of the potassium channel openers.
  5. ATP and ADP, which block K+ flux through mitochondrial ATP-sensitive K+ channels, inhibited the effects of potassium channel openers, without preventing the action of valinomycin.
  6. In intact cardiomyocytes, diazoxide also induced mitochondrial depolarization and decreased mitochondrial Ca2+ content. These effects were inhibited by the mitochondrial ATP-sensitive K+ channel blocker 5-hydroxydecanoic acid.
  7. Thus, potassium channel openers prevent mitochondrial Ca2+ overload by reducing the driving force for Ca2+ uptake and by activating cyclosporin-sensitive Ca2+ release. In this regard, modulators of an ATP-sensitive mitochondrial K+ conductance may contribute to the maintenance of mitochondrial Ca2+ homeostasis.

Potassium channel-opening drugs are powerful cardioprotective agents capable of limiting cardiac cell injury following ischaemia-reperfusion (Gross, 1995; Grover, 1997; Kloner et al. 1998). These chemically diverse compounds share the property of promoting K+ current through ATP-sensitive K+ (KATP) channels (Quast, 1992; Terzic et al. 1995; Edwards & Weston, 1997). It has been assumed that the cardioprotective action of potassium channel openers is mediated through sarcolemmal KATP channels which, when open, shorten the action potential duration and limit Ca2+ influx into cardiac cells. Growing evidence, however, indicates that the protective efficacy does not always correlate with the degree of action potential shortening, implying additional cellular sites of drug action (Garlid et al. 1997; Liu et al. 1998).

The prime candidates include mitochondria, which also harbour an ATP-sensitive K+ conductance, recognized as the mitochondrial KATP (mitoKATP) channel (Inoue et al. 1991; Paucek et al. 1992; Garlid, 1996; Szewczyk, 1997). Activation of mitoKATP channels by potassium channel openers has been associated with increased survival of cardiac cells following ischaemia and improved post-ischaemic recovery of heart muscle, although the precise mechanism responsible for such an outcome remains partially understood (Garlid et al. 1997; Liu et al. 1998). Mitochondrial dysfunction, secondary to excessive accumulation of Ca2+, is an important mediator of ischaemia-reperfusion injury in the heart (Park et al. 1990; Ferrari et al. 1993; Miyamae et al. 1996; Delcamp et al. 1998). Mitochondrial Ca2+ content is determined by the balance between mitochondrial Ca2+ uptake and release (Gunter et al. 1994; Zoratti & Szabo, 1995). It has been proposed that potassium channel openers, by virtue of their ability to dissipate mitochondrial membrane potential (Czyz et al. 1995; Holmuhamedov et al. 1998), may reduce the driving force for Ca2+ accumulation (Liu et al. 1998). Although maintenance of mitochondrial Ca2+ levels is expected to contribute to cardioprotection, direct evidence that potassium channel openers limit mitochondrial Ca2+ overload is lacking.

Therefore, we here examined the action of potassium channel openers on mitochondrial Ca2+ in isolated cardiac mitochondria and living cardiomyocytes. We report that prototype potassium channel openers impede mitochondrial Ca2+ uptake and promote mitochondrial Ca2+ release, thereby diminishing the amount of accumulated Ca2+ within the mitochondrial matrix.


The investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication no. 85-23, revised 1985), and was approved by the Institutional Animal Care and Use Committee.

Isolation of cardiac mitochondria

Mitochondria were isolated from rat hearts as described previously (Holmuhamedov et al. 1998). Animals (Harlan Sprague-Dawley, Indianapolis, IN, USA) were anaesthetized via intraperitoneal administration of pentobarbital sodium solution (100 mg kg−1). Following thoracotomy, hearts were removed into an ice-cold isolation buffer containing (mM): sucrose, 50; mannitol, 200; KH2PO4, 5; EGTA, 1; and Mops, 5 (pH 7.3 adjusted with KOH), with 0.2 % bovine serum albumin (BSA). Atria were excised, and ventricles cut into 1 mm3 pieces. Two 10 s-long homogenization cycles were performed using a PT 10/35 Polytron (Brinkman, Westbury, NY, USA), followed by a 10 min centrifugation at 750g (Sorvall II centrifuge equipped with a GSA rotor). The supernatant was stored on ice, and the pellet resuspended in isolation buffer, and homogenized using a Potter-Elvehjem tissue grinder with a Teflon pestle to release the remaining mitochondria. The homogenate was then centrifuged at 750 g, and the supernatant combined with that obtained from the previous step and further centrifuged at 7000 g for 15 min. The obtained pellet was suspended in 40 ml of isolation buffer containing no EGTA or BSA, and sedimented at 7000 g for 15 min. Finally, the mitochondrial fraction, resuspended in EGTA- and BSA-free isolation buffer at 20-30 mg protein ml−1, was kept on ice prior to experiments. Protein concentration was determined with a DC protein kit (Bio-Rad, Hercules, CA, USA).

Assessment of mitochondrial respiration

The quality of a mitochondrial preparation was determined using the respiratory control index calculated as the ratio of the rate of oxygen consumption by mitochondria in the presence of 100-500 μm ADP (state 3) to that measured following conversion of added ADP into ATP (state 4). Mitochondria (1 mg of protein) were placed into 1 ml of incubation medium containing (mM): KCl, 110; K2HPO4, 5; pyruvate, 5; malate, 5; and Mops, 10 (pH 7.3), and stirred at 30°C. Measurements of oxygen consumption were made in a multichannel chamber of an ESON-6CH computerized analyser equipped with an oxygen-sensitive minielectrode (Ichas et al. 1994; Holmuhamedov et al. 1998). Data acquisition and processing were performed using Bioquest software (Alekseev et al. 1998). Mitochondria with a respiratory control index exceeding 10 were used to study the effect of potassium channel openers on mitochondrial Ca2+.

Mitochondrial Ca2+ uptake and release

Changes in free Ca2+ concentration, within the mitochondrial suspension, were measured with a Ca2+-selective minielectrode, in conjunction with a reference electrode (Microelectrodes Inc., Bedford, NH, USA), using a computerized analyser (Holmuhamedov, 1986). The selectivity of the electrode for Ca2+ over other cations, such as Mg2+, K+ and Na+, was > 10 000. Electrodes were calibrated before each experiment in incubation medium containing known concentrations of CaCl2 and mitochondrial protein. Calibration of the Ca2+ electrode was done in the presence of mitochondria (1 mg of protein) and the Ca2+ uptake inhibitor Ruthenium Red (0.1 nmol (mg mitochondrial protein)−1) to prevent accumulation of Ca2+ in the mitochondria. The potassium channel openers diazoxide and pinacidil (1-600 μm) did not affect the properties of the Ca2+-selective minielectrodes. Ca2+ uptake was initiated by adding mitochondria to Ca2+-containing incubation medium (final 150 nmol Ca2+ (mg mitochondrial protein)−1). Mitochondrial Ca2+ release was induced 1-3 min following completion of Ca2+ accumulation.

Mitochondrial membrane potential

Mitochondrial membrane potential (ΔΨ) was measured using a tetraphenylphosphonium (TPP+)-sensitive minielectrode (Holmuhamedov et al. 1998). Mitochondria (1 mg of protein) were added to the incubation medium containing 200 nM tetraphenylphosphonium chloride, and ΔΨ was calculated according to the following equation: ΔΨ= 59 log(v/V) - 59 log(10(E -Eo)/59 - 1), where ΔΨ is the mitochondrial membrane potential (in mV), v is the mitochondrial matrix volume (1.6 μl (mg mitochondrial protein)−1; Petronilli et al. 1986), V is the volume of the incubation medium (1 ml), and Eo and E are electrode potentials (in mV) before and after addition of mitochondria, respectively (Kamo et al. 1979). Potassium channel openers, at concentrations of 1-600 μm, did not interfere with the properties of the TPP+-sensitive electrode. Electrodes which exhibited a Nernstian response were used.

Cardiomyocyte culture and laser confocal imaging

Cardiomyocyte cultures were prepared from hearts removed from 1- to 2-day-old rats, after ventricles had been separated and cut into 1 mm3 pieces (Perez-Terzic et al. 1999). Tissue was digested (20 min, 37°C) in ADS buffer (mM: NaCl, 116; Hepes, 20; NaH2PO4, 1; glucose, 5.5; KCl, 5.4; and MgSO4, 0.8, with 0.6 ml l−1 Phenol Red, pH 7.35) with collagenase Type II (0.5 mg ml−1; Worthington, Freehold, NJ, USA) and pancreatin (0.15 mg ml−1; Gibco BRL). Following sedimentation of digested tissue, the supernatant was mixed with newborn calf serum (NCS, 1 ml; Gibco), centrifuged (6 min, 1000 r.p.m.), and the pellet suspended in NCS and incubated at 37°C (5 % CO2). After five to six such cycles to achieve full digestion, cells were pooled in ADS, centrifuged (6 min, 1000 r.p.m.) and the pellet suspended in 8 ml ADS. Cells were then centrifuged (30 min, 3000 r.p.m.) through a discontinuous two-layer Percoll gradient. Cardiomyocytes, present in the first layer from the bottom, were collected, suspended in ADS buffer and centrifuged (6 min, 1000 r.p.m.). The pellet was suspended in plating medium: 67 % Dulbecco's modified Eagle's medium (DMEM; Gibco), 17.9 % medium 199 (Gibco), 10 % horse serum (Gibco), 5 % fetal bovine serum (Gibco), 0.1 % penicillin-streptomycin (50 U ml−1; Gibco) and 20 mM Hepes (pH 7.2). Cells were plated on laminin-coated coverslips, and incubated in serum-containing medium (37°C, 95 % O2-5 % CO2) until use. Cardiomyocytes were transferred to (mM): NaCl, 116; KCl, 4; MgCl2, 1.9; NaH2PO4, 1.7; NaHCO3, 4.3; CaCl2, 1; and Hepes, 16.8 (pH 7.4), and loaded for 20 min at room temperature, in the presence of 0.0025 % pluronic acid, with the potential-sensitive dye tetramethylrhodamine methyl ester (TMRM; 3 μm) or the Ca2+-sensitive dye rhod-2 AM (3 μm) (Molecular Probes). These probes predominantly accumulate within mitochondria (Ehrenberg et al. 1988; Trollinger et al. 1997). Myocyte contraction was prevented by 2,3-butanedione monoxime (20 mM) to limit possible artifacts associated with cell motion. Confocal images were obtained with an LSM 510 laser scanning confocal microscope using integrated software (Carl Zeiss Inc., Thornwood, NY, USA). The single-excitation, single-emission, fluorescent probes TMRM and rhod-2 were excited using the 568 nm line of the Ar-Kr laser, and emitted fluorescence was filtered through long-pass filter settings (LP 585). Recordings were made at room temperature. Digitized images were analysed using the image analysis software ANALYZE (Mayo Foundation, Rochester, MN, USA).


Diazoxide and pinacidil were from RBI or Sigma. Concentrated stock solutions were prepared in dry DMSO. The maximal concentration of DMSO within the incubation medium was kept under 0.5 %, and control experiments were performed with corresponding DMSO concentrations. Cyclosporin A was from Novartis (Basel, Switzerland), and Ruthenium Red from Fluka (Ronkonkoma, NY, USA). Other chemicals were from Sigma, unless otherwise indicated.

Statistical analysis

Data are expressed as means ±s.e.m., and n represents the number of mitochondrial isolations or the number of analysed cardiomyocytes. Comparison between groups was made using Student's t test, and a P value of < 0.05 was considered significant.


Inhibition of mitochondrial Ca2+ uptake by potassium channel openers

Cardiac mitochondria avidly accumulate Ca2+ into their matrices (Fig. 1A). Within ~1 min following addition to a medium containing 150 μm CaCl2, isolated mitochondria accumulated essentially all available extramitochondrial Ca2+ (Fig. 1A). Treatment with the potassium channel opener diazoxide reduced the rate of Ca2+ accumulation (Fig. 1A). In the absence of diazoxide, the initial rate was 617 ± 50 nmol Ca2+ min−1 (mg protein)−1 (n = 13), but it decreased to 585 ± 48 (n = 3), 562 ± 38 (n = 5) and 191 ± 20 nmol Ca2+ min−1 (mg protein)−1 (n = 7) in the presence of 10, 30 and 300 μm of the potassium channel opener, respectively. The IC50 for the inhibitory effect of diazoxide was estimated at 65 μm (Fig. 1B). Diazoxide also reduced the total amount of Ca2+ accumulated into the mitochondrial matrix (Fig. 1A). On average, untreated mitochondria accumulated 146 ± 4 nmol Ca2+ (mg protein)−1 (n = 13), and retained accumulated Ca2+ throughout the duration of recordings. At 30 and 300 μm, diazoxide decreased the total magnitude of mitochondrial Ca2+ uptake by 6 ± 2 (n = 5) and 36 ± 7 nmol Ca2+ (mg protein)−1 (n = 7), respectively (Fig. 1A). The structurally unrelated potassium channel opener pinacidil also slowed the rate and decreased the magnitude of mitochondrial Ca2+ accumulation (Fig. 1A, inset) in a concentration-dependent manner (IC50~128 μm; Fig. 1B, inset).

Figure 1
Potassium channel openers reduce Ca2+ uptake in isolated cardiac mitochondria

Effect of potassium channel openers on the driving force for Ca2+ uptake

Mitochondrial Ca2+ influx occurs through the Ruthenium Red-sensitive Ca2+ uniporter, and application of 100 nM Ruthenium Red prevented Ca2+ uptake (Fig. 2A). The rate of Ca2+ uptake through the uniporter is determined by the membrane potential, and slows down with mitochondrial depolarization (Nicholls & Akerman, 1982). Throughout the time course of mitochondrial Ca2+ uptake, diazoxide (300 μm) reduced both membrane potential and Ca2+ influx (n = 3, Fig. 2B). Following Ca2+ accumulation, the membrane potential of isolated mitochondria was -195 ± 17 mV (n = 6). Addition of 300 μm diazoxide (Fig. 2B) or 600 μm pinacidil (not illustrated) decreased mitochondrial membrane potential by 15 ± 2 (n = 3) and 19 ± 4 mV (n = 3), respectively. Thus, potassium channel openers reduce mitochondrial membrane potential at all stages of Ca2+ uptake, and thereby decrease the driving force responsible for mitochondrial Ca2+ accumulation.

Figure 2
Effect of diazoxide on mitochondrial membrane potential and Ca2+ uptake

Treatment of mitochondria with cyclosporin A (2 μm), an inhibitor of Ca2+ efflux through the permeability transition pore (Crompton et al. 1988), did not prevent diazoxide from reducing the rate of Ca2+ uptake (Fig. 2C). In cyclosporin A-treated mitochondria, the rate of Ca2+ uptake was 671 ± 22 and 337 ± 10 nmol Ca2+ min−1 (mg protein)−1, in the absence (n = 3) and presence (n = 3) of 300 μm diazoxide, respectively (P < 0.01). This cyclosporin-insensitive effect supports a potassium channel opener-mediated reduction in the driving force for Ca2+ uptake through the Ruthenium Red-sensitive mitochondrial uniporter.

However, cyclosporin A did prevent diazoxide from reducing the total amount of accumulated Ca2+ (Fig. 2C). On average, cyclosporin A-treated mitochondria accumulated 149 ± 3 nmol Ca2+ (mg protein)−1 in the absence (n = 3) and 148 ± 3 nmol Ca2+ (mg protein)−1 in the presence (n = 3) of diazoxide. This suggests that potassium channel openers could also promote Ca2+ efflux through a cyclosporin A-sensitive pathway.

Unloading of Ca2+ from mitochondria by potassium channel openers

Besides preventing Ca2+ accumulation (Fig. 1), diazoxide, in fact, also promoted release of pre-accumulated Ca2+ from mitochondria (Fig. 3A). In the absence of diazoxide, the amount of accumulated Ca2+ within the matrix was 149 ± 2 nmol Ca2+ (mg protein)−1 (n = 10), a Ca2+ load sustained for at least 3 min after initiation of mitochondrial Ca2+ uptake (Fig. 3A). Within 30 s of addition, diazoxide, used at 30 and 300 μm, reduced the total amount of loaded Ca2+ by 3 ± 1 (n = 4) and 17 ± 3 nmol Ca2+ (mg protein)−1 (n = 9), respectively (Fig. 3A). The rate of mitochondrial Ca2+ release was essentially non-detectable in the absence of the opener (n = 10), but increased to 6 ± 2 (n = 4) and 120 ± 15 nmol Ca2+ min−1 (mg protein)−1 (n = 6) in the presence of 30 and 300 μm diazoxide, respectively (Fig. 3A). The IC50 for the effect of diazoxide was estimated at 96 μm (Fig. 3B). Pinacidil also promoted release of pre-accumulated Ca2+, and stimulated mitochondrial Ca2+ unloading in a concentration-dependent manner with an apparent IC50 of 132 μm (Fig. 3B, inset).

Figure 3
Potassium channel openers unload Ca2+ from pre-loaded mitochondria

Potassium channel opener-mediated Ca2+ release and the permeability transition pore

The major route for massive mitochondrial Ca2+ release is through the mitochondrial permeability transition pore (Gunter et al. 1994). Here, the specific inhibitor of the permeability transition pore, cyclosporin A (Crompton et al. 1988), prevented potassium channel opener-induced Ca2+ release from mitochondria pre-loaded with 150 nmol Ca2+ (mg protein)−1 (Fig. 4). In the absence of cyclosporin A, diazoxide (300 μm) and pinacidil (600 μm), within 30 s of addition, reduced the total amount of loaded Ca2+ by 24 ± 6 (n = 3) and 20 ± 6 nmol Ca2+ (mg protein)−1 (n = 3) (Fig. 4A and B). In the presence of cyclosporin A (2 μm), diazoxide (300 μm) and pinacidil (600 μm) reduced the total amount of loaded Ca2+ by 0.3 ± 0.1 (n = 4) and 0.7 ± 0.1 nmol Ca2+ (mg protein)−1 (n = 4), respectively (Fig. 4A and B). Concomitantly, cyclosporin A (2 μm) did not prevent mitochondrial depolarization induced by 300 μm diazoxide (Fig. 4A) or 600 μm pinacidil (Fig. 4B). In the absence and presence of cyclosporin A, diazoxide induced mitochondrial depolarization of 15 ± 4 (n = 20) and 18 ± 6 mV (n = 7), respectively (P > 0.05). Similarly, pinacidil depolarized the mitochondrial membrane by 20 ± 5 (n = 6) and 23 ± 3 mV (n = 6) in the absence and presence of cyclosporin A, respectively (P > 0.05). Thus, potassium channel opener-induced Ca2+ discharge from pre-loaded mitochondria occurs, at least in part, through a cyclosporin A-sensitive pathway downstream from the mitochondrial recognition site for diazoxide or pinacidil.

Figure 4
Effect of cyclosporin A on potassium channel opener-induced membrane depolarization and Ca2+ release

Dependence on extramitochondrial K+

Activation of mitoKATP channels by potassium channel openers increases K+ flux across the mitochondrial membrane (Garlid et al. 1996; Jaburek et al. 1998). This is prevented by removal of extramitochondrial K+ (Holmuhamedov et al. 1998). In nominally K+-free medium (i.e. 110 mM KCl replaced by the osmotic equivalent, 220 mM mannitol), both diazoxide and pinacidil were unable to interfere with mitochondrial Ca2+ transport (Fig. 5). In the absence of openers, replacement with mannitol did not significantly modify mitochondrial Ca2+ accumulation, and the initial rate of uptake was 659 ± 52 nmol Ca2+ min−1 (mg protein)−1 (n = 3) compared with 617 ± 50 nmol Ca2+ min−1 (mg protein)−1 (n = 13) in KCl (Fig. 5A). However, in mannitol, neither diazoxide (300 μm) nor pinacidil (600 μm) could reduce Ca2+ uptake, which remained at 550 ± 29 (n = 3) and 605 ± 66 nmol Ca2+ min−1 (mg protein)−1 (n = 3), respectively (Fig. 5A). Similarly, replacement of KCl with mannitol also prevented diazoxide and pinacidil from promoting Ca2+ release in pre-loaded mitochondria (Fig. 5B). In mannitol-containing medium, the rate of Ca2+ discharge was 3.5 ± 0.7 (n = 3) and 5.0 ± 2.5 nmol Ca2+ min−1 (mg protein)−1 (n = 3) in the presence of diazoxide (300 μm) and pinacidil (600 μm), respectively (Fig. 5B). These values are significantly lower than the 142 ± 18 (n = 4) and 131 ± 16 nmol Ca2+ min−1 (mg protein)−1 (n = 7) recorded with these openers in the presence of KCl (Fig. 5B). Thus, the effects of potassium channel openers on mitochondrial Ca2+ uptake and discharge require extramitochondrial K+.

Figure 5
Replacement of K+ with mannitol abolishes the effects of potassium channel openers

The K+ ionophore valinomycin also reduces mitochondrial Ca2+ load

Valinomycin is an effective K+ ionophore which facilitates K+ flux across the mitochondrial membrane, and is an established tool for studying the consequences of increased K+ conductance in mitochondria (Kinnally & Tedeschi, 1982; Holmuhamedov, 1986). Here, much like potassium channel openers, valinomycin also promoted Ca2+ release from pre-loaded mitochondria (Fig. 6). Within 30 s of addition, valinomycin (1 ng (mg protein)−1) reduced the total amount of loaded Ca2+ by 35 ± 5 nmol Ca2+ (mg protein)−1 (n = 3) (Fig. 6A). Moreover, in the presence of valinomycin, the rate of Ca2+ release from pre-loaded mitochondria was increased to 110 ± 17 nmol Ca2+ min−1 (mg protein)−1 (Fig. 6B). The ability of valinomycin to reduce mitochondrial Ca2+ load was prevented by substitution of mannitol for KCl (Fig. 6B), and suppressed by cyclosporin A (not illustrated). Thus, the K+ ionophore mimicked the effect of potassium channel openers in reducing mitochondrial Ca2+ load through a K+-dependent mechanism.

Figure 6
Valinomycin induces release of Ca2+ from mitochondria

ATP and ADP inhibit the effect of potassium channel openers

Cytosolic adenine nucleotides, ATP and ADP, inhibit K+ flux through mitoKATP channels (Garlid, 1996). Here, mitochondria were pre-treated with atractyloside (20 μm), an inhibitor of the mitochondrial adenine-nucleotide translocator (Fiore et al. 1998), to prevent translocation of added ATP or ADP into the matrix. In the presence of 100 μm ATP (Fig. 7A) or 100 μm ADP (Fig. 7B), added to the mitochondrial suspension, diazoxide (300 μm) could no longer reduce the rate of Ca2+ uptake, diminish the total amount of accumulated Ca2+ or depolarize the mitochondrial membrane. Moreover, both ATP (250 μm) and ADP (150 μm) prevented potassium channel opener-induced release of Ca2+ from pre-loaded mitochondria and steady-state membrane depolarization (Fig. 8). These mitochondria maintained their sensitivity towards a K+ ionophore, and consequently valinomycin (1 ng (mg protein)−1) could depolarize mitochondria and release Ca2+ despite the presence of nucleotides ATP (Fig. 8A) and ADP (Fig. 8B). In contrast to ATP and ADP, other nucleotides which do not inhibit mitoKATP channels (Garlid, 1996), namely UTP, UDP, GDP and GTP (250 μm each), did not prevent the effects of diazoxide or pinacidil on mitochondrial membrane potential or Ca2+ release (Fig. 9).

Figure 7
ATP and ADP inhibit diazoxide-induced mitochondrial depolarization and reduction in Ca2+ uptake
Figure 8
ATP and ADP inhibit diazoxide- but not valinomycin-induced mitochondrial depolarization and Ca2+ release
Figure 9
Effect of nucleotides on diazoxide-induced Ca2+ release from pre-loaded cardiac mitochondria

Potassium channel openers depolarize mitochondria and reduce mitochondrial Ca2+ content in intact cardiomyocytes

Cardiomyocytes, loaded with the mitochondrial potential-sensitive fluorescent dye TMRM, displayed punctate staining of the cytosol (Fig. 10A) consistent with the mitochondrial localization of this probe (Ehrenberg et al. 1988). Diazoxide (300 μm), within 5 min following application, reduced the intensity of TMRM fluorescence (Fig. 10A). On average, the TMRM fluorescence decreased from 175 ± 9 arbitrary fluorescence units (a.u.) in untreated cardiomyocytes to 115 ± 11 a.u. following addition of diazoxide (n = 24, obtained from 3 separate cell isolations), in line with potassium channel opener-induced mitochondrial membrane depolarization. Treatment of cardiomyocytes with the mitoKATP channel blocker 5-hydroxydecanoic acid (5-HD; Sato et al. 1998) did not change the intensity of TMRM fluorescence, estimated at 168 ± 20 a.u. (Fig. 10B). However, in cells treated with 300 μm 5-HD, diazoxide failed to significantly decrease TMRM fluorescence, which remained at 155 ± 7 a.u. (Fig. 10B). In contrast, treatment with 5-HD did not prevent mitochondrial depolarization by the mitochondrial uncoupler carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP; Fig. 10B).

Figure 10
Diazoxide-induced mitochondrial depolarization in intact cardiomyocytes: prevention by the mitoKATP channel blocker 5-HD

Cardiac cells loaded with rhod-2 AM also displayed punctate staining of the cytosol in accordance with mitochondrial localization of this Ca2+-sensitive fluorescent dye (Trollinger et al. 1997). Rhod-2 fluorescence decreased following application of 300 μm diazoxide (Fig. 11A). On average, this decrease was from 184 ± 32 a.u. in the absence to 115 ± 13 a.u. in the presence of diazoxide (n = 24, obtained from 3 separate cell isolations), indicating an opener-induced decrease in mitochondrial Ca2+ content. As in the case with TMRM, pre-treatment of cardiac cells with 5-HD prevented the diazoxide-induced decrease in rhod-2 fluorescence (Fig. 11B). With 5-HD (300 μm) alone, rhod-2 fluorescence was 166 ± 6 a.u., a value not significantly different from the 162 ± 7 a.u. observed in the presence of 5-HD and 300 μm diazoxide. Treatment with 5-HD, however, did not prevent FCCP from decreasing mitochondrial Ca2+ levels (Fig. 11B).

Figure 11
Diazoxide-induced decrease in mitochondrial Ca2+ load in intact cardiomyocytes: prevention by the mitoKATP channel blocker 5-HD


The present study provides evidence at the level of isolated mitochondria, as well as in intact cardiomyocytes, that potassium channel openers can reduce mitochondrial Ca2+ load. This appears to be a consequence of potassium channel opener-induced mitochondrial depolarization leading to a reduction in the driving force responsible for Ca2+ uptake, and/or activation of a cyclosporin-sensitive pathway for Ca2+ release. The action of the potassium channel openers was K+ dependent and sensitive to inhibitors of mitoKATP channels. In this regard, modulators of mitoKATP channels may contribute to the regulation of cellular Ca2+ homeostasis through the maintenance of mitochondrial Ca2+ content.

Previously, it has been established that potassium channel openers target mitoKATP channels, promoting K+ influx, mitochondrial membrane depolarization, activation of respiration and oxidation of mitochondrial respiratory chain components (Beavis et al. 1993; Belyaeva et al. 1993; Czyz et al. 1995; Garlid et al. 1996; Szewczyk, 1997; Grimmsmann & Rustenbeck, 1998; Holmuhamedov et al. 1998; Liu et al. 1998). The present findings extend our understanding of the action of potassium channel openers on cardiac mitochondria by demonstrating the ability of these channel modulators to regulate mitochondrial Ca2+ transport.

In isolated mitochondria, parameters of mitochondrial Ca2+ metabolism were determined from changes in extramitochondrial Ca2+ concentration monitored by Ca2+-selective electrodes. A high selectivity for Ca2+ over other cations, and a Ca2+ sensitivity within the micromolar concentration range, has established this potentiometric approach as a reliable tool with which to study mitochondrial Ca2+ transport (Ferrari et al. 1993; Ichas et al. 1994; Holmuhamedov et al. 1998). Using this approach, we observed that diazoxide and pinacidil reduce mitochondrial Ca2+ uptake. It is known that mitochondrial Ca2+ uptake occurs through the Ca2+ uniporter, and is driven by the difference in membrane potential across the inner mitochondrial membrane (Gunter et al. 1994). The rate of Ca2+ uptake can be modulated either by direct inhibition of the uniporter, for example with Ruthenium Red, or through the dissipation of mitochondrial membrane potential with mitochondrial uncouplers (Nicholls & Akerman, 1982). Thus, in principle, potassium channel openers could have acted through one or both of these mechanisms and interfered with mitochondrial Ca2+ accumulation. Although we cannot exclude a direct Ruthenium Red-like action of diazoxide and pinacidil, three lines of evidence favour the latter mechanism. First, it is rather unlikely that such chemically distinct molecules as the hexavalent dye Ruthenium Red, the sulfonamide diazoxide and the cyanoguanidine pinacidil could share the ability to bind to the mitochondrial Ca2+ uniporter. Second, both potassium channel openers induce mitochondrial membrane depolarization, which is not observed with Ruthenium Red. Third, the potassium channel opener-mediated reduction in mitochondrial Ca2+ uptake and mitochondrial depolarization required the presence of extramitochondrial K+. Thus, potassium channel opener-induced membrane depolarization would be expected to be the main cause for reduced Ca2+ uptake, which, in turn, could translate into decreased mitochondrial Ca2+ load.

In addition, we here observed that potassium channel openers could promote Ca2+ discharge from pre-loaded mitochondria (see also Holmuhamedov et al. 1998). The primary mitochondrial Ca2+ release pathway includes the mitochondrial permeability transition pore, which can be activated by mitochondrial membrane depolarization and swelling (Gunter et al. 1994; Zoratti & Szabo, 1995; Bernardi & Petronilli, 1996). Activation of mitoKATP channels has been associated with both mitochondrial membrane depolarization and swelling (Garlid, 1996; Holmuhamedov et al. 1998). Thereby, opening of mitoKATP channels may activate the permeability transition pore, and thus contribute to opener-induced Ca2+ discharge from pre-loaded mitochondria. Such a possibility is supported by the observation that cyclosporin A, a recognized inhibitor of the mitochondrial permeability transition pore (Zoratti & Szabo, 1995), prevented potassium channel opener-mediated mitochondrial Ca2+ efflux. This effect of cyclosporin A cannot be attributed to direct inhibition of mitoKATP channels per se, since this agent did not prevent opener-induced mitochondrial depolarization nor did it preclude opener-mediated inhibition of Ca2+ uptake. Thus, potassium channel opener-induced K+ influx may also contribute to the reduction of mitochondrial Ca2+ load through activation of a cyclosporin A-sensitive Ca2+ release pathway, such as the permeability transition pore.

The precise site of action of diazoxide and pinacidil in mitochondria remains unknown. Previously, the action of potassium channel openers has been associated with modulation of a mitoKATP channel, recognized as a nucleotide-sensitive ion conductance expressed in the mitochondrial inner membrane (Inoue et al. 1991; Paucek et al. 1992; Garlid et al. 1996; Szewczyk, 1997). Although the molecular composition of this channel remains elusive, it may consist of an inwardly rectifying K+ channel subunit responsible for K+ permeance (Suzuki et al. 1997). That the effects of diazoxide and pinacidil were dependent on extramitochondrial K+ supports a role for a K+ conductance. Moreover, valinomycin, a K+ ionophore, mimicked the effect of diazoxide and pinacidil, further suggesting that promotion of K+ flux across the mitochondrial membrane may be implicated in the action of potassium channel openers. Also, the apparent IC50 describing the action of both potassium channel openers on Ca2+ in isolated mitochondria was within the range previously reported for activation of mitoKATP channels suggesting that, under our experimental conditions, the openers could bind to channel proteins (Liu et al. 1998). Finally, known inhibitors of mitoKATP channels, ATP and ADP, but not non-blocking nucleotides (Garlid, 1996), were effective in preventing the action of a potassium channel opener on mitochondrial Ca2+. The effect of added ATP and ADP was observed in the presence of atractyloside which prevents nucleotide translocation into the mitochondrial matrix (Zoratti & Szabo, 1995; Fiore et al. 1998), in accordance with a cytosolic site of action of these nucleotides on mitoKATP channels (Beavis et al. 1993; Garlid, 1996). Under our experimental conditions, known pharmacological blockers of plasmalemmal KATP channels, glyburide and 5-HD, demonstrated variable effects on mitochondrial function, and were not reliable as specific inhibitors of potassium channel openers (data not shown). This is consistent with the observation that the efficacy of such blockers is defined by the operative condition of the KATP channels (Alekseev et al. 1998; Brady et al. 1998; Jaburek et al. 1998), and their failure to inhibit mitochondrial K+ flux has been previously described (Beavis et al. 1993; Szewczyk, 1997; Grimmsmann & Rustenbeck, 1998).

Potassium channel opener-induced changes in mitochondrial membrane potential and Ca2+ were also observed in intact cardiomyocytes, indicating that the action of openers is preserved in mitochondria within their natural, cytosolic, environment. This is of significance since mitochondria are an integrated component in cellular Ca2+-regulating networks (Rizzuto et al. 1993, 1998; Jouaville et al. 1995), and have been implicated in the regulation of cytosolic Ca2+ in cardiomyocytes (Bassani et al. 1993; Bowser et al. 1998; Duchen et al. 1998). In intact cardiac cells, the mitoKATP channel blocker 5-HD (Sato et al. 1998) was a reliable inhibitor of both mitochondrial membrane depolarization and the reduction in mitochondrial Ca2+ level induced by diazoxide. The consistent efficacy of 5-HD supports the notion that the inhibitory action of this channel blocker depends on cytosolic factors governing the state of mitoKATP channels (Jaburek et al. 1998).

In summary, this study uncovers the ability of potassium channel openers to modulate one of the most fundamental functions of cardiac mitochondria, Ca2+ homeostasis, both in vitro and in vivo. Our data provide direct support to the recently proposed hypothesis that dissipation of mitochondrial membrane potential by potassium channel openers can lead to reduced mitochondrial Ca2+ influx, and thereby protect against ischaemic injury (Liu et al. 1998). That modulation of mitoKATP channels is involved in the regulation of mitochondrial Ca2+ content provides a rationale to further evaluate the cardioprotective potential of mitochondrial potassium channel openers under conditions that predispose mitochondria to Ca2+ overload and dysfunction.


This work was supported by NIH (HL-07111), American Heart Association, Miami Heart Research Institute, and the Bruce and Ruth Rappaport Program in Vascular Biology and Gene Delivery.


  • Alekseev AE, Brady PA, Terzic A. Ligand-insensitive state of cardiac ATP-sensitive K+ channels: Basis for channel opening. Journal of General Physiology. 1998;111:381–394. [PMC free article] [PubMed]
  • Bassani JW, Bassani RA, Bers DM. Ca2+ cycling between sarcoplasmic reticulum and mitochondria in rabbit cardiac myocytes. The Journal of Physiology. 1993;460:603–621. [PMC free article] [PubMed]
  • Beavis AD, Lu Y, Garlid KD. On the regulation of K+ uniport in intact mitochondria by adenine nucleotides and nucleotide analogs. Journal of Biological Chemistry. 1993;268:997–1004. [PubMed]
  • Belyaeva E, Szewczyk A, Mikolaek B, Nalecz M, Wojtczak L. Demonstration of glibenclamide-sensitive K+ fluxes in rat liver mitochondria. Biochemistry and Molecular Biology International. 1993;31:493–500. [PubMed]
  • Bernardi P, Petronilli V. The permeability transition pore as a mitochondrial Ca2+ release channel: A critical appraisal. Journal of Bioenergetics and Biomembranes. 1996;28:131–138. [PubMed]
  • Bowser DN, Minamikawa T, Nagley P, Williams D. Role of mitochondria in calcium regulation of spontaneously contracting cardiac muscle cells. Biophysical Journal. 1998;75:2004–2014. [PMC free article] [PubMed]
  • Brady PA, Alekseev AE, Terzic A. Operative condition-dependent response of cardiac ATP-sensitive K+ channels toward sulfonylureas. Circulation Research. 1998;82:272–278. [PubMed]
  • Crompton M, Ellinger H, Costi A. Inhibition by cyclosporin A of a Ca2+-dependent pore in heart mitochondria activated by inorganic phosphate and oxidative stress. Biochemical Journal. 1988;255:357–360. [PMC free article] [PubMed]
  • Czyz A, Szewczyk A, Nalecz MJ, Wojtczak L. The role of mitochondrial potassium fluxes in controlling the protonmotive force in energized mitochondria. Biochemical and Biophysical Research Communications. 1995;210:98–104. [PubMed]
  • Delcamp TJ, Dales C, Ralenkotter L, Cole PS, Hadley RW. Intramitochondrial [Ca2+] and membrane potential in ventricular myocytes exposed to anoxia-reoxygenation. American Journal of Physiology. 1998;275:H484–494. [PubMed]
  • Duchen MR, Leyssens A, Crompton M. Transient mitochondrial depolarizations reflect focal sarcoplasmic reticular calcium release in single rat cardiomyocytes. Journal of Cell Biology. 1998;142:975–988. [PMC free article] [PubMed]
  • Edwards G, Weston AH. Recent advances in potassium channel modulation. Progress in Drug Research. 1997;49:93–121. [PubMed]
  • Ehrenberg B, Montana V, Wei MD, Wuskel JP, Loew ML. Membrane potential can be determined in individual cells from the Nernstian distribution of cationic dyes. Biophysical Journal. 1988;53:785–794. [PMC free article] [PubMed]
  • Ferrari R, Pedersini P, Bongrazio M, Gaia G, Bernocchi P, Di Lisa F, Visioli O. Mitochondrial energy production and cation control in myocardial ischemia and reperfusion. Basic Research in Cardiology. 1993;88:495–512. [PubMed]
  • Fiore C, Trezeguet V, Le Saux A, Roux P, Schwimmer C, Dianoux AC, Noel F, Lauquin GJ, Brandolin G, Vignais PV. The mitochondrial ADP/ATP carrier: structural, physiological and pathological aspects. Biochimie. 1998;80:137–150. [PubMed]
  • Garlid KD. Cation transport in mitochondria: The potassium cycle. Biochimica et Biophysica Acta. 1996;1275:123–126. [PubMed]
  • Garlid KD, Paucek P, Yarov-Yarovoy V, Murray H, Darbenzio R, D'Alonzo A, Lodge N, Smith M, Grover G. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels. Circulation Research. 1997;81:1072–1082. [PubMed]
  • Garlid KD, Paucek P, Yarov-Yarovoy V, Sun X, Schindler PA. The mitochondrial KATP channel as a receptor for potassium channel openers. Journal of Biological Chemistry. 1996;271:8796–8799. [PubMed]
  • Grimmsmann T, Rustenbeck I. Direct effects of diazoxide on mitochondria in pancreatic B-cells and on isolated liver mitochondria. British Journal of Pharmacology. 1998;123:781–788. [PMC free article] [PubMed]
  • Gross GJ. ATP-sensitive potassium channels and myocardial preconditioning. Basic Research in Cardiology. 1995;90:85–88. [PubMed]
  • Grover GJ. Pharmacology of ATP-sensitive potassium channel openers in models of myocardial ischemia and reperfusion. Canadian The Journal of Physiology and Pharmacology. 1997;75:309–315. [PubMed]
  • Gunter TE, Gunter KK, Sheu SS, Gavin CE. Mitochondrial calcium transport: physiological and pathological relevance. American Journal of Physiology. 1994;36:C313–339. [PubMed]
  • Holmuhamedov EL. Oscillating dissipative structures in mitochondrial suspension. European Journal of Biochemistry. 1986;158:543–546. [PubMed]
  • Holmuhamedov EL, Jovanovic S, Dzeja P, Jovanovic A, Terzic A. Mitochondrial ATP-sensitive K+ channels modulate cardiac mitochondrial function. American Journal of Physiology. 1998;44:H1567–1576. [PubMed]
  • Ichas F, Jouaville LS, Sidash SS, Mazat JP, Holmuhamedov EL. Mitochondrial calcium spiking: A transduction mechanism based on calcium-induced permeability transition involved in cell calcium signaling. FEBS Letters. 1994;348:211–215. [PubMed]
  • Inoue I, Nagase H, Kishi K, Higuti T. ATP-sensitive K+ channel in the mitochondrial inner membrane. Nature. 1991;352:244–247. [PubMed]
  • Jaburek M, Yarov-Yarovoy V, Paucek P, Garlid KD. State-dependent inhibition of the mitochondrial KATP channel by glyburide and 5-hydroxydecanoic acid. Journal of Biological Chemistry. 1998;273:13578–13582. [PubMed]
  • Jouaville LS, Ichas F, Holmuhamedov EL, Camacho P, Lechleiter JD. Synchronization of calcium waves by mitochondrial substrates in Xenopus laevis oocytes. Nature. 1995;377:438–441. [PubMed]
  • Kamo N, Muratsugu M, Hongoh R, Kobatake Y. Membrane potential of mitochondria measured with an electrode sensitive to tetraphenylphosphonium and relationship between proton electrochemical potential and phosphorylation potential in steady state. Journal of Membrane Biology. 1979;49:105–121. [PubMed]
  • Kinnally KW, Tedeschi H. Adenosine triphosphate synthesis coupled to K+ influx in mitochondria. Science. 1982;216:742–744. [PubMed]
  • Kloner RA, Bolli R, Marban E, Reinlib L, Braunwald E, Reinlib LJ, Bonow RO, Borgers M, Downey J, Feigenbaum H, Ganote C, Gross G, Jennings R, Levitsky S, McCully JD, Mentzer R, Miller WP, Przyklenk K, Rahimtoola S, Vatner S, Yellon D. Medical and cellular implications of stunning, hibernation, and preconditioning. Circulation. 1998;97:1848–1867.
  • Liu Y, Sato T, O'Rourke B, Marban E. Mitochondrial ATP-dependent potassium channels. Novel effectors of cardioprotection? Circulation. 1998;97:2463–2469. [PubMed]
  • Miyamae M, Camacho A, Weiner M, Figueredo V. Attenuation of postischemic reperfusion injury related to prevention of [Ca2+]m overload in rat hearts. American Journal of Physiology. 1996;40:H2145–2153. [PubMed]
  • Nicholls D, Akerman K. Mitochondrial calcium transport. Biochimica et Biophysica Acta. 1982;683:57–88. [PubMed]
  • Park Y, Bowles DK, Kehrer JP. Protection against hypoxic injury in isolated-perfused rat heart by ruthenium red. Journal of Pharmacology and Experimental Therapeutics. 1990;253:628–635. [PubMed]
  • Paucek P, Mironova G, Mahdi F, Beavis AD, Woldegiorgis G, Garlid KD. Reconstitution and partial purification of the glibenclamide-sensitive ATP-dependent K+ channel from rat liver and beef heart mitochondria. Journal of Biological Chemistry. 1992;267:26062–26069. [PubMed]
  • Perez-Terzic C, Gacy AM, Bortolon R, Dzeja PP, Puceat M, Jaconi M, Prendergast FG, Terzic A. Structural plasticity of the cardiac nuclear pore complex in response to regulators of nuclear import. Circulation Research. 1999;84:1292–1301. [PubMed]
  • Petronilli V, Pietrobon D, Zoratti M, Azzone GF. Free energy coupling between H+-generating and H+-consuming pumps. European Journal of Biochemistry. 1986;155:423–431. [PubMed]
  • Quast U. Potassium channel openers: Pharmacological and clinical aspects. Fundamental and Clinical Pharmacology. 1992;6:279–293. [PubMed]
  • Rizzuto R, Brini M, Murgia M, Pozzan T. Microdomains with high Ca2+ close to IP3-sensitive channels that are sensed by neighboring mitochondria. Science. 1993;262:744–747. [PubMed]
  • Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T. Close contacts with the endoplasmic reticulum as determinants of mitochondrial Ca2+ responses. Science. 1998;280:1763–1766. [PubMed]
  • Sato T, O'Rourke B, Marban E. Modulation of mitochondrial ATP-dependent K+ channels by protein kinase C. Circulation Research. 1998;83:110–114. [PubMed]
  • Szewczyk A. Intracellular targets for antidiabetic sulfonylureas and potassium channel openers. Biochemical Pharmacology. 1997;54:961–965. [PubMed]
  • Suzuki M, Kotake K, Fujikura K, Inagaki N, Suzuki T, Gonoi T, Seino S, Takata K. Kir6.1: A possible subunit of ATP-sensitive K+ channels in mitochondria. Biochemical and Biophysical Research Communications. 1997;241:693–697. [PubMed]
  • Terzic A, Jahangir A, Kurachi Y. Cardiac ATP-sensitive K+ channels: regulation by intracellular nucleotides and K+ channel-opening drugs. American Journal of Physiology. 1995;269:C525–545. [PubMed]
  • Trollinger DR, Cascio WE, Lemasters JJ. Selective loading of Rhod2 into mitochondria shows mitochondrial Ca2+ transients during the contractile cycle in adult rabbit cardiac myocytes. Biochemical and Biophysical Research Communications. 1997;236:738–742. [PubMed]
  • Zoratti M, Szabo I. The mitochondrial permeability transition. Biochimica et Biophysica Acta. 1995;1241:139–176. [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...