INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE POTASSIUM ION is the major cytoplasmic and mitochondrial cation, and net flux of K⁺ across the inner mitochondrial membrane critically regulates mitochondrial activity (16). This includes regulation of energy production and maintenance of cellular Ca²⁺ homeostasis, two mitochondrial functions essential for cellular survival (12, 15, 23, 24). To allow for bidirectional K⁺ cycling across an otherwise K⁺-impermeable inner membrane, mitochondria express specialized K⁺ conduits (16). These include the well-characterized electroneutral K⁺/H⁺ antiporter responsible for K⁺ efflux, along with less known pathways for K⁺ influx (8, 15).

A candidate mechanism for K⁺ entry is an ATP-sensitive K⁺ (K_ATP) channel (15, 16). This channel, known as the mitochondrial K_ATP channel, has been recently identified within the inner mitochondrial membrane (29), and the molecular identity of channel subunits has been partially characterized (16, 59, 60). Purified channel proteins, when incorporated into artificial bilayers or liposomes, reconstitute an ATP-sensitive K⁺ conductance with properties similar to those previously described for other members of the K_ATP channel family (52). This channel family regroups nucleotide-gated inwardly rectifying K⁺ channels, which sense the metabolic state of a cell and regulate K⁺ flux and membrane excitability (1, 2, 31, 42, 46, 59, 67).

K_ATP channels have been originally described in the cardiac sarcolemma (50), and opening of sarcolemmal K_ATP channels has been associated with shortening of cardiac action potential, decrease in intracellular Ca²⁺ loading, and cardioprotection during ischemia (5, 14, 20, 21, 32, 33, 44). However, the consequences of activation of mitochondrial K_ATP channels on mitochondrial functions, including oxidative phosphorylation and Ca²⁺ transport, have not been determined.

K⁺ channel-opening drugs are the established tool used to activate K_ATP channels (21, 54, 66) and also target the mitochondrial K_ATP channels (17, 18, 61, 62). Therefore, in this study we used K⁺ channel openers to examine the effect of mitochondrial K_ATP channel opening on mitochondrial functions, including membrane potential, respiration, ATP generation, Ca²⁺ transport, and membrane integrity. We present evidence that activation of mitochondrial K_ATP channels modulates the function of isolated cardiac mitochondria.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Isolated cardiac mitochondria. Cardiac mitochondria were isolated from ether- or pentobarbital-anesthetized rats by differential centrifugation (25, 43). After thoracotomy, hearts were rapidly excised into an ice-cold isolation buffer (50 mM sucrose, 200 mM mannitol, 5 mM KH₂PO₄, 1 mM EGTA, 5 mM MOPS, and 0.1% BSA, pH 7.15 adjusted with KOH), atria were removed, and 1-mm³ pieces of ventricular myocardium were homogenized (30 ml of isolation buffer per heart) using a PT10/35 Polytron (Brinkman). Three 20-s homogenization cycles were performed on ice, and then the samples were centrifuged for 10 min (at 750 g) using a Sorvall II centrifuge equipped with a GSA rotor. The supernatant, containing the mitochondrial fraction, was further centrifuged at 7,000 g for 20 min, and the pellet was resuspended in 30 ml of isolation buffer (with no EGTA) and spun at 7,000 g for 20 min. Finally, mitochondria were resuspended in the isolation buffer (with no EGTA), and protein concentration was determined using a protein kit (Bio-Rad). Mitochondrial suspension (~30-40 mg protein/ml) was kept on ice before experiments.

Mitochondrial membrane potential. Measurements were made at 30°C under continuous stirring of the mitochondrial suspension (1 mg protein/ml incubation medium) placed into a multichannel chamber of an ESON-6ch computerized analyzer (25-28). The standard incubation medium contained (in mM) 110 KCl, 5 K₂HPO₄, 5 succinate, 5 pyruvate, and 10 MOPS (pH 7.15). Mitochondrial membrane potential was measured with tetraphenylphosphonium (TPP⁺)-sensitive minielectrodes manufactured and calibrated as described elsewhere (38). TPP⁺ (200 nM) was added to the incubation medium in the chamber before addition of mitochondria, and mitochondrial membrane potential was calculated according to the following equation:  = 59 × log (v/V)  59 × log [10^{(E  E₀)/59}  1], where is mitochondrial membrane potential (mV), v is mitochondrial matrix volume (1.6 µl/mg mitochondrial protein) (53), V is volume of incubation medium (1 ml), and E₀ and E are electrode potentials before and after addition of mitochondria, respectively (49). Membrane potential was measured in the presence of K⁺ channel openers [pinacidil (RBI), cromakalim (Sigma Chemical), and levcromakalim (a gift from SmithKline Beecham)], K⁺ ionophore (valinomycin), and/or K⁺ channel blockers [glyburide and 5-hydroxydecanoic acid (5-HD), both from RBI]. In addition, mitochondrial membrane potential was also measured after addition of ADP, in the absence and presence of a K⁺ channel opener.

Mitochondrial respiration. O₂ consumption of mitochondria was measured in the multichannel chamber using a calibrated Clark-type O₂ minielectrode (11, 25). For calibration purposes, depletion of O₂ content within the mitochondrial suspension was achieved by sodium hydrosulfite. Data acquisition and processing were the same as with the TPP⁺ electrode. The effects of drugs were assessed on "state 2" mitochondrial respiration (7). In addition, in certain experiments, ADP-stimulated mitochondrial respiration ("state 3") was measured in the absence and presence of a K⁺ channel opener.

Mitochondrial Ca²⁺ uptake. Mitochondrial Ca²⁺ uptake was measured as a change in the free Ca²⁺ concentration within the suspension using calibrated Ca²⁺-selective minielectrodes (Microelectrodes) (11, 26, 28). Data acquisition and processing were the same as with the TPP⁺ electrode. The effect of K⁺ channel openers on mitochondrial Ca²⁺ retention and release was studied in mitochondria preloaded with Ca²⁺ (3 consecutive 40-nmol Ca²⁺ pulses/mg mitochondrial protein).

Mitochondrial ATP synthesis. Mitochondrial ATP content was determined in mitochondrial extracts (9). Briefly, 200 µl of mitochondrial suspension were treated with 20 µl of 3.3 M HClO₄, and precipitated proteins were removed by centrifugation (60 s, 14,000 rpm, 4°C). After neutralization of the supernatant with 80 µl of a mixture containing 2.5 M K₂CO₃ in 1 M HEPES, the precipitate was separated by centrifugation (60 s, 14,000 rpm, 4°C), and the concentration of ATP within the extract was determined by coupled enzymatic analysis based on spectrophotometric detection of NADH (40) in the absence and presence of a K⁺ channel opener.

Mitochondrial volume and integrity. Light scattering of mitochondrial suspensions, a measure of matrix volume (18, 24, 25), was determined within a 10-mm cuvette (maintained at 22°C) in the presence of K⁺ channel openers or a K⁺ ionophore. The absorbance of the mitochondrial suspension was measured at 540 nm using a spectrophotometer (model DU-7400, Beckman).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect on mitochondrial membrane potential. Isolated cardiac mitochondria had a membrane potential of 180 ± 15 mV (n = 12), as demonstrated using a potential-sensitive probe, TPP⁺ (Fig. 1A). Mitochondria maintained a steady membrane potential throughout the duration of recordings (up to 40 min). Addition of the K⁺ channel opener pinacidil (100 µM) induced rapid depolarization (Fig. 1A). On average, mitochondrial membrane potential decreased by 10 ± 7 mV (n = 5) in the presence of 100 µM pinacidil. The effect of pinacidil was concentration dependent (Fig. 1B). The K⁺ ionophore valinomycin (12.5 ng/mg mitochondrial protein) also reduced mitochondrial membrane potential, indicating that depolarization can be triggered by promoting K⁺ flux across the mitochondrial membrane (Fig. 1A).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Effect of pinacidil (PIN) on mitochondrial membrane potential. Mitochondrial membrane potential was measured using the membrane potential-sensitive probe tetraphenylphosphonium (TPP+). Arrowhead, addition of isolated cardiac mitochondria to a solution containing TPP+. On addition of mitochondrial suspension, resting membrane potential was rapidly achieved. A: pinacidil (100 µM)-induced membrane depolarization. K+ ionophore valinomycin (Valino, 12.5 ng/mg mitochondrial protein) was added as a positive control. B: concentration-response curve for effect of pinacidil on mitochondrial membrane potential. Values are means ± SE of 3-5 measurements.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Cromakalim (CROM)-induced mitochondrial membrane depolarization and extramitochondrial concentration of K+. A: cromakalim (20 µM)-induced membrane depolarization. Valinomycin (12.5 ng/mg mitochondrial protein) was added as a positive control. B: in nominally K+-free solution (i.e., KCl replaced with 110 mM choline chloride, K2HPO4 with 5 mM Na2HPO4, and pH adjusted with Trizma base), cromakalim (20 µM) was without effect. Addition of KCl (in 4 mM steps) restored the depolarizing effect of cromakalim. C: in deenergized mitochondria treated with antimycin A (AntiA, 1 µg/mg mitochondrial protein) and oligomycin (Oligo, 1 µg/mg mitochondrial protein), cromakalim (in 25 µM steps) induced mitochondrial membrane hyperpolarization in nominally K+-free solution. Arrowheads, addition of isolated cardiac mitochondria to a solution containing TPP+. Mitochondrial membrane potential was measured as described in Fig. 1 legend.

View larger version (7K): [in this window] [in a new window]   Fig. 3.   Levcromakalim (LEV)-induced mitochondrial membrane depolarization and a blocker of ATP-sensitive K+ (KATP) channels. A: levcromakalim (25 µM)-induced membrane depolarization. Valinomycin (12.5 ng/mg mitochondrial protein) was added as a positive control. B: in mitochondria treated with KATP channel blocker 5-hydroxydecanoic acid (5-HD, 20 µM), levcromakalim (20 µM) did not induce membrane depolarization. Arrowheads, addition of isolated cardiac mitochondria to a solution containing TPP+. Mitochondrial membrane potential was measured as described in Fig. 1 legend.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Effect of pinacidil on ADP-induced mitochondrial membrane depolarization. A: ADP (500 µM)-induced mitochondrial membrane depolarization in absence (Control) and presence of pinacidil (100 µM). B: concentration-response curve for effect of pinacidil on duration of ADP (500 µM)-induced membrane depolarization. Values are means ± SE.

Effect on mitochondrial respiration. The rate of state 2 respiration of isolated cardiac mitochondria was 14.0 ± 1.3 nmol O₂ · min¹ · mg protein¹ (n = 20). Pinacidil (100 µM-1 mM), in a concentration-dependent manner, induced an increase in the rate of mitochondrial respiration (Fig. 5A). Pretreatment (3-5 min) of mitochondria with the K_ATP channel blocker glyburide (1 µM) inhibited the effect of pinacidil on mitochondrial respiration (Fig. 5B; n = 3).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Effect of pinacidil on mitochondrial O2 consumption. A: concentration-response curve for effect of pinacidil on O2 consumption. Values are means ± SE. B: glyburide (1 µM) inhibits 100 µM pinacidil-induced increase in mitochondrial O2 consumption. Values are means ± SE. * Significant difference (P < 0.01, Student's t-test) between absence and presence of glyburide.

1

1

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Effect of pinacidil on ADP-induced acceleration of mitochondrial respiration. Mitochondrial rate of respiration was measured using an O2-sensing electrode in pinacidil-untreated (A) and pinacidil (100 µM)-treated (B) mitochondria under control conditions (V2 vs. V'2) and in presence of ADP (V3 and V4 vs. V'3 and V'4). t, Time mitochondria spent in ADP-induced state 3 respiration; O2, magnitude of ADP-induced O2 consumption. Mitochondrial uncoupler dinitrophenol (DNP, 40 µM) was added at end of each experiment to verify mitochondrial responsiveness. Arrowheads, addition of isolated cardiac mitochondria.

Effect on mitochondrial ATP synthesis. Isolated cardiac mitochondria efficiently phosphorylated ADP (500 µM) so that ~90% of added ADP was converted to ATP within 100 ± 15 s (n = 3; Fig. 7). The initial rate of ATP synthesis was 401 ± 35 nmol ATP · min¹ · mg protein¹, and the half-time required for complete conversion of added ADP to ATP was 27 ± 5 s (n = 3; Fig. 7). Pinacidil (100 µM) decreased the rate of initial ATP synthesis to 187 ± 23 nmol ATP · min¹ · mg protein¹ and increased the half-time required for conversion of ADP to ATP to 87 ± 7 s (n = 3; Fig. 7). Although the rate of mitochondrial synthesis was slowed, pinacidil (100 µM) did not prevent conversion of ADP to ATP (Fig. 7).

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Effect of pinacidil on mitochondrial conversion of ADP to ATP. Time course of ATP synthesis after addition of 500 µM ADP was measured in absence and presence of 100 µM pinacidil. Values are means ± SE. [ATP], ATP concentration.

Effect on mitochondrial Ca²⁺ loading. Mitochondria are known to accumulate and to retain Ca²⁺ within the matrix (22). Isolated cardiac mitochondria efficiently took up and retained added Ca²⁺ when repeatedly challenged with extramitochondrial Ca²⁺ pulses (in total 120 nmol Ca²⁺/mg mitochondrial protein; Fig. 8). Addition of pinacidil (250 µM) induced rapid release of Ca²⁺ from preloaded mitochondria (Fig. 8, A-C). The efficacy of pinacidil to induce release of mitochondrial Ca²⁺ was dependent on the extramitochondrial concentration of K⁺ (Fig. 8, A-C). At 1, 10, and 110 mM extramitochondrial KCl, the magnitude of pinacidil-induced Ca²⁺ release was 1, 3, and 5 nmol Ca²⁺/mg mitochondrial protein, respectively (Fig. 8, A-C). The Ca²⁺-releasing effect of pinacidil (250 µM) was suppressed by pretreatment of mitochondria with the K_ATP channel blocker glyburide (1 µM; Fig. 8D).

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Effect of pinacidil on Ca2+ release from mitochondria. Original records of extramitochondrial Ca2+ concentration measured using a Ca2+-sensitive electrode are shown. Arrowheads, addition of mitochondrial suspension. Ca2+ was added as a series of 3 pulses (40 nmol each) to load mitochondria. A-C: pinacidil (100 µM)-induced release of Ca2+ from mitochondria. Magnitude of pinacidil-induced Ca2+ release was dependent on concentration of extramitochondrial K+ (i.e., 1, 10, and 110 mM). Treatment with glyburide (GLYB, 1 µM) suppressed pinacidil (100 µM) to induce release of mitochondrial Ca2+. In D, valinomycin (12.5 ng/mg mitochondrial protein) was added as a positive control.

Effect on mitochondrial matrix volume and integrity. The volume of mitochondria depends on the permeability of the inner membrane to osmotically active particles, such as K⁺ (6). The K⁺ channel opener cromakalim (100 µM) induced swelling of mitochondria, as revealed by a decrease in the absorbance of the mitochondrial suspension in K⁺-containing media by 0.1 ± 0.02 optical density unit (n = 3; Fig. 9A). Under this condition, the K⁺ ionophore valinomycin also induced significant mitochondrial swelling (not shown). In contrast, in K⁺-free medium, cromakalim had no significant effect on mitochondrial swelling (Fig. 9; n = 3). K⁺-dependent changes in mitochondrial volume were also observed with pinacidil (250 µM; Fig. 9B).

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Effect of K+ channel openers on mitochondrial matrix volume. A: original records of mitochondrial volume expressed as optical density (OD) of mitochondrial suspension in K+-free (sucrose: KCl was isotonically replaced with 220 mM sucrose and K2HPO4 with 5 mM Na2HPO4, and pH was adjusted with Trizma base) or K+-containing standard incubation medium (KCl). Arrows, addition of cromakalim (25 µM). B: average effect of K+ channel openers, 25 µM cromakalim and 100 µM pinacidil, on mitochondrial volume. Values are means ± SE. * Significantly different (P < 0.01, Student's t-test) from sucrose.

View larger version (42K): [in this window] [in a new window]   Fig. 10.   Effect of K+ channel openers on cytochrome c release from mitochondria. Dot blots of extramitochondrial medium were obtained using a monoclonal anti-cytochrome c antibody (Ab) under control conditions and after treatment of mitochondria with K+ channel openers, pinacidil (250 µM) and cromakalim (100 µM). Valinomycin (56 ng/mg mitochondrial protein) was used as a positive control.

View larger version (36K): [in this window] [in a new window]   Fig. 11.   Effect of K+ channel openers on adenylate kinase release from mitochondria. Adenylate kinase activity was measured under control conditions and after treatment of mitochondria with K+ channel openers, pinacidil (250 µM) and cromakalim (100 µM). Valinomycin (56 ng/mg of mitochondrial protein) was used as a positive control.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In the present study we report that, in mitochondria isolated from cardiac muscle, K⁺ channel openers induced membrane depolarization, accelerated respiration, slowed ATP production, triggered release of Ca²⁺, and produced swelling associated with efflux of intermembrane proteins. In accord with an activation of mitochondrial K_ATP channels, these effects were dependent on the extramitochondrial concentration of K⁺, were inhibited by blockers of K_ATP channels, and could be induced by structurally distinct K⁺ channel openers. These observations provide direct evidence for a role of mitochondrial K_ATP channels in the regulation of mitochondrial functions.

That K_ATP channels are present within the inner membrane of mitochondria was initially demonstrated on the basis of electrophysiological measurements of a K⁺-selective, ATP-gated ion conductance in fused giant mitoplasts (29). The presence of functional K_ATP channels in mitochondria was confirmed after isolation of an inner membrane protein fraction from rat liver and beef heart mitochondria, which, when reconstituted in liposomes or planar lipid bilayers, catalyzed electrophoretic ATP-sensitive K⁺ flux (52). This ATP-inhibited K⁺ flux was activated by K⁺ channel openers such as cromakalim (18) and suppressed by K⁺ channel blockers such as glyburide and 5-HD (3, 17, 51, 63). K⁺ influx through this channel was hypothesized to serve as a mechanism for osmotic regulation of mitochondrial volume (15, 23, 29, 62). In mitochondria from noncardiac tissue, such as liver and pancreatic -cells, openers of K_ATP channels were found to affect mitochondrial energy metabolism (16, 19, 63). Thus the present findings that opening of mitochondrial K_ATP channels can modulate multiple functions in isolated cardiac mitochondria provide direct evidence for a K_ATP channel-mediated regulation of cardiac mitochondrial behavior.

Pinacidil, cromakalim, and levcromakalim share the property to activate K_ATP channels (21, 54, 66) and were found here to depolarize the membrane potential of cardiac mitochondria. However, it has remained controversial whether opening of mitochondrial K_ATP channels can be associated with changes in mitochondrial membrane potential (19, 61). In pancreatic -cells, glyburide, a K_ATP channel blocker, failed to antagonize the depolarizing effect of diazoxide, a drug with multiple sites of action, including K⁺ channel activation, suggesting that diazoxide-induced membrane depolarization is not associated with activation of K_ATP channels, at least in this cell type (19). Moreover, in isolated liver mitochondria, other K⁺ channel openers, including aprikalim and nicorandil, were without effect on membrane potential, although in the same preparation, RP-66471, a more recently developed K_ATP channel opener, induced mitochondrial depolarization (61).

Here, in cardiac mitochondria, a K_ATP channel-mediated regulation of mitochondrial membrane potential is supported by three lines of evidence. First, structurally distinct K⁺ channel openers were effective in depolarizing mitochondria. Second, the effect of K⁺ channel openers on mitochondrial membrane potential was dependent on the electrochemical gradient for K⁺. Third, 5-HD, a selective K_ATP channel blocker (51), did not affect membrane potential by itself but prevented the depolarizing action of a K⁺ channel opener. Our findings are in accord with the property of cromakalim and other K⁺ channel openers to induce K⁺ flux through mitochondrial K_ATP channels (18). Under the present experimental condition (110 mM extramitochondrial K⁺ and  = 180 mV), opening of K_ATP channels is expected to lead to K⁺ influx and depolarization of mitochondria. In turn, a decrease in membrane potential would dissipate the driving force for ATP synthesis, triggering a compensatory activation of the mitochondrial respiratory chain. It has previously been shown that increase in the K⁺ permeability by valinomycin of the inner membrane decreases the rate of ATP synthesis and promotes respiration (53). The present study extends this concept and shows that activation of K_ATP channels not only induces membrane depolarization but can also reduce the rate of ATP synthesis and increase O₂ consumption in cardiac mitochondria.

In addition to energy production, mitochondria have the capacity to store Ca²⁺, a function critical in the maintenance of cellular Ca²⁺ homeostasis (13). In the present study, isolated cardiac mitochondria, when challenged with repeated Ca²⁺ pulses, displayed the property to accumulate and store Ca²⁺. Opening of mitochondrial K_ATP channels rapidly released Ca²⁺ from preloaded mitochondria, an effect also seen with K⁺ ionophores. Although the precise mechanism for such an effect on Ca²⁺ release is unknown, the sensitivity toward K_ATP channel blockade and the dependence on extramitochondrial K⁺ concentration are in line with a K_ATP channel-mediated membrane depolarization and promotion of Ca²⁺ release. Such interpretation also concurs with the notion that setting the mitochondrial membrane potential is critical for cyclic accumulation and release of Ca²⁺ (30).

In the present study we further demonstrated that opening of mitochondrial K_ATP channels increased matrix volume, which is manifested through mitochondrial swelling. This finding supports previous reports that mitochondrial K_ATP channels can regulate the osmotic status of the mitochondrial matrix (15, 16). Depending on the extent of swelling, an increase in matrix volume may also lead to a change in the permeability of the outer membrane to macromolecules that are located within the intermembrane (47, 58). In this regard, the present observation, that opening of K⁺ channels could induce release of intermembrane proteins, is suggestive of a significant increase in matrix volume and of a possible compromise in the integrity of the mitochondrial membrane.

In summary, the present study provides direct evidence that opening of mitochondrial K_ATP channels modulates essential functions of isolated cardiac mitochondria. The cellular consequences of mitochondrial K_ATP channel opening are not known. It has been proposed that this ion conductance may participate in the protection of cardiac muscle from ischemia-reperfusion injury (17). Because mitochondrial Ca²⁺ overload is implicated in cellular injury (13), our findings may suggest a possible beneficial effect of mitochondrial K_ATP channel opening through regulation of mitochondrial Ca²⁺ levels. The present study also identifies a K_ATP channel-mediated decrease in the rate of ATP synthesis and a release of mitochondrial proteins. Because a decrease in cellular ATP content coupled with the presence of cytochrome c within the cytosol has been associated with cellular apoptosis (48, 56), this further suggests that opening of mitochondrial K_ATP channels may participate in the regulation of cellular viability. The net effect of opening of mitochondrial K_ATP channels is, however, difficult to predict in view of the diversity in cellular protective mechanisms (20, 21, 34, 35, 44, 57) and the complex nature of K_ATP channel gating (2, 4, 16, 36, 65). Indeed, it is still controversial whether in excitable tissues opening of K⁺ channels leads to programmed cell death (68) or promotes cellular survival (17, 20, 41).