Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

K_ATP may serve an endogenous cardioprotective function, because myocardial preconditioning is thought to be mediated by activation of these channels (Gross and Auchampach, 1992; Tomai et al., 1994). Pharmacologic openers of K_ATP are cardioprotective in numerous experimental models of ischemia and reperfusion, suggesting they may simulate preconditioning (Ohta et al., 1991; Grover, 1994; Mizumura et al., 1995). These protective effects are exerted directly on ischemic myocardium and are independent of vasodilator activity, as well as APD shortening activity (Grover, 1994; Yao and Gross, 1994; Grover et al., 1995a,b). This vasodilator activity may be contraindicated, because blood pressure reductions can compromise diseased myocardium (Belin et al., 1996). The lack of correlation between APD shortening and cardioprotection also suggests that the relevant K_ATP for cardioprotection is distinct from cardiac sarcolemmal channels and may involve activation of mitochondrial K_ATP (Garlid et al., 1996, 1997; Liu et al., 1998). Involvement of mitochondrial K_ATP was deduced from the cardioprotective effects of diazoxide, which opens cardiac mitochondrial K_ATP while having little effect on cardiac sarcolemmal channels. Interestingly, diazoxide is a potent vasorelaxant, suggesting that it opens plasmalemmal channels in vascular smooth muscle and that tissue subtypes for K_ATP exist.

Structure-activity determinations in our laboratories showed a poor correlation between cardioprotective and vasorelaxant activities of K_ATP openers (Atwal et al., 1993). Further studies led to the identification BMS-180448, which shows greater selectivity for ischemic myocardium (lower vasodilator activity), compared with first generation agents such as cromakalim (Grover et al., 1995b). Interestingly, BMS-180448 is also relatively devoid of APD shortening activity in ischemic myocardium (Grover et al., 1995a), even though its cardioprotective effects are abolished by K_ATP blockers. Although BMS-180448 has a lower propensity to vasodilate at cardioprotective doses when compared with currently existing agents, we continued our efforts to further reduce vasodilator activity in novel agents. In addition to reducing vasodilator activity, it was also desirable to eliminate APD shortening activity to reduce proarrhythmic potential.

Further structure-activity studies confirmed that cardioprotection and vasorelaxant effects of K_ATP openers are not correlated (Rovnyak et al., 1997). These studies led to the identification of 4-(N-aryl)-substituted benzopyran derivatives having poor vasorelaxant potency, but retaining cardioprotective potency. One compound in this series, BMS-191095 (see structure in Fig. 1), has little measurable peripheral vasorelaxant activity up to 30 µM while retaining cardioprotective activity that was comparable with BMS-180448 and cromakalim. Due to its superior in vitro pharmacologic profile, BMS-191095 was selected for detailed pharmacologic studies. The goal of these studies was to elucidate the cardioprotective profile for BMS-191095. In addition to determining cardioprotective and vasodilator profile, we also determined if BMS-191095 can selectively open cardiac mitochondrial K_ATP.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Chemical structures of the three KATP openers used in this study.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cardioprotective Profile in Isolated Rat Hearts

Male Sprague-Dawley rats (400-500 g) were anesthetized using 100 mg/kg sodium pentobarbital (i.p.). The trachea was intubated, and then the jugular vein was injected with heparin (1000 U/kg). While being mechanically ventilated, their hearts were perfused in situ via retrograde cannulation of the aorta. The hearts were then excised and quickly moved to a Langendorff apparatus where they were perfused with oxygenated Krebs-Henseleit solution containing (in mM): 112 NaCl, 25 NaHCO₃, 5 KCl, 1.2 MgSO₄, 1 KH₂PO₄, 1.2 CaCl₂, 11.5 glucose, and 2 pyruvate at a constant perfusion pressure (85 mm Hg). A water-filled latex balloon attached to a metal cannula was then inserted into the left ventricle and connected to a Statham pressure transducer for measurement of left ventricular pressure. The hearts were allowed to equilibrate for 15 min, at which time EDP was adjusted to 5 mm Hg, and this balloon volume was maintained for the duration of the experiment. Preischemia or predrug function, heart rate, and coronary flow (extracorporeal electromagnetic flow probe; Carolina Medical Electronics, Inc., King, NC) were then measured. Contractile function was calculated by subtracting EDP from left ventricular peak systolic pressure, resulting in LVDP. Cardiac temperature was maintained throughout the experiment by submerging the hearts in 37°C buffer, which was allowed to accumulate in a stoppered, heated chamber.

After equilibration, the hearts were subjected to one of several treatments. Hearts were treated with vehicle (0.04% DMSO, n = 8), 0.3 to 10 µM BMS-191095 (n = 5 per group), 0.3 to 10 µM BMS-180448 (n = 5 per group), 1 to 10 µM cromakalim (n = 5 per group), 6 µM BMS-191095 + 1 µM glyburide (n = 5), 6 µM BMS-191095 + 100 µM 5-HD, 1 µM glyburide + 10 µM BMS-180448 (n = 5), 1 µM diltiazem (n = 5), and 1 µM glyburide + 1 µM diltiazem (n = 5). The respective drug treatments were given for 10 min before ischemia and were included in the perfusate. Diltiazem was used to confirm that K_ATP blockers are specific for abolishing the protective effects of K_ATP openers. After 10 min of drug treatment, the hearts were subjected to 25 min of global ischemia and 30 min of reperfusion. Ischemia was initiated by completely shutting off perfusate flow. At the end of the reperfusion period, contractile function, coronary flow, and LDH release were measured. The respective drugs were given only before global ischemia and were not given during reperfusion. Severity of ischemia was determined from time to the onset of ischemic contracture, recovery of contractile function at 30 min into reperfusion, and LDH release into the reperfusate. Time to the onset of ischemic contracture was defined as the time (min) during global ischemia in which the first 5 mm Hg increase in EDP was observed. Cardioprotective potency was calculated as the ED₂₅ for increasing time to contracture. ED₂₅ is the concentration of drug casuing a 25% increase in time to contracture relative to vehicle-treated hearts. Because a detailed cardioprotective profile for cromakalim and BMS-180448 has previously been published (Grover, 1994; Grover et al., 1995b), we are only showing the time to contracture and coronary flow data for this compound. For coronary vasodilator potency determinations, extra hearts were prepared to obtain complete concentration responses: 0.03, and 0.1 µM concentrations were added as necessary for each drug (n = 5 for each group done). The respective drug treatment was given for 10 min, and coronary flow was recorded as described above, except that the hearts were removed after the coronary flow determination and ischemia was not instituted.

Effect of BMS-191095 on APD in Guinea Pig Papillary Muscles and Isolated Hearts

Guinea Pig Papillary Preparation. Male guinea pigs (450-600 g) were sacrificed by cervical dislocation. Hearts were rapidly removed and rinsed in Krebs-Henseleit bicarbonate buffer solution (room temperature) equilibrated with 95% O₂/5% CO₂. The Krebs-Henseleit solution for all electrophysiologic studies was composed of the following (mM): 112 NaCl, 5.0 KCl, 11.5 glucose, 25 NaHCO₃, 1.2 MgSO₄, 1.25 CaCl₂, 5.0 mM HEPES, and 1.0 KH₂PO₄ at pH 7.4. A posterior papillary muscle, 3 to 5 mm in length and 1 mm or less in diameter, was removed from the right or left ventricle and was pinned to the base of a 5 ml tissue chamber. The papillary muscle was continuously stimulated through a pair of platinum wires with 1-ms square wave pulses set at 200% of threshold voltage. The frequency of stimulation was held constant at 6 Hz during the first hour of equilibration and then paced at 1 Hz during the remaining equilibration period. Muscles were allowed 2 to 3 h to equilibrate while being superfused with oxygenated buffer solution (15-25 ml/min at 37 ± 0.2°C).

Electrophysiologic Determinations in Isolated Perfused Hearts. Guinea pigs were anesthetized and hearts removed as described above. Hearts were quickly connected to a Langendorff apparatus where they were perfused with oxygenated Krebs-Henseleit solution without the addition of HEPES. Hearts were perfused horizontally with buffer at a constant pressure (85 mm Hg) and temperature (37°C).

Electrophysiological Recording: Whole-Cell K⁺ Currents

In addition to cardioprotection studies, the relative effect of BMS-191095 on K⁺ currents in guinea pig ventricular myocytes was studied to detemine if a dissociation between cardioprotection and sarcolemmal currents could be observed. Guinea pig ventricular myocytes were dissociated using previously described methods (Hamill et al., 1981; Lodge and Smith, 1996). Currents were recorded using the whole-cell configuration of the patch-clamp technique. Electrodes (2-5 M) were fabricated from borosilicate glass. Voltage-clamp protocols were generated and data acquired using Pulse software (HEKA) in conjunction with a HEKA EPC9 amplifier (Lambrecht, Germany). Voltage ramps from 100 to 40 mV were applied from a holding potential of 40 mV (to inactivate sodium current). Resultant currents were filtered at 3 KHz using an analog 4-pole Bessel filter. The bath solution contained (in mM): NaCl, 140; KCl, 4; MgCl₂, 1; CaCl₂, 2; glucose, 10; HEPES, 5; pH 7.4. Nisoldipine (1 µM) was added to the bath solution to inhibit L-type calcium current. The pipette solution contained (in mM): KCl, 125; MgCl₂, 2; CaCl₂, 2; NaCl, 10; EGTA, 10; HEPES, 5; glucose, 10; ATP, 1; pH 7.2 with KOH. All experiments were performed at 33 to 34.5°C. Stock solutions of BMS-191095 or cromakalim (0.01-0.03 M) were prepared using DMSO and then diluted in bath solution as required.

K⁺ currents were also recorded using the perforated patch technique (Horn and Marty, 1988; Korn and Horn, 1989). The patch pipette contained (in mM): 55 KCl, 75 K₂SO₄, 2 MgCl₂, pH 7.3, and 150 to 225 µg/ml nystatin. The external bathing solution contained in mM: 140 NaCl, 4.6 KCl, 1 MgCl₂, 10 dextrose, 10 HEPES, and 1.5 CaCl₂, pH 7.4 (28-31°C). Nifedipine (3 µM) was used to block calcium currents. Currents were activated by applying slow voltage ramps from 100 to 40 mV over a period of 33.75 s (holding potential = 40 mV).

Effect of BMS-191095 on K⁺ Flux through Mitochondrial K_ATP and on the Matrix Volume of Isolated Rat Heart Mitochondria

The purpose of this part of the study was to determine the effect of BMS-191095 on mitochondrial K_ATP in rat heart. Mitochondria were isolated by differential centrifugation from hearts of Sprague-Dawley rats, as described by Saks et al. (1986). The final mitochondrial pellet was resuspended at 20 mg protein/ml in 0.22 M mannitol and 0.07 M sucrose and buffered with 5 mM TES (K⁺ salt), pH 7, and maintained at 0°C. For assays of matrix volume, stock mitochondria were transferred to medium containing K⁺ salts of 130 mM chloride, 20 mM succinate, 5 mM phospate, 0.5 mM EGTA, and 1 mM MgCl, buffered to pH 7.2, 25°C. Changes in matrix volume, secondary to net salt transport across the inner membrane, were followed using a quantitative light-scattering technique, as described in Jaburek et al. (1998). This technique is based on the principal that reciprocal absorbance of the mitochondrial suspension, when corrected for the extrapolated absorbance at infinite protein concentration, is linear with matrix volume within well defined regions (Beavis et al., 1985).

Submitochondrial particles were also prepared, and mitochondrial K_ATP was extracted, purified, and reconstituted into lipid vesicles containing the potassium-sensitive fluorescent probe PBFI, as previously described (Paucek et al., 1992; Garlid et al., 1996). Electrophoretic K⁺ flux was catalyzed by carbonyl cyanide p-(trifluormethoxy)phenylhydrazone addition, to provide a counterion, and the resulting rates were obtained from calibrations. Internal medium contained 100 mM TEA-SO₄, 1 mM EDTA, 25 mM TEA-HEPES, pH 6.8, and 300 µM PBFI. Kinetic studies were performed on proteoliposomes suspended at 0.4 mg lipid/ml of external medium containing 150 mM KCl, 1 mM EDTA, 1 mM MgCl, and 25 mM TEA-HEPES, pH 7.2, at 25°C.

Determination of Vasorelaxant Activity

In Vivo Vasorelaxant Activity. Adult mongrel dogs of either sex weighing 12 to 19 kg were anesthetized with intravenous pentobarbital sodium (30 mg/kg), intubated, and respired with room air to maintain eucapnic conditions using a Harvard respirator (Harvard Apparatus, South Natick, MA). The right femoral artery was cannulated for measurement of blood pressure via a pressure transducer (Gould Statham, Oxnard, CA). The right femoral vein was cannulated for drug infusion. Arterial blood was sampled before and after drug administration for blood gas determinations using an ABL 4 blood gas analyzer (Radiometer, Westlake, OH) to ensure eucapnia and normoxia were maintained. A Millar Mikro-Tip pressure transducer (Millar Instruments, Houston, TX) was inserted into the left carotid artery and advanced into the left ventricle for the measurement of left ventricular pressure and dP/dt. A lead II ECG was also monitored. A left thoracotomy was performed at the 5th intercostal space, and a pericardial cradle was formed.

In Vitro Vasorelaxant Activity. In vitro vasorelaxant activity was determined using rat aortas. IC₅₀ values for relaxation of methoxamine (3 µM)-contracted aortas was used to compare with in vitro cardioprotective potency (time to onset of ischemic contracture). The experimental details of determination of vasorelaxant potency in vitro have been previously described (Grover et al., 1995b). The compounds tested were cromakalim, BMS-180448, and BMS-191095.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cardioprotective Profile of Action of K_ATP Openers

The effect of increasing concentrations of BMS-191095 on cardiac function is shown in Table 1. Before ischemia, BMS-191095 had a slight cardiodepressant effect, which was only significant at the 10 µM concentration. Because we had difficulty keeping BMS-191095 in solution at the 10 µM concentration, the reliability of the data at this concentration may be questionable. During reperfusion, LVDP was significantly reduced in vehicle-treated hearts, indicating severe ischemic/reperfusion damage. BMS-191095 significantly improved reperfusion function starting at the 1 µM concentration, and the peak effect was observed at 6 µM. This protective effect seemed to decrease at the 10 µM concentration, although this may have been due to the poor solubility at this concentration. The effects of BMS-191095 on LDH release are shown in Fig. 2. As can be seen in this figure, LDH release was reduced in a concentration- dependent manner by BMS-191095. Similar results were obtained for reperfusion EDP, although some of the protective effect of BMS-191095 was lost at 10 µM (data not shown). These data were comparable with data generated for cromakalim. BMS-191095 did not exert protective effects when given only during reperfusion (data not shown).

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Effect of increasing concentrations of BMS-180448 and BMS-191095 on reperfusion LDH release in isolated rat hearts. The hearts were treated for 10 min with their respective drug and were then rendered globally ischemic for 25 min and reperfused for 30 min. LDH values are cumulative over the 30-min reperfusion period. BMS-180448 and BMS-191095 significantly reduced LDH release relative to vehicle-treated hearts (*P < 0.05).

The effects of cromakalim, BMS-180448, and BMS-191095 on time to the onset of ischemic contracture are shown in Fig. 3. All compounds increased time to contracture in a concentration-dependent manner and were similar in potency. The EC₂₅ for the percent increase in time to contracture was 3.0 µM for BMS-180448, 1.4 µM for BMS-191095, and 5.0 µM for cromakalim. The data are expressed as the percent change in time to contracture relative to vehicle values, and EC₂₅ is the concentration of drug causing a 25% increase in time to contracture relative to vehicle-treated groups.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effect of cromakalim, BMS-180448, and or BMS-191095 on the time to the onset of ischemic contracture during global ischemia in isolated rat hearts. The values are expressed as percent change from vehicle-treated group values. All agents increased the time to the onset of contracture and the cardioprotective potencies were similar.

We compared the cardioprotective effects of these agents with their vasorelaxant activity, and these data are shown in Table 2. Vasorelaxant potencies in vitro for the K_ATP openers were measured using methoxamine-contracted rat aortas (IC₅₀). Cromakalim was a potent vasorelaxant in this model and was significantly more potent than BMS-180448 and BMS-191095. Using this methodology, BMS-191095 did not show vasodilator activity. By taking the ratio of cardioprotective versus vasorelaxant potencies (EC₂₅/IC₅₀), the relative selectivities of these compounds for ischemic heart can be ascertained. The ratio for cromakalim, BMS-180448, and BMS-191095 are 15.7, 1.0, and <0.05, respectively. Therefore, BMS-191095 is significantly more selective for ischemic myocardium (compared with vascular smooth muscle) relative to cromakalim. Similar results were observed in anesthestized dogs using blood pressure as an index of peripheral vasorelaxation (Table 2). While being significantly less potent as a peripheral vasorelaxant compared with BMS-180448, BMS-191095 was equivalent as a coronary dilator both in vitro and in vivo. The coronary dilator effect of BMS-191095 and BMS-180448 was inhibited by glyburide (data not shown). Although there appears to be some coronary selectivity for BMS-191095, it is nevertheless significantly less potent as a coronary dilator compared with cromakalim.

Data for the effect of K_ATP blockers on the cardioprotective action of BMS-191095 are shown in Fig. 4 and Table 1. The protective effect of BMS-191095 on reperfusion function was abolished by glyburide as was its weak preischemic coronary dilator effect (not shown). The preischemic coronary dilator effect of BMS-191095 was not abolished by 5-HD (data not shown), but the cardioprotective effects of this agent were abolished by 5-HD. These results are consistent with previously published data showing 5-HD to selectively block the cardioprotective effects of K_ATP openers (McCullough et al., 1991). Similar results were seen for BMS-180448 (data not shown). The preischemic and postischemic (Fig. 4) effects of diltiazem were not affected by glyburide as expected.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effect of 10 µM BMS-180448 (448), 6 µM BMS-191095 (095), or 1 µM diltiazem (DILT) with or without glyburide (GLY) or 5-HD on reperfusion LDH release in isolated rat hearts. The hearts were treated for 10 min with their respective drug and were then rendered globally ischemic for 25 min and reperfused for 30 min. LDH values are cumulative over the 30-min reperfusion period. Diltiazem, BMS-180448, and BMS-191095 were all significantly cardioprotective, compared with vehicle (VEH) (*P < 0.05), and the effects produced by the KATP openers were abolished by the KATP blockers, unlike diltiazem.

Effect of K_ATP Openers on Electrophysiologic Parameters

Effect on Guinea Pig Papillary Muscles. Baseline electrophysiologic data for all groups were similar. Average baseline resting membrane potential was 90 ± 2 mV, V_max was 202 ± 20 V/s, and APD₉₀ was 144 ± 7 ms. APD values expressed as percent change from baseline are shown in Fig. 5. Guinea pig papillary muscles treated with vehicle alone displayed no change in action potential characteristics under normoxic conditions. In the presence of hypoxia, a slight reduction in V_max (data not shown) and APD₉₀ was observed. Values returned to control levels during the re-oxygenation period. BMS-191095 did not significantly change action potential characteristics under normoxic conditions. It is important to note that BMS-191095 was evaluated up to 6 µM due to problems associated with solubility at higher concentrations, although 6 µM BMS-191095 and 10 µM cromakalim displayed similar cardioprotective activity. During hypoxia, BMS-191095 did not significantly alter APD₉₀ values relative to vehicle. APD₉₀ values decreased approximately 25% in the vehicle-treated muscles, compared with 20% in the BMS-191095 treated muscles at 15 min of hypoxia. In contrast, cromakalim (10 µM) significantly reduced APD₉₀ values under normoxic conditions from 140 ± 9 to 105 ± 9 ms (25%). During hypoxia, cromakalim reduced APD₉₀ values by 80% at 15 min of hypoxia. Following reoxygenation, both vehicle-treated and BMS-191095-treated muscles returned to near control levels. There were no changes in either resting membrane potential or action potential overshoots in any treatment group (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Effects of vehicle, BMS-191095, or cromakalim on APD90 values recorded intracellularly from guinea pig papillary muscles. Measurements of APD90 were taken at control (C), postdrug (D), during 15 min of hypoxia, and 30 min of reoxygenation. Values were expressed as a percent change from control, and statistical significance was determined at the P < 0.05 level and indicated with an asterisk. No change in APD90 was noted between vehicle and BMS-191095, whereas cromakalim produced significant shortening of APD90 following the drug. APD90 values further shortened during hypoxia and recovery were incomplete to postdrug levels.

Effect on Perfused Guinea Pig Hearts. Isolated perfused guinea pig hearts demonstrated no significant differences in APD₉₀ or QT interval between groups at control readings. Following administration of cromakalim (10 µM), there was a significant reduction in APD₉₀ and QT interval of 16% and 11%, respectively (Fig. 6). These changes were also significantly different from vehicle-treated hearts. BMS-191095 caused no significant change in APD₉₀ or QT-interval values. In contrast, BMS-180448 caused a slight but significant increase in APD₉₀ and QT-interval values of 9% and 11%, respectively. Under hypoxic conditions, APD₉₀ and QT intervals in vehicle-treated hearts decreased 25% and 28%, respectively. Cromakalim caused a precipitous decrease in APD₉₀ and QT-interval values during hypoxia of 66% and 58%, respectively. Unlike cromakalim, BMS-191095 in the presence of hypoxia had no effect on APD₉₀ (29%) and QT interval (30%), compared with vehicle-treated hearts. Similar results were observed for BMS-180448.

View larger version (33K): [in this window] [in a new window]   Fig. 6.   The effects of vehicle (0.1% DMSO), cromakalim (CROM; 10 µM), BMS-180448 (448; 10 µM), or BMS-191095 (095; 6 µM) on APD90 values (A) recorded epicardially and QT intervals (B) taken from isolated perfused hearts are expressed as a percent change from control. Statistical significance was determined at the P < 0.05 level and indicated with an asterisk. Reading were taken under normoxic and hypoxic conditions. BMS-191095 results were nearly identical to the vehicle group for both APD90 and QT-interval values.

Effect of BMS-191095 on Cardiac Myocyte Whole-Cell K⁺ Currents. Experimental conditions were controlled to exclude Na⁺ and Ca²⁺ currents from K⁺ currents. Using the standard whole-cell patch-clamp technique, BMS-191095 (30 µM) failed to activate K_ATP current (n = 9 cells) (Fig. 7). Interestingly, BMS-191095 partially inhibited delayed rectifier potassium currents at voltages greater than approximately 25 mV. Using the perforated patch technique, 3 and 10 µM BMS-191095 (n = 6 and 5 cells, respectively) failed to produce a detectable increase in current. Again, BMS-191095 partially inhibited the delayed rectifier current. Cromakalim activated K_ATP currents as previously described. BMS-191095 had no effects on either Na⁺ or Ca²⁺ (IC₅₀ ~30 µM) currents.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Top panel, effect of BMS-191095 (30 µM) on cell currents measured in isolated ventricular myocytes. Bottom panel, effect of cromakalim (30 µM) on whole-cell currents. For both, currents were evoked by a slow voltage ramp from 100 to +40 mV. Holding potential = 40 mV using the standard whole-cell (open patch) recording technique.

Effect of BMS-191095 on Reconstituted Mitochondrial K_ATP

Since little effect on sarcolemmal K⁺ currents was observed, we then tested the hypothesis that BMS-191095 is selectively opening mitochondrial K_ATP. We used two independent protocols to explore the pharmacology of BMS-191095 in mitochondrial K_ATP.

Figures 8 and 9 contain data from K⁺ flux measurements performed on proteoliposomes reconstituted with mitochondrial K_ATP and suspended in assay medium containing Mg²⁺ and ATP to inhibit the channel (Garlid et al., 1996). The data in Fig. 8 show that BMS-191095 is a potent opener of mitochondrial K_ATP (K_1/2 = 83 nM, n_H = 1). The data in Fig. 9 show that the open state induced by BMS-191095 is inhibited by either glyburide (K_1/2 = 77 nM, n_H = 1) or 5-HD (K_1/2 = 83 µM, n_H = 1). Similar results were obtained in three independent experiments. To verify these reconstitution results, BMS-191095 was evaluated for its effects on matrix volume of mitochondria (0.2 mg/ml) respiring in buffered KCl-succinate medium. Kinetic experiments gave results similar to those reported in Figs. 8 and 9 for the reconstituted channel (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   BMS-191095 activated ATP-inhibited K+ flux in the reconstituted cardiac mitochondrial system. The channel-active DEAE fraction was reconstituted into vesicles containing the fluorescent probe PBFI, and the K+ flux was monitored upon addition of 15 µl of proteoliposomes into 2 ml of assay medium. This graph shows the dose response for BMS-191095 activation in the presence of 0.5 mM ATP. Linear regression of Hill plots of these data (not shown) yielded a Hill slope of 1 and a K1/2 of 83 nM.

View larger version (16K): [in this window] [in a new window]   Fig. 9.   The pharmacologically open state, with ATP (0.5 mM) and BMS-191095 (0.5 mM), was inhibited by either glyburide (K1/2 = 77 nM, nH = 1) or 5-HD (K1/2 = 83 nM, nH = 1).

The trace in Fig. 10 shows the effect of BMS-191095 on mitochondrial steady-state volume during oxidative phosphorylation. Mitochondria were first brought to steady-state volume in K⁺ medium containing 0.5 mM ATP. Initiation of oxidative phosphorylation by addition of ADP caused a significant matrix contraction. This is caused by a normal depolarization under conditions of high electron flow, which perturbs the balance between diffusive K⁺ influx and K⁺ efflux via the mitochondrial K⁺/H⁺ antiporter (Garlid et al., 1996). Matrix volume was partly restored after addition of 3 µM BMS-191095, and this effect on matrix volume was reversed by 10 µM glyburide. The trace in Fig. 10 is representative of four independent experiments, in which the extent of restoration of matrix volume induced by BMS-191095 ranged between 50 to 80%.

View larger version (14K): [in this window] [in a new window]   Fig. 10.   Opening effect of BMS-191095 in intact mitochondria. Rat heart mitochondria (0.2 mg/ml) were assayed in K+ salts of 50 mM chloride, 10 mM succinate, 100 µM EGTA, 5 mM inorganic phosphate, MOPS pH 7.2 buffer, and 500 µM ATP + 1 mM MgCl2. In the presence of ATP (control trace), the mitochondrial KATP channel was closed, but when BMS-191095 was added, the channel was fully opened. This pharmacologic open state was reversed by glyburide (6 µM).  = P(A1  a)/Ps where P is mitochondrial protein concentration (mg/ml) in the assay medium, A is the absorbance ( = 520 nm), a is a machine constant (equal to the y-intercept of a plot of A1 versus P1), and Ps is defined as 1 mg/ml and is a normalizing constant to make  a dimensionless quantity

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacologic openers of K_ATP have been shown by numerous investigators to exert protective effects in experimental models of myocardial ischemia and reperfusion (Auchampach et al., 1991; Yao and Gross, 1994; Armstrong et al., 1995). The protective effects of these agents are characterized by conservation of ATP during ischemia, enhanced recovery of contractile function, and reduction of necrosis (Grover, 1994). The conservation of ATP is seen despite a lack of cardioplegic effects of K_ATP openers (Grover and Sleph, 1995). The cardioprotective effect is also independent of vasodilator activity and is most likely due to a direct protective effect on myocytes (Armstrong et al., 1995). Interest in K_ATP has been further heightened by the findings that preconditioning may be mediated by K_ATP opening (Gross and Auchampach, 1992; Tomai et al., 1994).

Since the protective effects of K_ATP openers are exerted directly on ischemic myocardium, their vasodilator effects are undesirable. A recent study showed that the vasodilator activity of aprikalim caused toxic effects (cardiac lesions) in animal models (Belin et al., 1996). Potent hypotensive activity during acute myocardial ischemia could further compromise coronary blood flow. With this in mind, we performed detailed structure-activity studies to determine whether cardioprotective activity could be separated from vasodilator activity for K_ATP openers and whether distinct structure-activity relationships exist for these two activities (Atwal et al., 1993; Rovnyak et al., 1997). Further investigation led to the discovery of BMS-180448, which is approximately 100- fold more selective for ischemic myocardium relative to vascular smooth muscle (D'Alonzo et al., 1995; Grover et al., 1995b). Although BMS-180448 could be successfully infused in vivo at cardioprotective doses without hemodynamic consequences, the separation between activities could have been greater (higher doses of this compound did cause hypotension). In this series of selective compounds, shortening of APD was also found to be weak, compared with cromakalim (Grover et al., 1995a). Further structure-activity studies revealed cardioprotective K_ATP openers that are devoid of peripheral vasodilator activity (Rovnyak et al., 1997), including BMS-191095.

BMS-191095 exerted concentration-dependent cardioprotective effects in isolated rat hearts, which were not accompanied by reduced cardiac work. Increased time to onset of contracture, however, is suggestive of ATP conservation consistent with findings for other K_ATP openers (McPherson et al., 1993; Grover, 1994). While being slightly more potent as a cardioprotectant relative to cromakalim, BMS-191095 was devoid of peripheral vasorelaxant activity within the concentration range tested. Cromakalim was more potent as a vasorelaxant than as a cardioprotectant.

Although BMS-191095 is not a peripheral vasorelaxant, it retained some coronary dilator activity albeit significantly less, compared with cromakalim, suggesting that coronary arteries may respond differently to K_ATP openers. Although the data were not shown, carotid blood flow and femoral blood flow were not affected by BMS-191095, similar to results for BMS-180448 (Weselcouch and Gomoll, 1997).

Since earlier studies showed a lack of correlation between APD shortening and cardioprotection for K_ATP openers (Yao and Gross, 1994; Hamada et al., 1998), we examined this for BMS-191095. As we showed for BMS-180448 (Grover et al., 1995a; D'Alonzo et al., 1996), BMS-191095 was devoid of APD shortening activity within its cardioprotective concentration range. Cromakalim slightly shortened APD under normoxic conditions, but this activity was potentiated during hypoxia which is consistent with previous reports (Cole et al., 1991; D'Alonzo et al., 1992). In addition to APD studies, whole-cell myocyte patch-clamp studies showed a lack of effects for BMS-180448 (Grover et al., 1995b) and BMS-191095 on sarcolemmal channels.

Despite the lack of correlation between APD shortening, sarcolemmal K_ATP activity, and cardioprotection, K_ATP opening appears to be involved in the cardioprotective effects of BMS-191095. Glyburide abolished both the pre- and postischemic activity of BMS-191095, whereas 5-HD was selective for abolishing its cardioprotective actions, as seen previously (McCullough et al., 1991). Recent evidence suggests that 5-HD may be a selective mitochondrial K_ATP blocker, and this is consistent with its pharmacologic activities (Garlid et al., 1997). Glyburide had no effect on the cardioprotective action of diltiazem, confirming previously published data (Sargent et al., 1991).

K_ATP are expressed in mitochondria (Inoue et al., 1991), and Garlid's laboratory showed that K_ATP openers such as cromakalim and diazoxide activate this channel (Garlid et al., 1996). Diazoxide opens mitochondrial K_ATP with low micromolar potency, while being relatively devoid of sarcolemmal K_ATP activity. Diazoxide was cardioprotective at 1 to 10 µM, while having no effect on sarcolemmal K_ATP current (Garlid et al., 1997). Since BMS-191095 was devoid of APD shortening activity, we determined its effect on mitochondrial K_ATP. BMS-191095 opened mitochondrial K_ATP with a potency comparable with values previously published for diazoxide and cromakalim (Garlid et al., 1997). This mitochondrial K⁺ flux was inhibited by both 5-HD and glyburide. Therefore, the pharmacologic profile of BMS-191095 in reconstituted mitochondrial K_ATP is consistent with the profile in isolated hearts and intact animals. These data strongly suggest BMS-191095 is protecting ischemic myocardium through selective activation of mitochondrial K_ATP. Although these data are consistent with the hypothesis that BMS-191095 protects through mitochondrial K_ATP activation, further work is still needed to prove this definitively.

BMS-191095 is devoid of peripheral vasodilator activity, which should increase its therapeutic window, compared with agents such as cromakalim. The lack of APD shortening by BMS-191095 may be advantageous because of a reduced propensity for reentrant arrhythmias. Our data also suggest that BMS-191095 may be a selective mitochondrial K_ATP opener. Cloning of various K_ATPs, suggest tissue heterogeneity that is consistent with pharmacologic data (Ashcroft, 1996; Inagaki et al., 1996; Wellman and Quayle, 1997). As we learn more about the regulation of mitochondrial K_ATP, the combination of pharmacologic and molecular tools may enable us to elucidate the cardioprotective mechanism of K_ATP openers. A clear understanding of the molecular mechanism for protection by K_ATP will enable us to develop structurally novel and selective agents for protecting ischemic myocardium.