Abstract
The activation of the ATP-sensitive potassium channel (KATP) during myocardial ischemia leads to potassium efflux, reductions in action potential duration and the formation of ventricular fibrillation (VF). Drugs that inactivate KATPshould prevent these changes and thereby prevent VF. However, most KATP antagonists also alter pancreatic channels, which promote insulin release and hypoglycemia. Recently, a cardioselective KATP antagonist, HMR 1883, has been developed that may offer cardioprotection without the untoward side effects of existing compounds. Therefore, VF was induced in 13 mongrel dogs with healed myocardial infarctions by a 2-min coronary artery occlusion during the last minute of a submaximal exercise test. On subsequent days, the exercise-plus-ischemia test was repeated after pretreatment with HMR 1883 (3.0 mg/kg i.v., n = 13) or glibenclamide (1.0 mg/kg i.v., n = 7). HMR 1883 (P < .001) and glibenclamide (P < .01) prevented VF in 11 of 13 and 6 of 7 animals, respectively. Glibenclamide, but not HMR 1883, elicited increases in plasma insulin and reductions in blood glucose. Glibenclamide also reduced (P < .01) both mean coronary blood flow and left ventricular dP/dt maximum as well as the reactive hyperemia induced by 15-sec coronary occlusions (−30.3 ± 11%), whereas HMR 1883 did not alter this increase in coronary flow (−3.0 ± 4.7%). Finally, myocardial ischemia (n= 10) significantly (P < .01) reduced refractory period (control, 121 ± 2 msec; occlusion, 115 ± 2 msec), which was prevented by either glibenclamide or HMR 1883. Thus, the cardioselective KATP antagonist HMR 1883 can prevent ischemically induced reductions in refractory period and VF without major hemodynamic effects or alterations in blood glucose levels. These data further suggest that the activation of KATPs may play a particularly important role in both the reductions in refractory period and lethal arrhythmia formation associated with myocardial ischemia.
Ventricular fibrillation (VF) has been identified as a leading cause of sudden death during myocardial ischemia in humans (Gillum, 1989). It has been proposed that alterations in cellular electrophysiological properties that culminate in the formation of malignant arrhythmias may result from abnormalities in the biochemical homeostasis of the cardiac cells (Opie et al., 1979). Alterations in extracellular potassium, in particular, have been linked to an increased propensity for malignant arrhythmias as a consequence of the interruption in CBF (for reviews, see Billman, 1994; Coronel, 1994). Extracellular potassium concentration rises rapidly during myocardial ischemia (Coronelet al., 1988; Harris, 1966; Kléber, 1984). The resulting depolarization of the surrounding tissue, decreases in action potential duration and nonuniformities of repolarization (refractory period) could all contribute to the induction of malignant arrhythmias (Janse and Wit, 1989).
Since the first characterization of KATP in cardiac tissue (Noma, 1983), an increasing body of evidence has implicated the activation of the channel in many of the electrical consequences of myocardial ischemia (for review, see Billman, 1994). For example, glibenclamide, a sulfonylurea drug that selectively blocks the KATP (Sturgess et al., 1985), has recently been shown to reduce the extracellular accumulation of potassium and reverse the shortening of the action potential duration provoked by hypoxia or myocardial ischemia (Nakaya et al., 1991; Nichols et al., 1991; Venkatesh et al., 1991, 1992). If the extracellular potassium accumulation contributes to the development of VF, then drugs that selectively block the KATP should also protect against these malignant arrhythmias. A limited number of experimental (Billman et al., 1993; Wollenben et al., 1989) and clinical (Cacciapuoti et al., 1991;Davis et al., 1996; Lomuscio and Fiorentini, 1996; Lomuscioet al., 1994) studies, in fact, provide preliminary support for this hypothesis. Glibenclamide, for example, has been shown to prevent VF in isolated ischemic rat hearts (Wollenben et al., 1989), as well as reduce the number and severity of arrhythmias during transient ischemia in diabetic patients with coronary artery disease (Cacciapuoti et al., 1991). In a similar manner, glibenclamide significantly reduced the incidence of VF in non-insulin-dependent diabetic patients with acute myocardial infarction (Lomuscio et al., 1994). Recently, glibenclamide also significantly reduced the incidence of VF induced by the combination of acute ischemia during exercise in dogs with previously healed myocardial infarctions (Billman et al., 1993). This drug also induced large reductions in left ventricular dP/dt maximum (an index of inotropic state) and mean CBF (Billman et al., 1993). Glibenclamide, however, is not selective for cardiac tissue. For example, this drug also blocks pancreatic KATPs, which promotes insulin release and often results in profound hypoglycemia. In addition, the activation of this channel may play an important role in the regulation of CBF (Aversano et al., 1991; Daut et al., 1990), and as such, glibenclamide has been shown to reduce coronary perfusion (Billman et al., 1993). Therefore, one would predict that compounds that selectively block cardiac KATPs should protect against arrhythmias induced by myocardial ischemia without compromising either CBF or blood glucose levels. Recently, a number of KATP subtypes have, in fact, been isolated from different tissues (Inagaki et al., 1996;Isomoto et al., 1996; Garlid et al., 1997). In addition, a novel sulfonylthiourea compound, HMR 1883, 1-[[5-[2-(5-chloro-o-anisamido)ethyl]-2-methoxyphenyl]sulfonyl]-3-methylthiourea (fig. 1) has been shown to block KATP channels in cardiac muscle cells with a much higher potency than in pancreatic β cells (Gögelein et al., 1998). However, the effects of this drug in intact preparations have not been fully characterized. Therefore, it was the purpose of this series of experiments to evaluate the effects of a novel cardioselective KATP antagonist, HMR 1883, on the susceptibility to ventricular fibrillation using an unanesthetized canine model of sudden death.
Methods
The principles governing the care and use of animals, as expressed by the Declaration of Helsinki and as adopted by the American Physiological Society, were followed at all times during this study. In addition, all procedures were approved by the Ohio State University Institutional Animal Care and Use Committee.
Surgical preparation.
Fifty heartworm-free mongrel dogs (Kaiser Lake Kennels, Kaiser Lake, OH), weighing 15.4 to 19.1 kg, were used in this study. The animals were anesthetized and instrumented to measure left circumflex CBF, left ventricular pressure and ventricular electrogram, as previously described (Billman and Hamlin, 1996; Billmanet al., 1993, 1997; Schwartz et al., 1984). Briefly, the animals were given Innovar Vet (0.02 mg/kg fentanyl citrate and 1 mg/kg hydroperidol i.v.; Pittman-Moore, Washington Crossing, NH) as a preanesthetic, whereas a surgical plane of anesthesia was induced with sodium pentobarbital (10 mg/kg i.v.; Ampro Pharmaceutical, Arcadia, CA). A left thoracotomy was made in the fourth intercostal space, and the heart was exposed and supported by a pericardial cradle. A 20-MHz pulsed Doppler flow transducer and a hydraulic occluder were placed around the left circumflex artery. A pair of insulated silver-coated wires were sutured to the epicardial surface of both the left and right ventricles. These electrodes were used for ventricular pacing (see below) or to record a ventricular electrogram from which HR was determined using a Gould Biotachometer (Gould Instruments, Cleveland, OH). A precalibrated solid-state pressure transducer (Konigsberg Instruments, Pasadena, CA) was inserted into the left ventricle via a stab wound in the apical dimple. Finally, a two-stage occlusion of the left anterior descending coronary artery was performed approximately one third the distance from the origin to induce an anterior wall myocardial infarction. This vessel was partially occluded for 20 min and then tied off. All leads from the cardiovascular instrumentation were tunneled under the skin to exit on the back of the animal’s neck.
A transdermal fentanyl patch that delivers 75 μg/hr for 72 hr (Duragesic; Jansen Pharmaceutical, Titusville, NJ) was placed on the back of the neck (secured with adhesive tape) to decrease postoperative discomfort. In addition, bupivacaine HCl, a long-acting local anesthetic (Abbott Laboratories, North Chicago, IL), was injected to block the intercostal nerves (i.e., pain fibers) in the area of the incision. Each animal was placed on prophylactic antibiotic therapy (amoxicillin 500 mg p.o.; IDE Interstate, Amityville, NY) three times daily for 7 days.
The animals were placed in an “intensive care” setting for the first 24 hr and placed on antiarrhythmic therapy as previously described (Billman and Hamlin, 1996; Billman et al., 1993,1997; Schwartz et al., 1984). Thirteen animals (26%) died acutely within the first 72 hours after myocardial infarction. Two additional animals could not be classified (see below) due to rupture of the coronary occluder, and 5 animals were not successfully defibrillated. Thus, studies were completed on 30 of the original 50 animals.
Exercise-plus-ischemia test.
The studies began 3 to 4 weeks after the production of the myocardial infarction. The animals were walked on a motor-driven treadmill and trained to lie quietly without restraint on a laboratory table during this recovery period. Susceptibility to VF was then tested, as previously described (Billman and Hamlin, 1996; Billman et al., 1993, 1997; Schwartzet al., 1984). Briefly, the animals ran on a motor-driven treadmill while workload was increased every 3 min for a total of 18 min. The protocol began with a 3-min warm-up period, during which the animals ran at 4.8 km/hr at 0% grade. The speed was increased to 6.4 km/hr, and the grade was increased every 3 min as follows: 0%, 4%, 8%, 12% and 16%. During the last minute of exercise, the left circumflex coronary artery was occluded, the treadmill was stopped and the occlusion was maintained for 1 additional min (total occlusion time, 2 min). Large metal plates (diameter, 11 cm) were placed across the animal’s chest so that electrical defibrillation could be achieved with minimal delay but only after the animal was unconscious (10–20 sec after VF began). The occlusion was immediately released if VF occurred. Eighteen animals developed VF (susceptible; 5 were not successfully defibrillated), and the remaining 17 did not (resistant).
The susceptible animals then received one or more of the following treatments: (1) the exercise-plus-ischemia test was repeated after pretreatment with glibenclamide (1.0 mg/kg i.v. dissolved in 1–2 ml of dimethylsulfoxide; Sigma Chemical, St. Louis, MO). This was the lowest dose to prevent VF. The drug was injected in a cephalic vein ∼3 min before exercise began. (2) The exercise-plus-ischemia test was repeated after pretreatment with the cardioselective KATP blocker HMR 1883 (Hoechst Marion Roussel, Frankfurt, Germany). This drug was dissolved in distilled water containing 0.48% NaCl and 0.6% NaHCO3 and heated to ∼80°C for 20 to 30 min. Eight animals received 3.0 mg/kg i.v. ∼3 min before exercise onset (i.e., 20 min before occlusion), and 5 animals received 3.0 mg/kg i.v. 1 hr before exercise onset (i.e., 78 min before occlusion). This dose of HMR 1883 produced a plasma concentration of 1.5 μg/ml (3 μmol/liter) and was the lowest dose to prevent VF. Preliminary studies demonstrated that this concentration elicited a half-maximal inhibition of the activation of KATP by the channel agonist rilmakalim. Furthermore, this concentration of HMR 1883 also nearly abolished the reductions in action potential duration induced by hypoxia without altering pancreatic cell membrane potential (Gögelein et al., 1998). (3) Finally, a second control (saline) exercise plus ischemia test was performed 1 week after the last drug test. At least 5 days elapsed between drug treatments and all drugs were given in a random order.
Refractory period determination.
On a subsequent day, the effective refractory period was determined as previously described (Billman and Hamlin, 1996), using a Medtronic model 5325 programmable stimulator, both at rest (n = 20) and during myocardial ischemia (n = 10). Briefly, the heart was paced for 8 beats (S1; intrastimulus interval, 300 msec; pulse duration, 1.8 msec at twice-diastolic threshold of ∼6 mA). The intrastimulus interval was progressively shortened between the last paced beat and a single extrastimulus (S2). The refractory period represented the shortest interval capable of generating a cardiac response and was measured using either the left or right ventricular electrodes. This procedure was completed within 30 sec. No differences were noted between either pair of electrodes. The data therefore were combined.
Once the control values were determined, refractory period measurements were repeated after glibenclamide (1.0 mg/kg i.v., n = 17), the KATP agonist pinacidil (0.5 mg/kg i.v. dissolved in 1 ml of ethanol, n = 14) and HMR 1883 (3.0 mg/kg i.v., n = 9). After the completion of these studies, refractory period was determined during myocardial ischemia (2-min occlusion of the left circumflex coronary artery) in 10 animals. The refractory period was determined ∼60 sec after the onset of the coronary occlusion. At least 24 hr after the control (i.e., no drug) occlusion, refractory period was determined before and after the following treatments: glibenclamide (1.0 mg/kg i.v.,n = 7), pinacidil (0.5 mg/kg i.v., n = 8) and HMR 1883 (3.0 mg/kg i.v., n = 7). These drugs were given in a random order 24 to 48 hr apart. On a subsequent day, the effects of HMR 1883 (n = 6) and glibenclamide (n = 4) on blood glucose and plasma insulin were evaluated. The animals were fasted at least 16 hr before the administration of the drugs. Blood samples (1 ml) were drawn from a cephalic vein before and 10, 20, 30, 60, 120 and 240 min after the injection of HMR 1883 (3.0 mg/kg i.v.) or glibenclamide (1.0 mg/kg). Time control samples (i.e., after the injection of saline) were obtained in the same animals the next day. Blood glucose levels were determined using a glucometer (Accu-Chek Instant Monitor; Boehringer Mannheim, Indianapolis, IN). Plasma insulin levels were measured using a commercially available radioimmunoassay (Serano, Freiburg, Germany).
Reactive hyperemia studies.
The KATP has been implicated in vascular regulation, particularly CBF (Aversano et al., 1991; Belloni and Hintze, 1991; Daut et al., 1990). Therefore, the effects of HMR 1883 and glibenclamide on the response to brief interruptions in CBF were also evaluated. Animals (n = 5) were placed on a laboratory table, and the left circumflex coronary was occluded three or four times for 15 sec. At least 2 min (or until CBF had returned to preocclusion base line) elapsed between occlusions. The occlusions were then repeated 5 min after either HMR 1883 (3.0 mg/kg i.v.) or glibenclamide (1.0 mg/kg i.v.). On the subsequent day, the studies were repeated with the drug that had not been given the previous day.
Data analysis.
All hemodynamic data were recorded on a Gould model 2800S eight-channel recorder (Cleveland, OH) and a Teac model MR-30 FM tape recorder (Tokyo, Japan). Coronary blood flow was measured with a University of Iowa Bioengineering flowmeter model 545 C-4 (Iowa City, IA). The rate of change of left ventricular pressure [d(LVP)/dt] was obtained by passing the left ventricular pressure through a Gould differentiator that has a frequency response linear to >300 Hz. The data were averaged over the past 5 sec of each exercise level. The coronary occlusion data were averaged over the last 5 sec before and at the 60-sec line point (or VF onset) after occlusion onset. The total area between the peak CBF and return to base line was measured for each 15-sec occlusion, and the percent repayment was calculated. The reactive hyperemia response to each occlusion was then averaged to obtain one value for each animal. The data were then analyzed using analysis of variance for repeated measures. When the F ratio was found to exceed a critical value (P < .05), Scheffé’s test was used to compare the mean values. The effects of the drug intervention on arrhythmia formation were determined using a χ2 test with Yates’ correction for continuity. All data are reported as mean ± S.E.M. Cardiac arrhythmias, PR interval and QT interval were evaluated at a paper speed of 100 mm/sec. QT interval was corrected for HR using Bazett’s method.
Results
Effects on susceptibility to VF.
The exercise-plus-ischemia test induced VF in 18 animals (5 were not successfully defibrillated). In agreement with previous studies (Billman and Hamlin, 1996; Billmanet al., 1993, 1997; Schwartz et al., 1984), VF was reproducibly induced in the susceptible animals with each presentation of both control exercise-plus-ischemia tests. The average time to the onset of VF was 62 ± 4 sec (range, 35–90 sec). The control exercise-plus-ischemia test elicited similar hemodynamic changes (e.g., HR: first occlusion control, 198.7 ± 8.6; occlusion, 224.2 ± 8 beats/min; second occlusion control, 181.6 ± 11.0; occlusion, 232.2 ± 8.1 beats/min) with a similar time to VF onset (59 ± 5.4 sec; range, 37.5–101 sec) as the first test. The exercise-plus-ischemia test was repeated after the following treatments: HMR 1883 (3.0 mg/kg i.v., n = 8), HMR 1883 (3.0 mg/kg i.v., 1 hr before the exercise-plus-ischemia test,n = 5) and glibenclamide (1.0 mg/kg i.v.,n = 7). Representative examples for the same animal before and after pretreatment with HMR 1883 are shown in figure2. HMR 1883 given 3 min before the onset of exercise prevented VF in 7 of 8 animals (χ2 = 9.1, P < .005). This drug completely suppressed ventricular arrhythmia formation in 6 animals. One animal exhibited a brief run of ventricular tachycardia (a run of four extra beats). In a similar manner, HMR 1883, 3.0 mg/kg i.v., given 1 hr before the exercise-plus-ischemia test, prevented ventricular fibrillation in 4 of 5 additional animals (χ2 = 3.8, P < .05). Thus, 3.0 mg/kg i.v. HMR 1883 prevented malignant arrhythmias in a total of 11 of 13 animals (χ2 = 15.8, P < .001, figure3). Finally, 1.0 mg/kg i.v. glibenclamide elicited a similar protection, protecting 6 of 7 animals (χ2 = 7.3, P < .01).
The hemodynamic response to exercise before and after HMR 1883 (3.0 mg/kg i.v.) is displayed in figure 4. Exercise elicited increases in HR, LVSP, left ventricular dP/dt maximum (an index of inotropic state) and mean CBF, which were not altered by HMR 1883. In a similar manner, HMR 1883 did not alter the hemodynamic response to coronary occlusion (table 1). The hemodynamic response to exercise before and after glibenclamide (1.0 mg/kg i.v.) is shown in figure 5. Glibenclamide did not alter the HR and LVSP responses to exercise, nor did this drug alter the response to the coronary occlusion (table 1). However, in contrast to HMR 1883, glibenclamide significantly (P < .01) reduced both mean CBF and left ventricular dP/dt maximum. These data confirm an earlier report using a higher dose of 10 mg/kg glibenclamide i.v. (Billman et al., 1993).
Electrophysiological and hemodynamic effects of KATPagonists and antagonists before and during myocardial ischemia.
The electrophysiological and hemodynamic effects of the KATP-modulating drugs before myocardial ischemia are displayed in table 2. HMR 1883 did not alter the resting values of any these of variables. Glibenclamide, however, provoked significant reductions in both left ventricular dP/dt maximum and mean CBF (table 2). In contrast, pinacidil significantly increased HR and mean CBF, which was accompanied by significant reductions in LVSP, PR interval and refractory period (table 2). Myocardial ischemia (n = 10) elicited significant increases in HR (control, 131 ± 7; occlusion, 151 ± 9 beats/min), which, in agreement with previous studies (Li and Ferrier, 1991; Billman and Hamlin, 1996), was accompanied by small but significant reductions in refractory period (fig.6). Pretreatment with either glibenclamide (n = 7) or HMR 1883 (n = 7) prevented the ischemia-induced reductions in refractory period. Pinacidil, as noted above, significantly reduced refractory period, which was not further reduced by ischemia.
Effects on reactive hyperemia.
Representative examples of CBF response to a 15-sec coronary artery occlusion before and after either glibenclamide or HMR 1883 are shown in figure7. The release of 15-sec coronary occlusion was accompanied by a large repayment of the flow deficit (control, 426.8 ± 102%), which was not significantly altered by HMR 1883 pretreatment (414.6 ± 133.3%, −3.0 ± 4.7% compared with control values). In contrast, glibenclamide provoked a large and significant reduction (−30.0 ± 11% compared with control values) in the flow repayment after occlusion release (310.6 ± 72.6%). Similar reductions in CBF have been reported after either glibenclamide (Aversano et al., 1991) or the administration of the adenosine antagonists aminophylline (Billman, 1987) and adenosine deaminase (Saito et al., 1981). It should be noted that it has been proposed that adenosine elicits coronary vasodilation via the activation of the KATP channel (Daut et al., 1990). Finally, glibenclamide provoked significant decreases in plasma insulin accompanied by large increases in plasma glucose (fig.8). When considered together, these data demonstrate that HMR 1883, in contrast to glibenclamide, does not inhibit the activation of either vascular or pancreatic KATPs.
Discussion
In the present study, both glibenclamide and a novel KATP antagonist, HMR 1883, prevented ischemically induced VF without altering the hemodynamic response to the coronary occlusion. Furthermore, both compounds prevented ischemically induced reductions in refractory period. Conversely, the activation of the KATP with pinacidil provoked reductions in refractory period similar to those noted during myocardial ischemia. Glibenclamide, in contrast to HMR 1883, significantly attenuated the reactive hyperemia associated with coronary occlusion release, reduced both mean CBF and left ventricular dP/dt maximum, increased plasma insulin concentration and promoted hypoglycemia. When considered together, these data suggest that activation of cardiac KATPs during myocardial ischemia may play an important role in both ischemically induced reductions in refractory period and the formation of malignant ventricular arrhythmias. The data further demonstrate that HMR 1883 preferentially blocks cardiac KATPs without adversely affecting either vascular or pancreatic KATPs.
HMR 1883 and selectivity for cardiac KATPs.
In the present study, HMR 1883 could prevent ischemically induced reductions in ventricular refractory period without altering either plasma insulin, blood glucose or coronary reactive hyperemia. In agreement with these findings, Gögelein et al. (1998)demonstrated that in isolated guinea pig papillary muscle or single cardiomyocytes, HMR 1883 could attenuate the reductions in action potential duration induced by either the potent KATP opener rilmakalim hypoxia or metabolic inhibition. A similar inhibition has been reported for glibenclamide (Gögelein et al., 1998; Krause et al., 1995; Nakaya et al., 1991;Takizawa et al., 1996). However, HMR 1883 was >3 orders of magnitude less potent than glibenclamide in blocking the KATP activation in pancreatic β cells (Gögeleinet al., 1998). The authors concluded that at therapeutic concentrations, HMR 1883 would preferentially block the activation of cardiac KATPs.
In a similar manner, KATPs have been shown to play a significant role in the regulation of vascular muscle tone (Aversanoet al., 1991; Belloni and Hintze, 1991; Daut et al., 1990). For example, in agreement with the present study, KATP agonists such as pinacidil promote smooth muscle relaxation and thereby reduce arterial pressure and increase CBF (Cavero et al., 1989). The activation of these channels by adenosine also participates in the regulation of CBF (Daut et al., 1990). Belloni and Hintze (1991) demonstrated that glibenclamide significantly increased the dose of adenosine required to achieve a half-maximal coronary vasodilation. In agreement with the present study, glibenclamide reduced the hyperemic response to coronary occlusion (Aversano et al., 1991) and attenuated the increase in coronary blood flow elicited by exercise (Billman et al., 1993). In the present study, HMR 1883 did not alter exercise-induced changes in CBF (fig. 4), nor, as previously noted, did this drug alter the duration of reactive hyperemia. Thus, HMR 1883, in contrast to glibenclamide, does not exert any negative actions on the CBF regulation (compare figs. 4 and 5). When considered together, the data described above suggest that in vivo, HMR 1883 has little or no effects on either vascular or pancreatic tissue at concentrations that inhibit cardiac KATPs.
HMR 1883 and susceptibility to VF.
In agreement with the present study, glibenclamide has been shown to prevent ventricular arrhythmias induced by myocardial ischemia in both isolated hearts (Bellemin-Baurreau et al., 1994; Chi et al., 1993; Gwilt et al., 1992; Tosaki and Hellegouarch, 1994) and intact preparations (Billman et al., 1993). For example,Wolleben et al. (1989) demonstrated that ischemia-induced VF in isolated rat heart was prevented by the sulfonylurea drugs glibenclamide and tolbutamide. In contrast, potassium channel agonists decreased the time to fibrillation (Chi et al., 1990, 1993;Wolleben et al., 1989). In a similar manner, Billmanet al. (1993) further demonstrated that glibenclamide significantly reduced the incidence of VF induced by the combination of exercise and acute myocardial ischemia in animals previously shown to be susceptible to life-threatening arrhythmias. This drug suppressed ventricular fibrillation in 13 of 15 animals. However, in agreement with the present study and in contrast to HMR 1883 (compare figs. 4 and5), glibenclamide provoked large reductions in both mean CBF and myocardial contractility as measured by left ventricular dP/dt maximum.
There are only a few clinical studies that also illustrate the antiarrhythmic potential of KATP antagonists (Cacciapuotiet al., 1991; Davis et al., 1996; Lomuscio and Fiorentini, 1996; Lomuscio et al., 1994). Cacciapuotiet al. (1991) found that glibenclamide significantly reduced the frequency and severity of ventricular arrhythmias recorded during transient ischemia in non-insulin-dependent diabetic patients with coronary artery disease. This drug did not alter nonischemic arrhythmias, nor did it change the number or length of the ischemic episodes. In a similar manner, glibenclamide substantially reduced the incidence of VF during acute myocardial infarction in non-insulin-dependent diabetic patients (Lomuscio et al., 1994). VF occurred in 1.9% of the glibenclamide-treated patients compared with 7.9% of patients treated with diet or other hypoglycemic drugs and 9.9% of nondiabetic patients. The mechanisms responsible for the protective actions of HMR 1883 and other sulfonylurea drugs on the ischemic heart remain to be determined.
Cardiac KATP and ischemically induced VF: Possible mechanisms.
Extracellular potassium has been known for some time to increase rapidly during ischemia (for reviews, see Billman, 1994;Coronel, 1994). The increased extracellular potassium promotes the depolarization of the tissue surrounding the ischemic regions, as well as reductions in action potential duration, which can provoke abnormalities of impulse conduction (Hicks and Cobbe, 1990; Nakayaet al., 1991; Venkatesh et al., 1991; Wildeet al., 1990). A major factor contributing to VF, particularly during myocardial ischemia, is a dispersion or inhomogeneity of refractory period resulting from regional differences in action potential duration (Janse and Wit, 1989). This allows for the fragmentation of impulse conduction during ensuing beats. The activation of the KATP produces large reductions in action potential duration, which are inhibited by glibenclamide (Bekheitet al., 1990; Kantor et al., 1990; Ruiz-Petrichet al., 1992) but exacerbated by pinacidil (Vanheel and de Hemptinne, 1992; Venkatesh et al., 1992). Glibenclamide has been reported to attenuate ischemically induced regional differences in refractory period (DiDiego and Antzelevitch, 1993; Kubota et al., 1993; Tweedie et al., 1993). In contrast, the KATP agonist pinacidil elicited a marked dispersion of repolarization and refractory period between the epicardium and endocardium, leading to the formation of extrasystoles (DiDiego and Antzelevitch, 1993). Therefore, HMR 1883, by preferentially blocking cardiac KATPs, could prevent potassium efflux and the resulting nonuniformities in refractory period between the ischemic and nonischemic tissue. As such, HMR 1883 could prevent VF by abolishing this substrate for reentrant arrhythmias induced by ischemia.
It should be noted that the activation of the KATP can also alter cardiac mechanical function. For example, a growing body of evidence suggests that activation of this channel during myocardial ischemia may conserve ATP levels and thereby preserve contractile function, particularly during subsequent ischemic events (Hearse, 1995;Cole et al., 1991). Thus, KATP antagonists could adversely affect cardiac contractility. Indeed, glibenclamide can exacerbate the impairment induced by ischemia (see Hearse, 1995) and provoked large reductions in left ventricular dP/dt maximum in the present study. However, the protection that results from the activation of the KATP does not correlate with alterations in action potential duration. Grover et al. (1995) demonstrated that the KATP agonists BMS-180448 and cromakalim had equipotent cardioprotective actions yet elicited markedly different changes in action potential duration. Cromakalim, but not BMS-180448, reduced action potential duration. The authors therefore concluded that these drugs could produce the preservation of mechanical function noted during myocardial ischemia by opening of sarcolemmal KATPs. In a similar manner, Garlid et al. (1997) demonstrated that the KATP agonist diazoxide maintained cardiac contractility during ischemia yet exerted minimal effects on sarcolemmal KATPs. However, they found that diazoxide as well as cromakalim elicited a potent activation of cardiac mitochondrial KATPs, an action blocked by glibenclamide (Garlid et al., 1997). They therefore proposed that diazoxide, and perhaps KATP agonists in general, protected the ischemic myocardiumvia actions on mitochondrial rather than sarcolemmal KATPs. If this hypothesis is correct, then it may be possible to develop agents that differentiate between the electrophysiological (i.e., sarcolemmal) and mechanical (i.e., mitochondrial) actions of the activation of the KATPs and thereby prevent malignant arrhythmias without untoward effects on cardiac contractility. The observation that HMR 1883, in contrast to glibenclamide, did not alter left ventricular dP/dt maximum (compare figs. 4 and 5) suggests that HMR 1883 may preferentially inhibit sarcolemmal KATPs rather than block mitochondrial KATPs. This intriguing possibility merits further investigation.
In summary, the KATP antagonist HMR 1883 abolished ischemically induced reductions in refractory period and prevented VF in animals known to be susceptible to malignant arrhythmias. In contrast to glibenclamide, HMR 1883 did not increase plasma insulin, reduce blood glucose, decrease ventricular contractility or alter CBF regulation. This drug may therefore selectively inhibit cardiac KATPs at therapeutically effective concentrations. This conclusion in fact can explain the antiarrhythmic properties of this drug. The activation of the KATP promotes potassium efflux with a corresponding reduction in action potential duration as ATP levels decline (Billman, 1994). The resulting regional differences in refractory period would create the conditions necessary for the formation of reentrant conduction and extrasystoles (Noma, 1983). Thus, by preventing the activation of the KATP HMR 1883 would prevent or reduce inhomogeneities in repolarization and thereby prevent the formation of a substrate for reentrant arrhythmias during ischemia. Because, as noted above, the KATP is only active as ATP levels decline, this drug has the added advantage that it only becomes active in ischemic tissue with little or no effect on the normal tissue. One would therefore predict that HMR 1883 should be particularly effective against ischemic arrhythmias and thereby would reduce cardiac mortality risk in patients with ischemic heart disease.
Footnotes
-
Send reprint requests to: George E. Billman, Ph.D., Department of Physiology, The Ohio State University, 302 Hamilton Hall, 1645 Neil Ave., Columbus, OH 43210-1218.
- Abbreviations:
- KATP
- ATP-sensitive potassium channel
- HR
- heart rate
- CBF
- coronary blood flow
- LVSP
- left ventricular systolic pressure
- VF
- ventricular fibrillation
- Received November 25, 1997.
- Accepted March 31, 1998.
- The American Society for Pharmacology and Experimental Therapeutics