Bimakalim: A Promising KATP Channel Activating Agent

Cardiovascular Drug Reviews Vol. 18, No. 1, pp. 25–46 © 2000 Neva Press, Branford, Connecticut

Bimakalim: A Promising KATP Channel Activating Agent Paolo Emilio Puddu, Keith D. Garlid,† Francesco Monti, Katsunori Iwashiro, Sandra Picard,‡ Amos Adeyemo Dawodu, Anna Criniti, Giovanni Ruvolo, and Pietro Paolo Campa Laboratory of Cardiovascular Pharmacology, Department of Cardiac Surgery and II Section of Cardiology, University “La Sapienza”; Rome, Italy; † Department of Chemistry, Biochemistry, and Molecular Biology, Oregon Graduate Institute of Science and Technology, Portland, OR USA; ‡ on leave from Laboratory of Experimental Anesthesiology and Cellular Physiology, University, Caen, France, in partial fulfillment of work performed during the tenure of a Doctorate exchange fellowship between the Universities of Poitiers, France and Rome “La Sapienza,” Italy

Key Words: ATP-dependent potassium channels—Bimakalim — Contractility—Electrophysiology—Myocardial protection.

INTRODUCTION This review focuses on selected aspects of the pharmacology of bimakalim. Bimakalim is probably the most intensively investigated, potent, and pure activator of adenosine triphosphate sensitive potassium channels (KATP), both at plasmalemmal (42,48) and mitochondnal (18) levels. It survived the initial developmental scrutiny with a substantial amount of data, including studies with human tissues (42,48) and even initial clinical studies in volunteers (11,52). Bimakalim is in many respects similar to other potassium channel openers (KCOs) reviewed in other publications (4,27,34,58). Bimakalim shares also its fate with other KCOs. Most pharmaceutical companies discontinued clinical development of KCOs. This decision is regrettable, particularly in view of the current interest in pharmacologically inducible preconditioning (PC) (23,25,56). Some recent data, discussed in this review, are pointing to the intracellular site of action of bimakalim, probably at the mitochondrial (mito-)-KATP (17,18,32). Address correspondence and reprint requests to: P. E. Puddu, MD, FESC, FACC, Istituto di Chirurgia del Cuore e Grossi Vasi, Universita degli Studi di Roma “La Sapienza,” Viale del Policlinico, 155, Rome 00161, Italy. Tel: +39 (06) 445-5291; Fax: +39 (06) 444-1600; E-mail: [email protected]

25

26

P. E. PUDDU ET AL.

FIG. 1. Chemical structure of bimakalim [(4-(1,2dihydro-2-oxo-1-pyridyl)-2,2-dimethyl-2H-1-benzopyran-6-carbonitrile): EMD 52692 and SR 44866]. Molecular weight: 278.33.

CHEMISTRY Bimakalim (Fig. 1) is 4-(1,2-dihydro-2-oxo-1-pyridyl)-2,2-dimethyl-2H-1-benzopyran-6-carbonitrile. Its molecular weight is 278.33, and its melting point is 144 – 146°C. It is sparingly soluble in water and is available as micronized white powder. Its solubility in water, saline, simulated gastric juice, or intestinal fluid is 0.07 g/100 mL. In acetonitrile, ethanol, and trichloroethane, its solubility is 29.2, 11.4, and 5.6 g/100 mL, respectively. The compound is nonhygroscopic and not sensitive to heat, humidity, or oxygen exposure. It is, however, sensitive to light, particularly in the solid state. Bimakalim is stable in acidic (pH 1) and neutral solutions but unstable at pH 10. Its n-octanol/pH 7.4 buffer solution partition coefficient is 27. Hence, bimakalim displays moderate lipophilicity (Merck Laboratories, Darmstadt, data on file). It might be estimated that, when appropriately kept, bimakalim powder stays unaltered for at least 5 years. For use with either normal (pH 7.35) or ischemia-simulating (pH 7.00) or modified Tyrode’s solutions and guinea pig or human myocardial tissue (42,43,48), we dissolved bimakalim in double-distilled water and used it at concentrations ranging from 10 nM to 10 mM. Other investigators used polyethylene glycol as solvent, diluted 0.5 mg bimakalim in 0.5 mL of solvent, and added saline to obtain the desired concentration of the compound (2,60).

PHARMACOLOGY Pharmacokinetics and Pharmacodynamics Auchampach and Gross (2) were first to measure bimakalim plasma concentrations in dogs using a sensitive method. They reported plasma levels of bimakalim as 11.4 ng/mL at 45 min after 3 mg/kg bolus injection followed by infusion at 0.1 mg/kg/min, i.v. (60). At 165 and 285 min after the bolus injection, mean plasma levels increased to 14.6 and 18.6 ng/mL suggesting slow clearence of bimakalim in the dog. There are little or no published data on the absorption, distribution, metabolism and excretion of bimakalim in humans. Senior et al. (52), referring to a personal communication from Merck Laboratories, reported that in 12 volunteers given 0.2 mg bimakalim maximum plasma concentration was 13.8 ± 3.8 ng/mL with a time to peak between 0.5 and 2 h and a mean plasma half-life of 14.7 ± 7.9 h. Based on these data, a time interval of 90 min was selected between oral administration and the time of the expected maximal

Cardiovascular Drug Reviews, Vol. 18, No. 1, 2000

BIMAKALIM

27

pharmacological effect of the drug. We estimated the plasma concentration of bimakalim after the maximal oral dose of 1 mg to be ~50 nM (48). It is not clear whether canine pharmacokinetics of bimakalim are relevant to that in humans. It is, however, a crucial issue in the interpretation of most in vivo data with bimakalim. Myocardial protection with bimakalim is likely to be its major and clinically relevant pharmacodynamic action. In the dog, a mean bimakalim plasma level of 1 ng/mL was reached at 45 min after injection of 4 mg/kg of the drug (2,60). In humans, at 90 min after oral administration of 2.7 mg/kg of bimakalim the plasma level of the drug was 14 ng/mL (52). Since there are no comparative pharmacokinetic studies with bimakalim in dogs and humans, the relevance of pharmacodynamic studies in dogs to the clinical setting is questionable. Also, plasma concentrations of bimakalim may not represent the desired therapeutic target. In fact, in dogs a significant cardioprotective effect without any arrhythmias was observed at doses of bimakalim that produced plasma concentrations at a low nanomolar range (60). However, in human, myocardial tissues both action potential duration (APD) shortening (48) and negative inotropic effect (42) were linearly related to superfusing concentrations of bimakalim and no significant effect was seen with the drug at concentration below 100 nM.

Mode of Action Bimakalim and other KCOs act to inhibit cell function mainly by hyperpolarizing the plasma membrane, and they are effective in many tissues, including cardiac, skeletal and smooth muscles, neurons, and secretory cells (1). Their inhibitory actions are antagonized by sulphonylureas such as glibenclamide (1,16). Glibenclamide has a receptor site on the protein, named Sulphonyl Urea Receptor (SUR), which is distinct but intimately associated with the KATP channel (1). The site of action of KCOs is likely to be at the level of the pore-forming protein of KATP channel, most probably at the internal pore level. As recently reviewed (48), KCOs access their receptor at the inner surface of the cell membrane where a phosphorylation-dephosphorylation mechanism might be involved. However, it is not certain whether KCOs bind to the channel protein or to the SUR or whether they promote an indirect competition with the binding site for ATP on cyclic nucleotide-dependent protein kinases. In cardiac myocytes the effects of cromakalim may be blocked by 5-hydroxydecanoic acid (36), whereas this is not true in smooth muscle cells. The clear-cut diversity in the molecular structure of most KCOs (7) suggests that the nature of the interaction between the KCOs and KATP channel is still largely unknown. The salient properties of all KATP channels are: 1) selectivity for the potassium current; 2) inhibition by ATP with unusually high affinity; 3) regulation of ATP inhibition by other nucleotides; and 4) inhibition by glibenclamide with high affinity. A certain degree of differentiation exists among KATP channels. The EC50 for ATP inhibition of plasma membrane KATP channels varies among heart (25 – 500 mM), skeletal muscle (17 – 135 mM), and pancreatic b cells (10 – 70 mM) (41); the EC50 for the potent inhibitor glibenclamide also varies among heart (20 – 300 nM), skeletal muscle (3 – 100 nM), and pancreatic b cells (4 – 20 nM). In mito-KATP channel EC50s are 45 mM for ATP and 60 nM for glibenclamide, respectively, with quantitatively indistinguishable figures in rat liver or beef heart mitochondrial preparations (41).


28

P. E. PUDDU ET AL.

At the level of mito-KATP channel, electrophoretic K+ uptake, a result of electrogenic H+ ejection by the mitochondrial redox chain, will dissipate energy and uncouple oxidative phosphorylation. Therefore, the mito-KATP channel must perforce be regulated (41). The observed high affinity for ATP meets the expectation that the mito-KATP channel should normally be closed but raises the obverse question — how can it be opened under physiological conditions? There is a possibility to justify the physiological mission of mito-KATP channel, namely to expand matrix volume during mitochondrial biogenesis in association with synthesis of proteins and expansion of membranes. As outlined by Paucek et al., volume expansion would require net K+ uptake, mediated by an imbalance between the activity of the mito-KATP channel and the K+/H+ antiporter (41). These assumptions represent the basis for the working hypothesis on mito-KATP channel activation and KCO-induced myocardial protection (Garlid et al. unpublished). As a rule, all synthetic KCOs discovered so far are much more potent in smooth muscle than in cardiac myocytes (15). This is also true for bimakalim (12,20). At relatively high concentrations (> 1 mM), KCOs open KATP channels in the normally oxygenated myocardium (15), causing a dramatic decrease in the APD (5,40,48). This effect on APD can be viewed as potentially toxic. It is important, however, to realize that at clinically used plasma concentrations of bimakalim (~50 nM) APD shortening (and, therefore, possibly arrhythmogenesis) was not seen in normoxic human atrial and ventricular fibers at 37°C (48). Under conditions of impaired cellular metabolism, such as ischemia, the situation is entirely different since cardiac KATP channels become much more sensitive to KCOs (9,14,15,40). This striking behavior is the basis of the cytoprotective properties of KCOs, making them potentially useful in ischemic heart disease (8). This behavior is currently the subject of intensive research (23,25,28). The clinical consequences of these effects, should they occur also in humans, might be indeed far-reaching in respect to the prevention of acute myocardial infarction, or of myocardial damage during percutaneous transluminal coronary artery dilation, coronary bypass, or heart transplantation. It will be essential, however, to administer KCOs before the ischemic event, a significant limitation in finding an appropriate clinical use for these compounds. Of paramount importance in reference to the mode of action was the observation that KCOs, and bimakalim in particular, can dramatically reduce infarct size (IS) at very low doses in anesthetized dogs following acute coronary artery ligation and reperfusion. KCOs can also induce postischemic arrhythmias and even ventricular fibrillation (60). It is, however, far from being firmly established whether this dichotomy of cytoprotective versus profibrillatory actions of KCOs involves modulation of APD shortening (9,15,26,60), is dose-related (43,48), is mediated through an intracellular site that is independent of any electrophysiologic effects (17,42). In our experiments with human atrial fibers subjected to hypoxia, we found that bimakalim, at 100 nM, had no effect on APD shortening or its kinetics (42).

Hemodynamic Effects in Animals and Man Bimakalim, at doses ranging from 300 ng to 1.2 mg/kg/min, in either anesthetized (51) or consious pigs (11), showed characteristics of a potent vasodilator, with a particularly pronounced effect on the vasculature of the brain. The hemodynamic profile of bimakalim or its vehicle were first investigated after intravenous and intracoronary administration in


BIMAKALIM

29

anesthetized pigs by Sassen et al. (51). By consecutive intravenous infusions, bimakalim (0.15, 0.30, 0.60, 1.20 mg/kg/min; for 10 min) dose-dependently decreased mean arterial blood pressure by up to 50%. This effect was entirely due to peripheral vasodilatation, since cardiac output did not change. Heart rate increased by up to 50%, while left ventricular end-diastolic pressure and stroke volume decreased dose-dependently. Left ventricular dP/dtmax was not affected. Although cardiac output did not change, bimakalim caused a redistribution of blood flow from the arteriovenous anastomoses to the capillary channels. Blood flow to the adrenals, small intestine, stomach, bladder, spleen, and brain increased, while renal blood flow decreased and blood flow to several muscle groups and skin were not altered. Vascular conductance was increased dose-dependently in all organs, except in the kidneys, where after the initial increase vascular conductance returned to the baseline, even at the highest dose. Particularly striking were the effects on the vasculature of the brain. At the highest dose used, bimakalim more than doubled the blood flow, while vascular conductance increased four-fold. Transmural myocardial blood flow increased slightly, which was entirely due to an increase in subepicardial blood flow. Myocardial O2-consumption and segment length shortening were not significantly affected. After consecutive 10-min intracoronary infusions (9.5, 19, 37.5 and 75 ng/kg/min) into the left anterior descending (LAD) coronary artery, mean arterial blood pressure was maintained after the two lowest doses, but decreased by up to 15% at the higher doses, whereas heart rate increased by up to 24%. Blood flow in the LAD coronary artery doubled at the highest dose with the subepicardium benefiting the most. Coronary venous O2-saturation increased dose-dependently, while myocardial O2-consumption was not affected by the drug. Van Woerkens et al. (57) evaluated the cardiovascular profile of bimakalim in conscious pigs with and without chronic left ventricular dysfunction and compared the effects of bimakalim with those of nicorandil. In normal conscious pigs, bimakalim (37.5–300 ng/kg/min) or nicorandil (10–80 ng/kg/min) increased cardiac output to a comparable extent, primarily due to an increase in heart rate. The mean arterial blood pressure decreased gradually with either drug due to a similar decrease in systemic vascular resistance. Left ventricular (LV) dP/dtmax increased with either drug. LV end-diastolic pressure remained unchanged after bimakalim and gradually decreased after nicorandil. In pigs with a 3- to 4-week-old myocardial infarction, the vasodilator responses to either bimakalim or nicorandil were similar to those in normal animals but increases in heart rate and in LV dP/dtmax were attenuated. Senior et al. (52) investigated the cardiac hemodynamic effects of bimakalim in 12 normal volunteers by echocardiography (ECHO)/Doppler in a placebo-controlled, randomized, double-blind, cross-over, dose-ranging study. A single oral dose (0.25–1 mg) was given at weekly intervals. Hemodynamic measurements were made at 0, 90, 120, and 240 min after drug intake and ECHO/Doppler was performed at 0 and 90 min. Reproducibility of the ECHO/Doppler study was assessed by comparing pretreatment baseline values in the four different treatment groups (placebo and 0.25, 0.5, and 1 mg) by analysis of variance (ANOVA), which showed no significant differences between groups for the left ventricular ejection fraction (LVEF). Doppler-derived stroke volume (SV), total peripheral resistance (TPR), and peak mitral early-to-late velocity ratio (PEV/PAV) were determined. ANOVA showed significant increases in LVEF and SV and decreases in TPR and PEV/PAV after bimakalim treatment. Heart rate increased dose-dependently, but sys-


30

P. E. PUDDU ET AL.

tolic and diastolic blood pressure (SBP, DBP) did not change after bimakalim. Despite vasodilatory headaches, none of the volunteers dropped from the study.

Electrophysiological Effects in Animal and Human Tissues Early studies suggested that the mechanism of the myocardial protective effect of KCOs, including bimakalim, is manifested by APD shortening and subsequent reduction of Ca2+ entry and cardiac work (9,14). This hypothesis was supported by studies in dogs where aprikalim-induced APD shortening occured concomitantly with infarct size reduction (59). With bimakalim, however, infarct size reduction in dogs and APD shortening have not been correlated (60). In human atrial tissue subjected to hypoxia and reoxygenation, bimakalim had cardioprotective effects at concentrations that produced neither APD shortening nor a negative inotropic effect (42). An in vivo dissociation between cardioprotective (infarction size) and electrophysiological (APD shortening and arrhythmias) effects of KCOs was also reported. Yao and Gross (60) used barbital-anesthetized, open-chest dogs to infuse bimakalim (0.1 to 10 mg/min) into the LAD coronary artery. The changes in coronary blood flow and in monophasic APD (MAPD) were used as indices of coronary vascular and myocyte KATP channel activities, respectively. In subsequent infarct studies, dogs were subjected to 60 min of LAD occlusion followed by 4 h of reperfusion. Bimakalim was infused at two doses (0.1 and 0.3 mg/min) that did not shorten MAPD during nonischemic conditions and one (3.0 mg) that shortened it markedly. Transmural myocardial blood flow was measured at 5 and 30 min after occlusion by the radioactive microsphere technique, and infarct size was determined at the end of 4 h of reperfusion by triphenyltetrazolium staining. The MAPD at 50% repolarization (MAPD50) was measured by an epicardial probe placed in the center of the ischemic area. Bimakalim had an approximately 10-fold higher affinity for the coronary vascular than the myocardial KATP channel (EC50 coronary, approximately 0.3 mg/min; EC50 myocyte, approximately 3.0 mg/min). All three doses of bimakalim (0.1, 0.3, and 3.0 mg/min) markedly reduced infarct size expressed as percent of the area at risk to nearly equal extent. Subsequently, it was found that the two higher doses of bimakalim markedly accelerated the ischemia-related shortening of the MAPD during the initial 5 min of occlusion; the 0.1 mg/min dose bimakalim had no significant effect. In addition, neither 0.1 nor 0.3 mg/min bimakalim increased the incidence of ventricular fibrillation during the 60 min of occlusion (0 of 7 and 0 of 8 dogs, respectively), whereas at 3.0 mg/min bimakalim had a profibrillatory effect (6 of 6) as compared with the control group (1 of 8). There were no significant differences between groups in systemic hemodynamics, myocardial oxygen demand, ischemic bed size, or collateral blood flow to the ischemic region. These results showed that the two smaller doses of bimakalim, administered to dogs only during the initial 10-min period of ischemia, are capable of markedly reducing myocardial infarct size to an extent equal to that produced by a higher profibrillatory dose. These data suggest that bimakalim and other KCOs may exert their cardioprotective effects partially by accelerating activation of KATP channels during early ischemia as evidenced by an enhanced rate of ischemic myocyte APD shortening. The results suggest also that other cellular mechanisms may mediate the cardioprotection observed after administration of a low dose of bimakalim (0.1 mg/min). At this dose bimakalim did not accelerate APD shortening, but reduced infarct size as effectively as at the two higher doses.


BIMAKALIM

31

We investigated the role of KATP channel activation and repolarization dispersion on the “border zone” arrhythmias induced by ischemia-reperfusion by assessing the effects of glibenclamide and bimakalim in an in vitro model of myocardial border zone (43). The electrophysiologic effects of 10 mM glibenclamide or 1 mM bimakalim were investigated in guinea pig ventricular strips at normal conditions (Normal Zone, NZ) and at conditions that simulated ischemia followed by reperfusion (Altered Zone, AZ). Glibenclamide prevented ischemia-induced APD shortening, reduced dispersion of APD 90% (APD90) between NZ and AZ, and concomitantly inhibited the border zone arrhythmias induced by an extrastimulus (ES). Bimakalim, which also reduced the APD90 dispersion due to differential AP shortening in normal and ischemic tissues, decreased the incidence of myocardial conduction block and favored border zone spontaneous arrhythmias. During reperfusion glibenclamide, unlike bimakalim, inhibited the ES-induced arrhythmias and reduced the incidence of the spontaneous ES. These results suggested that KATP blockade with glibenclamide may concomitantly protect the ischemic-reperfused myocardium from border zone arrhythmias and reduce APD90 dispersion between normal and ischemic regions. Conversely, KATP channel activation with bimakalim may modify the incidence of conduction block and exacerbate the ischemia-induced border zone arrhythmias. It is important to interprete these data (43) in relation to the concentration of bimakalim used (1 mM). Further studies are needed to assess whether lower and more clinically relevant concentrations of bimakalim (< l00 nM), which have been shown by our group to have significant cardioprotective properties in superfused hypoxic models (42), are still arrhythmogenic during ischemia. Based on data by Gautier et al. (19) and those of other investigators (48, 60) the probability to obtain myocardial protection with bimakalim at concentrations < 100 nM (as after 0.25–1 mg per os) is high, whereas the probability of proarrhythmic action is quite low (43). Finally, we investigated the in vitro effects of bimakalim (10 nM–10 mM) on transmembrane AP in human and guinea pig myocardial tissue (48) (Fig. 2). The frequency relation of human atrial APD90 shortening was determined (Fig. 3). A parallel study was performed with nicorandil at concentrations from 10 nM to 1 mM. Resting membrane potential and maximal upstroke velocity of AP were not modified by bimakalim at maximal concentration, whereas the AP amplitude was decreased in guinea pig preparations. APD90 was shortened in all tissues. EC50 of bimakalim for APD90 shortening at 37°C was 540 nM and 2.74 mM in atrial and ventricular human tissue, respectively and 8.55 mM and 890 nM in atrial and ventricular guinea pig tissue, respectively. In human atrial tissue at 31°C, the EC50 of bimakalim was 390 nM compared to 210 mM for nicoradil. Evidence was provided for species (human vs. guinea pig) and tissue (atrium vs. ventricle) differential AP sensitivity to bimakalim, thus confirming the previous data by Gautier et al. (19). Bimakalim was approximately 500-fold more potent than nicorandil in shortening human atrial APD90. At 10 mM of bimakalim this effect was use-dependent.

Effect on Contractility The negative inotropic effects of KCOs have been reviewed in detail, and it has been concluded that they mimic metabolic inhibition and that APD shortening is a concomitant hallmark (5). During ischemia, block of KATP channel (by glibenclamide) reduces APD shortening; this observation led to the proposal that KATP channel block has a potential


32

P. E. PUDDU ET AL.

FIG. 2. Concentration-dependent effects of bimakalim on APD90 of human and guinea pig cardiac tissues stimulated at 1000 msec of cycle lenght (37°C). Values expressed as percent changes of APD90 measured before bimakalim superfusion (mean ± S.E.M.). APD90 action potential duration at 90% of repolarization;EC50 drug concentration inducing 50% of the maximal effect. Note: there is a differential sensitivity of atrial and ventricular muscles to bimakalim effects and a tissue-relationship of APD90 shortening opposite between the two species. Modified from Rouet et al. (48).

antiarrhythmic role (43). The converse was also proposed in elegant in vitro studies whereby APD and contractility effects were investigated (9,39). It was suggested that KCOs may protect myocardium during ischemia by reducing metabolic requirements due to a direct negative inotropic effect. The cause and effect relationship between APD shortening and negative inotropism was not ascertained during these studies. Therefore, APD effects are in an uncertain equilibrium with inotropic effects with either KATP channel activation or block. Since inotropism is an ATP consuming process (5), it might not be safe to try to prevent arrhythmias by prolonging APD under conditions of ischemia. Such an at-


BIMAKALIM

33

FIG. 3. Use-dependence of action potential duration (APD) shortening by agents increasing gK via KATP channel activation. Data are compiled from experiments performed in guinea pig right ventricular papillary muscles (APD90, n = 6, u) by Kojima and Ban (33), in normal canine Purkinje fibers (APD95, n = 7, ¨) and canine Purkinje fibers surviving anterior descending coronary occlusion (APD95, n = 8, n) by Bril and Man (6), and (48) with right human atrial pectinate muscles (APD90, n = 7, p) by Rouet et al. Data were calculated at corresponding cycle lengths (CL) by subtraction from average initial values. At CL 1000 msec there was an expected species-related large variation of APD, which was 180 ± 9 msec in guinea-pig papillary muscles, 298 ± 5 and 345 ± 7 msec in normal and surviving-to-infarction canine Purkinje fibers, respectively, and 431 ± 52 msec in human atrial tissue. Accordingly, the amount of APD shortening after high concentration of structurally different KCOs greatly varied (from –30 to –60 msec), the largest effect being observed in tissues with the longest APD. Note normal use-dependence (see ref. 48) by all agents (with r 2 of polynomial fits ranging from 0.97 to 1.0). It is noteworthy that at the highest CL there was an overlap of APD shortening effects independent of agents, tissues, or species. Modified from Rouet et al. (48).

tempt may have a detrimental effect on contractility. The converse is also true: it would not be safe to reduce contractility (for the sake of ATP preservation); APD will be shortened to the extent that arrhythmias will be favored by a mechanism related to reentry due to APD dispersion (43). Our contribution was the demonstration of the concentration-related effects of bimakalim (10 nM–10 mM) on both APD and developed tension (DT) in normoxic human atrial preparations (42) (Fig. 4). The fact that EC50s were 492 and 273 nM, respectively, led us to conclude that APD shortening (rightward shifted curve) was not a necessary cause of negative inotropism in this preparation. Furthermore, bimakalim at either 10 or 100 nM had myocardial protectant effects during hypoxia, whereas glibenclamide at 1 mM was detrimental (42).

Cardioprotective Action Myocardial protection by KCOs has been repeatedly demonstrated in various animal and human tissue models. The common observation in all these studies was that KCOs


34

P. E. PUDDU ET AL.

FIG. 4. Concentration-effect relationships of bimakalim on developed tension (DT) and action potential duration at 90% of repolarization (APD90) of human atrial trabeculae stimulated at 1600 msec of CL (37°C). Values expressed as percent of baseline DT (mean ± S.E.M.). Modified from Picard et al. (42).

must be present prior to ischemia to be effective. The myocardial protective effects of KCOs are much less pronounced or are even absent if the compounds are added at the time of reperfusion. This finding limited further clinical development of most KCOs. Even the observation that KCOs might simulate PC did not facilitate clinical development since KCOs have to be present before an ischemic event in order to act. Gross et al. (22) investigated the effects of nicorandil, a nicotinamide nitrate with KATP channel-opening activity, in several models of ischemia-reperfusion injury in conscious and anesthetized dogs or isolated buffer-perfused rat hearts. In several models of reversible ischemic injury in dogs, nicorandil enhanced recovery of regional systolic shortening during reperfusion after a single episode of coronary artery occlusion. This beneficial action of nicorandil was not shared by the nitrovasodilator sodium nitroprusside, but was mimicked by bimakalim. In a dog model of irreversible ischemia-reperfusion injury,


BIMAKALIM

35

nicorandil produced a marked reduction of myocardial infarct size. At an equihypotensive dose, the calcium antagonist nifedipine had no significant effect, while bimakalim produced the same reduction in infarct size as nicorandil. In another study, Auchampach and Gross (2) examined the effect of bimakalim on myocardial infarct size (IS) in dogs. Dogs anesthetized with sodium barbital were subjected to a 90-min occlusion of the left circumflex coronary artery (LCX) followed by 5-h reperfusion. Bimakalim (3 mg/kg bolus followed by 0.1 mg/kg/min i.v.) was administered starting either 15 min before LCX occlusion and continued throughout the experiment or starting 5 min before and infused throughout reperfusion. A third group of dogs received i.v. vehicle (control) 15 min before LCX occlusion and throughout the remainder of the experiment. IS was determined by histochemical staining with triphenyltetrazolium, regional myocardial blood flow (RMBF) by the radioactive microsphere technique, and neutrophil migration by measurement of tissue myeloperoxidase (MPO) activity. Bimakalim reduced mean aortic blood pressure during the occlusion and reperfusion periods in the group of dogs that received the drug throughout the experiment. In addition, bimakalim increased LCX coronary artery blood flow and increased RMBF primarily during reperfusion in both drug-treated groups with the greatest increase to the subepicardial region. During occlusion, however, bimakalim had no effect on collateral blood flow to the ischemic region. In all three groups, LV mass, area at risk (AAR) mass, and percentage of the left ventricle at risk were similar. Compared to the control group, bimakalim reduced IS when administered during both the occlusion and reperfusion periods, although it had no beneficial effect when administered during the reperfusion period only. Rohmann et al. (46) studied the effect of bimakalim on IS, coronary blood flow, regional wall function, and oxygen consumption in open chest pigs subjected to a 60-min occlusion of a branch of the LAD coronary artery and to 2 h reperfusion. Five groups of 7 animals each were studied. In the first group, bimakalim infusion (3 mg/kg bolus over 5 min followed by 0.1 mg/kg/min) was started after 45 min of coronary occlusion and continued during 60 min of reperfusion (group A), while in the second group bimakalim infusion was started 15 min before occlusion and ended after 60 min of reperfusion (group B). In the third group bimakalim infusion was started 15 min before coronary occlusion, but stopped at the onset of ischemia (group C). In the fourth group, hydralazine was infused (0.2 mg/kg over 15 min) starting at 15 min before the occlusion and terminated at the start of occlusion. The chosen dose of hydralazine lowered arterial pressure to the same extent as bimakalim. A fifth group of animals received the vehicle and served as controls. At the end of the protocol, IS (as percent of risk region) was determined by incubating myocardium with p-nitrobluetetrazolium. Regional myocardial oxygen consumption (MVO2) was calculated as the product of coronary blood flow and the difference between the oxygen contents in the aorta and interventricular vein. Regional wall function was quantified with ultrasonic crystals aligned to measure wall thickening (% DWT). In all pigs in which bimakalim treatment was started prior to the 60-min coronary occlusion, IS was significantly reduced compared to IS in pigs subjected to 60 min of ischemia only; during reperfusion drug-induced potassium channel opening had no effect. Treatment with hydralazine did not reduce infarct size. Neither drug altered % DWT; however, all drugs reduced MVO2. These data are consistent with the hypothesis that bimakalim might reduce IS by activating cardiac KATP channels and not by unloading the heart because of its vasodilator effects.


36

P. E. PUDDU ET AL.

Yao and Gross (61) determined whether bimakalim can reduce the time required for ischemic PC to develop the protective effect and whether this effect is mediated by accelerating the rate of APD shortening. Dogs anesthetized with sodium barbital were subjected to 60 min of LAD coronary artery occlusion followed by 4 h of reperfusion. Ten minutes of PC was found to markedly reduce IS. Subsequently, it was observed that either 3 min of LAD coronary artery occlusion or a 3-min intracoronary infusion with 0.3 mg/min of bimakalim did not reduce IS. However, intracoronary infusion with bimakalim during the 3-min PC period reduced IS to a similar extent as ischemic PC. In addition, it was observed that bimakalim markedly accelerated the ischemia-induced shortening of APD during PC. This was the first demonstration that activation of KATP channels with a KCO reduces the time necessary to produce PC in anesthetized dogs. These data suggested also that KATP channel activation may produce this effect by enhancing the rate of ischemic myocardial APD shortening during PC. Mizumura et al. (38) studied the IS-reducing effect of PC and its association with adenosine release from the ischemic myocardium. They compared also the effects of ischemic PC with those of bimakalim. Dogs anesthetized with sodium barbital were subjected to 60 min of LAD coronary artery occlusion followed by 3 h of reperfusion. In the PC group, 5 min of LAD coronary artery occlusion followed by 10 min of reperfusion was elicited before the 60-min occlusion period. In two other groups, bimakalim 1 mg/kg bolus followed by a 0.05 mg/kg/min infusion or an equivalent volume of saline was administered intravenously 15 min before occlusion and continued until the time of reperfusion. In a final group, bimakalim was administered 10 min before reperfusion and continued until the end of the experiment. To measure the release of adenosine from the ischemic region, coronary venous blood samples were collected at various times during ischemia and after reperfusion, and the concentration of adenosine was measured. IS was determined by triphenyl tetrazolium chloride, and transmural myocardial blood flow was determined by radioactive microspheres. Transmural myeloperoxidase (MPO) activity was also measured. PC produced a marked reduction in IS, adenosine release at 5, 10, 15, and 30 min of the 3-h reperfusion period, and transmural MPO activity in the risk area. Similarly, pretreatment with bimakalim resulted in reductions in IS, adenosine release, and transmural MPO activity to an extent almost identical to that of PC. When bimakalim was administered 10 min before reperfusion, the drug also produced a significant reduction in IS and transmural MPO activity; however, no significant reduction in coronary venous adenosine concentrations was observed. There were no significant differences in collateral blood flow between groups. These results indicate that myocardial PC in the canine heart, produced by a short period of ischemia or a KCO, is not mediated by an increase in adenosine release. A significant reduction in the transmural MPO activity in the ischemic area appears also to result from KATP channel activation and may play a role, at least in part, in the observed reduction in IS, particularly when a KCO is administered just before reperfusion. Yao et al. (62) determined whether occupation of A1 receptors and/or the opening of KATP channels is involved in the time delay between the PC stimulus and the prolonged ischemic insult or the “memory” of PC to reduce IS. Barbital sodium-anesthetized dogs were subjected to 1 h of LAD coronary artery occlusion followed by 4 h of reperfusion. Ischemic PC was elicited by 10 min of LAD occlusion followed by 1 h reperfusion (1-h memory) before the 1-h occlusion period. Either adenosine (800 mg/min), bimakalim


BIMAKALIM

37

(3 mg/min), a combination of two lower doses of each agent (400 mg/min of adenosine and 0.3 mg/min of bimakalim), or an equivalent volume of saline were infused into the LAD for 10 min followed by a 1-h drug-free period before the 1-h ischemic insult. In another senes, glibenclamide, 8-cyclopentyl-1,3-dipropylxanthine (DPCPX: a selective A1-receptor blocker), or PD-115199 (a nonselective adenosine-receptor antagonist) were administered 50 min after ischemic PC (10 min before the 1-h occlusion penod). IS was expressed as a percentage of the area at risk. PC with 1 h of reperfusion resulted in a marked reduction in IS. Administration of adenosine or bimakalim followed by a 1-h drug-free period had no effect on IS; however, simultaneous administration of adenosine and bimakalim resulted in a marked decrease in IS. One h after ischemic PC, glibenclamide blocked the protective effect of ischemic PC, whereas DPCPX or PD-115199 did not affect it. These results provided evidence that the opening of myocardial KATP channels may play an important role in the memory of ischemic PC in the canine heart and suggest also that adenosine and the KATP channel may have a synergistic interaction that is important for the memory phase of PC. Gross et al. (24) addressed the issue of whether KATP channel activation protects hearts through adenosine A1-receptor activation by determining the effect of DPCPX on the cardioprotective activity of bimakalim. In isolated rat hearts subjected to 25 min of global ischemia and 30 min of reperfusion, bimakalim significantly reduced lactate dehydrogenase release and improved postischemic recovery of contractile function. Bimakalim increased the time to the onset of ischemic contracture, while DPCPX (10 mM) had no effect. At the same concentration, DPCPX abolished the bradycardic and cardioprotective effects of the adenosine A1-receptor agonist, (R )-(–)-N6-(2-phenylisopropyl)adenosine. DPCPX alone had no effect on the severity of ischemia/reperfusion damage. Glibenclamide completely abolished the cardioprotective effects of bimakalim. In anesthetized dogs bimakalim (1 mg/kg, intracoronary, given over 4 periods of 5 minutes, interspersed with 10-min drug-free periods before a 60-min occlusion and 3-h reperfusion period, significantly reduced IS. DPCPX had no effect on the infarction-sparing activity of bimakalim. The protective effect of bimakalim was not accompanied by marked hemodynamic changes or by changes in regional myocardial blood flow. These results suggest that the cardioprotective effect of bimakalim does not depend on adenosine A1-receptor activation in the rat or dog models of ischemia. Baker et al. (3) determined whether increased tolerance to ischemia in immature rabbits compared with age-matched normoxic rabbits was due to an alteration in the KATP channel and whether increased KATP channel activation was associated with increases in intracellular lactate. Isolated immature rabbit hearts (7–10 days old) were perfused with bicarbonate buffer at 39°C and a constant pressure. Saline-filled latex balloons were placed in the left and right ventricles for measurement of developed pressure. Bimakalim or glibenclamide were added 15 min before a global ischemic period of 18 minutes, followed by 35 min of reperfusion. Rabbits raised from birth in hypoxic conditions displayed significantly enhanced recovery of developed pressure. In normoxic or hypoxic hearts, the right ventricle was more tolerant to ischemia than the left ventricle. Bimakalim (1 mM) increased the recovery of left ventricular developed pressure in normoxic hearts to values not different from those of hypoxic controls and slightly increased developed pressure in hypoxic hearts. Glibenclamide (3 mM) abolished the cardioprotective effect of hypoxia. Constant-flow studies indicated that the effects of bimakalim and glibenclamide were in-


38

P. E. PUDDU ET AL.

dependent of their actions on coronary flow. Ventricular lactate and lactate dehydrogenase concentrations were elevated in hypoxic vs. normoxic control hearts. The increased tolerance to ischemia, exhibited by chronically hypoxic rabbit hearts, is likely to be associated with increased activation of the KATP channel. This increased KATP activity may be the result of increased intracellular concentrations of lactate. Mei et al. (37) determined the role of adenosine in mediating the cardioprotective effects of PC produced by 5 min of ischemia, hypoxia, or by a 5-min intracoronary (i.c.) infusion of bimakalim (1 mg/min). A single microdialysis probe was implanted into the midwall of the ischemic area to sample interstitial fluid adenosine and its breakdown products during the PC stimulus, prolonged occlusion (60 min), and the first 30 min of the reperfusion (3-h) period. Ischemic or hypoxic PC or pretreatment with bimakalim significantly reduced infarct size compared to control values. Both ischemic and hypoxic PCs produced similar and significant increases of adenosine concentration in dialysate, which persisted during the brief 10-min reperfusion period following PC; however, i.c. bimakalim significantly decreased adenosine concentration in dialysate, an effect that persisted during the 10-min drug-free period. All three PC protocols resulted in similar decreases in adenosine, inosine, and uric acid concentrations throughout the prolonged ischemic period. These results suggest that an increase in interstitial adenosine may be necessary for the initiation of the protective effect of PC and that an increase in adenosine concentration in dialysate is not required for the cardioprotective effect of a KATP channel opener. In human atrial myocardium, we investigated whether bimakalim, at concentrations devoid of the negative inotropic and APD-shortening effects, might exhibit myocardial protection after hypoxia and reoxygenation (42). The recovery of human atrial trabeculae contractility subjected to either short-duration (5 min) or long-duration (60 min) and severe (high pacing rate) hypoxia followed by reoxygenation was assessed by challenging with dobutamine. Treated preparations were exposed to 10 or 100 nM bimakalim, 1 mM glibenclamide, or both prior to hypoxia. Variations of isometric-developed tension (% DT) or APD90 were studied. At concentrations under 100 nM, bimakalim showed no negative inotropic effects and did not significantly modify APD90 in either normoxia or hypoxia. In the short-duration hypoxia protocol using preparations treated with bimakalim, dobutamine-induced % DT increase was significantly higher than in controls (Fig. 5) and similar to that observed in normoxia. This bimakalim effect was blocked by glibenclamide. In the long-duration hypoxia protocol (Fig. 6), % DT after dobutamine was 50% of what was observed in normoxic preparations. In preparations treated with bimakalim, dobutamine % DT was more than two-fold above controls, whereas in the glibenclamide group dobutamine effect remained 50% of that observed in normoxia. In conclusion, exposure to hypoxia (either short or long lasting) and reoxygenation affected contractility of human atrial myocardium with a pronounced reduction of the positive inotropic action of dobutamine. Pretreatment with bimakalim restored the response to a level observed with dobutamine at normoxic conditions. Glibenclamide blocked the effect of bimakalim or further impaired the response to dobutamine. Evidence for the protective effects of bimakalim on human myocardial contractile function at the conditions of hypoxia-reoxygenation is presented here. This effect was observed at concentrations at which bimakalim had no negative inotropic or APD-shortening effects.


BIMAKALIM

39

FIG. 5. Human atrial contractility changes during a short-duration hypoxia protocol in control preparations (n = 6) and 2 groups treated with bimakalim either 10 (n = 6) or 100 nM (n = 6) just before the second 5-min. hypoxic period. Values are expressed as percent of baseline DT (mean ± S.E.M.). P values refer to analysis of variance (ANOVA) for repeated measures with analysis of covariance. Note the significant P values for time, group ´ time and no significant differences for group and absolute baseline DT. Also, note a positive inotropic action of dobutamine in pretreated groups (also present at wash-out). Pretreatment with glibenclamide 1 mM in another 4 preparations also given bimakalim 100 nM (data not shown) abolished the difference from controls in the inotropic response to dobutamine. Modified from Picard et al. (42).

Cardioplegia Cardioplegic arrest provides essential cardioprotection during open heart surgery and the advantages of continuous warm blood cardioplegia have been advanced to minimize ischemia and to avoid risks related to hypothermia; however, postoperative ventricular dysfunction and injury due to prolonged cardiac arrest led to the development of additional cardioprotective strategies (10). Experimental studies carried out in rabbits (44), pigs (13), guinea pigs (54,55), and rats (45) have demonstrated that KCOs improve myocardial preservation. Stowe et a1. (53) administered bimakalim alone or with 2,3-butanedione monoxime (BDM) a reversible uncoupler of contractility, to protect myocardial function during 1 day of hypothermia. Left ventricular pressure (LVP), coronary flow (CF), percent O2 extraction (%O2E), and cardiac efficiency were measured in 96 isolated, perfused guinea pig hearts divided into 7 groups: 1) cold control (no drugs); 2) BDM; 3) bimakalim; 4) BDM + bimakalim; 5) BDM + glibenclamide; 6) BDM + bimakalim + glibenclamide; and 7) time control (6-h warm perfusion only). Drugs were given before, during, and initially after 22 h of low CF at 3.8°C. At 26 h (cold groups) or 4 h (warm group), LVP was similar for time control and BDM + bimakalim groups, lower and equivalent in the BDM and


40

P. E. PUDDU ET AL.

FIG. 6. Human atrial contractility changes during a long-duration hypoxia protocol. Values are expressed as percent of baseline DT (mean ± S.E.M.). P values refer to ANOVA for repeated measures. Note the significant P values for time, group ´ time interaction and group. Also, note a positive inotropic action of dobutamine in both bimakalim (100 nM) (42) and the dihydropyridine Ca2+-entry blocker, felodipine (100 nM) (31) pretreated groups. Impaired responsiveness to dobutamine was present in the glibenclamide (1 mM; data not shown) pretreated bimakalim group (42).

BDM + bimakalim + glibenclamide groups, but LVP was higher than in the bimakalim group and lowest in the cold control group. In addition, basal CF, %O2E, and cardiac efficiency returned to control values only in the BDM + bimakalim group. Epinephrine increased LVP to that of the control group only in the BDM + bimakalim group after hypothermia. Aso, CF increases with adenosine, 5-hydroxytryptamine, and nitroprusside after hypothermia were similar to that of the control group only in the BDM + bimakalim group. All of the effects of bimakalim were reversed by glibenclamide. These results indicate that bimakalim, given with BDM, effectively preserves myocardial function and metabolism as well as inotropic and vasodilatory reserve during long-term hypothermic preservation. Addition of bimakalim or possibly other KCOs might prove useful in preserving cardiac function during cardiac transplantation. Lathrop et al. (35) determined the concentration-dependent effect of bimakalim on transmembrane APD changes induced by mild (27°C) or moderate (20°C) hypothermia in isolated guinea pig ventricular muscle in the presence or absence of bimakalimi (0.1–30 mM). Hypothermia alone increased APD and depolarized the resting membrane potential (RMP). At 37°C, bimakalim (0.1–10 mM) shortened APD in a concentration-dependent fashion; this effect was blocked by glibenclamide. RMP was hyperpolarized by bimakalim. More bimakalim was required to shorten APs during mild and moderate hypothermia than at 37°C. Bimakalim hyperpolarized RMP toward drug-free values obtained at 37°C. Bimakalim reversed also hypothermia-induced AP lengthening and tended to reverse hypothermia-induced decrease in RMP.


BIMAKALIM

41

To study the effects of cardioplegia in human myocardium, we obtained human atrial trabeculae from adult patients undergoing open heart surgery. The preparations were subjected to 60 min hypoxia at high stimulation rate either in Tyrode’s solution or in St. Thomas Hospital solution (STHS) without additives or with 100 nM bimakalim or 1 mM glibenclamide. Hypoxia was followed by 60 min of reoxygenation and 15 min of positive inotropic stimulation with 1 mM dobutamine at 37°C. Atrial developed tension (DT) was reduced by hypoxia. It recovered incompletely after reoxygenation, whereas dobutamine restored contractility to 74% of the basal values. STHS with or without bimakalim improved DT after reoxygenation and dobutamine, whereas glibenclamide inhibited these protective effects. After reoxygenation at high-pacing rate, the protective effect of bimakalim disappeared but recovered during dobutamine superfusion. We concluded that KATP channels are likely to be involved in the cardioprotective effects of cardioplegia in human atrial trabeculae and that KATP channel activation by bimakalim, used as an additive to cardioplegia, enhanced this protection (Monti et al., unpublished). Since most studies have found that the effect of KCOs as cardioplegia additives was abolished by glibenclamide, KCOs were thought to have a mode of action that is distinct from that of cardioplegia and that does not involve an enhanced sarcolemmal K+ current (25). Moreover, since arrested hearts or preparations were electrically quiescent, APD shortening could not account for protection, which further points toward a putative intracellular site of action for KCOs (25).

THE POTENTIAL ROLE OF BIMAKALIM IN THERAPY Currently available KCOs, including aprikalim, bimakalim, cromakalim, emakalim, nicorandil, and pinacidil, display a high affinity for potassium channels of vascular smooth muscle. Vasodilation and a reduction in systemic vascular resistance are their prominent pharmacological effects. Coronary and cerebral arteries are highly sensitive to KCO-induced dilation. Apart from the treatment of hypertension, potassium channel activators appear to have therapeutic potential in coronary heart disease. They reduce cardiac afterload, increase native and collateral coronary blood flow, and reduce the size of experimental myocardial infarcts. This last effect cannot be satisfactorily explained entirely by hemodynamic or coronary vascular actions of KCOs and, as discussed above, a cardioprotective effect is postulated. While currently existing KCOs mimic PC in some respects, it remains unknown whether the cardioprotective potency and selectivity have been optimized. Indeed, there is virtually no correlation between vasorelaxant activity and glibenclamide-reversible cardioprotection and very little correlation between APD shortening and glibenclamide-reversible cardioprotection (25). A critical issue concerning potential clinical applications of KCOs in general, and bimakalim, in particular, relates to the necessity of giving the drug before acute myocardial ischemia. Furthermore, the ischemia has to be large enough to see any difference from controls. When considered at the clinical level, these concepts render phase II clinical trials with most KCOs in patients with chronic angina pectoris extremely difficult and likely to produce inconclusive results. Pharmaceutical companies can not be expected to undertake large and extremely expensive phase III trials in the prevention of acute myocardial infarction, which may prove the existence of favorable cardioprotective effects of


42

P. E. PUDDU ET AL.

KCOs in humans. Secondary endpoints, such as acute myocardial ischemia during percutaneous coronary angioplasty or coronary artery surgery (with or without cardioplegia), may not be regarded as substitutes of phase III trials in coronary patients whereby the full potential of KCOs as cardioprotectants could be tested. An area where myocardial protection with KCOs, in general, and bimakalim, in particular, might be tested is cardiac surgery, since in the surgical setting the compounds are given before ischemia. The studies in cardioplegia, described above, suggest another possible indication for bimakalim and other KCOs — preservation of hearts for transplantation. A neglected area where future clinical indications for KCOs might be found is congestive heart failure (7, 21) or composite endpoints that include also stroke (25) and peripheral vascular disease (7). It is suggested that, only congestive heart failure patients with an ischemic etiology should be considered for such studies and that the vasorelaxant KCOs should be excluded from trials since they may interfere with the effects of ACE inhibitors frequently used by these patients. Bimakalim at very low dose might represent a suitable candidate.

PERSPECTIVES Ongoing investigations are directed toward understanding the role of mito-KATP channels in normal and ischemic cardiac function. The current understanding is summarized here. In high-work states, ATP production occurs at rates that will sharply reduce the mitochondrial membrane potential. This will reduce K+ influx by diffusion with the result that matrix volume will contract (due to efflux through K+/H+ antiporter) and volume of the intermembrane space (IMS) will expand. According to our hypothesis, the signal for high work sends simultaneously a signal to open mito-KATP channels. Thus, an added K+ conductance will compensate for the lower driving force so that matrix and IMS volumes remain constant. This is important because, as Saks et al. have shown, the correct organization of mitochondrial creatine phosphokinase (mito-CPK) is essential for the production of ATP at a high rate (47,49,50). Myofibrillar creatine phosphokinase (cyto-CPK) efficiently converts ADP to creatine, which is present at high concentrations and is competent as a carrier of the “low-energy” charge. Creatine crosses the outer membrane and reacts with mito-CPK. Mito-CPK is self-associated (probably as a dimer) and is functionally associated also with the ATP/ADP translocator. Thus, creatine is phosphorylated by mito-CPK with simultaneous production of ADP, which is taken up immediately into the matrix and phosphorylated. Phosphocreatine, simultaneously produced from exchanged ATP, shuttles back to the myofibril. In this manner, the ADP produced by the myofibrils is efficiently transmitted to the mitochondrial matrix. If the IMS was allowed to expand during high ATP-production, the mito-CPK assembly would become dissociated precisely when it is most needed. Thus, mito-KATP is postulated to open in order to preserve the architecture of the IMS. To approach this issue experimentally, we examined increased work imposed at maximal inotropic stimulation, a situation where mitochondrial matrix volume may well contract. In 12 human normoxic atrial preparations (Fig. 7) subjected consecutively at 37°C to two incremental superfusions with the b-agonist dobutamine (from 1 nM–10 mM, each for 15 min) dobutamine, as expected, increased work (% developed tension). Pretreatment with 1 mM glibenclamide, a nonspecific blocker of mito-KATP, and with 300 mM


BIMAKALIM

43

FIG. 7. Requirement for open mito-KATP. positive inotropy: bar graph contains contractility data from 12 human atrial strips, three for each treatment. In each condition, isometric tension was stimulated by 10 mM dobutamine. C = control (100%); no other treatment. Developed tension increased 60% after superfusion of dobutamine, which is typical of this preparation. GLY preparation pretreated with 1 mM glyburide, a nonspecific blocker of KATP channels. Glyburide caused the positive inotropic effect of dobutamine to drop by 30%. 5-HD: preparation pretreated with 300 mM 5-OH-decanoate, a specific blocker of the mitochondrial KATP channel. 5-HD caused the positive inotropic effect of dobutamine to drop by 60%. BIM: preparation pretreated with 100 nM bimakalim. Bimakalim had no significant effect on the positive inotropic effect of dobutamine, suggesting that mito-KATP already opened in response to dobutamine (Puddu and Garlid, unpublished).

5-hydroxydecanoic acid, a specific blocker of mito-KATP (18) significantly reduced this effect by 30%. At 100 nM, bimakalim had no significant effects, suggesting that mito-KATP was indeed opened in response to the dobutamine. These data support the view that blockade of mito-KATP may interfere with the capacity of mitochondria to sustain high-work states. Ischemia presents another situation in which membrane potential will drop, in this case due to anoxia, with consequent matrix contraction, IMS expansion, and disruption of the mito-CPK assembly. We hypothesize that in this state opening of mito-KATP to preserve this architecture will help the cell to preserve ATP. ATP hydrolysis by mitochondria will be retarded by mito-CPK because the kinetics of the enzyme complex are highly unfavorable for the reverse reaction. In this way, the entry of ATP into the matrix is retarded, and the phosphorylation potential seen by the ATPase and the ATP hydrolysis are reduced. This hypothesis is currently being examined in ischemic hearts. Whether diverse stimuli (adenosine, bradykinin, and opioids), acting as triggers, and elevated cytosolic calcium and/or PKC activation, acting as mediators, impinge upon mito-KATP as an end-effector, are subjects of intensive investigation (23). Investigation of mito-KATP pharmacology (18,29,30) will add a new dimension to the understanding of the mechanism of action of bimakalim. If the central role of mito-KATP in cardioprotection is revealed by these studies, it will be important to search for a mitochondrium-specific agent that may be cytoprotective in clinical situations associated with ischemia.


44

P. E. PUDDU ET AL.

SUMMARY This review focuses on selected aspects of the pharmacology of bimakalim, including pharmacokinetics, pharmacodynamics, mode of action, and hemodynamic and electrophysiological effects in animal preparations, human tissue, and normal volunteers. This review emphasizes cardioprotective action of bimakalim, its mechanism of action, and the therapeutic potential. Data are pointing toward an intracellular site of action of bimakalim, probably at the mito-KATP. Aknowledgments. We express our gratitude to Dr. Wagner from Merck Laboratories, Darmstadt, for the gift of bimakalim used in expenments performed in our laboratory in Rome. The expert advise of Icilio Cavero, PhD, from Rhône-Poulenc Laboratories, Paris, was also appreciated along with his kindness in reading the drafts of this article. This work was supported by MURST projects 147/97, 611/98, and 1017/98 (Rome “La Sapienza”) and by CardioRicerca, Rome, Italy.

REFERENCES 1. Ashford MLJ. Sulphonylureas: A receptor and a potassium channel? In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;161–174. 2. Auchampach JA, Gross GJ. Reduction in myocardial infarct size by the new potassium channel opener bimakalim. J Cardiovasc Pharmacol 1994;23:554–561. 3. Baker JE, Curry BD, Olinger GN, Gross GJ. Increased tolerance of the chronically hypoxic immature heart to ischemia. Contribution of the KATP channel. Circulation 1997;95:1278–l285. 4. Berdeaux A, Lablanche JM. Potassium channel openers in coronary heart disease. In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;293–303. 5. Boyett MR, Harrison SM, White E. Potassium channels and cardiac contraction. In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;69–91. 6. Bril A, Man RYK. Effects of the potassium channel activator, BRL 34915, on the action potential characteristics of canine cardiac Purkinje fibers. J Pharmacol Exp Ther 1990;253:1090–l096. 7. Cavero I, Guillon JM. Pharmacological profile of potassium channel openers in the vasculature. In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;193–223. 8. Cavero I, Premmereur J. ATP sensitive potassium channel openers are of potential benefit in ischaemic heart disease. Cardiovasc Res 1994;28:32–33. 9. Cole WC, McPherson CD, Sontag D. ATP-regulated K+ channels protect the myocardium against ischemia/reperfusion damage. Circ Res 1991;69:571–581. 10. Damiano RJ. The electrophysiology of ischemia and cardioplegia: Implications for myocardial protection. J Card Surg 1995;10:445–453. 11. De Mey C, Erb K, Schroeter V, Belz GG. Differentiation of inodilatory responses by non-invasive measures of cardiovascular performance in healthy man. Int J Clin Pharmacol Ther 1996;34:525–532. 12. de Peyer JE, Lues I, Gericke R, Haeusler G. Characterization of a novel K+ channel activator, EMD 52692, in electrophysiological and pharmacological experiments. Pflugers Arch 1989;414(Suppl 1):S191. 13. Dorman BH, Cavallo MJ, Hinton RB, Roy RC, Spinale FG. Preservation of myocyte contractile function after hypothermic, hyperkalemic cardioplegic arrest with 2,3-butanedione monoxime. J Thorac Cardiovasc Surg 1996;111:621–629. 14. Escande D, Cavero I. K+ channel openers and “natural” cardioprotection. Trends Pharmacol Sci 1992;13:269–272. 15. Escande D, Cavero I. Potassium channel openers in the heart. In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;225–244. 16. Fosset M, De Weille JR, Green RD, Schmid-Antomarchi H, Lazdunski M. Antidiabetic sulfonylureas control action potential properties in heart cells via high affinity receptors that are linked to ATP-dependent K+ channels. J Biol Chem 1988;263:7933–7936.


BIMAKALIM

45

17. Garlid KD, Paucek P, Yarov-Yarovoy V, et al. Cardioprotective effects of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels. Possible mechanism of cardioprotection. Circ Res 1997;81:1072–1082. 18. Garlid KD, Paucek P, Yaro. v-Yarovoy V, Sun X, Schindler PA. The mitochondrial KATP channel as a receptor for potassium channel openers. J Biol Chem 1996;271:8796–8799. 19. Gautier P, Bertrand JP, Guiraudou P. Effects of SR 44866, a potassium channel opener, on action potentials of rabbit, guinea pig, and human heart fibers. J Cardiovasc Pharmacol 1991;17:692–700. 20. Gericke R, Lues I, de Peyer JE, Haeusler G. Electrophysiological and pharmacological characterization of EMD 52692, a new potassium channel activator. Naunyn-Schmiedeberg’s Arch Pharmacol 1989;339: R62. 21. Gopalakrishnan M, Janis RA, Triggle DJ. ATP-sensitive K+ channels: Pharmacologic properties, regulation, and therapeutic potential. Drug Develop Res 1993;28:95–l27. 22. Gross GJ, Auchampach JA, Maruyama M, Warltier DC, Pieper GM. Cardioprotective effects of nicorandil. J Cardiovasc Pharmacol 1992;20(Suppl 3):S22–S28. 23. Gross GJ, Fryer RM. Sarcolemmal versus mitochondrial ATP-sensitive K+ channels and myocardial preconditioning. Circ Res 1999;84:973–979. 24. Gross GJ, Mei DA, Sleph PG, Grover GJ. Adenosine A1 receptor blockade does not abolish the cardioprotective effects of the adenosine triphosphate-sensitive potassium channel opener bimakalim. J Pharmacol Exp Ther 1997;280:533–540. 25. Grover GJ. Pharmacology of ATP-sensitive potassium channel (KATP) openers in models of myocardial ischemia and reperfusion. J Physiol Pharmacol 1997;75:309–315. 26. Grover GJ, D’Alonzo AJ, Hess T, Sleph PG, Darbenzio RB. Glyburide-reversible cardioprotective effects of BMS-180448 is independent of action potential shortening. Cardiovasc Res 1995;30:731–738. 27. Haverkamp W, Borggrefe M, Breithardt G. Electrophysiologic effects of potassium channel openers. Cardiovasc Drugs Ther 1995;9:195–202. 28. Hearse DJ. Activation of ATP-sensitive potassium channels: A novel pharmacological approach to myocardial protection? Cardiovasc Res 1995;30:1–17. 29. Holmuhamedov EL, Jovanovic S, Dzeja PP, Jovanovic A, Terzic A. Mitochondrial ATP-sensitive K+ channels modulate cardiac mitochondrial function. Am J Physiol 1998;275:H1567–H1576. 30. Hu H, Sato T, Seharaseyon J, et al. Pharmacological and histochemical distinctions between molecularly defined sarcolemmal KATP channels and native cardiac mitochondrial KATP channels. Mol Pharmacol 1999;55:1000–1005. 31. Iwashiro K, Criniti A, Sinatra R, et al. Felodipine protects human atrial muscle from hypoxia-reoxygenation dysfunction: A force-frequency relationship study in an in vitro model of stunning. Int J Cardiol 1997;62:107–132. 32. Jaburek M, Yarov-Yarovoy V, Paucek P, Garlid KD. State-dependent inhibition of the mitochondrial KATP channel by glyburide and 5-hydroxydecanoate. J Biol Chem 1998;273:13578–13582. 33. Kojima M, Ban T. Nicorandil shortens action potential duration and antagonists the reduction of Vmax by lidocaine but not by disopyramide in guinea-pig papillary muscles. Naunyn-Schmiedebergs Arch Pharmacol 1988;337:203–212. 34. Kovacs RJ. Potassium channel openers in hypertension. In: Escande D, Standen N, eds. K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;285–291. 35. Lathrop DA, Contney SJ, Bosnjak ZJ, Stowe DF. Reversal of hypothermia-induced action potential lengthening by the KATP channel agonist bimakalim in isolated guinea pig ventricular muscle. Gen Pharmacol 1998;31:125–131. 36. McCullough JR, Normandin DE, Conder ML, Sleph PG, Dzwonczyk S, Grover GJ. Specific block of the anti-ischemic actions of cromakalim by sodium 5-hydroxydecanoate. Circ Res 1991;69:949–958. 37. Mei DA, Nithipatikom K, Lasley RD, Gross GJ. Myocardial preconditioning produced by ischemia, hypoxia, and a KATP channel opener: Effects on interstitial adenosine in dogs. J Mol Cell Cardiol 1998;30:1225–1236. 38. Mizurnura T, Nithipatikom K, Gross GJ. Bimakalim, an ATP-sensitive potassium channel opener, mimics the effects of ischemic preconditioning to reduce infarct size, adenosine release, and neutrophil function in dogs. Circulation 1995;92:1236–1245. 39. Nakaya H, Takeda Y, Tohse N, Kanno M. Effects of ATP-sensitive K+ channel blockers on the action potential shortening in hypoxic and ischaemic myocardium. Br J Pharmacol 1991;103:1019–1026. 40. Nichols CG, Ripoll C, Lederer WJ. ATP-sensitive potassium channel modulation of the guinea pig ventricular action potential and contraction. Circ Res 1991;68:280–287.


46

P. E. PUDDU ET AL.

41. Paucek P, Mironova G, Mahdi F, Beavis AD, Woldengiorgis G, Garlid KD. Reconstitution and partial purification of the glibenclamide-sensitive, ATP-dependent K+ channel from rat liver and beef heart mitochondria. J Biol Chem 1992;267:26062–26069. 42. Picard S, Criniti A, Iwashiro K, et al. Protection of human myocardium in vitro by KATP activation with low concentrations of bimakalim. J Cardiovasc Pharmacol 1999;34:162–172. 43. Picard S, Rouet R, Ducouret P, et al. KATP channels and ‘border zone’ arrhythmias: Role of the repolarization dispersion between normal and ischaemic ventricular regions. Br J Pharmacol 1999;127:1687–1695. 44. Pignac J, Bourgouin J, Dumont L. Cold cardioplegia and the K+ channel modulator aprikalim (RP 52891): Improved cardioprotection in isolated ischemic rabbit hearts. Can J Physiol Pharmacol 1994;72:126–132. 45. Qiu Y, Galinanes M, Hearse DJ. Protective effect of nicorandil as an additive to the solution for continuous warm cardioplegia. J Thorac Cardiovasc Surg 1995;110:1063–1072. 46. Rohmann S, Weygandt H, Schelling P, et al. Effect of bimakalim (EMD 52692), an opener of ATP sensitive potassium channels, on infarct size, coronary blood flow, regional wall function, and oxygen consumption in swine. Cardiovasc Res 1994;28:858–863. 47. Rossi A, Kay L, Saks VA. Early ischemia-induced alterations of the outer mitochondrial membrane and the intermembrane space: A potential cause for altered energy transfer in cardiac muscle? Mol Cell Biochem 1998;184:401–408. 48. Rouet R, Picard S, Criniti A, et al. Effects of bimakalim on human cardiac action potentials: Comparison with guinea pig and nicorandil and use-dependent study. J Cardiovasc Pharmacol 1999;33:255–263. 49. Saks VA, Dos Santos P, Gellerich FN, Diolez P. Quantitative studies of enzyme-substrate compartmentation, functional coupling and metabolic channeling in muscle cells. Mol Cell Biochem 1998;184:291–307. 50. Saks VA, Veksler VI, Kuznetsov AV, et al. Permeabilized cell and skinned fiber techniques in studies of mitochondrial function in vivo. Mol Cell Biochem 1998;184:81–100. 51. Sassen LM, Duncker DJ, Gho BC, Diekmann HW, Verdouw PD. Haemodynamic profile of the potassium channel activator EMD 52692 in anaesthetized pigs. Br J Pharmacol 1990;101:605–614. 52. Senior R, Buchner-Moell D, Raftery E, Lahiri A. Potent hemodynamic effects of bimakalim, a new potassium channel opener, in humans. J Cardiovasc Pharmacol 1993;22:717–721. 53. Stowe DF, Graf BM, Fujita S, Gross GJ. One-day cold perfusion of bimakalim and butanedione monoxime restores ex situ cardiac function. Am J Physiol 1996;271:H1884–H1892. 54. Sugimoto S, Puddu PE, Monti F, et al. Activation of ATP-dependent K+ channels enhances myocardial protection due to cold high potassium cardioplegia: A force-frequency relationship study. J Mol Cell Cardiol 1995;27:1867–1881. 55. Sugimoto S, Puddu PE, Monti F, Schiariti M, Campa PP, Marino B. Pretreatment with the adenosine triphosphate-sensitive potassium channel opener nicorandil and improved myocardial protection during high-potassium cardioplegic hypoxia. J Thorac Cardiovasc Surg 1994;108:455–466. 56. Tomai F, Crea F, Chiariello L, Gioffrè PA. Ischemic preconditioning in humans: models, mediators, and clinical relevance. Circulation 1999;100:559–563. 57. van Woerkens LJ, Baas NRA, van der Giessen WJ, Verdouw PD. Cardiovascular effects of the novel potassium channel opener bimakalim in conscious pigs after myocardial infarction: a comparative study with nicorandil. Cardiovasc Drugs Ther 1992;6:409–417. 58. Williams AJ. Potassium channel openers: Clinical aspects. In: Weston AH, Hamilton TC, eds. Potassium Channel Modulators: Pharmacological, Molecular and Clinical Aspects. Oxford: Blackwell Scientific Publications, 1992;486–501. 59. Yao Z, Cavero I, Gross GJ. Activation of cardiac KATP channels: An endogenous protective mechanism during repetitive ischemia. Am J Physiol 1993;264:H495–H504. 60. Yao Z, Gross G. Effects of the KATP channel opener bimakalim on coronary blood flow, monophasic action potential duration, and infarct size in dogs. Circulation 1994; 89: 1769–1775. 61. Yao Z, Gross GJ. Activation of ATP-sensitive potassium channels lowers threshold for ischemic preconditioning in dogs. Am J Physiol 1994;267:H1888–H1894. 62. Yao Z, Mizumura T, Mei DA, Gross GJ. KATP channels and memory of ischemic preconditioning in dogs: Synergism between adenosine and KATP channels. Am J Physiol 1997;272:H334–H342.
