Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Bepridil, a highly effective antiarrhythmic agent with antianginal properties, is primarily classified as a calcium antagonist (Prystowsky, 1985; Hollingshead et al., 1992). In addition to this class IV action, two other actions have been reported: 1) a lidocaine-like fast kinetic block of inward sodium current (class Ib; Yatani et al., 1986; Nawada et al., 1995; Sato et al., 1996) and 2) a block of delayed rectifier K⁺ current and prolongation of QT interval (class III; Berger et al., 1989; Prystowsky, 1992). Such electrophysiological properties of bepridil may, at least in part, account for the mechanisms of antiarrhythmic action. Bepridil has been reported to be useful in the management of ventricular tachycardia and fibrillation encountered during ischemia (Kane and Winslow, 1980; Marshall and Muir, 1981; Lynch et al., 1985). In ischemic myocardium, ATP-sensitive K⁺ (K_ATP) channels are activated as a result of a decrease in the intracellular ATP concentration ([ATP]_i) and shorten the action potential duration (APD; Noma, 1983). Furthermore, it is reasonable to speculate that decreased [ATP]_i and decreased Na⁺,K⁺- ATPase activity depress the function of the Na⁺-pump, increase the intracellular Na⁺ concentration, and activate Na⁺-activated K⁺ (K_Na) channels (Kameyama et al., 1984; Luk and Carmeliet, 1990). Although, the physiological and pathophysiological implications of K_Na channels are still poorly understood, the activation of K_Na channels may contribute to APD shortening during ischemia. Accordingly, it is of interest to know whether bepridil modulates K_ATP and K_Na channels. The acquisition of such information is mandatory to fully understand the mode of action of bepridil on ventricular arrhythmias encountered during ischemia.

In the present study, we first examined the effects of bepridil on K_ATP and K_Na channels in isolated guinea pig ventricular myocytes, using the patch-clamp technique. Next, we recorded action potentials with a microelectrode in coronary-perfused right ventricular myocardium (Shigematsu et al., 1995) and examined the effect of bepridil on the metabolic inhibition-induced shortening of the APD. Our results show that bepridil inhibits both K_ATP and K_Na channels and blunts the APD shortening during metabolic inhibition.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

All procedures conformed to the guidelines stipulated by the Physiological Society of Japan and the Animal Ethics Committee of Oita Medical University.

Chemicals

Bepridil (a kind gift from Sankyo, Tokyo, Japan), glibenclamide (a kind gift from Hoechst Japan, Tokyo, Japan), R56865 (N-[1-(4-(4-fluorophenoxy)butyl)-4-piperidinyl]-N-methyl-2-benzothiazolamine; a kind gift from Janssen Research Foundation, Beerse, Belgium), FCCP [carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone; Sigma Chemical Co., St. Louis, MO] were dissolved in dimethyl sulfoxide as stock solutions. Each stock solution was added to the experimental solution immediately before use to produce the final concentration given in the text. Control experiments were performed with 0.08% dimethyl sulfoxide, which was the maximal concentration used and had no effect on the ionic currents, action potentials, and contraction.

Isolated Patches from Single Ventricular Myocytes

Cell Isolation. Single guinea pig ventricular myocytes were isolated enzymatically using a modified procedure described by Taniguchi et al. (1981). In brief, guinea pigs weighing 250 to 300 g were stunned by a blow on the neck, and the heart was quickly dissected and perfused through the coronary arteries with modified Tyrode's solution. The composition of modified Tyrode's solution was 137 mM NaCl, 3 mM NaHCO₃, 5.4 mM KCl, 0.16 mM NaH₂PO₄, 0.5 mM MgCl₂, 1.8 mM CaCl₂, 5.5 mM glucose, and 5 mM HEPES (pH 7.4 with NaOH). After 5 min of perfusion, the hearts were perfused without Ca²⁺ for an additional 5 min. The perfusate was switched to Ca²⁺-free modified Tyrode's solution containing 0.005% collagenase (Type I; Yakult, Tokyo, Japan). After a 3- to 5-min perfusion, the heart was immersed in KB solution composed of 5 mM KCl, 70 mM glutamic acid, 10 mM taurine, 10 mM oxalic acid, 5 mM KH₂PO₄, 5 mM HEPES, 11 mM glucose, and 0.5 mM EGTA (pH 7.4 with KOH). The temperature of these perfusates was maintained at 35-36°C. A small piece of tissue was detached from the ventricles, and the cells were dispersed by stirring in the recording chamber (0.8 ml in volume) and mounted on the stage of an inverted microscope (TMD; Nikon, Tokyo, Japan). Rod-shaped cells with a clear margin and striation were used for the experiments.

Electrophysiological Measurements. Conventional patch-clamp techniques (Hamill et al., 1981) were used to record K_ATP and K_Na channel currents from inside-out membrane patches. The resistance of the patch electrodes ranged from 3 to 5 M. The composition of the pipette solution (extracellular medium) was 140 mM KCl, 1.8 mM CaCl₂, 0.53 mM MgCl₂, 5.5 mM glucose, and 5 mM HEPES (pH 7.4 with KOH). After a gigaohm-seal formation, the patch membrane was excised to make inside-out patches in bath solution (intracellular medium). For K_ATP channel current recording, the composition of the bath solution was 140 mM KCl, 5 mM HEPES, 5 mM EGTA, and 5.5 mM glucose (pH 7.3 with KOH). For K_Na channel current recordings, 2 mM ATP was added to the bath solution to block the K_ATP channel activity completely after recording this channel current; then, the bath solution was switched to a high-sodium solution to evoke K_Na currents, which contained 100 mM NaCl, 40 mM KCl, 2 mM MgCl₂, 2 mM Na₂ATP, 5 mM HEPES, 5 mM EGTA, and 5.5 mM glucose (pH 7.4 with KOH). All experiments were carried out at room temperature (~22°C).

Coronary-Perfused Right Ventricular Myocardium

Preparations. The isolated coronary-perfused guinea pig right ventricular free wall was prepared as described previously (Shigematsu et al., 1995). In brief, the isolated right ventricular free wall preparation, in which the coronary artery was cannulated via the aorta, was mounted in the recording chamber and pinned to the floor of the chamber. The coronary artery was perfused with oxygenated Tyrode's solution composed of 136.7 mM NaCl, 11.9 mM NaHCO₃, 5.4 mM KCl, 0.42 mM NaH₂PO₄, 0.5 mM MgCl₂, 1.8 mM CaCl₂, and 11 mM glucose (pH 7.35-7.40 when gassed with 97% O₂/3% CO₂). The flow rate was maintained at 1.0 ± 0.2 ml/min/g wet weight using a roller pump (MP-3; Tokyo Rikakikai, Tokyo, Japan), with an intra-aortic pressure ranging from 40 to 50 mm H₂O. The surface of the preparation was superfused with glucose-free hypoxic Tyrode's solution (10 ml/min) to minimize direct O₂ diffusion from the surface of the preparations into the muscles. The composition of the hypoxic Tyrode's solution was the same as above, except that it contained no glucose and was gassed with 97% N₂ and 3% CO₂. The temperatures of these solutions were maintained at 37 ± 0.5°C.

Electromechanical Measurements. The basal portion of the preparation was stimulated at 3 Hz throughout the experiment with the use of a pair of platinum electrodes connected to the isolation unit of an electrical stimulator (SS-302J; Nihon Kohden, Tokyo, Japan). Action potentials were recorded from the epicardial site of the ventricular muscle fiber that was located deep (usually five or six cell layers) in the subepicardial surface with the use of a flexibly mounted microelectrode. Microelectrodes (tip resistance, 20-30 M) were filled with 3 M KCl. A direct current preamplifier (MEZ-7101; Nihon Kohden) with capacitance compensation was used to record the transmembrane potential. Contractile tension was recorded using a force transducer (TB-612T; Nihon Kohden) connected to the apical end of the preparation. Resting tension was adjusted to obtain the optimal developed tension. The membrane potential and contractile tension were monitored on a multibeam oscilloscope (VC-9A; Nihon Kohden) and recorded on a multichannel thermal arraycorder (WT-645G; Nihon Kohden).

Data Analysis

All data are expressed as mean ± S.E., and the number of cells tested was given (n) for patch-clamp experiments (all individual patches were taken from different cells; the number of patches used is equivalent to the number of cells used) and the number of preparations for the experiments with coronary-perfused ventricular muscles. ANOVA with the Fisher post hoc test and paired or unpaired t tests were used to assess statistical significance. A value of P  .05 was considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Bepridil on K_ATP Channels. We first examined the effects of bepridil on K_ATP channel currents recorded in inside-out membrane patches. Figure 1A shows the representative effects of bepridil on outward K_ATP channel currents. After the inside-out patch was prepared from ventricular myocytes, the membrane potential was held first at +60 mV to generate the outward current through the K_ATP channel. Multiple channels with a unitary channel conductance of ~80 pS were opened. Bepridil at a concentration of 10 µM decreased the K_ATP channel current without affecting the unitary current amplitude. After removal of the drug from the perfusate, the current was restored. Concentration-dependent effects of bepridil on the outward K_ATP channel current are summarized in Fig. 2. The relative channel activity (relative NPo) measured 1.5 min after the application of bepridil was plotted against the drug concentrations tested (1-100 µM). The IC₅₀ value of NPo was attained at a bepridil concentration of 10.5 µM with a Hill coefficient of 1.01. 

View larger version (40K): [in this window] [in a new window]   Fig. 1.   Effects of bepridil and glibenclamide on outward KATP channel currents in inside-out membrane patches. A, effect of bepridil on outward KATP channel currents (unitary current and NPo of the channels in control = 1.67 pA and 2.68, respectively). B, effect of glibenclamide on outward KATP channel currents (NPo of the channels in control = 2.74). Outward currents were evoked at a holding potential of +60 mV. Dotted lines indicate the zero current level. A and B are recorded from different patches.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Concentration-response relationship for bepridil inhibition of outward KATP channel currents at +60 mV. The relative NPo of the channels was plotted against the bepridil concentration. Data are mean ± S.E. with the number of patches tested indicated in parentheses. The curve was drawn to fit the Hill equation. The IC50 value for the outward current was estimated to be 10.5 µM.

View larger version (44K): [in this window] [in a new window]   Fig. 3.   Effects of bepridil and glibenclamide on inward KATP channel currents in inside-out membrane patches. A, effect of bepridil on inward KATP channel currents (NPo of the channels in control = 1.65). B, effect of glibenclamide on inward KATP channel currents (NPo of the channels in control = 1.38). Inward currents were evoked at a holding potential of 60 mV. Dotted lines indicate the zero current level.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Concentration-response relationship for bepridil inhibition of inward KATP channel currents at 60 mV. The relative NPo of the channels was plotted against the bepridil concentration. Data are mean ± S.E. with the number of patches tested indicated in parentheses. The curve was drawn to fit the Hill equation. The IC50 value for the outward current was estimated to be 6.6 µM.

Effect of Bepridil on K_Na Channels. To test the effect of bepridil on K_Na channels, a K_Na channel current was elicited by exposing an inside-out patch to a high-sodium solution. As shown in Fig. 5A, after the K_ATP channel current was recorded in the high-potassium solution (Fig. 5A-a), ATP at a concentration of 2 mM was added to the bath solution to block the K_ATP channel activity, where the inward rectifier K⁺ current (I_K1) remained unaffected (Fig. 5A-b). When the bath solution was switched to the high-sodium solution ([Na⁺]_i = 100 mM), the K_Na channel current was eventually activated (Fig. 5B-a). The K_Na channel current was observed in ~25% of the patches excised under our experimental conditions (n = 108). The mean slope conductance of the K_Na channel was 216 ± 15 pS (n = 5), a value consistent with a previous report (Wang et al., 1991). As shown in Fig. 5, B-b and B-c, bepridil (10 µM) reversibly inhibited the K_Na channel currents. The blocking effect of bepridil on K_Na channel currents was concentration dependent, and the IC₅₀ value and the Hill coefficient were 2.2 µM and 0.85, respectively (Fig. 6). Because the slope conductance of K_Na channels was considerably decreased (to 123 ± 17 pS; n = 5) at potentials more positive than 20 mV as reported by Wang et al. (1991), we did not systematically examine the effect of bepridil on the outwardly directed K_Na channel current. We verified that R56865 (10 µM), an alleged K_Na channel inhibitor (Carmeliet and Tytgat, 1991), blocked the inwardly directed K_Na channel currents (Fig. 5B-d) and decreased the mean relative NPo by 92 ± 3% (n = 4).

View larger version (50K): [in this window] [in a new window]   Fig. 5.   Effects of bepridil and R56865 on KNa channel currents in inside-out membrane patches. A, KATP channel currents recorded in zero ATP bath solution (a); the application of 2 mM ATP inhibited KATP channels (b). Note that inward rectifier K+ currents indicated by brackets below the trace (a) were not suppressed by the application of ATP as shown by a bracket below the trace (b). B, KNa channel current recorded from the same patch in A, where the bath solution was switched to a solution containing 100 mM Na+. The membrane potential was held at 60 mV in each panel. a, control (NPo of the channels = 1.39). b, effect of 10 µM bepridil. c, after washout of bepridil. d, effect of 10 µM R56865. e, after washout of R56865. Dotted lines indicate zero current level. Traces in A and B were recorded from the same patch.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Concentration-response relationship for bepridil inhibition of KNa channel currents at 60 mV. The relative NPo of the channels was plotted against the bepridil concentration. Data are mean ± S.E. with the number of patches tested in parentheses. The curve was drawn to fit the Hill equation. The IC50 value was estimated to be 2.2 µM.

Effects of Bepridil on APD and Contractile Tension during Metabolic Inhibition (MI). In the next series of experiments, with coronary-perfused right ventricular myocardium, we examined the effects of bepridil on the APD and the contractile tension during MI and compared the results with those for glibenclamide and R56865. MI was achieved by the addition of 0.1 µM FCCP, a mitochondrial uncoupler of oxidative phosphorylation, and by omitting glucose from the coronary-perfused Tyrode's solution. As shown in Fig. 7, MI induced the shortening of the APD in a time-dependent manner. After a 20-min exposure to MI (control MI), the APD was shortened to 14.3 ± 2.0% of the pre-MI value (from 179.3 ± 2.7 to 26.4 ± 3.6 ms, n = 5). When the application of bepridil (10 µM) was started 5 min after the introduction of MI, the shortening of the APD was dramatically blunted and remained as high as 84.5 ± 1.2% of the pre-MI values after 20 min of MI. Based on our observations mentioned above, it is reasonable to speculate that bepridil inhibited both K_ATP channels and K_Na channels and attenuated the MI-induced shortening of the APD. We therefore examined whether glibenclamide and/or R56865 mimics the effect of bepridil. Glibenclamide at a concentration of 1 µM (started also 5 min after introduction of MI) mitigated the APD shortening, albeit this effect was limited to only the initial 6 to 8 min of MI. In contrast, R56865 at a concentration of 10 µM significantly attenuated the shortening of the APD only in the later phase (10-20 min) of MI. Concomitant application of glibenclamide (1 µM) with R56865 (10 µM) prevented the APD shortening most effectively and significantly during almost the entire period of MI (from 8 to 20 min), although the effect was still apparently smaller in comparison with that seen in the presence of bepridil alone.

View larger version (30K): [in this window] [in a new window]   Fig. 7.   Effects of bepridil, R56865, and glibenclamide on MI-induced shortening of APD. The time course of the APD measured at the 90% repolarization level is expressed as a percent of the pre-MI value (179.3 ± 2.7, 173.8 ± 2.6, 173 ± 1.9, 175.8 ± 2.2, and 172.6 ± 2.8 ms for Control, MI + Bepridil, MI + R56865, MI + Glibenclamide, and MI + Glibenclamide + R56865, respectively). Drugs were applied 5 min after the introduction of MI (n = 5/group). *P < .05 versus control (MI).

View larger version (25K): [in this window] [in a new window]   Fig. 8.   Effects of bepridil, R56865, and glibenclamide on MI-induced changes in developed tension. The time course of the changes in the developed tension is expressed as a percentage of the pre-MI value (13.7 ± 0.9, 13.3 ± 0.6, 12.9 ± 1.1, 13.1 ± 1.0, and 13.7 ± 0.8 millinewtons for Control, MI + Bepridil, MI + R56865, MI + Glibenclamide, and MI + Glibenclamide + R56865, respectively). The drugs were applied 5 min after the introduction of MI (n = 5/group). *P < .05 versus control (MI).

View larger version (37K): [in this window] [in a new window]   Fig. 9.   Summarized data for the resting tension measured 20 min after the introduction of MI. *P < .05 versus control (MI).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of the present study demonstrate that: 1) in inside-out membrane patches from guinea pig ventricular cells, bepridil inhibited both K_ATP and K_Na channels in a concentration-dependent manner, and 2) in coronary-perfused right ventricular preparations, bepridil significantly blunted the shortening of the APD caused by MI.

The K_ATP channels are activated by a decrease in intracellular ATP concentration (Noma, 1983) and selectively inhibited by antidiabetic sulfonylureas (Fosset et al., 1988). Recent molecular-biological studies reveal that cardiac K_ATP channels are heteromultimers of inwardly rectifying potassium channel subunits (Kir6.2) and sulfonylurea receptors (SUR2A; Aguilar-Bryan et al., 1998). Other than intracellular ATP and sulfonylureas, several modulators of K_ATP channels have been documented (Edwards and Weston, 1993). We previously reported that class Ia antiarrhythmic drugs such as cibenzoline, disopyramide, and procainamide inhibit the K_ATP channel current in guinea pig ventricular cells (Wu et al., 1992; Sato et al., 1993). In addition, class Ic antiarrhythmic drugs such as flecainide inhibit K_ATP channels when the currents are directed outward but not when they are directed inward (Wang et al., 1995). In the present study, we found that bepridil, a class IV antiarrhythmic drug, blocked both outward and inward K_ATP channel currents in a concentration-dependent manner. On the other hand, the K_Na channels are activated by an increase in intracellular Na⁺ concentration (Kameyama et al., 1984; Wang et al., 1991; Mistry et al., 1997). The molecular structure of this channel is still unknown. R56865 is reported to be a potent inhibitor of K_Na channels (Carmeliet and Tytgat, 1991), which we verified in the patch-clamp study. More recently, it has been reported that several antiarrhythmic drugs inhibit K_Na channels in guinea pig ventricular myocytes (Mori et al., 1996). In the present study, we found that bepridil inhibited K_Na channels with an IC₅₀ value of 2.2 µM, and K_ATP channels with an IC₅₀ value of 6.6 to 10 µM; these concentrations are within or close to the therapeutic concentrations for humans of ~3 µM (Hollingshead et al., 1992).

Since the discovery of K_ATP channels in heart cells by Noma (1983), it has been accepted that K_ATP channels can be opened via hypoxic and ischemic conditions and that they shorten the APD in cardiac myocytes (Faivre and Findlay, 1990; Deutsch et al., 1991; Nakaya et al., 1991). In contrast, the physiological and pathophysiological significance of the K_Na channel is poorly understood. Nevertheless, it is reasonable to speculate that the activation of K_Na channels occurs during cardiac ischemia or digitalis intoxication and contributes to the shortening of the APD (Luk and Carmeliet, 1990; Veldkamp et al., 1994). In the present study, the MI (achieved with 0.1 µM FCCP plus glucose removal) shortened the APD in a time-dependent manner in coronary-perfused right ventricular preparations. When glibenclamide (a K_ATP channel blocker) alone was applied 5 min after the introduction of MI, the APD shortening was significantly attenuated only during the early phase (6-8 min) of MI. In contrast, R56865 (a K_Na channel inhibitor) alone attenuated the APD shortening only during the relatively late phase (10-20 min) of MI. Based on these results, it is reasonable to assume that 1) the decrease in the subsarcolemmal ATP concentration results in the activation of K_ATP channels before the inhibition of Na⁺-K⁺ ATPase and 2) further depletion of [ATP]_i inhibits Na⁺,K⁺-ATPase, thereby leading to eventual increases in the intracellular Na⁺ concentration that activates K_Na channels. Consistent with this notion, Abe et al. (1999) reported that the inhibition of Na⁺,K⁺-ATPase preserves [ATP]_i and leads to blockade of the K_ATP channels. In fact, dihydro-ouabain (an Na⁺,K⁺-ATPase inhibitor) attenuated the MI-induced shortening of the APD via inhibition of the K_ATP channels, when the drug application was started 5 min, but not 10 min, after the introduction of MI, using the same experimental design as used in the present study. These results imply that the Na⁺/K⁺ pump is still operating 5 min after introduction of MI, whereas it is inhibited after a relatively longer period of MI. These findings taken together suggest that K_Na channels are activated during the late phase of MI and principally contribute to the APD shortening in this phase.

In the present study, bepridil blunted the MI-induced shortening of the APD, when application was started 5 min after introduction of MI. Furthermore, the attenuation of the APD shortening was observed in early as well as late phases of MI. The concomitant application of glibenclamide with R56865 also attenuated the shortening of the APD during both early and late phases of MI. Although glibenclamide and R56865 only partially mimicked the effect of bepridil, it is reasonable to consider that the inhibition of K_ATP and K_Na channels by these agents contributes to the attenuation of the APD shortening during MI.

Activation of K_ATP and K_Na channels accelerates repolarization, shortens the effective refractory period, and confers deleterious consequences (e.g., provocation of reentrant arrhythmias; Rosen, 1995). Therefore, blockade of these channels by bepridil and subsequent prolongation of the APD (i.e., prolongation of the effective refractory period) could reduce the incidence of reentrant arrhythmias. On the other hand, it can be postulated that the prolongation of the APD by bepridil may enhance myocardial damage during ischemia and reperfusion (Shigematsu et al., 1995). However, unlike glibenclamide or R56865, bepridil did not increase the resting tension during MI (Fig. 9), presumably because of its inhibitory effect on L-type Ca²⁺ channels (Yatani et al., 1986). Indeed, in common with other Ca²⁺ channel antagonists, the cardioprotective effects of bepridil have been reported (Reifart et al., 1986; Watts et al., 1987; van Amsterdam et al., 1990).

In conclusion, bepridil blocks K_ATP and K_Na channels in cardiac ventricular cells in its therapeutic concentrations and blunts the APD shortening during MI. This property, along with its alleged cardioprotective effect, may be useful for the management of tachyarrhythmias encountered during ischemia.