Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Three Ca²⁺ channel blocker prototypes of (dihydropyridines, phenylalkylamines, and benzothiazepines) are well known to have their own binding sites on ₁ subunits of L-type Ca²⁺ channels, which may explain some of the observed clinical differences, including tissue selectivity. From the pharmacological and clinical aspects, it has been documented that Ca²⁺ channel blockers can be divided roughly into those of the dihydropyridine and nondihydropyridine families (Opie, 1997). The Ca²⁺ channel blockers have been demonstrated to protect the myocardium from injury caused by ischemia or reperfusion in various experiments. Watts and coworkers (1986) showed that diltiazem, verapamil, and nifedipine, at concentrations that exerted equal negative inotropic effects, had the same cardioprotective effects and concluded that cardiodepression was an important mechanism underlying the cardioprotective effects of these blockers. On the other hand, Hamm and Opie (1983) and Grover and Sleph (1989) reported that diltiazem, but neither verapamil nor nifedipine, at doses that did not result in cardiodepression, inhibited lactate dehydrogenase release during the reperfusion period.

It is known that K⁺ and H⁺ accumulate extracellularly (Harris et al., 1954; Gebert et al., 1971) and that the tissue ATP content decreases (Jennings and Steenbergen, 1985) during myocardial ischemia. Of interest is the elevation of extracellular K⁺ concentrations ([K⁺]_e) during ischemia, because this elevation is characteristically triphasic (Hill and Gettes, 1980; Weiss and Shine, 1981, 1982; Kléber, 1984). This phenomenon reflects the temporal changes in cellular electrophysiological events occurring during ischemic periods. In fact, it has been demonstrated that opening of ATP-sensitive K⁺ (K_ATP) channels, at least in part, contributes to the early phase of [K⁺]_e elevation (Bekheit et al., 1990; Wilde et al., 1990; Mitani et al., 1991). The late phase of [K⁺]_e elevation is considered to reflect the cessation of anaerobic glycolysis and reduced Na⁺-K⁺ pump activity (Sakamoto et al., 1997, 1998). However, the effects of Ca²⁺ channel blockers on ionic movement, levels of metabolic markers, and cardiac function have not been fully investigated comparatively in terms of different ischemic phases.

It is plausible that some of the time-dependent factors that affect the potency of Ca²⁺ channel blockers specifically alter mechanical and metabolic parameters during ischemic and postischemic periods. One of these factors may be cell membrane depolarization, which has been shown to increase the potencies of Ca²⁺ channel blockers with varying voltage dependence (Ehara and Daufmann, 1978; Kanaya and Katzung, 1983; Uehara and Hume, 1985; Okuyama et al., 1994). Furthermore, enhancement of the effects of these agents by acidosis has been reported (Briscoe and Smith, 1982; Smith and Briscoe, 1985). Thus, temporal changes in L-type Ca²⁺ channel blocking activity under various conditions of [K⁺]_e and pH may contribute to their different protective effects on the ischemic myocardium.

In the present study, we evaluated the protective effects of three Ca²⁺ channel blockers, diltiazem, verapamil, and nifedipine, on changes in [K⁺]_e, myocardial pH, diastolic pressure, postischemic cardiac function, and levels of high-energy phosphates in globally ischemic guinea pig hearts. Interestingly, we recognized time-dependent inhibitory effects on these parameters, which differed among the Ca²⁺ channel blockers. Therefore, we further explored the possibility that the different potencies of Ca²⁺ channel blockers under condition of high [K⁺]_e account for their different protective effects on the ischemic myocardium.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Langendorff Preparations. Hearts from male, Hartley strain, guinea pigs (6-8 weeks old, weighing 380-450 g) were removed quickly and immersed immediately in ice-cold Krebs-Henseleit solution (KHS) of the following composition: 118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, and 10 mM glucose. When contraction ceased, an aortic arch was cannulated and the heart was perfused retrogradely at a constant pressure of 55 mm Hg with KHS aerated with 95% O₂ + 5% CO₂ at 37°C (pH 7.5). The O₂ tension of the perfusate was maintained above 600 mm Hg throughout the experiment except for an ischemic period. The heart was kept in the humidified atmosphere (~150 mm Hg O₂) in a temperature (37°C)-controlled chamber. The left ventricular developed pressure (LVDP) was measured using a thin latex balloon, which was inserted via the left atrium into the LV cavity. The balloon was connected to a pressure transducer (MPU-0.5, Toyo Baldwin, Tokyo, Japan) and the signal was amplified (AP-621G, Nihon Kohden, Tokyo, Japan). The LV end-diastolic pressure (LVEDP) was adjusted to 5 to 10 mm Hg by inflating the balloon. Heart rate (HR) was counted by a tachometer (AT-601G, Nihon Kohden) into which LVDP signal was fed. The first derivative of LVDP was calculated by a pressure processor (EQ-601G, Nihon Kohden). These parameters were simultaneously recorded in a multichannel recorder (WR3701, Graphtec, Japan).

Measurement of [K⁺]_e and H⁺ Concentration. We used a K⁺-sensitive electrode (Hill et al., 1978) and a pH electrode (Johnson et al., 1990) to measure [K⁺]_e and extracellular H⁺ concentration ([H⁺]_e), respectively. Briefly, a thin polyethylene tube (1 mm i.d.; 1.5 mm o.d.) was pulled in a flame to reduce the i.d. to 0.2 mm, then filled with 500 mM KCl solution saturated with AgCl and sealed with polyvinyl chloride membrane containing valinomycin for the K⁺ electrode or tridodecylamine for the pH electrode. The K⁺- and pH-selective electrodes, via an Ag-AgCl wire, and a reference Ag-AgCl electrode were connected to a high-impedance amplifier (MEZ-7200, Nihon Kohden). The voltage difference was amplified with a bioelectric amplifier (AB-621B, Nihon Kohden) and displayed by a pen recorder (FBR-252A, Toa Electric, Tokyo, Japan). Before and after each experiment, the K⁺ electrode was calibrated using isotonic KCl-NaCl solution ([K⁺] = 3, 6, 12, and 24 mM, total ionic concentration = 149 mM) and the pH electrode was calibrated using HEPES-NaOH buffers (pH 6.5, 7.0, 7.5, and 8.0). Electrodes that showed a voltage change of 56 to 62 mV/decade of change in [K⁺] or pH were used. The K⁺ and pH electrode tips were inserted into the LV free wall to a depth of 1 to 1.5 mm.

Experimental Protocol. In the studies on Ca²⁺ channel blockers and low Ca²⁺ concentration ([Ca²⁺]), the hearts were equilibrated under spontaneous beating conditions for 30 min. Then the hearts were perfused with KHS containing the required drug or low [Ca²⁺] (1.0 mM) KHS for 10 min, followed by global ischemia for 30 min. [K⁺]_e, [H⁺]_e, the time to contractile arrest (considered as a LVDP < 2 mm Hg) and ischemic contractures were measured during ischemia. The latency of [K⁺]_e elevation was expressed as the time it took for the [K⁺]_e to reach 7 mM. After ischemia for 30 min, the hearts were reperfused with drug-free KHS and postischemic functional parameters were measured after reperfusion for 30 min.

Measurement of High-Energy Phosphates and Lactate Contents. Before the onset of ischemia or after 3, 15, or 30 min of ischemia, the hearts were quickly frozen by precooled Wollenberger's clamp and immediately put into liquid nitrogen. Frozen hearts were homogenized in 0.6 N perchloric acid with a polytron homogenizer. Tissue ATP, creatine phosphate (CrP) and lactate levels were measured according to the methods previously reported (Lamprecht and Trautschold, 1974; Lamprecht et al., 1974; Gutmann and Wahlefeld, 1974).

Statistical Analysis. All data are expressed as means ± S.E.M. The time-dependent data were subjected to two-way ANOVA to determine whether there were any significant differences among the groups. The pH and ischemic contracture data obtained from just after the onset of ischemia to cessation of ischemia were analyzed. To estimate each phase of the triphasic pattern of [K⁺]_e elevation, the [K⁺]_e data were divided into three phases, i.e., the early (0-8 min), plateau (8-15 min), and late (15-30 min) phases, and each of these was analyzed. One-way ANOVA was conducted to evaluate the one-way layout data. If a significant difference was found, Bonferroni's post hoc test was conducted to determine which group differed significantly. In some experiments, an unpaired or paired t test was performed to compare two groups. A difference at a p of < 0.05 was considered statistically significant.

Chemicals. Diltiazem HCl was provided by Tanabe Seiyaku Co., Ltd. (Saitama, Japan). Verapamil HCl and nifedipine HCl were purchased from Sigma Chemical Co. (St. Louis, MO). All the other drugs and chemicals used were purchased from commercial sources. Dimethyl sulfoxide was used (final concentration, 0.02%, w/v) to dissolve nifedipine. At this concentration, dimethyl sulfoxide did not affect any of the cardiac function parameters evaluated (data not shown). All the experiments involving nifedipine were performed under dim light conditions to avoid drug decomposition.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Ca²⁺ Channel Blockers on [K⁺]_e Elevation and Acidosis during Ischemia

Table 1 shows the effects of Ca²⁺ channel blockers on the LVDP and HR before ischemia. No difference was observed in the parameters before adding drugs in each group from those in the control group. In comparison with the LVDP value after adding vehicle in the control group, diltiazem (0.1-10 µM), verapamil (0.03-1 µM), and nifedipine (0.01-1 µM) significantly (p < .05) reduced the LVDP in a concentration-dependent manner. Diltiazem, verapamil, and nifedipine reduced the HR significantly as well.

Figure 1 shows the time course of the [K⁺]_e elevation during ischemia. In the control hearts, [K⁺]_e elevation showed a triphasic pattern. The [K⁺]_e (initial level = 5.9 mM) started to rise within 60 s of cessation of perfusion (early phase). Thereafter, [K⁺]_e was maintained for 5 to 15 min after ischemia onset ([K⁺]_e = 12.1 ± 0.4 mM, plateau phase), after which it tended to increase gradually and reached about 27 mM at the end of the 30-min ischemic period (late phase).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Extracellular potassium accumulation during global ischemia for 30 min of guinea pig perfused hearts in the absence () and presence (solid symbols) of diltiazem (top), verapamil (middle), and nifedipine (bottom). Each point represents mean ± S.E.M. of number of experiments shown in Table 1. Statistical analysis of [K+]e concentration was performed at each phase of its elevation (I, early phase; II, plateau phase; III, late phase).

The effects of drugs on [K⁺]_e during the 30-min ischemic period are shown in Fig. 1. Diltiazem (0.1-10 µM), verapamil (0.03-1 µM), and nifedipine (0.01-1 µM) delayed the onset of the early phase of [K⁺]_e elevation in a concentration-dependent manner (Fig. 1 and Table 2). These results correlated closely with their negative inotropic effects (see Table 1; r = 0.98, n = 50) and no differences among these three agents were detected. [K⁺]_e was increased to approximately 12 mM within 8 min after ischemia onset in the presence of each of these agents. [K⁺]_e in plateau phase did not differ from control. The late phase of [K⁺]_e elevation was inhibited by diltiazem and verapamil in a concentration-dependent manner (Fig. 1, top and middle). In contrast, it is notable that nifedipine, at any concentration that delayed the early phase of [K⁺]_e elevation, did not inhibit the late phase (Fig. 1, bottom).

Figure 2 shows the effects of Ca²⁺ channel blockers on acidosis during ischemia. The preischemic extracellular pH value of control hearts was 7.28 ± 0.03 (n = 10). After cessation of perfusion, the extracellular pH decreased gradually (pH = 6.27 ± 0.09 at 10 min) and finally reached a steady phase (pH = 5.76 ± 0.08) from approximately 23 min until the cessation of ischemia (Fig. 2, open symbols). Although diltiazem (0.1 and 1 µM) tended to inhibit the pH reduction during early phase, the pH declined to the control level by the end of the ischemic period. A higher concentration of diltiazem (10 µM) significantly inhibited acidosis throughout the ischemic period and the pH continued to decline until the end of ischemia (Fig. 2, top). Verapamil at concentrations that exerted weak negative inotropic effects (0.03-1 µM), inhibited acidosis significantly during ischemia (Fig. 2, middle). The pH continued to decline till the end of ischemia in the presence of verapamil (0.1 and 1 µM). The pH change by nifedipine exhibited less concentration dependence. The effect of nifedipine appears to be biphasic; 0.1 µM nifedipine inhibited the pH decrease significantly (p < .05 versus control), although the effect of this compound was attenuated at a concentration of 1 µM (p > .05 versus control; Fig. 2, bottom).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Extracellular acidosis during global ischemia for 30 min of guinea pig perfused hearts in the absence () and presence (solid symbols) of diltiazem (top), verapamil (middle), and nifedipine (bottom). Each point represents mean ± S.E.M. of number of experiments shown in Table 1.

Effects of Ca²⁺ Channel Blockers on Ischemic Contracture and Postischemic Cardiac Function

Cardiac contractions declined rapidly after cessation operfusion and the time to arrest (LVDP < 2 mm Hg) of the control hearts (n = 10) was 2.6 ± 0.1 min. None of the Ca²⁺ channel blockers influenced the time to arrest.

Ischemic contracture was assessed by measurement of LVEDP during ischemia. In control hearts, LVEDP started to increase 18.6 ± 1.1 min after ischemia onset and reached 74.0 ± 5.9 mm Hg at the end of the ischemic period (ischemic contracture; Fig. 3). Diltiazem (0.1-10 µM) and verapamil (0.03-1 µM) postponed the time of ischemic contracture onset in a concentration-dependent manner and inhibited the contractures during ischemia for 30 min. Nifedipine caused a biphasic action on ischemic contracture. Nifedipine at 0.01 µM appeared to accelerate the onset of contracture although the values of LVEDP at each time point were varied (Fig. 3, S.E.M. bars). This blocker at 0.1 and 1 µM inhibited the contractures significantly at the end of the 30-min ischemic period, although the effect was less than that of verapamil (p < .05). Unlike diltiazem and verapamil, nifedipine (0.01-1 µM) failed to delay the time of contracture onset (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Elevation of LVEDP during global ischemia for 30 min of guinea pig perfused hearts in the absence () and presence (solid symbols) of diltiazem (top), verapamil (middle), and nifedipine (bottom). Each point represents mean ± S.E.M. of number of experiments shown in Table 1.

Effects of Ca²⁺ channel blockers on LVEDP elevation after a 30-min reperfusion period are shown in Table 3. The LVEDP of the control group after reperfusion was 66.3 ± 4.4 mm Hg (n = 10). Diltiazem and verapamil significantly inhibited LVEDP elevation after reperfusion in a concentration-dependent manner. Nifedipine did not exhibit a significant inhibition of LVEDP elevation.

Effects of Ca²⁺ Channel Blockers on Tissue ATP, CrP, and Lactate Levels

We examined the effects of diltiazem (10 µM), verapamil (1 µM), and nifedipine (0.3 µM) on levels of tissue high-energy phosphates, ATP, and CrP, and an anaerobic glycolytic product, lactate (Fig. 4), at concentrations producing nearly equivalent effects on LVDP (control, 101.0 ± 1.1%; diltiazem, 20.3 ± 1.1%; verapamil, 17.6 ± 1.0%; nifedipine, 20.9 ± 2.1%; each n = 16; compared with the values before drug or vehicle).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Comparison of effects of diltiazem (10 µM), verapamil (1 µM), and nifedipine (0.3 µM) on tissue ATP, CrP, and lactate levels before the onset of ischemia or after 3, 15, or 30 min of ischemia. Each datum represents mean ± S.E.M. of four experiments. *p < .05; compared with corresponding value in control group.

ATP. In the control group, the tissue ATP level gradually fell from a peak of 18, reaching 6 µmol/g d.wt. by the end of the ischemic period. Diltiazem and verapamil significantly attenuated a decrease in the tissue ATP level at 15 min of ischemia. Nifedipine, however, failed to alter this ATP level decrease.

CrP. In the control group, the CrP level declined rapidly from 22 to 5 µmol/g d.wt. during the first 3 min of ischemia and continued to fall until the end of ischemic period. Diltiazem and verapamil clearly maintained the CrP level at 3 min of ischemia, although nifedipine appeared not to exert such a protective effect.

Lactate. The tissue lactate level in the control group increased rapidly within 3 min of ischemia and continued to increase up to 170 µmol/g d.wt. Before the ischemic period, all three blockers had decreased the lactate level to a similar extent, probably due to the decreased preischemic function. However, diltiazem and verapamil, but not nifedipine, significantly attenuated the lactate increase at 15 min of ischemia.

Effects of Negative Inotropy and Chronotropy on [K⁺]_e Elevation and Acidosis during Ischemia

To test whether either negative inotropy or chronotropy can inhibit the late phase of [K⁺]_e elevation and acidosis during ischemia, the effect of reducing LVDP in low [Ca²⁺] KHS (Fig. 5) or of reducing HR in paced hearts was examined (Fig. 6).

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Effects of the Ca2+ concentration (2.5 or 1.0 mM) of perfusate on extracellular potassium accumulation (top), acidosis (middle), and LVEDP (bottom) during global ischemia for 30 min of guinea pig perfused hearts. Each point represents mean ± S.E.M. of four experiments.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Effects of pacing rate (3 or 4 Hz) on extracellular potassium accumulation (top), acidosis (middle), and LVEDP (bottom) during global ischemia for 30 min of guinea pig perfused hearts. Each point represents mean ± S.E.M. of four experiments.

The preischemic LVDP and HR of the control hearts (n = 4) were 87.2 ± 6.4 mm Hg and 251 ± 11 beats/min, respectively. The LVDP was significantly reduced to 24.2 ± 3.7 mm Hg (35% of the control value) by the low [Ca²⁺] (1.0 mM) perfusate, but the HR was not affected (227 ± 7 beats/min). Although the early and plateau phases of [K⁺]_e were not affected, the late phase was significantly inhibited by the low [Ca²⁺] (Fig. 5). The pH decreases and ischemic contractures were also inhibited significantly by the low [Ca²⁺], but the time to arrest, 2.5 ± 0.13 min, was not significantly different from that of the control group. The LVEDP of the low [Ca²⁺] group after a 30-min reperfusion was 19.8 ± 6.4 mm Hg (n = 4), which is significantly less than that of the control group (62.1 ± 3.3 mm Hg, n = 4, p < .01).

The preischemic LVDP of hearts at a pacing rate of 4 Hz was 74.0 ± 5.7 mm Hg (n = 4). The LVDP was not altered (66.7 ± 3.5 mm Hg) when the pacing rate was reduced to 3 Hz (p > .05, n = 4). Although the early and plateau phases of [K⁺]_e were not affected, the late phase was significantly inhibited in the hearts paced at 3 Hz (Fig. 6). The pH decrease was also inhibited significantly in the hearts paced at 3 Hz, yet attenuation of ischemic contracture (Fig. 6) or LVEDP after a 30-min reperfusion (3 Hz, 78.5 ± 4.7 mm Hg, n = 4; 4 Hz, 82.3 ± 6.0 mm Hg, n = 4; p > .05) was not affected.

Negative Inotropic Effects of Ca²⁺ Channel Blockers in Various [K⁺]_e

There was an apparently contradictory effect in the response of ischemic guinea pig hearts to nifedipine compared with the other Ca²⁺ channel blockers. Nifedipine potently attenuated LVDP even though its negative chronotropic action was less than the other blockers (see Table 1). As shown in the previous section, a decrease in LVDP due to a decrease in external [Ca²⁺] inhibited [K⁺]_e elevation, acidosis, and postischemic cardiac dysfunction. Nifedipine, however, did not exert definite inhibitory actions on [K⁺]_e or postischemic dysfunction, nor caused a concentration-dependent inhibition of acidosis during ischemia. One possibility to explain this conflict is that in ischemic guinea pig hearts the negative inotropic effect of nifedipine may be attenuated whereas that of the other blockers remains the same. Alternatively, the negative inotropic effect of nifedipine is the same and that of the others is enhanced. Because depolarization of the cell membrane is one of the factors that may affect the potency of Ca²⁺ channel blockers (Ehara and Daufmann, 1978; Kanaya and Katzung, 1983; Uehara and Hume, 1985), the increase in [K⁺]_e during the early period of ischemia (5.9 mM to 8 or 9 mM) may be sufficient to depolarize the membrane (Okuyama et al., 1994). Therefore, we tested the possibility that the inhibition of contractility by these blockers under higher [K⁺]_e may be involved in their different protective actions.

When the [K⁺]_e of the perfusate was reduced from 5.9 to 2.9 mM, the LVDP increased to 113 ± 3.4% (n = 4) of the control value ([K⁺]_e=5.9 mM) and in the presence of 8.9 mM [K⁺]_e, it declined significantly to 64.5 ± 5.9% (p < .01, n = 4) of the control value. The effects of the [K⁺]_e of the perfusate on the negative inotropic effects of each Ca²⁺ channel blocker are shown in Fig. 7. Each response was expressed as a percentage of the value obtained just before drug application. The estimated pIC₅₀ [log I(₅₀)] values, calculated from the concentration-response curve, are shown in Table 4. The pIC₅₀ value for diltiazem in the presence of 8.9 mM [K⁺]_e was about 12 times higher than that in the presence of 2.9 mM [K⁺]_e, showing its negative inotropic effect was potentiated 12-fold and that of verapamil was potentiated 7-fold. In contrast, the pIC₅₀ value of nifedipine was not affected by increasing [K⁺]_e.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Comparison of negative inotropic effects of diltiazem (top), verapamil (middle), and nifedipine (bottom) on guinea pig perfused hearts. Each point represents mean ± S.E.M. of four experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The time course of the [K⁺]_e elevation in control hearts evoked by ischemia showed a triphasic pattern consisting of an early, a plateau, and a late phase, a pattern consistent with results reported previously (Weiss and Shine, 1981; Kléber, 1984). The major significance of the findings reported herein is that differential effects among these Ca²⁺ channel blockers were observed in the late phase; diltiazem and verapamil, but not nifedipine, inhibited the late phase of [K⁺]_e elevation. However, the mechanism underlying this inhibition of late phase of [K⁺]_e elevation remains obscure. Nonspecific K⁺ release from the myocardium has been suggested to occur during the 20- to 30-min period following ischemia onset as a result of cell membrane injury caused by reduced tissue ATP levels (Sakamoto et al., 1997). Similarly, ischemic contracture has been reported to reflect intracellular ATP level reduction (Koretsune and Marban, 1990). The present results demonstrate differences in the effects of the three blockers in ischemic contracture; diltiazem and verapamil inhibited ischemic contracture dose-dependently during the late phase of ischemia, whereas the effect of nifedipine was less clear. Cascio et al. (1990) previously demonstrated a close association between cell-to-cell electrical coupling, development of ischemic contracture, and the second rise in [K⁺]_e, all of which started to develop after approximately 15 min of no-flow ischemia in the rabbit papillary muscle. In their study, verapamil was shown to delay changes in electrical and mechanical function. This observation suggests that these functional and ionic changes may be triggered by a common event, i.e., a critical increase in free intracellular [Ca²⁺] due to the energy imbalance and decreased ATP content, which is of relevance in explaining irreversible ischemic damage (Cascio et al., 1990).

In the present experiment focusing on high-energy phosphates, diltiazem and verapamil were shown to preserve tissue ATP and CrP to a greater degree than nifedipine at 3 or 15 min of ischemia, an effect possibly attributable to reduced cardiac function during the early ischemic period. Diltiazem and verapamil, but not nifedipine, were also shown to decelerate an increase in lactate. Our previous studies have shown that diltiazem preserves high-energy phosphates during the early phase of ischemia and prolongs the period of glycolytic ATP synthesis, judging from delayed saturation of the increase in an anaerobic glycolytic product, lactate (Sakamoto et al., 1997). Glycolytic ATP synthesis is likely to result in inhibition of the late phase of [K⁺]_e elevation, for the following reasons. First, the Na⁺-K⁺ pump inhibitor ouabain selectively enhances the late phase of [K⁺]_e elevation (Sakamoto et al., 1998). Second, the Na⁺-K⁺ pump has been shown to be preferentially fueled by glycolytic ATP production in sheep cardiac Purkinje fibers (Glitsch and Tappe, 1993). In fact, the present study demonstrates that maneuvers with low [Ca²⁺] and low frequency pacing, which are expected to preserve high-energy phosphates during the early phase of ischemia, exerted remarkable inhibitory effects on the late phase of [K⁺]_e. Therefore, it is postulated that diltiazem and verapamil are capable of preserving ATP and CrP and prolonging the glycolytic period, which results in attenuation of the late phase of [K⁺]_e elevation and ischemic contracture, presumably through maintenance of Na⁺-K⁺ ATPase activity. In contrast, nifedipine, even at the concentration causing an equipotent effect on preischemic cardiac function, may possess less ability to preserve high-energy phosphates than the other blockers. Thus, nifedipine is likely to show differential effects during the late phase (20-30 min) of ischemia.

The inhibitory effects of these Ca²⁺ channel blockers on acidosis are also supported by findings of the late [K⁺]_e elevation, ischemic contracture, and levels of high-energy phosphates. Diltiazem and verapamil dose dependently blocked acidosis, whereas the dose-dependence of the effect of nifedipine was not clear. The plateau phase of pH was considered to reflect the cessation of glycolysis (Kingsley et al., 1991). In the present study, the plateau phase in the control group was observed from approximately 23 min until the cessation of ischemia. In contrast, pH apparently continued to decline until the end of the ischemic period in the diltiazem- and verapamil-treated groups, suggesting prolongation of the glycolytic period. Previously, diltiazem (Ichihara and Abiko, 1982), but neither nifedipine nor nisoldipine (Ichihara et al., 1986), was reported to reverse the pH decrease in a regional ischemia model. Furthermore, the inhibitory effect of verapamil on acidosis was shown to be stronger than that of nifedipine (Moo et al., 1988). The manner in which these agents affect acidosis during ischemia may, however, be complicated because [K⁺]_e elevation is thought to be linked to production of lactate or phosphate (Kléber 1983).

The early phase through the plateau phase of the [K⁺]_e elevation is shown to be inhibited by glibenclamide in the previous reports (Bekheit et al., 1990; Wilde et al., 1990; Mitani et al., 1991), indicating that opening of K_ATP channels plays a role in the early phase of [K⁺]_e. Because K_ATP channels have been reported to open in response to a decrease in the concentration of intracellular ATP (Noma and Shibasaki, 1985; Nichols and Lederer, 1990), agents with an inhibitory effect on ATP consumption are expected to prevent the elevation in [K⁺]_e as a result of inhibiting K_ATP channel opening etc. Our data demonstrate that three Ca²⁺ channel blockers (diltiazem, verapamil, and nifedipine) share the same property in that the delay of [K⁺]_e elevation correlates with the negative inotropic effects in the preischemic period (inhibition of LVDP). These results are consistent with the previous finding by Heijnis et al. (1991) that the ability to slow the rise in [K⁺]_e is a specific characteristic of calcium channel blockers.

It still remains uncertain what causes the differential effect of Ca²⁺ channel blockers on preservation of ATP and CrP and subsequent attenuation of the late phase of [K⁺]_e and ischemic contracture. Several studies have shown that ischemia alters ionic and energy parameters, such as elevating [K⁺]_e (Harris et al., 1954), causing acidosis (Gebert et al., 1971) and reducing the tissue ATP content (Jennings and Steenbergen 1985). These factors may affect the blocking effects of nifedipine, verapamil, and diltiazem on the L-type Ca²⁺ current. One of the possible mechanisms is depolarization of the cell membrane (Ehara and Daufmann, 1978; Kanaya and Katzung, 1983; Uehara and Hume, 1985), which occurs as [K⁺]_e increases in ischemic hearts (Kléber, 1983; Okuyama et al., 1994). To test the possibility that the blocking action of these three Ca²⁺ channel blockers is different under depolarizing conditions, we designed the experiment in which contractility was measured in the beating hearts at a constant heart rate in the perfusate with different [K⁺]_e. The negative inotropic effects of diltiazem and verapamil were potentiated when the [K⁺] of the perfusate was increased. However, nifedipine did not exhibit any noticeable potentiation of the negative inotropic effects in elevated [K⁺]_e. A previous whole-cell patch clamp study by Okuyama et al. (1994) demonstrated that the use-dependent blocking effect of diltiazem on the L-type Ca²⁺ current was potentiated by depolarization, and that of verapamil was enhanced to a lesser extent but that of nifedipine was not affected. From all of the above considerations, we speculate that the negative inotropic effects of diltiazem and verapamil, which are potentiated in elevated [K⁺]_e, contribute to the more potent protective effects of those compounds in globally ischemic hearts.

Of pathophysiological importance is an uneven distribution of [K⁺]_e in regional cardiac ischemia. The presence of inhomogeneities in myocardial [K⁺]_e within a restricted location of ischemic zone has been attributed to electrophysiological abnormalities responsible for ventricular arrythmias (Coronel et al., 1995). Previous studies (Fleet et al., 1986; Johnson et al., 1991) have shown that verapamil reduces the heterogeneity of cellular K⁺ and H⁺ loss in ischemic myocardium of pigs. The present results suggest that verapamil and diltiazem, which have a more potent effect at depolarized potential, may serve to homogenize the accumulation of [K⁺]_e and [H⁺]_e in the regional ischemia. For instance, the protective effects of these blockers may be most marked in the center of the ischemic zone where K⁺ and H⁺ efflux is greatest; the effects could be less marked in the marginal zone where the efflux are less. Consequently, the overall effect would be to homogenize [K⁺]_e and pH. Therefore, it may be reasonable to speculate that a diminution of ionic inhomogeneities by some calcium channel blockers accounts for their antiarrythmic effect, because the inhomogeneities contribute to the early arrythmias after coronary occlusion (Coronel et al., 1995).

In conclusion, we demonstrated temporal differences among the inhibitory effects of three representative types of Ca²⁺ channel blockers on ischemia-induced [K⁺]_e elevation in guinea pig isolated perfused hearts. Differences in their abilities to block the L-type Ca²⁺ channel during the ischemia-induced elevation in [K⁺]_e are likely to contribute to the differential effects of these compounds on [K⁺]_e, pH, level of high-energy phosphates, and cardiac function.