Since the initial description of the ATP-sensitive K+ (KATP) channel in cardiac myocytes by Noma (1), many studies have indicated that KATP channel plays an important role in the action potential shortening in hypoxic or ischemic myocardium (2-5). It has been suggested that activation of KATP channels during ischemia is one of the important adaptive mechanisms for the protection of ischemic myocardium (6,7). In addition, adenosine A1 and muscarinic M2 receptor stimulation activate KATP channels (8,9) and may contribute to ischemic preconditioning (10,11). During myocardial ischemia, large quantities of endogenous catecholamines are released and stimulate adrenergic receptors (12). Because adenosine also is accumulated in ischemic myocardium, the β-adrenoceptor-mediated cyclic adenosine monophosphate (cAMP)-generating response may be partly inhibited by adenosine via A1 receptor-inhibitory guanosine triphosphate (GTP)-binding protein pathway (13). Therefore an unopposed response to α-adrenergic stimulation by endogenous catecholamines may predominate in ischemic myocardium.
It has been reported that α1-adrenergic stimulation modulates several cardiac potassium currents. The transient outward current (Ito) and the inward rectifier potassium current (IK1) are inhibited (14-16), whereas the delayed rectifier potassium current (IK) is activated by α1-adrenergic stimulation (17). Contribution of α1-adrenoceptor-mediated changes in these potassium currents to the action potential configuration appears to depend on the cardiac tissues and the animal species studied. In a hypoxic or ischemic condition, the activated KATP current (IK.ATP) plays a dominant role in the repolarization of the action potential in any heart cells (2-5). However, effects of α1-adrenergic stimulation on IK.ATP, which is activated in hypoxic and ischemic heart cells, have not been evaluated. Therefore our study was conducted to determine whether α1-adrenergic stimulation modulates IK.ATP in guinea pig ventricular cells by using patch-clamp techniques. In addition, we also evaluated the influence of α1-adrenergic stimulation on the action potential shortening produced by KCOs or hypoxia in isolated papillary muscles by using conventional microelectrode techniques. We hoped that by so doing we would gain greater insight into the direct effect of α1-adrenergic stimulation on ischemic myocytes.
Action potential and tension recordings from guinea pig papillary muscles
Guinea pigs of either sex weighing 200-300 g were stunned with a blow on the head, and the hearts were quickly removed. Papillary muscles having a diameter <1 mm were dissected from the right ventricle. The preparations were transferred to a tissue bath and superfused with a modified Tyrode solution of the following composition (in millimoles per liter): NaCl, 125; KCl, 4; NaHCO3, 25; NaH2PO4, 1.8; MgCl2, 0.5; CaCl2, 2.7; and glucose, 5.5. The solution was gassed with 95% O2 plus 5% CO2, and the bath temperature was kept constant at 33.0 ± 1.0°C. One end of the preparation was hooked to an extension of the lever arm of a force transducer (Nihon Kohden TB651T, Tokyo, Japan), and the other end was pinned to bottom of the tissue bath. The length of muscle was adjusted until the resting tension was 1.96 mN. The preparation was electrically stimulated at 0.5 or 1 Hz through platinum field electrodes with rectangular pulses of 1 ms duration at twice the diastolic threshold, delivered from an electronic stimulator (Nihon Kohden S-7272B). Transmembrane potentials were recorded with 3 M KCl-filled microelectrodes (10-30 MΩ), which were connected to a high-impedance amplifier (W-P Instruments FD223, New Haven, CT, U.S.A.) with an input capacity neutralization unit (W-P Instruments FC-23). These amplified signals were displayed on an oscilloscope (Nihon Kohden VC-11), photographed by a Polaroid camera, and recorded on a chart recorder (Watanabe Sokki Mark VII, Tokyo, Japan).
In the first series of experiments, effects of methoxamine, an α1-adrenoceptor agonist, on the KCOs-induced action potential shortening were evaluated. After an equilibration period of 2 h, control recordings were made from the preparations stimulated at 0.5 Hz. The preparations were then exposed to a solution containing 1 mM nicorandil or 30 μM cromakalim. After 60-min exposure to a KCO, which made the action potential shortening reach a steady-state level, they were exposed to 100 μM methoxamine for 40 min. In part of the experiments, effects of methoxamine on the action potential and twitch tension were evaluated in the absence of any KCO. Antagonizing effects of prazosin, a nonselective α1-adrenoceptor antagonist, or WB4101, an α1A-selective antagonist, on the methoxamine-induced electrophysiologic changes were evaluated in preparations pretreated with the antagonists for 30 min. In the experiments using chloroethylclonidine (CEC), an α1B-selective antagonist, preparations were exposed to the antagonist for 30 min and subsequently superfused with a drug-free solution for 10 min before the exposure to nicorandil. It is known that CEC is an alkylating irreversible α1B-antagonist, and free CEC antagonizes α1A-receptors (18). In part of the experiments, norepinephrine (10 μM) was used instead of methoxamine (100 μM).
In the second series of experiments, effects of the α1-adrenoceptor agonist methoxamine on the hypoxia-induced action potential shortening were evaluated. Preparations were superfused with a hypoxic, glucose-free solution for 30 min under a constant electrical stimulation at 1 Hz. To obtain the hypoxic, glucose-free solution, glucose was omitted from the modified Tyrode solution, and the solution in a reservoir was gassed with 95% N2 plus 5% CO2, which resulted in Po2 values of <50 mm Hg, as previously described (5). Changes in action-potential configuration and twitch tension were observed in the hypoxic, substrate-free condition in the presence or absence of 100 μM methoxamine. In part of the experiments, glibenclamide (10 μM) was used instead of methoxamine (100 μM).
The following action potential variables were analyzed by magnification of the photographs: resting membrane potential (RMP), action potential amplitude (APA), action potential duration at 50% (APD50), and 90% repolarization levels (APD90). All the experiments were conducted in the presence of the β-adrenoceptor antagonist propranolol (1 μM). Only the experiments in which a stable impalement was maintained were used for the data analysis.
Whole-cell membrane current recordings from isolated guinea pig ventricular myocytes
Single ventricular myocytes of the guinea pig heart were obtained by enzymatic dissociation, as previously described (17). Guinea pigs (200-300 g) were anesthetized with pentobarbital sodium and ventilated with an artificial respirator. The heart was rapidly removed from the open-chest guinea pig and mounted on a modified Langendorff apparatus. The heart was retrogradely perfused with a normal HEPES-Tyrode solution (37°C). The perfusate was then changed to a nominally Ca2+-free Tyrode solution and changed to the solution containing 0.02% wt/vol collagenase (Wako, Osaka, Japan). After digestion, the heart was perfused with a high K+, low Cl-, solution [modified kraftbrühe (KB) solution] (17). Ventricular tissue was cut into small pieces in the modified KB solution. The cell suspension was filtered through a 100-μm-pore stainless-steel mesh and stored in a refrigerator (4°C) for later use. The composition of the normal HEPES-Tyrode solution was (in millimoles per liter): NaCl, 143; KCl, 5.4; CaCl2, 1.8; MgCl2, 0.5; NaH2PO4, 0.33; glucose, 5.5; and HEPES-NaOH buffer (pH 7.4), 5.0. The nominally Ca2+-free Tyrode solution was prepared by omitting CaCl2 from the normal Tyrode solution. The composition of the modified KB solution was (in millimoles per liter): KOH, 70; L-glutamic acid, 50; KCl, 40; taurine, 20; KH2PO4, 20; MgCl2, 3; glucose, 10; ethylene glycol-bis-(β-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA), 1.0; and HEPES-KOH buffer (pH 7.4), 10.
Whole-cell membrane current recordings were performed by the patch clamp method. Single ventricular myocytes were placed in a recording chamber (1 ml volume) attached to an inverted microscope (Olympus IMT-2, Tokyo, Japan) and superfused with the HEPES-Tyrode solution at a rate of 3 ml/min. The temperature of the external solution was kept constant at 36.0 ± 1.0°C. Glass patch pipettes with a tip diameter of 2-3 μm were heat-polished and filled with an internal solution composed of (in millimoles per liter): KOH, 100; L-aspartate, 100; KCl, 20; MgCl2, 1; ATP-K2, 1; phosphocreatine-K2, 1; GTP-Na, 0.1; EGTA, 10; and HEPES-KOH (pH 7.4), 5. The free Ca2+ concentration in the pipette solution was adjusted to pCa 8. The resistance of the pipette filled with the internal solution was 1-3 MΩ. After the gigaohm seal between the tip and the cell membrane was formed, the membrane patch was disrupted by applying more negative pressure to make the whole cell voltage-clamp mode.
The electrode was connected to a patch/whole cell clamp amplifier (Nihon Kohden CEZ-2300). Voltage command pulses were generated, and data were acquired by an IBM compatible computer (Compaq Prolinea 4/50 with a 200 Mbytes hard disk, Houston, TX, U.S.A.) by using pCLAMP software (Axon Instruments, Foster City, CA, U.S.A.). Current signals were digitized with a sampling interval of 2 kHz and stored on the hard disk of the computer. A liquid junction potential between the internal solution and the bath solution of-5 mV was corrected. All experiments were carried out in the presence of the β-adrenoceptor antagonist propranolol (1 μM) and the Ca2+ channel blocker nifedipine (1 μM).
A ramp pulse protocol was used to record the quasisteady state membrane current. The membrane potential was held at -40 mV and depolarized first to +50 mV at a rate of 1.2 mV/ms. It was then repolarized or hyperpolarized to -100 mV with a slope of -1.2 mV/ms, during which time the change in the membrane current was automatically plotted against the membrane potential. The voltage protocol was repeated at a frequency of once every 20 s. After the stabilization of the steady-state current induced by the ramp pulse, cells were exposed to 1 mM nicorandil or 30 μM cromakalim. After several minutes exposure to nicorandil or cromakalim, which made the current reach a steady-state level, they were exposed to 100 μM methoxamine. To examine the influence of WB4101, cells were treated with 100 nM WB4101 before the application of methoxamine, and the antagonist was present throughout the experiments. In some experiments, cells were pretreated with CEC for 30 min and subsequently washed out for 10 min before the commencement of the experiment. In part of the experiments, effects of methoxamine on the nicorandil-induced current were examined in the presence of staurosporine, an inhibitor of protein kinase C. Effects of 4β-phorbol 12-myristate 13-acetate (PMA; 100 nM), an activator of protein kinase C, on the nicorandil-induced current were also examined. Some experiments were conducted by using patch electrodes containing 20 μM inositol 1,4,5-trisphosphate (IP3) plus 5 μM inositol 1,3,4,5-tetrakisphosphate (IP4) or 100 μg/ml heparin, which inhibits the IP3-induced Ca2+ release from Ca2+ pool in vascular smooth muscle cells(19).
Drugs and chemicals
The following drugs were used: nicorandil (2-nicotinamidoethylnitrate; Chugai Pharmaceutical Co., Tokyo, Japan), glibenclamide, methoxamine hydrochloride, nifedipine, cromakalim, d-myoinositol 1,4,5-trisphosphate hexasodium (Sigma Chemical, St. Louis, MO, U.S.A.), 4β-phorbol 12-myristate 13-acetate (PMA; Nacalai Tesque Inc., Kyoto, Japan), WB4101 hydrochloride (2-(2,6-dimethoxy-phenoxyethyl)-aminomethyl-1,4-benzodioxane hydrochloride), chloroethylclonidine dihydrochloride (CEC; Funakoshi, Tokyo, Japan), prazosin hydrochloride (Pfizer, Tokyo, Japan), dl-propranolol hydrochloride, l-norepinephrine hydrogen tartrate, staurosporine, heparin sodium, dl-myoinositol 1,3,4,5-tetrakisphosphate tetrapotassium (Wako, Osaka, Japan). Glibenclamide and PMA were prepared as a stock solution of 1 mM in dimethylsulfoxide. Staurosporine and cromakalim were prepared as a stock solution of 100 μM and 100 mM, respectively, in dimethylsulfoxide. Nifedipine was dissolved in ethanol at a concentration of 1 mM, and the other drugs were dissolved in distilled water.
All values are presented in terms of mean ± SEM. Statistical analyses by Student's t test were performed for paired and unpaired observations; p values of <0.05 were considered significant.
Effects of α1-adrenergic stimulation on the action potential shortening induced by KCOs
In guinea-pig papillary muscles stimulated at 0.5 Hz, methoxamine (100 μM) did not change APD, but slightly increased twitch tension (TT), as shown in Fig. 1. In eight preparations, methoxamine per se did not affect APD, although it significantly increased TT (Table 1). Nicorandil, an ATP-sensitive K+ channel opener (KCO) (20), markedly shortened APD and decreased TT (Fig. 1). After 60-min exposure to 1 mM nicorandil, APD at 50 and 90% repolarization levels (APD50, APD90) and TT of seven preparations were significantly decreased to 46 ± 2%, 51 ± 2%, and 41 ± 5% of the control values (222 ± 11 ms, 260 ± 12 ms, and 0.44 ± 0.12 mN), respectively. Subsequent addition of 100 μM methoxamine partially reversed the nicorandil-induced decreases in APDs and TT (Fig. 1). Methoxamine increased APD50, APD90, and TT to 67 ± 3% (p < 0.005), 69 ± 3% (p < 0.005), and 92 ± 15% (p < 0.05) of the control values (Table 1). Similar results were obtained when we used cromakalim instead of nicorandil. Cromakalim at a concentration of 30 μM markedly decreased APD50, APD90, and TT to 35 ± 5% (p < 0.001), 40 ± 5% (p < 0.001), and 32 ± 4% (p < 0.05) of the control values in four preparations. Addition of 100 μM methoxamine similarly produced increases in APD50, APD90, and TT to 46 ± 4%, 50 ± 4%, and 70 ± 15% of the control values, respectively. When we stimulated α-adrenergic receptors by 10 μM norepinephrine, we could observe increases in APD50, APD90, and TT. Nicorandil (1 mM) decreased APD50, APD90, and TT to 42 ± 5%, 46 ± 4%, and 40 ± 14% of the control values, and addition of 10 μM norepinephrine produced increases in these variables to 62 ± 2%, 66 ± 1%, and 123 ± 30% of the control values, respectively (n = 3). Thus α-adrenergic stimulation prolonged APD in the presence of KCOs.
The methoxamine-induced increases in APDs and TT in nicorandil-treated preparations were inhibited by pretreatment with prazosin (1 μM), an α1-adrenoceptor antagonist (Table 1). To define further the α1-adrenoceptor subtype involved in the electrophysiologic response, additional experiments were performed by using subtype-selective antagonists, WB4101 and CEC. WB4101 at a concentration of 100 nM antagonized the effect of 100 μM methoxamine (Fig. 2). However, CEC (10 μM) failed significantly to affect the electrophysiologic responses to 100 μM methoxamine (Fig. 2). Influences of WB4101 and CEC on the methoxamine-induced increases in APDs and TT in nicorandil-treated preparations are summarized in Table 1. Thus α1A- adrenoceptor stimulation partially but significantly antagonized the action potential shortening induced by KATP channel activation.
Effects of methoxamine on the hypoxia-induced action potential shortening
The baseline values of RMP, APA, APD50, APD90, and TT of papillary muscles stimulated at 1.0 Hz were -89 ± 0 mV, 130 ± 0 mV, 179 ± 5 ms, 212 ± 5 ms, and 0.70 ± 0.12 mN (n = 29), respectively. Superfusion of hypoxic, glucose-free solution produced gradual decreases in APDs and TT (Table 2). APD50 and APD90 were significantly decreased to 30 ± 3% and 38 ± 2% of the control values after 30-min hypoxia in untreated preparations (n = 10). In the presence of 100 μM methoxamine, the hypoxia-induced APD90 shortening was significantly retarded up to 10 min, although the APD90 shortening became close to that of the control group at 30 min (Table 2). In the presence of 10 μM glibenclamide, the hypoxia-induced APD shortening was significantly retarded up to 30 min. In control preparations, TT was significantly decreased to 16 ± 4% of the control value at 30 min. In methoxamine-treated (n = 9) and glibenclamide-treated preparations (n = 10), TT was decreased to 35 ± 14% and 52 ± 14% of the control values, respectively. The decrease in TT in the presence of methoxamine or glibenclamide was significantly less than that in control preparations (Table 2). Thus α1-adrenergic stimulation attenuated the action potential shortening in the hypoxic and substrate-free condition, although the effect was less marked than that of glibenclamide.
Effects of methoxamine on the KCOs-induced outward current
To examine effects of α1-adrenergic stimulation on IK.ATP, whole-cell membrane currents were recorded by using patch-clamp techniques in isolated guinea pig ventricular cells. A ramp pulse protocol of 125 ms from 50 mV to -100 mV was used to record the quasi-steady-state current. As shown in Fig. 3A, nicorandil (1 mM) produced a marked increase in the quasi-steady-state outward current at potentials positive to ≈-40 mV. The outward current induced by nicorandil was completely inhibited by 1 μM glibenclamide, indicating that the current was IK.ATP. The outward current at potentials negative to ≈-40 mV and the inward current below the reversal potential were slightly inhibited by nicorandil, suggesting that nicorandil might inhibit the inward rectifier K+ current (IK1), as reported by Hiraoka and Fan (20). Methoxamine (100 μM) significantly inhibited the nicorandil-induced outward current. The exposure to methoxamine reduced the outward current at 0 mV by 18 ± 5% in 14 nicorandil-treated cells (from 1.23 ± 0.18 nA to 1.00 ± 0.16 nA, p < 0.005) (Fig. 3B). Similar results were obtained when we used cromakalim instead of nicorandil. Cromakalim at a concentration of 30 μM markedly increased the glibenclamide (1 μM)-sensitive outward current and methoxamine (100 μM) significantly inhibited the cromakalim-induced outward current. The exposure to methoxamine reduced the outward current at 0 mV by 16 ± 2% in 22 cromakalim-treated cells (from 1.28 ± 0.12 nA to 1.10 ± 0.10 nA, p < 0.001).
As observed in the experiments using papillary muscles, the inhibition of the nicorandil-induced current by methoxamine was antagonized by the α1A-antagonist WB4101 but not by the α1B-antagonist CEC. In the presence of 100 nM WB4101, the inhibition by 100 μM methoxamine of the nicorandil-induced current at 0 mV was 5 ± 4% (n = 16, NS), which was statistically different from the inhibition (18 ± 5%) by methoxamine alone (p < 0.05). In the CEC (10 μM)-treated cells, the inhibition of the nicorandil-induced current was 8 ± 3% (n = 15, p < 0.05), which was not significantly different from the inhibition by methoxamine alone.
To evaluate the intracellular mechanism(s) involved in methoxamine-induced inhibition of IK.ATP, further experiments were conducted in isolated ventricular myocytes. Staurosporine (30 nM), a protein kinase C inhibitor, failed to affect the inhibition of the nicorandil-induced outward current by methoxamine (Fig. 3C). Even in the presence of staurosporine methoxamine inhibited the nicorandil-induced IK.ATP at 0 mV level by 20 ± 3% (n = 12, p < 0.001), which was not significantly different from the methoxamine-induced inhibition in the absence of the protein kinase C inhibitor (Table 3). PMA (100 nM), a protein kinase C activator, also failed significantly to inhibit the nicorandil-induced outward current in 12 cells (Fig. 3D and Table 3). Intracellular perfusion of cells with a combination of 20 μM IP3 and 5 μM IP4 failed to affect the methoxamine-induced inhibition of nicoradil-induced IK.ATP in 14 cells (Fig. 3E and Table 3). In the presence of IP3 and IP4 in the pipette solution, methoxamine inhibited the IK.ATP by 20 ± 3%, which was not significantly different from the inhibition without these substances. Intracellular loading of heparin (100 μg/ml), an inhibitor of IP3 dependent Ca2+ release, failed to affect the inhibitory action of methoxamine on the nicorandil-induced outward current either (Fig. 3F). In the presence of heparin in the pipette solution (100 μg/ml), methoxamine still produced a significant inhibition of the nicorandil-induced IK.ATP (n = 14, 19 ± 2%, p < 0.001; Table 3). Thus neither protein kinase C activation nor IP3 formation appeared to be involved in the α1A-adrenoceptor-mediated inhibition of IK.ATP.
α1A-Adrenoceptor-mediated inhibition of IK.ATP
Increasing evidence suggests that activation of KATP channels plays an important role in the action potential shortening and cellular K+ loss in the ischemic myocardium (2,3,21). The KATP channel is a target protein for various vasorelaxant agents such as cromakalim, pinacidil, and nicorandil (20,22,23), and activation of the KATP channel by these drugs is supposed to be responsible for the vasodilation and the action potential shortening (24). In our study, either nicorandil or cromakalim shortened APD and increased a steady state outward current. Because the outward current was blocked by glibenclamide, the outward current could be designated as IK.ATP. It has been demonstrated that these KCOs activate KATP channels in guinea-pig ventricular cells (20,22).
In our study, α1-adrenergic stimulation by methoxamine partially reversed the action potential shortening produced by nicorandil or cromakalim, although methoxamine per se hardly affected APD in the absence of any KCO. Similar results were obtained with changes in TT. α1-Adrenergic stimulation by methoxamine partially reversed the decreases in TT by KCOs. It can be interpreted that KCOs might reduce Ca2+ entry through L-type Ca2+ channel by shortening APD, and methoxamine partially offset the effect of KCOs. The APD-prolonging effect of methoxamine was inhibited by the α1-antagonist prazosin. Such an APD-prologing effect in the presence of nicorandil was also shared with norepinephrine. These findings suggest that α1-adrenergic stimulation inhibits IK.ATP. However, we must take into account the nonspecific actions of α1-adrenoceptor agonists and antagonists. It has been reported that the α-antagonist phentolamine blocks KATP channels in cardiac ventricular cells independent of its effects on the α-adrenoceptors (25). One may argue that methoxamine might prolong APD directly not via α1-adrenoceptors. However, it is likely that the effect of methoxamine occurs via the stimulation of α1-adrenoceptors because the methoxamine effect was antagonized by prazosin and shared with norepinephrine.
More direct evidence was obtained from our patch-clamp experiments in which methoxamine inhibited the whole-cell outward current induced by nicorandil or cromakalim in guinea pig ventricular cells. Recent molecular cloning (26), radioligand binding (27), and functional studies (28) indicated that there are at least two pharmacologically distinct subtypes, α1A-and α1B-adrenoceptors. α1A-Adrenoceptors show a high affinity for WB4101, a competitive antagonist, whereas α1B-adrenoceptors are readily blocked by CEC, an irreversible alkylating antagonist. In this study, the inhibitory effect of methoxamine on IK.ATP was attenuated by WB4101 but not by CEC. Therefore the inhibition of IK.ATP would be mediated by α1A-adrenoceptors.
Either nicorandil or cromakalim markedly increased the outward current positive to -40 mV. The outward current was readily blocked by 1 μM glibenclamide, a sulfonylurea KATP channel blocker, suggesting that the activated current positive to -40 mV would be IK.ATP. However, the outward current for potentials negative to -50 mV was reduced rather than augmented by KCOs. The decrease in the outward current negative to -50 mV might be due to a direct inhibitory action of KCOs on the inward rectifier K+ current (IK1). Similar inhibition of IK1 by KCO was also reported by others (20). The repolarization of action potential was markedly accelerated by KCOs, even though IK1 negative to -50 mV was depressed. One possible explanation may be that a marked increase in the outward current positive to -40 mV, where the inward Ca2+ current also flows, plays a more prominent role in the repolarization of the action potential than the outward current negative to -50 mV. However, this consideration is only speculative, and further experimentation using the action potential clamp method may be needed precisely to correlate the change in the membrane current with the alterations of action potential configuration.
Subcellular mechanisms underlying α1-adrenoceptor-mediated modulation of cardiac K+ channels
α1-Adrenergic stimulation is known to modulate several kinds of cardiac K+ channels. It has been reported that the transient outward current (I10) is inhibited by α1-agonists in rabbit atrial cells and rat ventricular cells (14,15). α1-Adrenergic stimulation was also reported to inhibit IK1 in rabbit atrial and ventricular cells (16,29). On the other hand, the delayed rectifier K+ current (IK) was enhanced by α1-adrenergic stimulation in guinea-pig ventricular cells (17). In terms of muscarinic K+ channel current (IK.ACh), conflicting results have been reported; Kurachi et al. (30) reported the α1-adrenoceptor-mediated activation of IK.ACh in guinea-pig atrial cells, whereas Braun et al. (29) reported the α1-adrenoceptor-mediated inhibition of the current in rabbit atrial cells. Among these K+ channel currents, only the intracellular mechanism for α1-adrenergic regulation of IK, which is mediated by protein kinase C activation, has been clarified.
The present study has shown for the first time that α1-adrenergic stimulation partially inhibits IK.ATP in guinea-pig ventricular cells. Recently it was reported that both α1A- and α1B-receptors are coupled to the hydrolysis of phosphoinositide (28). Therefore we tried to determine the intracellular mechanism(s) by which α1-adrenergic stimulation inhibits the nicorandil-induced IK.ATP. The protein kinase C inhibitor staurosporine could not abolish the methoxamine-induced inhibition of IK.ATP. In addition, PMA failed to mimic the inhibitory action of methoxamine. The methoxamine-induced inhibition of IK.ATP was not affected by intracellular perfusion of 20 μM IP3 plus 5 μM IP4. In addition, the inhibition of IK.ATP by methoxamine was not affected by intracellular perfusion of heparin, a blocker of IP3-induced Ca2+ release from the pool (19). Therefore, neither protein kinase C activation nor IP3 formation might be involved in the α1-adrenoceptor-mediated inhibition of IK.ATP. Braun et al. (31) also examined underlying mechanism(s) for the α1-adrenoceptor-mediated inhibition of Ito and reached a similar conclusion that the Ito inhibition could not be attributed to the protein kinase C activation or the formation of inositol polyphosphates. Further studies are needed to determine the underlying signal-transduction system by which α1-adrenergic stimulation inhibits these K+ currents.
It has been demonstrated that cardiac KATP channels are modulated by several receptor mechanisms. Kirsch et al. (8) reported that adenosine A1 receptor stimulation activates KATP channels via pertussis toxin-sensitive G protein pathway in neonatal rat ventricular myocytes. Ito et al. (9) also demonstrated that activation of A1 receptors or muscarinic M2 receptors results in an increase in KATP channel activity via Gi proteins in guinea pig atrial and ventricular myocytes. Opening of KATP channels mediated by adenosine A1 receptors or muscarinic M2 receptors is suggested to be an important mechanism of ischemic preconditioning (10,11). A recent report has indicated that β-adrenergic stimulation enhances IK.ATP in feline ventricular cells dialyzed with ATP-free solution via Gs- and adenylate cyclase-dependent, cAMP- and protein kinase A-independent pathway (32). In addition, it has been also reported that in canine ventricular cells dialyzed with high levels of intracellular ATP β-agonists enhance the pinacidil-induced IK.ATP via a cAMP-dependent mechanism (33). Therefore it is of interest that α- and β-adrenergic stimulation produce directionally opposite effects on cardiac KATP channels.
Pathophysiologic significance of α1-adrenoceptor-mediated IK.ATP inhibition
In our study, methoxamine partially retarded the action potential shortening in guinea pig papillary muscles superfused with a hypoxic, glucose-free solution. In addition, glibenclamide, a KATP channel blocker, also inhibited the action potential shortening from the early phase of the hypoxic, glucose-free condition in guineapig papillary muscles. These findings are consistent with the concept that KATP channels are activated from the early phase of myocardial hypoxia or ischemia (3). However, the prolonging effects of methoxamine and glibenclamide on APD under hypoxic condition were rather limited in our experiments. Although glibenclamide almost completely inhibited the KCOs-induced IK.ATP in isolated ventricular cells, it incompletely prevented the hypoxia-induced action potential shortening. It has been reported that the inhibitory effect of glibenclamide on the opening of cardiac KATP channels is impaired under severe metabolic stress (34). Changes in other membrane currents might also be involved in the action potential shortening under the hypoxic, glucose-free condition (5).
It is acknowledged that KATP channel activation may be an endogenous adaptive mechanism and a cardioprotective mechanism of KCOs (6,7,35). The action potential shortening due to activation of KATP channels is expected to reduce Ca2+ entry through L-type Ca2+ channels and mechanical contraction, leading to energy sparing. Therefore α1-adrenoceptor-mediated inhibition of KATP channels may increase ischemic damage and partly offset the cardioprotective effect of KCOs because of an indirectly enhanced Ca2+ influx. In fact, methoxamine partially antagonized the KCOs-induced and hypoxia-induced decreases in TT in our study.
Recently Kitakaze et al. (36) reported that α1-adrenergic stimulation can mimic ischemic preconditioning, resulting in a decrease in infarct size in anesthetized dogs. According to their hypothesis, α1-adrenergic stimulation increases 5′-nucleotidase activity and adenosine production. As already discussed, adenosine A1-receptor stimulation may activate cardiac KATP channels (8,9) and protect ischemic myocardium. Therefore during α1-adrenergic stimulation, indirect activation of KATP channels through adenosine production might predominate over direct inhibition of the channels, at least, in dogs.
This study has demonstrated for the first time that α1A-adrenoceptor stimulation partially inhibits IK.ATP. α1A-Adrenergic stimulation may produce a direct deleterious effect on the ischemic myocardium and partly offset the beneficial effect of KCOs.
Acknowledgment: We thank Mr. M. Tamagawa and Mr. I. Sakurada for their excellent technical assistance and Ms. I. Sakashita for her secretarial assistance. This work was supported in part by Grant-in-Aid from the Ministry of Education, Science and Culture of Japan to H.N.
1. Noma A. ATP-regulated K+
channels in cardiac muscle. Nature
2. Nichols CG, Ripoll C, Lederer JW. ATP-sensitive potassium channel modulation of the guinea pig ventricular action potential and contraction. Circ Res
3. Venkatesh N, Lamp ST, Weiss JN. Sulfonylureas, ATP-sensitive K+
channels, and cellular K+
loss during hypoxia, ischemia, and metabolic inhibition in mammalian ventricle. Circ Res
4. Benndorf K, Friedrich M, Hirche HJ. Anoxia opens ATP-regulated K channels in isolated heart cells of the guinea pig. Pflugers Arch
5. 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
6. Grover GJ, Mccullough JR, Henry DE, Conder ML, Sleph PG. Antiischemic effects of the potassium channel activators pinacidil and cromakalim and the reversal of these effects with the potassium channel blocker glyburide. J Pharmacol Exp Ther
7. Cole WC, Mcpherson CD, Sontag D. ATP-regulated K+
channels protect the myocardium against ischemia/reperfusion damage. Circ Res
8. Kirsch GE, Codina J, Birnbaumer L, Brown AM. Coupling of ATP-sensitive K+
channels to A1
receptors by G proteins in rat ventricular myocytes. Am J Physiol
9. Ito H, Tung RT, Sugimoto T, et al. On the mechanism of G protein βγ subunit activation of the muscarinic K+
channel in guinea pig atrial cell membrane. J Gen Physiol
10. Auchampach JA, Gross GJ. Adenosine A1
channels, and ischemic preconditioning in dogs. Am J Physiol
11. Yao Z, Gross GJ. Acetylcholine mimics ischemic preconditioning via a glibenclamide-sensitive mechanism in dogs. Am J Physiol
12. Schömig A, Dart AM, Diets R, Mayer E, Kübler W. Release of endogenous catecholamines in the ischemic myocardium of the rat. Part A: locally mediated release. Circ Res
13. Belardinelli L, Linden J, Berne RM. The cardiac effect of adenosine. Prog Cardiovasc Dis
14. Apkon M, Nerbonne JM. α1
-Adrenergic agonists selectively suppress voltage-dependent K+
current in rat ventricular myocytes. Proc Natl Acad Sci U S A
15. Tohse N, Nakaya H, Hattori Y, Endou M, Kanno M. Inhibitory effect mediated by α1
-adrenoceptors on transient outward current in isolated rat ventricular cells. Pflugers Arch
16. Fedida D, Braun AP, Giles WR. α1
-Adrenoceptors reduce background K+
current in rabbit ventricular myocytes. J Physiol
17. Tohse N, Nakaya H, Kanno M. α1
-Adrenoceptor stimulation enhances the delayed rectifier K+
current of guinea pig ventricular cells through the activation of protein kinase C. Circ Res
18. Han C, Abel PW, Minneman KP. Heterogeneity of α1
-adrenergic receptors revealed by chlorethylclonidine. Mol Pharmacol
19. Kobayashi S, Somlyo AV, Somlyo AP. Heparin inhibits the inositol 1,4,5,-trisphosphate-dependent, but not independent, calcium release induced by guanine nucleotide in vascular smooth muscle. Biochem Biophys Res Commun
20. Hiraoka M, Fan Z. Activation of ATP-sensitive outward K+
current by nicorandil (2-nicotinamidoethyl nitrate) in isolated ventricular myocytes. J Pharmacol Exp Ther
21. Wilde AAM, Escande D, Schumacher CA, et al. Potassium accumulation in the globally ischemic mammalian heart: a role for the ATP-sensitive K+
channel. Circ Res
22. Sanguinetti MC, Scott AL, Zingaro GJ, Siegel PKS. BRL34915 (cromakalim) activates ATP-sensitive K+
current in cardiac muscle. Proc Natl Acad Sci U S A
23. Arena JP, Kass RS. Enhancement of potassium-sensitive current in heart cells by pinacidil: evidence for modulation of the ATP-sensitive potassium channel. Circ Res
24. Edwards G, Weston AH. The pharmacology of ATP-sensitive potassium channels. Annu Rev Pharmacol Toxicol
25. Wilde AAM, Veldkamp MW, Van Ginneken ACG, Opthof T. Phentolamine blocks ATP sensitive potassium channels in cardiac ventricular cells. Cardiovasc Res
26. Lefkowitz RJ, Caron MG. The adrenergic receptors. In: Nishizuka Y, Endoh M, Tanaka C, eds. The biology and medicine of signal transduction. New York: Raven Press, 1990:1-8.
27. Morrow AL, Creese I. Characterization of α1
-adrenergic receptor subtypes in rat brain: a reevaluation of [3
H]WB4101 and [3
H]prazosin binding sites. Mol Pharmacol
28. Yang HT, Endoh M. Dissociation of the positive inotropic effect of methoxamine
from the hydrolysis of phosphoinositide in rabbit ventricular myocardium: a comparison with the effects of phenyle-phrine and the subtype of the alpha-1 adrenoceptor involved. Pharmacol Exp Ther
29. Braun AP, Fedida D, Giles WR. Activation of α1
-adrenoceptors modulates the inwardly rectifying potassium currents of mammalian atrial myocytes. Pflugers Arch
30. Kurachi Y, Ito H, Sugimoto T, Shimizu T, Miki I, Ui M. α-Adrenergic activation of the muscarinic K+
channel is mediated by arachidonic acid metabolites. Pflugers Arch
31. Braun AP, Fedida D, Clark RB, Giles WR. Intracellular mechanisms for α1
-adrenergic regulation of the transient outward current in rabbit atrial myocytes. J Physiol
32. Schackow TE, Ten Eick RE. Enhancement of ATP-sensitive potassium current in cat ventricular myocytes by β-adrenoreceptor stimulation. J Physiol
33. Tseng GN, Hoffman BF. Action of pinacidil on membrane currents in canine ventricular myocytes and their modulation by intracellular ATP and cAMP. Pflugers Arch
34. Findlay I. Sulphonylurea drugs no longer inhibit ATP-sensitive K+
channels during metabolic stress in cardiac muscle. J Pharmacol Exp Ther
35. Yao Z, Gross GJ. Effects of the KATP
channel opener bimakalim on coronary blood flow, monophasic action potential duration, and infarct size in dogs. Circulation
36. Kitakaze M, Hori M, Kamada T. Role of adenosine and its interaction with α-adrenoceptor activity in ischemic and reperfusion injury of myocardium. Cardiovasc Res