Aprikalim Reduces the Na -Ca 2 Exchange Outward Current Enhanced by Hyperkalemia in Rat Ventricular Myocytes

Transcription

1 Aprikalim Reduces the Na -Ca 2 Exchange Outward Current Enhanced by Hyperkalemia in Rat Ventricular Myocytes Hong-Yu Li, PhD, Song Wu, PhD, Guo-Wei He, MD, PhD, and Tak-Ming Wong, PhD Department of Physiology, Faculty of Medicine, University of Hong Kong, and Department of Surgery, Faculty of Medicine, Chinese University of Hong Kong, Hong Kong SAR, China Bacground. Aprikalim, an adenosine triphosphate (ATP) sensitive K (K ATP ) channel opener, attenuates the elevation of intracellular Ca 2 concentration ([Ca 2 ] i ) and improves the contractile functions after hyperkalemic and hypothermic cardioplegia. There is evidence that cardioplegia increases the Na -Ca 2 exchange activity without affecting Ca 2 influx through L-type Ca 2 channels or Ca 2 content in the sarcoplasmic reticulum, the intracellular Ca 2 store. Methods. We measured the Na -Ca 2 exchange outward current with the patch-clamp technique in single rat ventricular myocytes exposed to hyperkalemia and hypothermia in the presence of aprikalim. The intracellular calcium concentration ([Ca 2 ] i ) during cardioplegia, and the contractile function and [Ca 2 ] i transients induced by electrical stimulation or caffeine during rewarming and reperfusion in single ventricular myocytes were also determined. Contraction and [Ca 2 ] i were determined with video tracking and spectrofluorometry, respectively. Results. Aprikalim, 100 mol/l, the effect of which was blocked by glibamclamide, a K ATP inhibitor, significantly attenuated the hyperkalemia-elevated Na -Ca 2 exchange current by 26% and 11% at 22 C and 4 C, respectively. Aprikalim also attenuated significantly the [Ca 2 ] i elevated during cardioplegia. Furthermore aprikalim significantly attenuated the reduction in amplitude and prolongation in duration of contraction of myocytes after cardioplegia. The effects of aprikalim mimicked those of nickle (Ni 2 ),ana -Ca 2 exchange blocker. The electrically or caffeine-induced [Ca 2 ] i transients were unaltered by cardioplegia or aprikalim. Conclusions. Aprikalim attenuates the Na -Ca 2 exchange outward current elevated by hyperkalemia, which may attenuate the [Ca 2 ] i elevation during hyperkalemia and improve the contractile function after cardioplegia in the ventricular myocyte. The study provides further support that addition of a K ATP channel opener to the cardioplegic solution may produce beneficial effects in open heart surgery. (Ann Thorac Surg 2002;73: ) 2002 by The Society of Thoracic Surgeons Accepted for publication Dec 18, Address reprint requests to Dr Wong, Department of Physiology, University of Hong Kong, 4/F Laboratory Block, Faculty of Medicine Building, 21 Sassoon Rd, Hong Kong SAR, China; wongtakm@hkucc.hku.hk. Although hyperkalemic/hypothermic cardioplegia has been a standard means to induce cardiac arrest for cardiac surgery for many years, ventricular dysfunction occurs during the subsequent rewarming/reperfusion [1]. Previous studies in single ventricular myocytes have shown that after hyperkalemic/hypothermic cardioplegic arrest, the contractile function of the myocytes is reduced upon rewarming/reperfusion [2]. The impaired contractile function is most likely related to alterations in intracellular calcium homeostasis, as intracellular calcium concentration ([Ca 2 ] i ) is significantly higher in cardiomyocytes exposed to hyperkalemia/ hypothermia [3, 4]. The increase in [Ca 2 ] i was prevented by nickle (Ni 2 ), known to block the Na -Ca 2 exchange [3, 5], which pumps Ca 2 in or out of the ventricular myocyte in exchange with Na. On the other hand blockade of the L-type Ca 2 channel had no effect on [Ca 2 ] i during cardioplegia [3]; nor was there any alteration in [Ca 2 ] i transient induced by caffeine [3], known to deplete Ca 2 of sarcoplasmic reticulum (SR) [6], the intracellular Ca 2 store. The observations suggest that the increase in [Ca 2 ] i after cardioplegia may be due to an increased Na -Ca 2 exchange activity, but not to alterations in influx of Ca 2 through the L-type Ca 2 channel or mobilization of Ca 2 from SR. Recently there is evidence that aprikalim (APK), an ATP-sensitive potassium (K ) channel opener (PCO) [7], improved contractile function during rewarming/ reperfusion after cardioplegia [4]. It has also been shown that under room temperature (22 C) [8] or at 4 C [4], APK inhibited the elevation in [Ca 2 ] i by hyperkalemia in ventricular myocytes, suggesting that the ATP-sensitive PCO may improve the contractile recovery by attenuating the elevation in [Ca 2 ] i. As the increase in [Ca 2 ] i after cardioplegia may be due to an increased Na -Ca 2 exchange activity, we hypothesized that APK may attenuate the Na -Ca 2 exchange activity. To test this hypothesis we investigated the effects of APK, at a cardioprotective concentration [4], on Na -Ca 2 exchange outward current during hyperkalemia at 22 C or 4 C in single 2002 by The Society of Thoracic Surgeons /02/$22.00 Published by Elsevier Science Inc PII S (02)

2 1254 LI ET AL Ann Thorac Surg APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 2002;73: ventricular myocytes. We used the patch-clamp technique, which measured the ionic fluxes across the sarcolemma, according to Kimura and associates [5, 9]. To correlate the Na -Ca 2 exchange activity with the function, we determined the effects of APK on contractile recovery, and [Ca 2 ] i during cardioplegia in Krebs solution or in a solution with compositions comparable to those for the patch-clamp. Results showed that APK attenuates the outward Na -Ca 2 exchange current enhanced by hyperkalemia, which may be mainly responsible for the beneficial effects of the ATP-sensitive PCO on [Ca 2 ] i during, and contractile recovery after, cardioplegia. Material and Methods Preparation of Single Ventricular Myocytes Ventricular myocytes were isolated from hearts of adult male Sprague-Dawley rats weighing 190 to 230 g using a collagenase method described previously [10]. Permission for the use of animals was obtained from the Department of Health, Hong Kong Regional Office, the Government of the Hong Kong SAR. Measurement of Na -Ca 2 Exchange Current Na -Ca 2 exchange current was recorded with the whole-cell patch clamp technique according to the method of Kimura and associates [5, 9]. Membrane potential was controlled with a patch-clamp amplifier (Axopatch-1C; Axon Instruments Inc, Foster City, CA). Ramp clamp pulses were used [5, 10]. The pulse shape was composed of three phases: an initial 70 mv depolarizing phase from the holding potential 70 mv, second hyperpolarization of 120 mv and then returning to the holding potential at a speed 90 mv/0.5 to 1.0 second. Myocytes were randomly assigned to one of the four groups: (1) APK 22 C (n 15); (2) APK 4 C (n 6); (3) Ni 2 at 22 C (n 3); and (4) Ni 2 at 4 C (n 6). All recordings were successively made from one single cell with the sequence: Control 3 High [K ] o 3 High [K ] o APK (or Ni 2 ). The experimental protocols are shown in Figure 1A. The internal solution composition (mmol/l) was NaCl 20, CsCl 60, MgCl 2 1.0, CaCl 2 1.0, HEPES 5.0, MgATP 10, and aspartic acid 50; ph 7.4 with CsOH. MgATP was added before experiment. The bath solution contained (mmol/l): NaCl 140, CaCl 2 1.0, CsCl 2.0, BaCl 2 1.0, MgCl 2 2.0, HEPES 5.0, and glucose 5.0; ph 7.4 with NaOH. In the external solution, 2 mmol/l CsCl (cesium chloride) and 1 mmol/l BaCl 2 (barium chloride), 20 mol/l ouabain and 2 mol/l verapamil were used to block K channels (K IR, K V ), the Na -K ATPase, and the L-type Ca 2 channel, respectively. In the internal solution 20 mmol/l tetraethylammonium chloride (TEA) was included to block the K ca channel. With the blockade of most of the timedependent currents, the remaining conductance was Na -Ca 2 exchange current, which showed a linear I-V relation (Fig 1B) similar to those reported previously [5, 9]. To prevent the cells from contracture by Ca 2 entering in exchange for Na, 42 mmol/l EGTA was added into the pipette solution, thus increasing the buffering capacity to Ca 2. Hyperkalemic cardioplegic-solution was imitated by simply increasing the extracellular K ([K ] o )to 20 mmol/l. Voltage-clamp protocols, data acquisition, and data storage were accomplished using pclamp 6.0 (Axon Instruments). Membrane currents were sampled at 16 khz by a 12-bit A/D converter (Digidat 1200B, Axon Instruments). Solution changes were completed within 2 seconds as measured previously with dyes. Protocols for Determining [Ca 2 ] i and Contractile Function For measuring [Ca 2 ] i during cardioplegia, myocytes were randomly assigned to one of the six groups (Fig 2A). (1) Krebs 22 C control (n 25): Krebs solution (mmol/l) with Na 143, Cl - 125, K 5, Ca 2 1, Mg 2 1.2, HCO , Glucose 11, ph 7.4 for 1 hour at 22 C. (2) Cardioplegia (n 16): Krebs solution containing 20 meq/l K for 1 hour at 4 C. (3) PCO/cardioplegia (n 15): same as group 2 except that 100 mol/l APK was added. (4) Ni 2 / cardioplegia (n 9): same as group 2 except that 5 mmol/l NiCl 2 was added. (5) PCO/cardioplegia/patch (n 6): same as group 3 except that 2 mmol/l CsCl, 1 mmol/l BaCl 2,20 mol/l ouabain, and 2 mol/l verapamil were added. (6) PCO/cardioplegia/glibenclamide (n 8): same as group 3 except that 1 mol/l glibenclamide was added. After cardioplegia myocytes were reperfused with normal Krebs at 22 C, rewarming/ reperfusion. [Ca 2 ] i was determined before, during, and after cardioplegia whereas [Ca 2 ] i transients and contraction were determined only before and after cardioplegia. For determination of contractile recovery, experiments were performed at 37 C to provide data in normothermic conditions as well as in 22 C to provide data under comparable conditions as in experiments of Na -Ca 2 exchange current. There were three groups at 37 C: (1) Krebs 37 C control; (2) Krebs 37 C after cardioplegia 4 C; and (3) Krebs 37 C after PCO/cardioplegia 4 C. For experiments at 22 C, in addition to these three groups, there were other two groups, namely Krebs 22 C after Ni 2 /cardioplegia 4 C and Krebs 22 C after PCO/ cardioplegia/patch 4 C (Fig 3). Measurement of Contraction in Single Ventricular Myocytes At 5 minutes into reperfusion the contraction in single myocytes was measured by an optical video system with an automatic video-analyzer as previously described [10]. Briefly, a rapid imaging was employed to determine changes in length of myocytes placed under an inverted microscope and stimulated with 0.2 Hz. The image of the contracting cell was recorded with a video-camera and displayed on a television monitor, while the twitch traces were recorded on a chart recorder. The amplitude and duration of contraction were taken as indices of contractile function.

3 Ann Thorac Surg LI ET AL 2002;73: APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 1255 Fig 1. Effects of aprikalim (APK) and Ni 2 on the Na -Ca 2 exchange outward current activated by high potassium, in single isolated rat ventricular myocytes. (A) The experimental protocols of the study. Recording was successively made throughout the perfusion. (APK aprikalim 100 mol/l; HK high potassium solution, [K ] o 20 mmol/l.) (B) Typical current traces showing the effects of APK on the Na -Ca 2 exchange outward current elicited by a ramp pulse during hyperkalemia ([K ] o 20 mmol/l) and hypothermia (4 C). Recordings were made successively from one single cell. (Solid box is high [K ] o at 4 C; hatched box is high [K ] o APK at 4 C; and open box is control at 4 C.) (C) Group results showing effects of APK (hatched bars) on the outward Na -Ca 2 exchange current activated by high [K ] o (solid bars) at 22 C (n 15) and 4 C (n 6). *p versus control; #, ## p 0.01, versus corresponding high [K ] o ; p 0.05 versus high [K ] o at 22 C. (D) Group result showing effects of 5 mmol/l Ni 2 (hatched bars) on the Na -Ca 2 exchange outward current activated by high potassium (solid bars) at 22 C (n 3) and 4 C (n 6). *p versus control; #, ## p 0.05, 0.01 versus corresponding high [K ] o ; p 0.05 versus high [K ] o at 22 C. 4 Measurement of [Ca 2 ] i The control and rewarming/reperfusion was in room temperature (22 C). Ventricular myocytes were loaded with a Ca 2 indicator Fura-2/AM (4 mol/l). [Ca 2 ] i concentration was measured with a spectrofluorometric method described previously [10], and the [Ca 2 ] i level was recorded successively throughout each experiment (Figure 2A). For caffeine-induced [Ca 2 ] i transient, 10 mmol/l caffeine was applied directly to the myocyte [11]. Both electrically and caffeine-induced [Ca 2 ] i transient were measured during the control and rewarming/ reperfusion periods. Drugs and Chemicals Fura-2/AM, type-i collagenase, caffeine, ouabain, verapamil, nickel-chloride (NiCl 2 ), glibenclamide, aspartic acid, TEA, and MgATP were purchased from Sigma. APK was generously provided by Rhodes-Poulenc Rorer Research-Development, France. All chemicals were dissolved in distilled water except Fura-2/AM and glibenclamide were first dissolved in dimethyl sulfoxide (DMSO) and APK in ethanol, respectively, and then diluted with normal MEM or Krebs before experiment. The final concentrations of DMSO and ethanol in this study had no effect on the myocyte. The concentrations of APK [4, 7], caffeine [11], ouabain [5, 9], NiCl 2 [3, 5], glibenclamide [4], and verapamil [12] chosen in this study were according to the previous studies. Statistical Analysis All values were presented as means SEM. Paired Student s t test was used to determine the difference in the same cells with different treatments and unpaired Student s t test was used between two groups. One-way analysis of variance (ANOVA) was used for determining the difference among the groups. Software of Excel (97 SR-2, Microsoft Corporation) was used for all of the statistics.

4 1256 LI ET AL Ann Thorac Surg APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 2002;73: Fig 2. Effects of aprikalim (APK) and Ni 2 on [Ca 2 ] i level during hyperkalemic/hypothermic cardioplegia in single isolated rat ventricular myocytes. (A) The experimental protocols. Recording was successively made throughout the perfusion. Patch is a solution with CsCl 2 mmol/l, BaCl 2 1 mmol/l, ouabain 20 mol/l, and verapamil 2 mol/l, which were present in the bath solution for the patch clamp experiments. (HK high potassium solution.) (B) Representative traces of successive recording of [Ca 2 ] i before, during, and after 1 hour of hyperkalemic (20 mmol/l [K ] o ) and hypothermic (4 C) cardioplegia. (C) Group results showing the effects of 100 mol/l APK and 5 mmol/l Ni 2 on [Ca 2 ] i during hyperkalemic/ hypothermic cardioplegia. The effects of APK were also determined in the presence of 1 mol/l glibenclamide (GBD) or in a solution with CsCl 2 mmol/l, BaCl 2 1 mmol/l, ouabain 20 mol/l, and verapamil 2 mol/l, which were present in the bath solution for the patch clamp experiments. Values are expressed as percentage of control (mean SEM). (n 25 in the Krebs 22 C group; n 16 in the high [K ] o 4 C group; n 9 in the high [K ] o 4 C Ni 2 group; n 15 in the high [K ] o 4 C APK group; n 6inthe high [K ] o 4 C APK patch solution group; and n 8inthe high [K ] o 4 C APK GBD group. *p versus Krebs 22 C; #, ##, ### p 0.05, 0.01, 0.001, respectively, versus high [K ] o at 4 C; p 0.05 versus high [K ] o 4 C APK.) 4 (p 0.01; Fig 1B and C). The effect was similar to that of 5 mmol/l Ni 2 (Fig 1D), known to block the Na -Ca 2 exchange activity [3, 5]. APK produced greater inhibitory effect at 22 C (26.2% 5.5%) than at 4 C (11.1% 2.3%), and the difference was significant (p 0.05). This is in agreement with the previous finding that the effect of PCOs is temperature dependent [8, 13, 14]. Results Effects of APK on Na -Ca 2 Exchange Outward Current During Hyperkalemic/hypothermic Cardioplegia in Single Ventricular Myocytes As shown in Figure 1B and C, the Na -Ca 2 exchange outward current was increased significantly under the cardioplegic condition with 20 mmol/l [K ] o. The current was lower at 4 C than at 22 C (p 0.05), which is in keeping with the previous finding [5]. APK, 100 mol/l, which itself did not produce any effect in normal Krebs solution at 22 C (data not shown), significantly attenuated the Na -Ca 2 exchange outward current enhanced by hyperkalemia at both 22 C (p 0.001; Fig 1C) and 4 C Effects of APK on [Ca 2 ] i During Hyperkalemic/Hypothermic Cardioplegia in Single Ventricular Myocytes APK did not affect the [Ca 2 ] i in Krebs solution at 22 C (data not shown) but attenuated the elevation in [Ca 2 ] i during hyperkalemic/hypothermic cardioplegia (p 0.01; Fig 2B and C) in agreement with the previous observation in isolated ventricular myocytes [4, 8]. The effect of APK was completely abolished by 1 mol/l of glibenclamide, an ATP-sensitive K channel antagonist (Fig 2C). In order to correlate the effects of APK on Na -Ca 2 exchange activity, the effect of APK on [Ca 2 ] i in the cardioplegic solution with 2 mmol/l CsCl, 1 mmol/l BaCl 2,20 mol/l ouabain, and 2 mol/l verapamil, which were present in the external solution for the patch-clamp study, was determined; the effect of APK was exactly the same as in cardioplegic solution (Fig 2C). The effect of APK on [Ca 2 ] i mimicked that of 5 mmol/l Ni 2, which itself did not exhibit any effect in Krebs solution at 22 C (data not shown), but significantly attenuated the increase in [Ca 2 ] i induced by hyperkalemic/ hypothermic cardioplegia (p 0.001; Fig 2C). Similar observations have been made previously in rat [8] and pig [4] ventricular myocytes. The electrically induced [Ca 2 ] i transient, which represents influx of Ca 2 upon membrane depolarization and release of Ca 2 from SR [15], and caffeine-induced [Ca 2 ] i transient, which represents the Ca 2 content in

5 Ann Thorac Surg LI ET AL 2002;73: APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 1257 Table 1. Effects of Aprikalim on Contractile Functions of Single Isolated Rat Ventricular Myocytes During Rewarming/Reperfusion at 37 C After Hyperkalemic/Hypothermic Cardioplegia Krebs 37 C HK 4 C HK 4 C APK Number Cell shortening (%) Time 50% RE (ms) a p 0.01 Total duration (ms) Values are presented as mean SEM. a Versus Krebs 37 C. b Versus HK 4 C. HK4 C myocytes incubated at 4 C in 20 mmol/l [K ] o ; APK 100 mol/l aprikalim; Time 50% RE time to 50% relaxation. Fig 3. The experimental protocols for the study of the contractile recovery in single ventricular myocyte. Recording was made during normal superfusion, ie, before cardioplegia, and during reperfusion, ie, after cardioplegia. For results, see Tables 1 and 2. (APK aprikalim; HK high potassium solution.) the SR [6] were also determined during the reperfusion period after cardioplegia. In agreement with the previous observations [3], hyperkalemic/hypothermic cardioplegia did not alter the [Ca 2 ] i transients to either electrical stimulation or caffeine (data not shown); neither did APK alter the [Ca 2 ] i transients to hyperkalemic/hypothermic cardioplegia (data not shown). Effects of APK on Contractile Function During Rewarming/Reperfusion After Cardioplegia in Single Ventricular Myocytes To correlate the findings on the Na -Ca 2 exchange activity and [Ca 2 ] i, the contractile function during rewarming/reperfusion at both 37 C and 22 C was determined; the experimental protocols are expressed in Figure 3. Hyperkalemia/hypothermia for 1 hour reduced the amplitude of recovery contraction and prolonged the duration of the contraction and 50% relaxation (Tables 1 and 2). However, both the normal contraction in normothermic Krebs and the contractile recovery after 1 hour of hyperkalemic/hypothermic cardioplegia were faster and the shortening was smaller at 37 C (Table 1) than that at 22 C (Table 2) in agreement with the previous observations [13]. The reduction in shortening was 45.2% of the control when rewarming/reperfusion at 37 C, which was greater than 23.3% at 22 C (Tables 1 and 2). APK, 100 mol/l, improved percent shortening, time 50% relaxation, and total duration of contraction by 64.8%, 18.3%, and 24.5%, respectively, when rewarming/reperfusion at 37 C, which were greater than the corresponding values of 15.0%, 10.3% and 9.2%, respectively at 22 C (Tables 1 and 2). Five (5) mmol/l Ni 2 in cardioplegic solution also restored the values near the control at 22 C (Table 2). A very important observation is that APK in cardioplegic solution with 2 mmol/l CsCl, 1 mmol/l BaCl 2,20 mol/l ouabain, and 2 mol/l verapamil, which were present in the bath solution for the patch-clamp study, produced the same effect as in cardioplegic solution only (Table 2). This is in agreement with the observations on [Ca 2 ] i. Neither APK nor Ni 2 alone had any effect on recovery contraction in normal Krebs solution (data not shown). This is in agreement with the previous observations that PCOs do not affect the normal contraction but improve contractile recovery after cardioplegia [4, 13]. Table 2. Effects of Aprikalim on Contractile Functions of Single Isolated Rat Ventricular Myocytes During Rewarming/Reperfusion at 22 C After Hyperkalemic/Hypothermic Cardioplegia Krebs 22 C HK 4 C HK 4 C APK HK 4 C Ni HK 4 C APK Patch Solution Number Cell shortening (%) b p 0.05 Time 50% RE (ms) b p 0.01 Total duration (ms) Values are presented as means SEM. a Versus Krebs 22 C. b Versus HK 4 C. HK 4 C myocytes incubated at 4 C in 20 mmol/l [K ] o ; Ni 5 mmol/l NiCl 2 ; APK 100 mol/l aprikalim; Patch solution bath solution for whole cell patch-clamp experiments including CsCl 2 mmol/l, BaCl 2 1 mmol/l, Ouabain 20 mol/l, and verapamil 2 mol/l; Time 50% RE time to 50% relaxation.

6 1258 LI ET AL Ann Thorac Surg APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 2002;73: Comment The most important and novel finding of the present study is that APK, an ATP-sensitive PCO, attenuated the outward Na -Ca 2 exchange current enhanced by hyperkalemia, an effect the same as that of Ni 2,aNa - Ca 2 exchange blocker. It was also observed that both Ni 2 and APK attenuated the elevation of [Ca 2 ] i during, and improved contractile recovery after hyperkalemic/ hypothermic cardioplegia in rat ventricular myocytes. The results are in agreement with previous studies, which showed that APK is protective in recovery contractility at 37 C [4] and decreases [Ca 2 ] i elevation by hyperkalemia at room temperature [8] and at 4 C [4]. More importantly, the results indicate that APK may provide beneficial effects to the heart subjected to cardioplegia by attenuating the reversed Na -Ca 2 exchange activity, which is activated by hyperkalemia. In the present study we found similar effects of hyperkalemia or APK on [Ca 2 ] i and on contractile recovery in the presence and absence of 2 mmol/l CsCl and 1 mmol/l BaCl 2, which block most of the outward K currents except the ATP-sensitive K channel. The observation suggests that the effects of hyperkalemia and APK are most likely not related to efflux of K through these K channels. We also found that the effects of APK and hyperkalemia are also similar with presence or absence of ouabain, which blocks the Na -K ATPase activity, suggesting that the Na -K pump, which affects the intracellular Na concentration, is not altered by hyperkalemia. In addition both previous [3] and present studies have also shown that neither the electrically induced nor the caffeine-induced [Ca 2 ] i transient is altered after hyperkalemic cardioplegia, indicating the alterations in [Ca 2 ] i and contractile recovery are not due to alterations in either Ca 2 influx or mobilization of Ca 2 from SR. It is therefore likely that hyperkalemia activates the Na -Ca 2 exchange, leading to an increased outward current. This results in increased [Ca 2 ] i and impaired contractile recovery. APK, which opens the ATP-sensitive K channel and inhibits the Na -Ca 2 exchange, thus reversing the undesirable effects of hyperkalemia on [Ca 2 ] i and contractile recovery. Further studies are needed to delineate how hyperkalemia activates the Na -Ca 2 exchange activity and how opening of the ATP-sensitive K channel attenuates the hyperkalemia-enhanced Na -Ca 2 exchange activity. It should be noted that the outward Na -Ca 2 exchange current was reduced by 11% and the [Ca 2 ] i by 19% in the presence of APK during 4 C cardioplegia. The observation may suggest that the effect of APK on Na - Ca 2 exchange current does not solely account for the effect on [Ca 2 ] i. It should be noted, however, that the experimental conditions for determination of outward Na -Ca 2 exchange current and [Ca 2 ] i were different. For the determination of Na -Ca 2 exchange current, the patch-clamp technique was used, and [Na ] i was 20 mmol/l, which enhanced the outward Na -Ca 2 exchange current. The increase in basal current may make the change in percentage smaller than that in condition with normal Na -Ca 2 exchange current. On the other hand, the possibility that APK may have effects on Ca 2 -homeostasis other than that on Na -Ca 2 exchange should not be excluded. Further studies are needed to determine the other possible effects of APK. This study demonstrated that the contractile recovery was improved by APK during rewarming/reperfusion at both 37 C and 22 C. The contractile recovery in terms of amplitude of contraction, duration of 50% relaxation and contraction in the rat ventricular myocyte during rewarming/reperfusion at 37 C are comparable to the corresponding values obtained in the isolated pig ventricular myocyte in normothermia [4] and are greater than those in the rat ventricular myocyte at 22 C. The observations indicate that PCOs produce more beneficial effects at body temperature than at room temperature in terms of contractile function in agreement with the previous observations [16]. This is an important property for the use of PCOs in open heart surgery. It is well established that high [Ca 2 ] i could be a load on the energy-dependent Ca 2 homeostasis in ischemia [17] and thus predispose cell injury during reperfusion [18]. As demonstrated in this study, APK inhibits the Na -Ca 2 exchange activity, thus suppressing the excessive elevation of [Ca 2 ] i during cardioplegia and improving contractile recovery after cardioplegic arrest. In addition, it has been suggested that PCO-induced membrane hyperpolarization might also reduce Ca 2 - sensitivity of the contractile apparatus [19]. It is therefore reasonable to speculate that APK might reduce the responsiveness of myofibrils to a sudden increase in [Ca 2 ] i during cardioplegia, thus preserving energy and improving the contractile function during reperfusion. This action may explain why in the same temperature the effect of APK was greater on [Ca 2 ] i than on contractile responses observed in this study. Further study is needed to verify this. PCOs have been suggested to be used in cardioplegia [7, 20, 21]. It has also been shown that PCOs added into the hyperkalemic cardioplegia may attenuate the sideeffects of the hyperkalemia such as inhibition of the endothelium-derived hyperpolarizing factor-mediated relaxation in the coronary arteries [21]. A previous [4] and the present study demonstrate that in the presence of hyperkalemia, PCOs may improve the cardiac contractile during the reperfusion period. A recent study with overexpression of Na -Ca 2 exchanger in the transgenic mouse showed that the Na -Ca 2 exchanger works in the Ca 2 influx mode during the ischemia/reperfusion, which leads to increased injury and lower recovery [22]. The observations strongly support that Ca 2 influx through the Na -Ca 2 exchanger produces detrimental effects on cardiac tissue incurred upon ischemia/ reperfusion. So inhibition of the Na -Ca 2 exchanger by ATP-sensitive K opener as APK should confer protection against myocardial injury induced by ischemia/ reperfusion in ischemic heart diseases by attenuating Ca 2 influx induced by high K in the hyperkalemic cardioplegia. However, the PCOs may produce undesirable effects, which make them unsuitable for clinic use.

7 Ann Thorac Surg LI ET AL 2002;73: APRIKALIM INHIBITS K -ENHANCED NA -CA 2 EXCHANGE 1259 PCOs have been shown to produce proarrhythmic [23] or antiarrhythmic [24] actions or both [25]. On the other hand there is also a report that PCOs have no undesirable effects [26]. Further studies on the undesirable effects of PCOs including APK are needed. This study was supported by the Cardiovascular Research Fund donated by LCST (Holdings), Ltd, and the Institute of Cardiovascular Science and Medicine, Faculty of Medicine, University of Hong Kong. We thank Dr C. M. Wong for advice on statistical tests, Dr G. R. Li for helpful discussion, and Mr C. P. Mok for assistance. References 1. Weisel RD. Myocardial stunning after coronary bypass surgery. J Cardiovasc Surg 1993;8: Handy JR, Spinale FG, Mukherjee R, et al. Hypothermic potassium cardioplegia impairs myocyte recovery of contractility and inotropy. J Thorac Cardiovasc Surg 1994;107: Ahn DS, Lee YH, Kang DH, et al. Effect of hypothermic cardioplegia on cardiac protection. 1. Effect of hypothermic cardioplegia on the cytosolic Ca 2 concentration in rat ventricular myocytes. Yonsei Med J 1994;35: Dorman BH, Hebbar L, Clair MJ, et al. Potassium channel opener-augmented cardioplegia-protection of myocyte contractility with chronic left ventricular dysfunction. Circulation 1997;96(Suppl 2): Kimura J, Miyamae S, Noma A. Identification of sodiumcalcium exchange current in single ventricular cells of guinea-pig. J Physiol 1987;384: Weber A. The mechanism of the action of caffeine on sarcoplasmic reticulum. J Gen Physiol 1968;52: He GW. Potassium-channel opener in cardioplegia may restore coronary endothelial function. Ann Thorac Surg 1998;66: Lopez JR, Jahangir R, Jahangir A, et al. Potassium channel openers prevent potassium- induced calcium loading of cardiac cells: possible implications in cardioplegia. J Thorac Cardiovasc Surg 1996;112: Kimura J, Noma A, Irisawa H. Na -Ca 2 exchange current in mammalian heart cells. Nature 1986;319: Wu S, Li HY, Wong TM. Cardioprotection of preconditioning by metabolic inhibition in the rat ventricular myocyte involvement of -opioid receptor. Circ Res 1999;84: Sheng JZ, Wong TM. Chronic U50,488H abolishes inositol 1,4,5-trisphosphate and intracellular Ca 2 elevation evoked by -opioid in rat myocytes. Eur J Pharmacol 1996;307: Wang YX, Zheng YM. Ionic mechanism responsible for prolongation of cardiac action-potential duration by berberine. J Cardiovasc Pharmacol 1997;30: Ripoll C, Lederer WJ, Nichols CG. Modulation of ATPsensitive K channel activity and contractile behavior in mammalian ventricle by the potassium channel openers cromakalim and RP J Pharmacol Exp Ther 1990;255: Saito W, Noguchi K, Okazaki K, et al. Temperature-sensitive effects of potassium channel openers on isolated guinea pig myocardium, and aorta. J Cardiovasc Pharmacol 1998;31: Bers DM, Bridge JH, Spitzer KW. Intracellular Ca 2 transient during rapid cooling contracture in guinea-pig ventricular myocytes. J Physiol 1989;417: McCullough JR, Normandin DE, Conder ML, et al. Specific block of the anti-ischemic actions of cromakalim by sodium 5-hydroxydecanoate. Circ Res 1991;69: Tsukube T, McCully JD, Metz KR, et al. Amelioration of ischemic calcium overload correlates with high-energy phosphates in senescent myocardium. Am J Physiol 1997; 273:H McCully JD, Tsukube T, Ataka K, et al. Myocardial cytosolic calcium accumulation during ischemia/reperfusion: the effects of aging and cardioplegia. J Card Surg 1994;9(Suppl 3): Quast U. Do the K channel openers relax smooth muscle by opening K channels? Trends Pharmacol Sci 1993;14: Jayawant AM, Lawton JS, Hsia PW, et al. Hyperpolarized cardioplegic arrest with nicorandil: advantages over other potassium channel openers. Circulation 1997;4:96(Suppl 2): Cohen NM, Wise RM, Wechsler AS, et al. Elective cardiac arrest with a hyperpolarizing adenosine triphosphatesensitive potassium channel opener. A novel form of myocardial protection? J Thorac Cardiovasc Surg 1993;106: Cross HR, Lu L, Steenbergen C, Philipson KD, Murphy E. Overexpression of the cardiac Na -Ca 2 exchanger increases susceptibility to ischemia/reperfusion injury in male but not female transgenic mice. Circ Res 1998;83: Tosaki A, Szerdahelyi P, Engelman RM, et al. Potassium channel openers and blockers: do they possess proarrhythmic or antiarrhythmic activity in ischemic and reperfused rat hearts? J Pharmacol Exp Ther 1993;267: Carlsson L, Abrahamsson C, Drews L, et al. Antiarrhythmic effects of potassium channel openers in rhythm abnormalities related to delayed repolarization. Circulation 1992;85: Lynch JJ Jr, Sanguinetti MC, Kimura S, et al. Therapeutic potential of modulating potassium currents in the diseased myocardium. FASEB J 1992;6: Asirvatham S, Sebastian C, Thadani U. Choosing the most appropriate treatment for stable angina. Drug Safety 1998; 19: INVITED COMMENTARY Potassium channel openers (KCOs) have been used to treat ischemic heart disease and hypertension by reducing the oxygen demand and promoting vasodilation and bradycardia KCOs hyperpolarize the cell and mitochondrial membranes, operating through the potassium (K ) ATP-sensitive channels. Because some of these actions mimic those following hyperkalemic-induced cardioplegic arrest during open-heart surgery, it has been suggested that KCOs may have a role as cardioplegic agents. KCO induced cardiac arrest has been followed by improved myocardial contractile function and a reduction of the extracellular matrix damage compared with conventional K cardioplegia. Potassium channel openers (KCOs) may also play a role with cerebral protection after hypothermic cardiac arrest. The heart and brain are very dependent on oxidative phosphorylation and are particularly sensitive to ischemic injury. Both are rich in mitochondrial potassium ATP-sensitive channels. KCOs decrease membrane potential and the Na -Ca 2 calcium exchange. They 2002 by The Society of Thoracic Surgeons /02/$22.00 Published by Elsevier Science Inc PII S (02)