Coordinate Interaction Between ATP-Sensitive K+ Channel and Na+,K+-ATPase Modulates Ischemic Preconditioning
Background—We reported that digoxin abolishes the infarct size (IS)-limiting effect of ischemic preconditioning (IPC). Because ATP-sensitive K+ (KATP) channels are involved in IPC, we studied whether Na+,K+-ATPase and KATP channels functionally interact, thereby modulating IPC.
Methods and Results—Rabbits received 30 minutes of coronary artery occlusion followed by 3 hours of reperfusion. IPC was elicited by 5 minutes of occlusion followed by 10 minutes of reperfusion. The IS, expressed as a percentage of the area at risk, was 40.2±2.8% in control and 39.8±5.0% in digoxin pretreatment rabbits. Both IPC and pretreatment with cromakalim, a KATP channel opener, reduced IS to 11.8±1.8% and 13.4±2.6% (P<0.05 versus control). Digoxin abolished the reduction in IS induced by IPC (33.5±3.3%), whereas it did not change that induced by cromakalim (18.8±3.0%). In patch-clamp experiments, digoxin was found to inhibit the opening of KATP channels in single ventricular myocytes in which ATP depletion had been induced by metabolic stress. In contrast, digoxin had little effect on the channel opening induced by cromakalim. Moreover, the inhibitory action of digoxin on channel activities was dependent on subsarcolemmal ATP concentration.
Conclusions—The IS-limiting effect of IPC is modulated by an interaction between KATP channels and Na+,K+-ATPase through subsarcolemmal ATP.
Exposure of the myocardium to ≥1 brief periods of ischemia, known as ischemic preconditioning (IPC),1 produces a marked resistance to subsequent prolonged episodes of ischemic stress in several mammalian species.2 3 4 It is now recognized that ATP-sensitive K+ (KATP) channels are ubiquitous target effectors of the cardioprotective process by IPC, independent of species difference.2 3 4
We have demonstrated in rabbit hearts that the prevention of ischemia-induced reduction of Na+,K+-ATPase activity during the early phase of sustained ischemia contributes to the infarct size (IS)-limiting effect of IPC because pretreatment with digoxin, an inhibitor of Na+,K+-ATPase, abolished the cardioprotective effect of IPC.5 In guinea pig cardiac myocytes, 2 other cardiac glycosides, strophanthidin6 and ouabain,7 suppress the opening of KATP channels by minimizing the consumption of subsarcolemmal ATP by Na+,K+-ATPase. This notion led us to the hypothesis that the IS-limiting effect of IPC and the inhibitory action of digoxin reflect a functional interaction between Na+,K+-ATPase and KATP channels through the ATP concentration ([ATP]) in the subsarcolemmal compartment.
The purpose of the present study was to test this potential interaction. We compared the effect of digoxin on the IS limitation induced by IPC and that by a KATP channel opener, cromakalim, in in vivo rabbit hearts, because it has been shown that KATP channel openers evoke the opening of KATP channels in the presence of relatively high concentrations of subsarcolemmal ATP.8 9 In addition, using patch-clamp techniques on single ventricular myocytes, we examined the effects of digoxin on KATP channel activation.
Female Japanese white rabbits (2.5 to 3.2 kg) were anesthetized with 30 mg/kg sodium pentobarbital IV, intubated, then mechanically ventilated with 100% oxygen by use of a volume-cycled respirator. A fluid-filled catheter was placed in the right femoral artery and was connected to a transducer for measurement of arterial pressure. Ventilation was maintained at 15 to 20 breaths per minute with ≈40 mL of tidal volume. A left thoracotomy was performed in the fourth intercostal space, and the pericardium was opened. A 4-0 silk thread was then passed around the circumflex branch of the left coronary artery, with its ends being threaded through a small polyethylene tube. Precordial electrocardiography was monitored with bipolar chest leads. Rabbits were allowed ≥20 minutes to reach a steady state after surgical preparation. Coronary occlusion was produced by pulling the snare and clamping it with a mosquito hemostat. Reperfusion was produced by releasing the clamp.
Determination of IS
Animals were subjected to a 30-minute period of coronary artery occlusion followed by 3 hours of reperfusion. To achieve IPC, the coronary artery was occluded for 5 minutes, followed by 10 minutes of reperfusion, before the sustained occlusion and reperfusion. Rabbits were assigned randomly to 1 of 6 groups, including the control group, consisting of control animals given vehicle only; the IPC group, animals given vehicle before IPC; the digoxin group, animals given digoxin 0.3 mg/kg as a bolus injection 30 minutes before sustained ischemia; the digoxin+IPC group, animals treated with digoxin before IPC; the cromakalim group, animals given cromakalim 4 μg · kg−1 · min−1 over a period of 10 minutes as an intravenous infusion 10 minutes before sustained ischemia; and the digoxin+cromakalim group, animals pretreated with digoxin 30 minutes before sustained ischemia before cromakalim treatment. The digoxin and digoxin+IPC groups consisted of animals from our previous study.5
In each protocol, the heart was rapidly excised at the end of the 3 hours of reperfusion and mounted by the aortic root on a Langendorff apparatus. The snare was retightened, and 0.5% phthalocyanine blue pigment was infused into the perfusate to demarcate the risk zone as the tissue without blue dye. The heart was then removed, and the left ventricle was isolated and cut into transverse slices 2 mm thick. The area at risk (nonblue area) was separated from each slice, weighed, and incubated at 37°C for 15 minutes in 1% triphenyl tetrazolium chloride (TTC) in phosphate buffer, pH 7.4. TTC stained the noninfarcted myocardium deep red. The sections were fixed in 10% formalin solution for >3 hours. Their digital images were then captured with a CCD camera and entered into an Apple Power Macintosh computer. The area at risk and the area of infarction (TTC-negative) were determined with image analysis software, corrected for the weight of each tissue slice, and summed for each heart. The IS was expressed as a percentage of the area at risk.
Single-Cell Preparation and Electrophysiology
Left ventricular myocytes were isolated from Japanese White rabbits by the collagenase perfusion method. Single ventricular myocytes were dispersed in a recording chamber (0.5 mL in volume) on the stage of an inverted microscope (Nikon, TMD-2) and were superfused with Tyrode’s solution containing (in mmol/L) NaCl 143, NaH2PO4 0.3, KCl 5.4, MgCl2 0.5, CaCl2 1.8, and HEPES/NaOH 5 (pH 7.4 adjusted by NaOH). For experiments using metabolic inhibition, single cells were preincubated with a glucose-free Kraft Brühe solution containing 2-deoxyglucose (5.5 mmol/L; Nacalai Tesque), an inhibitor of glycolysis, at 37°C for >30 minutes.
The pipette solution for KATP current measurement contained (in mmol/L) aspartic acid 130, KCl 10, NaCl 10, MgCl2 1.0, EGTA 10, and K2ATP 0.1 (pH 7.4 adjusted with HEPES/KOH 5). KATP current was activated either by metabolic stress (glucose- and Ca2+-free Tyrode’s solution containing 1 mmol/L NaCN) or by cromakalim (100 μmol/L), recorded by ramp pulses applied every 10 seconds at −40-mV holding potential to obtain quasi-instantaneous current-voltage (I-V) relations. All the current data measured by an amplifier (List EPC-7) were directly displayed on a chart recorder (Graphtec, Linerecorder, WR 3320) with on-line acquisition to an NEC personal computer at a Bessel-type cutoff frequency of 1 kHz.
Single KATP channel activity was recorded in either cell-attached, inside-out, or open cell–attached mode. Glucose-free Tyrode’s solution was used as a pipette solution. To virtually eliminate the membrane potential, a K+-rich solution was used as a bathing solution. Its composition was (in mmol/L) KCl 150, EGTA 0.5, and HEPES 5 (pH 7.4 adjusted by KOH). In the cell-attached mode, the KATP channels were activated by metabolic inhibition (1 mmol/L NaCN, glucose-free) after incubation with 2-deoxyglucose (>20 minutes at 37°C) or by cromakalim (100 μmol/L). The inhibition of the glycolysis pathway was applied to the myocytes only for metabolic stress–induced opening of the channel.
Brief (2 to 5 minutes) exposure of the myocytes to streptolysin O (0.02 to 0.08 U/mL, Nacalai Tesque) produced tiny holes on the cell membrane and allowed us to control the [ATP] just beneath the membrane by changing the [ATP] of the bathing solution.7 The inside-out mode was achieved by excising the patch membrane into ATP-free bath solution. Patch pipettes were prepared by pulling borosilicate glass capillaries (Hilgenberg) at 2 to 3 MΩ for whole-cell and at 5 to 7.5 MΩ for single-channel recordings when filled with pipette solution.
The mean patch currents (I) for single-channel events were measured as the average difference between baseline (all channels closed) and open channel currents. The unitary amplitude (i) was estimated by the analysis of the amplitude distribution by use of pCLAMP software version 6.1 (Axon Instruments Inc) at a Bessel-type cutoff frequency of 100 Hz at a sample frequency of 1 kHz. The mean number of open channels (NP) was given by the product of the number of available channels in the patch (N) and the probability of their being open (P). NP was estimated by dividing the mean patch currents (I=NPi) by unitary amplitude (i) as I/i.
Digoxin (Sigma Chemical Co) was dissolved in distilled water at 1 mmol/L (stock solution) and glibenclamide (Sigma) in DMSO at 1 mmol/L (stock solution). Streptolysin O was dissolved in the bathing solution at a final concentration of 0.08 U/mL immediately before every experiment and was used within 3 to 4 hours after preparation.7 All experiments were carried out at ≈36°C.
The differences in time courses of hemodynamics change between groups, and time courses were analyzed by 2-way ANOVA with repeated measurements. In the IS study, the differences between groups were compared by 1-way ANOVA with Scheffé’s post hoc test. A level of P<0.05 was accepted as statistically significant. In the electrophysiological study, statistical analyses were performed with an unpaired Student’s t test.
The hemodynamic data are summarized in Table 1⇓. No significant differences were observed in mean arterial pressure or heart rate among the 6 groups at any of the experimentally determined time points.
Data from the study of IS are summarized in Table 2⇓. Body weights, left ventricular weights, and area-at-risk weights did not differ significantly among any of the experimental groups. However, IS weights were significantly smaller in the IPC, cromakalim, and digoxin+cromakalim groups than in the control group.
Figure 1⇓ shows the IS as a percentage of the area at risk in each experiment. IPC and cromakalim reduced the size of the infarct compared with control (11.8±1.8% and 13.4±2.6% versus 40.5±2.8%, respectively; P<0.05). The administration of 0.3 mg/kg digoxin had no effect on the IS compared with controls (39.8±5.0%) but significantly reduced the beneficial IS-limiting effects of IPC (33.5±3.3%). In contrast, it did not affect such effects induced by cromakalim (18.8±3.0%).
Effects of Digoxin on KATP Channel Activities
Exposure of a myocyte preincubated with 2-deoxyglucose to glucose- and Ca2+-free Tyrode’s solution containing 1 mmol/L NaCN activated outward conductance at a holding potential of −40 mV, usually within 10 minutes (Figure 2A⇓). Quasi-instantaneous I-V relationships (Figure 2B⇓-1) showed that the conductance induced by metabolic stress reversed at −83 mV and was inhibited by extracellular glibenclamide 1 μmol/L) (Figure 2A⇓, d and e). Difference currents for metabolic stress–induced and glibenclamide-sensitive components (b-a in Figure 2C⇓-1 and d-e in Figure 2C⇓-3) showed reversal potentials of −85 mV, close to the estimated equilibrium potential for K+ (EK) under these experimental conditions. These findings suggest that the current induced by metabolic inhibition is carried by the activation of KATP channels.
External application of digoxin reversibly suppressed the conductance induced by metabolic stress (Figure 2A⇑, b and c) without altering the reversal potential (Figure 2B⇑-2). Current components sensitive to digoxin (b-c in Figure 2C⇑-2) reversed at a potential of −82 mV, which is also close to EK. Similar findings were observed in 3 other experiments. Percentage inhibitions of outward current at 0 mV were calculated as 100×(b-c)/b as in Figures 2B⇑ and 2C⇑-2 and were 89.6±3.3% (n=4). Thus, digoxin appeared to inhibit the KATP channel current.
In the experiment shown in Figure 3A⇓, cromakalim 100 μmol/L evoked an outward current at a holding potential of −40 mV (Figure 3A⇓). Because of the reversal potential (−83 mV; Figure 3A⇓, a and b; 3B; and 3C-1) and block by glibenclamide (Figure 3A⇓, e and f), the cromakalim-induced conductance was considered to be due to the activation of KATP channels. Extracellular digoxin 1 and 10 μmol/L did not significantly suppress the KATP current induced by cromakalim (Figure 3A⇓, b and d; 3B; and 3C-2). Failure of digoxin to inhibit the current was also observed in a total of 4 experiments, and percent inhibitions at 0 mV were 10.7±4.0% (P=NS versus control).
The effect of digoxin on single KATP channel activity was also examined in the cell-attached mode (Figure 4⇓). KATP channels were opened by either metabolic stress (A) or cromakalim 100 μmol/L (B). KATP channel currents were recorded as an upward deflection at 0 mV under these experimental conditions. Exposure of myocytes preincubated with 2-deoxyglucose 5.5 mmol/L to NaCN 1 mmol/L for 15 to 20 minutes produced the opening of single channels (Figure 4A⇓), as did application of cromakalim into both pipette and external solutions (Figure 4B⇓). Glibenclamide 0.5 μmol/L inhibited both channel activities. Figure 4C⇓ summarizes the data obtained in both protocols at various concentrations of digoxin and glibenclamide 0.5 μmol/L. Digoxin applied outside the pipette blocked the channel activities by metabolic inhibition in a reversible and concentration-dependent manner. In contrast, it had no effect on the channel activities evoked by cromakalim.
Inside-Out and Open Cell–Attached Experiments
Figure 5⇓ shows the data obtained in inside-out (A) and open cell–attached modes (B and C). Excision of patch membrane into an ATP-free solution produced a robust opening of those single-channel currents that were sensitive to glibenclamide (1 μmol/L; Figure 5A⇓). In the absence of internal ATP, the channels ran down gradually. The findings indicated that single-channel events were openings of KATP channels. Digoxin 1 μmol/L applied to the internal side of the patch membrane had no effect (only 5.5±2.5% inhibition of control).
In the open cell–attached mode (Figure 5B⇑), we were able to open the channel by reducing extracellular [ATP] through tiny membrane holes made by streptolysin O.7 In fact, raising extracellular ATP from 0 to 1 mmol/L reversibly closed the channel (Figure 5B⇑). As in the experiments with cromakalim, digoxin 1 μmol/L was without effect on the channel activity thus obtained. Figure 5C⇑ summarizes the percentage of residual patch current in digoxin 1 μmol/L and glibenclamide 1 μmol/L in the open cell–attached mode.
Effects of Digoxin on IS in In Vivo Rabbit Hearts
We have previously demonstrated that pretreatment with digoxin, an inhibitor of Na+,K+-ATPase, attenuated the KATP channel–mediated IS-limiting effect of IPC in the rabbit heart.5 Interestingly, however, we found in the present study that digoxin did fail to block the salutary IS-limiting effect of cromakalim. To explain this discrepancy, we focused on the interaction of Na+,K+-ATPase and KATP channels, because sarcolemmal molecules consuming subsarcolemmal ATP, such as Na+,K+-ATPase, have been shown to alter the channel function through the change in subsarcolemmal [ATP].6 7 10 11 The activation of KATP channels is dependent on subsarcolemmal [ATP] during ischemia,12 13 whereas KATP channel openers can open the channel in the presence of relatively high concentrations of ATP.8 9 It is therefore likely that an interaction between the Na+,K+-ATPase and the channel via subsarcolemmal ATP would affect the IS-limiting action of IPC as well as its blockade by digoxin. This idea is not contradictory to the findings in previous studies14 15 using nuclear magnetic resonance spectroscopy15 that myocardial [ATP] levels were relatively preserved in the preconditioned compared with nonpreconditioned hearts. This method cannot measure intracellular gradient of ATP because of the nonhomogeneous distribution of mitochondria,16 glycolysis-dependent ATP production,10 or membrane molecule–induced consumption of ATP.6 7 10 11 During sustained ischemia, subsarcolemmal [ATP] could be low enough to activate KATP channels in the preconditioned heart even though gross [ATP] remains relatively preserved, because the activities of sarcolemmal enzymes, especially Na+,K+-ATPase, which consume subsarcolemmal ATP, are not attenuated during the early phase of sustained ischemia.5
We used a single 5-minute IPC stimulus in the present study. It has been reported that bradykinin antagonists can block cardioprotection only when 1 cycle, but not 4 cycles, of IPC is used.17 In contrast, we confirmed that digoxin inhibited even the IS-limiting effect induced by 3 cycles of IPC (data not shown). It appears that the IS-limiting effect by IPC requires Na+,K+-ATPase activity to open KATP channels.
Effects of Digoxin on KATP Channel Activities in Isolated Rabbit Myocytes
In the present study, as in the findings in in vivo rabbit for myocardial ischemia, whole-cell experiments using patch-clamp techniques demonstrated that digoxin reversibly inhibits the KATP current induced by metabolic stress, but its inhibitory action was largely attenuated when the current was activated by cromakalim. In the cell-attached mode, digoxin also inhibited single KATP channel currents activated by metabolic stress in a reversible and concentration-dependent manner, but not those activated by cromakalim.
Therefore, the digoxin-induced inhibition of KATP channels appeared to depend on subsarcolemmal [ATP]. This was also supported by the experimental finding that digoxin was without effect in the inside-out or open cell–attached modes. In the former, excision of patch membrane into an ATP-free solution produced opening of KATP channels. The inhibition by digoxin was then negligible, indicating that the drug does not affect the channel directly. In the latter mode, through tiny membrane holes, subsarcolemmal [ATP] could be “concentration-clamped” at a desired level by [ATP] in the bathing extracellular solution.7 Under these conditions, digoxin was also unable to inhibit the channel. The findings again suggest that Na+,K+-ATPase and KATP channel functionally interact through subsarcolemmal ATP.
We have previously reported that angiotensin II mediates the closure of KATP channels through inhibition of adenylate cyclase, which transits from ATP to cAMP in guinea pig myocytes.7 Conversely, Liu et al18 showed that pretreatment with angiotensin II caused an IS-limiting effect in the rabbit heart. Angiotensin II potentially opens KATP channels by stimulating protein kinase C activation, although the relationship between protein kinase C and KATP channels still remains controversial.19 20 In angiotensin II–induced cardioprotection, this stimulatory pathway may act on KATP channels more prominently over the inhibitory pathway via adenylate cyclase.
Cardioprotection by KATP Channel Activation
In the present study, we assumed that both IPC- and cromakalim-mediated IS-limiting effects were derived from the activation of sarcolemmal KATP channels. These beneficial effects have been thought to be due to the shortening of myocardial action potential duration (APD) and the subsequent reduction in Ca2+ entry into myocytes.12 13 However, Yao and colleagues21 demonstrated in dogs that bimakalim, a KATP channel opener, conferred the IS-limiting effect at a dose that produced no shortening of APD. Likewise, Grover et al22 23 reported that the IS-limiting effects of cromakalim and IPC are preserved despite concomitant treatment with dofetilide, an antiarrhythmic agent with class III action, at a concentration sufficient to abolish the cromakalim- and IPC-induced shortening of APD.
Recently, mitochondrial KATP channels have drawn considerable attention as targets for the action of KATP channel openers.24 Garlid et al25 have shown that diazoxide, a potent mitochondrial but weak sarcolemmal KATP channel opener, improves postischemic functional recovery and decreases lactate dehydrogenase release during reperfusion in isolated perfused rat and rabbit hearts subjected to global ischemia. However, the theory that mitochondrial KATP channels alone contribute to the cardioprotection by IPC and KATP channel openers cannot explain the dissociation of the inhibitory effects of digoxin on IS between IPC and cromakalim that we showed in this study. The potency of diazoxide to achieve cardioprotection was reported to be comparable to that of cromakalim in rabbit heart.25 However, we found that diazoxide given in the same protocol as cromakalim did not induce any IS-limiting effect (IS, 33.4±5.9%, n=7) even at a 10-fold higher dose (40 μg · kg−1 · min−1) than that of cromakalim used in the present study (4 μg · kg−1 · min−1). Thus, sarcolemmal KATP channels also play an important role in the IS-limiting effect of IPC and cromakalim in the in vivo rabbit model of regional myocardial ischemia. Because there was little relationship between APD shortening and cardioprotection in terms of KATP channel openers,21 22 23 further studies are necessary to examine the precise mechanism(s) underlying the KATP channel–mediated cardioprotection.
This work was supported by Grants-in-Aid on Priority Areas of Channel-Transporter Correlation and research grants (09670713) from the Japanese Ministry of Education, Science, and Culture.
- Received March 30, 1998.
- Revision received August 7, 1998.
- Accepted August 20, 1998.
- Copyright © 1998 by American Heart Association
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;74:1124–1136.
Gross GJ, Auchampach JA. Blockade of ATP-sensitive potassium channels prevents myocardial preconditioning in dogs. Circ Res. 1992;70:223–233.
Schultz JE, Yao Z, Cavero I, Gross GJ. Glibenclamide-induced blockade of ischemic preconditioning is time-dependent in intact rat heart. Am J Physiol. 1997;272:H2607–H2615.
Kouchi I, Murakami T, Nawada R, Akao M, Sasayama S. KATP channels are common mediators of the cardioprotection induced by ischemic and calcium preconditioning in rabbits. Am J Physiol. 1998;274:H1106–H1112.
Nawada R, Murakami T, Iwase T, Nagai K, Morita Y, Kouchi I, Akao M, Sasayama S. Inhibition of sarcolemmal Na+,K+-ATPase activity reduces the infarct size–limiting effect of preconditioning in rabbit hearts. Circulation. 1997;96:599–604.
Tsuchiya K, Horie M, Albrecht CA, Obayashi K, Fujiwara H, Sasayama S. Functional compartmentalization of ATP is involved in the angiotensin II–mediated closure of cardiac ATP-sensitive K+ channels. Circulation. 1997;96:3129–3135.
Ripoll C, Lederer WJ, Nichols CG. Modulation of ATP-sensitive K+ channel activity and contractile behavior in mammalian ventricle by the potassium channel openers cromakalim and RP49356. J Pharmacol Exp Ther. 1990;255:429–435.
Thuringer D, Escande D. Apparent competition between ATP and the potassium channel opener RP 49356 on ATP-sensitive K+ channels of cardiac myocytes. Mol Pharmacol. 1989;36:897–902.
Weiss JN, Lamp ST. Glycolysis preferentially inhibits ATP-sensitive K+ channels in isolated guinea-pig cardiac myocytes. Science. 1987;238:67–69.
Nichols CG, Ripoll C, Lederer WJ. ATP-sensitive potassium channel modulation of the guinea pig ventricular action potential and contraction. Circ Res. 1991;68:280–287.
Murry CE, Richard VJ, Reimer KA, Jennings RB. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990;66:913–931.
de Albuquerque CP, Gerstenblith G, Weiss RG. Importance of metabolic inhibition and cellular pH in mediating preconditioning contractile and metabolic effects in rat hearts. Circ Res. 1994;74:139–150.
Jones DP. Intracellular diffusion gradients of O2 and ATP. Am J Physiol. 1986;250:C663–C675.
Goto M, Liu Y, Yang XM, Ardell JL, Cohen MV, Doweney JM. Role of bradykinin in protection of ischemic preconditioning in rabbit hearts. Circ Res. 1995;77:611–621.
Hu K, Duan D, Li GR, Nattel S. Protein kinase C activates ATP-sensitive K+ current in human and rabbit ventricular myocytes. Circ Res. 1996;78:492–498.
Light PE, Sabir AA, Allen BG, Walsh MP, French RJ. Protein kinase C-induced changes in the stoichiometry of ATP binding activate cardiac ATP-sensitive K+ channels: a possible mechanistic link to ischemic preconditioning. Circ Res. 1996;79:399–406.
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. 1994;89:1769–1775.
Grover GJ, D’Alonzo AJ, Dzwonczyk S, Parham CS, Darbenzio RB. Preconditioning is not abolished by the delayed rectifier K+ blocker dofetilide. Am J Physiol. 1996;271:H1207–H1214.
Garlid KD, Paucek P, Yarov-Yarovoy V, Sun X, Schindler PA. The mitochondrial KATP channel as a receptor for potassium channel openers. J Biol Chem. 1996;271:8796–8799.
Garlid KD, Paucek P, Yarov-Yarovoy V, Murray HN, Darbenzio RB, D’Alonzo AJ, Lodge NJ, Smith MA, Grover GJ. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels. Circ Res. 1997;81:1072–1082.KATP channels mediate the cardioprotection of ischemic preconditioning (IPC). In the present study, digoxin, an inhibitor of Na+,K+-ATPase, was found to reduce the infarct size–limiting effect caused by IPC, whereas it did not change that caused by cromakalim, a KATP channel opener, in the in vivo rabbit heart. In single rabbit ventricular myocytes, digoxin suppressed KATP channels opened by ATP depletion with metabolic stress but not those opened by cromakalim. This inhibitory action by digoxin on the channel activities was dependent on subsarcolemmal ATP concentration. Thus, the infarct size–limiting effect of IPC is modulated by an interaction between KATP channels and Na+,K+-ATPase through subsarcolemmal ATP.