Unanticipated Lessening of the Rise in Extracellular Potassium During Ischemia by Pinacidil
Background The efflux of potassium (K) through the ATP- sensitive K channel is considered an important cause of the rise in extracellular K ([K+]e ) during no-flow ischemia. We postulated that agents that enhance K conductance in this channel would enhance the rise in [K+]e.
Methods and Results We studied the effects of 10 and 25 μmol/L pinacidil, an ATP-sensitive K channel opener that provides metabolic protection to the ischemic myocardium, on the rise in [K+]e recorded by K-sensitive electrodes, the change in action potential duration (APD) recorded by microelectrodes, and the changes in activation during ischemia in in situ pig hearts and Tyrode-perfused rabbit interventricular septa. Pinacidil 25 μmol/L unexpectedly lessened the rise in [K+]e and the activation delay in both preparations. Pinacidil 10 μmol/L had no effect in the rabbit and only a slight effect in the pig. Both concentrations significantly exaggerated the APD shortening induced by ischemia. By varying stimulation frequency, we demonstrated that the rise in [K+]e during ischemia, both before and after pinacidil, correlated with the time that the action potential was at its plateau voltage.
Conclusions Our results indicate that the rise in [K+]e during ischemia is due to multiple factors, including K conductance across membrane channels, K driving force as reflected by the time that the action potential is at its plateau voltage, and the metabolic effects of ischemia. The unanticipated lessening of the rise in [K+]e by pinacidil reflects the balance of its effects on these several parameters.
The rise in extracellular potassium ([K+]e) during acute ischemia1 2 3 is a major cause of the electrophysiological changes and ventricular arrhythmias that occur in this setting.4 5 The precise cause of this rise has not been well defined, although both anionic- and nonanionic-linked factors are believed to be important.6 7 8 Studies using the sulfonylureas glibenclamide and tolbutamide suggest that K+ efflux via the ATP-sensitive K+ channel8 9 10 11 may be one such factor. Thus, agents that enhance K+ conductance via this channel might be expected to enhance the rise in [K+]e during ischemia. To test this possibility, we studied the effects of 10 and 25 μmol/L pinacidil, a drug known to open ATP-sensitive K+ channels,12 on the rise in [K+]e and the associated changes in action potential duration (APD) and activation induced by acute no-flow ischemia in the in situ pig heart and the perfused rabbit interventricular septum. Pinacidil unexpectedly caused a decrease in the rise in [K+]e. However, analysis of the factors contributing to this unanticipated result provides insight into the causes of the ischemia-induced rise of [K+]e.
The care of animals used in this study conformed to the Position of the American Heart Association on Research Animal Use and was done in accordance with accepted guidelines for the care and treatment of experimental animals at the University of North Carolina.
In Vivo Pig Experiments
Domestic swine of either sex weighing 30 to 50 kg were anesthetized with thiopental sodium (25 mg/kg) followed by α-chloralose as needed. Mechanical ventilation and supplemental oxygen were supplied via an endotracheal tube and a Harvard respirator. Arterial blood gas samples were monitored and appropriate ventilator adjustments made to maintain arterial Po2 >80 mm Hg and pH at 7.35 to 7.45. Heating blankets were used to maintain body temperature at 36° to 37°C.
The heart was exposed via a median sternotomy and suspended in a pericardial cradle. A site midway in the left anterior descending coronary artery (LAD) was dissected and briefly occluded to define the epicardial margin between ischemic and nonischemic tissue. Multiple ion-selective electrode groups were placed in the center of the ischemic zone and in the normal (nonischemic) zone, and in some experiments, pairs of piezoultrasonic crystals were placed in the mid-myocardial wall as previously described3 to record changes in segment length with each beat. After electrode placement, systemic heparin (10 000 U followed by 2000 U/h) was administered, and a carotid-to-LAD shunt was created as previously described.3 13 Perfusion of the distal LAD was maintained at 1.2 mL/kg body wt per minute, which provides 1.2 to 1.5 mL of blood per gram of heart tissue per minute.3 13 A sidearm in the shunt was used for infusion of pinacidil. Flow through this sidearm was controlled by a second roller pump.
The heart rate was maintained between 100 and 120 bpm by atrial pacing in the experiments using a single rate. In those designed to determine the effects of increasing rates, rates of 100 and 150 bpm were used. Arterial blood pressure and a lead II ECG were continuously monitored on a 12-channel strip-chart recorder (WR3101, Western Graphtec Inc).
K+- and pH-sensitive electrodes were constructed and calibrated as previously described.14 Only electrodes demonstrating a stable baseline (drift <1 mV/h) and 95% to 105% of the predicted nernstian slope (56 to 62 mV shift per decade change in K+ activity or pH at room temperature) were used. Pairs of Teflon-coated silver wires, used to record bipolar electrograms, were placed with one or two K+ or pH electrodes and a reference electrode in a 19-gauge needle and inserted into the mid-myocardium. The needle was then withdrawn, leaving the electrode group embedded in the myocardium. As many as six such groups were placed at various locations within the ischemic zone, and one or two were placed outside of the ischemic zone. After electrode insertion, the in vivo performance of the K+ electrodes was tested as previously described.14 At the end of each experiment, the electrodes were removed from the hearts and recalibrated.
Transmembrane action potentials were recorded by “floating” 3 mol/L KCl–filled microelectrodes fabricated as previously described,15 using the tips of standard microelectrodes mounted on a tungsten wire 0.002 mm in diameter connected to a high-input resistance buffer amplifier. The reference electrode consisted of a silver/silver chloride wire placed as close as possible to the recording site. The action potentials were continuously monitored on an oscilloscope and recorded on the strip-chart recorder at a paper speed of 50 mm/s.
No intervention was performed for 50 minutes after electrode placement. The first in a series of 8-minute episodes of myocardial ischemia was induced by the abrupt cessation of flow through the LAD shunt. Each episode was separated by 50 minutes of reperfusion. We first performed two control ischemic periods because our earlier studies showed that the rise in [K+]e and the changes in activation in the first ischemic episode differed from those occurring during subsequent ischemic episodes.16 We then infused a solution of 500 μmol/L pinacidil into the sidearm of the shunt at a rate calculated to produce a drug concentration of 10 or 25 μmol/L in the blood supplying the LAD. The infusion of pinacidil was initiated after 30 minutes of reperfusion and maintained for 20 minutes, at which time coronary flow was discontinued for 8 minutes. After the pinacidil ischemic period, the LAD was reperfused with pinacidil-free blood for 20 minutes and a post–pinacidil ischemic period was performed.
Data Collection and Analysis
The signals from all electrodes were individually amplified by high-impedance amplifiers, digitized (Phoenix Data analog-to-digital converter), and simultaneously sampled (1000 samples per second) every 60 seconds during the pinacidil infusion and every 15 seconds during ischemia by a Micro Vax II/GPX computer (DEC). Values of [K+]e and pH were calculated from measured millivolt changes using the calibration curve of each electrode, the systemic [K+]e or pH determined from an arterial blood sample obtained immediately before the event and, for the K+ electrode, an activity coefficient of .746.17 APD was determined at 50% and 90% repolarization (APD50, APD90). APD50 served as a measure of the action potential plateau or phase 2, and the difference between APD90 and APD50 (APD90−APD50) served as a measure of phase 3. Local activation delay was determined from the high-frequency deflection of the bipolar electrogram filtered between 50 and 500 Hz and referenced to an electrode in the nonischemic zone. Local activation delay was determined only at sites where data from the K+ electrodes were acceptable. The number of acceptable K+ and pH electrodes in each experiment ranged from two to four. Mean values for [K+]e, pH, and activation delay were determined for each experiment and then used to calculate a mean for the experimental group as a whole.
In Vitro Rabbit Experiments
Preparations of the arterially perfused rabbit interventricular septum were adapted from previously reported studies.2 18 19 New Zealand White rabbits (weight, 2 to 3 kg) of either sex were heparinized (200 U/kg IV) and anesthetized with sodium thiopental (70 mg/kg IV). The heart was rapidly excised and placed in cold (4°C) Tyrode solution. The atrial and left ventricular free wall were removed and the ventricular septum was attached, left side down, to a wax platform containing a large silver-silver chloride grounded electrode. The septal branch of the left coronary artery was cannulated and perfused with solution containing the following (mmol/L): Na+ 149, K+ 4.5, Mg2+ 0.49, Ca2+ 1.8, Cl− 133, HCO3− 25, HPO42− 0.4, glucose 20, albumin 2 g/L, dextran (MW, 70 000) 40 g/L, insulin 1 U/L, heparin 400 U/L. The total ischemic time before starting perfusion was <5 minutes. The perfused preparation was placed in a recording chamber having a temperature of 36° to 37°C and surrounded by a humidified atmosphere consisting of 95% O2–5% CO2. Perfusion pressure, monitored by a transducer (Millar Instruments Inc), was maintained between 40 and 50 mm Hg by adjustment of the perfusion flow rate (0.8 to 1 mL−1·min−1 · g−1) via a roller pump (Digistaltic Cole, Parmer Instrument Co). The preparation was stimulated by 2-ms stimuli of twice diastolic threshold strength from a bipolar stainless steel electrode. The stimulation frequency was 2 Hz in experiments using only a single rate and 1 Hz and 3 Hz in those designed to study the effects of increasing rate. The partial pressures of O2 and CO2 in the perfusate were controlled with a membrane gas exchanger and continuously monitored by a pH glass electrode (Fisher Scientific) connected to the perfusion line. The relative amounts of O2, CO2, and N2 were adjusted by individual gas flowmeters (Cole Palmer Instrument Co) to yield a Po2 of 470±7.2 mm Hg (mean±SEM) and a pH of 7.35 to 7.45. The Po2, Pco2, and [HCO3−] in the perfusate were measured with a blood gas analyzer (system 1304 pH·Blood Gas Analyzer Instrumentation Laboratory, Lexington, Mass).
Microelectrodes and K+-Sensitive Electrodes
Transmembrane action potentials were measured with a 3 mol/L KCl–filled floating glass microelectrode referenced to a closely spaced extracellular electrode as described above. The action potentials were continuously monitored on an oscilloscope (Tektronix) and recorded on a Graphtec strip-chart recorder (WR3310, Western Graphtec Inc) at a paper speed of 250 mm/s. The K+-sensitive electrodes were fabricated using a polyethylene tube filled with 150 mmol/L KCl and coated with a polyvinylchloride-valinomycin–based membrane.19 In vitro and in situ calibration procedures were as described above. The tip of the K+-sensitive electrode was positioned on the surface of the myocardium as close as possible to the microelectrode. In some experiments, peak developed tension during isometric contraction was determined by attaching the tip of the papillary muscle to a force transducer, as previously described.19
In each experiment, the preparation stabilized in the chamber for at least 50 minutes before collection of control data. No-flow ischemia was induced by stopping perfusion and by replacing the 95% O2–5% CO2 gas mixture inside the chamber with 95% N2–5% CO2. After 6 minutes of ischemia, the preparation was reperfused and the normoxic surrounding atmosphere reestablished. After 40 minutes of reperfusion the perfusate was changed to one containing 1, 10, or 25 μmol/L pinacidil. After 20 minutes of perfusion with the pinacidil perfusate, a second 6-minute period of no-flow ischemia was created.
Data Collection and Analysis
Determination of [K+]e from K+ electrodes was as described for the pig. APD was measured at 50% and 90% repolarization. Activation delay was determined as the interval from the stimulus artifact to the peak of the action potential upstroke. Although more than one microelectrode penetration was often required in each experiment, the region from which action potentials were recorded in each experiment was very small, and no differences in activation time were noted with different penetrations before the onset of ischemia.
The statistical significance of the effect of pinacidil on the rise in [K+]e, shortening of APD, and the activation delay associated with no-flow ischemia was assessed for averages over time using a repeated-measures ANOVA with a two-way model for animal and treatment. We compared minutes 2 to 8 versus time zero for the pigs and minutes 2 to 6 versus time zero for the rabbits. The two-way repeated-measures ANOVA for averages over time as the primary focus enabled potential issues concerning multiple comparisons to be addressed appropriately through global assessments. To assess the effects of stimulation frequency on the rise in [K+]e, the shortening of APD, or the cumulative APD, we used the Wilcoxon signed ranks test on the averaged paired differences between the two stimulation frequencies. All results are expressed as mean±SEM. Statistical significance was considered at a value of P≤.05.
We performed 14 experiments in the pigs and 21 in the rabbit interventricular septa. Pinacidil 10 μmol/L was the only concentration tested in four pigs and 25 μmol/L the only concentration tested in six. In the remaining four animals, both concentrations were tested. In the rabbit we tested only a single concentration in each preparation. A concentration of 1 μmol/L was tested in five experiments, 10 μmol/L in seven, and 25 μmol/L in eight. Pinacidil 1 μmol/L had no effect on the ischemia-induced changes in the rise in [K+]e, the change in APD, or the change in activation.
Fig 1⇓ illustrates the changes in [K+]e, APD90, and activation associated with no-flow ischemia before and after the administration of 10 and 25 μmol/L pinacidil in both experimental models. In the pig, [K+]e rose from 3.3±0.1 to 10.1±0.3 mmol/L after 8 minutes of ischemia before pinacidil. Pinacidil 10 μmol/L caused a slight but significant lessening in the rise in [K+]e (P<.05). After 25 μmol/L pinacidil, the rise in [K+]e was significantly less than control or after 10 μmol/L (P<.04). In the rabbit, [K+]e rose from 4.5 (the value in the perfusate) to 9.4±0.3 mmol/L after 6 minutes of ischemia before pinacidil. Pinacidil 10 μmol/L did not significantly affect the [K+]e rise. After 25 μmol/L, the rise in [K+]e from 4.6±0.1 to 7.6±0.4 mmol/L was significantly less than the control (P<.01).
Before pinacidil, APD lengthened during the first 2 minutes of ischemia in both the rabbit and the pig (Figs 1⇑ and 2⇓). This was due to lengthening of phase 3 of the action potential as reflected by an increase in the difference between APD90 and APD50 (APD90−APD50). In the examples shown in Fig 2⇓, APD90−APD50 increased from 30 to 60 ms in the pig and from 29 to 49 ms in the rabbit. Thereafter, APD shortened due primarily to a shortening of the plateau as reflected by the shortening of APD50 from 300 to 200 ms in the pig and from 125 to 108 ms in the rabbit.
Pinacidil 10 μmol/L caused only minimal shortening of the APD before ischemia. However, it prevented the initial lengthening of APD induced by ischemia and exaggerated the subsequent shortening (P<.01) in both preparations. In the pig experiment shown in Fig 2⇑, 25 μmol/L pinacidil administered for 20 minutes before the onset of no-flow ischemia shortened APD50 from 260 to 200 ms but had no effect on APD90−APD50. These effects were similar in the rabbit. After ischemia, APD shortened immediately in both species, due in large part to an effect on the plateau. In the pig, APD50 shortened from 200 to 60 ms while phase 3 (APD90−APD50) was unchanged. In the rabbit, the plateau of the action potential was virtually nonexistent after 6 minutes of ischemia, and APD90−APD50 shortened from 33 to 16 ms. The shortening of APD was significantly greater after 25 μmol/L pinacidil and then after 10 μmol/L in the pig (P<.01) but not in the rabbit.
The effect of pinacidil on the activation delay induced by no-flow ischemia paralleled its effect on the rise in [K+]e. Before pinacidil in the pig, activation began to change ≈90 seconds after the onset of no-flow ischemia and lengthened by a maximum of 72±5 ms. Pinacidil 10 μmol/L did not significantly influence this activation delay. However, after 25 μmol/L pinacidil, activation did not change until 4 minutes after the initiation of ischemia and lengthened by only 22±4 ms. This difference was statistically significant. Similar results were observed in the rabbit. Neither 10 nor 25 μmol/L pinacidil influenced the incidence of ventricular fibrillation in the pigs (2 of 14 in control animals; 1 of 8 after 10 μmol/L pinacidil and 2 of 10 after 25 μmol/L pinacidil).
We next determined the effect of increasing stimulation frequency on the rise in [K+]e during ischemia. We reasoned that the accentuated shortening of the action potential plateau during ischemia induced by pinacidil might have lessened the time available for K+ efflux to occur. By increasing stimulation frequency, the total time during ischemia when the action potential would be at its plateau level, and thus the total time for K+ efflux to occur via the various plateau K+ channels, would increase. Fig 3⇓ illustrates the effect of increasing driving rate from 100 to 150 bpm in 6 pig experiments. Panels A show the rise in [K+]e; panels B, the change in APD50; and panels C, the cumulative time in seconds, during which the transmembrane voltage was at the plateau level. This value was determined by multiplying APD50 by the number of beats in each 30-second interval throughout the ischemic period and then summing the results.
Before pinacidil (top panels), the ischemia-induced rise in [K+]e was similar at both driving rates; ie, there was no rate-dependent effect (panel A). APD50 shortened from 331±31 to 193±34 ms when the rate was 100 bpm and from 234±15 to 160±24 ms when the rate was 150 bpm (panel B). There was no significant difference in the cumulative time during which the transmembrane potential was at the plateau level during ischemia at the two rates (panel C).
After pinacidil (bottom panels), the rise in [K+]e during ischemia was less at each rate than before the drug and the rise in [K+]e at 150 bpm was significantly greater than at 100 bpm (P<.05); ie, there was a rate-dependent effect. Pinacidil shortened APD before and during ischemia at both rates, but after pinacidil there was no significant difference between APD50 during ischemia at the two rates. The cumulative number of seconds during ischemia when the transmembrane voltage was at the plateau level was less at each rate than before pinacidil, and the cumulative value at 150 bpm was significantly greater than at 100 bpm (P<.05).
Fig 4⇓ illustrates these effects in the rabbit (n=5). Before pinacidil, the [K+]e rise during ischemia was significantly greater at 3 Hz than at 1 Hz (P<.05); ie, there was a rate-dependent effect (panel A, top). APD50 shortened from 102±3 to 71±5 ms during 6 minutes of ischemia at 1 Hz and from 99±5 to 45±5 ms at 3 Hz (P=.0625, the minimum possible two-sided probability value with 5 animals) (panel B, top). The cumulative number of seconds during which the transmembrane voltage was at that plateau level during ischemia was also consistently greater at 3 Hz (P=.0625, the minimum possible two-sided probability value) (panel C, top).
After pinacidil, the ischemia-induced rise in [K+]e at each rate was less than before pinacidil, but the rate-dependent effect persisted (P<.03) (panel A, bottom). Pinacidil shortened APD50 before ischemia at both frequencies. During ischemia, APD50 decreased from 65±6 to 15±4 ms at 1 Hz and from 52±9 to 8±2 ms at 3 Hz (panel B, bottom). The cumulative number of seconds during which the transmembrane voltage was at the plateau level during ischemia remained greater at 3 Hz than at 1 Hz (P=.0625).
Examination of the rabbit data reveals that the ischemia-induced rise in [K+]e at 1 Hz before pinacidil and at 3 Hz after 25 μmol/L pinacidil were similar (Fig 5⇓, upper panel). The cumulative number of seconds during which the transmembrane voltage was at the plateau level was also similar at 1 Hz before pinacidil and at 3 Hz after pinacidil. In the pig, examination of the data obtained during similar conditions, ie, at a rate of 100 bpm before and 150 bpm after 25 μmol/L pinacidil (Fig 5⇓, lower panels), reveals that both the [K+]e rise and the cumulative number of seconds at the plateau voltage were significantly greater (P<.05) at the slower rate before pinacidil than at the more rapid rate after pinacidil. The results, shown in Figs 3⇑, 4⇑, and 5⇓, indicate that the total time during ischemia when the transmembrane voltage was at the plateau level of the action potential correlates with the rise of [K+]e. This correlation is further illustrated in Fig 6⇓, which shows correlation coefficients of .6 to .9 under the various conditions explored in these studies.
The significantly greater decrease in the rise in [K+]e after 25 μmol/L, then 10 μmol/L pinacidil, despite only a slightly greater decrease in cumulative plateau duration, suggests that the higher concentration may also have provided greater metabolic protection. Support for this hypothesis is shown in Fig 7⇓, which illustrates the recorded fall in pHe during ischemia before and after pinacidil (solid lines) and the fall anticipated in the second of the two sequential pinacidil-free occlusions (broken line). These data are derived from earlier studies16 that showed that the decrease in pHe decreased progressively in sequential ischemic periods although the rise in [K+]e remained constant. The figure illustrates that the fall in pHe after 10 μmol/L was similar to that anticipated by our earlier studies,16 whereas the fall in pHe after 25 μmol/L was less than anticipated. This suggests that in the in situ pig heart, 25 μmol/L pinacidil afforded greater metabolic protection than did 10 μmol/L.
The triphasic rise in [K+]e that characterizes no-flow ischemia1 has been attributed to anionic-linked factors involving the movement of K+ ions out of the cell in association with anions generated during ischemia6 and to nonanionic factors, which include K+ efflux via voltage and ligand-gated K+ channels, ionic pumps, and transporters.5 The voltage-gated K+ channels include the channels responsible for the transient outward current (Ito), the delayed rectifying K+ current (IK), and the inward rectifying K+ current (IK1). The ligand-gated K+ channels include the ATP-sensitive K+ channel,20 the sodium-activated K+ channel,21 and the K+ channel activated by arachidonic acid.22 K+ efflux through these channels is determined by K+ conductance and K+ electrical driving force, which, in turn, is determined by the difference between the transmembrane potential and the K+ equilibrium potential (Em−Ek).
Inhibition of the ATP-sensitive K+ channel with glibenclamide and tolbutamide lessens the rise in [K+]e during acute ischemia.8 9 10 11 Thus, agents that enhance conductance of the ATP-sensitive K+ channel might reasonably be expected to accentuate the rise in [K+]e. Venkatesh et al23 found that 5 μmol/L cromakalin accelerated the shortening of the APD induced by ischemia in the perfused rabbit interventricular septum but did not significantly modify the initial rise in [K+]e. They postulated that the decreased driving force for K+ associated with the shortening of the action potential plateau balanced the increased efflux of K+ through the ATP-sensitive K+ channel. Mitani et al24 reported that in the rat ventricle, nicorandil caused an initial increase in the rise in [K+]e. However, these investigators did not measure APD, and it is possible that the shorter APD of the rat ventricle may have led to a different relationship between APD shortening and the rise in [K+]e.
The purpose of our study was to determine whether pinacidil would enhance the rise in [K+]e during acute ischemia by increasing K+ conductance or would decrease the rise in [K+]e by shortening APD and by providing metabolic protection. We used two concentrations of pinacidil (10 and 25 μmol/L) and two preparations, the in situ porcine heart and the isolated, perfused rabbit interventricular septum, to determine which if any of our observations might be concentration or species dependent. Our results demonstrate that 10 μmol/L pinacidil does not significantly alter the initial rise in [K+]e in the rabbit but that 25 μmol/L pinacidil significantly lessens the rise in [K+]e during 8 minutes of no-flow ischemia in the pig and 6 minutes of ischemia in the rabbit.
The primary effect of pinacidil in both preparations was to shorten the action potential plateau before ischemia, to prevent the lengthening of APD during the first 2 minutes of ischemia, and to accentuate the subsequent shortening of the action potential plateau. This effect on the plateau would have lessened the time for K+ efflux via the ATP-sensitive K+ channel to occur and was the mechanism proposed to explain the lack of effect by cromakalin on the rise in [K+]e.23 The importance of plateau duration on the rise in [K+]e during ischemia is suggested by the correlation of the rate-related changes in the ischemia-induced [K+]e rise to the cumulative time during which the transmembrane potential was at its plateau voltage both in the presence and absence of pinacidil.
The role of plateau duration on K+ efflux during ischemia should also apply to K+ channels other than the ATP-sensitive K+ channel and influence the rise in [K+]e, if in fact K+ efflux via these other channels contributes to the rise. Mitani and Shattock25 showed that inhibition of the sodium-activated K+ channel in the rat lessened the rise in [K+]e during ischemia, and Hicks and Cobbe26 showed that d-sotalol, a putative blocker of IK, lessened the ischemia-induced rise in [K+]e in the rabbit. Studies in our laboratory suggest that inhibition of Ito by 4-aminopyridine and IKr by E-4031 also lessens the rise in [K+]e during ischemia.27 Thus, we believe that the shortening of the plateau duration by pinacidil lessened the time during which K+ efflux occurred not only through the ATP-sensitive K+ channel but also through other voltage- and ligand-gated K+ channels.
The shortening of the plateau induced by 10 μmol/L pinacidil was only slightly less than that induced by 25 μmol/L pinacidil, suggesting that the increase in K+ conductance in the ATP-sensitive K+ channel may have been only slightly greater at the higher dose. An increase in K+ conductance would be expected to enhance the rise in [K+]e during ischemia. However, 10 μmol/L pinacidil had no effect on the ischemia-induced rise in [K+]e in the rabbit and only a slight effect in the pig, while 25 μmol/L pinacidil lessened the rise in [K+]e in both. This suggests that factors other than the increase in K+ efflux through the ATP-sensitive K+ channel and plateau duration contributed to the [K+]e rise.
Noma20 postulated that activation of the ATP-sensitive K+ channel, with its effect on APD, would decrease the time available for calcium to enter the cell, reduce contractility, lessen the rate of decline of high-energy phosphates, thereby providing metabolic protection, and lessen the onset of irreversible ischemic injury. Such effects have been demonstrated in the arterially perfused guinea pig right ventricular wall.28 29 Our studies of contractility in a few animals are consistent with this scenario. Evidence of the metabolic protection afforded by 25 μmol/L pinacidil is provided by our studies of the changes in extracellular pH, illustrated in Fig 7⇑. This metabolic protection would not only have lessened the rise in [K+]e as a result of its effect on anion-linked factors but might also have led to a reduced increase in intracellular sodium and thereby reduced K+ efflux via the sodium-activated K channel. Additional nonspecific effects of pinacidil might also have contributed to its effect on the rise in [K+]e.
In many ways, the effects of 25 μmol/L pinacidil on the rise in [K+]e13 and the depletion in high-energy phosphates30 resemble the effects of the L-type calcium channel–blocking agents verapamil and diltiazem.13 30 The effects of the calcium channel–blocking agents and pinacidil on intracellular pH and intracellular calcium might also be similar, even though verapamil has been reported to block the ATP-sensitive K+ channel.31 The similarity of the effects of verapamil and pinacidil on the rise in [K+]e provides additional support for the concept that this rise is multifactorial, influenced by both anionic- and nonionic-linked mechanisms, and that the ATP-sensitive K+ channel is but one of these mechanisms. The influence of any intervention on the rise in [K+]e will be determined by the balance of its effects on K+ conductance through the various voltage- and ligand-gated K+ channels, on K+ driving force as reflected by the time during which the action potential is at its plateau voltage, and on the metabolic changes induced by ischemia. Pinacidil appears to influence all of these parameters. In a concentration of 25 μmol/L, its effects on action potential plateau duration and on the metabolic changes of no-flow ischemia appear to overwhelm its effects on the conductance of the ATP-sensitive K+ channel and cause an unexpected lessening of the ischemia-induced rise in [K+]e.
This research was supported in part by grants PO1-HL-27430 and 5R-37-HL-38885 from the National Heart, Lung, and Blood Institute, Bethesda, Md. Pinacidil was provided by the Eli Lily Co, Indianapolis, Ind. We are grateful to Dr Wayne Cascio for his insightful comments and criticisms, to Zoe Sherman, Weili Sun, and James Holmes for their technical assistance, and to Anne Middleton for her expert secretarial support.
- Received July 23, 1996.
- Revision received November 14, 1996.
- Accepted November 25, 1996.
- Copyright © 1997 by American Heart Association
Hill JL, Gettes LS. Effect of acute coronary artery occlusion on local myocardial extracellular K+ active in swine. Circulation. 1980;61:768-778.
Weiss JN, Shine KI. [K+]o accumulation and electrophysiological alternations during early myocardial ischemia. Am J Physiol. 1982;243:H318-H327.
Watanabe I, Johnson TA, Buchanan J, Engle CL, Gettes LS. Effect of graded coronary flow reduction on ionic, electrical, and mechanical indexes of ischemia in the pig. Circulation. 1987;76:1127-1134.
Gettes LS, Cascio WE. Effect of acute ischemia on cardiac electrophysiology. In: Fozzard HA, ed. The Heart and Cardiovascular System. New York, NY: Raven; 1992:2021-2054.
Shieh RC, Goldhaber JI, Stuart JS, Weiss JN. Lactate transport in the mammalian ventricle: general properties and relation to K+ fluxes. Circ Res. 1994;74:829-838.
Wilde AAM, Escande D, Schumacher CA, Thuringer D, Meestre M, Fiolet JWT, Janse MJ. Potassium accumulation in the globally ischemic mammalian heart: a role for the ATP-sensitive potassium channel. Circ Res. 1990;67:835-843.
Kantor PF, Coetzee WA, Carmeliet EE, Dennis SC, Opie LH. Reduction of ischemic K+ loss and arrhythmias in rat hearts: effect of glibenclamide, a sulfonylurea. Circ Res. 1990;66:478-485.
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. 1991;69:623-637.
Bekheit SS, Restivo M, Boutjdir M, Henkin R, Gooyandeh K, Assadi M, Khatibs S, Gough WB, El-Sherif N. Effects of glyburide on ischemia-induced changes in extra potassium and local myocardial activation: a potential new approach to the management of ischemia-induced malignant arrhythmias. Am Heart J. 1990;119:1025-1033.
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. 1989;65:436-445.
Fleet WF, Johnson TA, Graebner CA, Engle CL, Gettes LS. Effects of verapamil on ischemia-induced changes in extracellular K+, pH, and local activation in the pig. Circulation. 1986;73:837-846.
Johnson TA, Engle CL, Jusy RP, Kinsley SB, Graebmer CA, Gettes LS. Fabrication, evaluation and use of extracellular [K+]e and H+ ion-selective electrodes. Am J Physiol. 1990;258:H1224-H1231.
Gettes LS, Surawicz B, Shiue JC. Effect of high K, low K, and quinidine on QRS duration and ventricular action potential. Am J Physiol. 1962;203:1135-1140.
Fleet WF, Johnson TA, Graebner CA, Gettes LS. Effect of serial brief ischemic episodes on extracellular K+ pH and activation in the pig. Circulation. 1985;72:922-932.
Cascio WE, Yan G-X, Kléber AG. Passive electrical properties, mechanical activity and extracellular potassium in artery-perfused and ischemic rabbit ventricular muscle: effects of calcium entry blocked or hypocalcemia. Circ Res. 1990;66:1461-1473.
Kim D, Claphan DE. Potassium channel in cardiac cells activated by arachidonic acid and phospholipids. Science. 1989;244:1174-1179.
Venkatesh N, Stuart J, Lamp ST, Alexander LD, Weiss JN. Activation of ATP-sensitive K+ channels by cromakalim: effects on cellular K+ loss and cardiac function on ischemic and reperfused mammalian ventricle. Circ Res. 1992;71:1324-1333.
Mitani A, Kinoshita K, Fukamachi K, Sakamoto M, Kurisu K, Tsuruhara Y, Fukumura F, Nakashima A, Tokunaga K. Effects of glibenclamide and nicorandil on cardiac function during ischemia and reperfusion in isolated perfused rat hearts. Am J Physiol. 1991;261:H1864-H1871.
Mitani A, Shattock MJ. Role of Na-activated K channel, Na-K-Cl cotransport, and Na-K pump in [K]e changes during ischemia in the rat heart. Am J Physiol. 1992;263:H333-H340.
Hicks MN, Cobbe SM. Attenuation of the rise in extracellular potassium concentration during myocardial ischemia by dL-sotalol and d-sotalol. Cardiovasc Res. 1990;24:404-410.
Kanda A, Watanabe I, Gettes LS. Ito and IKr but not IK1 contribute to rise in extracellular K during ischemia. PACE. 1995;18:934. Abstract.
Cole WC, McPherson CD, Sontag D. ATP-regulated K+ channels protect the myocardium against ischemia-reperfusion damage. Circ Res. 1991;69:571-581.
McPherson CD, Pierce GN, Cole WC. Ischemic cardioprotection by ATP-sensitive K+ channels involves high-energy phosphate preservation. Am J Physiol. 1993;265:H1809-H1818.
Steenbergen C, Murphy E, Watts JA, London RE. Correlation between cytosolic free calcium, contracture, ATP, and irreversible ischemic injury in perfused rat heart. Circ Res. 1990;66:135-146.
Kimura S, Bassett AL, Xi H, Myerberg RJ. Verapamil diminishes action potential changes during metabolic inhibitor by blocking ATP-regulated potassium currents. Circ Res. 1992;71:87-95.