Sarcolemmal and Mitochondrial Adenosine Triphosphate– dependent Potassium Channels: Mechanism of Desflurane-induced Cardioprotection

All experimental procedures and protocols used in this investigation were reviewed and approved by the Animal Care and Use Committee of the Medical College of Wisconsin. All procedures were in conformity with the Guiding Principles in the Care and Use of Animals  of the American Physiologic Society 24and were performed in accordance with the Guide for the Care and Use of Laboratory Animals . 25 

The experimental methods have been previously described in detail. 9Briefly, mongrel dogs (weight = 26 ± 1 kg, mean ± SEM) were anesthetized with sodium barbital (200 mg/kg) and sodium pentobarbital (15 mg/kg) and ventilated with an air and oxygen mixture (fraction of inspired oxygen = 0.25) after tracheal intubation. Tidal volume and respiratory rate were adjusted to maintain arterial blood gas tensions within a physiologic range. After calibration, a double pressure transducer–tipped catheter was inserted into the aorta and left ventricle (LV) via  the left carotid artery to measure aortic and LV pressures, respectively. The maximum rate of increase of LV pressure (+dP/dt^max) was obtained by electronic differentiation of the LV pressure waveform. The femoral artery and vein were cannulated for the withdrawal of reference blood flow samples and fluid administration, respectively. A thoracotomy was performed at the left fifth intercostal space. A heparin-filled catheter was inserted into the left atrial appendage for administration of radioactive microspheres. A 1.0-cm segment of the left anterior descending (LAD) coronary artery was dissected immediately distal to the first diagonal branch, and a silk ligature was positioned around this vessel for production of coronary artery occlusion and reperfusion. Regional myocardial perfusion was measured in the ischemic (LAD) and normal (left circumflex coronary artery) zones using radioactive microspheres. 9Myocardial infarct size was determined with triphenyltetrazolium chloride staining at the completion of each experiment. 26End-tidal concentrations of desflurane were measured at the tip of the endotracheal tube by an infrared anesthetic gas analyzer that was calibrated with known standards before and during experimentation. The canine minimum alveolar concentration value of desflurane used in the present investigation was 7.2%. 27Hemodynamic data were continuously monitored throughout the experiment, recorded on a polygraph, and digitized using a computer interfaced with an analog-to-digital converter.

The experimental design used in the present investigation is illustrated in figure 2. Ninety minutes after completion of the surgical preparation, dogs were randomly assigned to one of eight experimental groups. All dogs underwent a 60-min LAD occlusion followed by 3-h reperfusion. In four groups of experiments, dogs received 0.9% saline (control) or the nonspecific K^ATPchannel antagonist glyburide (0.1 mg/kg intravenously) in the presence and absence of 1.0 minimum alveolar concentration desflurane (end-tidal concentration). These experiments tested the hypothesis that desflurane reduces myocardial infarct size by K^ATPchannel activation. To determine further whether the myocardial protection produced by desflurane was related to sarcolemmal or mitochondrial K^ATPchannels, four additional groups of dogs were pretreated with intracoronary infusions of HMR 1098 (1 μg · kg⁻¹· min⁻¹in 10 ml 0.9% saline over 45 min, a dose comparable to that used previously via  an intravenous route, 19) or 5-HD (150 μg · kg⁻¹· min⁻¹in 10 ml 0.9% saline over 45 min, a dose previously used to block ischemic preconditioning in dogs 1) in the presence or absence of 1.0 minimum alveolar concentration desflurane. Infusions of HMR 1098 and 5-HD were initiated 10 min before, continued during, and discontinued 5 min after the administration of desflurane.

Statistical analysis of data within and between groups was performed with multiple analysis of variance for repeated measures with post hoc  analysis by Student-Newman-Keuls test. Changes within and between groups were considered statistically significant at P < 0.05. All data are expressed as mean ± SEM.

Eighty-eight dogs were instrumented, and 63 successful experiments were completed. Four dogs were excluded because of intractable ventricular fibrillation during LAD occlusion or reperfusion (one control, one desflurane, one glyburide, one 5-HD). Fourteen dogs were excluded from analysis because transmural coronary collateral blood flow exceeded 0.2 ml · min⁻¹· g⁻¹(four desflurane, one glyburide, four 5-HD, one HMR 1098, one desflurane + 5-HD, three desflurane + HMR 1098). Four dogs were excluded because of technically difficult intracoronary catheter insertion (one desflurane, one 5-HD, one HMR 1098, one desflurane + HMR 1098). Two dogs were excluded because of the presence of heart worms (one desflurane, one HMR 1098). One dog was excluded because of profound hypotension throughout the experiment (HMR 1098).

No differences in baseline systemic hemodynamics were observed between experimental groups (table 1). Glyburide produced no hemodynamic effects. Desflurane caused significant (P < 0.05) decreases in heart rate, mean arterial and LV systolic pressures, rate-pressure product, and LV +dP/dt^max, and an increase in LV end-diastolic pressure. Desflurane produced similar hemodynamic effects in the presence and absence of glyburide. Intracoronary administration of HMR 1098 or 5-HD alone did not cause hemodynamic effects. The cardiovascular actions of desflurane were not affected by HMR 1098 or 5-HD pretreatment. LAD occlusion increased LV end-diastolic pressure in all groups, and there were no hemodynamic differences between groups during coronary artery occlusion or reperfusion.

Transmural myocardial blood flow in the ischemic (LAD) region is summarized in table 2. There were no intergroup differences in myocardial blood flow before or during LAD occlusion or reperfusion.

The area at risk was similar between groups (control, 44 ± 3%; desflurane, 42 ± 3%; glyburide, 43 ± 2%; desflurane + glyburide, 46 ± 1%; HMR, 46 ± 2%; desflurane + HMR, 46 ± 2%; 5-HD, 49 ± 2%; desflurane + 5-HD, 43 ± 1% of the LV). Desflurane significantly reduced myocardial infarct size to 10 ± 2% of the area at risk (fig. 3) compared with control experiments (25 ± 3%). Glyburide abolished the protective effects of desflurane (25 ± 2%) but had no effect on infarct size when administered alone (24 ± 2%). HMR 1098 and 5-HD did not affect infarct size (21 ± 4% and 24 ± 2%, respectively;figs. 4 and 5) but blocked the protective effects of desflurane (19 ± 3% and 22 ± 2%, respectively).

Experimental evidence accumulated in recent years indicates that K^ATPchannels play a central role in volatile anesthetic–induced preconditioning. 3,5,6Isoflurane and sevoflurane have been shown to reduce reversible and irreversible ischemic injury by activating K^ATPchannels. 3,5,6Isoflurane also produces an acute memory phase of myocardial protection by a K^ATPchannel–mediated mechanism, an action similar to that observed with a brief ischemic stimulus. 3We have recently demonstrated that sevoflurane reduces the time threshold of ischemic preconditioning in vivo , 6demonstrating the additive actions of a brief ischemic episode and a volatile anesthetic agent at the K^ATPchannel. The present results with desflurane confirm and extend findings with other volatile anesthetics and indicate that this agent also exerts cardioprotective effects against irreversible ischemic injury. These beneficial effects were blocked by glyburide, indicating that desflurane-induced myocardial protection ultimately occurs through K^ATPchannels. Furthermore, the reduction in infarct size produced by desflurane also occurred independent of alterations in systemic hemodynamics and transmural coronary collateral blood flow.

Adenosine triphosphate–dependent potassium channels are clearly involved in anesthetic-induced myocardial protection, but the subcellular location of these channels has not been defined. Noma 28originally suggested that sarcolemmal K^ATPchannel opening may hyperpolarize the cardiac myocyte in the presence of reduced intracellular concentrations of ATP during ischemia. Such sarcolemmal hyperpolarization reduces myocyte action potential duration 15,29,30and partially inhibits voltage-dependent calcium (Ca²⁺) channel activity. Subsequent reductions in myocardial contractility 31and intracellular Ca²⁺overload 30may preserve intracellular energy stores 32for vital processes during ischemia and reperfusion. Sarcolemmal K^ATPchannel opening would maintain the normal function of the sodium (Na⁺)–Ca²⁺exchanger, further reducing intracellular Ca²⁺accumulation. 30Sarcolemmal K^ATPchannels may also exert protective effects independent of action potential duration, 33,34because protective effects of K^ATPchannel openers have been observed without concomitant changes in action potential duration 15,16and in unstimulated 35,36and electrically inactive 33cardiac myocytes. Furthermore, the specific sarcolemmal K^ATPchannel subunit confers protection to transfected nonmyocyte cells in the absence of a developed action potential. 33Alternatively, mitochondrial K^ATPchannels 37have recently been proposed as a site of action for K^ATPchannel openers, 17,38,39and these channels may play a central role in ischemic preconditioning. 18,21Diazoxide, a selective mitochondrial K^ATPchannel opener, reduced myocardial injury in isolated rat hearts subjected to ischemia and reperfusion. 21This beneficial action was blocked by pretreatment with 5-HD. These data indicated that K^ATPchannels located in the mitochondria may be involved in reducing ischemic injury. The precise mechanism through which mitochondrial K^ATPchannels mediate such protective effects has yet to be determined. Opening of these channels causes transient mitochondrial K⁺uptake and matrix swelling, effects that seem to favorably modulate mitochondrial metabolism. 38–40The importance of sarcolemmal versus  mitochondrial K^ATPchannel opening during ischemic preconditioning is unresolved. Previous investigation is not unequivocal in favor of the sarcolemmal versus  the mitochondrial K^ATPchannel, and evidence for the involvement of both channels during ischemic preconditioning has recently been presented. 41,42 

The current investigation is the first to examine the subcellular K^ATPchannel sites responsible for anesthetic-induced preconditioning. The results indicate that specific blockade of sarcolemmal K^ATPchannels with HMR 1098 abolishes the protective actions of desflurane. This finding suggests that sarcolemmal K^ATPchannel activation by volatile agents plays a role in reducing myocardial ischemic injury. In addition, 5-HD blocked the decrease in infarct size produced by desflurane, findings that also implicate a role for mitochondrial K^ATPchannel activation in anesthetic-induced preconditioning. It is unknown if interactions exist between sarcolemmal and mitochondrial K^ATPchannels during ischemic- or anesthetic-induced preconditioning.

The present findings must be interpreted within the constraints of several potential limitations. Desflurane-induced decreases in the rate–pressure product may have produced favorable alterations in myocardial oxygen supply–demand and contributed to a reduction in infarct size. However, K^ATPchannel blockade with selective and nonselective antagonists completely abolished the protective effect of desflurane without affecting the hemodynamic actions of this agent. Volatile anesthetics also mediate protective effects during mechanical arrest produced by cardioplegia, 43indicating that preferential alterations in myocardial metabolism are not solely responsible for the antiischemic actions of these drugs. Nevertheless, coronary venous oxygen tension was not measured, and myocardial oxygen consumption was not directly quantified in the present investigation; thus, favorable changes in myocardial metabolism during administration of desflurane cannot be completely excluded as a mechanism for the beneficial effect of this drug. HMR 1098 is the water-soluble salt of the selective sarcolemmal K^ATPchannel antagonist HMR 1883. 19,20Recent data using several different models suggest that HMR 1098 demonstrates a similar high degree of selectivity for sarcolemmal K^ATPchannels (E. Marban, personal communication) at a concentration similar to that achieved in the current investigation (1 μm; calculated assuming coronary blood flow = 40 ml/min). However, the specificity of HMR 1098 has not been confirmed in canine myocardium; therefore, it is also possible that this drug abolished the protective effects of desflurane by an indirect mechanism or by blockade of mitochondrial K^ATPchannels. 5-HD has been shown to abolish the cardioprotective effects of the selective mitochondrial K^ATPchannel opener diazoxide 17,18,21,22and to inhibit mitochondrial flavoprotein oxidation produced by the K^ATPchannel agonist pinacidil while leaving sarcolemmal K^ATPcurrent unaffected. 225-HD has also been shown to antagonize the beneficial actions of the K^ATPchannel agonist cromakalim without influencing action potential duration. 23Although the specificity of 5-HD for mitochondrial K^ATPchannels has not been confirmed in canine myocardium, these findings suggest that 5-HD preferentially blocks mitochondrial K^ATPchannels at a concentration 1,17,22,23similar to that achieved (450 μm) in the current investigation. It is unknown if higher concentrations of HMR 1098 and 5-HD may be incompletely selective for specific subcellular K^ATPchannel locations. The actions of anesthetics to specifically enhance activation of sarcolemmal or mitochondrial K^ATPchannels will require future investigation using patch clamp and flavoprotein fluorescence techniques. Interpretation of the present findings should also be qualified because only a single end-tidal concentration of desflurane was used. Higher inspired concentrations of desflurane may have produced more pronounced reductions in infarct size and may have altered the subcellular locus of action of this anesthetic.

In summary, the present results indicate that desflurane reduces experimental myocardial infarct size after prolonged coronary artery occlusion and reperfusion. Desflurane-induced cardioprotection is dependent on K^ATPchannel activation, and selective antagonists of both sarcolemmal and mitochondrial K^ATPchannels block reductions of infarct size afforded by this drug in vivo .

The authors thank Drs. Werner List and Helfried Metzler (University of Graz, Austria) and Raul Trillo (Director of Medical Services, Baxter Pharmaceutical Products Inc.) for their gracious support, and John Tessmer and David Schwabe for technical assistance. The authors also thank Hoechst-Marion-Roussel, Frankfurt, Germany for providing HMR 1098.