Distinct Roles for Sarcolemmal and Mitochondrial Adenosine Triphosphate-sensitive Potassium Channels in Isoflurane-induced Protection against Oxidative Stress

The animal use and experimental protocols of this study were approved by the Animal Use and Care Committee of the Medical College of Wisconsin, Milwaukee, Wisconsin.

Ventricular myocytes were isolated from hearts of adult male Wistar rats (150–250 g) by enzymatic dissociation. The rats were injected with heparin (1,000 U intraperitoneally) and anesthetized with thiobutabarbital (Inactin, 100 mg/kg intraperitoneally; Sigma-Aldrich, St. Louis, MO). After thoracotomy, the hearts were excised, mounted on a Langendorff apparatus, and retrogradely perfused with heparinized Joklik medium (Gibco BRI; Invitrogen, Grand Island, NY) at the flow rate of 7 ml/min. After the blood had been washed out, the perfusate was replaced with an enzyme solution containing Joklik medium supplemented with 0.5 mg/ml collagenase type II (Invitrogen, Carlsberg, CA), 0.25 mg/ml protease XIV (Sigma-Aldrich), and 1 mg/ml bovine serum albumin (Serologicals, Kankakee, IL) at pH 7.23. All solutions were continuously gassed with a mixture of 95% O²–5% CO²and were kept at 37°C. After 25 min of enzyme digestion, the ventricles were excised, minced, and incubated in the enzyme solution for additional 8–10 min in a shaker bath at 37°C. The cell suspension was filtered through 200 μm mesh and centrifuged. The cell pellet was then washed twice in modified Tyrode solution (132 mm NaCl, 10 mm HEPES, 5 mm glucose, 5 mm KCl, 1 mm CaCl², and 1.2 mm MgCl², adjusted to pH 7.4). After isolation, the myocytes were stored in Tyrode solution at room temperature and allowed to recover for 1 h before the cell survival experiments. All experiments were performed within 5 h after the cell isolation procedure.

The suspension of isolated cardiomyocytes (1 ml) was placed in a chamber on the stage of inverted microscope (Diaphot 300; Nikon, Tokyo, Japan), and cells were allowed to settle for 10 min. The myocytes were then stained with 1 ml trypan blue solution, 0.4% (Sigma-Aldrich), for 2 min followed by a dye washout with glucose-free Tyrode solution (132 mm NaCl, 10 mm HEPES, 5 mm KCl, 1 mm CaCl², and 1.2 mm MgCl², adjusted to pH 7.4). Cells that were rod-shaped and excluded trypan blue were considered living21and were counted. The counting time was monitored and was kept uniform in all experiments (10 min). In each experiment, approximately 250 myocytes were counted. After cell counting, perfusion with glucose-free Tyrode was started. Thirty minutes into the experiment, the myocytes were exposed to oxidative stress by perfusion with 200 μm H²O²and 100 μm FeSO⁴·7H²O (both from Sigma-Aldrich) for 17 min. The mixture of H²O²and Fe²⁺yields a highly reactive hydroxyl radical (OH·) via  the Fenton reaction.22The 17-min duration was chosen because it was optimal for damaging approximately 50% of myocytes. After oxidative stress, the myocytes were reperfused with glucose-free Tyrode solution for another 20 min and stained with trypan blue, after which the remaining living cells were counted. The percentage of cell death was calculated from the cell count before and after oxidative stress. The location of the myocytes was monitored using a chamber bottom with a labeled grid. This enabled counting of the same myocytes before and after oxidative stress.

The time between cell counts before and after stress was exactly 67 min in all experiments. In the experimental groups that underwent APC, the myocytes were exposed to isoflurane for 20 min and 5 min anesthetic washout before oxidative stress. To investigate the effects of the sarcK^ATPand mitoK^ATPchannels on cell survival, the specific inhibitor HMR-1098 (50 μm; a gift from Aventis Pharma, Frankfurt am Main, Germany) or 5-hydroxydecanoic acid (5-HD, 200 μm; Sigma-Aldrich, St. Louis, MO) was added to the superfusing solution at different time points. In all experimental groups, n indicates the number of animals. The protocols for all experimental groups are illustrated in figure 1.

Isoflurane (Abbott Laboratories, North Chicago, IL) was dispersed in glucose-free Tyrode solution by sonication and delivered to cardiomyocytes from the airtight glass syringes. At the end of each experiment, samples of isoflurane-containing solution were collected from the chamber, and the concentrations of isoflurane were analyzed by gas chromatography (Shimadzu, Kyoto, Japan). The average concentration of isoflurane used in this study was 0.51 ± 0.09 mm, equivalent to 1.2 vol% at 22°C. The 5-min washout period was sufficient to remove all isoflurane from the chamber as confirmed by isoflurane measurements after the washout. HMR-1098 and 5-HD were kept as stock solution in double-distilled water. All stock solutions were diluted to required concentration in the superfusing buffer immediately before application.

Results are expressed as mean ± SD. Data were analyzed using Origin 7 software (OriginLab, Northampton, MA). Statistical analysis was performed using one-way analysis of variance with Scheffépost hoc  test. Differences were considered significant when the two-tailed P  value was less than 0.05.

Oxidative stress was used to mimic the reperfusion injury and investigate isoflurane-induced myocyte protection. In the time control group, 67 min of perfusion with glucose-free Tyrode solution had no significant effect on cell death (7 ± 5%, n = 5; fig. 2, TC group). During exposure to 200 μm H²O²and 100 μm FeSO⁴, some of the otherwise nonbeating cardiomyocytes started contracting, which resulted in cell hypercontracture and death in 47 ± 14% of cardiomyocytes (n = 11, stress group). When cardiomyocytes were pretreated with isoflurane before oxidative stress (APC group), cell death was markedly attenuated to 26 ± 8% (P < 0.05, n = 10). These results demonstrated that in vitro  pretreatment with isoflurane protects the cardiomyocytes from damage by oxidative stress.

To investigate the role of sarcK^ATPchannels in protection by isoflurane, the selective sarcK^ATPchannel inhibitor HMR-1098 was used. HMR-1098 (50 μm) applied throughout the experiment had no effect on myocyte death in the time control group (5 ± 3%, n = 6; fig. 3A, TC + HMR group). Application of HMR-1098 did not potentiate the deleterious effects of oxidative stress, and cell damage in this group was 48 ± 13% (n = 9; fig. 3A, stress + HMR group). However, inhibition of the sarcK^ATPchannel completely abolished the protective effects of isoflurane, and cell death after isoflurane pretreatment in the presence of HMR-1098 was 50 ± 21% (n = 9; APC + HMR group).

To determine the exact time period in which the sarcK^ATPchannel activation is protective, HMR-1098 was applied only during isoflurane pretreatment or only during stress. When sarcK^ATPchannel inhibitor was applied during isoflurane exposure, the cardioprotective effects of isoflurane were still present and cell damage by oxidative stress was lower than without isoflurane pretreatment (29 ± 13%, n = 7; fig. 3B, APC + HMR^ISOgroup). However, when HMR-1098 was applied only during oxidative stress, the protective effects of isoflurane were completely abolished and cell death was 50 ± 12% (n = 7; APC + HMR^stressgroup). These results show that activation of the sarcK^ATPchannel during the stress period is necessary for isoflurane-induced myocyte protection.

To assess contribution of the mitoK^ATPchannel to isoflurane-induced protection against oxidative stress, 5-HD (200 μm), a selective inhibitor of the mitoK^ATPchannels, was used. 5-HD was applied either throughout the experiment or during specific parts of the experiment as described previously. 5-HD did not affect cell damage in the time control group (7 ± 4%, n = 6; fig. 4A, TC + 5-HD group). In addition, inhibition of mitoK^ATPchannel had no effect on myocyte damage by oxidative stress, and cell death in the stress + 5-HD group was 44 ± 17% (n = 9; fig. 4A). However, when applied throughout the experiment, 5-HD completely abolished cellular protection by isoflurane, and cell death in APC + 5-HD group was 42 ± 23% (n = 8). When 5-HD was included during only the isoflurane exposure or only the stress period, cell death was 45 ± 23% and 46 ± 22% in the APC + 5-HD^APCand APC + HD^stressgroups, respectively (both n = 6; fig. 4B). These results indicate that activation of the mitoK^ATPchannel during both isoflurane pretreatment and oxidative stress plays a role in protection by isoflurane.

In this study, freshly isolated adult cardiomyocytes were used to investigate specific roles of sarcK^ATPand mitoK^ATPchannels in cardioprotective effects of APC. It was found that both sarcK^ATPand mitoK^ATPchannels are essential for the protection of myocytes from damage by oxidative stress because their inhibition completely abolished the protective effects of isoflurane preconditioning. Specifically, activation of the sarcK^ATPchannel was required for the cardioprotective effects of APC during the stress period but not during the preconditioning period. In contrast, activation of mitoK^ATPchannel was found to be necessary during both isoflurane preconditioning and during exposure to oxidative stress. From these results, it seems that although activation of both channels is important for the cardioprotective effects of preconditioning, each channel plays a distinct role: The sarcK^ATPchannel acts as an effector, whereas the mitoK^ATPchannel plays a dual role as a trigger and an effector.

In our study, we found that isoflurane protects isolated cardiomyocytes from oxidative stress via  both sarcK^ATPand mitoK^ATPchannel activation, because channel inhibition by HMR-1098 and 5-HD, respectively, abolished cardioprotective effects of APC. Similar results were reported by Toller et al .,14who demonstrated in a dog model in vivo  that infarct size reduction achieved by desflurane preconditioning is abolished in the presence of HMR-1098 and 5-HD. Further, using the same inhibitors, both channels were found to mediate ketamine-induced protection of the force of contraction of human right atrial trabeculae during hypoxia and reoxygenation.15In contrast to these studies that demonstrated an equally important role for both sarcK^ATPand mitoK^ATPchannels, there are studies that demonstrate a more significant role of the mitoK^ATPchannels. For example, Zaugg et al .16found that mitoK^ATPchannel inhibition by 5-HD, but not sarcK^ATPchannel inhibition by HMR-1098, completely abolished the cardioprotective effects of isoflurane and sevoflurane in isolated adult rat cardiomyocytes exposed to ischemia. Similarly, Uecker et al .17showed that 5-HD completely blocked the cardioprotection by ischemic or isoflurane-induced preconditioning in isolated perfused rat hearts, whereas HMR-1098 had no effect. MitoK^ATPactivation was found to be essential, whereas sarcK^ATPchannel was found to play no role or have only a partial role in desflurane- and sevoflurane-induced preconditioning of isolated human right atria, respectively.18,19 

Pronounced differences in findings from these studies that investigated the role of the sarcK^ATPchannel in APC may result from differences in the experimental models used (isolated cardiomyocytes, isolated hearts, atrial trabeculae, in vivo  preparation). There are also differences in the insults on the heart (ischemia and reperfusion in vivo , simulated ischemia without reperfusion, hypoxia, oxidative stress, metabolic inhibition). Moreover, the measured endpoints of cardiac injury also differ (cell death and infarcted area vs . functional parameters such as developed force of contraction). Finally, there are differences in the time and duration of application of pharmacologic inhibitors (application during preconditioning stimulus vs . stress period). In our preparation, the timing of application of HMR-1098, but not of 5-HD, was found to be crucial. If HMR-1098 was applied only during the preconditioning stimulus period (isoflurane exposure), cytoprotection was still present, but when HMR-1098 was applied during oxidative stress, cytoprotection was abolished. Therefore, in evaluating the studies that test the role of the sarcK^ATPchannel in preconditioning, it is important to consider the timing of application of the inhibitors. Further, the endpoint variable used to evaluate the myocardial injury is also an important factor. For example, the sarcK^ATPchannel was found not to have a role in the infarct size reduction by adenosine-enhanced ischemic preconditioning, but it was important for the improvement of functional recovery.23In the same study, the mitoK^ATPchannel had effect primarily on the infarct size without affecting the functional recovery. Another important condition is the type of insult. Both sarc and mitoK^ATPchannels were shown to have distinct roles in ischemia-reperfusion injury. Activation of the mitoK^ATPchannel mediates the phorbol 12-myristate 13-acetate-induced protection against cell death during chemically induced hypoxia, but not during reoxygenation.11In contrast, activation of the sarcK^ATPchannel was protective only during reoxygenation.11Therefore, it is possible that if only hypoxia is used as an insult,16involvement of the sarcK^ATPchannel in protection against ischemia-reperfusion injury can be overlooked. In addition, the effectiveness of the inhibitors can be altered during application of certain insults. For example, HMR-1098 was found to be ineffective in blocking the sarcK^ATPcurrents during metabolic inhibition by NaCN and iodoacetate.24This may also explain the often reported lack of HMR-1098 effects on the cardioprotection by ischemic and pharmacologic preconditioning.

One of the most remarkable characteristics of the preconditioning phenomenon is the memory phase, when cardioprotection persists despite withdrawal of the preconditioning stimulus (such as anesthetic in APC). In the complex mechanism of preconditioning, one key element of the preconditioning pathway is triggers, activated during preconditioning stimulus and responsible for initiating the downstream signaling cascade. Another key component is the effectors, which are at the end of the preconditioning signaling pathway and are directly responsible for cardioprotective effects of preconditioning during the memory phase. Some of the identified triggers are Gi-coupled receptors, such as adenosine receptors and a small burst of reactive oxygen species.10,25The downstream signaling cascade includes activation of intracellular kinases, with protein kinase C playing a central role. Kinase activation results in phosphorylation and priming of the effectors and may potentially result in establishing the cardioprotection memory.26Both sarcK^ATPand mitoK^ATPchannels have been implicated as triggers and effectors of preconditioning, but studies have yielded controversial results. In our study, application of HMR-1098 during the stress phase, but not during the preconditioning phase, completely abrogated cardioprotective effects of APC. This suggests that the sarcK^ATPchannel acts as an effector, but not as a trigger of APC, being activated and acting protectively only during the memory phase. These results support our previous studies, which demonstrated that opening of the sarcK^ATPchannel is potentiated during the memory phase of APC and that protein kinase C-δ mediates this effect.27In the current study, application of 5-HD during both the preconditioning phase and the stress phase reversed cardioprotective effects of APC to the same extent, indicating that mitoK^ATPchannel acts as both trigger and effector. The findings were similar to those of Fryer et al .,28who demonstrated that the mitoK^ATPchannel acts as both trigger and effector of preconditioning by ischemia, because 5-HD administration before or after the preconditioning phase completely abolished cardioprotection in the rat hearts. In contrast, Pain et al .29demonstrated that the mitoK^ATPchannel acts only as a trigger of ischemic preconditioning in isolated rabbit hearts. Similarly, the mitoK^ATPchannel opening was found to trigger isoflurane-induced preconditioning of rabbit hearts by generating reactive oxygen species.30Interestingly, the mitoK^ATPchannel was found to be an effector, but not a trigger, whereas the sarcK^ATPchannel was shown to act as a trigger of the delayed cardioprotection by opioid-induced preconditioning in rats.31Furthermore, the role of the sarcK^ATPchannel as a trigger was demonstrated in the delayed ischemic preconditioning of the rat heart.32However, these studies investigated the delayed phase of preconditioning 24 h after the preconditioning stimulus, whereas our study was focused on the early phase of preconditioning.

The role of the mitoK^ATPchannel as both a trigger and an effector and the role of the sarcK^ATPchannel as the effector of preconditioning could be explained by the following sequence of events. Exposure to isoflurane can directly activate mitoK^ATPchannels.33Opening of the mitoK^ATPchannel results in changes in the mitochondrial bioenergetics, which may result in a small burst of reactive oxygen species,29that can further affect mitochondrial bioenergetics and activate cytosolic mediators such as protein kinase C.34–37Activated protein kinase C translocates to the sarcolemma and can then phosphorylate the sarcK^ATPchannel and sensitize it to opening.27,38The primed sarcK^ATPchannel opens sooner and/or more during subsequent metabolic stress (ischemia-reperfusion, oxidative stress, metabolic inhibition, and others), resulting in greater K⁺efflux, a more rapid repolarization of the cell membrane, and action potential shortening with subsequent contractile failure, which may thereby decrease cytosolic Ca²⁺loading during ischemia-reperfusion.12,13,39A similar sequence of events occurs in quiescent cardiomyocytes, where sarcK^ATPchannel opening prevents diastolic depolarization of the cell membrane and decreases Ca²⁺loading.40This can reduce or prevent mitochondrial Ca²⁺overload, a major trigger of the cell death pathway.41The mitoK^ATPchannel can also open during ischemia-reperfusion, which may result in depolarization of the inner mitochondrial membrane and thus further decrease driving force for the mitochondrial Ca²⁺entry and loading during ischemia-reperfusion.42Evidence that pretreatment with the mitoK^ATPchannel opener diazoxide causes early activation of the sarcK^ATPchannel during metabolic inhibition and protects the cells from Ca²⁺loading via  indirect mechanism supports in part such a sequence of events.39From ours as well as other studies, it seems that both the sarcK^ATPand mitoK^ATPchannel are crucial components of cardioprotection that can interact and potentiate each other's protective effects. For example, Tanno et al .43showed that HMR-1098 can partially block the protection of isolated rabbit hearts from ischemia-reperfusion injury induced by diazoxide. In the same study, 5-HD completely abolished the cardioprotective effects of pretreatment with a low-dose pinacidil (10 μm) that is considered to open primarily sarcK^ATPchannels but not mitoK^ATPchannels.

It is important to acknowledge some limitations of our study. This study relies on specificity of pharmacologic openers and blockers of the sarcK^ATPand mitoK^ATPchannels. For example, both 5-HD and diazoxide were shown to have the mitoK^ATPchannel-independent effects in mitochondria,44and we cannot exclude the possibility that abolishment of APC by 5-HD in our study could have been due to the nonspecific effects. In addition, certain limitations are inherent to all in vitro  models, including the isolated myocyte model. Disaggregated myocytes are in an artificial environment that is different from that in the whole organ, and in vitro  conditions cannot replicate all of the complexity of the in vivo  setting. However, one advantage of our model, which contains only myocytes without the presence of other cell types, is that we are able to isolate the effects of drugs (isoflurane, the channel inhibitors) on the cardiomyocytes without vascular and neuronal influences. Also, although the use of oxidative stress does not strictly mimic the conditions during ischemia-reperfusion, both H²O²and OH· were shown to be a relevant component of ischemia-reperfusion injury in vivo . Therefore, keeping in mind the limitations of the isolated myocyte model, this model might still provide information relevant to the in vivo  setting.

From our results, we conclude that both the sarcK^ATPand the mitoK^ATPchannel play essential and distinct roles in APC in rat heart. The sarcK^ATPchannel acts as an effector of anesthetic preconditioning, whereas the mitoK^ATPchannel plays a dual role as a trigger and an effector.

The authors give special thanks to Martin Bienengraeber, Ph.D. (Assistant Professor, Departments of Anesthesiology and Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee, Wisconsin), for helpful discussions. The authors also thank Mary Ziebell (Research Technologist, Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin) for isoflurane measurements.