This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by O’Rourke, B. Search for Related Content PubMed PubMed Citation Articles by O’Rourke, B. Related Collections Apoptosis Energy metabolism Ischemic biology - basic studies Acute myocardial infarction Chronic ischemic heart disease

(Circulation Research. 2000;87:845.)
© 2000 American Heart Association, Inc.

Reviews

Myocardial K_ATP Channels in Preconditioning

Brian O’Rourke

From the Institute of Molecular Cardiobiology, Division of Cardiology, Department of Medicine, Johns Hopkins University, Baltimore, Md.

Correspondence to Brian O’Rourke, Johns Hopkins University, 720 Rutland Ave, 844 Ross Building, Baltimore, MD 21205. E-mail bor{at}jhmi.edu

	Abstract

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Abstract—We are on the brink of harnessing the cell’s natural defenses against ischemia and reperfusion injury after years of research into the destructive and protective mechanisms involved. Since the discovery of ischemic preconditioning, the surface receptors and signal transduction pathways underlying this phenomenon have been clarified, but many questions remain about the downstream targets that ultimately protect the cell. ATP-sensitive K⁺ (K_ATP) channels are thought to play a role in protection, but their mechanism of action has been unclear. Accumulating evidence now suggests that the location of the K_ATP channels relevant to cytoprotection may be on the mitochondrial inner membrane instead of on the sarcolemma of the cardiac cell. This review discusses recent findings and unanswered questions about the role of K_ATP channels in preconditioning and protection.

Key Words: K_ATP • protection • mitochondria • ischemia • apoptosis

	Introduction

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Efforts to mitigate or prevent ischemic injury have traditionally focused on finding ways to block events associated with irreversible ischemic injury. More recently, the discovery of the endogenous cellular protective mechanism known as ischemic preconditioning (IPC) has raised hopes that natural pathways or target proteins could be activated to help cells stave off commitment to necrosis or apoptosis. IPC in the heart refers to the finding that brief periods of ischemia confer protection against infarction produced by a subsequent long ischemia.^{1 2 3 4} Both acute protection, in which the brief ischemia must be applied within a 1- to 2-hour window before the long ischemia, and second window protection,^{5 6} occurring 24 to 48 hours after the preconditioning ischemia, have been described.

As summarized in Figure 1, a variety of intracellular signaling pathways have been implicated in the protective mechanism of IPC. These include the activation of G protein–linked phospholipase C–coupled receptors, tyrosine kinase pathways, protein kinase C (PKC), and the generation of reactive oxygen species. Protection can be blocked at several steps in each cascade; however, redundancy is built in, so protection can be preserved through the activation of alternative pathways. The identification of a common end-effector for protection has been elusive.

View larger version (59K): [in this window] [in a new window]   Figure 1. Figure 1. Mechanisms of preconditioning. Locally released adenosine, produced by the breakdown of adenine nucleotides during ischemia, or other agonists of G protein–coupled receptors lead to the activation of phospholipase C (or phospholipase D) and the generation of diacylglycerol (DAG), which activates and translocates PKC to target membranes. Phospholipase C also releases phosphatidyl inositol bisphosphate (PIP2), which is known to influence surfaceKATP channel activity. Stimulation of PKC-dependent or tyrosine kinase–dependent signaling pathways confers cardioprotection. Multiple pathways, including intracellular reactive oxygen species (ROS) generation, converge on PKC activation. MitoKATP channel opening may be a common downstream effector leading to protection but may also provide positive feedback by altering upstream components such as ROS or PKC. Tyr. kinase R. indicates tyrosine kinase receptor.

Among the candidates as a mediator of protection is the ATP-sensitive potassium channel (K_ATP). This channel is normally inhibited by intracellular ATP and opens during periods of energy depletion.^{7 8 9 10 11} K_ATP channels are found on a wide variety of tissues, and one of their most prominent functions is to modulate insulin release from pancreatic ß cells by setting the resting membrane potential (E_r) of the cell. An effect on E_r also underlies the mechanism of K_ATP action in vascular smooth muscle.¹² In the heart, K_ATP channels are present on the sarcolemma of cardiac myocytes, where they were first described,^{7 8} but their purpose remains unclear. The opening of surface K_ATP (surfaceK_ATP) channels in cardiomyocytes has little effect on E_r, because it is already close to the equilibrium potential for K⁺, but the outward current carried by K_ATP shortens the action potential and, if large enough, can render the cell inexcitable. Thus, it has been suggested that suppression of excitability spares energy by reducing that required for active ion cycling because of membrane depolarization and Ca²⁺ handling.¹³

Another K_ATP channel isoform, which is presumed to be ubiquitously distributed in all cells with mitochondria, is found on the mitochondrial inner membrane (mitoK_ATP). Recent evidence indicates that this channel may be both a trigger and effector of IPC. This review will compare the pharmacology of mitoK_ATP with surfaceK_ATP and discuss the evidence implicating mitoK_ATP in cellular protection.

	Role of KATP Channels in Acute IPC

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
IPC was first described in 1986 by Murry et al,¹ who showed that four 5-minute cycles of circumflex artery occlusion and reperfusion reduced infarct size produced by a long ischemia (40 minutes) from 30% to 7% of the area at risk. This acute protection is preserved for 1 to 2 hours after the preconditioning period. Although other endpoints, such as ventricular fibrillation, stunning, or functional recovery, are also influenced by preconditioning, infarct size limitation remains the gold standard for acute IPC studies. A leading mechanistic hypothesis, supported by findings that adenosine receptor antagonists abrogated and agonists mimicked protection, was that adenosine released locally from cells in the ischemic zone could activate receptors on cardiomyocytes in an autocoid manner, triggering a signaling cascade fortifying the cell against injury.² Protection can also be conferred by stimulating other phospholipase C–linked receptors, including bradykinin, endothelin, or acetylcholine. The activation of PKC was implicated in the response, because direct activation of the kinase could mimic IPC and inhibitors could eliminate protection induced by receptor stimulation.

Reports that the sulfonylurea receptor antagonists could diminish IPC-induced protection suggested that K_ATP channels may be effectors of protection, and this idea was reinforced by the finding that K_ATP channel openers like cromakalin, bimakalim, or pinacidil could mimic protection.^{3 4} It was also found that 5-hydroxydecanoate (5-HD), a putative ischemia-selective K_ATP channel inhibitor, could block protection.^{14 15 16 17}

On the basis of these studies, a mechanistic hypothesis was developed proposing that the opening of surfaceK_ATP channels during ischemia was somehow facilitated by the activation of IPC signaling pathways, such that the action potential shortening that occurs in the early phase of the long ischemia was enhanced. The result would be better preservation of cellular energy stores and suppression of deleterious downstream events, such as cellular Ca²⁺ overload.

This hypothesis was critically tested in several studies. Yao and Gross¹⁸ showed a lack of correlation between the extent of action potential shortening and the reduction of infarct size. Doses of bimakalim,¹⁸ cromakalim,¹⁹ or BMS-180448²⁰ that had little or no effect on the action potential could still significantly reduce infarct size. Furthermore, preventing ischemic action potential shortening by concomitant treatment with dofetilide did not eliminate protection.²¹ Finally, in simulated ischemia models of isolated cardiomyocytes, protection was conferred by K_ATP channel openers, even though the cells were quiescent and no action potentials were being generated.²² Thus, it was necessary to consider alternative explanations for the effects of the K_ATP channel openers and inhibitors in IPC.

	MitoKATP

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Although the mitochondrial inner membrane was traditionally thought to be relatively impermeable to ions, K-selective cation transport has been widely observed in mitochondria, as supported by studies using light scattering to measure mitochondrial swelling or fluorescent dyes to measure K⁺ influx.^{23 24 25 26} Suppression of K⁺ flux by ATP and a sensitivity to K⁺ channel openers and blockers led to the suggestion that a K_ATP channel similar to surfaceK_ATP existed on the inner mitochondrial membrane.^{23 26 27} This was strongly supported by patch-clamp studies of isolated mitoplasts, which revealed a channel with gating properties similar to K_ATP but with a conductance (10 pS in 100 mmol/L K⁺) smaller than the surface variety.²⁸ The mitoK_ATP channel was inhibited by 100 µmol/L ATP and blocked by glibenclamide or 4-aminopyridine. In reconstitution studies, mitoK_ATP was shown to be competitively inhibited by ATP, ADP, and palmitoyl- or oleyl-CoA, and relief from inhibition was mediated by GTP and GDP.^{23 25 29} Modulator effects require the presence of Mg²⁺. Although the excised patch studies indicated that the ATP inhibitory site was on the matrix face of the channel, other evidence suggests that it may be on the cytoplasmic (technically the intermembrane) face. This conclusion was based on the findings that the channel’s regulatory sites were all located on one side of the channel and that GTP and palmitoyl CoA are maximally active only when applied to the external medium of intact mitochondria and are presumably not transported into the matrix.³⁰

Studies of proteins reconstituted into proteoliposomes permitted a direct comparison of the pharmacology of mitoK_ATP and surfaceK_ATP channels. Garlid et al²⁴ showed that diazoxide was 1000 to 2000 times more potent in opening the mitoK_ATP channel compared with the sarcolemmal K_ATP channel (K_1/2 0.5 to 0.8 µmol/L for mitoK_ATP versus 840 µmol/L for cardiac sarcolemmal K_ATP). Furthermore, the effect of diazoxide could be blocked by 5-HD with a K_i of 45 to 85 µmol/L. Glibenclamide (K_i 1 to 6 µmol/L) and 5-HD block mitoK_ATP in a state-dependent manner, having no effect on channels activated by complete removal of ATP, but inhibiting channels activated by GTP or K⁺ channel openers in the presence of ATP.³¹

Exploiting the selectivity of diazoxide for the mitochondrial isoform, Garlid et al²⁴ compared the efficacy of diazoxide and cromakalim in protecting Langendorff-perfused rat hearts from ischemic contracture. Diazoxide significantly prolonged the time to ischemic contracture with a half-maximal effect at 8.8 µmol/L, whereas the K_1/2 for cromakalim was 11 µmol/L. Importantly, at equally protective doses of the 2 drugs, diazoxide produced markedly less action potential shortening. Both 5-HD (at 100 µmol/L) and glibenclamide abolished the protective effect of diazoxide. Garlid et al²⁴ concluded that the protective effect of K_ATP openers may be mediated by mitochondrial rather than sarcolemmal K_ATP channels.

Contemporaneously, Liu et al³² made a similar connection between mitoK_ATP activation and cardioprotection in isolated cardiomyocytes. By developing a method for monitoring the opening of mitoK_ATP using the native autofluorescence of mitochondrial flavoproteins, the opening of mitoK_ATP by diazoxide was detected for the first time in intact cells as a reversible increase in flavoprotein oxidation. At concentrations up to 100 µmol/L, diazoxide dose-dependently increased mitochondrial matrix oxidation (K_1/2 27 µmol/L) without activating sarcolemmal K_ATP currents, and 5-HD inhibited the redox effect. The latter finding provides important support for the argument that mitoK_ATP channels are involved, because there is substantial evidence that 5-HD is selective for mitoK_ATP over sarcolemmal K_ATP.^{15 24 33 34} The link between mitoK_ATP and protection in cardiomyocytes was demonstrated by examining the extent of cell killing in response to simulated ischemia using a cell pelleting method.³² Diazoxide decreased cell killing to about one half of that in controls, and this protection was blocked by 5-HD. The latter result was similar to that of Armstrong et al,²² who showed that 5-HD inhibited preconditioning induced by 1 to 3 short pelleting episodes before the long ischemia in the same quiescent cell model.

Using the flavoprotein fluorescence method, the pharmacological profile of several K_ATP openers and inhibitors have been characterized, and compounds have been found that specifically act on either the mitochondrial or sarcolemmal K_ATP channels (Table). In isolated cell models of ischemia, protection is afforded by compounds capable of activating mitoK_ATP, and this protection can be inhibited by compounds effective at blocking mitoK_ATP.^{35 36} Compounds selective for sarcolemmal channels have no such actions and serve as convincing evidence that sarcolemmal channels play much less of a role in protection. It should be noted that the specificities shown in the Table primarily apply to the case of intact cardiomyocytes or isolated mitochondria, usually determined with ATP present in the cytoplasmic solution, and may vary depending on the conditions of the experiment. For example, the pinacidil derivative P-1075 is quite selective for surfaceK_ATP in the isolated cells³⁵ but potently activates mitoK_ATP in mitochondrial preparations (P. Paucek, personal communication, June 2000) and confers protection in intact hearts with an EC₂₅ of 57 nmol/L.³⁷ The reason for this discrepancy is unknown.

View this table: [in this window] [in a new window]   Table 1. Selectivity of KATP Channel Openers and Inhibitors Toward Cardiac Mitochondrial KATP (MitoKATP) or Sarcolemmal KATP (SarcKATP)

Similarly, 5-HD was originally reported to block surfaceK_ATP currents activated by high ADP (1 mmol/L), low pH of 6.6,³⁸ or metabolic inhibition,³⁹ so conclusions about selectivity need to reexamined under many different conditions. Also, under conditions of high ADP or severe metabolic inhibition, surfaceK_ATP channels are more readily opened by diazoxide;⁴⁰ however, activation of heterologously expressed SUR2A/Kir6.2 channels by diazoxide is not inhibited by up to 500 µmol/L 5-HD.³⁴ Thus, extrapolation of these pharmacological specificities to other situations should be made with caution, particularly in intact hearts where many cell types are present, and diazoxide is a potent vasodilator through its action on surfaceK_ATP channels of vascular smooth muscle. Still, by using combinations of old and new selective agonists and antagonists, the hypothesis that mitoK_ATP is the effector of protection has received widespread support.

Acknowledging the caveat about pharmacological specificity mentioned above, there is remarkably good agreement between isolated cell data and results obtained in whole heart. Protection against infarction by diazoxide or low doses of cromakalim does not seem to be correlated with effects on coronary flow, action potential shortening, or ST segment elevation⁴¹ during ischemia. Almost universally, the protection is inhibited by 5-HD. Furthermore, HMR1883 (or its salt form HMR1098), an antagonist selective for surfaceK_ATP, has no effect on IPC^{41 42 43 44 45 46} but is effective in blocking reperfusion arrhythmias,^{43 47} presumably by reducing the dispersion of repolarization caused by the activation of surfaceK_ATP. 5-HD, on the other hand, does not influence ischemia-related electrical dispersion⁴⁸ or arrhythmias.⁴⁹ HMR1883 may also be more selective toward cardiomyocyte, rather than vascular, surfaceK_ATP channels, because reperfusion hyperemia was blocked by glibenclamide but not HMR1883 in rat hearts.⁴³

A dichotomous recent result suggested that HMR1883 could block diazoxide, but not conventional IPC, in rabbit hearts.⁴¹ Interestingly, diminution of ischemic ST segment elevation with successive occlusions was unaffected by doses of 5-HD known to block IPC, but abolished by HMR1883. Thus, this study confirmed the dissociation between surfaceK_ATP activation and protection while at the same time supporting a surfaceK_ATP role in diazoxide-induced protection (either that or the HMR1883 was not completely surface selective). These results will need to be reconciled with a report showing no effect of HMR1098 on diazoxide-induced protection in isolated myocytes.³⁵

Assuming that the primary action of 5-HD is on mitoK_ATP, it is quite remarkable that this compound blocks protection induced by an endless variety of preconditioning protocols, including protection induced by IPC,^{14 50 51} adenosine,^{52 53 54 55 56} endothelin,⁵⁷ nicorandil,^{36 58} opioids,⁵⁹ acetylcholine,⁶⁰ diazoxide,^{32 44 61 62 63 64} heat shock,^{65 66 67} Ca²⁺ preconditioning,⁶² HpETE,⁶⁸ renal ischemia,⁶⁹ monophosphoryl lipid A,^{70 71} phorbol myristic acid,⁷² BMS-180448,⁷³ RP52891,¹⁶ and volatile anesthetics.⁷⁴

Although perfect isoform selectivity of K_ATP openers and blockers has yet to be achieved, all of the pharmacological and electrophysiological results taken together favor mitoK_ATP rather than surfaceK_ATP as the relevant effector of preconditioning. This conclusion will be bolstered by the future development of highly isoform-specific and potent drugs. In this regard, a new K⁺ channel opener, BMS-191095,⁷⁵ has recently been described that is extremely potent (K_D 83 nmol/L) and selective for mitoK_ATP, has no effect on ischemic action potential shortening, will not open vascular or sarcolemmal K_ATP channels, and is cardioprotective.⁷⁶

Another important question is the time frame for mitoK_ATP channel opening necessary for protection. In a recent study in a rat model of IPC by Fryer et al,⁴⁴ 5-HD was applied either before or after the preconditioning ischemia or diazoxide application to examine whether mitoK_ATP channel opening exerted its protective effect during the trigger or long ischemia phases or both. In controls, IPC reduced infarct size (normalized to area at risk) from 56% to 7%. When 5-HD was applied 5 minutes before IPC, protection was largely eliminated (infarct size was 40%), whereas in the absence of preconditioning, 5-HD had no significant effect on infarct size. Diazoxide given 15 minutes before the long ischemia reduced infarct size (to 36%), whereas 5-HD applied either before or after diazoxide exposure completely eliminated this protection. This suggests that mitoK_ATP opening is important as both a preconditioning trigger and effector of protection during the long ischemia (discussed in the context of delayed preconditioning in the next section). Also noted in this study was a diminished effect of 5-HD with longer pretreatment periods, suggesting that 5-HD was being actively metabolized. This result highlights the difficulty of verifying the local concentration of a pharmacological agent at an effector site in vivo, particularly when the site is an intracellular one like the mitochondria and coronary flow is changing during the experiment.

A recent study by Pain et al⁷⁷ supported the role of mitoK_ATP as a trigger of IPC but challenged its role as an effector during the long ischemia in perfused rabbit hearts. They found that 5-HD (or glibenclamide) administered during a 5-minute preconditioning ischemia or a short diazoxide pretreatment could block the infarct limiting effect of IPC but was ineffective when given after the preconditioning period. Tyrosine kinase inhibition abrogated diazoxide-induced protection, whereas PKC inhibition did not. In addition, free radical scavengers applied during diazoxide pretreatment were capable of blocking protection, suggesting that mitoK_ATP opening is an upstream trigger of radical production and tyrosine kinase activation. As discussed by Gross and Fryer,⁷⁸ these findings will need to be reconciled with several earlier studies demonstrating an effect of 5-HD applied after the preconditioning period.

Although preconditioning in humans is more difficult to verify, several studies have noted a significant effect of K_ATP channel blockers or openers either in vivo or in explanted human tissues.^{45 56 79 80 81 82 83} It is reassuring to find that the pharmacological profile described in animal models has been replicated in human myocardium. As reported recently by Ghosh et al,⁴⁵ hypoxic preconditioning (assessed by prevention of creatine kinase loss) of human atrial tissue could be blocked by previous treatment with glibenclamide or 5-HD, but not HMR1883, and protection was mimicked by diazoxide.

	MitoKATP in Delayed IPC

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
In addition to acute IPC, which lasts for 1 to 2 hours, a second window of protection or delayed preconditioning phase reappearing 24 to 72 hours after the brief ischemic period has been described.^{5 6} Protection by delayed preconditioning is usually less effective than acute IPC and requires more cycles of brief ischemia for maximal activation. The activation of tyrosine kinases and PKC, the expression of heat shock protein 72 and inducible NOS, and the release of nitric oxide have been shown to be involved in delayed preconditioning.

Recent findings indicate that mitoK_ATP channel opening may be a component of delayed protection. In a rabbit model of delayed preconditioning, protection at 24 hours (conferred by either four 5- or 10-minute ischemia/reperfusion cycles⁸⁴ or by adenosine treatment⁸⁵ ) could be eliminated by treatment with glibenclamide or 5-HD given just before the long ischemia. As is consistently found in acute IPC, these blockers did not worsen the extent of injury in the absence of preconditioning, indicating that K_ATP channels are probably involved in the IPC recruitable protection but do not limit infarct size under normal circumstances.

It is noteworthy that the selectivity of 5-HD for mitoK_ATP was disputed in the report described above⁸⁴ and in others^{85 86 87} on the basis of the observation that action potential shortening was significantly diminished by 5-HD. Similarly, early studies of 5-HD showed inhibition of whole-cell and single-channel K_ATP currents by 5-HD.^{38 39 88} Although a direct effect of 5-HD on sarcolemmal K_ATP may be possible under some conditions, a result demonstrating an effect of 5-HD on action potential shortening or surfaceK_ATP currents does not necessarily imply that the drug is acting on the surface channel. It is possible, and perhaps to be expected, that a drug acting on mitoK_ATP, by influencing cellular energy metabolism, could have secondary effects on surfaceK_ATP. Conversely, agents that are intended to alter subsarcolemmal energy balance, through tight coupling between sarcolemmal and mitochondrial membranes, could potentially influence mitoK_ATP activity as well. This could provide an alternative explanation to the conclusions of Haruna et al,⁸⁹ who recently showed that inhibition of Na⁺/K⁺ ATPase with digoxin could inhibit the activation of surfaceK_ATP and abrogate preconditioning but could not inhibit K_ATP channel opener-mediated protection.

These results were interpreted as evidence that selective surfaceK_ATP channel inhibition could block protection. Even taking into account possible cross-reactivity of 5-HD with surfaceK_ATP, its mitochondrial site of action is likely to be the relevant one for protection, because 5-HD can block protection without an effect on ischemic action potential shortening,^{48 90} and when 5-HD does have an electrophysiological effect, there is still no correlation between the action potential shortening and protection.⁸⁴ These findings have also been confirmed in a cellular model of IPC. Delayed protection (elicited by brief ischemia or adenosine) at 24 hours can be blocked by 5-HD given just before the long ischemia in a human cardiac cell line.⁵⁶

Also in support of mitoK_ATP as an effector of delayed preconditioning is that a single dose of diazoxide, applied 24 hours before a long ischemia, is enough to afford significant protection in rats.⁹¹ This protection can be blocked by 5-HD, applied either before diazoxide exposure or 10 minutes before the long ischemia, again supporting the hypothesis that mitoK_ATP channels are involved in both the trigger phase and in protection during the lethal ischemia. In another study in rabbits, diazoxide mimicked both early and delayed IPC, and 5-HD applied before the long ischemia blocked protection, as did L-NAME, leading to the conclusion that both mitoK_ATP and nitric oxide (NO) were mediators of protection induced by diazoxide.⁶³

The NO pathway and mitoK_ATP activation have been linked in a recent study by Sasaki et al,⁹² who showed that NO donors facilitate the activation of mitoK_ATP opening by diazoxide and partially activate the channel directly. A similar facilitation of mitoK_ATP opening by PKC activation was earlier noted by Sato et al.³³ Thus, modulation of mitoK_ATP could play a role in both the short- and long-term memory of IPC. Little is known about how signal transduction pathways modify mitochondrial targets, but a recent study by Wang et al⁹³ reported that one isoform of PKC (PKC-) is specifically translocated to the mitochondria after diazoxide treatment. How this finding fits in with the known participation of other PKC isoforms² in protection has not been determined.

	MitoKATP and Apoptosis

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Several studies have shown that apoptosis contributes to cardiac cell death after ischemia, and emerging evidence indicates that preconditioning may suppress apoptosis in addition to necrosis in intact hearts.^{94 95 96 97 98} Regarding the role of mitoK_ATP in protecting against apoptosis, several recent studies show that diazoxide pretreatment decreases the appearance of apoptotic markers resulting from ischemia/reperfusion injury.⁹¹ The protection against apoptosis was also blocked by PKC inhibitors or 5-HD, reminiscent of the general mechanism of protection already discussed, and by the same reasoning implicating mitoK_ATP in the inhibition of apoptosis.

A recent study by Nakamura et al⁹⁹ showed that IPC blunted the upregulation of Bax protein expression associated with ischemia/reperfusion without altering the levels of Bcl-2. The role of Bax in inducing or Bcl-2 in preventing cytochrome c release from the mitochondrial intermembrane space begs the question of whether mitoK_ATP activation influences this early trigger of apoptosis. The only published evidence thus far paradoxically suggests that cytochrome c release may be induced by mitoK_ATP opening. Holmuhamedov and colleagues^{100 101} reported that in Ca-loaded mitochondria, diazoxide treatment caused mitochondrial swelling, Ca²⁺ release, and loss of cytochrome c. However, recent findings in cultured neonatal rat myocytes suggest the opposite. Apoptosis induced by H₂O₂ treatment [assessed by cytochrome c translocation, caspase activation, mitochondrial membrane potential loss, and poly(ADP-ribose)polymerase cleavage] can be inhibited by diazoxide, and 5-HD blocks this protection.¹⁰²

	Mechanisms of Protection

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Now that substantial evidence has accumulated in support of mitoK_ATP as a trigger and late effector of cardioprotection, attention is focussing on how the opening of a K⁺ influx pathway on the mitochondrial inner membrane may be protective. At face value, it may seem paradoxical that the opening of an energy dissipating cation conductance on the mitochondrial inner membrane would be beneficial. Because tight coupling between proton pumping and ATP production requires a relatively impermeable inner membrane, mitoK_ATP opening must lead to changes in mitochondrial function that supersede the loss of energy attributable to uncoupling. A least 3 main mechanistic hypotheses have been proposed to explain the protective effect of mitoK_ATP channel opening (Figure 2). These hypotheses are not mutually exclusive and may all contribute to protection, but it is still undetermined whether they play a role in either acute or delayed preconditioning in vivo.

View larger version (37K): [in this window] [in a new window]   Figure 2. Figure 2. Putative mechanisms of protection mediated by mitoKATP channel opening. A, Mitochondrial volume, determined by the balance between salt influx and efflux from the matrix, may be adjusted to optimize energy production (or perhaps minimize energy loss) during ischemia and reperfusion.25 105 B, Mitochondrial Ca2+ overload during ischemia or reperfusion may be slowed by depolarization of the mitochondrial membrane potential (m), and Ca2+ release may be initiated by permeability transition pore opening.101 C, ROS production by the mitochondria may be enhanced during early ischemia to trigger protection but inhibited during reperfusion to mitigate damage106 113 (see text for details).

	Mitochondrial Swelling and Optimization of Respiration

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
A well-described consequence of mitoK_ATP opening in studies of isolated mitochondria is matrix swelling. According to the model of Garlid et al,²⁵ the electrophoretic uptake of K⁺, in concert with the operation of a K⁺/H⁺ antiporter, are the components of a volume regulatory K⁺ cycle. K⁺ influx is accompanied by the movement of diffusible weak acids to maintain electroneutrality and water movement attributable to osmotic forces. The net result is matrix swelling. Because the changes in matrix K⁺ and H⁺ concentrations are initially insufficient to directly activate K⁺/H⁺ exchange, some initial swelling is required before indirect activation of the antiporter (by Mg²⁺ or H⁺) is invoked, and a higher steady-state matrix volume is maintained until the K⁺ influx ceases. In normally respiring mitochondria, the K⁺ fluxes are small relative to the rate of proton pumping, so mitochondrial membrane potential and pH is largely maintained.

Matrix swelling by itself may improve the rate of oxidative metabolism, as previously suggested by Halestrap et al,¹⁰³ who showed that swelling activated fatty acid oxidation, respiration, and ATP production. If swelling plays a role in improving mitochondrial function, it is worthwhile to consider other factors that may be involved in preventing excess swelling and loss of function. In an earlier model of mitochondrial volume homeostasis described by Garlid and Beavis,^{104 105} respiration-driven cation influx through mitoK_ATP might also be accompanied by anion efflux through an inner membrane anion channel, or IMAC, in coordination with the operation of the K⁺/H⁺ antiporter. IMAC was therefore suggested to be a safety valve that prevents excessive matrix swelling. In light of recent findings that chloride channel inhibitors may play a role in cardioprotection,^{106 107} it is intriguing to speculate that a balance between K⁺ influx and anion efflux may tune the mitochondrion to the optimal volume for preserving function during ischemia and reperfusion.

On the basis of thermodynamic considerations, it has also been theorized that optimal efficiency of oxidative phosphorylation is achieved when mitochondria are partially uncoupled,¹⁰⁸ and this idea has been put forward as the purpose of mitochondrial uncoupling proteins.¹⁰⁹ By analogy, optimization of respiration may also be the result of the partial uncoupling induced by mitoK_ATP opening.

Similarly, Fryer et al⁴⁴ showed that mitochondria isolated from the area at risk of preconditioned hearts had higher rates of ATP synthesis than those of hearts subjected to long ischemia alone and that this preservation of mitochondrial function could be partially inhibited by 5-HD, supporting improved ATP production as a common feature of IPC and K_ATP channel openers.

	Mitochondrial Ca2+ Handling

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
A decrease in the extent of mitochondrial Ca²⁺ overload may be a consequence of opening mitoK_ATP. Holmuhamedov et al¹⁰¹ demonstrated that the rate of Ca²⁺ uptake by isolated mitochondria in suspension (with 150 µmol/L Ca²⁺ in the bath) is dose-dependently suppressed by diazoxide or pinacidil. Both compounds also activated Ca²⁺ release from preloaded mitochondria through a mechanism that was apparently separate from the mitoK_ATP channel. The effect on Ca²⁺ release (but not uptake) was cyclosporin A–sensitive and accompanied by cytochrome c release,¹⁰⁰ suggesting that the permeability transition pore may be involved. Although these investigators were unable to obtain consistent block of either response by 5-HD or glybenclamide in the isolated organelles, mitochondrial Ca²⁺ concentration was reduced by diazoxide in intact neonatal myocytes, and 5-HD inhibited this effect.

It was suggested that the mechanism of decreased Ca²⁺ uptake by mitoK_ATP opening was attributable to a decreased driving force for Ca²⁺ entry. From a baseline resting potential of -195 mV in the isolated mitochondria, the K_ATP openers depolarized by 15 to 20 mV, and this effect was not inhibited by cyclosporin A.¹⁰¹

These interesting results raise several questions that will require additional investigation. Why was the K_1/2 for inhibition of Ca²⁺ uptake 65 µmol/L for diazoxide, 100 times higher than the K_1/2 for K⁺ uptake via mitoK_ATP as reported by Garlid et al²⁴ ? What is the nature of the Ca²⁺ release mechanism? Why is there a permeability transition and cytochrome c release if diazoxide protects against apoptosis?

	Free Radicals and Redox State

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Reactive oxygen species (ROS) have long been implicated in the cellular damage associated with ischemia and reperfusion.^{110 111} On the other hand, ROS generation is thought to be a trigger of signaling pathways mediating preconditioning.^{77 112} Mitochondria not only set the redox balance of the cell, but also are a source of ROS production because of electron leakage from the electron transport chain. Recent work has suggested that the opening of mitoK_ATP channels may alter the rate of mitochondrial ROS production and contribute to cardioprotection. In models of hypoxic preconditioning in embryonic chick myocytes, IPC-, adenosine-, or acetylcholine-mediated protection was associated with an early increase in ROS production during the preconditioning period, as determined by the accumulation of 2',7'-dichlorofluorescein produced by oxidation of its precursor.^{60 106 113} Protection and ROS production were inhibited by 5-HD, the thiol reductant 2-mercaptoproionyl glycine, or the mitochondrial site III inhibitor myxothiazol, indicating that mitochondria were the source of ROS production and that the opening of mitoK_ATP channels could stimulate ROS accumulation. Vanden Hoek et al¹⁰⁶ showed that the chloride channel blocker diisothiocyanato-stilbene-2,2'-disulfonate abrogated protection, and they proposed a model in which superoxide generated in the mitochondria is exported through anion channels to the cytoplasm, where it is dismutated to hydrogen peroxide.

Interestingly, the generation of ROS during reperfusion, presumed to contribute to irreversible cellular injury, was attenuated by adenosine preconditioning or pinacidil applied just at the time of reperfusion. This protection was eliminated in the presence of 5-HD or an inhibitor of PKC.¹¹³ Thus, mitoK_ATP opening could either enhance or attenuate mitochondrial ROS production, depending on the phase of preconditioning, ischemia, or reperfusion. Using a ROS-sensing microprobe, Obata et al¹¹⁴ also demonstrated that cromakalim or nicorandil increased hydroxyl radical production in rat myocardium. The effect was inhibited by 5-HD or glibenclamide. The link between mitoK_ATP opening, ROS generation, and PKC remains incompletely defined at present. MitoK_ATP opening seems to increase ROS production, which may in turn activate PKC, but the mitoK_ATP channel is also modulated by these factors.^{33 92} In a recent study by Wang et al,⁹³ diazoxide induced PKC translocation and protection in Langendorff-perfused rat hearts and these responses could be blocked by PKC inhibitors. In contrast, Miura et al¹¹⁵ reported that the PKC inhibitor calphostin C could block adenosine-, but not diazoxide-mediated cardioprotection.

	Other Mechanisms

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Other effects of K_ATP channel openers have been reported, which may also contribute to protection in the intact heart. Oe et al¹¹⁶ reported that diazoxide or cromakalim attenuated resting and stimulated norepinephrine release in perfused guinea pig hearts, but not in human right atrium, with the effect being antagonized by glibenclamide. Glibenclamide by itself increased NE release in both preparations. Sakamoto et al¹¹⁷ showed that 5-HD, like glibenclamide, blunts the early, but not late myocardial K⁺ efflux from ischemic guinea pig hearts, indicating potential crosstalk between the mitochondrial and sarcolemmal isoforms of the K_ATP channel during ischemia (assuming mitochondrial selectivity of 5-HD).

Another possible mechanism involving mitoK_ATP and the actin cytoskeleton was suggested by Baines et al,¹¹⁸ who found that cytochalasin D, a disrupter of the cytoskeleton, could eliminate protection conferred by diazoxide, pinacidil, or IPC in isolated adult cardiomyocytes subjected to simulated pelleting ischemia. Furthermore, anisomysin, a p38/JNK activator, decreased osmotic fragility in a 5-HD–sensitive manner.

	KATP Isoforms in Heart

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
Elucidation of the molecular structure of mitoK_ATP is thus far limited by the lack of an identified channel clone. Present working models of surfaceK_ATP channels suggest that they are composed of a tetramer of core inward rectifier K⁺ channels (Kir6.x) surrounded by 4 sulfonylurea receptor subunits (SUR), which confer sensitivity not only to glibenclamide but also to K_ATP channel openers.^{9 10} There are 3 SUR isoforms encoded by 2 genes: SUR1 and the splice variants SUR2A and SUR2B. The predominate pancreatic isoform is thought to be SUR1 paired with Kir6.2, whereas the myocyte sarcolemmal isoform has been identified as SUR2A/Kir6.2.^{119 120} In smooth muscle, SUR2B/Kir6.1 is thought to be the small-conductance, diazoxide-sensitive, ATP-insensitive isoform observed in patch-clamp recordings.¹²¹ Similarities between mitoK_ATP and surfaceK_ATP have led to the assumption that mitoK_ATP will also consist of SUR and Kir components. This is reinforced by the tentative identification of a 63 kDa sulfonylurea binding protein purified from mitochondria and a putative pore forming channel subunit of 55 kDa reported by Grover and Garlid.⁴

A study by Susuki et al¹²² suggested that antibodies raised against a 12 amino acid stretch of Kir6.1 immunolocalize to mitochondria, implying that this isoform may be present on the inner membrane. In a recent test of whether Kir6.1 is an essential component of mitoK_ATP, dominant-negative adenoviruses, in which the pore signature sequence GFG was mutated to AFA in Kir6.1 and Kir6.2, were constructed to knock out current carried by SUR/Kir6.x channels.¹²³ The AFA mutation eliminates the ability of the channel to conduct K⁺ but does not prevent it from coassembling with native K_ATP channels. These constructs selectively suppressed surface currents carried by their wild-type counterparts when coinfected into A549 cells.¹²³ The 6.2AFA virus, but not the 6.1AFA virus, could also suppress native sarcolemmal K_ATP current in isolated rabbit cardiomyocytes. Notably, infection of adult myocytes with the dominant-negative Kir6.1AFA subunit had no effect on the mitochondrial redox response to diazoxide, indicating that it is unlikely that Kir6.1 is part of mitoK_ATP.¹²⁴ However, this does not preclude the possibility that a Kir6.1-like subunit, with antigenic similarity to 6.1 but without the ability to heteromultimerize with its surface counterpart, could still contribute to the mitoK_ATP channel structure. In this regard, recent work by Liu et al¹²⁵ indicates that among the combinations of known SUR and Kir subunits expressed heterologously as surface channels, only SUR1/Kir6.1 fits the pharmacological profile for mitoK_ATP (as defined in intact rabbit myocytes; see the Table).

Isoform-specific localization of the binding sites for K_ATP channel openers and inhibitors may provide some insight into the structure of mitoK_ATP. To fit the pharmacological profile of mitoK_ATP, one would look for a combination of SUR/Kir subunits that is readily opened by diazoxide, nicorandil, cromakalin, and pinacidil (and, perhaps, P1075) and is sensitive to inhibition by glibenclamide and 5-HD. In this context, recent evidence using chimeras of SUR1 (the pancreatic isoform potently activated by diazoxide) and SUR2A (insensitive to diazoxide), both coexpressed with Kir6.2, indicates that diazoxide sensitivity may be conferred by a region comprised of SUR1 transmembrane domains 6 to 11 interacting with its first nucleotide binding fold.¹²⁶ Cromakalin and pinacidil, which readily open SUR2A/Kir6.2 channels, interact with a region spanning transmembrane domains 12 to 17 of SUR2A,¹²⁶ and the inhibitory binding site of sulfonylureas is also in this area.¹²⁷ The critical residues for pinacidil binding were additionally narrowed by Uhde et al,¹²⁸ who showed that P1075 binding involved regions including amino acids 1059 to 1087 and 1218 to 1320 of SUR2.

The relatively small 42 amino acid carboxy-terminal divergence between the splice variants SUR2A and SUR2B¹²⁹ contributes to major differences in K_ATP opener sensitivity. Not only is SUR2B sensitive to diazoxide, but nicorandil is more than 100 times more potent at opening this isoform as compared with SUR2A, whereas both are equally sensitive to pinacidil.¹³⁰

Regarding the effects of different Kirs on the K_ATP pharmacology, Russ et al¹³¹ reported that glibenclamide binding to SUR2B is enhanced by coexpression of Kir6.1, although the K_D for binding was about 7 times lower than the K_i for channel inhibition. Koster and colleagues^{132 133} showed that mutations of Kir6.2 altered the nucleotide sensitivity of the channel and activation by phosphatidyl inositol 4,5-bisphosphate. The N-terminus of Kir has been implicated in the coupling of sulfonylurea-bound SUR to the stabilization of the channel-closed state.¹²⁷ Conductance and gating of the channel is also conferred by the Kir subunit. In a study by Kondo et al¹³⁴ using chimeras of Kir6.1 and Kir6.2, the extracellular linker domain between the membrane-spanning regions determined conductance (80 pS in Kir6.2 and 35 pS in Kir6.1 when coexpressed with SUR2A), whereas the N- and C-terminal regions were implicated in spontaneous opening in the absence of intracellular ATP (not present in Kir6.1).

Reconstruction of the known properties of mitoK_ATP from structure-function studies of surface channels may provide clues toward the molecular structure of mitoK_ATP but will require additional elucidation of the isoform-specific differences in 5-HD sensitivity, conductance, and regulation.

	Summary

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 
The hypothesis that the mitoK_ATP channel plays a major role in both acute and delayed preconditioning has been well supported by an accumulating body of evidence. Questions remain regarding the selectivity of some of the pharmacological tools used, and it is unwise to rely on only a single compound to justify a particular target, especially when intracellular conditions are changing markedly during ischemia. Still, when the results of many studies are taken together, protection is poorly correlated with drug action on surfaceK_ATP channels and well correlated with effects on mitoK_ATP channels. These questions will undoubtedly be resolved with a new generation of highly isoform-specific compounds.

If the mitoK_ATP hypothesis holds up to future scrutiny, the question remains as to what the role of surfaceK_ATP is in the cardioprotection. Because mitoK_ATP seems to mediate the extra protection associated with preconditioning but does not seem to affect injury in the absence of preconditioning, it is possible that surfaceK_ATP provides a baseline level of resistance to damage. Because surfaceK_ATP channel opening may also be arrhythmogenic on reperfusion, defining its true role in ischemia and reperfusion is critical.

The mechanism of mitoK_ATP protection may involve alterations in mitochondrial Ca²⁺ handling, the optimization of energy production, and modulation of ROS production during ischemia or reperfusion. It is not known which of these factors is important in vivo. Also to be resolved is the relationship between mitoK_ATP opening and protein kinase action. Recent findings indicate that mitoK_ATP channels are modulated by PKC and NO but may also trigger the translocation and activation of PKC or activate tyrosine kinases.

The molecular cloning of mitoK_ATP has not yet been achieved and has been hampered by uncertainty about the molecular determinants of intracellular targeting of transmembrane proteins. Until it is cloned, it is only possible to draw structural inferences about mitoK_ATP from known properties of surfaceK_ATP.

Despite many unanswered questions, recent experimental findings and drug developments support the feasibility of specifically recruiting the natural defenses of the cell and will likely lead to new therapeutic strategies in the near future.

	Footnotes

 
This manuscript was sent to Eugene Braunwald, Consulting Editor, for review by expert referees, editorial decision, and final disposition.

Received August 4, 2000; revision received September 29, 2000; accepted September 29, 2000.

	References

Top Abstract Introduction Role of KATP Channels... MitoKATP MitoKATP in Delayed IPC MitoKATP and Apoptosis Mechanisms of Protection Mitochondrial Swelling and... Mitochondrial Ca2+ Handling Free Radicals and Redox... Other Mechanisms KATP Isoforms in Heart Summary References

 

1. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;74:1124–1136.[Abstract/Free Full Text]
2. Cohen MV, Baines CP, Downey JM. Ischemic preconditioning: from adenosine receptor of K_ATP channel. Annu Rev Physiol. 2000;62:79–109.[Medline] [Order article via Infotrieve]
3. Gross GJ, Fryer RM. Sarcolemmal versus mitochondrial ATP-sensitive K⁺ channels and myocardial preconditioning. Circ Res. 1999;84:973–979.[Abstract/Free Full Text]
4. Grover GJ, Garlid KD. ATP-sensitive potassium channels: a review of their cardioprotective pharmacology. J Mol Cell Cardiol.;2000;32:677–695.
5. Marber MS, Latchman DS, Walker JM, Yellon DM. Cardiac stress protein elevation 24 hours after brief ischemia or heat stress is associated with resistance to myocardial infarction. Circulation. 1993;88:1264–1272.[Abstract/Free Full Text]
6. Kuzuya T, Hoshida S, Yamashita N, Fuji H, Oe H, Hori M, Kamada T, Tada M. Delayed effects of sublethal ischemia on the acquisition of tolerance to ischemia. Circ Res. 1993;72:1293–1299.[Abstract/Free Full Text]
7. Noma A. ATP-regulated K⁺ channels in cardiac muscle. Nature. 1983;305:147–148.[Medline] [Order article via Infotrieve]
8. Trube G, Hescheler J. Inward-rectifying channels in isolated patches of the heart cell membrane: ATP-dependence and comparison with cell-attached patches. Pflügers Arch. 1984;401:178–184.
9. Babenko AP, Aguilar-Bryan L, Bryan J. A view of sur/KIR6.X, K_ATP channels. Annu Rev Physiol. 1998;60:667–687.[Medline] [Order article via Infotrieve]
10. Seino S. ATP-sensitive potassium channels: a model of heteromultimeric potassium channel/receptor assemblies. Annu Rev Physiol. 1999;61:337–362.[Medline] [Order article via Infotrieve]
11. Aguilar-Bryan L, Clement JP, Gonzalez G, Kunjilwar K, Babenko A, Bryan J. Toward understanding the assembly and structure of K_ATP channels. Physiol Rev. 1998;78:227–245.[Abstract/Free Full Text]
12. Quayle JM, Nelson MT, Standen NB. ATP-sensitive and inwardly rectifying potassium channels in smooth muscle. Physiol Rev. 1997;77:1165–1232.[Abstract/Free Full Text]
13. 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.[Abstract/Free Full Text]
14. Grover GJ, Gross GJ. Blockade of ischaemic preconditioning in dogs by the novel ATP dependent potassium channel antagonist sodium 5-hydroxydecanoate. Cardiovasc Res. 1992;26:1054–1062.[Abstract/Free Full Text]
15. McCullough JR, Normandin DE, Conder ML, Sleph PG, Dzwonczyk S, Grover GJ. Specific block of the anti-ischemic actions of cromakalim by sodium 5-hydroxydecanoate. Circ Res. 1991;69:949–958.[Abstract/Free&nbs