FOCUSSED ISSUE: K ATP CHANNELS AND CARDIOPROTECTION Basic Res Cardiol 95: 272 – 274 (2000) © Steinkopff Verlag 2000 James M. Downey Michael V. Cohen Do mitochondrial K ATP channels serve as triggers rather than end-effectors of ischemic preconditioning’s protection? should be so protective. In a sense the answer that was long sought after actually resulted in more questions than it addressed. We also became interested in mK ATP ’s effect in precondi- tioning. Indeed, using infarct size as the end-point we found that diazoxide mimicked preconditioning and that 5HD blocked preconditioning’s anti-infarct effect (4). However in subsequent studies we began to see behavior that suggested mK ATP might not be the end-effector. In a recently published abstract (14), we reported that a 5-min exposure to diazoxide in lieu of brief preconditioning ischemia continued to protect the heart even if drug administration was followed by up to 30 min of washout before the index ischemia. Yet data from Liu et al. (11) have demonstrated that mK ATP close quickly after diazoxide is washed out. Thus, the presence or absence of a protected state did not seem to hinge on whether the chan- nel was open or not. Additionally, the tyrosine kinase inhibitor genistein completely eliminated diazoxide’s anti-infarct effect in the rabbit. Ashraf’s group found that chelerythrine, a PKC blocker, would also block diazoxide’s protective effect in rat hearts (16). If protection were dependent on kinases down- stream of the mK ATP , then the channel could not be an end- effector. Finally we noted that bracketing the diazoxide pulse with an infusion of 5HD (starting 5 min prior to and continu- ing until 5 min after diazoxide administration) blocked pro- tection. If, however, 5HD were administered after diazoxide was washed out, starting 5 min prior to the long ischemic period, protection was not affected. Similar behavior was observed with preconditioning: 5HD infused at the time of the preconditioning ischemia aborted protection, whereas it was ineffective if present only during the index ischemia. There- fore, the mK ATP had to open prior to ischemia in order to pro- tect, and it seemed irrelevant whether they were open during the index ischemia. The mK ATP channels have also been implicated in the second window of protection. Infusion of opioid agonist in rats induces a delayed protection 48 h later and that protection can J. M. Downey () · M. V. Cohen University of South Alabama College of Medicine Mobile, AL 36688, USA Ever since the first description of ischemic preconditioning by Murry et al. (12) there has been an ongoing search for the elusive end-effector of ischemic preconditioning’s protection. For the past decade attention has focused on the K ATP channel following the initial discovery by Gross and colleagues that K ATP channels appeared to be involved in preconditioning’s protection (8). It was first assumed that sarcolemmal channels were responsible for the protection. Indeed, the presumption that the opening of these channels would shorten the action potential and reduce calcium entry (a prime suspect as the perpetrator of lethal injury in the ischemic heart) seemed completely plausible. That hypothesis was soon shown to be flawed, however, when it was observed that the potassium channel openers could protect even in the absence of shorten- ing of the action potential (10, 18). Additionally dofetilide, a blocker of the inward rectifying K + current, could prevent any shortening of the action potential seen during ischemia with- out abolition of preconditioning's protection (9). Furthermore, potassium channel openers could easily mimic precondition- ing in isolated, non-beating cardiomyocytes not generating any action potentials (1). This conundrum seemed to be solved when two simultaneously published papers proposed that it was mitochondrial rather than sarcolemmal K ATP channels that mediated the protection of preconditioning (7, 11). Diazoxide was identified as a putative selective opener of mitochondrial K ATP channels (mK ATP ), while 5-hydroxydecanoate (5HD) was noted to be a selective blocker of the channels. Since diazoxide mimicked preconditioning and 5HD blocked it, the hypothe- sis that the mK ATP mediated protection and might even be the end-effector fit the data nicely. The only problem was that there was no rational explanation of why opening the mK ATP BRC 228