- Top of page
Diazoxide and 5-hydroxydecanoate (5-HD; C10:0) are reputed to target specifically mitochondrial ATP-sensitive K+ (KATP) channels. Here we describe KATP channel-independent targets of diazoxide and 5-HD in the heart. Using submitochondrial particles isolated from pig heart, we found that diazoxide (10-100 μm) dose-dependently decreased succinate oxidation without affecting NADH oxidation. Pinacidil, a non-selective KATP channel opener, did not inhibit succinate oxidation. However, it selectively inhibited NADH oxidation. These direct inhibitory effects of diazoxide and pinacidil cannot be explained by activation of mitochondrial KATP channels. Furthermore, application of either diazoxide (100 μm) or pinacidil (100 μm) did not decrease mitochondrial membrane potential, assessed using TMRE (tetramethylrhodamine ethyl ester), in isolated guinea-pig ventricular myocytes. We also tested whether 5-HD, a medium-chain fatty acid derivative which blocks diazoxide-induced cardioprotection, was ‘activated’ via acyl-CoA synthetase (EC 126.96.36.199), an enzyme present both on the outer mitochondrial membrane and in the matrix. Using analytical HPLC and electrospray ionisation mass spectrometry, we showed that 5-HD-CoA (5-hydroxydecanoyl-CoA) is indeed synthesized from 5-HD and CoA via acyl-CoA synthetase. Thus, 5-HD-CoA may be the active form of 5-HD, serving as substrate for (or inhibiting) acyl-CoA dehydrogenase (β-oxidation) and/or exerting some other cellular action. In conclusion, we have identified KATP channel-independent targets of 5-HD, diazoxide and pinacidil. Our findings question the assumption that sensitivity to diazoxide and 5-HD implies involvement of mitochondrial KATP channels. We propose that pharmacological preconditioning may be reelated to partial inhibition of respiratory chain complexes.
Adenosine triphosphate (ATP)-sensitive K+ (KATP) channels are thought to be present in the sarcolemma of cardiac myocytes as well as in the inner mitochondrial membrane. The structure and function of the sarcolemmal KATP channel has been extensively characterized using molecular biological and electrophysiological techniques. Evidence for the existence of a mitochondrial KATP channel, however, is largely pharmacological (Hu et al. 1999). In particular, diazoxide has been inferred to be a selective opener of this channel, whereas the unsaturated fatty acid derivative 5-hydroxydecanoate (5-HD) has been inferred to be a selective inhibitor. Moreover, since diazoxide also mimics, whereas 5-HD blocks, ischaemic preconditioning, mitochondria have been implicated as effectors of cardioprotection (Gross & Fryer, 1999; Hu et al. 1999).
Evidence that diazoxide selectively opens, and 5-HD selectively blocks, mitochondrial KATP channels has come from work using isolated mitochondrial preparations (Garlid et al. 1996) and intact cardiac myocytes. For example, using rabbit cardiac myocytes, Liu et al. (1998) have shown that diazoxide increases flavoprotein fluorescence, used as an index of mitochondrial KATP channel activation, while exerting no effect on the simultaneously measured sarcolemmal KATP current. However, in the presence of high concentrations of ADP, diazoxide can activate cardiac sarcolemmal KATP channels as well (Matsuoka et al. 2000). The KATP channel opener pinacidil has been reported to increase both flavoprotein fluorescence and sarcolemmal KATP current (Sato et al. 1998). The fatty acid derivative 5-HD was shown to block only the former effect.
The mechanism and extent by which mitochondrial KATP channels could contribute to cardiac protection against ischaemia is not clear. Work with isolated cardiac mitochondria has shown that putative mitochondrial KATP channel openers depolarize the inner membrane, which could stimulate respiration and promote Ca2+ efflux from the matrix (Garlid et al. 1996; Holmuhamedov et al. 1999). It has been postulated that depolarization of the inner mitochondrial membrane by activation of KATP channels, or other means, may protect these organelles from the deleterious effects of Ca2+ overload, which include the mitochondrial permeability transition (Holmuhamedov et al. 1999). An alternative hypothesis is that opening of mitochondrial KATP channels is not the end-effector of the preconditioning cascade but acts as an initial trigger of cardioprotection by inducing release of reactive oxygen species (Pain et al. 2000).
Although they are putative modulators of mitochondrial KATP channels, diazoxide and 5-HD may have other targets in the heart. Diazoxide was reported three decades ago to inhibit succinate oxidation in liver mitochondria (Schäfer et al. 1969). More recently, 100 μm diazoxide has been shown to decrease the rate of succinate oxidation in heart mitochondria (Ovide-Bordeaux et al. 2000). Hence, complex II may be a specific target of diazoxide. Moreover, 5-HD is a hydroxy (-OH) derivative of decanoate (C10:0) and, in principle, it may be metabolized like other medium-chain fatty acids in the heart. In the present study, we tested (i) whether diazoxide and pinacidil target the electron transport chain and (ii) whether 5-HD serves as substrate for acyl-CoA synthetase, an enzyme which thioester-links fatty acids to CoA (coenzyme A) with broad substrate specificity. Some of the results have been published in preliminary form (Hanley et al. 2001).
- Top of page
We have shown that diazoxide, pinacidil and 5-hydroxydecanoate have KATP channel-independent targets in the heart. We found that diazoxide inhibits succinate oxidation and succinate dehydrogenase activity without affecting NADH oxidation, suggesting that complexes I, III and IV are unaffected. These results agree with the previous observations that diazoxide inhibits succinate-supported respiration in mitochondria isolated from liver (Schäfer et al. 1969) and in cardiac mitochondria in situ (Ovide-Bordeaux et al. 2000). We also found that pinacidil, albeit less potently than diazoxide, inhibits the electron transport chain. In the case of pinacidil, NADH oxidation, and not succinate oxidation, is inhibited.
In our experiments with freshly isolated cardiac ventricular myocytes of the guinea-pig we observed no effect of diazoxide (100 μm) on flavoprotein fluorescence. These findings are in line with a recent study in rat cardiomyocytes (Lawrence et al. 2001), but disagree with the report of Liu et al. (1998), who did observe flavoprotein fluorescence changes (with glucose-free physiological salt solution) in rabbit cardiomyocytes kept in culture medium for up to 2 days. This discrepancy may be related to the different experimental conditions or, less likely, to species differences.
Furthermore, we observed no effect of diazoxide (100 μm) or pinacidil (100 μm) on the mitochondrial membrane potential of guinea-pig cardiomyocytes, which confirms and extends the results of Lawrence et al. 2001. Our findings suggest that diazoxide and pinacidil, at concentrations used for preconditioning, do not induce opening of mitochondrial K+ channels. Since pharmacological preconditioning with diazoxide or pinacidil is usually attributed to activation of mitochondrial KATP channels, an alternative explanation for the cardioprotective effect of these drugs is required. We propose that there may be a mechanistic link between partial inhibition of electron transport and pharmacological preconditioning. Several lines of evidence support this link. (i) It has recently been shown that partial inhibition of complex II with a low dose of 3-nitropropionic acid confers ischaemic protection in the rabbit heart (Ockaili et al. 2001). (ii) In the brain, pre-treatment with selective inhibitors of either complex I or complex II has been shown to afford ischaemic protection (Riepe & Ludolph, 1997). (iii) Volatile anaesthetics, which have been deduced to inhibit complex I (Berman et al. 1974), confer ischaemic-like preconditioning in the heart, which is sensitive to 5-HD (Piriou et al. 2000). (iv) Nicorandil, which can also produce pharmacological preconditioning, is known to produce nitric oxide (Sakai et al. 2000), a potent complex IV inhibitor.
How could partial inhibition of the electron transport chain protect the myocardium during subsequent periods of ischaemia? Recent literature suggests that generation of moderate concentrations of reactive oxygen species (ROS) plays an important role in pharmacological preconditioning (Tritto et al. 1997; Pain et al. 2000; Forbes et al. 2001). Under physiological conditions, about 1–2 % of electron flow through the respiratory chain generates ROS (Turrens, 1997). This basal rate of production of ROS is augmented in the presence of complex I and complex III inhibitors (Turrens, 1997; Ide et al. 1999), and during ischaemia (Becker et al. 1999). Diazoxide and pinacidil have been shown to promote ROS production during the conditioning periods (Forbes et al. 2001; Han et al. 2002), and the cardioprotective effect of diazoxide was abolished by free-radical scavengers (Pain et al. 2000; Forbes et al. 2001). Furthermore, inhibition of protein kinase C (PKC), which is known to be activated by ROS (Cohen et al. 2000), also abolished diazoxide-induced cardioprotection (Tritto et al. 1997; Wang et al. 1999; Pain et al. 2000). Taken together, these findings suggest that generation of ROS and activation of PKC represent important upstream mechanisms in pharmacological preconditioning. Interestingly, the same mechanisms, generation of ROS and activation of PKC, have been proposed to play a role in ischaemic preconditioning (Baines et al. 1997; Vanden Hoek et al. 1998; Pain et al. 2000).
Since 5-hydroxydecanoate blocks all forms of preconditioning the elucidation of its mechanism(s) of action could provide a key to the understanding of the molecular basis of preconditioning. We found that 5-HD serves as substrate for acyl-CoA synthetase. This observation is potentially important since it identifies an intracellular target of 5-HD and opens the possibility that the acyl-CoA ester 5-HD-CoA may represent the active form of 5-HD. One possible mechanism of action is that metabolism of 5-HD-CoA via the four-step β-oxidation pathway could provide a limited means for 5-HD to bypass partial inhibition of either complex I or complex II. At the first step, catalysed by acyl-CoA dehydrogenase, electrons are transferred directly to ubiquinone via ETF and ETF dehydrogenase (ETF-QO), as illustrated in Fig. 4. This effect could compensate for the partial inhibition of the respiratory chain by diazoxide or pinacidil. Alternatively, 5-HD-CoA may target other sites, such as sarcolemmal KATP channels (Liu et al. 2001), various PKC isoforms or the ADP/ATP translocase, where acyl-CoA esters are known to exert potent stimulatory or inhibitory actions (reviewed by Knudsen et al. 1999), or even the putative mitochondrial KATP channels.
Figure 4. Schematic diagram showing mitochondrial sites of action of diazoxide, pinacidil and 5-HD
Diazoxide and pinacidil, as well as volatile anaesthetics, inhibit the electron transport chain at the sites depicted. Nicorandil may also inhibit the electron transport chain via the production of NO (nitric oxide). 5-Hydroxydecanoate serves as substrate for the enzyme acyl-CoA synthetase. The principal product of this reaction, 5-HD-CoA (an acyl-CoA ester), may serve as substrate for acyl-CoA dehydrogenase or, possibly, inhibit this enzyme.
Download figure to PowerPoint
In conclusion, we have shown that diazoxide and pinacidil have KATP channel-independent targets. Diazoxide inhibits succinate oxidation (and succinate dehydrogenase activity) whereas pinacidil inhibits NADH oxidation. We have also shown that 5-hydroxydecanoate serves as substrate for acyl-CoA synthetase, which synthesizes 5-hydroxydecanoyl-CoA from 5-HD and CoA. 5-Hydroxydecanoyl-CoA may act directly on some intracellular target or, indirectly, by supporting the electron transport chain via β-oxidation. These findings point towards inhibition of the respiratory chain as a possible primer for the cardioprotective effects of pharmacological preconditioning and question the hypothesis that processes activated by diazoxide and inhibited by 5-HD are necessarily related to mitochondrial KATP channels.