Inhibiting mitochondrial permeability transition pore opening: a new paradigm for myocardial preconditioning?

Abstract

Objective: We propose that ischemic preconditioning (IPC) and mitochondrial K_ATP channel activation protect the myocardium by inhibiting mitochondrial permeability transition pore (MPTP) opening at reperfusion. Methods: Isolated rat hearts were subjected to 35 min ischemia/120 min reperfusion and assigned to the following groups: (1) control; (2) IPC of 2×5 min each of preceding global ischemia; (3,4,5) 0.2 μmol/l cyclosporin A (CsA, which inhibits MPTP opening), 5 μmol/l FK506 (which inhibits the phosphatase calcineurin without inhibiting MPTP opening), or 20 μmol/l atractyloside (Atr, a MPTP opener) given at reperfusion; (6,7) pre-treatment with 30 μmol/l diazoxide (Diaz, a mitochondrial K_ATP channel opener) or 200 nmol/l 2 chloro-N⁶-cyclopentyl-adenosine (CCPA, an adenosine A1 receptor agonist); (8) IPC+Atr; (9) Diaz+Atr; (10) CCPA+Atr. The effect of mitochondrial K_ATP channel activation on calcium-induced MPTP opening in isolated calcein-loaded mitochondria was also assessed. Results: IPC, CsA when given at reperfusion, and pre-treatment with diazoxide or CCPA all limited infarct size (19.9±2.6% in IPC; 24.6±1.9% in CsA, 18.0±1.7% in Diaz, 20.4±3.3% in CCPA vs. 44.7±2.0% in control, P<0.0001). Opening the MPTP with atractyloside at reperfusion abolished this cardio-protective effect (47.7±1.8% in IPC+Atr, 42.3±3.2% in Diaz+Atr, 51.2±1.6% in CCPA+Atr). Atractyloside and FK506, given at reperfusion, did not influence infarct size (45.7±2.1% in Atr and 43.1±3.6% in FK506 vs. 44.7±2.0% in control, P = NS). Diazoxide (30 μmol/l) was shown to reduce calcium-induced MPTP opening by 52.5±8.0% in calcein-loaded mitochondria. 5-Hydroxydecanoic acid (100 μmol/l) was able to abolish the cardio-protective effects of both diazoxide and IPC. Conclusion: One interpretation of these data is that IPC and mitochondrial K_ATP channel activation may protect the myocardium by inhibiting MPTP opening at reperfusion.

Time for primary review 34 days.

1 Introduction

Ischemic preconditioning (IPC) describes a phenomenon whereby single or multiple brief periods of non-lethal ischemia protect the heart from a subsequent prolonged lethal ischemic period [1]. The mechanism by which IPC protects the myocardium from ischemia–reperfusion injury is not clear, although opening of the mitochondrial K_ATP channel [2], and activation of G-protein coupled receptors by ligands such as adenosine have been implicated [3]. Pre-treatment with mitochondrial K_ATP channel openers or adenosine A1 receptor agonists, prior to lethal ischemia–reperfusion, have been shown to induce ischemic preconditioning-like protection [2,3]. Conversely, blockers of the mitochondrial K_ATP channel, such a 5-hydroxydecanoic acid, have been shown to abolish cardio-protection induced by both ischemic [4], and pharmacological preconditioning [2]. Both, the mechanism by which mitochondrial K_ATP channel activation protects the myocardium and whether it acts as a trigger or mediator/effector of preconditioning are not clear [5].

The mitochondrial permeability transition pore (MPTP) which has been implicated in ischemia–reperfusion injury has not previously been examined in the context of myocardial preconditioning. The MPTP is a non-specific pore of the inner mitochondrial membrane [6], which on opening causes cell death by apoptosis/necrosis [7]. Opening the MPTP allows water and solutes to enter the mitochondria, increasing matrix volume, and rupturing the outer mitochondrial membrane, leading to release of intermembrane cytochrome C release, which in turn initiates apoptosis. In addition, its opening uncouples mitochondria, leading to ATP hydrolysis and collapse of the mitochondrial membrane potential. The MPTP has been shown to stay closed during ischemia and only open in the first few minutes of reperfusion [8], when the conditions for its opening are present: high mitochondrial [Ca²⁺], ATP depletion, increased [Pi], oxidative stress and increased matrix pH [9]. In support of its role as a mediator of ischemia-reperfusion injury, inhibiting MPTP opening at reperfusion, using cyclosporin A, has been shown to be cardio-protective [10]. Furthermore, inhibition of MPTP opening has been shown to inhibit cytochrome C release and apoptosis [7].

The aim of this study was to examine the role of the MPTP in the context of myocardial preconditioning and test whether IPC or mitochondrial K_ATP channel opening induce cardio-protection by inhibiting MPTP opening at reperfusion. In isolated perfused heart studies, using infarct size as the end-point of injury, we assessed the effect of pharmacologically opening the MPTP at reperfusion, to determine whether the cardio-protection observed with both ischemic or pharmacological preconditioning could be abolished. In isolated calcein-loaded mitochondria we investigated the effect of mitochondrial K_ATP channel activation on calcium-induced MPTP opening.

2 Methods

2.1 Animals

Ninety-seven male Sprague–Dawley rats (400±50 g body weight) were used. All animals were obtained from Charles River UK Limited, (Margate, UK), and received humane care in accordance with The Guidance on the Operation of the Animals (Scientific Procedures) Act 1986 (The Stationery Office, London, UK). The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.2 Isolated heart perfusion

Rats were anesthetized with sodium pentobarbital (55 mg/kg intraperitoneally) and given heparin sodium (300 IU). Hearts were rapidly excised and placed in ice-cold buffer, and mounted on a constant pressure (100 mmHg) Langendorff-perfusion apparatus. They were perfused retrogradely with modified Krebs–Henseleit bicarbonate buffer containing (in mmol/l): NaCl 118.5, NaHCO₃ 25.0, KCl 4.8, MgSO₄ 1.2, KH₂PO₄ 1.2, CaCl₂ 1.7, and glucose 12. All solutions were filtered through a Whatman 2.0-μm microfilter and gassed with 95% O₂/5% CO₂ (pH 7.35–7.50 at 37 °C). Temperature was continuously monitored by a thermoprobe inserted into the pulmonary artery and maintained between 36.5 °C and 37.5 °C. A latex, fluid-filled, isovolumic balloon was introduced into the left ventricle through the left atrial appendage and inflated to give a preload of 8 to 10 mmHg. Left ventricular developed pressure, heart rate and coronary flow were noted at regular intervals. A surgical needle was passed under the left main coronary artery, and the ends of the suture were passed through a pipette tip to form a snare. Regional ischemia was induced by tightening the snare and reperfusion was started by releasing the ends of the suture.

2.3 Materials

Diazoxide (Sigma) was dissolved in dimethyl sulfoxide (DMSO) and added to the Krebs–Henseleit buffer such that the final DMSO concentration was less than 0.02%. 2 chloro-N⁶-cyclopentyl-adenosine (CCPA, Sigma), atractyloside (Atr, Sigma), and 5-hydroxydecanoic acid (5-HD, Sigma) were dissolved in distilled water and added to the Krebs–Henseleit buffer. Cyclosporin A (Sigma), and FK506 (Fujisawa) were dissolved in 50% ethanol and added to the Krebs–Henseleit buffer such that the final ethanol concentration was less than 0.005%. All other reagents were of standard analytical grade.

2.4 Treatment protocols for infarct studies

The experiment protocols for the infarct studies are presented in Fig. 1. All hearts received 35 min regional ischemia and 120 min reperfusion. The hearts were randomly assigned to one of the following 13 treatment groups: (1) Control hearts were perfused with 0.02% DMSO (n = 3) prior to lethal ischemia or 0.005% ethanol (n = 3) at reperfusion for 15 min or Krebs–Henseleit buffer alone throughout (n = 6); (2) IPC hearts (n = 6) were treated with two periods of 5 min each of global ischemia with a 10-min intervening reperfusion before the lethal ischemia; (3,4,5) hearts were perfused with cyclosporin A (0.2 μmol/l, n = 6), FK506 (5 μmol/l, n = 6), or atractyloside (20 μmol/l, n = 12) 5 min before reperfusion for 20 min. This concentration of CsA has been previously shown to protect the isolated rat heart from ischemia–reperfusion injury when given at reperfusion [10]. This concentration of atractyloside has been previously shown to abolish the beneficial effects of calcium preconditioning in adult myocytes [11]. Perfusion of drugs to be given at reperfusion were begun 5 min prior to reperfusion for 20 min to ensure that the drug was perfusing the heart immediately at onset of reperfusion; (6) hearts (n = 6) were perfused with diazoxide (30 μmol/l) for 10 min followed by 10 min perfusion with Krebs–Henseleit buffer prior to regional ischemia. This concentration has been previously shown to precondition the isolated perfused rat heart; [2] (7) hearts (n = 7) were perfused with CCPA (200 nmol/l) for 10 min (during which time the hearts were paced at 300 beats/min due to CCPA-induced bradycardia) followed by 10 min perfusion with Krebs–Henseleit buffer prior to regional ischemia. This concentration of CCPA has been shown to precondition the isolated perfused rat heart; [12] (8) hearts (n = 6) underwent IPC (as in group 2) and were then perfused with atractyloside at reperfusion (as in group 5); (9) hearts (n = 6) were pre-treated with diazoxide (as in group 6) and were then perfused with atractyloside at reperfusion (as in group 5); (10) hearts (n = 6) were pre-treated with CCPA (200 nmol/l) and then perfused with atractyloside at reperfusion; (11) hearts (n = 4) were perfused with 100 μmol/l 5-HD prior to and during regional ischemia. This concentration of 5-HD has been shown to block preconditioning in the isolated perfused rat heart; [13] (12) hearts (n = 6) underwent IPC (as in group 2) in the presence of 5-HD (100 μmol/l) prior to and during regional ischemia; (13) hearts (n = 6) were co-perfused with diazoxide (30 μmol/l) prior to regional ischemia and perfused with 5-HD (100 mol/l) alone during regional ischemia.

2.5 Infarct size measurement

At the end of the 120 min reperfusion period, the snare was pulled tight and the heart was slowly perfused with saline solution containing 0.125% Evans blue, to delineate the non-ischemic zone of the myocardium as a dark blue area. After 1–4 h at −20 °C, the hearts were sliced into 2 mm-thick transverse sections and incubated in triphenyltetrazolium chloride solution (1% in phosphate buffer, pH 7.4) at 37 °C for 10–15 min. The tissue slices were then fixed in 10% formalin. In the risk zone the viable tissue was stained red and the infarcted tissue appeared pale. The slices were drawn onto acetate sheets and with the use of a computerised planimetry package (Summa Sketch III, Summagraphics, Seymour, CT, USA), the percentage of infarcted tissue within the volume of myocardium at risk was calculated.

2.6 Studies in isolated cardiac mitochondria

Mitochondria were isolated from rat hearts using a previously described technique [14]. After extraction, the mitochondria were kept on ice, in respiratory buffer containing (in mmol/l): sucrose 250, Hepes 10, K₂ HPO₄ 2, KCl 80, Mg acetate 2, pH 7.5. Aliquots (0.5 mg/ml) were incubated with calcein-AM (1 μmol/l) for 15 min at room temperature. Calcein-AM readily enters the mitochondria and is de-esterified and trapped in the matrix in its free form, which is fluorescent. Once trapped within the mitochondria, calcein leaves the mitochondria if the MPTP opens, resulting in a reduction in calcein fluorescence, thus indicating MPTP opening [15]. The mitochondria were then washed twice with KCl buffer containing (in mmol/l) KCl 120.0, TES 5.0, and MgCl₂ 0.1, with added ATP 0.2 and sodium succinate 10. In order to confirm that a reduction in mitochondrial calcein fluorescence represented MPTP opening, cyclosporin A, was used to inhibit MPTP opening and abolish any observed reduction in calcein fluorescence from atractyloside or calcium-induced MPTP opening. Calcein-loaded mitochondria in KCl buffer were incubated for 10 min at room temperature with the following treatments: (1) 0.1 μmol/l Ca²⁺; (2) 500 μmol/l Ca²⁺ to induce opening of the MPTP; (3) 500 μmol/l Ca²⁺ in the presence of either diazoxide (30 μmol/l), CCPA (200 nmol/l), cyclosporin A (0.2 μmol/l) or 5-HD (100 μmol/l); (4) 0.1 μmol/l Ca²⁺ in the presence of atractyloside (5 mmol/l); (5) 0.1 μmol/l Ca²⁺ in the presence of atractyloside (5 mmol/l) and cyclosporin A (0.2 μmol/l). We used a concentration of 20 μmol/l atractyloside for the isolated heart preparation, the same as that used in a myocyte preparation [11], but for the cardiac mitochondria we used a concentration of 5 mmol/l, a standard concentration used in isolated mitochondrial preparations [16]. It has been suggested that to open the MPTP using atractyloside, both BAX and atractyloside need to be present [17]. This finding is supported in a study [18], where 5 μM atractyloside alone was only able to open the MPTP in isolated mitochondria, when in the presence of Bax. Where Bax is not present, as in the isolated mitochondrial preparations, a much higher concentration of atractyloside is required to open the MPTP. Cytofluorometric analysis was performed on a Partec flow cytometer (Partec, Münster, Germany) equipped with a 488-nm argon laser. The calcein signal was analyzed in the FL1 channel equipped with a band-pass filter at 520 nm; the photo-multiplier value of the detector was 631 V. Data were acquired on the logarithmic scale. Arithmetic mean values of the median fluorescence intensities were determined for each sample for graphic representation. Experiments were performed on mitochondria isolated from six individual rats.

2.7 Statistical analysis

All values are expressed as mean±S.E.M. Infarct size and mitochondrial calcein fluorescence data were analyzed by one-way ANOVA and Fisher's protected least significant difference test for multiple comparisons. Differences were considered significant when P<0.05.

3 Results

3.1 Exclusions

We used 90 rat hearts for the infarct size experiments of which four were excluded owing to poor function during stabilization, and seven rat hearts were used for mitochondrial extraction of which one was excluded owing to poor mitochondrial function.

3.2 Hemodynamic data

Baseline data relating to cardiac function and coronary flow before regional ischemia were similar in all experimental groups. However, CCPA decreased heart rate such that in experimental groups 5 and 9, pacing at 300 beats/min was required. During regional ischemia, coronary flow and left ventricular developed pressure decreased to a similar extent in all groups. An increase in coronary flow upon reperfusion was indicative of successful re-flow, but coronary flow subsequently declined in all groups during the 120 min reperfusion period. In reperfusion the left ventricular developed pressure recovered gradually, though never reaching stabilization values, a feature of ‘run-down’ in this experimental procedure.

3.3 Infarct size data

Infarct size is represented as the percentage of tetrazolium-negative tissue in the ischemic risk zone. IPC significantly reduced infarct size compared with control hearts (19.9±2.6 vs. 44.7±2.0%, P<0.0001; Fig. 2), as did pre-treatment with diazoxide (18.0±1.7 vs. 44.7±2.0%, P<0.0001; Fig. 3) and pre-treatment with CCPA (20.4±3.3 vs. 44.7±2.0%, P<0.0001; Fig. 4). Inhibition of MPTP opening with cyclosporin A at reperfusion also reduced infarct size (24.6±1.9 vs. 44.7±2.0%, P<0.0001; Fig. 4). However, FK506 at reperfusion did not influence infarct size (43.1±3.5 vs. 44.7±2.0% in control; Fig. 4). Atractyloside given alone at reperfusion did not influence infarct size (43.9±1.5 vs. 44.7±2.0% in control; Fig. 2). However, when atractyloside was administered after 35 min ischemia and immediately at reperfusion for 15 min, it abolished the protective effects of IPC (47.7±1.8% in IPC+atrac vs. 19.9±2.6% in IPC, P<0.0001; Fig. 2), diazoxide (42.3±3.2% in diazoxide+atractyloside vs. 18.0±1.7% in diazoxide, P<0.0001; Fig. 3) and CCPA (51.2±1.6% in CCPA+atractyloside vs. 20.4±3.3% in CCPA, P<0.0001; Fig. 4). 5-HD given prior to and during the lethal ischemic period abolished the cardio-protective effects of both IPC (45.8±3.7% in IPC+5-HD vs. 19.9±2.6% in IPC, P<0.0001; Fig. 2) and diazoxide (46.6±2.5 vs. 18.0±1.7% in diazoxide, P<0.0001; Fig. 3). 5-HD given alone prior to and during the lethal ischemic period did not influence infarct size (49.3±4.1 vs. 44.7±2.0% in control; Fig. 3).

3.4 Mitochondrial calcein fluorescence

The mitochondrial calcein fluorescence values after treatment with Ca²⁺ (0.1 μmol/l) plus or minus the treatment drug were taken as the values from which any reduction in fluorescence was measured. Exposure of mitochondria to Ca²⁺ (500 μmol/l) induced MPTP opening, represented by a reduction in calcein fluorescence of 41.9±3.5% (Figs. 5 and 6). This Ca²⁺ -induced reduction in calcein fluorescence was attenuated in the presence of 30 μmol/l diazoxide (18.4±4.1 vs. 41.9±3.5% with 500 μmol/l Ca²⁺ alone, P<0.0001; Figs. 5 and 6). This effect of diazoxide was abolished in the presence of 100 μmol/l 5-HD (41.5±2.4 vs. 18.4±4.1% with diazoxide alone, P<0.0001; Figs. 5 and 6), where the reduction in calcein fluorescence was similar to that with 500 μmol/l Ca²⁺ alone (41.5±2.4 vs. 41.9±3.5% with 500 μmol/l Ca²⁺ alone; Figs. 5 and 6). 5-HD alone did not influence the reduction in calcein fluorescence (44.5±3.7 vs. 41.9±3.5% with 500 μmol/l Ca²⁺ alone; Figs. 5 and 6). CCPA had no effect on Ca²⁺-induced MPTP opening, as shown by similar reductions in calcein fluorescence (44.9±10.2 vs. 41.9±3.5% with 500 μmol/l Ca²⁺ alone; Figs. 5 and 7). Atractyloside (5 mmol/l) was shown to induce MPTP opening, represented by a reduction in calcein fluorescence of 40.3±3.1% (Figs. 5 and 7). Cyclosporin A inhibited opening of the MPTP induced by both 500 μmol/l Ca²⁺, as shown by the reduction in calcein fluorescence being abolished (0.7±0.7 vs. 37.5±4.0% with 500 μmol/l Ca²⁺ alone, P<0.0001; Figs. 5 and 6), and atractyloside-induced MPTP opening, as shown by an attenuated reduction in calcein fluorescence (7.0±2.0 vs. 40.3±3.1% with atractyloside alone, P<0.0001; Fig. 7). Hence, opening the mitochondrial K_ATP channel with diazoxide, decreases Ca²⁺-induced MPTP opening, represented by an attenuated reduction in calcein fluorescence from 37.5±4.0 to 18.4±4.1%, which is equivalent to a 52.5±8.0% reduction.

4 Discussion

The present study suggests that cardio-protection induced by both IPC and pharmacological preconditioning may be associated with inhibition of MPTP opening at reperfusion. Opening the MPTP, pharmacologically, using atractyloside, resulted in an increased infarct size in the hearts preconditioned with ischemia, diazoxide or CCPA, suggesting involvement of the MPTP in the mechanism of protection of these preconditioning treatments. Conversely, inhibiting MPTP opening at reperfusion, using cyclosporin A, resulted in a protective effect similar to that seen with preconditioning. The isolated mitochondrial data, which assessed MPTP opening directly, demonstrated that activation of the mitochondrial K_ATP channel inhibited calcium-induced MPTP opening in calcein-loaded mitochondria.

As expected, CCPA had no effect on MPTP opening in the isolated mitochondria, confirming that it requires the sarcolemmal G protein-coupled receptor to mediate its cardio-protective action. Both atractyloside and Ca²⁺ were shown to open the MPTP to a similar level, as shown by similar reductions in calcein fluorescence and this action was blocked by cyclosporin A, verifying that the observed reduction in calcein fluorescence was due to MPTP opening.

The MPTP is believed to be made up of three core components: the voltage dependant anion channel (VDAC) in the outer mitochondrial membrane; the adenine nucleotide translocase (ANT) in the inner mitochondrial membrane; and cyclophilin D in the mitochondrial matrix [19]. Cyclophilin D, is a peptidyl prolyl cis–trans isomerase, which in the presence of high [Ca²⁺], has been suggested to alter the conformation of the ANT from the m-conformation, where it functions as a nucleotide transporter, to the c-conformation (when adenine nucleotide binding is antagonised), where it functions as the channel of the MPTP [19,20]. According to this model, when the ANT is in the pore-forming c-conformation, as in the first few minutes of reperfusion, adenine nucleotide binding is antagonised, and the ANT is not translocating. Ligands that bind to the m-conformation of the ANT, such as bongkrekic acid, inhibit MPTP opening, whereas ligands that bind to the c-conformation of the ANT, such as atractyloside, open the MPTP [19].

Cyclosporin A is believed to inhibit MPTP opening by preventing the binding of cyclophilin D to ANT, thereby preventing the ANT from changing into the pore-forming c-conformation [20]. Although, cyclosporin A when given at reperfusion, has been shown to protect the isolated heart with respect to recovery of LV function and preservation of ATP levels [10], our study shows, for the first time, that inhibiting MPTP opening at reperfusion with cyclosporin A also protects the heart from cell death as shown by it limiting myocardial infarction. To exclude the possibility of cyclosporin-A protecting via inhibition of the phosphatase, calcineurin [21], FK506 (which inhibits calcineurin but does not inhibit MPTP opening) [10,22] was given at reperfusion, and shown, not to protect the hearts from infarction. Therefore cyclosporin A-mediated protection was most likely due to inhibition of MPTP opening rather than calcineurin inhibition.

To open the MPTP we used the ANT ligand, atractyloside [16,23,24], which has previously, been shown in myocytes to decrease the protective effect of calcium-preconditioning when given prior to hypoxia [11]. Bearing in mind that the MPTP is believed to open in the first few minutes of reperfusion, we only perfused the heart with atractyloside for the first 15 min of reperfusion. As well as opening the MPTP, atractyloside inhibits translocation of adenine nucleotides, a situation that occurs in the first few minutes of reperfusion, when according to the proposed model of the MPTP, the ANT is no longer translocating when in the pore-forming configuration [25]. One may have expected the infarct size to be increased in the hearts given atractyloside alone, but one possible explanation for why this was not the case, may be that atractyloside only opens pores that have been potentially closed by the effect of preconditioning. This is speculation and requires further investigation. We have also conducted preliminary isolated perfused rat heart experiments using phenylarsenine oxide (PAO), another agent that opens the MPTP. This agent induces MPTP opening by cross-linking sulphydryl groups in the mitochondrial membrane [26–28]. We administered 2 μM of PAO at reperfusion for 15 min, following 35 min of ischemia to hearts that had been ischaemically preconditioned (IPC). We were again able to show that opening the MPTP at reperfusion with PAO completely abolished the protective effect of IPC (data not shown). This finding lends further support to our interpretatation of a potential link between IPC and inhibition of MPTP opening at reperfusion.

It is not clear how mitochondrial K_ATP channel activation could be linked to inhibition of MPTP opening, though we can speculate that it could be through three possible mechanisms, which are not mutually exclusive (Fig. 8): (1). Mitochondrial calcium load: Both IPC and opening the mitochondrial K_ATP channel have been shown to decrease mitochondrial Ca²⁺ load either by reducing the electrochemical gradient for Ca²⁺ to enter the mitochondria (from partial mitochondrial membrane depolarisation) or by causing Ca²⁺ efflux from the mitochondria [13,29–31]. Lowering mitochondrial matrix Ca²⁺ load would prevent opening of the MPTP at reperfusion. (2) ATP preservation: Opening the mitochondrial K_ATP channel has been shown to improve mitochondrial energy production [32–34], possibly via mitochondrial matrix swelling [35,36]. Improved oxidative phosphorylation and maintenance of ATP levels would prevent MPTP opening at reperfusion. (3) Reactive oxygen species (ROS): The production of ROS within the first few minutes of reperfusion may mediate the cell damage due to ischemia/reperfusion injury [37]. Oxidative stress has been shown to promote MPTP opening whereas antioxidants have been shown to prevent MPTP opening [38]. In myocytes, it has been shown that hypoxic preconditioning [39], pre-treatment with diazoxide [40], CCPA [41] or the adenosine receptor agonist AMP 579 [42] attenuates the production of ROS at time of reperfusion. This reduction in the production of ROS at time of reperfusion would prevent MPTP opening. However, studies have also shown that opening the mitochondrial K_ATP channel releases ROS which can act as a trigger for preconditioning [5,43,44]. Therefore, ROS display dual roles, both triggering preconditioning and causing ischemia–reperfusion damage. The disparity may relate to the quantity of ROS released with smaller amounts of ROS triggering preconditioning and larger amounts, released at reperfusion, causing cell death, perhaps by opening of the MPTP. Alternatively, the action of ROS may be dependent on the stage they are produced: acting as a trigger for preconditioning if produced prior to ischemia or causing cell death if released at reperfusion.

In conclusion, the results of this study support the possibility that both IPC and pharmacological preconditioning via the mitochondrial K_ATP channel may lead to inhibition of MPTP opening at reperfusion. However, further studies are required to show directly, that IPC and mitochondrial K_ATP channel activation inhibit MPTP opening at reperfusion. The mechanism by which IPC or activating the mitochondrial K_ATP channel could prevent MPTP opening is not known and requires further study, but it may involve reducing mitochondrial Ca²⁺ loading, maintaining mitochondrial ATP levels or reducing the production of ROS at reperfusion. We postulate that inhibition of MPTP opening at reperfusion therefore could represent a distal effector mechanism of myocardial preconditioning with mitochondrial K_ATP channel activation acting as a trigger or an intermediate step. At present, the technical limitation associated with work in the intact heart and isolated mitochondrial preparations and the possible non-specificity of pharmacological tools obviate making a definitive link between mitochondrial K_ATP channel activation and inhibition of MPTP opening. Further studies, investigating the effects of preconditioning and mitochondrial K_ATP channel activation on MPTP opening in the intact cell model would provide further insight. However, even with these limitations, recognition that the mechanism of IPC may involve a critical event at reperfusion may have important implications for potential clinical applications. If our interpretation of these results is correct, the development of agents that inhibit MPTP opening in the first few minutes of reperfusion would offer the cardio-protective effects of IPC and could be used, at time of reperfusion, in the clinical settings of thrombolysis, coronary artery angioplasty and cardiac surgery.

Acknowledgements

Dr Derek Hausenloy is supported by a British Heart Foundation Clinical PhD Studentship (FS/02017). We thank the Hatter Foundation for continuing support.

References