INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DETERMINATION of the molecular mechanism of myocardial preconditioning is being actively pursued in many laboratories. Studies have suggested several distinct mechanisms, but it is unclear how these diverse systems fit together. Adding to the confusion are the findings that rat hearts seem to be preconditioned via a mechanism(s) different from those of dogs, pigs, and rabbits (7, 8, 15, 16). In early studies done by Murry et al. (18), preconditioning in dogs was associated with a reduced ATP concentration immediately after preconditioning, but the rate of ATP decline during a subsequent prolonged ischemia was slowed, such that ATP concentrations were higher in preconditioned than in sham-treated hearts early into ischemia (although ATP levels eventually became similar later into ischemia). Studies in rats have yielded more diverse results, with investigators showing preconditioning to reduce (1, 14), increase (27), or have no effect (3) on the ATP reduction observed during prolonged ischemia relative to nonpreconditioned hearts.

A recent study by Vuorinen et al. (27) showed that preconditioning in an isolated rat heart model is associated with ATP conservation, and these investigators suggested an inhibition of mitochondrial F₁F₀ adenosinetriphosphatase (ATPase) during ischemia to explain these data. Experimental evidence suggests that F₁F₀ ATPase may be an important consumer of ATP during ischemia (10, 13, 22). This multimeric enzyme is localized to the inner membrane and synthesizes ATP by harnessing the energy from the proton motive force created by the proton pumping of electron transport proteins (10, 17). The ATPase runs in reverse when the electrochemical (proton) gradient collapses, as seen during ischemia, and therefore hydrolyzes ATP at a time when this is least desired. This is interesting, inasmuch as it is proposed that the ATP hydrolysis by the mitochondrial ATPase does not result in useful work and therefore is "wasted." Previous work suggested that mitochondria become inefficient during ischemia and/or reperfusion (19). Inhibitors of this ATPase, such as oligomycin, have been shown to reduce the rate of ATP depletion during ischemia (13, 22, 27), further suggesting that the mitochondrial ATPase consumes ATP during ischemia. An endogenous inhibitor of this enzyme (IF₁) has been found to selectively inhibit under conditions of low pH and/or deenergized state, although inhibition during ischemia is certainly not complete (10, 21, 22, 30). It was therefore interesting that Vuorinen et al. found the mitochondrial ATPase to be inhibited by preconditioning in the rat heart. Vander Heide et al. (25) recently showed that preconditioning in dogs is not associated with inhibition of the mitochondrial ATPase, although differences between species in terms of regulation of the ATPase have been reported (21).

Several studies, including our own, have shown preconditioning in isolated rat hearts to be associated with a paradoxical decrease in the time to onset of ischemic contracture, while reperfusion recovery of contractile function is enhanced (1, 14, 23). Ischemic contracture is usually associated with ATP depletion, which is thought to cause rigor bond formation. Asimakis et al. (1) showed that the time to ischemic contracture is reduced and ATP is depleted more rapidly in preconditioned rat hearts. The study by Vuorinen et al. (27) did not report times to contracture, and it would be interesting to determine the effect of preconditioning on ATP depletion and mitochondrial ATPase activity in a model in which contracture is known to occur sooner. Therefore, the goal of this study was to determine the effect of preconditioning on ATP depletion during prolonged ischemia in isolated rat hearts and how this relates to mitochondrial ATPase activity.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Preparation of isolated rat hearts. Male Sprague-Dawley rats (400-500 g) were anesthetized using pentobarbital sodium (100 mg/kg ip). The trachea was intubated, and then the jugular vein was injected with heparin (1,000 U/kg). While the hearts were mechanically ventilated, they were perfused in situ via retrograde cannulation of the aorta. The hearts were then excised and quickly moved to a Langendorff apparatus, where they were perfused with oxygenated Krebs-Henseleit solution containing (in mM) 112 NaCl, 25 NaHCO₃, 5 KCl, 1.2 MgSO₄, 1 KH₂PO₄, 1.2 CaCl₂, and 11.5 glucose at a constant perfusion pressure (85 mmHg). A water-filled latex balloon attached to a metal cannula was then inserted into the left ventricle and connected to a Statham pressure transducer for measurement of left ventricular pressure. The hearts were allowed to equilibrate for 15 min, at which time end-diastolic pressure (EDP) was adjusted to 5 mmHg, and this balloon volume was maintained for the duration of the experiment. Preischemia or predrug function, heart rate, and coronary flow (extracorporeal electromagnetic flow probe, Carolina Medical Electronics, King, NC) were then measured. Contractile function was calculated by subtracting EDP from left ventricular peak systolic pressure, resulting in left ventricular developed pressure (LVDP). Cardiac temperature was maintained throughout the experiment by submerging the hearts in 37°C buffer, which was allowed to accumulate in a stoppered, heated chamber.

Preconditioning protocol. Preconditioned hearts were subjected to three 5-min periods of global ischemia separated by 5 min of reperfusion (3 × 5 protocol) followed by a prolonged ischemic period. Sham hearts were subjected to buffer perfusion for an equal amount of time, but without global ischemia before prolonged ischemia. For the determination of the effect of preconditioning on reperfusion contractile function and lactate dehydrogenase (LDH) release, the hearts were subjected to 20 min of global ischemia (prolonged ischemia) starting 5 min after the final preconditioning interval. The hearts were then reperfused for 30 min. A pilot study was performed to show that the three 5-min periods of ischemia were necessary to observe significant preconditioning. Two 5-min periods of ischemia produced only marginal protection in our model, and one period of ischemia provided no protection.

Effect of preconditioning on myocardial ATP concentration and recovery of contractile function. Hearts were collected after preconditioning (3 × 5 protocol, n = 8, after final 5 min of reperfusion following the 3rd period of ischemia) or sham treatment (n = 8), 5 min into prolonged ischemia for preconditioned hearts (n = 8) or sham-treated hearts (n = 8), 10 min into prolonged ischemia for preconditioned hearts (n = 8) or sham-treated hearts (n = 8), 15 min into prolonged ischemia for preconditioned hearts (n = 8) or sham-treated hearts (n = 8), 20 min into prolonged ischemia for preconditioned hearts (n = 8) or sham-treated hearts (n = 8), or 20 min of ischemia + 30 min of reperfusion for preconditioned hearts (n = 8) or sham-treated hearts (n = 8). At each of these time points, the hearts were rapidly frozen with Wallenberger tongs that were cooled in liquid nitrogen and then stored in liquid nitrogen. The hearts were analyzed for ATP using a previously described spectrophotometric technique (9). Cardiac function and coronary flow were measured throughout the protocols, although the data are reported only for the group of hearts (sham treated or preconditioned) subjected to 20 min of global ischemia + 30 min of reperfusion. In these latter groups, cumulative LDH release was also measured as an index of cell viability. LDH release was measured using an LDH assay kit purchased from Boehringer Mannheim (New York, NY) that is based on the method of Wroblewski and La Due (29). LDH was measured in the coronary effluent collected for the 30 min of reperfusion and expressed as a cumulative concentration. The time to the onset of ischemic contracture was defined as the time (min) during the prolonged global ischemia in which the first 5-mmHg increase in EDP was observed.

Effect of oligomycin on myocardial ATP concentration. Another group of hearts was generated to determine the effect of the mitochondrial ATPase inhibitor oligomycin B (Sigma Chemical, St. Louis, MO) on ischemic ATP concentrations and the time to the onset of ischemic contracture. A group of hearts (prepared as described above) was treated with 10 µM oligomycin (n = 8) or vehicle (0.04% dimethyl sulfoxide, n = 8) for 2 min before the onset of global ischemia. A 2-min oligomycin perfusion period was used to minimize the time the heart was exposed to ATP synthase inhibition (the synthase is normally inhibited under nonischemic conditions). The time to the onset of ischemic contracture was determined as described above. Another group of hearts was generated to determine the effect of oligomycin on ATP concentrations during global ischemia. Rat hearts were prepared and isolated as described above and pretreated for 2 min with vehicle (n = 8) or 10 µM oligomycin (n = 8). The hearts were rendered globally ischemic for 10 min, at which time they were rapidly frozen in liquid nitrogen and assayed for ATP.

Effect of preconditioning on mitochondrial F₁F₀ ATPase activity during ischemia. The effect of preconditioning on F₁F₀ ATPase activity was also determined. Rat hearts were preconditioned (or sham treated) as described above, then subjected to various periods of global ischemia and/or reperfusion (n = 4/group). Mitochondria were prepared according to Gasnier et al. (6), and submitochondrial particles (SMP) were prepared according to Matsuno-Yagi and Hatefi (17). Four hearts from each group (preconditioning or sham treatment: baseline and 3 × 5 preconditioning + 5 min of reperfusion 5 and 10 min into prolonged ischemia, and 30 min of reperfusion after 20 min of prolonged ischemia; 10 µM oligomycin: 10 min of ischemia in sham-treated hearts) were immediately put on ice and individually homogenized in 50 mM tris(hydroxymethyl)aminomethane (Tris) · HCl, pH 7.4, 250 mM sucrose, and 1 mM EDTA (2 ml/g tissue) using a Kinematica PT1200 Polytron. The insoluble material was pelleted at 960 g for 15 min (4°C), and the supernatant was used to prepare the crude mitochondrial pellet (8,600 g for 15 min at 4°C). The pellet was redissolved in 10 mM Tris · HCl, pH 7.4, 250 mM sucrose, and 1 mM EDTA and repelleted. The washing and pelleting were repeated twice. The final pellet was resuspended in ~1 ml of buffer and layered onto a mixture of 2.2 ml of 2.5 M sucrose, 12.25 ml of 10 mM Tris · HCl, pH 7.4, and 1 mM EDTA, and 6.55 ml of Percoll. The mixture was centrifuged at 60,000 g for 45 min at 4°C in a Beckman R 50:2 Ti rotor. The second band from the top of the tube was removed with a pasteur pipette, diluted to 30 ml with 10 mM Tris · HCl, pH 7.4, 250 mM sucrose, and 10 mM EDTA, and pelleted at 10,000 g to remove the Percoll. The mitochondrial pellet was resuspended in 10 mM Tris · acetate, pH 7.5, 250 mM sucrose, and 10 mM MgCl₂ and could be stored at 7°C. The frozen mitochondria were thawed at room temperature and homogenized in storage buffer containing 1 mM potassium succinate and 10 mM MnCl₂. The suspension was sonicated using a Branson sonifier at maximum output and 50% pulse mode for 1 min at 4°C. The suspension was allowed to cool for 3-5 min before the sonication was repeated. The pH of the suspension was adjusted to 7.4 with 1 N KOH, and the suspension was centrifuged at 15,000 g for 6 min before the supernatant was decanted and the SMP were pelleted at 100,000 g for 40 min at 4°C. The SMP were washed (0.25 M sucrose and 10 mM Tris · acetate, pH 7.5) and repelleted before they were stored at 70°C or assayed directly. The yield of mitochondria was routinely 5 mg (protein) per rat heart (rat heart wet weight typically 1 g). SMP from sham-treated hearts were found to have a specific ATP synthesis activity of ~100 nmol ATP · min¹ · mg¹ with 200 mM succinate and 5.0 mM ADP at pH 8.0 and ambient temperature using the coupled assay method of Cross and Kohlbrenner (4). ATP synthesis and hydrolysis (assayed as described below) in SMP or intact mitochondria could be completely inhibited by 100 nM oligomycin.

1

1

1

Effect of preconditioning on myocardial lactate. The effect of preconditioning on myocardial lactate accumulation was determined during the prolonged global ischemia. Isolated rat hearts were prepared and subjected to one of several treatments: sham-treated hearts that were not made ischemic (n = 4); sham-treated hearts subjected to 5, 10, 15, or 20 min of global ischemia (n = 4-8/group); sham-treated hearts subjected to 20 min of ischemia + 30 min of reperfusion (n = 4); 3 × 5 preconditioned hearts subjected to 5, 10, 15, or 20 min of global ischemia (n = 4-8/group); and 3 × 5 preconditioned hearts subjected to 20 min of ischemia + 30 min of reperfusion (n = 4). At the end of each study the hearts were rapidly frozen in liquid nitrogen and analyzed for lactate concentration.

Statistics. Differences with respect to time and treatment were discerned using a repeated-measures analysis of variance. A Newman-Keuls post hoc test was used. When appropriate, a Student's t-test was performed to ascertain differences when only two groups were compared. Values are means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effect of preconditioning on functional recovery and ATP concentrations. The effect of preconditioning on cardiac function and coronary flow in isolated rat hearts subjected to a subsequent 20-min ischemia episode and 30 min of reperfusion is shown in Table 1. Baseline cardiac function and coronary flow were similar for both groups. Preconditioning had no effect on heart rate before the prolonged ischemia compared with sham-treated hearts. Preconditioning significantly reduced LVDP compared with the sham-treated group before the prolonged ischemia. During reperfusion, sham-treated hearts displayed a poor recovery of LVDP and significant bradycardia. Reperfusion coronary flow was significantly reduced compared with baseline in sham-treated hearts. Preconditioning significantly improved the recovery of contractile function in terms of heart rate and LVDP. Reperfusion LVDP was improved approximately threefold by preconditioning. Although we showed the data for only 30 min into reperfusion, the rate of recovery of function was significantly enhanced by preconditioning at earlier times as well. Reperfusion coronary flow was also improved by preconditioning, although this is most likely due to the enhanced reperfusion function. The time to the onset of contracture during global ischemia was significantly reduced by preconditioning, paradoxically suggesting enhanced severity of ischemia during this period (11.1 ± 1.8 and 7.0 ± 0.7 min for sham-treated and preconditioned hearts, respectively). Although the data are not shown, reperfusion EDP was significantly reduced by preconditioning, suggesting a protective effect on this parameter at this time. Cumulative LDH release during reperfusion was significantly reduced by preconditioning, indicating significant protection (26 ± 2 and 10 U/g for sham-treated and preconditioned hearts, respectively).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of 3 periods of 5 min of global ischemia separated by 5 min of reperfusion (3 × 5 preconditioning) on myocardial ATP concentrations in isolated rat hearts. Samples were taken after 5 min of reperfusion following final preconditioning episode (POST-PREC) and 5, 10, 15, or 20 min into prolonged ischemia (5', 10', 15', and 20' ISCH) and 30 min of reperfusion (30' REPER) after 20 min of ischemia. * Significantly different from vehicle (P < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effect of 3 × 5 preconditioning on myocardial ATP concentrations at end of final 5-min ischemic period (3 × 5 PREC) or after 5 min of reperfusion following 3 × 5 preconditioning (3 × 5 PREC + REPER). * Significantly different from vehicle (P < 0.05).

Effect of preconditioning on mitochondrial ATPase hydrolase activity. The effect of preconditioning on mitochondrial ATPase activity was addressed by measuring the specific activity of the enzyme in SMP prepared from preconditioned hearts at baseline, after 3 × 5 preconditioning and 5 min of reperfusion, 5 and 10 min into the prolonged ischemia, and 30 min into reperfusion (after prolonged ischemia) and compared with sham-treated hearts (Fig. 3). The comparisons for the hydrolase activity indicate no significant difference between the sham-treated and preconditioned groups at any time point measured. The magnitude of the differences between preconditioned and sham-treated hearts is certainly less than that observed between oligomycin-treated and sham-treated hearts 10 min into ischemia (Fig. 4), where the hydrolase activity was inhibited by >50%. The incomplete inhibition of activity by oligomycin could be due to a suboptimal effective concentration in mitochondria from treated hearts and/or washout of the compound during preparation of the SMP. The ATP hydrolase activity in SMP prepared from sham-treated hearts could be completely inhibited by 100 nM oligomycin, indicating that the activity in these preparations was attributable to the F₁F₀ ATPase and not contaminating ATPases. Thus there appears to be no significant difference in the ATPase hydrolase activity between sham-treated and preconditioned samples during ischemia and reperfusion. Ischemia did reduce the intrinsic hydrolase activity, and this appeared to be a time-dependent phenomenon.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Effect of preconditioning on ATP hydrolase activities of rat heart mitochondrial F1F0 ATPase in submitochondrial particles. Baseline, before preconditioning or sham treatment; Precond, after preconditioning; 5' and 10' Ischemia, 5 and 10 min into prolonged ischemia; 30' Reperfusion, 30 min into reperfusion.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Top: effect of oligomycin on time to onset of contracture. Middle and bottom: effect of 10 µM oligomycin B on myocardial ATP concentrations and mitochondrial ATPase activity, respectively, at 10 min into global ischemia in isolated rat hearts. * Significantly different from vehicle (P < 0.05).

Effect of preconditioning on myocardial lactate accumulation. Because of the observation that preconditioning aggravated ATP depletion, we determined the effect of preconditioning on myocardial lactate accumulation (Fig. 5). Myocardial lactate concentration was significantly increased during the prolonged ischemia in preconditioned and nonpreconditioned hearts. However, starting at 10 min into ischemia, myocardial lactate was increased more in the sham-treated than in the preconditioned hearts. The trend continued at 15 min, although this was not statistically significant. Myocardial lactate was significantly reduced by preconditioning at 20 min into ischemia. On reperfusion, lactate levels dropped back to baseline by 30 min, and no differences between groups were observed.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effect of preconditioning on myocardial lactate concentrations in isolated rat hearts. Data from nonischemic (Baseline) and sham-treated or preconditioned hearts are shown at various times after initiation of prolonged global ischemia. Lactate was increased above baseline levels by ischemia and at 10 and 20 min into ischemia. * Preconditioning significantly (P < 0.05) reduced myocardial lactate. At 30 min into reperfusion, lactate was washed out to equivalent levels in both groups.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Much effort has been devoted to determining the mechanism of myocardial preconditioning. Several potential pathways for mediating the cardioprotective effects of preconditioning have been identified. Many investigators have shown that adenosine A₁ receptors and ATP-sensitive potassium channels interact in preconditioned hearts in a manner that has yet to be clarified (7, 16). Although this appears to be true in species such as dogs and rabbits, it is less clear in rat hearts. Rat hearts can be effectively preconditioned, but the protective effects are not abolished by ATP-sensitive potassium blockers or adenosine A₁ antagonists (8, 15). Studies have suggested that protein kinase C activation is a critical step in preconditioning rat hearts, but it is not known what ion channel or other protein is being phosphorylated (2). It has been suggested that catecholamines are responsible for protein kinase C activation during preconditioning in rat hearts (2).

A recent study by Vuorinen et al. (27) demonstrated that preconditioning in isolated rat hearts is associated with significant inhibition of mitochondrial F₁F₀ ATPase activity during the prolonged ischemia. This enzyme catalyzes the synthesis of ATP under conditions of normal oxygenation, but on collapse of the electrochemical gradient (as may be seen during ischemia) this enzyme catalyzes the hydrolysis of ATP. It has been hypothesized that this hydrolytic activity can account for a high percentage of ATP utilized during ischemia, and it is doubtful whether this energy is efficiently used (13, 22). The study by Vuorinen et al. was designed to address whether preconditioning can inhibit the ATP hydrolase activity of this enzyme and whether this will cause a reduced depletion of myocardial ATP during the prolonged ischemia. These investigators showed that the ATPase activity was inhibited by preconditioning and that ATP was significantly conserved during the prolonged ischemia. In addition, ATP recovered significantly better during reperfusion in preconditioned hearts. Preconditioning did not cause an ATP depletion before the prolonged ischemia in their model. These exciting results suggested a mechanism for the protective effects of preconditioning in rat hearts as well as ATP conservation. Also of interest is the naturally occurring inhibitor protein of the ATP hydrolase activity of the F₁F₀ ATPase, IF₁. Although Vuorinen et al. postulated that IF₁ may be involved in the preconditioning mechanism, its role in conserving ATP during ischemia in rat hearts is not believed to be significant (21). In addition, a recent study by Vander Heide et al. (25) showed that preconditioning in a canine model was not dependent on mitochondrial ATPase inhibition, although differences in preconditioning as well as in the regulation of ATPase function have been observed between species with slow heart rates, such as dogs, and those with fast heart rates, such as rats (21).

Early studies by Murry et al. (18) showed that ischemic preconditioning in canine hearts can cause some degree of ATP depletion before the initiation of the prolonged ischemia. Despite this, the rate of ATP depletion was slowed, such that early into the prolonged ischemia, ATP levels were slightly higher than in sham-treated animals. This conservation of ATP was maintained for only a short period during ischemia, however. Studies from other laboratories have shown preconditioning to reduce ATP before the prolonged ischemia or to have no effect (1, 3, 5). A study by Asimakis et al. (1) showed that ATP is not conserved during prolonged ischemia in rat hearts after preconditioning, despite a significant protective effect on parameters such as postischemic functional recovery. In this interesting study, ATP was depleted faster in preconditioned rat hearts. A recently published study by Chen et al. (3) showed that a preconditioning protocol of four 5-min periods of ischemia caused ATP depletion before the prolonged ischemia and also caused a slightly reduced ATP during the prolonged ischemia compared with sham-treated hearts, although this was not statistically significant. Similar results were observed by Kolocassides et al. (14). Despite the variability of results from these studies, the trend in the data is consistent with our finding that absolute levels of ATP may not be necessary for the protective effect of preconditioning to be expressed.

Previous studies in isolated rat hearts showed that preconditioning in this model is generally associated with a paradoxical decrease in the time to ischemic contracture, whereas reperfusion function is enhanced and necrosis is inhibited (1, 21). Because ischemic contracture is thought to represent rigor bond formation, ATP depletion may be a prime mediator of this effect. Asimakis et al. (1) found that preconditioning in isolated rat hearts was associated with a reduced time to contracture and reduced ATP levels during the prolonged ischemia. Therefore, it is difficult to reconcile the results of Asimakis et al. with the findings of Vuorinen et al. (27), in which significant inhibition of the mitochondrial hydrolase by preconditioning caused ATP conservation in isolated rat hearts. Schjott et al. (24) found preconditioning to decrease the time to contracture, despite ATP conservation, suggesting that ATP levels may not be the only mediator of contracture.

Previous studies from our laboratory showed that preconditioning in rat hearts was associated with a reduced time to ischemic contracture (23). For this reason, we thought that it was possible that preconditioning can be associated with ATP depletion, which would not be consistent with mitochondrial hydrolase inhibition as a protective mechanism of preconditioning in rat hearts. For this reason, we tested the hypothesis that inhibition of the F₁F₀ ATPase was responsible for preconditioning in isolated rat hearts. We found that preconditioning significantly reduced the time to onset of contracture, suggesting a paradoxical aggravation of ischemia at this time. ATP was reduced during the preconditioning protocol but was repleted by the final reperfusion period just before prolonged ischemia. ATP was depleted significantly faster in preconditioned hearts, despite an equivalent ATP concentration before the prolonged ischemia, which is comparable to studies in rat hearts by Asimakis et al. (1) and Kolocassides et al. (14). If one compares the ATP concentration for sham-treated and preconditioned groups at the approximate time of contracture formation, the ATP concentration was ~10 µmol/g dry weight, and therefore contracture formation is associated with a given degree of ATP depletion.

Measurement of ATPase activity showed that preconditioning did not inhibit this enzyme before or during the prolonged ischemia. An inherent difficulty in these measurements is that any "factor" that might regulate the activity of the ATPase must be retained during the preparation of SMP for enzyme assay. Therefore, an interpretation for no difference in activity between samples from preconditioned and sham-treated hearts could be that a factor or factors were purified away before the enzyme was assayed. Although we cannot discount this possibility, we do observe significant inhibition of the ATPase activity when SMP are prepared from hearts treated with oligomycin, and the protocol we follow for preparation of SMP has been shown to retain the inhibitor protein (21). Oligomycin produced physiological effects in our rat heart model that were consistent with mitochondrial ATPase inhibition. A reduced hydrolysis of ATP and an increased time to ischemic contracture were observed with oligomycin treatment, and these results are consistent with previously published studies using oligomycin (13, 22). These data show that our model will detect mitochondrial ATPase inhibition and reacts in a way that will clearly show such activity. The profile of preconditioning in our model is not consistent with ATPase inhibition.

We do not understand the mechanism of the rapid preconditioning-induced ATP depletion and how this can be reconciled with the profound protective effect of preconditioning on reperfusion function and LDH release. It is possible that the ATP can be utilized more effectively in preconditioned hearts so that ionic homeostasis is better maintained. Interestingly, reperfusion function is enhanced by preconditioning, and this is consistent with the enhanced recovery of ATP during reperfusion. Previous studies have shown no improvement of reperfusion ATP repletion or an enhanced repletion in preconditioned hearts (3, 24, 26, 27). It is difficult to understand how reperfusion function can be enhanced with poor ATP repletion, unless the ATP is somehow more accessible or is being used more efficiently. Previous studies have suggested that preconditioning is associated with reduced glycolysis (perhaps due to glycogen depletion) and, therefore, reduced lactate production during prolonged ischemia (12, 26, 28). The reduced acidosis may be expected to manifest a protective effect. Our study shows that glycolysis is inhibited (reduced lactate), and this is in agreement with the glycogen depletion (or reduced glycogenolysis) hypothesis of preconditioning and is also consistent with the greater reduction in ATP in preconditioned hearts.

The potential problem of model differences, of course, must be explored. The study by Vuorinen et al. (27) used a single 3-min period of ischemia to induce preconditioning, whereas we could not achieve preconditioning with this protocol. This is difficult to understand, inasmuch as our Langendorff preparations appeared to be similar. We need a minimum of three 5-min periods of ischemia to achieve preconditioning, although we see slight protection with two periods of ischemia (data not shown). Our times of prolonged global ischemia were nevertheless similar. Despite the fact that we both used global ischemia, ATP depletion was not as great in the study of Vuorinen et al., suggesting a lower severity of ischemia. These investigators used ³¹P nuclear magnetic resonance for measurement of ATP, in which the ATP data were acquired over 3 min. Therefore, their ATP values may be higher than ours, because we froze our hearts and the ATP values were not averaged over 3 min, although this would not explain why preconditioned ATP values were higher than time-matched sham group values during prolonged ischemia. Our rat heart model and that of Vuorinen et al. responded in a similar manner to oligomycin, suggesting that both models are sensitive to agents inhibiting the mitochondrial hydrolase.

Determination of the mechanism of preconditioning has yielded confusing results, perhaps suggesting the possibility of multiple mechanisms. It has been reported that glycolytic flux is slowed by preconditioning, and perhaps this can explain the reduced ATP levels in preconditioned hearts. Even this is uncertain, inasmuch as other investigators have shown that glycolysis is enhanced by preconditioning (11). With the matrix of data collected in our study, we believe that inhibition of mitochondrial ATPase is not responsible for the protective effects of preconditioning in our model. Our results are in agreement with a recently published study of Vander Heide et al. (25).