Mechanisms of Ischemic Preconditioning in Rat Myocardium
Roles of Adenosine, Cellular Energy State, and Mitochondrial F1Fo-ATPase
Background Adenosine has been proposed as one mediator for the preconditioning effect in the myocardium of some animals, but recent investigations have shown that this may not be the mechanism in the rat heart, although the effect itself is clearly demonstrable. The cellular energy state has been shown to be better in preconditioned hearts, and the role of ATP consumption has been discussed. The role of inhibition of mitochondrial F1Fo-ATPase as a mechanism for the preservation of ATP in preconditioned hearts remains controversial.
Methods and Results Three-minute global ischemia followed by 9 minutes of reperfusion was used to precondition Langendorff-perfused rat hearts, and control hearts were perfused under normoxic conditions for the same time. The duration of sustained ischemia in both groups of hearts was 21 minutes, after which the hearts were reperfused for 15 minutes to evaluate their mechanical and metabolic recovery. Separate experiments were performed for tissue metabolite determinations, mitochondrial ATPase activity measurements, and 31P nuclear magnetic resonance studies. The recovery of the rate-pressure product was better in the preconditioned group. Three-minute preconditioning ischemia caused inhibition of the mitochondrial ATPase that persisted throughout the 9-minute intervening reperfusion so that at the early stages of sustained ischemia the enzyme activity was still more inhibited in preconditioned hearts. ATP was better preserved in preconditioned hearts than in control hearts during sustained ischemia. The accumulation of adenosine and its degradation products during sustained ischemia was greater in the control group. More lactate and H+ ions accumulated in this group, indicating higher anaerobic glycolysis. Also, inhibition of mitochondrial ATPase by oligomycin slowed ATP depletion during ischemia.
Conclusions The results indicate that preconditioning causes inhibition of rat heart mitochondrial ATPase that persists during reperfusion so that the enzyme is inhibited from the very beginning of the sustained ischemia. This inhibition leads to sparing of high-energy phosphates and improves the time-averaged energy state during ischemia. Although a causal relationship is difficult to prove, this reversible inhibition may contribute to postischemic recovery of the heart.
The phenomenon of preconditioning was first described in 1986 by Murry et al,1 who discovered the protective effect of a brief ischemic period with regard to the detrimental consequences of subsequent prolonged ischemia. Since then, preconditioning has been shown to limit infarct size2 and to reduce ventricular arrhythmias during sustained ischemia and reperfusion.3 4 5 Brief repetitive periods of ischemia have been shown to retard cardiac energy metabolism during sustained ischemia in dogs,6 and 31P nuclear magnetic resonance (NMR) has been used as a noninvasive method to demonstrate that the decreases in high-energy phosphate concentrations and intracellular pH during prolonged ischemia are attenuated in the preconditioned pig myocardium.7
Adenosine A1 receptor activation has been proposed as a mechanism for preconditioning in some animal species,8 9 but it has recently been reported that postischemic dysfunction in the rat heart is not attenuated by adenosine agonists or exogenous adenosine.10 11 Experiments with isolated rat cardiomyocytes have also failed to demonstrate a beneficial effect of adenosine and A1-selective agonists on cell survival after ischemia.12 Whatever its mechanism, the clinical significance of preconditioning has been emphasized, and some evidence has been found for it in the human myocardium.13 14
The work of Rouslin15 suggests that mitochondrial ATP synthase, a reversible H+-ATPase (F1Fo-ATPase), may be a major ATP consumer in ischemia. ATP synthesis is driven by the proton gradient and membrane potential of the mitochondrial inner membrane (for a review of the structure and regulation of F1Fo-ATPase, see Reference 1616 ). When this electrochemical gradient collapses during ischemia, the reaction catalyzed by the ATP synthetase begins to run in reverse, resulting in ATP hydrolysis. This imposes a need for a physiological inhibitor of F1Fo-ATPase. Indeed, modulation of F1Fo-ATPase by an inhibitory protein (IF1) has been demonstrated in isolated mitochondria and cultured rat cardiomyocytes subjected to a variation in mechanical work load,17 and it has been proposed that the metabolic role of IF1 is to prevent ATP wastage by the backward reaction of ATP synthase in deenergized mitochondria.17 Inhibition of F1Fo-ATPase during ischemia has also been proposed as one mechanism for ATP preservation due to preconditioning,18 although this has never been studied properly, probably because of methodological difficulties.
Despite the tremendous effort put into experimentation, the mechanism of preconditioning remains controversial. We therefore set out to study further the regulation of the mitochondrial F1Fo-ATPase in preconditioned hearts. Langendorff-perfused isolated rat hearts were used as an experimental model; we measured the activity of mitochondrial ATPase during and after preconditioning and during sustained ischemia. Both noninvasive NMR methodology and a conventional biochemical approach were used to evaluate whether metabolic differences exist between preconditioned and control myocardium during sustained global no-flow ischemia and subsequent reperfusion. The rapidity and reversibility of the regulation of F1Fo-ATPase necessitated a new method for rapid tissue fractionation and activity measurement. This allowed us to demonstrate that reversible regulation of the mitochondrial ATP synthase occurs in the rodent heart and that this may be involved in myocardial preconditioning.
Animals and Perfusion Methods
The work was approved by the local ethical committee in accordance with recognized standards for the care and use of laboratory animals.
Three-month-old male Sprague-Dawley rats from the Department of Physiology, University of Oulu, were used in all of the experiments. They were anesthetized with intraperitoneal sodium pentobarbital (60 mg/kg body wt); 500 IU heparin was injected into the inferior vena cava; and 1 minute later, the hearts were excised and rinsed in phosphate-free ice-cold Krebs-Henseleit perfusion medium containing 120 mmol/L NaCl, 4.7 mmol/L KCl, 2.5 mmol/L CaCl2, 0.25 mmol/L Ca-EDTA, 1.2 mmol/L MgSO4, 25 mmol/L NaHCO3, 10 mmol/L glucose, and 12 IU/L insulin. The hearts were perfused in a modified Langendorff apparatus at a pressure of 90 cm H2O (8.83 kPa) with the medium maintained by thermostat at 37°C and gassed with O2-CO2 (19:1). In the NMR experiments, a 4-m thermostated perfusion line was used to keep the perfusion equipment away from the superconducting magnet of the NMR spectrometer. Total, global ischemia was induced by discontinuation of the perfusate flow.
The hearts were divided into a preconditioning group and a control group for measurement of heart rate, left ventricular systolic pressure, coronary flow, and oxygen consumption and for collection of the effluent perfusate samples. All the experiments were preceded by a 15-minute normoxic stabilization period. After a 9-minute normoxic period, the preconditioning group was subjected to 3 minutes of preconditioning ischemia, a 9-minute intervening reperfusion period, and 21 minutes of ischemia followed by a 15-minute reperfusion. The control group underwent a 21-minute normoxic period and 21 minutes of ischemia followed by a 15-minute reperfusion. The same protocol was used in the NMR experiments except the final reperfusion time was 18 minutes.
Two separate sets of experiments with the above perfusion protocol were performed to obtain tissue samples for both biochemical determination of metabolites (Table 1⇓) and measurement of the activity of F1Fo-ATPase at various time points (as shown in Fig 4⇓). When we tested the effects of oligomycin on the cellular energetics of perfused or ischemic hearts, the antibiotic was dissolved in dimethyl sulfoxide (DMSO) and added to the perfusate to give a final concentration of 3 μg/mL (and 0.3% DMSO). Control perfusions were performed with the same concentration of DMSO alone.
An additional experiment was performed for estimation of the time course of inhibition of mitochondrial ATPase during sustained ischemia. Rat hearts were perfused for 36 minutes, and the apex of each was then cut into three slices. One slice was immediately assayed for ATPase activity, and the other two were placed into plastic self-sealing bags that were incubated in a water bath for either 12 or 21 minutes, after which the ATPase assay was performed. In this way, it was possible to estimate the time course of inhibition using the heart as its own control (see Fig 3⇓).
Left Ventricular Pressure, Heart Rate, and Oxygen Consumption
Left ventricular pressure was monitored by inserting a water-filled latex balloon into the left ventricle and connecting it to a Statham P231D pressure transducer and Statham SP1400 pressure monitor. End-diastolic pressure was adjusted to 15 to 20 mm Hg. Heart rate was measured from the pressure curve with a digital frequency counter. Coronary flow was monitored throughout the experiments using a drop counter with an analog frequency output. Oxygen consumption was calculated as the arteriovenous concentration difference multiplied by the coronary flow.
31P NMR Spectroscopy
A Bruker AM-200 Fourier transform NMR spectrometer was used to record the 31P NMR spectra. The field strength of the superconducting magnet was 4.7 T; phosphorus has a resonance frequency of 81.02 MHz. The heart and a glass capillary containing methylenediphosphonic acid (DMPA) to serve as an internal standard were inserted into a 16-mm tube, which was put into a laboratory-built probe. Before the experiment, the magnet was shimmed to a line width of 0.15 ppm using an H3PO4 solution. Because the insertion of the heart with the auxiliary perfusion equipment caused a minor change in the magnetic field, the homogeneity of the latter was readjusted to the same line width by reference to the water proton resonance readily detectable in the phosphorus-tuned probe.
Blocks of 64 free induction decays were accumulated over consecutive 3-minute periods using a pulse width of 20 μs resulting in a 55-degree tilting angle. A pulse interval of 2.7 seconds and an acquisition time of 0.85 second were used. Multiplication by an exponential function resulted in a line broadening of 10 Hz. Partial saturation of inorganic phosphate (Pi) was observed at this pulse-repetition rate. In calculation of concentrations of the phosphorus-containing metabolites, correction for saturation was applied on the basis of a calibration curve obtained with varying pulsing rate. Because all measurements were made in reference to the DMPA internal standard, only saturation factors relative to DMPA saturation were needed; these were 0.781 for Pi, 1.01 for creatine phosphate (CrP), and 1.202 for ATP-β for the 2.712-second pulse-repetition rate used.
The spectra were manipulated using Bruker win-nmr software (Bruker Spectrospin AG) on a personal computer, with the areas and positions of the resonance peaks calculated after phase correction, baseline correction, and deconvolution. Typical NMR spectra are presented in Fig 6⇓.
The H+ ion concentration in the cytosol was calculated from the chemical shift of Pi by the following equation19 :
where δPi denotes the chemical shift of Pi in ppm by reference to CrP. During sustained ischemia, when the peak area of CrP reached very low values, the DMPA peak (20.9 ppm by reference to CrP) served as a standard for calibration of the ppm scale.
The relative cytosolic concentrations of ATP, CrP, and Pi were calculated from the area under the ATP-β-phosphorus, CrP, and Pi peaks, respectively.
Effluent perfusate samples were collected over consecutive 1-minute periods for the measurement of adenosine and its degradation products (inosine, hypoxanthine, and xanthine). The samples were quenched with perchloric acid (final concentration, 1%) and kept at −20°C until analyzed. Nucleosides and free purines were measured by high-pressure liquid chromatography (HPLC), essentially as described by Raatikainen et al.20
The frozen myocardial tissue samples were pulverized and extracted with 8% (wt/vol) HClO4 in 40% (vol/vol) ethanol precooled to −20°C, and the extraction was repeated with 6% (wt/vol) HClO4. The filtrates were neutralized to pH 6 with 3.75 mol/L K2CO3 containing 0.5 mol/L triethanolamine hydrochloride.21
Lactate and pyruvate were determined by enzymatic methods, measuring the appearance or disappearance of NADH in an Aminco DW-2 dual-wavelength spectrophotometer using an ε340 minus ε385 value of 5.33 mmol/L per centimeter. Tissue adenosine and its degradation products were measured as described by Raatikainen et al.20
Tissue Fractionation and Measurement of F1Fo-ATPase Activity
After perfusion, the hearts were immersed in ice-cold buffer containing 20 mmol/L HEPES, 1 mmol/L MgCl2, and 2 mmol/L EGTA, pH 7.0. The heart was immediately homogenized in an Ultra-Turrax homogenizer (Ika Werk) for 30 seconds. The homogenate (1.5 mL) was sonicated for a total of 25 seconds in five burst time periods separated by 7 seconds of cooling in an ice-water bath. The sonication power and time were optimized to obtain maximal ATPase activity and retention of the in situ regulatory state of the enzyme. The homogenate was then centrifuged at 5300 rpm in a Hereus Biofuge A centrifuge for 30 seconds. The supernatant was diluted in the above-described homogenization-sonication buffer to a protein concentration of approximately 0.5 mg/mL.
ATPase was measured on a Shimadzu-3000 dual-wavelength spectrophotometer at 28°C. The sample (10 μL) was added to a cuvette containing 33 mmol/L Tris acetate, 83 mmol/L sucrose, 10 mmol/L MgCl2, 1 mmol/L KCN, 1 mmol/L EDTA, 2 mmol/L ATP, 1.5 mmol/L phosphoenolpyruvate, 0.17 mmol/L NADH, 6 U pyruvate kinase, and 12 U lactate dehydrogenase at pH 7.2; the decline in the NADH absorbance difference at 340 and 385 nm was observed for 5 minutes, after which oligomycin was added to the reaction cuvette to inhibit F1Fo-ATPase to estimate the contribution of the other ATPase activities.22 Eighty-five percent of the total activity was found to be oligomycin sensitive. The time interval between the end of the perfusion and the beginning of measurement of the enzyme activity varied from 3.5 to 4 minutes. Protein was measured using the Bio-Rad assay.23
Cytochrome c oxidase was assayed essentially according to Cooperstein and Lazarow,24 measuring the oxidation of reduced cytochrome c by the change in the absorbance difference between 550 and 540 nm.
The mechanical function of the hearts during the final reperfusion period was evaluated using summary measures,25 ie, the area under curve was integrated in each experiment for the time period of interest, the average values of the groups were calculated, and Student’s t test for independent mean values was applied. The same test was also used to evaluate the final mechanical recovery of the hearts (last data point) and differences in tissue metabolites and ATPase activities between the preconditioned and control groups at various time points. In statistical evaluation of the recovery of mechanical function, fibrillating hearts were included by giving them a pulse rate value of zero. ANOVA for repeated measurements followed by paired t test applying the Bonferroni method was used to evaluate the time course of mitochondrial ATPase inhibition. A one-way ANOVA followed by the least significant difference test was used for differences in the ratio of ATPase to cytochrome c oxidase and in cytochrome c oxidase activity.
An ANOVA for repeated measurements was used for the data obtained by NMR spectroscopy using the spss program package (SPSS, Inc). Student’s t test served as a post hoc test. Data are expressed as mean±SEM. A value of P≤.05 was considered statistically significant.
Mechanical Performance, Coronary Flow, and Oxygen Consumption
A rapid reduction in left ventricular systolic pressure occurred during the 3-minute preconditioning ischemia, and the pressure remained depressed for the subsequent 9-minute reperfusion compared with the preischemic values. The hemodynamic parameters during sustained ischemia were similar in both groups, but the systolic pressure was higher in the preconditioned hearts (P=.034 by the summary method) during the entire final perfusion; the heart rate also was higher in these hearts during the same period (P=.019 by the summary method). The data on the mechanical performance of the hearts are summarized in Fig 1⇓, in which values are presented in fractional units with reference to the preischemic basal values. The absolute preischemic heart rate, systolic pressure, and rate-pressure product values, respectively, were 3.6±0.2 Hz, 118±10 mm Hg, and 424±35 mm Hg · Hz in the preconditioned group and 3.5±0.2 Hz, 122±12 mm Hg, and 432±47 mm Hg · Hz in the control group; there were no statistically significant differences.
All of the preconditioned hearts recovered from the sustained ischemia during the final reperfusion, but the response of the control hearts to reperfusion was unpredictable. One heart showed an overshoot in mechanical performance; three recovered in the same way as the preconditioned hearts; five showed only partial recovery; and two started a slow, irregular beat and went into ventricular fibrillation. Two hearts remained in asystole.
The preconditioning ischemia caused a marked reactive increase in coronary flow and oxygen consumption during the 9-minute reperfusion, but the temporal patterns of coronary flow and oxygen consumption during the final reperfusion period were similar in the two groups (Fig 2⇓).
Activity of F1Fo-ATPase
The activity of the mitochondrial oligomycin-sensitive ATPase became inhibited in a time-dependent manner during continuous ischemia (P<.05), as depicted in Fig 3⇓. ATPase activities during the ischemia and reperfusion protocol are depicted in Fig 4⇓. The 3-minute preconditioning ischemia caused an inhibition of the enzyme activity that persisted throughout the 9-minute reperfusion period so that the activity of the enzyme was lower in the preconditioned group than in the control group in the early phase of sustained ischemia. There were no differences in ATPase activity from 12 minutes of ischemia on, however. During the final 15-minute reperfusion, enzyme activity returned toward preischemic levels in both groups.
Magnetic Resonance Spectroscopy: High-Energy Phosphates and Cytosolic pH
The CrP-to-Pi ratio, which qualitatively reflects the cellular energy state (cytosolic [ATP]/[ADP] · [Pi]), decreased sharply during the preconditioning ischemia, and an overshoot phenomenon could be seen during the following reperfusion period (Fig 5⇓). In the first 3-minute block of sustained ischemia, the CrP-to-Pi ratio in the preconditioned group was higher than in the control group (P<.05). The overshoot in the CrP-to-Pi ratio could also be seen in the preconditioned group during the final reperfusion period. As shown in Fig 6⇓, this overshoot was mainly caused by a decrease in the Pi concentration. Because the changes in high-energy phosphates and Pi are stoichiometric and the Pi concentration is small compared with that of CrP or ATP, changes in the cellular energy state mainly reflect those of Pi, whether expressed as the CrP-to-Pi or the ATP-to-ADP · Pi ratio.
Despite the decrease in the energy state, the ATP level remained almost unchanged during the 3-minute preconditioning ischemia period due to buffering by the creatine kinase reaction (Fig 5⇑). The ATP level appeared to decrease more slowly in the preconditioned group at the beginning of the sustained ischemia (P≤.05 in the third and fourth 3-minute periods). The difference vanished toward the end of the ischemia period, but the ATP concentration reached higher levels in the preconditioned group than in the control group during the final reperfusion.
pH declined to 6.74±0.04 during the preconditioning ischemia. The rate of decrease in cytosolic pH was then similar in both groups for the first 12 minutes of sustained ischemia, but afterward it settled to a steady level in the preconditioned hearts, whereas in the control hearts it declined further, reaching 5.43±0.14 during the final 3 minutes compared with 5.95±0.09 in the preconditioned hearts (mean±SEM, P=.014 for the difference). When perfusion was commenced, the cytosolic pH rapidly returned to the basal level in both groups.
Because the ATPase measurements suggested that inhibition of the mitochondrial F1Fo-ATPase might be involved in the observed improvement in postischemic survival, the paradigm was tested by infusing oligomycin, an inhibitor of F1Fo-ATPase, into one group of hearts before global ischemia was applied. The 5-minute preischemic infusion of oligomycin did not affect the heart rate or systolic pressure, and the ischemic arrest occurred with a similar time course in oligomycin-treated and control hearts.
NMR measurements showed that although the cellular energy state began a decline on oligomycin infusion due to inhibition of ATP synthesis, further decrease in the ATP concentration during the subsequent ischemia was slowed (P<.05) by oligomycin (Fig 5⇑). However, because the oligomycin inhibition is irreversible, it was not possible to test the effect of oligomycin on functional and metabolic recovery during reperfusion. In hearts infused with plain DMSO at the same 0.3% final concentration, ATP depletion occurred with the same kinetics as in noninfused hearts, so the observed oligomycin effect was not due to the vehicle DMSO. An inefficiency of low concentrations of DMSO in affecting performance or adenine nucleotide content in isolated, perfused rat hearts has also been reported.26
Tissue Metabolite Content
The lactate-to-pyruvate (L-P) ratio, which reflects the redox state of the cytosolic NADH/NAD pool through mediation of the lactate dehydrogenase reaction, responded to the preconditioning ischemia as expected, but the values at the end of the intervening reperfusion were similar in both groups of hearts (Table 1⇑). It is significant that the lactate concentration (P=.079) and the L-P ratio (P<.05) were higher in the nonpreconditioned group at the end of the sustained ischemia.
Tissue concentrations of adenosine and adenosine compounds (adenosine plus inosine plus hypoxanthine plus xanthine) at various time points are presented in Table 1⇑. Although the expected formation of adenosine compounds was observed during the 3-minute preconditioning ischemia, the levels at the end of the 9-minute reperfusion tended to be lower in the preconditioned hearts. The sustained 21-minute ischemia caused a marked rise in adenosine and adenosine compounds, but the increase was smaller in the preconditioned group. There were no differences in tissue adenosine compounds between the groups at the end of the final reperfusion.
An enhanced output of adenosine could be observed in effluent perfusate after 3 minutes of preconditioning ischemia (12.2±3.2 versus 1.79±0.5 nmol · min−1 · g dry wt−1), but the adenosine levels in the efflux had returned to preischemic levels (2.1±0.4 nmol · min−1 · g dry wt−1) by the end of the 9-minute reperfusion period. It is surprising that although the tissue levels of adenosine compounds differed significantly after a long period of ischemia, no difference could be seen in the total efflux levels of adenosine compounds (preconditioned, 4.1±0.5 μmol/g dry wt; control, 4.3±0.4 μmol/g dry wt). The slopes of the adenosine output curves were similar during the final reperfusion period (data not shown).
Yield of Mitochondria
As the ATPase activity was measured in a rather crude tissue fraction, it was important to determine that the observed activity changes were not due to variations in the yield of mitochondria. For this purpose, we measured the activity of cytochrome c oxidase, a mitochondrial inner membrane enzyme, in the tissue fraction. Das and Harris27 used NADH oxidase as a mitochondrial marker enzyme, but since it has been shown that NADH oxidase activity decreases markedly during the first hour of ischemia,28 we used cytochrome c oxidase, an intrinsic membrane enzyme that is fairly stable during ischemia.
Table 2⇓ shows that the observed changes (on a protein basis) persist when expressed relative to the inner membrane marker and that the yield of mitochondria remains fairly constant throughout the experimental protocol.
Results from the present study suggest that preconditioning of the myocardium in the rat heart may be mediated by inhibition of F1Fo-ATPase, since inhibition of the mitochondrial ATPase activity resulted in better preservation of the tissue energy state and pH and in better final recovery of the preconditioned hearts. These results are also in accordance with those of Asimakis et al,10 Cave et al,11 and Ganote et al,12 showing that adenosine probably is not the mediator of the preconditioning effect in the rat heart.
Mitochondrial ATP synthase (F1Fo-ATPase) catalyzes a reversible reaction of ATP hydrolysis that is coupled to proton pumping across the mitochondrial inner membrane. Under normoxic conditions, the mitochondrial respiratory chain maintains an electrochemical H+ gradient across the mitochondrial inner membrane sufficient for driving the reaction toward ATP synthesis, but this gradient drops in ischemia and hypoxia, so that the F1Fo-ATPase reaction runs toward ATP hydrolysis.
F1Fo-ATPase is estimated to be a major ATP consumer during ischemia in a number of animal species.29 Mitochondrial ATPase can be actively modulated in submitochondrial particles and intact mitochondria on manipulation of the activity of the mitochondrial respiratory chain.30 31 Rouslin15 was the first to demonstrate an inhibition of mitochondrial ATPase in the ischemic heart. Ischemia-induced ATPase inhibition has been regarded as a property of the cardiac myocytes of large mammals,32 but Das and Harris17 demonstrated that work transitions in cultured rat cardiomyocytes also lead to changes in mitochondrial ATPase when measured after rapid cell disruption.
Although the role of the inhibition of mitochondrial ATPase in preconditioning effect has been discussed by several authors,6 18 33 34 no direct ATPase activity data have been published. In our rat heart model, we found that 3 minutes of ischemia were capable of causing enzyme inhibition that persisted throughout the intervening reperfusion period, so that the enzyme was still inhibited when the sustained ischemia began. The ATPase activity in the preconditioned hearts had decreased to the same level as in the nonpreconditioned hearts after 10 minutes of sustained ischemia.
The time course of ATPase inhibition must be interpreted in the context of the metabolic energetics and mechanical variables observed. As can be seen in Fig 5⇑, there is a significant overshoot in the charging of the cellular energy reserves after the preconditioning ischemia despite the inhibition of mitochondrial ATPase. It is most probable that this exaggerated energization is due to subdued work output (stunning) of the myocardium (Fig 1⇑), resulting in an increase in the ATP supply-to-consumption ratio. This overshoot on reperfusion in rat heart has been observed previously, and its magnitude decreases when the length of ischemia increases.35 The short, 3-minute ischemia in the present case produces an appreciable overshoot, which more than doubles the CrP-to-Pi ratio. Despite the observed partial inhibition, the capacity of mitochondrial ATP synthase must still be sufficient to meet metabolic needs in normoxia. This is in accord with the current views on cardiac energetics. A large body of evidence indicates that under physiological conditions, the oxidative phosphorylation in myocardium works in near-equilibrium.36 This is an advantage during normoxia because it contributes to the high efficiency of oxidative phosphorylation in heart muscle but also means that the net flux of ATP synthase is far from maximal. During ischemia, this turns to a disadvantage, because during reversal of the reaction the extra capacity of the enzyme results in a high rate of ATP wastage. On the other hand, a noticeable percentage of inhibition of F1Fo-ATPase would be beneficial in ischemia but still tolerable during the early stages of reperfusion because of the basal surplus capacity, which decreases the metabolic control coefficient of the enzyme.37 During sustained ischemia, the high-energy phosphate concentration (CrP plus ATP) declines at the same pace in the preconditioned and control hearts. The energy state, which is determined by the ratio of [ATP]f to [ADP]f · [Pi]f, is higher in preconditioned hearts than in control hearts during the first few minutes of ischemia and throughout the final reperfusion period, as has also been reported by others.38 Moreover, during the final reperfusion, the preconditioned hearts showed both an increased energy state and enhanced mechanical performance compared with the control hearts. Because the degrees of inhibition of ATPase were similar in both groups at the beginning and end of the final reperfusion period, it is obvious that the state of the ATPase in the early stages of ischemia is most important for recovery.
31P NMR is a rather insensitive method, and some of the phosphate compounds that are the main determinants of the cellular energy state are present in very low concentrations, below the detection limit. Therefore, although not apparent in Fig 5⇑, the energy state was probably higher in the preconditioned hearts than in the control hearts when estimated on the basis of the data presented in Table 1⇑, which shows a significant diminution in the output of adenosine and its degradation products. It is known that adenosine formation in the heart is regulated by the concentration of the precursor molecule AMP, which is in turn connected with the cellular energy state through the adenylate kinase reaction.39 If this holds good in the present case, the time-averaged energy state during the sustained ischemia must have been higher in the preconditioned hearts.
The fact that the tissue lactate concentration and L-P ratio were lower in the preconditioned hearts at the end of the sustained ischemia indicates retardation of glycolysis. This also fits with the higher cytosolic pH at this time point (Fig 5⇑). The ATPase reaction is also an H+ producer, and the ATPase inhibition may have contributed to the maintenance of the H+ balance. The reason for the decrease in the glycolytic rate is not evident, but the preconditioning ischemia-induced depletion of glycogen may contribute to it along with the effect of energy sparing by ATPase inhibition.6 Preservation of cytosolic pH by preconditioning has also been shown by others.7 40
The concept of ATPase inhibition during ischemia in small rodent hearts has been under debate, because the IF1 content of the heart of an animal with a fast heart rate has been considered to be low.32 According to Das and Harris,27 no reliable estimates of the IF1 content of rat mitochondrial membranes are available, and the conjectures regarding their low IF1 content have been based on preparations that have not shown downregulation of F1Fo-ATPase.32 The present data show a clear, reversible downregulation of F1Fo-ATPase in rat heart during ischemia. Its mechanism remains unknown, but its reversibility excludes ischemic autolysis. The F1Fo-ATPase values in Table 2⇑ are expressed in reference to cytochrome oxidase, an intrinsic membrane enzyme. The catalytic site of F1Fo-ATPase is in an external arm of the enzyme and can be removed by urea treatment.41 Therefore, it is possible that the observed effect is due to differential recovery of the enzyme in the submitochondrial particles used in the enzyme assay. Also, the reversibility of the phenomenon makes this less probable. It is significant that Rouslin and Broge42 have reported that they have not been able to observe ATPase activity modulation in Langendorff-perfused rat hearts on pacing, perfusate Ca2+ elevation, or isoproterenol stimulation in experiments (unpublished data) where ATPase activity measurements were begun 5 minutes after the conclusion of perturbation. This lack of effect is difficult to explain, although it is true that the method they used for ATPase assay used acid extraction and the determination of Pi,43 which means that the oligomycin sensitivity of the reaction must be tested with a separate sample. The coupled-enzyme assay of Rosing et al22 was used in the present study, so the oligomycin sensitivity could be tested in the same cuvette. The time delay between finishing the biological experiment and beginning the enzyme assay is critical and was reduced here to only 3 minutes. The present data argue that mitochondrial ATPase inhibition by ischemia or oligomycin contributes to ATP sparing in small rodent hearts, although this has been previously demonstrated only in dog hearts.29
Thus, there is a discrepancy between the data reported by Rouslin32 and those reported by Das and Harris17 from a study in cultured cells and reported in the present study in isolated hearts. Although a difference certainly exists between the fast-beating hearts of small rodents and the slow-beating hearts of larger mammals, it would be strange if this were to represent an all-or-none phenomenon despite the presence of a low but measurable amount of IF1 in the former. If the proton conductance of the mitochondrial inner membrane were low and stable during ischemia, the hydrolysis of ATP by F1Fo-ATPase would be self-limiting even in the absence of inhibition by IF1. The physiological significance of IF1 in humans has been elegantly proven by the recent demonstration of its absence in Luft’s disease,44 a condition that results in euthyroid hypermetabolism caused by loosely coupled mitochondria with high ATPase activity.45 The symptoms of Luft’s disease can thus be interpreted as meaning that IF1 is needed to maintain the appropriate level of proton conductance in the mitochondrial inner membrane or the stoichiometry of the proton-pumping ATPase. This mechanism may also be important for cell survival in ischemia.
The role of adenosine in the preconditioning effect remains controversial. It has been suggested by several authors that mechanisms of preconditioning may be species specific, and our experiments show that the formation of adenosine and its degradation products was enhanced by 3 minutes of preconditioning ischemia, as estimated from tissue adenosine and effluent perfusate adenosine measurements, but all the adenosine products had been washed away before the sustained ischemia began. In light of the latter observation and some previous studies,10 12 it therefore appears that adenosine does not play a significant role in the phenomenon, as seen in the rat heart.
It is also interesting that brief low-flow ischemia, zero-flow ischemia, and hypoxia improve the recovery of rat heart from subsequent, more prolonged ischemia to the same extent.46 47 All three conditions would result in mitochondrial deenergization, which is known to lead to ATPase inhibition.17 It is therefore possible that the protection is linked to the inhibition of F1Fo-ATPase in these cases, too.
We conclude that preconditioning ischemia induced a slowly reversible inhibition of mitochondrial ATPase in the rat myocardium and that this may play an important role in the preservation of high-energy phosphates during sustained ischemia.
This study was supported by grants from the Medical Research Council of the Academy of Finland; the Sigrid Juselius Foundation, Helsinki, Finland; and the Ida Montin Foundation, Finland. We thank Maija-Leena Lehtonen for her technical assistance.
- Received October 31, 1994.
- Accepted December 13, 1994.
- Copyright © 1995 by American Heart Association
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;74:1124-1136.
Shiki K, Hearse DJ. Preconditioning of ischemic myocardium: reperfusion-induced arrhythmias. Am J Physiol. 1987;253:H1470-H1476.
Osada M, Sato T, Komori S, Tamura K. Protective effect of preconditioning on reperfusion induced arrhythmias of isolated rat hearts. Cardiovasc Res. 1991;25:441-444.
Murry CE, Richard VJ, Reimer KA, Jennings RB. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990;66:913-931.
Kida M, Fujiwara H, Ishida M, Kawai C, Ohura M, Miura I, Yabuuchi Y. Ischemic preconditioning preserves creatine phosphate and intracellular pH. Circulation. 1991;84:2495-2503.
Liu GS, Thornton J, Van Winkle DM, Stanley AWH, Olsson RA, Downey JM. Protection against infarction afforded by preconditioning is mediated by A1 adenosine receptors in rabbit heart. Circulation. 1991;84:350-356.
Thornton JD, Liu GS, Olsson RA, Downey JM. Intravenous pretreatment with A1-selective adenosine analogues protects the heart against infarction. Circulation. 1992;85:659-665.
Asimakis GK, Inners-McBride K, Conti VR. Attenuation of postischaemic dysfunction by ischaemic preconditioning is not mediated by adenosine in the isolated rat heart. Cardiovasc Res. 1993;27:1522-1530.
Cave AC, Collis CS, Downey JM, Hearse DJ. Improved functional recovery by ischaemic proconditioning is not mediated by adenosine in the globally ischaemic isolated rat heart. Cardiovasc Res. 1993;27:663-668.
Ganote CE, Armstrong S, Downey JM. Adenosine and A1 selective agonists offer minimal protection against ischaemic injury to isolated rat cardiomyocytes. Cardiovasc Res. 1993;27:1670-1676.
Deutsch E, Berger M, Kussmaul WG, Hirshfeld JW Jr, Herrmann HC, Laskey WK. Adaptation to ischemia during percutaneous transluminal coronary angioplasty: clinical, hemodynamic, and metabolic features. Circulation. 1990;82:2044-2051.
Rouslin W. Protonic inhibition of the mitochondrial oligomycin-sensitive adenosine 5′-triphosphatase in ischemic and autolyzing cardiac muscle. J Biol Chem. 1983;258:9657-9661.
Senior AE. ATP synthesis by oxidative phosphorylation. Physiol Rev. 1988;68:177-231.
Katz LA, Swain JA, Portman MA, Balaban RS. Intracellular pH and inorganic phosphate content of heart in vivo: a 31P-NMR study. Am J Physiol. 1988;255:H189-H196.
Williamson JR, Corkey B. Assays of intermediates of citric acid cycle and related compounds by fluorometric enzymate methods. In: Colowick SP, Kaplan NO, eds. Methods in Enzymology, vol 13. New York, NY: Academic Press; 1969:439.
Cooperstein SJ, Lazarow A. A microspectrophotometric method for the determination of cytochrome oxidase. J Biol Chem. 1951; 189:665-670.
Matthews JMS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. Br Med J. 1990; 300:230-235.
Das AM, Harris DA. Regulation of the mitochondrial ATP synthase in intact rat cardiomyocytes. Biochem J. 1990;266:355-361.
Rouslin W, Erickson JL, Solaro RJ. Effects of oligomycin and acidosis on rates of ATP depletion in ischemic heart muscle. Am J Physiol. 1986;250:H503-H508.
Ferguson SJ, Harris DA, Radda GK. Influence of the inhibitor on adenosine triphosphate-dependent reactions. Biochem J. 1977; 162:351-357.
Rouslin W. The mitochondrial adenosine 5′-triphosphatase in slow and fast heart rate hearts. Am J Physiol. 1987;252:H622-H627.
Liu Y, Downey JM. Ischemic preconditioning protects against infarction in rat heart. Am J Physiol. 1992;263:H1107-H1112.
Yellon DM, Alkhulaifi AM, Browne EE, Pugsley WB. Ischaemic preconditioning limits infarct size in the rat heart. Cardiovasc Res. 1992;26:983-987.
Bailey IA, Seymour AM. The effect of reperfusion on the 31P n.m.r. spectrum of ischemic rat hearts. Biochem Soc Trans. 1981;9:234-235.
Brown GC. Control of respiration and ATP synthesis in mammalian mitochondria and cells. Biochem J. 1992;284:1-13.
Miyamae M, Fujiwara H, Kida M, Yokota R, Tanaka M, Katsuragawa M, Hasegava K, Ohura M, Koga K, Yabuuchi Y, Sesayama S. Preconditioning improves energy metabolism during reperfusion but does not attenuate myocardial stunning in porcine hearts. Circulation. 1993;88:223-234.
Albuquerque CP, Gerstenblith G, Weiss RG. Importance of metabolic inhibition and cellular pH in mediating preconditioning contractile and metabolic effects in rat hearts. Circ Res. 1994; 74:139-150.
Kagawa Y, Racker E. Partial resolution of the enzymes catalyzing oxidative phosphorylation, IX: reconstruction of oligomycin-sensitive adenosine triphosphatase. J Biol Chem. 1966; 241:2467-2474.
Tzagoloff A, Byington KH, MacLennan DH. Studies of the mitochondrial adenosine triphosphatase system, II: the isolation and characterization of an oligomycin-sensitive adenosine triphosphatase from bovine heart mitochondria. J Biol Chem. 1968; 243:2405-2412.
Yamada EW, Huzel NJ. Distribution of the ATPase inhibitor proteins of mitochondria in mammalian tissues including fibroblasts from patient with Luft’s disease. Biochim Biophys Acta. 1992; 1139:143-147.
Di Mauro S, Bonille E, Lee CP, Schotland DL, Scarpe A, Conn H, Chance B. Luft’s disease: further biochemical and ultrastructural studies of skeletal muscle in the second case. J Neurol Sci. 1976; 27:217-232.
Shizukuda Y, Mallet RT, Lee S-C, Downey HF. Hypoxic preconditioning of ischaemic canine myocardium. Cardiovasc Res. 1992; 26:534-542.