Differences in Protection and Susceptibility to Blockade With Single-Cycle Versus Multicycle Transient Ischemia
Background We compared ischemic preconditioning (IP) induced with a single cycle of transient ischemia and reperfusion with that induced by multiple cycles in terms of (1) efficacy of protection against myocardial necrosis and (2) susceptibility to pharmacological blockade by inhibition of protein kinase C (PKC) or elevation of cAMP.
Methods and Results All rabbits were subjected to 30 minutes of regional ischemia and 90 minutes of reperfusion in vivo. IP was induced with either one or three cycles of 5-minute transient ischemia and 10-minute reperfusion given before the 30-minute ischemia. Drug-treated hearts received a bolus dose of one of the following just before the 30-minute ischemia: (1) the PKC inhibitor chelerythrine (3.8 mg/kg), (2) the PKC inhibitor polymyxin B (10 mg/kg), or (3) the cAMP-increasing agent NKH477 (45 μg/kg). IP induced with either one or three cycles of transient ischemia and reperfusion significantly protected the heart against infarction, although the extent of protection was significantly greater with three-cycle IP. Chelerythrine, polymyxin B, or NKH477 alone did not alter infarct size in control hearts, nor did they increase infarct size in hearts preconditioned with three-cycle IP. In contrast, when IP was induced with only a single cycle, all three of these drugs significantly increased infarct size above that of the untreated one-cycle IP group. However, infarct size in all three of these drug-treated one-cycle IP groups was still significantly lower than that in the corresponding drug-treated controls, indicating a partial block of IP.
Conclusions Three-cycle IP provided more effective protection against myocardial necrosis than one-cycle IP and was less susceptible to blockade by inhibitors of PKC or an agent that increases cAMP levels. However, single-cycle IP was only partially blocked by either inhibition of PKC or stimulation of cAMP production. Neither activation of the PKC pathway nor reduced formation of cAMP alone fully accounted for the necrosis protection by IP even when induced with only a single cycle of transient ischemia.
- myocardial infarction
- signal transduction
- receptors, adrenergic, alpha
- receptors, adrenergic, beta
Brief episodes of myocardial ischemia interspersed with reperfusion render the heart remarkably resistant to the injury produced by a subsequent episode of ischemia and reperfusion. This acute adaptive response, called ischemic preconditioning (IP), has been shown to protect against several damaging sequelae of ischemia and reperfusion, including necrosis,1 2 3 arrhythmias,4 5 and postischemic contractile dysfunction.6 7 Because IP appears to be the most powerful myocardial protection strategy investigated to date, much attention has focused on understanding the characteristics and mechanism(s) of IP, ultimately to harness its potential for clinical use.
For clinical application of IP, it is desirable to maximize the extent of protection attained. Increasing the number of cycles of transient ischemia and reperfusion used to induce IP may achieve this goal. However, the conventional wisdom, as stated in several review articles, is that the protection afforded by IP against necrosis is all or none and that increasing the number of cycles used to induce IP does not result in a cumulative increase in protection.4 8 9 The protection by IP against arrhythmias, however, is not all or none. Two studies4 10 have reported a cumulative increase in the protection by IP against arrhythmias when the number of cycles used to induce IP is increased from one to three. Furthermore, three cycles of transient ischemia and reperfusion attenuate norepinephrine release10 and cAMP levels11 during the subsequent sustained episode of ischemia to a significantly greater extent than IP induced with only a single cycle.
In addition to differences in the extent of protection achieved with single-cycle versus multicycle IP, differences in susceptibility to pharmacological blockade may also exist. Many investigators have sought to elucidate the mechanism of IP by attempting to block its protection with drugs targeted at sites along the signal transduction pathway suspected of being involved in IP. To induce IP, these studies have generally used experimental protocols with either one cycle or multiple cycles of transient ischemia and reperfusion but not both protocols for IP in the same study, with the implicit assumption that IP induced with a single cycle of transient ischemia is as susceptible to blockade as that induced with multiple cycles. However, this assumption has not been directly tested.
In light of the evidence suggesting that at least some aspects of IP show clear dose dependency,4 10 11 the first goal of this study was to reexamine the postulate that the protection by IP against necrosis is all or none. The second goal was to compare the susceptibility to blockade of IP induced with a single cycle of transient ischemia and reperfusion versus that induced with three cycles by manipulation of two different signal transduction pathways implicated in the mechanism of IP. The signal transduction pathways investigated were activation of PKC and inhibition of PKA through a reduction in cAMP formation. We used the PKC inhibitor chelerythrine or polymyxin B to inhibit PKC and the adenylyl cyclase activator NKH477 to raise cAMP levels in the heart and thus increase activation of PKA. Our results show that IP induced with three cycles of transient ischemia more effectively protects the rabbit heart against necrosis than that induced with a single cycle and is less easily blocked either by inhibitors of PKC or by an agent that elevates cAMP levels. Furthermore, when IP was induced with a single cycle of transient ischemia, it was only partially blocked by inhibitors of PKC or by an agent that elevates cAMP levels, suggesting that both of these pathways are involved in the mechanism of single-cycle IP but that neither alone can fully account for the observed protection.
We used New Zealand White rabbits of either sex for this study (weight range, 3 to 4 kg). Rabbits were treated in accordance with the Guide to the Care and Use of Laboratory Animals published by the US National Institutes of Health. The experimental protocol was approved by the Animal Care Committee of The Hospital for Sick Children, Toronto, Ontario.
To induce anesthesia, we placed each rabbit in a restrainer and cannulated the marginal ear vein of the right ear to administer sodium pentobarbital (28 mg/kg) and heparin (285 IU/kg). During the experiment, anesthesia was maintained with sodium pentobarbital (25 mg·kg−1·h−1). Anticoagulation was maintained with heparin (100 IU·kg−1·h−1 IV). We administered both drugs diluted in 5% dextrose (25 mL/h).
After the induction of surgical anesthesia, we opened the neck with a ventral midline incision and performed a tracheotomy. We ventilated the rabbits with 100% oxygen using a respirator (Harvard Apparatus). The respiratory rate was set at 35 breaths per minute; the tidal volume was maintained between 18 and 24 mL. We took arterial blood samples for blood-gas determinations periodically during the experiments and adjusted the tidal volume to keep the Pco2, pH, and Po2 within normal physiological limits. A heat lamp was placed over the animals to ensure that body temperature was maintained between 37°C and 38°C during the experiment. We used a catheter inserted into the left carotid artery to monitor blood pressure and heart rate. We cannulated the marginal ear vein of the left ear for infusion of chelerythrine and polymyxin B.
To expose the heart, we performed a left thoracotomy through the fourth intercostal space and opened the pericardium. We then passed a 2-0 polypropylene thread attached to a curved needle around the left main coronary artery. After the needle was removed, we passed the two ends of the thread through the lumen of a flanged piece of polyethylene tubing. To create ischemia, we pushed the flange against the coronary artery to produce a snare and clamped it in position with a mosquito clamp. We confirmed the presence of ischemia by observing cyanosis distal to the occlusion site. Reperfusion was achieved by releasing the snare. If fibrillation occurred during the course of the experiment, a defibrillator (Burdick Co) was used. If the heart fibrillated more than three times, the animal was euthanized and the experiment was terminated.
The experimental protocols for all groups used in the necrosis studies are presented in Fig 1⇓. All animals used for the necrosis studies received 30 minutes of regional myocardial ischemia and 90 minutes of reperfusion.
Study 1: Does IP Induced With Three Cycles of Transient Ischemia and Reperfusion Provide More Protection Against Necrosis Than IP Induced With Only a Single Cycle?
Animals preconditioned with three cycles of transient myocardial ischemia had an initial 30-minute stabilization period followed by three episodes of 5 minutes of ischemia and 10 minutes of reperfusion before the 30-minute ischemia and 90-minute reperfusion. Animals preconditioned with one cycle of transient myocardial ischemia were subjected to the same initial 30-minute stabilization followed by an additional 30-minute stabilization before receiving a single episode of 5 minutes of ischemia and 10 minutes of reperfusion before the 30-minute ischemia and 90-minute reperfusion. The purpose of the additional stabilization period in these one-cycle IP experiments was to compensate for the two fewer cycles of transient ischemia and reperfusion. The corresponding control group also received the same initial 30-minute stabilization period, followed by an additional 45 minutes of stabilization (total stabilization period, 75 minutes) before the 30-minute ischemia and 90-minute reperfusion. Thus, for these three groups (three-cycle IP, one-cycle IP, and controls), the total length of the experiment was 3 hours and 15 minutes.
In initial experiments, it became clear that in hearts preconditioned with the one-cycle IP protocol, the protection against necrosis was weak (the median necrosis measured 36.7% of the area at risk, compared with a median of 61.9% of the area at risk in controls). We reasoned that with such weak protection, the dynamic range within which either a partial or full block of one-cycle IP could be resolved with the drugs used in study 2 (described below) would be small and therefore could potentially lead to erroneous conclusions. Consequently, we attempted to achieve greater protection with one-cycle IP. We found that shortening the stabilization period from 60 minutes (ie, 30-minute initial stabilization plus 30-minute additional stabilization) to only 5 minutes before giving one cycle of transient ischemia and reperfusion significantly increased the effectiveness of one-cycle IP (median of 10.4% necrosis). As a comparison, a corresponding control group was generated in which animals received an initial 5-minute stabilization period followed by an additional 15-minute stabilization period (equivalent to the time required for one-cycle IP) before 30-minute ischemia and 90-minute reperfusion. Therefore, for both of these groups (ie, the short-stabilization one-cycle IP group and the corresponding short-stabilization control group), the total length of the experiment was 2 hours and 20 minutes.
Study 2: Is There a Difference in the Susceptibility of One- and Three-Cycle IP to Pharmacological Blockade With the Use of PKC Inhibitors or an Adenylyl Cyclase Activator?
Because shortening the stabilization period improved the effectiveness of one-cycle IP, we used the short stabilization protocol in all one-cycle IP and corresponding control groups in this study. All animals in this study received one of these four basic protocols before receiving 30 minutes of regional ischemia and 90 minutes of reperfusion: (1) 30 minutes of initial stabilization+three cycles of 5-minute transient ischemia and 10-minute reperfusion (long-stabilization three-cycle IP), (2) 30 minutes of initial stabilization+45 minutes of additional stabilization for a total of 75 minutes of stabilization (long-stabilization control), (3) 5 minutes of initial stabilization+one cycle of 5-minute transient ischemia and 10-minute reperfusion (short-stabilization one-cycle IP), or (4) 5 minutes of initial stabilization+15 minutes of additional stabilization for a total of 20 minutes of stabilization (short-stabilization control).
For each of the chelerythrine, polymyxin B, and NKH477 treatments, 4 groups were generated (ie, a total of 12 groups). The protocols for each of these 4 groups were identical to those for the untreated control and IP groups (protocols 1 through 4), except that these animals received either chelerythrine (3.8 mg/kg, Research Biochemicals), polymyxin B (10 mg/kg, Calbiochem), or NKH477 (45 μg/kg, Nippon Kyauka Pharmaceuticals) administered over a period of 3 minutes, 5 minutes before the sustained ischemia. All three of the drugs were diluted in normal sterile saline; chelerythrine and polymyxin B were delivered intravenously and NKH477, directly into the left atrium.
To determine whether chelerythrine and polymyxin B did in fact inhibit PKC activity in the rabbit myocardium, we examined the effects of both of these PKC inhibitors on in vitro PKC activity using the Pierce Colorimetric PKC assay kit. Protein kinase C was first purified by phosphatidylserine/polyacrylamide affinity chromatography. Briefly, we obtained ventricular tissue from the untreated, nonischemic rabbit heart, which was subsequently minced and homogenized in 7 times the volume (wt/vol) of ice-cold buffer containing 0.25 mol/L sucrose, 20 mmol/L Tris-HCl (pH 7.5), 2 mmol/L EDTA, 5 mmol/L dithiothreitol, 0.1 mmol/L PMSF, and 0.5 mg/mL leupeptin. The homogenate was centrifuged for 15 minutes at 800g, and the supernatant was subsequently centrifuged at 35 000g for 60 minutes. The crude extract (supernatant) from the 35 000g centrifugation was mixed with the same volume of buffer containing 10 mmol/L MES (pH 6.5), 300 mmol/L KCl, 14 mmol/L CaCl2, 5 mmol/L dithiothreitol, 0.1 mmol/L PMSF, and 0.5 mg/mL leupeptin. The mixture was then loaded on the affinity column as described by Uchida and Filburn.12 Elutions were collected and assayed for PKC activity. The purified PKC was stabilized by addition of BSA (1 mg/kg) and stored at −80°C until the time of PKC assay.
After purification of PKC, the working reagents for the Pierce colorimetric assay were prepared by mixing of appropriate quantities of reaction buffer, fluorescently labeled PKC synthetic peptide substrate (pseudosubstrate peptide), and activator according to the Pierce assay protocol. The assay was carried out by incubating 10 μL of the sample with 15 μL of the premixed working reagent at 30°C for 30 minutes. The reaction mixture was then loaded on the Pierce SpinZyme separation units, and the phosphorylated peptide was bound to the membrane of these units. After the nonphosphorylated peptide was washed off, the phosphorylated peptide was eluted with elution buffer. The absorbance of the elutions at 570 nm was measured for calculating PKC activity in the samples against the PKC standard curve. One unit of PKC activity defined by the Pierce assay protocol equals the amount that would transfer 1 nmol of phosphate to histone H1 per minute at 30°C. The assay of the inhibitory effect of chelerythrine and polymyxin B on rabbit myocardial PKC was carried out by preincubation of these drugs with the enzyme at room temperature for 20 minutes, followed by assay of the remaining PKC activity. Each sample was assayed in duplicate.
We have previously confirmed that the dose of NKH477 used in the present study (45 μg/kg) increased cAMP levels within the area at risk during the 30-minute period of regional ischemia in the three-cycle IP group to levels not different from those for untreated control hearts.11
Study 3: Can PKC Inhibitors Block the Protection by Three-Cycle IP When They Are Administered Before the First Cycle of Transient Ischemia?
Either chelerythrine (3.8 mg/kg) or polymyxin B (10 mg/kg) was administered as a bolus dose over a period of 3 minutes, 5 minutes before the first episode of transient ischemia. In the chelerythrine-treated group, the bolus dose was followed by a constant infusion of chelerythrine (34 μg·kg−1·min−1) given right up until the start of the 30-minute period of sustained ischemia. In the polymyxin B–treated group, no constant infusion followed the bolus dose because polymyxin B has an in vivo half-life of at least 3 hours.13 14 Therefore, it was expected that no appreciable elimination of the drug would have occurred during the three cycles of transient ischemia and reperfusion.
Hearts were excluded from the study only if they met the following a priori exclusion criteria: (1) absence of regional cyanosis after the induction of ischemia, (2) persistence of regional cyanosis after the tissue has been reperfused, (3) ventricular fibrillation more than three times, and (4) technical difficulties associated with delineation of the area at risk.
Infarct and Area-at-Risk Measurement
At the end of the experiment, we euthanized the rabbits, excised the hearts, and mounted them quickly onto a Langendorff apparatus. We then perfused the hearts with normal saline to wash out the blood. Subsequently, we resnared the left main coronary artery at the in vivo occlusion site and perfused the heart with fluorescent particles (Duke Scientific) suspended in normal saline at a perfusion pressure of 75 mm Hg. This technique causes the fluorescent particles to lodge only in the myocardium that is not part of the area at risk, thus delineating it. We then removed the hearts from the perfusion apparatus, sliced the ventricles transversely into slices 3 mm thick, and then incubated the slices at 37°C for 10 minutes with 1.25% TTC solution in 0.2 mol/L Tris buffer (pH 7.4). TTC stains viable myocardium brick red, whereas areas of necrosis appear a tan or brown color. After TTC staining, we placed transparent sheets of acetate over each slice and traced its outline, as well as the outline of necrotic areas, onto the acetates. We then examined the slices under ultraviolet light to see the fluorescent particles and traced the region of nonfluorescence (the area at risk) onto the acetates. These areas (ie, total slice area, area at risk, and necrotic area) were digitized with a digitizing tablet interfaced with a personal computer (Intel 486 microprocessor) and analyzed with Sigma Scan software (Jandel Scientific).
The area at risk, necrosis, and hemodynamic data were first analyzed for normality with the Shapiro-Wilks test (SPSSX version 6.1.2) and for homogeneity of variance with Levene’s test (SPSSX version 6.1.2) to determine whether the assumptions for parametric testing (ie, normal distributions and homogeneous variances between groups compared) were met. These assumptions were met for the hemodynamic data and the area-at-risk data, which are expressed as mean±SEM. Hemodynamic variables were compared with repeated-measures ANOVA with a post hoc Duncan’s multiple range test (SPSSX version 6.1.2). Area-at-risk data were compared by factorial ANOVA with a post hoc Scheffé’s test (Statview version 4.01).
The assumptions for parametric testing were not met for the necrosis data. We therefore used the Mann-Whitney rank sum test, a nonparametric test, to analyze these data. For the same reason, we use box plots to present the necrosis data in the figures. Because the median is a more appropriate measure of central tendency with nonnormal data, we have reported the necrosis data for all groups as median values. Because infarct size data are traditionally expressed as mean±SEM, we have also provided this information in the appropriate figure legends.
The Table⇓ shows the mean arterial blood pressure and heart rate for all of the groups used for the necrosis studies. There were no differences in the baseline values for blood pressure and heart rate between any of the groups compared. NKH477 had no effect on blood pressure but significantly increased heart rate throughout the 30-minute ischemic period in all groups. However, by the end of the final 90-minute reperfusion period, the heart rate–elevating effect of NKH477 had subsided. Chelerythrine had no significant effect on either heart rate or blood pressure. Polymyxin B also had no significant effect on the blood pressure but reduced the heart rate during the 30-minute ischemia in the long-stabilization control group (P<.05 versus untreated control).
Study 1: Does IP Induced With Three Cycles of Transient Ischemia and Reperfusion Provide More Protection Against Necrosis Than IP Induced With Only a Single Cycle?
As shown in Fig 2⇓, three cycles of IP significantly protected the heart from infarction: the median necrosis value was only 0.6% of the area at risk (necrosis values in all groups are expressed as the median) compared with 61.9% in the corresponding control group (P<.0001 versus control). One-cycle IP (with long stabilization) also protected the heart from infarction (36.7% of the area at risk infarcted, P=.05 versus corresponding control), but the extent of protection was significantly less than that obtained with three-cycle IP (P<.001 versus three-cycle IP). Shortening the stabilization period before inducing transient ischemia in hearts preconditioned with one-cycle IP from a total of 60 minutes to only 5 minutes significantly improved the protection obtained with one-cycle IP (10.4% necrosis in short-stabilization one-cycle IP versus 36.7% of the area at risk infarcted in the long-stabilization one-cycle IP, P<.05). Shortening of the stabilization period had no effect on necrosis in control hearts (60.0%, P=NS versus 61.9% of the area at risk infarcted in the long-stabilization control).
Study 2: Is There a Difference in the Susceptibility of One- and Three-Cycle IP to Pharmacological Blockade With PKC Inhibitors or an Adenylyl Cyclase Activator?
Effect of PKC inhibitors on the necrosis protection by one- and three-cycle IP. As shown in Fig 3⇓, the PKC inhibitor chelerythrine administered just before the sustained ischemia did not block the protection by three-cycle IP against necrosis (41.3% in long-stabilization control+chelerythrine versus 7.1% in three-cycle IP+chelerythrine, P<.01). Furthermore, chelerythrine did not significantly increase necrosis above that observed in untreated hearts preconditioned with three cycles of transient ischemia (0.6% in untreated three-cycle IP versus 7.1% in three-cycle IP+chelerythrine, P=NS).
In contrast, when chelerythrine was administered just before the sustained ischemia in the one-cycle IP group, it significantly increased necrosis (10.4% in untreated one-cycle IP versus 24.2% in one-cycle IP+chelerythrine, P<.05). However, necrosis in these chelerythrine-treated one-cycle ischemic preconditioned hearts was still significantly lower than in the corresponding chelerythrine-treated controls, indicating only a partial block of one-cycle IP (54.7% in the short-stabilization control+chelerythrine versus 24.2% in one-cycle IP+chelerythrine, P<.01).
As shown in Fig 4⇓, the results with polymyxin B were similar to those with chelerythrine. Polymyxin B administered just before the sustained ischemia also did not block the protection by three-cycle IP (50.1% in the long-stabilization control+polymyxin B versus 1.7% in three-cycle IP+polymyxin B, P<.01), nor did it increase necrosis in hearts preconditioned with three cycles of transient ischemia (0.6% in untreated three-cycle IP versus 1.7% in three-cycle IP+polymyxin B). However, when polymyxin B was administered just before sustained ischemia in the one-cycle IP group, necrosis was significantly increased (10.4% in untreated one-cycle IP versus 42.8% in one-cycle IP+polymyxin B, P<.01). The polymyxin B–treated one-cycle IP hearts still had significantly less necrosis than their corresponding polymyxin B–treated controls (P<.05), indicating that like chelerythrine, polymyxin B only partially blocked one-cycle IP. Neither chelerythrine nor polymyxin B had a significant effect on infarct size in the control groups.
Effect of PKC inhibitors on PKC activity in rabbit myocardium in vitro. Purification of rabbit cardiac PKC by affinity chromatography resolved a single PKC activity peak. Specific activity of purified PKC (2650 U/mg) was 4500-fold higher than in the crude extract (0.58 U/mg) and comparable to that obtained from commercially available rat brain PKC (2480 U/mg, Calbiochem).
Polymyxin B inhibited rabbit myocardial PKC activity in a dose-dependent manner in vitro. The polymyxin B concentration that inhibited 50% of the PKC activity (IC50) was found to be 4.4 μmol/L in our assay conditions.
Although we found that chelerythrine also inhibited PKC in vitro (data not shown), the extent of inhibition was highly variable, and we were unable to obtain reproducible inhibition curves with this agent despite use of different suppliers of chelerythrine (Biomole, Research Biochemicals, and LC Laboratories).
Effect of an activator of adenylyl cyclase on the necrosis protection by one- and three-cycle IP. As described in our previous work11 and shown in Fig 5⇓, the adenylyl cyclase activator NKH477 also did not block the protection by IP against necrosis when it was induced with three cycles of transient ischemia (62.9% in long-stabilization control+NKH477 versus 0.8% in three-cycle IP+NKH477, P<.05), nor did it increase necrosis above that observed in untreated hearts subjected to three cycles of IP (0.6% in untreated three-cycle IP versus 0.8% in three-cycle IP+NKH477, P=NS). In contrast, when IP was induced with a single cycle of transient ischemia, NKH477 significantly increased necrosis above that observed in the untreated ischemic preconditioned hearts (10.4% in untreated one-cycle IP versus 31.6% in one-cycle IP+NKH477, P<.01). However, necrosis in these one-cycle IP+NKH477 hearts was still significantly less than that in the corresponding controls, indicating a partial block of one-cycle IP (60.8% in NKH477+short-stabilization control versus 31.6% in one-cycle IP+NKH477, P<.01). NKH477 had no significant effect on necrosis in either the long- or short-stabilization control groups.
The absolute difference in the median necrosis between the ischemic preconditioned hearts treated with either chelerythrine, polymyxin B, or NKH477 and the corresponding untreated ischemic preconditioned hearts (ie, the median necrosis in drug-treated ischemic preconditioned hearts minus the median necrosis in the untreated ischemic preconditioned hearts) was 13.9% for one-cycle IP+chelerythrine, 32.4% for one-cycle IP+polymyxin B, and 21.3% for NKH477+one-cycle IP. For hearts subjected to three cycles of IP, chelerythrine, polymyxin B, and NKH477 produced 6.5%, 1.1%, and 0.2% increases in necrosis, respectively.
Study 3: Can PKC Inhibitors Block the Protection by Three-Cycle IP When They Are Administered Before the First Cycle of Transient Ischemia?
Chelerythrine administration before the first cycle of transient ischemia followed by a constant infusion throughout the three cycles of transient ischemia and reperfusion had no effect on the necrosis protection by three-cycle IP (1.6% of the area at risk necrotic in chelerythrine+three-cycle IP versus 0.6% in untreated three-cycle IP, P=NS). Similarly, when polymyxin B was administered before the first cycle of transient ischemia, it also did not significantly increase necrosis in the three-cycle IP model (10.7% in polymyxin B+three-cycle IP versus 0.6% in untreated three-cycle IP, P=NS).
There were no significant differences in the percentage of the ventricular area at risk between any of the groups compared in these studies.
Efficacy of Single-Cycle Versus Multicycle IP in Necrosis Protection
In this study, we show that IP induced with three cycles of transient ischemia is significantly more effective in reducing necrosis than IP induced with one cycle of transient ischemia. This finding contradicts the previously held notion that the protection by IP against necrosis is all or none. That notion was based, in part, on previous findings by Li and coworkers,3 who demonstrated that in an in vivo porcine model of regional myocardial ischemia and reperfusion, IP induced with 1, 6, or 12 cycles of 5 minutes of transient ischemia and 10 minutes of reperfusion produced similar reductions in infarct size. These investigators concluded that increasing the number of cycles of transient ischemia used to induce IP provided no additional protection against necrosis. However, because of the substantial protection afforded by IP in their study (only 3.9±1.3% of the area at risk was necrotic with one-cycle IP), further reductions in necrosis would have been difficult to detect. Similarly, in the rabbit, Miura et al15 detected no significant difference in infarct size in hearts preconditioned with either 1, 2, or 4 cycles of IP. However, examination of their data reveals a trend toward successively lower infarct sizes in hearts preconditioned with multiple cycles.
Whether further protection can be achieved by increasing the number of cycles beyond three is not known and can be tested only in a model of more severe ischemic injury than that used in the present study, because three-cycle IP produced almost complete myocardial salvage in our model, which used 30 minutes of regional ischemia.
Other aspects of IP show a similar dose dependency. In the rat heart, a cumulative increase in the protection by IP against both arrhythmias4 and diastolic dysfunction16 has been reported when the number of cycles used to induce IP is increased from one to three. Similarly, three-cycle IP has been shown to produce significantly greater attenuation in norepinephrine release10 and cAMP levels11 during sustained ischemia than single-cycle IP.
From a clinical standpoint, these collective observations suggest that it may be possible to achieve greater protection against arrhythmias, postischemic diastolic dysfunction, and necrosis in humans when multiple cycles of transient ischemia are used to induce IP. From the point of view of an experimentalist, these same observations suggest that the use of multiple cycles of transient ischemia makes it possible to obtain a larger dynamic range (ie, difference between control and IP groups) and thus greater statistical power to discriminate between control and IP groups for these same three end points.
Effect of PKC Blockade on the Protection by IP Against Necrosis
Activation of PKC is thought to play a critical role in the protection by IP. Much of the evidence implicating PKC activation in IP comes from pharmacological studies demonstrating that inhibitors of PKC block the protection by IP17 18 19 20 21 22 23 and activators of PKC18 19 20 induce cardioprotection.
Our results demonstrate that administration of either the PKC inhibitor chelerythrine or polymyxin B just before the sustained ischemia had no effect on the necrosis protection by three-cycle IP. However, both drugs partially blocked the necrosis protection by one-cycle IP.
Chelerythrine and polymyxin B inhibit PKC by binding to distinctly different binding sites on PKC. Chelerythrine binds to the phosphate acceptor site of the catalytic subunit,24 whereas polymyxin B binds to the phosphatidylserine binding site on the regulatory subunit.25 We also demonstrated that polymyxin B was an effective inhibitor of PKC in the rabbit myocardial enzyme, with an IC50 of 4.4 μmol/L. The volume of distribution of polymyxin B depends on the dose administered; however, in the rat, a 2.2-mg/kg dose has been found to have a 45% volume of distribution.13 Given this volume of distribution and the IC50 of polymyxin B in our rabbits, the dose of polymyxin B used in the present study (10 mg/kg) should have been sufficient to inhibit >90% of the myocardial PKC. Although we clearly observed PKC inhibition with chelerythrine in our in vitro assay system, we were unable to obtain reproducible PKC inhibition curves with this agent. A previous study reported that chelerythrine has a low IC50 for PKC (0.67 μmol/L)24 in rat brain. Chelerythrine has been reported to be a potent inhibitor of PKC-mediated effects both in isolated myocytes26 27 and in vivo.28 29 Assuming a distribution in total body water and based on an IC50 of 0.67 μmol/L, the dose of chelerythrine used in the present study (3.8 mg/kg) should have been sufficient to inhibit all of the PKC.
Our observation that neither chelerythrine nor polymyxin B affected necrosis in ischemic preconditioned hearts when given 5 minutes before sustained ischemia in the three-cycle model demonstrates that PKC activation during the sustained ischemia is not a requirement for IP protection against necrosis in the three-cycle IP model. However, since both of these PKC inhibitors partially blocked the necrosis protection by one-cycle IP, this suggests that the PKC activation that occurred subsequent to the administration of the PKC inhibitor was an important contributor to necrosis protection in the one-cycle model.
An explanation for why the PKC inhibitors had no effect on the necrosis protection by three-cycle IP when administered before the sustained ischemia could be that the processes responsible for the protection may have moved farther downstream from these signal transduction pathways by the time the PKC inhibitors were introduced (ie, PKC translocation, activation, and phosphorylation of the target proteins involved in the protection by IP may have already occurred). However, our results do not support this explanation. Neither polymyxin B nor chelerythrine had any effect on necrosis when initiated before the first cycle of transient ischemia in the three-cycle IP model. This suggests that activation of PKC occurring during the three cycles of transient ischemia and reperfusion was also not necessary for the necrosis protection by three-cycle IP.
In our one-cycle model, we were able to demonstrate only partial blockade of the necrosis protection by IP using either chelerythrine or polymyxin B. This is in contrast to previous studies that have claimed that both polymyxin B18 and chelerythrine23 completely block the protection by one-cycle IP against necrosis in the rabbit.
One possible explanation for the difference between our findings and those of previous reports18 23 may relate to differences in the statistical power of the studies. In most studies that use pharmacological agents to block IP, the convention is to base the claim that an agent blocks preconditioning on the absence of a difference between the drug-treated control group and the corresponding drug-treated preconditioned group. Statistically, it is difficult to arrive at the conclusion that there is no difference between groups unless the statistical power of the comparison is high (ie, >.80). Statistical power is defined as the ability of a statistical test to detect a true difference between groups being compared, and it increases if the number of animals in each group is higher, the variability within the groups being compared is lower, and the dynamic range (ie, difference between control and preconditioned groups) is larger (see Glantz30 for discussion). In infarct-size studies, the generally small numbers of animals in each group and the high variability in infarct sizes result in lower statistical power. Therefore, statements indicating that a block of IP has been achieved by pharmacological intervention may simply reflect a lower statistical power in detecting a difference between the drug-treated control and drug-treated preconditioned hearts. This issue is especially important when it is apparent that either the protective effect being blocked is weak, the sample size is small, or the within-group variability is large.
Effect of Adenylyl Cyclase Activation on the Protection by IP Against Necrosis
We have observed that IP induced with three cycles of transient ischemia completely prevented the nearly twofold rise in cAMP that normally occurs between 10 and 30 minutes of sustained myocardial ischemia.11 Prevention of the rise in cAMP levels during myocardial ischemia should reduce sarcolemmal calcium entry,31 activation of cardiac lipases,32 and energy consumption,33 all effects that are consistent with the observed protective effects of IP.7 34 35 Further support of the hypothesis that lowering cardiac cAMP levels contributes to the protection by IP against necrosis is the observation that the protective effect of IP against necrosis can be blocked by inhibition of G-inhibitory (Gi) proteins,36 an effect that would also be expected to increase cAMP.
Despite the presence of multiple lines of evidence implicating reduced formation of cAMP in the mechanism of IP, in a previous study11 we were unable to block the protection by three-cycle IP with the adenylyl cyclase activator NKH477 administered at a dose that increased cAMP levels in these hearts to levels similar to those observed in untreated control hearts during sustained ischemia. On the basis of this finding, we concluded that the attenuation of the rise in cAMP levels observed with IP during sustained ischemia was not necessary for its protection against necrosis in the three-cycle model. In the present study, however, we demonstrate that when IP is induced with only a single cycle of transient ischemia and reperfusion, raising cAMP levels produces a partial block of IP, suggesting that the reduced formation of cAMP during sustained ischemia is an important contributor to the protective effect of IP against necrosis in the one-cycle model.
Differences in Susceptibility to Blockade in Single-Cycle Versus Multicycle IP
For all of the drugs used in this study (chelerythrine, polymyxin B, and NKH477), IP induced with a single cycle was more susceptible to pharmacological manipulation than that induced with three cycles. One concept that has recently gained popularity is that IP is not produced by a single mediator but rather by multiple mediators that are liberated in response to the transient ischemia and act in concert to stimulate cell-surface receptors to produce the protection by IP. This concept is based largely on data that demonstrate that the blockade of any one of multiple receptor types, including (but not limited to) α1,6 adenosine A1,1 and bradykinin B2,37 blocks the protection by IP and, furthermore, that administration of agonists acting on these receptors protects the heart to an extent similar to that of IP.1 6 37 Because activation of these diverse receptor types affects different signal transduction pathways (eg, α1 and bradykinin B2 activate PKC, and adenosine A1 inhibits adenylyl cyclase38 ), these observations could be extrapolated to suggest that a multiplicity of signal transduction pathways are also collectively altered by IP. Data from the present study demonstrate that two different inhibitors of PKC or an activator of adenylyl cyclase can modify the protection achieved by a single cycle of IP, consistent with a multiple pathway phenomenon. Furthermore, Murphy and coworkers39 also suggested that activation of the lipoxygenase pathway is involved in IP, thus supporting the notion that IP involves activation of multiple signal transduction pathways.
We suggest that when IP is induced with multiple cycles, the cumulative dose of mediators released by episodes of transient ischemia and, consequently, the alterations in downstream signal transduction pathways are much greater than those achieved with a single cycle. Therefore, blockade of a single pathway may be insufficient to block the protection by IP induced with multiple cycles, because alterations in the remaining pathways are sufficient to produce protection. However, when IP is induced with only a single cycle of transient ischemia, the collective alterations of all pathways are relatively more important in the observed protection. If this explanation is correct, it would imply that data obtained with pharmacological inhibitors in multicycle models would reveal only those factors that are nonredundant and therefore necessary for IP to exert its protective effect.
Consistent with the findings of the present study, Goto and coworkers40 recently reported that in the in vivo model of regional ischemia and reperfusion in the rabbit, HOE 140, an inhibitor of the bradykinin B2 receptor, also blocked the necrosis protection by one-cycle IP but had no effect on the protection by four-cycle IP. However, in contrast to the findings of the present study, Goto et al40 did not report significantly greater necrosis protection with four-cycle IP than with one-cycle IP. However, it should be noted that in the Goto et al40 study, the one-cycle and four-cycle IP groups did not receive matched protocols before sustained ischemia, which makes direct comparisons between these two groups more difficult.40
Critique of Methods
In the present study, we used a shorter stabilization period in the one-cycle IP experiments than in the three-cycle IP experiments to test the hypothesis that one-cycle IP was more susceptible to pharmacological blockade than three-cycle IP, because this improved the effectiveness of one-cycle IP and increased the dynamic range of our data (ie, the difference between the one-cycle IP and corresponding control groups). We do not know why shortening the stabilization period improved the effectiveness of the protection by a single cycle of IP in our model. However, we have hypothesized that the stress associated with surgery may result in the systemic liberation of catecholamines. Because catecholamine release has been implicated in the mechanism of IP,6 this effect may contribute to the protection by IP when a short stabilization period is used. A sufficiently prolonged stabilization period may allow time for catecholamine levels to decrease, reducing their contribution to the IP phenomenon. In the control hearts, we think that the higher levels of catecholamines observed in hearts subjected to a short stabilization period was without effect on necrosis because these hearts were not subjected to transient ischemia. On this basis, the high levels of catecholamines by themselves would be insufficient to trigger protection in the absence of transient ischemia.
We used different stabilization periods for the one- and three-cycle IP experiments, but this does not in any way affect the conclusions reached in our studies of the susceptibility of one- and three-cycle IP to blockade, because these conclusions are based on statistical comparisons made between groups in which matched protocols were used. For example, to test the hypothesis that one-cycle IP is more susceptible to blockade than three-cycle IP, comparisons were made between the untreated IP group and the corresponding drug-treated IP group that received the same stabilization period.
The protection by IP against necrosis shows a definite dose dependency. Three cycles of transient ischemia produced significantly greater salvage of ischemic myocardium than preconditioning induced with only a single cycle. Furthermore, IP induced with three cycles is less susceptible than IP induced with one cycle to blockade by inhibitors of PKC or by an agent that elevates cAMP levels. Nevertheless, even with one-cycle IP, manipulation of either of these pathways alone does not fully block the protection by IP against necrosis.
Selected Abbreviations and Acronyms
|PKA||=||protein kinase A|
|PKC||=||protein kinase C|
This study was supported by Ontario Heart and Stroke Foundation grant T-2687. Dr Sandhu is a recipient of an Ontario Ministry of Health Fellowship. Dr Wilson is a recipient of a Career Investigator Award of the Ontario Heart and Stroke Foundation. The manuscript was prepared with the assistance of Editorial Services, The Hospital for Sick Children, Toronto, Ontario, Canada. We gratefully acknowledge Nippon Kyauka Pharmaceuticals for providing NKH477. We thank Usha Thomas for her technical assistance and Dr A.K. Sen for advice.
- Received June 19, 1996.
- Revision received February 3, 1997.
- Accepted February 11, 1997.
- Copyright © 1997 by American Heart Association
Liu G, Thornton J, Van Winkle D, Stanley A, Olsson R, Downey J. Protection against infarction afforded by preconditioning is mediated by A1 adenosine receptors in rabbit heart. Circulation. 1991;84:350-356.
Liu Y, Downey J. Ischemic preconditioning protects against infarction in rat heart. Am J Physiol. 1992;263(4 pt 2):H1107-H1112.
Li G, Vasquez J, Gallagher K, Lucchesi B. Myocardial protection with preconditioning. Circulation. 1990;82:609-619.
Lawson C, Hearse D. Preconditioning and ischemia- and reperfusion-induced arrhythmias. In: Przyklenk K, Kloner R, Yellon D, eds. Ischemic Preconditioning: The Concept of Endogenous Cardioprotection. Norwell, Mass: Kluwer Academic Publishers; 1994:19-40.
Hagar J, Hale S, Kloner R. Effect of preconditioning ischemia on reperfusion arrhythmias after coronary artery occlusion and reperfusion in the rat. Circ Res. 1991;68:61-68.
Banerjee A, Locke-Winter C, Rogers K, Mitchell MB, Bensard DD, Brew EC, Cairns CB, Harken AH. Preconditioning against myocardial dysfunction after ischemia and reperfusion by an α1-adrenergic mechanism. Circ Res. 1993;73:656-670.
Steenbergen C, Perlman M, London R, Murphy E. Mechanism of preconditioning: ionic alterations. Circ Res. 1993;72:112-125.
Walker D, Yellon D. Ischaemic preconditioning: from mechanisms to exploitation. Cardiovasc Res. 1992;26:734-739.
Jenkins D, Baxter G, Yellon D. The pathophysiology of ischemic preconditioning. Pharmacol Res. 1995;31:219-224.
Seyfarth M, Münch G, Schreieck J, Kurz T, Richardt G, Schömig A. Release of norepinephrine is suppressed by preconditioning in rat ischemic hearts. Circulation. 1994;90(suppl I):I-108. Abstract.
Sandhu R, Diaz R, Thomas U, Wilson G. Effect of ischemic preconditioning of the myocardium on cyclic AMP. Circ Res. 1996;78:137-147.
Uchida T, Filburn CR. Affinity chromatography of protein kinase C-phorbol ester receptor on polyacrylamide-immobilized phosphatidylserine. J Biol Chem. 1984;259:12311-12314.
Al-Khayyat A, Aronson A. Pharmacologic and toxicologic studies with the polymyxins. Chemotherapy. 1973;19:82-97.
Hoeprich P. The polymyxins. Med Clin North Am. 1970;54:1257-1265.
Miura T, Adachi T, Ogawa T, Iwamoto T, Tsuchida A, Iimura O. Myocardial infarct size: limiting effect of ischemic preconditioning: its natural decay and the effect of repetitive preconditioning. Cardiovasc Pathol. 1992;1:147-154.
Cave A, Collis C, Downey J, Hearse D. Improved functional recovery by ischaemic preconditioning is not mediated by adenosine in the globally ischaemic isolated rat heart. Cardiovasc Res. 1993;27:663-668.
Cave A, Apstein C. Inhibition of protein kinase C abolishes preconditioning against contractile dysfunction in the isolated blood perfused rat heart. Circulation. 1994;90(suppl I):I-206. Abstract.
Ytrehus K, Liu Y, Downey J. Preconditioning protects ischemic rabbit heart by protein kinase C activation. Am J Physiol. 1994;266(3 pt 2):H1145-H1152.
Speechly-Dick M, Mocanu M, Yellon D. Protein kinase C: its role in ischemic preconditioning in the rat. Circ Res. 1994;75:586-590.
Mitchell M, Meng X, Ao L, Brown J, Harken A, Banerjee A. Preconditioning of isolated rat heart is mediated by protein kinase C. Circ Res. 1995;76:73-81.
Bugge E, Ytrehus K. Ischaemic preconditioning is protein kinase C dependent but not through stimulation of alpha adrenergic or adenosine receptors in the isolated rat heart. Cardiovasc Res. 1995;29:401-406.
Li Y, Kloner R. Does protein kinase C play a role in ischemic preconditioning in rat hearts? Am J Physiol. 1995;268(1 pt 2):H426-H431.
Liu Y, Cohen M, Downey J. Chelerythrine, a highly selective protein kinase C inhibitor, blocks the anti-infarct effect of ischemic preconditioning in rabbit hearts. Cardiovasc Drugs Ther. 1994;8:881-882. Letter.
Herbert J, Augereau J, Maffrand J. Chelerythrine is a potent and specific inhibitor of protein kinase C. Biochem Biophys Res Commun. 1990;172:993-999.
Casnellie J. Protein kinase inhibitors: probes for PKC and the function of protein phosphorylation. Adv Pharmacol. 1991;22:167-205.
Ward C, Moffat M. Role of protein kinase C in mediating effects of hydrogen peroxide in guinea-pig ventricular myocytes. J Mol Cell Cardiol. 1995;27:1089-1097.
Light P, Allen B, Walsh M, French R. Regulation of adenosine triphosphate-sensitive potassium channels from rabbit ventricular myocytes by protein kinase C and type 2A protein phosphatase. Biochemistry. 1995;34:7252-7257.
Serrano P, Rodriguez W, Pope B, Bennett E, Rosenzweig M. Protein kinase C inhibitor chelerythrine disrupts memory formation in chicks. Behav Neurosci. 1995;2:278-284.
Yashpal K, Pitcher G, Parent A, Quirion R, Coderre T. Noxious thermal and chemical stimulation induced increases in 3H-phorbol 12,13-dibutyrate binding in spinal cord dorsal horn as well as persistent pain and hyperalgesia, which is reduced by inhibition of protein kinase C. J Neurosci. 1995;15:3263-3272.
Glantz S. What does ’not significant’ really mean? In: Kaufman B, White B, White J, eds. Primer of Biostatistics. 2nd ed. New York, NY: McGraw-Hill; 1987:138-159.
Tsien R. Calcium channels in excitable cell membranes. Annu Rev Physiol. 1983;45:341-358.
Opie L. Role of cyclic nucleotides in heart metabolism. Cardiovasc Res. 1982;16:483-507.
Katz A. Interplay between inotropic and lusitropic effects of cyclic adenosine monophosphate. Circulation. 1990;82(suppl I):I-7-I-11.
Simkhovich B, Hale S, Ovize M, Przyklenk K, Kloner R. Ischemic preconditioning and long-chain acyl carnitine in the canine heart. Coron Artery Dis. 1993;4:387-392.
Murry C, Richard V, Reimer K, Jennings R. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990;66:913-931.
Thornton J, Liu G, Downey J. Pretreatment with pertussis toxin blocks the protective effects of preconditioning: evidence for a G-protein mechanism. J Mol Cell Cardiol. 1993;25:311-320.
Brew E, Rehring T, Banerjee A. Bradykinin receptors mediate preconditioning through protein kinase C. Circulation. 1994;90(suppl I):I-205. Abstract.
Watson S, Arkinstall S. The G-Protein Linked Receptor Facts Book. San Diego, Calif: Academic Press Ltd; 1994:7-18 and 19-31.
Murphy E, Glasgow W, Fralix T, Steenbergen C. Role of lipoxygenase metabolites in ischemic preconditioning. Circ Res. 1995;76:457-467.
Goto M, Liu Y, Yang X, Ardell J, Cohen M, Downey J. Role of bradykinin in protection of ischemic preconditioning in rabbit hearts. Circ Res. 1995;77:611-621.