Objective Ischemic postconditioning protects the reperfused heart from infarction, and this protection is dependent on the occupancy of adenosine receptors. We further explored the role of adenosine receptors in this salvage.
Methods In situ rabbit hearts underwent 30min of regional ischemia and 3h of reperfusion, and postconditioning was effected with four cycles of 30-s reperfusion/30-s coronary artery occlusion at the end of ischemia.
Results Postconditioning reduced infarct size from 40.2±3.4% of the risk zone in untreated hearts to 15.5±2.5%. Protection by postconditioning was blocked by either the non-selective adenosine receptor blocker 8-p-(sulfophenyl)theophylline or the A2b-selective antagonist MRS 1754, injected intravenously 5min before reperfusion. The protein kinase C (PKC) antagonist chelerythrine also aborted postconditioning's salvage, indicating a PKC-dependent mechanism. Neither the A1-selective antagonist 8-cyclopentyl-1,3-dipropylxanthine nor the A2a-selective antagonist 8-(13-chlorostyryl)caffeine had an effect on protection. The non-selective but A2b-potent adenosine agonist 5′-(N-ethylcarboxamido)adenosine (NECA) infused from 5min before to 1h after reperfusion mimicked postconditioning's effect on infarct size (17.2±2.7% infarction) and MRS 1754 blocked the NECA-induced cardioprotection, confirming that A2b activation was protective. The PKC activator phorbol 12-myristate 13-acetate delivered just before reperfusion also duplicated the protective effect of postconditioning (16.3±4.1% infarction), and co-administration of the PKC antagonist chelerythrine aborted PMA's protection, confirming that the protection was the result of PKC activation. NECA's protective effect was not affected by chelerythrine, but rather MRS 1754 blocked PMA's salutary effect (42.8±1.0% infarction), suggesting that the A2b receptor's effect is under control of PKC. Finally, wortmannin, a blocker of phosphatidylinositol 3-kinase, also abrogated protection by PMA.
Conclusions Salvage of ischemic myocardium by postconditioning is dependent on activation of A2b receptors, which in turn depends on activation of PKC. It is still unclear why PKC activation is required to make the heart's adenosine become protective.
Recently, it has been shown that the canine heart can be effectively protected from infarction with brief cycles of ischemia and reperfusion during the early reperfusion period following a lethal ischemic insult, a phenomenon termed postconditioning . We were also able to demonstrate sparing of infarction by postconditioning in hearts of open-chest rabbits and showed that this protection was dependent on signal transduction elements leading to nitric oxide (NO) production, ERK activation and opening of the mitochondrial KATP channel . Tsang et al.  extended those observations in a rat model and showed that postconditioning protects myocardium by activating the prosurvival kinases phosphatidylinositol 3-kinase (PI3-K), Akt, and p70S6K.
It is not known what triggers these pathways in the postconditioned heart. One obvious candidate is adenosine which we have previously shown activates PI3-K to increase Akt phosphorylation in the rabbit heart . Furthermore, like postconditioning, protection from the adenosine agonist NECA has been shown to also be dependent on PI3-K, ERK and NO . Indeed, we  and Kin et al.  reported that postconditioning's cardioprotection in an isolated rabbit heart model can be blocked by the non-selective adenosine receptor antagonist 8-p-(sulfophenyl)theophylline (SPT). Ischemic preconditioning was also found to be dependent on adenosine receptor stimulation in the reperfusion phase following the index ischemia . Moreover in that study we presented evidence that the adenosine A2b receptor subtype was largely responsible for preconditioning's protection and that the A2a subtype was not. Although Kin et al.  concluded that A2a receptors were responsible for postconditioning's protection, ZM241385, the antagonist they used, is not highly selective between A2a and A2b receptors and it is possible that their conclusion may have been erroneous. We therefore designed experiments to determine whether the A2b receptor subtype might be responsible for postconditioning's protection. In addition, we also explored the mechanism whereby postconditioning causes adenosine receptors to become protective.
All procedures were approved by the Institutional Animal Care and Use Committee and were conducted in accordance with recommendations published in the Guide for Care and Use of Laboratory Animals.
New Zealand White rabbits of either sex weighing between 1.7 and 2.7kg were anesthetized with intravenous sodium pentobarbital (30mg/kg). Throughout the experiment additional anesthesia was administered as needed (5–15mg pentobarbital/15min). A heating pad maintained rectal temperature between 38.5 and 39.5°C. Animals were intubated through a tracheotomy and ventilated with 100% O2 with the aid of a mechanical ventilator. Arterial pH, pO2, and pCO2 were measured with blood gas analyzer ABL 5 (Radiometer, Copenhagen, Denmark). Respiratory rate was adjusted to keep pH between 7.35 and 7.45. In this preparation arterial pO2 averages 180–200mm Hg and pCO2 25–30mm Hg. A PE 50 catheter filled with heparinized saline (10U/ml) was placed in the right carotid artery to monitor arterial blood pressure and to measure arterial pH and gases. To administer drugs, a PE 50 catheter filled with heparinized saline was placed in an ear vein.
After a left thoracotomy, a prominent branch of the left coronary artery was surrounded with a suture (2–0 silk) to form a snare. The rabbits were allowed to stabilize for 15min after surgery before the protocols were begun.
Hearts of 21 experimental groups experienced 30min of regional ischemia (Fig. 1). In all hearts reperfusion following the occlusion lasted for 3h. Control hearts had only the index occlusion followed by reperfusion. In all postconditioned groups, four cycles of 30-s reperfusion/30-s occlusion were started immediately after release of the index ischemia. In drug-treated animals SPT (7.5mg/kg), MRS 1754 (9.5μg/kg), an adenosine A2b-selective antagonist, 8-cyclopentyl-1,3-dipropylxanthine (DPCPX, 1mg/kg), a selective A1 adenosine receptor antagonist, chelerythrine (5mg/kg), a highly selective antagonist of protein kinase C (PKC), or wortmannin (60μg/kg), an inhibitor of PI3-K, was infused as an intravenous bolus 5min before the onset of reperfusion with or without subsequent postconditioning cycles. A 6-min infusion of 8-(13-chlorostyryl)caffeine (CSC, 660μg/kg), a selective A2a adenosine receptor antagonist, was started 11min before the onset of reperfusion and was also combined with postconditioning.
NECA (2μg/kg), a non-selective but potent A2b adenosine agonist, was infused 5min before the onset of reperfusion. After administration of the bolus, an infusion of 0.2μg/kg/min was continued for 65min. To investigate if the adenosine A2b receptor were responsible for the effects of NECA, we administered a bolus of MRS 1754 (9.5μg/kg) at the same time as NECA 5min before the onset of reperfusion and again after 15min of reperfusion. To determine whether PKC activation was critical to NECA's effect on ischemic myocardium, chelerythrine (5mg/kg) was infused as an intravenous bolus 5min before the onset of reperfusion.
The effect of a left atrial infusion of the PKC activator phorbol 12-myristate 13-acetate (PMA) (0.5μg/kg/min) beginning 5min before reperfusion and extending for 20min in lieu of NECA was examined. Co-infusion of either chelerythrine or MRS 1754 as detailed above was also tested in 2 other groups of rabbit hearts.
Finally, the effects of the tool drugs themselves, SPT, MRS 1754, DPCPX, CSC, chelerythrine, and wortmannin infused 5min before reperfusion were evaluated.
Doses of drugs were selected mainly because of our prior experience of effective action or blockade. DPCPX at a dose of 1mg/kg had previously been shown to block the negative chronotropic effect of the selective adenosine A1 receptor agonists 2-chloro-N6-cyclopentyladenosine  and GR79236 . Preliminary observations with CSC demonstrated that a dose of 660μg/kg infused over 6min completely prevented the hypotensive effect of an intravenous infusion of 0.4mg/kg/min adenosine.
Determination of area at risk and infarct size
After completion of the studies all in situ hearts were excised and the aortic root was perfused with 0.9% saline. The coronary artery was reoccluded, and 2–9μm diameter green fluorescent microspheres (Duke Scientific Corp., Palo Alto, CA) were infused into the perfusate to demarcate the ischemic zone as the area of tissue without fluorescence (region at risk). Hearts were weighed, frozen, and then cut into 2-mm-thick transverse slices. The slices were incubated in 1% triphenyltetrazolium chloride (TTC) in sodium phosphate buffer (pH 7.4) at 38°C for 8min. TTC stains the noninfarcted myocardium brick red indicating the presence of dehydrogenase enzymes. The slices were then immersed in 10% formalin to enhance the contrast between stained (viable) and unstained (necrotic) tissue. The risk zone was revealed as the non-fluorescent region by illumination with UV light. The areas of infarct and risk zone were determined by planimetry of each slice and volumes were calculated by multiplying each area by its slice thickness and summing them for each heart. Infarct size is expressed as a percentage of the risk zone.
Wortmannin, chelerythrine, PMA, SPT, NECA, MRS 1754, CSC, and DPCPX were obtained from Sigma Chemical Co. (St. Louis, MO). They were dissolved in DMSO. The solution was then further diluted in 1.5ml of 0.9% saline before administration to the rabbit. This amount of DMSO had no effect in control hearts.
All data are expressed as mean±S.E.M. One-way analysis of variance (ANOVA) combined with Tukey's post hoc test was performed on baseline hemodynamic and infarct data. Temporal differences in hemodynamic variables in any given group were analyzed with one-way repeated measures ANOVA, and Tukey's post hoc test was used to examine differences between measurements at any given time point and baseline. Differences were considered to be significant if the p value was below 0.05.
No group differences in heart rate, developed pressure, or coronary flow were observed at baseline (Table 1 in the online supplement). No substantial changes in heart rate or blood pressure were observed after administration of drugs. Blood pressure declined mildly during coronary occlusion in all groups.
Although there was some variability in heart weights and risk zone volumes in the groups (Table 2 in the online supplement), there were no significant differences. In control hearts infarct size was 40.2±3.4% of the risk zone (Fig. 2). Infarct size in the postconditioned group was nearly 70% less than that in the control group (15.5±2.5%, p<0.001 versus control). SPT, a non-selective adenosine receptor blocker, aborted protection confirming the involvement of adenosine receptors. Furthermore MRS 1754, a highly selective adenosine A2b receptor antagonist, also blocked postconditioning's protection, thus identifying the A2b receptor subtype as being responsible for postconditioning's protection. Neither DPCPX nor CSC, highly selective A1 and A2a antagonists, respectively, blocked postconditioning's protection, thus making involvement of the A1- and A2a-receptors unlikely.
We next examined whether PKC activation played a role in postconditioning's protection. As demonstrated in Fig. 3, chelerythrine blocked salvage of ischemic myocardium (42.6±3.0% infarction). Additionally wortmannin, an antagonist of PI3-K, also aborted postconditioning's protective effect (35.1±4.6% infarction).
It was reasoned that if postconditioning protects by activating A2b receptors then administration of an A2b agonist should mimic the protection. Fig. 4 shows that an infusion of the non-selective but A2b-potent adenosine receptor agonist NECA beginning just before reperfusion significantly decreased infarction to 17.2±2.7% of the risk zone (p<0.001 vs. control). As predicted by our hypothesis this protection was aborted when A2b receptors were blocked by MRS 1754, thus documenting the importance of A2b receptor occupancy which is presumably the first step in NECA's signaling. Chelerythrine had no effect when protection was initiated by NECA, thus excluding PKC as a necessary step in NECA's downstream signaling pathway.
Next we tried to activate cardiac PKC by directly infusing PMA into the rabbit just prior to reperfusion and found that it was equally protective as NECA (16.3±4.1% infarction) (Fig. 5). Chelerythrine blocked PMA's salutary effect, thus supporting our assumption that PMA protected by activating PKC. The A2b blocker MRS 1754 also blocked PMA's protection indicating that PKC activation acts to control adenosine's activation of the A2b receptor in the signaling cascade. Finally, wortmannin also abrogated PMA's protection.
None of the tool drugs administered at reperfusion had any independent effect on infarct size (Figs. 2 and 3).
These studies implicate involvement of the adenosine A2b receptor in postconditioning's signaling pathways in our rabbit model. Furthermore for the first time we provide evidence that PKC activation is upstream of the critical binding of adenosine A2b receptors. These observations provide a link back to PKC which for many years has been a pivotal kinase in the protection from ischemic preconditioning [12,13].
Although ischemic preconditioning introduced almost 20 years ago is a powerful anti-infarct intervention , it has not been possible to apply it in patients with acute myocardial infarction because of the requirement for pretreatment. Furthermore, restoration of blood flow to ischemic myocardium, a requisite for long-term survival of tissue, appears to hasten the transition from ischemic but viable to necrotic myocardium [15–18]. Zhao et al.  provided a new tool for protection of the heart from ischemia/reperfusion injury when they introduced the concept of postconditioning. Because of its timing postconditioning could be applied in a clinical setting of reperfusion by direct angioplasty, and indeed a recent clinical trial of postconditioning yielded encouraging results .
Because a highly selective A2b antagonist blocked postconditioning's protection we tested whether an A2b agonist would mimic the protection. Unfortunately a selective A2b agonist is unavailable. Although not selective for any adenosine receptor, NECA is a very potent A2b receptor agonist, and is frequently employed to activate this receptor subtype. As previously demonstrated in isolated rabbit hearts , NECA was quite protective in in situ hearts. Therefore, it was not surprising that blockade of A2b receptors with MRS 1754 aborted NECA's protection. Because PKC is known to play an important mediator role in preconditioning, and because a PKC antagonist blocked postconditioning's protection in the present study, we tested whether this intracellular kinase would also be important in NECA's signaling. To our surprise it was not, suggesting that PKC resides upstream of the step where the postconditioned heart employs A2b receptors in its protective pathway.
We pursued the above hypothesis by testing whether activating PKC with PMA would be protective. Intraatrial PMA just prior to reperfusion was equally protective as NECA. We excluded the possibility that PMA was protecting ischemic myocardium through some unknown, nonspecific effect by documenting abolition of PMA's salvage with the highly selective PKC antagonist chelerythrine. The observation that blockade of A2b receptors with MRS 1754 also prevented PMA from sparing ischemic myocardium corroborated our hypothesis that PKC resides upstream of the A2b step in this pathway. Therefore, we propose that PKC activation by postconditioning somehow facilitates the activation of protective signal transduction pathways that are coupled to the A2b receptor.
Kin et al.  recently reported that adenosine receptors were involved in a rat model of postconditioning, but they concluded that it was the A2a and A3 receptor subtypes that were important. AMP579, an adenosine A1/A2 agonist, is quite protective when delivered just prior to or at reperfusion, and its cardioprotection was not affected by the potent and selective A1 adenosine receptor antagonist DPCPX (Ki=10nM for A1 and 1440nM for A2 receptors), but was effectively abrogated by the adenosine A2a antagonist 8-(13-chlorostyryl) caffeine (CSC) which is 520-fold more selective for A2a than A1 receptors [20,21]. Yet we were unable to duplicate its protection with the highly selective A2a agonist CGS21680 . These observations led us to suspect the A2b receptor might be the important one since NECA, a very similar compound with known adenosine A2b agonist potency, mimicked AMP579's protection  (no A2b affinity data are available for AMP579). Kin et al.  used ZM241385 to block A2a receptors in their rat hearts. However, if the affinity of rat receptors for ZM241385 is similar to that of cloned human adenosine receptors , then nearly all A2b receptors would also have been blocked as well at the dose used by Kin et al. We were unable to find affinity data for ZM241385 against rat A2b receptors.
Affinity studies for MRS 1754 have been performed for human and rat, but unfortunately not for rabbit, adenosine receptors . The Ki for hA2b is 1.97nM which is more than 200-fold lower than that for either hA1 or hA2a receptors. However, the Ki for rat A1 receptors is 16.8nM, only 8.5 times that for hA2b receptors. Because of the possible collateral effect of MRS 1754 on A1 receptors in the rabbit, we specifically examined the result of DPCPX administration at a dose known to block the heart rate effects of a highly selective adenosine A1 agonist in rabbits [10,11]. Indeed, the A2b antagonist MRS 1754, but not the A1 antagonist DPCPX, aborted postconditioning's protection, strongly suggesting that cardioprotection from postconditioning in the open-chest rabbit requires occupancy of A2b receptors during reperfusion. Our observations with CSC exclude involvement of the A2a receptor. Unfortunately, there is no selective A3 antagonist in the rabbit. Interestingly, Kilpatrick et al.  have shown that DPCPX blocks cardioprotection triggered by A3 receptor stimulation in rabbit hearts. Because of the inability of DPCPX to block postconditioning's protection, a role for A3 receptors is doubtful.
Olanrewaju and Mustafa  have demonstrated in coronary artery endothelial cells that occupancy of both adenosine A2a and A2b receptors stimulates NO production and increases cGMP levels. We have already shown that cardioprotection triggered by NECA at reperfusion in vitro  as well as postconditioning  are dependent on NO. Mitochondrial KATP opening has also been implicated in postconditioning . Increased PKG activation resulting from an increased cGMP level would be expected, and we have recently reported that activated PKG can open mitochondrial KATP.
Tsang et al.  reported that PI3-K is an important signaling step in postconditioning in the isolated rat heart. We now confirm this observation in the in situ heart, and also show that PMA's protection is effectively blocked by wortmannin. Previously, we had noted that salvage triggered by NECA infused shortly before reperfusion was blocked by wortmannin . Furthermore, NECA increased Akt phosphorylation. Therefore, adenosine receptor occupancy results in PI3-K activation and Akt phosphorylation, although the intermediate steps are not yet known.
It is unclear why only the A2b receptors appear to protect the postconditioned heart. The untreated heart produces and releases adenosine during the prolonged coronary occlusion, and, therefore, one might wonder why these hearts are not protected. The A2b receptor is a very low-affinity receptor and the adenosine concentration even during ischemia may not be sufficient to populate this receptor to the degree required for protection. NECA, on the other hand, is a more potent agonist of the critical A2b receptor. Kin et al.  measured adenosine in the effluent of their postconditioned hearts and found it to be decreased. They proposed that the postconditioned heart might somehow be retaining adenosine in the tissues, but the mechanism for this is unprecedented. Perhaps postconditioning somehow activates pathways that increase the affinity of the receptors so that they can be occupied by the heart's endogenous adenosine. Indeed, PKC activation has been reported to increase A2b responses in some cell systems [27,28] and we now find that PKC activation is required for postconditioning to elicit protection. Although PKC may activate a myriad of kinases, an intriguing possibility is that it phosphorylates the A2b receptor to increase its sensitivity.
Inagaki et al. [29,30] have reported that specific inhibition of PKC-δ at reperfusion decreases infarct size. Our present observation that PKC activation at reperfusion is protective is obviously not consistent with that paradigm. Perhaps activation of all isoforms by PMA activates a previously unsuspected protective isoform, while inhibition of all isoforms with chelerythrine blocks some additional protective isoform resulting in the deleterious effect that we observed.
In conclusion, our study demonstrates for the first time that at least some protective strategies introduced shortly before reperfusion and ischemic postconditioning are acting through the adenosine A2b receptor. Furthermore, binding of the A2b receptor at reperfusion results in activation of at least one prosurvival kinase, PI3-K. Finally, PKC must also be activated during postconditioning to elicit protection, perhaps by increasing the sensitivity of the A2b adenosine receptor. Hence, PKC activation is upstream of A2b receptor occupancy.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.cardiores.2006.02.014.
This study was supported in part by grants HL-20468 and HL-50688 from the Heart, Lung and Blood Institute of the National Institutes of Health. S. Philipp was supported by a grant from the Deutsches Institut für Bluthochdruckforschung, Heidelberg, Germany.
- myocardial ischemia
- protein kinase c
- coronary occlusion
- adenosine receptors
- reperfusion therapy
- physiologic reperfusion
- protein kinase
- oryctolagus cuniculus
- transcriptional activation
- postconditioning, ischemic