In vivo adenosine receptor preconditioning reduces myocardial infarct size via subcellular ERK signaling

Easton A. Reid, Gentian Kristo, Yukihiro Yoshimura, Cherry Ballard-Croft, Byron J. Keith, Robert M. Mentzer Jr, Robert D. Lasley


The protective effects of adenosine receptor acute preconditioning (PC) are well known; however, the signaling mechanism mediating this effect has not been determined in in vivo models. The purpose of this study was to determine the role of the extracellular signal-regulated kinase (ERK) pathway in mediating adenosine PC in in vivo rat myocardium. Open-chest rats were submitted to 25 min of coronary artery occlusion and 2 h of reperfusion. ERK activation was assessed by measuring total and dually phosphorylated p44/42 ERK isoforms in nuclear and/or myofilament, mitochondrial, cytosolic, and membrane fractions. Adenosine receptor PC with the A1/A2a agonist 1S-[1a,2b,3b,4a(S*)]-4-[7-[[2-(3-chloro-2-thienyl)-1-methylpropyl]amino]-3H-imidazo[4,5-b]pyridyl-3-yl]cyclopentane carboxamide (AMP-579) reduced infarct size from 49 ± 3% to 29 ± 3%, an effect that was blocked by the mitogen-activated protein kinase-ERK inhibitor U-0126. ERK isoforms were present in all fractions, with the greatest expression in the cytosolic fraction and the least in the mitochondrial fraction. AMP-579 treatment increased preischemic p44/42 ERK phosphorylation in all fractions 2.7- to 6.9-fold. Reperfusion increased ERK isoform activation in all fractions, but there were no differences between control and AMP-579 hearts. Preischemic increases in phospo-p44/p42 ERK with AMP-579 were blunted by U-0126, although only in mitochondrial and membrane compartments. The PC effects of AMP-579 on infarct size and ERK were blunted by both the A1 antagonist 8-cyclopentyl-1,3-dipropylxanthine and, surprisingly, the A2a antagonist ZM-241385. These results indicate that the unique adenosine receptor agonist AMP-579 exerts its beneficial effects in vivo via both A1 and A2a receptor modulation of subcellular ERK isoform signaling.

  • myocardium
  • ischemia-reperfusion
  • compartmentation
  • adenosine

adenosine receptor activation before ischemia, also referred to as preconditioning (PC), protects the heart against reversible and irreversible ischemic injury in multiple species and preparations (13, 22). Despite significant evidence for adenosine receptor PC, there is little definitive information on its mechanism of action. Initial studies examined the modulation of protein kinase C (5, 9). The results of subsequent studies implicated the involvement of mitochondrial ATP-dependent K+ (KATP) channels, with some evidence that this signaling pathway involves protein kinase C isoforms (5, 14, 22).

Most recently, studies on the mechanisms of ischemic and pharmacological PC have focused on mitogen-activated protein kinases (MAPKs) (2, 20). MAPKs are a family of serine and threonine kinases that include extracellular signal-regulated kinase (ERK), p38 MAPK (p38), and c-Jun NH2-terminal kinase. The ERK MAPK family, which consists of p44 and p42 isoforms, is activated by various stimuli, including oxidative stress and several G protein-coupled receptor agonists (19). There are numerous reports of ERK activation in cardiac tissue (4, 6, 7, 15, 21, 23, 24, 26, 32, 3337), including in vivo evidence that ischemic PC and δ-opioid PC activate at least one ERK isoform (7, 24). Adenosine receptor subtypes, including the A1 and A2a receptors, have been reported to couple to the ERK pathway in both cardiac and noncardiac tissues (8, 25, 29). It has also been reported that the beneficial effects of adenosine A2a receptor activation during reperfusion appear to be due to ERK activation (11, 36). However, to date, there have been no studies examining whether adenosine receptor PC is mediated via ERK.

Although the results of a significant number of studies suggest that ERK activation during ischemia-reperfusion is beneficial (15, 24, 35, 36), there are additional reports suggesting otherwise (4, 23, 26, 33). The discrepancies in the exact role of ERK activation in myocardial ischemia-reperfusion may be due to the specific stimulus and the duration of ERK activation. Another factor contributing to the lack of consensus of the role of ERK MAPK may be the subcellular compartmentation of ERK isoforms. Ping et al. (24) observed that ischemic PC in conscious rabbits was associated with significant activation of ERK isoforms in nuclear, but not cytosolic, fractions. In contrast, Fryer et al. (7) reported that opioid receptor and ischemic PC in in vivo anesthetized rats increased ERK activation in cytosolic and nuclear fractions, but these authors concluded that only the increase in p44 isoform activity in the cytosol was involved in the mechanism of protection. Despite the key role for mitochondria in mediating the beneficial effects of PC, as well as contributing to reperfusion oxidative stress, and a report that ERK isoforms are present in murine heart mitochondrial fractions (3), there have been no reports examining mitochondrial ERK activation during PC or myocardial ischemia-reperfusion. The purpose of this study was to determine whether acute adenosine receptor agonist PC against in vivo myocardial infarction is due to the modulation of phospho-p44/p42 ERK in myocardial subcellular fractions.


All animals in this study received humane care according to the guidelines set forth in The Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 86-23, Revised 1996). In addition, animals were used in accordance with the guidelines of the University of Kentucky Institutional Animal Care and Use Protocol.

In vivo ischemia-reperfusion protocol.

Adult Sprague-Dawley rats weighing 350–399 g were used. Rats were anesthetized with ketamine-xylazine (60 mg/kg + 6–9 mg/kg ip) with supplemental doses of ketamine as needed. The right jugular vein was cannulated for fluid and drug administration; the right femoral artery was used for the measurement of blood pressure, heart rate, and blood gases. A tracheotomy was then performed, and the animals were connected to a small animal ventilator (model 683, Harvard Apparatus, South Natick, MA). Room air ventilation (with positive end-expiratory pressure) was supplemented with 100% O2. Respiration rate and tidal volume were adjusted to maintain normal arterial blood gases, which were determined every 30 min. Body temperature was monitored with a rectal temperature probe and maintained at 38°C with heating pads.

A median sternotomy was performed and the pericardium removed. A 6-0 prolene suture was then passed below the left coronary artery in the area immediately below the left atrial appendage. The ends of the suture were then fed through a short length of propylene tubing to form a snare. After the animals recovered from the surgical procedures for 30 min, the experimental protocols were initiated. Regional ischemia was induced by pulling up on the snare and clamping it onto the epicardial surface by using a small hemostat. Coronary artery occlusion was confirmed by epicardial cyanosis and a decrease in blood pressure. After 25 min of regional ischemia, the occlusion was released and the heart was reperfused for 2 h.

Study groups and experimental protocols.

Rats were randomly divided among eight groups. The control group (n = 8) received vehicle (10% DMSO in 1 ml normal saline iv) 30 min before ischemia. Adenosine receptor PC was induced by administering the agonist 1S-[1a,2b,3b,4a(S*)]-4-[7-[[2-(3-chloro-2-thienyl)-1-methylpropyl]amino]-3H-imidazo[4,5-b]pyridyl-3-yl]cyclopentane carboxamide (AMP-579, 50 μg/kg iv; n = 9) 30 min before ischemia. To confirm the involvement of the A1 receptor, another group of rats was given the A1 antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX, 100 μg/kg iv) 5 min before AMP-579 (n = 6). To determine A2a receptor involvement, additional rats were given the A2a antagonist ZM-241385 (1.5 mg/kg iv bolus) 15 min before AMP-579 (n = 7). The MAPK-ERK (MEK) inhibitor U-0126 (0.2 mg/kg iv) was administered 30 min before AMP-579 (n = 6) or vehicle (n = 6). Additional rats (n = 3 in each group) were treated with DPCPX or ZM-241385 30 min before ischemia to determine whether these agents alone altered infarct size.

Determination of infarct size.

After 2 h of reperfusion, the ligature at the coronary occlusion site was permanently tied off, and Evans blue solution (1%) was injected into the venous line to demarcate the left ventricular (LV) area at risk (AAR). The animal was then euthanized with a pentobarbital overdose, the heart was excised, and the atria and great vessels were removed. The heart was sliced into three to four pieces (>2 mm thickness) from base to apex for staining with triphenyltetrazolium chloride solution to measure infarct size as previously described (12). The AAR was devoid of Evans blue dye while the infarcted tissue within the AAR was the triphenyltetrazolium chloride -negative stained region. The areas were then quantified by computerized planimetry. Infarct size was expressed as a percentage of the AAR.

In vivo myocardial subcellular fractionation.

Four additional rat hearts per group were used to generate subcellular fractions 10 min after vehicle or AMP-579 treatment, after 15 min of ischemia, and at 10 min of reperfusion. For ischemia and reperfusion samples, the ischemic zone was demarcated with an epicardial 4-0 prolene suture around its perimeter. The tissue from both ischemic and nonischemic zones located within a 2- to 3-mm distance of this suture was discarded when samples were collected. On rapid excision, hearts were placed in ice-cold saline, and the ischemic and nonischemic zones were carefully separated. Tissue from the ischemic and nonischemic zones was homogenized with a polytron (Tekmar; Cincinnati, OH) in ice-cold homogenization buffer (250 mM sucrose, 20 mM HEPES, 10 mM KCl, 1.5 mM MgSO4, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 0.1 mg/ml phenylmethylsulfonyl fluoride, 45 μg/μl aprotinin, 0.5 mM β-glycerophosphate, and 1 mM sodium vanadate).

Differential centrifugation was used to generate subcellular fractions. For isolation of the nuclear-myofilament fraction, the heart homogenate was centrifuged at 750 g for 10 min. The 750-g supernatant was removed and centrifuged at 10,000 g for 10 min to generate the mitochondria-enriched fraction. The initial mitochondrial pellet was resuspended in homogenization buffer and centrifuged at 10,000 g for 10 min. This step was repeated a second time to further enrich the mitochondrial pellet. The initial 10,000-g supernatant was centrifuged at 100,000 g for 30 min to form the cytosol supernatant and membrane pellet. The subcellular fractions were aliquoted, frozen with liquid nitrogen, and stored at −80°C until analysis.

Western blot analysis.

Protein fractions (50–100 μg) were resolved on an 8% SDS-PAGE and transferred to nitrocellulose membranes. The membranes were then stained with Ponceau S to confirm equal protein loading. All membranes were blocked in 5% milk. Primary antibodies were incubated overnight (4°C), and secondary antibodies were diluted in 5% milk and incubated at room temperature (1 h). The phospho-ERK (pERK) primary antibody concentration was 1:800 in 5% IgG-free, protease-free BSA (Jackson Immunoresearch Laboratories, West Grove, PA). The horseradish peroxidase-goat anti-rabbit IgG (H + L) secondary antibody was diluted 1:6,000 (Zymed, San Fancisco, CA). Protein band antibody conjugates on membranes were detected with ECL chemiluminescence (Amersham, Piscataway, NJ). After pERK immunoreactivity was determined, the membranes were stripped and reprobed for total ERK. The total ERK primary antibody (1:1,000; Santa Cruz, CA) was placed in 5% milk. Immunoreactive band intensity was quantified with the SCION Image program.

Nuclear-myofilament, cytosol, membrane, and mitochondrial fraction purity was determined by α-actinin, histone deacetylase-1 (HDAC-1), caveolin-3, and cytochrome-c oxidase antibody markers, respectively. For each antibody marker, 15 μg of subcelluar fraction protein was used except for HDAC-1 where 30 μg was used. A positive control for pERK and total ERK was included on each blot to normalize the immunointensities across membranes. The positive control was a whole cell lysate protein sample generated from adult rat ventricular cardiomyocytes submitted to H2O2 (100 μM, 15 min, 37°C). Myocytes were isolated from one heart by collagenase perfusion as previously described (10).


AMP-579 was obtained from Aventis Pharmaceuticals (Bridgewater, NJ). U0126 was purchased from LC Laboratories (Woburn, MA), and DPCPX was purchased from Sigma-Aldrich (St. Louis, MO). 4-(-2-[7-Amino-2-{2-furyl}-{1,2,4}triazolo-{2,3-a}-{1,3,5}triazin-5-yl-amino]ethyl)phenol (ZM-241385) was acquired from Tocris (Ellisville, MI). All of these agents were initially dissolved in DMSO and then diluted in saline to reduce the final DMSO concentration to 10%.

Statistical analysis.

Results are expressed as means ± SE. Differences in infarct size were determined by one-way ANOVA, followed by Tukey's post hoc test. Effects of AMP-579 on hemodynamics were analyzed by two-way ANOVA with repeated measures. Western blot band intensity was measured using Scion image analysis software. Differences in preischemia band intensity were determined by unpaired Student's t-test. Ischemic and nonischemic bed band intensities from ischemia and reperfusion samples were evaluated using the paired Student t-test. Differences between groups were analyzed by unpaired Student's t-test. A P value <0.05 was considered statistically significant.


The hemodynamic data are summarized in Table 1. The adenosine agonist AMP-579 significantly decreased both heart rate and blood pressure, with these effects persisting 5 and 30 min, respectively. Pretreatment with the adenosine A1 receptor antagonist DPCPX blocked AMP-579-induced bradycardia but did not alter the hypotension. The adenosine A2a receptor blocker ZM-241385 did not affect the AMP-579 reduction in heart rate but blunted the decrease in blood pressure 30 min post-AMP-579. The MEK inhibitor U-0126 did not alter the hemodynamic effects of AMP-579. DPCPX, ZM-241385, and U-0126 alone did not alter hemodynamics.

View this table:
Table 1.

Hemodynamics during in vivo regional myocardial ischemia

Figure 1 shows the infarct size results. Preconditioning with AMP-579 decreased infarct size (29 ± 3% of the AAR) compared with vehicle treatment (49 ± 3%). Treatments with the MEK inhibitor U-0126 (42 ± 3%), the A1 receptor antagonist DPCPX (54 ± 2%), and the A2a receptor antagonist ZM-241385 (45 ± 2%) blocked the cardioprotective effect of AMP-579. None of the inhibitors alone altered infarct size. There were no differences in the ischemic AAR among the groups (range from 34 ± 3% to 44 ± 1%).

Fig. 1.

Left ventricle (LV) infarct size as a percentage of the area at risk (AAR). Hearts were submitted to 25 min regional ischemia by left coronary artery occlusion, the occlusion was released, and the heart was reperfused for 2 h. Infarct size was determined by triphenyltetrazolium chloride staining as a percentage of the AAR and AAR as the percentage of the LV. *P < 0.05 control group. See text for compound names.

Identification of the subcellular fractions is shown by the Western blot results in Fig. 2. The 750-g pellet contained both HDAC-1 and α-actinin, markers for nuclear membranes and myofilaments, respectively. The 10,000-g washed pellet was enriched in the mitochondrial marker cytochrome-c oxidase, which was absent in the 100,000-g supernatant (cytosol). Caveolin-3 was expressed in the membrane pellet, but not in the cytosol.

Fig. 2.

Determination of subcellular fraction purity. Western blot analysis was used to identify nuclear-myofilament (N/M), cytosol (C), membrane (Me), and mitochondrial (Mi) fractions by α-actinin, histone deacetylase-1 (HDAC-1), caveolin-3, and cytochrome-c oxidase antibody markers, respectively.

Figure 3 shows a representative Western blot (Fig. 3A) and averaged preischemia phospho-p44 (Fig. 3B) and phospho-p42 ERK isoform (Fig. 3C) band intensities in vehicle control and AMP-579-treated rats. As shown in Fig. 3B, treatment with the adenosine agonist increased nuclear-myofilament phospho-p44 intensity 5.5-fold (P = 0.0004), mitochondrial phospho-p44 2.7-fold (P = 0.01), cytosolic phospho-p44 3.8-fold (P = 0.002), and membrane phospho-p44 intensity 3.4-fold (P = 0.02). Figure 3C shows similar effects of AMP-579 on presischemic phospho-p42 ERK isoform immunoreactivity in cytosolic (3.0-fold, P = 0.03), mitochondrial (3.7-fold, P = 0.04), and membrane (4.3-fold, P = 0.03) fractions. The increase in the nuclear-myofilament fraction did not attain statistical significance.

Fig. 3.

Phospho-p44/p42 extracellular signal-regulated kinase (ERK) during preischemia. A: representative Western blot of preischemic ERK activation in vehicle and AMP-579-treated hearts. B: phospho-p44 ERK 10 min after AMP-579 administration relative to controls in the Nu/My, Mi, Me, and Cy fractions before ischemia (n = 4). C: phospho-p42 ERK in the Nu/My, Mi, Me, and Cy fractions during preischemia (n = 4). *P < 0.05 vs. control group.

The results in Fig. 4 show a representative Western blot (Fig. 4A) and averaged band intensities of total p44- and p42-ERK isoforms (Fig. 4B). There were no differences in the expression of the p44- and p42-ERK isoforms between the vehicle and AMP-579-treated animals. However, there were differences in the distribution of these isoforms among the subcellular fractions. The expression of both p44- and p42-ERK isoforms was greatest in the cytosol and least in the mitochondrial fraction. When the results of Figs. 3 and 4 were compared, the expression of total ERK isoforms in the mitochondrial fraction was nearly threefold less than in the cytosol; however, the phospho-ERK intensities in the two fractions were similar.

Fig. 4.

Relative density of subcellular p44/p42 total ERK Western blots during preischemia (n = 4). Total (phosphorylated and unphosphorylated) p44/p42 ERK immunoreactivity between control and AMP-579 groups

The effects of ischemia on ERK immunoreactivity are summarized in Fig. 5, A and B. There were no differences in phospho-p44 ERK between ischemic and nonischemic beds in any subcellular fraction, nor were there any differences between vehicle and AMP-579-preconditioned rats (Fig. 5A). The phospho-p42-ERK results shown in Fig. 5B revealed a statistically significant increase in the ischemic zone cytosolic fraction compared with the nonischemic bed in the AMP-579-treated animals (416 ± 127 vs. 137 ± 36 arbitrary units; P = 0.03). In vehicle control hearts, phospho-p42-ERK in the mitochondrial fraction of the ischemic bed was significantly reduced compared with the nonischemic bed. Ischemia did alter the distribution of p42 and p44 isoforms (data not shown).

Fig. 5.

Compartmental phospho-p44/p42 ERK immunodensity of Western blots during ischemia (n = 4; Nu/My n = 2). A: phospho-p44 ERK content in the subcellular fractions. B: phospho-p44 ERK content in the subcellular fractions. *P < 0.05 vs. nonischemic bed.

Reperfusion had no effect on the distribution of ERK isoforms (data not shown), but it was associated with changes in ERK phosphorylation. Cytosolic phospho-p44-ERK (Fig. 6A) was significantly increased in the ischemic zone relative to the nonischemic bed in both control (822 ± 142 vs. 232 ± 87 arbitrary units; P = 0.02) and AMP-579 groups (1,124 ± 279 vs. 321 ± 140 arbitrary units; P = 0.02). In the membrane fraction ischemic zone, phospho-p44-ERK was also increased in both groups, but this effect was significant only in the AMP-579 group. The phospho-p42 isoform increased in the ischemic zone relative to the nonischemic zone in all of the fractions in both groups, although the AMP-579 effect in the membrane fraction did not attain statistical significance (P value = 0.06).

Fig. 6.

Western blot subcellular phospho-p44/p42 ERK relative density during reperfusion (n = 4; Nu/My n = 2). A: phospho-p44 ERK in subcellular fractions for AMP-579 and control. B: phospho-p42 ERK in subcellular fractions for AMP-579 and control. *P < 0.05 vs. nonischemic bed.

Figure 7 displays the Western blots of subcellular fractions obtained from hearts treated with AMP-579 in the presence and absence of U-0126 (Fig. 7A), DPCPX (Fig. 7B), and ZM-241385 (Fig. 7C). The phospho-ERK band intensities in the mitochondrial and membrane fractions isolated from the U-0126 + AMP-579-treated hearts were reduced relative to AMP-579 alone. The cytosolic and nuclear-myofilament fractions from these same hearts did not reveal differences in band intensities. Treatments with DPCPX and ZM-241385 before AMP-579 PC showed decreased nuclear-myofilament and cytosolic band intensities compared with AMP-579 alone, but mitochondrial and membrane bands appeared unchanged. Total ERK expression in the various subcellular fractions was not altered by these treatments.

Fig. 7.

Western blot inhibitor studies with relative phospho-p44/p42 ERK immunodensity during preischemia. A: control (C), AMP-579 (A)-, and AMP-579 plus U-1026 (A+U)-treated hearts are shown in myocardial subcellular compartments. B: AMP-579 (A)- and AMP-579 plus DPCPX (A+D)-treated hearts are shown in myocardial subcellular compartments. C: AMP-579 (A)- and AMP-579 plus ZM-241385 (A+Z)-treated hearts are shown in myocardial subcellular compartments.


The results from this study demonstrate that in vivo myocardial adenosine receptor PC is mediated by phosphorylation (thereby activation) of p44- and p42-ERK isoforms before ischemia. This conclusion is based on the observations that: 1) AMP-579 treatment was associated with significant phosphorylation of ERK before ischemia, 2) the MEK inhibitor U-0126 blocked the AMP-579 infarct reduction, and 3) U-0126 also blunted AMP-579-induced ERK activation. Another novel aspect of our findings was the activation of ERK in mitochondrial and membrane subcellular fractions. The adenosine A1 receptor antagonist DPCPX reversed the infarct-sparing effects and the preischemic increase in ERK phosphorylation due to AMP-579, but, surprisingly, the adenosine A2a receptor antagonist ZM-241385 exerted similar inhibitory effects.

The protective effects of acute adenosine receptor cardioprotection have been reported in multiple species and preparations (13, 22); however, the mechanisms involved have not been unequivocally identified. This is the first study to demonstrate that adenosine receptor agonist acute PC is mediated via phosphorylation of p44/p42 ERK-MAPK. Our results indicate that AMP-579 treatment before ischemia stimulated p44/p42 ERK isoforms in nuclear-myofilament, mitochondrial, membrane, and cytosolic fractions three- to fivefold. Total p44/p42 ERK immunoreactivity in each fraction was not different among control and treatment groups, thus increased ERK activity with AMP-579 treatment did not appear to be due to translocation. Though there were no differences in the expression of the two ERK isoforms within each fraction, there were differences in the subcellular distribution of this kinase, with the least expression in the mitochondrial fraction and the greatest amount in the cytosol. Although AMP-579 treatment was associated with significant increases in subcellular p44/p42 ERK phosphorylation before ischemia, there were only two notable phospho-ERK effects during ischemia. Ischemic zone mitochondrial p42 ERK decreased 1.5-fold in control hearts, whereas AMP-579 ischemic bed cytosolic p42 ERK increased threefold. Reperfusion, however, showed increased ischemic bed phospho-p44 and -p42 ERK in all fractions, effects that were not altered by AMP-579.

An important aspect of our findings is the observation that although AMP-579 increased ERK activation in all subcellular fractions before ischemia, ERK isoform phosphorylation was blunted by U-0126 only in the mitochondrial and membrane fractions. In the only one other in vivo study examining the role of ERK in acute pharmacological PC, subcellular-dependent changes in ERK were also reported by Fryer et al. (7). These authors reported that δ-opioid receptor PC in rats increased nuclear fraction phospho-p44 and -p42 levels fourfold before ischemia. However, opioid receptor-induced increases in nuclear ERK activation were not blocked by the MEK inhibitor PD-098059. We observed similar increases with AMP-579 in our nuclear-myofilament fractions; however, ERK activation in this fraction did not appear to be consistently blocked by the MEK inhibitor U-0126. However, there are differences between our findings and those of Fryer et al. (7). Adenosine receptor, but not opioid receptor, PC was associated with increased ERK activity in preischemic cytosolic fractions. In both studies, reperfusion was associated with robust increases in nuclear ERK activity, but in the present study this activation was not altered by AMP-579 PC, whereas opioid receptor PC was associated with a further increase in ERK activity (7). Whether these dissimilarities are due to methodological differences or variations in opioid and adenosine receptor PC remain to be determined.

Our present findings, and those of Fryer et al. (7), indicate that a complete understanding of MAPK signaling during myocardial ischemia-reperfusion requires the examination of subcellular fractions. Given the important role of both mitochondria and the plasma membrane in the modulation of ischemia-reperfusion injury, we felt it was important to examine these fractions for ERK activity. Baines et al. (3) reported the presence of ERK in murine heart mitochondrial fractions, but ERK isoform activation in this compartment during myocardial ischemia-reperfusion has not been previously studied. Treatment with AMP-579 was associated with a threefold increase in phospho-ERK isoforms before ischemia. More importantly, we observed that this increase in mitochondrial ERK activation before ischemia was blunted with the MEK inhibitor U-0126. These results are consistent with reports linking mitochondrial reactive oxygen species, mitochondrial KATP channels, and ERK in cardioprotection (35).

Another novel aspect of our findings is the significant expression and activation of ERK isoforms in the 100,000-g membrane fraction. Adenosine agonist treatment increased phospho-ERK levels greater than threefold before ischemia, and in both control and AMP-579 groups reperfusion was associated with ERK activation. Finally, the preischemic increase in phospho-ERK in this fraction with AMP-579 treatment was blunted by U-0126, implicating the importance of this compartment in adenosine PC. Although not well recognized, there are reports (16, 27) of ERK expression and activity in myocardial membranes, with this kinase possibly being compartmentalized in caveolae. In contrast to our findings, Ping et al. (24) reported no ERK expression in their myocardial membrane fractions. However, these authors examined only detergent-solubilized membrane fractions, whereas we used no detergents.

The adenosine A1 receptor antagonist DPCPX blocked AMP-579 PC and blunted preischemic ERK phosphorylation, consistent with the hypothesis that adenosine receptor PC is mediated via the A1 receptor (13, 22). There are also reports that adenosine A1 receptor activation increases ERK activity in several tissues, including neonatal cardiac myocytes (8, 25, 29). However, an unexpected finding in this study was that the adenosine A2a antagonist ZM-241385 also blocked AMP-579 PC and blunted preischemic ERK phosphorylation. Although there are reports that adenosine A2a receptor stimulation can couple to ERK (8, 25, 29, 30), the selective adenosine A2a receptor agonist CGS-21680, administered before ischemia, does not protect the ischemic heart (5, 13, 22).

An explanation for these surprising findings may be due to the fact that AMP-579 is not a selective A1 agonist but rather an adenosine receptor agonist that has high affinities for both A1 [inhibitory constant (Ki) = 5 nM] and A2a (Ki = 56 nM) receptors (31). Several other investigators (11, 34, 37) have concluded that cardioprotection with AMP-579 reperfusion treatment is mediated via A2a receptor activation (and ERK activation) based on blockade by the A2a antagonist ZM-241385. Despite this support for A2a receptor involvement, the selective A2a agonist CGS-21680 was unable to mimic the protective effects of AMP-579 (11, 34, 37). Thus our present findings may be due to the unique effects of AMP-579 to simultaneously activate both A1 and A2a receptors.

An alternative explanation for the apparent dual receptor-mediated effects of AMP-579 is the possibility that there is an interaction between A1 and A2a receptors. Lopes et al. (18) reported that although ZM-241385 did not significantly inhibit radioligand binding to A1 receptors, it did block A1 receptor-mediated electrophysiological effects in rat hippocampal neurons. A subsequent study by the same authors indicated crosstalk between A2a and A1 receptors in the same tissue (17). Abe et al. (1) reported that neural stunning of symapathetic activation in the coronary circulation in in vivo canine myocardium was due to the combined effects of A1 and A2a receptor activation. Whether the effects of AMP-579 PC are due to dual receptor activation or a unique effect of this agonist remain to be determined.

In summary, the results of this study provide the first evidence for the role of ERK signaling in in vivo myocardial adenosine receptor PC. Although AMP-579 treatment increased preischemic p44/42 ERK phosphorylation in nuclear-myofilament, mitochondrial, cytosolic, and membrane fractions, these increases were blunted by U-1026 only in mitochondrial and membrane compartments. The PC effects of AMP-579 on infarct size and ERK were blunted by both the A1 antagonist DPCPX and, surprisingly, the A2a antagonist ZM-241385. These results indicate that the unique adenosine receptor agonist AMP-579 exerts its beneficial effects in vivo via both A1 and A2a receptor modulation of subcellular ERK isoform signaling.


This work was supported by National Heart, Lung, and Blood Institute Grants R01 HL-34759 (to R. M. Mentzer, Jr.) and R01 HL-66132 (R. D. Lasley). E. A. Reid is the recipient of an National Institutes of Health National Research Service Award (F32 HL-075998).


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract