Following myocardial infarction (MI), contractile dysfunction develops not only in the infarct zone but also in noninfarcted regions of the left ventricle remote from the infarct zone. Inflammatory activation secondary to MI stimulates inducible nitric oxide synthase (iNOS) induction with excess production of nitric oxide. We hypothesized that the anti-inflammatory effects of selective A2A-adenosine receptor (A2AAR) stimulation would suppress inflammation and preserve cardiac function in the remote zone early after MI. A total of 53 mice underwent 60 min of coronary occlusion followed by 24 h of reperfusion. The A2AAR agonist (ATL146e, 2.4 μg/kg) was administered intraperitoneally 1, 3, and 6 h postreperfusion. Because of the 1-h delay in treatment after MI, ATL146e had no effect on infarct size, as demonstrated by contrast-enhanced cardiac MRI (n = 18) performed 24 h post-MI. ATL146e did however preserve global cardiac function at that time by limiting contractile dysfunction in remote regions [left ventricle wall thickening: 51 ± 4% in treated (n = 9) vs. 29 ± 3% in nontreated groups (n = 9), P < 0.01]. RT-PCR, immunohistochemistry, and Western blot analysis indicated that iNOS mRNA and protein expression were significantly reduced by ATL146e treatment in both infarcted and noninfarcted zones. Similarly, elevations in plasma nitrate-nitrite after MI were substantially blunted by ATL146e (P < 0.01). Finally, treatment with ATL146e reduced NF-κB activation in the myocardium by over 50%, not only in the infarct zone but also in noninfarcted regions (P < 0.05). In conclusion, A2AAR stimulation after MI suppresses inflammatory activation and preserves cardiac function, suggesting the potential utility of A2AAR agonists against acute heart failure in the immediate post-MI period.
- magnetic resonance imaging
- nitric oxide synthase
- nuclear factor-κB activation
- acute heart failure
- left ventricular function
early after a large myocardial infarction (MI), regional contractile dysfunction develops not only at the infarct site but also in nonischemic regions of the left ventricle (LV) remote from the infarcted territory. With the use of cardiac MRI, this remote LV dysfunction has been demonstrated in patients shortly after reperfused MI (4, 15, 16). Remote LV dysfunction has also been described in canine (20, 33), ovine (14), and murine (10, 25) models of coronary occlusion using a variety of techniques. Because acute heart failure poses a major threat to patient survival immediately after large transmural MI (5), therapies aimed at preserving LV function in this setting may have clinical potential.
Myocardial ischemia/reperfusion injury is associated with an acute inflammatory process accompanied by the release of cytokines that can depress myocardial contractility. The inflammatory process may be beneficial in initiating tissue repair and scar formation, but it is also known to extend myocardial injury (6). The elevated production of nitric oxide (NO) by local induction of inducible nitric oxide synthase (iNOS) has been reported to contribute to myocardial injury and dysfunction in a variety of clinical settings (13). Furthermore, selective inhibition or abrogation of iNOS has been shown to improve global LV performance and long-term survival following experimental MI (11, 24, 34, 36, 37). Although tissue necrosis and proinflammatory responses have been linked to functional impairment in the infarct-related myocardium, the potential impact of inflammatory activation on contractile function in remote, nonischemic regions of the LV has not been fully explored.
The present study was undertaken to investigate the role of inflammatory activation and iNOS induction in the genesis of remote LV dysfunction early after large MI. ATL146e (ATL) is a potent and selective A2A-adenosine receptor (A2AAR) agonist that displays rapid anti-inflammatory properties in a variety of in vitro and in vivo models (12, 19, 22, 24). Accordingly, we tested the anti-inflammatory effects of subvasoactive doses of ATL in a mouse model of reperfused MI using in vivo cardiac MRI to assess both myocardial infarct size and regional LV function. The mechanisms of action of ATL on remote LV dysfunction early after MI were then investigated with emphasis on the potential involvement of iNOS.
Mouse model of MI.
This study conformed to the “Guide for the Care and Use of Laboratory Animals” (NIH publication 85-23, 1985) and was conducted under protocols approved by the Institutional Animal Care and Use Committee. A standard myocardial ischemia-reperfusion protocol was employed, as described previously (38). Mice were anesthetized with intraperitoneal injection of pentobarbital sodium (100 mg/kg) and intubated. Artificial respiration was maintained with a rodent ventilator. A para-sternal incision was made, and coronary occlusion was imposed by passing a 7-0 silk suture beneath the left anterior descending artery just inferior to the left auricle and then tightening it over a length of PE-20 tubing. Sixty minutes later, reperfusion was achieved by removing the PE-20 tubing, and the chest was closed in layers. Core body temperature was monitored with a rectal probe and maintained at 37.0 ± 0.5°C with a heat lamp throughout surgical procedure.
Male C57Bl/6 mice (10–12 wk old) were purchased from Hilltop Lab Animals (Scottdale, PA) and subjected to sham operation or 60 min of coronary occlusion followed by 24 h of reperfusion. Intraperitoneal injections of A2AAR agonist (ATL) were performed at 1, 3, and 6 h postreperfusion in the MI+ATL group (n = 27) at a dose of 2.4 μg/kg. We have previously determined that the intraperitoneal injection of ATL at a dose of 10 μg/kg has no significant effects on LV hemodynamics in mice (39). An equal volume of vehicle (saline) was similarly injected in the MI+saline group (n = 26). Sham-operated animals underwent the same procedure without coronary occlusion (n = 16). Twenty-four hours after reperfusion, in vivo cardiac MRI was performed on the MI+saline and MI+ATL groups (n = 9/group) to determine LV mass, infarct size, LV chamber volumes, global cardiac function, and regional wall thickening (38). The remaining mice were euthanized 24 h after reperfusion, and hearts were excised for use in protein and mRNA assays. For immunohistochemistry (n = 3/group), hearts were fixed in 3.7% paraformaldehyde and embedded in paraffin. For the analysis of mRNA (n = 5–6/group) and protein (n = 6/group), tissue samples taken from infarcted and noninfarcted regions of the LV were snap-frozen for storage at −70°C. For the analysis of NF-κB activation (n = 3/group), nuclear protein extracts were prepared 24 h postreperfusion from freshly isolated infarcted and noninfarcted regions of the LV.
Cardiac MRI was performed at baseline (before surgery) and 24 h after reperfusion (post-MI) using a Varian Inova 4.7T magnetic resonance (MR) scanner with Magnex gradients as previously described (38). In preparation for scanning, mice were sedated with diazepam (10–15 mg/kg ip) and fitted with ECG electrodes for cardiac gating. For contrast-enhanced imaging on day 1 post-MI, 0.3–0.6 mmol/kg Gd-diethylenetriamine pentaacetic acid (DTPA) was injected intraperitoneally 20 min before imaging. During imaging, core body temperature was monitored with a rectal probe and maintained at 37.0 ± 0.5°C by circulating warm water within the birdcage RF coil.
Determination of LV mass and global cardiac function.
Cardiac MR images were processed using the ARGUS cardiac analysis package (Siemens Medical Systems, Iselin, NJ). After the endocardial borders were planimetered, the end-diastolic and end-systolic phases were identified for determination of LV end-systolic volume, LV end-diastolic volume, stroke volume, and LV ejection fraction. Heart rate, as detected by ECG surface electrodes on the two forelimbs, was used to calculate cardiac output (cardiac output = stroke volume × heart rate). Epicardial contours were also traced at the end-diastolic and end-systolic phases, and LV mass was calculated as described previously (25).
Determination of infarct size.
Infarct size was determined from MR images obtained 24 h after reperfusion as described previously (38). Briefly, areas of delayed hyperenhancement (defined as regions with signal intensity >2 SD above the mean of the remote region) were planimetered on contiguous, 1-mm-thick, end-diastolic, short-axis images, and total infarct size was expressed as percentage of the total LV mass.
Determination of segmental LV wall thickening.
Analysis of regional contractile function was determined as previously described (38). Briefly, two contiguous, 1-mm-thick, short-axis slices acquired from the midventricular level of each heart were divided into eight equal radial segments indexed vertically to the apex and radially to the anterior right ventricular insertion point. The area thickness of each segment was then computed, and absolute wall thickening and percent wall thickening were calculated from the corresponding segment thicknesses at the end-diastolic and end-systolic phases. The segments were then classified into three categories based on the Gd-enhanced images acquired 24 h postreperfusion. Segments were classified as “infarcted” if >50% of the LV wall in that segment was enhanced by contrast agent. Segments immediately adjacent to infarcted segments were classified as “adjacent,” and the remaining segments were categorized as “remote” (38). For the baseline MR studies, the segments corresponding to infarcted, adjacent, and remote segments were identified by first classifying the segments on the contrast-enhanced MR images acquired 24 h postreperfusion and then mapping the classifications back onto the images acquired at baseline.
Real-time RT-PCR analysis.
Total cellular RNA was extracted from myocardial tissue samples using TRIZOL reagent (Invitrogen Life Technologies, Carlsbad, CA) according to the manufacturer's recommendations. One microgram of total RNA was reverse transcribed into cDNA using Superscript II (Invitrogen Life Technologies), and samples were then diluted 1:5 with sterile water. Real-time PCR experiments were performed using an I-Cycler IQ system (Bio-Rad Laboratories, Hercules, CA) and the QuantiTect SYBR green PCR kit (Qiagen, Valencia, CA). Primer sequences for TNF-α, iNOS, and GAPDH were as previously described (23). Each real-time PCR reaction (50 μl total volume) contained 25 μl of PCR Master Mix, 20 μl of primers (0.3 μmol/l each), and 5 μl of diluted cDNA template. Serial dilutions of cloned TNF-α, iNOS, or GAPDH plasmid cDNA (23) were run alongside the test samples. A melting curve analysis was performed at the end of each PCR reaction. PCR products from the test samples and cDNA standards were further analyzed by electrophoresis on a 2% agarose gel to verify the size and specificity of the amplicons. Negative control reactions consisted of appropriately diluted, nonreverse-transcribed RNA extracts. Transcript abundance was quantitated using gene-specific standard curves. Minor variations in cDNA quality and in total input RNA were corrected by normalizing the results to the housekeeping gene GAPDH. Equivalent results were obtained when the data were normalized to the housekeeping gene hypoxanthine phosphoribosyltransferase (data not shown).
Western blot analysis.
Frozen tissue samples were homogenized, and equal amounts of protein (70 μg/well) were electrophoresed and then transferred onto polyvinyldine difluoride membranes. The membranes were incubated overnight at 4°C with anti-iNOS rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA) followed by 1-h incubation with donkey anti-rabbit IgG (Pierce, Rockford, IL) at room temperature. After film was exposed to the chemiluminescent signal (SuperSignal West Pico, Pierce), optical densitometry was performed. Results were expressed relative to the average signal intensity obtained from sham-operated hearts.
Immunostaining for iNOS protein and myoglobin was performed on serial 5-μm paraffin sections. After incubation with hydrogen peroxide (0.5%) followed by avidin blocking, the sections were incubated overnight at 4°C with goat anti-iNOS antibody (Santa Cruz Biotechnology) or rabbit anti-myoglobin (DAKO, Carpinteria, CA). Biotinylated secondary antibodies (Vector Laboratories, Burlingame, CA) were then applied for 1 h at room temperature. After incubation with avidin-biotin complex (Vector Laboratories), immunoreactivity was visualized by incubating the sections with the chromogen 3,3-diaminobenzidine tetrahydrochloride (DAKO) to produce a brown precipitate. Immunostained sections were counterstained with hematoxylin before they were coverslipped for photography.
Nitrate and nitrite assay.
After euthanasia before removal of the hearts at 24 h postreperfusion, blood samples (n = 6/group) were withdrawn from the right ventricle and plasma samples were prepared for nitrate + nitrite (NOx) assay. Plasma nitrate was first reduced to nitrite by incubation of 50 μl plasma with 25 μl nitrate reductase (600 mU/ml) and 25 μl NADPH (160 μmol/l) at 37°C for 90 min. NOx levels were then assayed using a modified Griess reagent (Szechrome NIT; Polyscience, Warrington, PA) according to the manufacturer's recommendations. Briefly, plasma samples and sodium nitrite standards were incubated with 50 μl of NIT reagent for 10 min. Absorbance was read at 542 nm, and plasma NOx levels were expressed in micromoles per milliliter.
Electrophoretic mobility shift assay.
EMSA was performed according to standard methods (9). Briefly, double-stranded oligonucleotides 5′-AGTTGAGGGGACTTTCCCAGGC-3′ (Santa Cruz Biotechnology) contained a consensus binding site for NF-κB. Nuclear protein extracts (0.8 μg) from noninfarcted and infarcted regions were incubated with of 32P-labeled oligonucleotides (20 fmol) for 20 min in 10 mM Tris (pH 7.9), 50 mM KCl, 1.0 mM EDTA, 1 mM DTT, 4 μg poly d(I-C), 100 μg/ml BSA, 0.1% IGEPAL CA 630, and 10% glycerol. The protein-DNA complexes were resolved by electrophoresis through 6% acrylamide gels in 0.5× Tris-borate EDTA (TBE). Specificity of the protein-DNA complex was confirmed by competition with a 200-fold excess of unlabeled probe. Gels were fixed, dried, visualized by autoradiography, and quantitated by densitometric scanning.
Data are presented as means ± SE. Comparisons were performed using Systat software (SPSS, Chicago, IL). Comparisons were performed using one-way ANOVA. When appropriate, post hoc testing was performed using the Bonferroni test. P values <0.05 were considered statistically significant.
MRI measurement of cardiac mass, infarct size, and global function.
Table 1 summarizes the structural and global functional parameters measured by cardiac MRI. Compared with baseline, LV mass was significantly increased after MI in both MI+saline and MI+ATL groups (P < 0.01 vs. baseline). We have previously established that treatment of mice with ATL at the time of reperfusion after coronary occlusion can reduce infarct size by >30% (39). However, in this series of experiments, treatment with ATL was intentionally delayed for 1 h after reperfusion. As determined by contrast-enhanced MRI, there was no statistical difference in infarct size between MI+saline and MI+ATL groups [33 ± 2% vs. 34 ± 1%, respectively; P = not significant (NS)]. However, the deficit in LV global function observed in the MI+saline group 24 h after reperfusion was significantly attenuated by delayed ATL treatment. The improvement in cardiac function attributable to ATL was evident both in measurements of LV ejection fraction (MI+saline: 25.5 ± 1.7% vs. MI+ATL: 38.4 ± 1.5%; P < 0.01) and cardiac output (MI+saline: 5.2 ± 0.7 μl/min vs. MI+ATL: 8.0 ± 0.5 μl/min; P < 0.01).
MRI measurement of regional function.
As shown in Fig. 1, LV wall thickening was markedly decreased throughout the heart in MI+saline mice 24 h post-MI (P < 0.01 vs. baseline). In mice treated with ATL, LV wall thickening remained depressed in the infarcted and adjacent zones (P < 0.01 vs. baseline); however, percent wall thickening in the remote zone of ATL-treated mice was significantly improved compared with the MI+saline group (51 ± 4 vs. 29 ± 3% respectively; P < 0.01). More notably, contractile function in the remote myocardium of ATL-treated mice was completely preserved relative to baseline measurements taken before MI (P = NS vs. baseline). A brief period of highly selective A2AAR stimulation early after reperfusion thus sufficed to improve global cardiac function 24 h post-MI by preserving contractile function in regions of the LV remote to the site of infarction.
TNF-α and iNOS mRNA levels.
Myocardial abundance of transcripts encoding the cytokine TNF-α and the inflammatory mediator iNOS were measured by real-time RT-PCR (Fig. 2). As illustrated in Fig. 2A, there was a sevenfold increase in TNF-α mRNA in the infarcted region after 60 min of occlusion and 24 h of reperfusion compared with sham-operated animals (P < 0.01). MI also induced a small but significant increase in TNF-α mRNA in the noninfarcted region (MI+saline vs. sham; P < 0.05). ATL treatment significantly decreased TNF-α mRNA in the ischemic region (P < 0.05 vs. MI+saline). In the noninfarcted zone, TNF-α transcript levels were not significantly reduced by ATL compared with MI+saline animals (P = NS). Figure 2B shows that MI induced an increase in the abundance of iNOS mRNA in both the infarcted and noninfarcted regions compared with sham-operated animals (MI+saline vs. sham, P < 0.01 and P < 0.05, respectively). In the ATL-treated animals, the induction of iNOS mRNA was significantly attenuated in infarcted as well as noninfarcted remote regions of the LV compared with saline-treated animals (MI+ATL, P < 0.01 and P < 0.05 vs. MI+saline, respectively). In particular, ATL treatment reduced iNOS transcript levels in the noninfarcted region to a level comparable to that found in sham-operated animals.
Effects of MI on iNOS protein expression.
In agreement with the changes observed in iNOS mRNA levels, Western blot analysis indicated marked increases in the 130-kDa iNOS protein 24 h post-MI, not only in infarcted regions but also in noninfarcted regions of the heart (Fig. 3A). The results of quantitative densitometry indicated a 15-fold induction of iNOS in infarcted regions (P < 0.01 vs. sham) and a 10-fold induction of iNOS in noninfarcted regions (P < 0.05 vs. sham) of the myocardium. Transient treatment with ATL dramatically attenuated the accumulation of iNOS protein in the infarcted area (P < 0.05 vs. sham or vs. MI+saline) and completely normalized iNOS protein levels in remote regions of the LV (P < 0.05 vs. MI+saline, P = NS vs. sham).
Immunohistochemisty for iNOS and myoglobin.
Immunohistochemistry for iNOS and myoglobin was performed in serial sections to define the spatial extent of iNOS protein expression relative to the infarct region. An antibody recognizing myoglobin was used to define viable tissue, since this protein is quickly lost from lethally compromised cardiomyocytes. As illustrated in Fig. 4E, immunoreactivity for iNOS was readily detected in the murine hearts 24 h post-MI. In the infarcted region (Fig. 4E1), diffuse iNOS immunostaining was observed in the necrotic area, and strong immunoreactivity was found in spared cardiomyocytes adjacent to infarcted tissue. Cardiomyocytes in noninfarcted regions of the heart also displayed robust immunoreactivity for iNOS (Fig. 4E2). As shown in Fig. 4F, treatment with ATL dramatically decreased the amount of iNOS immunoreactivity present, although low levels of iNOS signal could still be detected in infarcted and noninfarcted areas of the LV (Figs. 4F1 and 4F2, respectively). Essentially no iNOS immunoreactivity was detected in hearts from sham-operated animals (Fig. 4D).
Plasma NOx levels.
Plasma NOx levels were used as a composite indicator of systemic iNOS activity and NO production. As shown in Fig. 5, when compared with iNOS activity in sham-operated animals (10.9 ± 1.1 μmol/l), mice subjected to MI showed significantly higher levels of plasma NOx (27.4 ± 5.3 μmol/l; P < 0.01 vs. sham). The increase in plasma NOx attributable to MI was completely normalized by brief treatment with ATL (8.1 ± 1.3 μmol/l; P < 0.01 vs. MI+saline). Moreover, in treated animals, NOx levels were not significantly different from those measured 24 h after sham surgery (P = NS vs. sham).
Effects of MI on NF-κB activation.
As anticipated from the changes observed in iNOS mRNA levels, EMSA analysis indicated marked activation of NF-κB 24 h post-MI, not only in infarcted regions but also in noninfarcted regions of the heart (Fig. 6A). Although nuclear-localized NF-κB activity could barely be detected in sham-operated animals, myocardial NF-κB activity was readily detected 24 h post-MI (Fig. 6A) in both infarcted and noninfarcted regions of the heart. The results of quantitative densitometry (Fig. 6B) indicated a 20-fold increase in myocardial NF-κB activity in noninfarcted myocardium and a 40-fold increase in myocardial NF-κB activity in the infarcted myocardium compared with sham-operated hearts (both P < 0.05 vs. sham). Transient treatment with ATL dramatically attenuated NF-κB activation, in both the infarcted and noninfarcted regions. Compared with saline-treated controls, myocardial NF-κB activity was reduced by >50% in the infarct zone and by >80% in noninfarcted zones (P < 0.05 for both comparisons; Fig. 6B).
The present study investigated the role of inflammatory activation in the genesis of remote LV dysfunction early after MI. The major findings of this study were that 1) following large transmural MI, specific stimulation of A2AARs using ATL significantly improved global LV function 24 h post-MI by preventing the contractile dysfunction that normally afflicts noninfarcted remote regions of myocardium, 2) the protective effects of ATL were associated with an inhibition of myocardial iNOS induction and NO production, and 3) the protective effects of ATL were associated with an inhibition of myocardial NF-κB activation.
Cardiac MRI showed that, 24 h post-MI, regional LV function was depressed not only in the infarcted region but also in remote, nonischemic regions of the LV (Fig. 1). LV wall thickening in the infarcted regions of both treated and untreated mice approached zero but did not enter negative values (indicating that dyskinesis in the form of systolic wall thinning was not a major contributor to the decreases observed in global LV function; Table 1). Wall thickening in the adjacent regions was also depressed to a similar extent in both groups, as might be expected due to tethering effects. In contrast, a significant difference between groups was observed in the remote region, where ATL treatment was effective in preserving baseline levels of wall thickening (Fig. 1). The transient and reversible forms of contractile dysfunction that afflicted remote regions of the LV in untreated mice have been described in a variety of animal models (10, 14, 20, 25, 33) and detected by MR tagging studies in patients early after MI (4, 15–16). Remote zone dysfunction may contribute importantly to acute heart failure in patients immediately after large, anterior MI. Hence, strategies for preserving LV function may well enhance patient survival in this setting. However, the mechanisms underlying contractile dysfunction in the remote LV remain the subject of debate. For example, cardiac MR tagging studies indicate that mechanical stress contributes importantly to contractile dysfunction in the remote LV early after MI (4). It would seem, however, that, if mechanical stress were directly and solely responsible for remote LV dysfunction, then the resulting dysfunction would be entirely dependent on the size and location of MI and totally unresponsive to anti-inflammatory therapy. The efficacy of the potent anti-inflammatory agent (ATL) in the present study indicates that inflammatory activation contributes importantly to contractile dysfunction in the remote LV early after large MI. It remains entirely plausible that mechanical stress contributes to inflammatory activation in the pathway leading from MI to remote LV dysfunction.
The stimulation of A2AARs during ischemia (or early during reperfusion) has consistently been shown to reduce infarct size in animal models, and the inhibition of proinflammatory responses is the primary mechanism underlying this cardioprotective effect (6, 17, 21, 29, 31, 39). Moreover, the specific A2AAR agonist ATL has previously been shown to protect a wide variety of organ systems (lung, vasculature, and kidney) against ischemia-reperfusion injury (12, 19, 22). In contrast, the present study focused on the preservation of contractile function after MI. Thus the administration of ATL was intentionally delayed until 1 h after reperfusion to prevent the agonist from reducing infarct size. The application of contrast-enhanced MRI was critical in demonstrating that infarct size was indeed equivalent in both the MI+saline- and MI+ATL-treated groups at 24 h post-MI. As assessed by segmental LV wall thickening analysis, ATL did not alter contractile function in infarcted or adjacent regions of the LV at this time point after MI, but it did significantly improve regional contractile function in myocardial segments remote from infarcted segments. Thus the beneficial effects of A2AAR stimulation on global cardiac performance early after MI were largely attributable to the attenuation of remote LV dysfunction. In this study, ATL was used at a subvasoactive dose that has no significant effects on LV hemodynamics in mice (39). Furthermore, the period of time between the last injection of the A2AAR agonist and the MRI assessment of myocardial function (18 h) is many times longer than the known half-lives of adenosine or ATL activity in vivo. We can therefore exclude the possibility that the long-term effects of ATL on cardiac function were somehow attributable to direct hemodynamic effects of the drug.
Recent studies have shown that rodent blood rapidly metabolizes ATL146e to ATL146a and that this form of ATL has equal potency for the A2AAR and A3AR subtypes (8). Although the present study does not explicitly exclude the possibility of A3AR involvement, we believe that this is unlikely in light of evidence showing that A3AR stimulation is actually proinflammatory, given that it induces both iNOS expression and NF-κB activation in the heart (40). Furthermore, previous work from our laboratory used both A2AAR knockout mice and a selective A2AAR antagonist (ZM241385) to confirm that, in this same mouse model of reperfused coronary occlusion, the cardioprotective effects of ATL146e against MI were directly attributable to the stimulation of A2AAR, as opposed to A3AR (39).
Production of large amounts of NO via iNOS induction has been studied in a wide variety of proinflammatory settings, including LPS or cytokine stimulation, experimental models of ischemia-reperfusion injury, and human cardiac disease states (13). These studies have shown that increases in iNOS expression and activity are associated with myocardial dysfunction and/or cell death (1–3, 27). There are a number of molecular mechanisms by which the unregulated production of NO by iNOS could negatively impact cardiac function. These include disruption of the endogenous NO signaling pathways responsible for modulating basal myocardial contractility and direct protein modification through S-nitrosylation of specific thiol residues and nitration of tyrosine residues as a result of peroxynitrite formation (7). In the present work, which used a mouse model of reperfused MI, LV contractile dysfunction was associated with marked increases in myocardial iNOS mRNA and protein, as shown by real-time RT-PCR, Western blot analysis, and immunohistochemistry. The resulting increase in NO production was of such a magnitude that a 2.5-fold systemic elevation in the end products of NO metabolism (NOx) could be detected in the bloodstream. Conversely, the salutary effect of ATL treatment on post-MI LV function was associated with a dramatic inhibition of NF-κB activation and iNOS expression in both the infarcted and noninfarcted (remote) regions of the LV, indicating that the anti-inflammatory effects of ATL were responsible for preserving LV function early after MI. These findings are entirely consistent with previous studies reporting that inhibition of iNOS with S-methylisothiourea or aminoguanidine improves global LV performance post-MI (34, 36, 37).
In a rabbit model of permanent coronary occlusion, Wildhirt et al. (34, 35) identified macrophages in the infarct-related myocardium as the major source of iNOS expression after acute MI. However, Takimoto et al. (30) reported that, after nonreperfused MI, iNOS expression was spatially and temporally regulated with early increases in mRNA and protein levels in the infarcted area and a delayed induction of the enzyme in the noninfacted region starting 2 wk after MI.
In the present study, iNOS expression was dramatically increased in both the infarcted and noninfarcted LV as early as 1 day after reperfused MI. Reperfusion is known to accelerate and to potentiate the proinflammatory responses initiated by ischemia. Thus spatial and temporal differences in iNOS expression between this and previous studies may largely be because the present study was undertaken in a reperfused model of transient coronary occlusion, whereas the aforementioned studies involved permanent coronary occlusion.
Interestingly, macrophages were not identified as the exclusive source of iNOS expression in the present study. Indeed, immunohistochemistry confirmed that iNOS was markedly expressed by cardiomyocytes throughout the heart, including those in the noninfarcted remote LV (a region that is essentially free of macrophage infiltration). In support of this finding, a number of in vitro studies undertaken in cell culture have established that iNOS expression is induced in cardiac myocytes after exposure to proinflammatory stimuli (2–3, 28). Moreover, Wang et al. (32) recently showed in a model of late preconditioning that a brief ischemic episode is followed by significant iNOS expression in cardiomyocytes in the nonischemic region 24 h after the preconditioning stimulus.
Myocardial levels of TNF-α mRNA were examined by real-time RT-PCR because this cytokine is known to induce iNOS expression, and the attenuation of myocardial iNOS production by ATL treatment may have been, in part, mediated through an inhibitory effect of the agonist on this proinflammatory signal. Indeed, a recent study reported that the pharmacological activation of A2AARs can inhibit the secretion of cytokines by activated macrophages and prevent accumulation of TNF-α in vivo (21). In the present study, ATL treatment significantly attenuated the eightfold increase in TNF-α expression observed 24 h post-MI in infarcted regions of the reperfused murine heart. However, the magnitude of TNF-α induction was not nearly as pronounced in the noninfarcted remote LV as it was in the infarcted region (only 2-fold vs. 8-fold), suggesting that iNOS induction in the remote LV does not depend entirely on an increase in TNF-α mRNA levels in that same region.
EMSA analysis was used to assess myocardial activation of NF-κB because this transcription factor is known to directly activate the transcription of iNOS mRNA. The results summarized in Fig. 6 suggest that the induction of iNOS mRNA in both the infarcted and noninfarcted regions can largely be accounted for by the activation of NF-κB. Not only did the accumulation of iNOS mRNA in the infarcted and noninfarcted regions coincide closely with the activation of NF-κB but the effect of ATL on iNOS mRNA in the infarcted and noninfarcted regions also coincided with the suppression of NF-κB activity. These results complement previous studies showing that A2AAR-deficient mice have enhanced NF-κB activity (18) and thus provide additional in vivo evidence for the ability of A2AAR stimulation to control the proinflammatory transcriptional activity of NF-κB.
In conclusion, the present study demonstrates that A2AAR stimulation early after MI inhibits both cardiac NF-κB activation and iNOS induction, which is associated with significant improvements in remote LV function independent of any effect on the size of MI. These results suggest that anti-inflammatory treatment with A2AAR agonists may find utility against acute heart failure immediately after large MI.
This work was supported by National Heart, Lung, and Blood Institute Grants R01 HL-69595 and R01 HL-37932 and by the Falk Fund for Adenosine Research. B. A. French is a Fellow of the American Heart Association.
J. Linden owns equity in Adenosine Therapeutics, LLC which provided ATL146e for these studies. However, J. Linden did not participate in data acquisition or analysis, nor did Adenosine Therapeutics, LLC have any control over the decision to publish.
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.
- Copyright © 2006 by the American Physiological Society