Adiponectin protects against myocardial ischemia-reperfusion injury through AMPK- and COX-2-dependent mechanisms

Abstract

Obesity-related disorders are associated with the development of ischemic heart disease. Adiponectin is a circulating adipose-derived cytokine that is downregulated in obese individuals and after myocardial infarction. Here, we examine the role of adiponectin in myocardial remodeling in response to acute injury. Ischemia-reperfusion in adiponectin-deficient (APN-KO) mice resulted in increased myocardial infarct size, myocardial apoptosis and tumor necrosis factor (TNF)-alpha expression compared with wild-type mice. Administration of adiponectin diminished infarct size, apoptosis and TNF-alpha production in both APN-KO and wild-type mice. In cultured cardiac cells, adiponectin inhibited apoptosis and TNF-alpha production. Dominant negative AMP-activated protein kinase (AMPK) reversed the inhibitory effects of adiponectin on apoptosis but had no effect on the suppressive effect of adiponectin on TNF-alpha production. Adiponectin induced cyclooxygenase (COX)-2-dependent synthesis of prostaglandin E(2) in cardiac cells, and COX-2 inhibition reversed the inhibitory effects of adiponectin on TNF-alpha production and infarct size. These data suggest that adiponectin protects the heart from ischemia-reperfusion injury through both AMPK- and COX-2-dependent mechanisms.

Figure 1

Increased myocardial infarction, myocardial apoptosis and TNF-α expression in APN-KO mice subjected to ischemia-reperfusion injury. (a) Representative pictures of myocardial tissues from wild-type (left) and APN-KO (right) mice at 48 h after ischemia-reperfusion. The nonischemic area is indicated by blue, the AAR by red and the infarct area by white. (b) Quantification of infarct size in wild-type (n = 7) and APN-KO (n = 7) mice. AAR/LV, ratio of AAR to left ventricular area; IA/AAR, ratio of infarct area to AAR; IA/LV, ratio of infarct area to left ventricular area. (c) Release of CPK from wild-type (n = 4) and APN-KO mice (n = 4) after sham operation or ischemia-reperfusion. (d) Representative photographs of TUNEL-stained heart sections from wild-type and APN-KO mice at 48 h after sham operation or ischemia-reperfusion. Apoptotic nuclei were identified by TUNEL staining (green) and total nuclei by DAPI counterstaining (blue). (e) Quantitative analysis of apoptotic nuclei from wild-type (n = 5) and APN-KO mice (n = 5) hearts after sham operation or ischemia-reperfusion. TUNEL-positive nuclei are expressed as a percentage of the total number of nuclei. (f) Phosphorylation of AMPK in heart tissues from wild-type and APN-KO mice at 48 h after sham operation or ischemia-reperfusion. (n = 4 hearts per group). (g) Serum levels of TNF-α, determined by ELISA, in wild-type (n = 5) and APN-KO (n = 5) mice at 48 h after sham operation or ischemia-reperfusion. (h) Myocardial levels of Tnf transcript in wild-type (n = 5) and APN-KO (n = 5) mice. Tnf mRNA in myocardium of wild-type and APN-KO mice were quantified by RT-PCR. Results are presented as mean ± s.d. WT, wild-type; I/R, ischemia-reperfusion.

Figure 2

Adenovirus-mediated expression of adiponectin diminishes infarct size, apoptosis and TNF-α production after ischemia-reperfusion in wild-type (WT) and APN-KO mice. Ad-APN (2 × 10⁸ plaque-forming units total) was delivered intravenously through the jugular vein 3 d before ischemia-reperfusion injury. (a) Quantification of infarct size in wild-type (n = 4) and APN-KO (n = 4) treated with Ad-APN or Ad-βgal (control) at 48 h after surgery. AAR/LV, ratio of AAR to left ventricular area; IA/AAR, ratio of infarct area to AAR; IA/LV, ratio of infarct area to left ventricular area. (b) Quantitative analysis of apoptotic nuclei from wild-type (n = 4) and APN-KO (n = 4) mice hearts treated with Ad-APN or Ad-βgal after ischemia-reperfusion. TUNEL-positive nuclei were counted in several randomly selected fields and expressed as a percentage of the total number of nuclei. (c) Phosphorylation of AMPK in heart tissues of wild-type and APN-KO mice treated with Ad-APN or Ad-βgal at 48 h after sham operation or ischemia-reperfusion. AMPK phosphorylation (P-AMPK) and total AMPK in myocardium was analyzed by western blotting. Phosphorylation levels of AMPK were quantified and expressed relative to untreated WT. Immunoblots were normalized to total loaded protein (n = 4 hearts per experimental group). (d) Serum levels of TNF-α, in wild-type (n = 4) and APN-KO (n = 4) mice treated with Ad-APN or Ad-βgal at 48 h after ischemia-reperfusion. (e) Myocardial levels of Tnf transcript in wild-type (n = 4) and APN-KO (n = 4) mice treated with Ad-APN or Ad-βgal at 48 h after ischemia-reperfusion. Results are presented as mean ± s.d.

Figure 3

AMPK-dependent inhibition of cardiac myocyte and fibroblast apoptosis by adiponectin. Rat neonatal myocytes and fibroblasts were treated with adiponectin (30 μg/ml) or vehicle in serum-free media for 48 h under normoxic conditions or 12 h under hypoxic conditions followed by 24 h of reoxygenation. (a) Representative photomicrographs of TUNEL-positive cardiac myocytes and fibroblasts. Apoptotic nuclei were identified by TUNEL staining (green) and total nuclei by DAPI counterstaining (blue). (b) Quantitative analysis of TUNEL-positive cells under conditions of normoxia or hypoxia-reoxygenation with adiponectin (APN) or vehicle. TUNEL-positive nuclei were counted in several randomly selected fields and expressed as a percentage of the total number of nuclei. (c) Effect of Ad-dnAMPK on adiponectin inhibition of myocyte and fibroblast apoptosis under conditions of normoxia or hypoxia-reoxygenation. Cells were transduced with Ad-dnAMPK or Ad-βgal (control) for 24 h and then treated with adiponectin or vehicle under conditions of normoxia or hypoxia-reoxygenation. Results are presented as mean ± s.d. (n = 3).

Figure 4

Adiponectin suppresses LPS-induced secretion of TNF-α from neonatal myocytes through a COX-2–dependent pathway. (a) Effect of adiponectin (APN) on LPS-induced production of TNF-α. TNF-α levels in media were determined by ELISA. Mock-transduced cells were compared with cells that were transduced with Ad-dnAMPK or Ad-βgal for 24 h, and then treated with adiponectin or vehicle, and stimulated with or without LPS. (b) Adiponectin stimulates secretion of PGE₂ from cardiac myocytes. Pretreatment with the COX-2 inhibitor NS398 blocked adiponectin-stimulated production of PGE₂ in the presence or absence of LPS. Transduction with Ad-dnAMPK did not affect adiponectin-induced production of PGE₂. Transduction with Ad-βgal had no effect on production of PGE₂. (c) Effect of adiponectin on expression of COX-2 in myocytes. Transduction with dominant negative AMPK did not affect induction of COX-2 in cultured cardiac myocytes. (d) Contribution of the COX-2–PGE₂ pathway to adiponectin inhibition of LPS-induced production of TNF-α from myocytes. Cells were pretreated with a EP₄-selective antagonist, AH23848, NS398 or vehicle, treated with adiponectin or vehicle and stimulated with or without LPS. Results are presented as mean ± s.d. (n = 3–5). N.S., not statistically significant.

Figure 5

Inhibition of COX-2 partially prevents the protective actions of adiponectin on myocardial infarct size after ischemia-reperfusion injury in wild-type (WT) and APN-KO mice. (a) The COX-2 inhibitor NS398 was injected intraperitoneally from 3 d before ischemia-reperfusion injury until mice were killed, and Ad-APN or Ad-βgal was injected into the jugular vein 3 d before surgery. Infarct size in the heart tissue was quantified (n = 5 mice per experimental group). IA/AAR, ratio of infarct area to AAR. (b) NS398 reversed the suppressive effect of adiponectin on serum levels of TNF-α, after infarction in both wild-type (WT) and APN-KO mice. (c) NS398 increased the frequencies of TUNEL-positive cells after ischemia-reperfusion in Ad-APN-treated wild-type and APN-KO mice. (d) Administration of recombinant adiponectin minimized the effects of ischemia-reperfusion on infarct size and heart function. Quantification of infarct size in wild-type mice treated with recombinant adiponectin before, during and after ischemic injury (n = 5 per group). *P < 0.01, **P < 0.05 versus vehicle. (e) Effect of recombinant adiponectin on left ventricular end-diastolic pressure (LVEDP) in wild-type mice at 24 h after sham operation or ischemia-reperfusion. Recombinant adiponectin (1.0 μg/g) or vehicle was injected in wild-type mice before ischemia-reperfusion (n = 5). (f) Left ventricular dP/dt in wild-type mice (n = 5) treated with adiponectin (1.0 μg/g) or vehicle at 24 h after sham operation or ischemia-reperfusion. (g) Left ventricular fractional shortening assessed by echocardiography in wild-type (n = 4) treated with recombinant adiponectin (1.0 μg/g) at 7 d after ischemia-reperfusion. (h) Adiponectin protects the myocardium from cardiac injury in response to ischemia by protecting cardiac cells from apoptosis through activation of AMPK signaling and by the suppression of cardiac production of TNF-α by the activation of the COX-2–PGE₂ pathway.