Intercellular channels formed by connexins (Cx) have been shown to play a critical role in cardiovascular disease. For instance, vascular Cx43 influences atherosclerosis development and plaque stability, and cardiac Cx43 may determine the development of arrhythmias after myocardial infarction. In the heart, Cx40 is expressed in atrial cardiomyocytes, the conduction system and in endothelial cells, whereas Cx37 is only found in the endothelium. Here, we investigated the implication of the two endothelial connexins during ischemia and reperfusion in mice. For this purpose, we used the Cre-loxP system to create a mouse line in which Cx40 is deleted from the endothelium only. Immunostainings on TIE2-Cre+Cx40fl/flApoE-/- mice confirmed the absence of Cx40 in the endothelium, whereas the protein was normally expressed in the atria and cardiac conduction system. Moreover, Cx40 was normally expressed in the endothelium, atria and conduction system of TIE2-Cre+Cx40wt/wtApoE-/- and TIE2-Cre-Cx40fl/flApoE-/- control mice. Sixteen-week-old mice were subjected to in vivo left coronary artery occlusion for 30 minutes and sacrificed 24-hours after reperfusion for analysis of infarct size and infiltration of inflammatory cells. Myocardial surface areas at risk and infarcted areas were measured from computed images using NIH Image software. Areas at risk, normalized to total left ventricle surface areas, were similar between the experimental groups, i.e TIE2-Cre+Cx40wt/wtApoE-/-: 29.2±1.2% (N=11), TIE2-Cre-Cx40fl/flApoE-/-: 33.7±3.7% (N=9), and TIE2-Cre+Cx40fl/flApoE-/- 31.7±2.2% (N=11, n.s.). Interestingly, the infarct area, normalized to areas at risk, was significantly increased in TIE2-Cre+Cx40fl/flApoE-/- (20.2±3.1%, P<0.05) mice as compared to controls (10.1±2.0% and 11.3±1.8%). Moreover, we performed immunofluorescent stainings against Ly6G to detect neutrophils in the infarcted heart. We showed that neutrophil infiltration was significantly higher in TIE2-Cre+Cx40wt/wtApoE-/- mice (24.2±3.1%, P<0.01, N=4) in comparison with controls (9.8±1.8%, N=5 and 8.5±1.7%, N=5). In a previous study, we demonstrated that endothelial-specific deletion of Cx40 promotes atherosclerosis by increasing CD73-dependent leukocyte adhesion to the endothelium. To determine whether this interaction also plays a role in ischemia-reperfusion injury, CD73-/-ApoE-/- mice have been submitted to in vivo ischemia-reperfusion. Whereas the area at risk in this group of mice was similar to those measured in the previous groups (31.3±3.6%, N=8), the infarcted area was significantly increased in comparison with control mice (18.2±1.6%, P<0.01). Treatment of TIE2-Cre+Cx40wt/wtApoE-/- mice with methotrexate, a CD73 activator, significantly reduced the infarcted area (8.5±1.3%, P<0.05, N=10) as compared to TIE2-Cre+Cx40wt/wtApoE-/- mice without treatment. Finally, hearts from TIE2-Cre+Cx40wt/wtApoE-/- mice and control mice have been submitted to ex vivo ischemia-reperfusion by using a Langendorff perfusion system. Areas at risk and infarct size were similar between the three groups. To investigate the possible implications of another endothelial connexin in myocardial infarction, Cx37-/-ApoE-/- mice were submitted to the same protocol of in vivo ischemia-reperfusion. Areas at risks were similar between Cx37-/-ApoE-/- and control ApoE-/- mice, and the infarcted area appeared also not affected by the deletion of Cx37 (Cx37-/-ApoE-/-: 13.8±1.3%, ApoE-/-: 14.7±2.7%, N=12, n.s.). We conclude that endothelial Cx40, but not Cx37, is implicated in resistance of the heart to ischemia-reperfusion injuries, such as cell death and inflammatory infiltration, by a CD73-dependent pathway. These findings may point towards a specific novel therapeutic target to limit the cardiac injury after coronary interventions.