Cyclooxygenase-2 mediates ischemic, anesthetic, and pharmacologic preconditioning in vivo

Background. Cyclooxygenase-2 (COX-2) mediates the late phase of ischemic preconditioning (IPC), but whether this enzyme modulates early IPC, anesthetic-induced preconditioning (APC), or other forms of pharmacologic preconditioning (PPC) is unknown. The authors tested the hypothesis that COX-2 is an essential mediator of IPC, APC, and PPC in vivo. Methods: Barbiturate-anesthetized dogs (n = 91) were instrumented for measurement of hemodynamics and randomly assigned to receive IPC (four 5-min coronary occlusions interspersed with 5-min reperfusions), APC (1.0 minimum alveolar concentration of isoflurane for 30 min), or PPC (selective mitochondrial K-ATP channel opener diazoxide, 2.5 mg/kg intravenous) in the presence or absence of pretreatment with oral aspirin (650 mg), the selective COX-2 inhibitor celecoxib (200 mg), or acetaminophen (500 mg) administered 24,12, and 2 It before experimentation in 12 separate experimental groups. All dogs were subjected to a 60-min coronary artery occlusion followed by 3 h of reperfusion. Myocardial infarct size and coronary collateral blood flow were quantified with triphenyltetrazolium staining and radioactive microspheres, respectively. Myocardial 6-keto-prostaglandin F-1alpha, a stable metabolite of prostacyclin, was measured (enzyme immunoassay) in separate experiments (n = 8) before and after isoflurane administration, in the presence or absence of celecoxib. Results: No significant differences in baseline hemodynamics or the left ventricular area at risk for infarction were observed between groups. IPC, isoflurane, and diazoxide all decreased myocardial infarct size (9 +/- 1, 12 +/- 2, and 11 +/- 1%, respectively) as compared with control (30 1%). Celecoxib alone had no effect on infarct size (26 +/- 3%) but abolished IPC (30 +/- 3%), APC (30 +/- 3%), and PPC (26 +/- 1%). Aspirin (24 +/- 3%) and acetaminophen alone (29 +/- 2%) did not alter infarct size or abolish APC-induced protection (18 +/- 1 and 19 +/- 1%, respectively). Isoflurane increased myocardial 6-keto-prostaglandin F-1alpha to 463 +/- 267% of baseline in the absence but not in the presence (94 +/- 13%) of celecoxib. Conclusions: The results indicate that COX-2 is a critical mediator of IPC, APC, and PPC in dogs. The role of cyclooxygenase enzymes as obligatory mediators of myocardial protection produced by diverse preconditioning stimuli may have implications for the clinical use of COX-2 inhibitors.