Whether volatile anesthetics produce a second window of preconditioning is unclear. The authors tested the hypothesis that isoflurane causes delayed preconditioning against infarction and, further, that cyclooxygenase (COX)-2 mediates this beneficial effect.
Rabbits (n = 43) were randomly assigned to receive 0.9% intravenous saline, the selective COX-2 inhibitor celecoxib (3 mg/kg intraperitoneal) five times over 2 days before coronary artery occlusion and reperfusion, or isoflurane (1.0 minimum alveolar concentration) 24 h before acute experimentation in the absence or presence of celecoxib pretreatment. Two additional groups of rabbits received a single dose of celecoxib either 30 min before or 21.5 h after administration of isoflurane. Rabbits were then instrumented for measurement of hemodynamics and underwent 30 min of coronary occlusion followed by 3 h of reperfusion. Myocardial infarct size was measured using triphenyltetrazolium staining. Western immunoblotting to examine COX-1 and COX-2 protein expression was performed in rabbit hearts that had or had not been exposed to isoflurane.
Isoflurane significantly (P < 0.05) reduced infarct size (22 +/- 3% of the left ventricular area at risk) as compared with control (39 +/- 2%). Celecoxib alone had no effect on infarct size (36 +/- 4%) but abolished isoflurane-induced cardioprotection (36 +/- 4%). A single dose of celecoxib administered 2.5 h before coronary occlusion and reperfusion also abolished the delayed protective effects of isoflurane (36 +/- 4%), but celecoxib given 30 min before exposure to isoflurane had no effect (22 +/- 4%). Isoflurane did not alter COX-1 and COX-2 protein expression.
The results indicate that the volatile anesthetic isoflurane produces a second window of preconditioning against myocardial ischemia and reperfusion injury. Furthermore, COX-2 is an important mediator of isoflurane-induced delayed preconditioning.
A BRIEF period of myocardial ischemia imposed shortly before a more prolonged coronary artery occlusion protects myocardium against infarction. 1This process is known as ischemic preconditioning (IPC). IPC is characterized by an acute early phase limited to 1–3 h after the brief ischemic stimulus, 2and a delayed late phase (i.e. , a second window) that emerges after 24 h and may persist for up to 72 h after the ischemic episode. The protective effects of delayed IPC are mimicked by a variety of drugs, including nitric oxide donors, 3selective δ-opioid agonists, 4adenosine receptor agonists, 5and mitochondrial adenosine triphosphate–sensitive potassium (KATP) channel openers. 6,7Volatile anesthetics have been shown to protect myocardium against reversible and irreversible ischemic injury. 8–11This anesthetic-induced preconditioning (APC) is mediated by many of the same signal transduction elements involved in acute and delayed IPC. Whether volatile anesthetics produce delayed preconditioning remains unclear. Our laboratory recently demonstrated that isoflurane did not produce a second window of preconditioning in dogs, 12but the effects of volatile agents in other species is unknown.
Delayed IPC is characterized by complex signal transduction cascades in which several protein kinases, 13–18nuclear factor κB, 19nitric oxide synthase, 20,21and cyclooxygenase (COX)-2 22,23have been implicated. In the current investigation, we tested the hypothesis that exposure to isoflurane 24 h before ischemia and reperfusion produces a second window of myocardial protection in vivo . In addition, we tested the hypothesis that COX-2 is involved in isoflurane-induced delayed preconditioning and, further, examined whether COX-2 functions as a trigger or mediator of this delayed protective effect.
Materials and Methods
All experimental procedures and protocols used in this investigation were reviewed and approved by the Institutional Animal Care and Use Committee of the Medical College of Wisconsin (Milwaukee, Wisconsin). Furthermore, all conformed to the Guiding Principles in the Care and Use of Animals 24of the American Physiologic Society and were in accordance with the Guide for the Care and Use of Laboratory Animals . 25
Administration of Isoflurane
Male New Zealand white rabbits weighing between 2.5 and 3.0 kg were placed in an induction chamber 24 h before acute experimentation. After inhalational induction, anesthesia was maintained with 1.0 minimum alveolar concentration (MAC) isoflurane (2.1%) 26in 100% oxygen during spontaneous ventilation. The concentration of isoflurane inside the induction chamber was continuously measured by an infrared anesthetic analyzer that was calibrated with known standards before and during experimentation. Ten rabbits were randomly chosen for arterial blood gas analysis during administration of isoflurane. A 22-gauge catheter was inserted into a middle ear artery approximately 15 min after anesthetic induction, and arterial blood gas tensions were measured at selected intervals during administration of isoflurane. Anesthesia was discontinued after 2 h, and emergence was allowed to occur. Each rabbit was then housed overnight before experimentation on the next day.
General Preparation
Rabbits with and without previous exposure to isoflurane were anesthetized with intravenous sodium pentobarbital (30 mg/kg) as previously described. 10Briefly, a tracheostomy was performed through a ventral midline incision, and the trachea was cannulated. The rabbits were ventilated with positive pressure using an air-and-oxygen mixture. Heparin-filled catheters were inserted into the right carotid artery and the left jugular vein for measurement of arterial blood pressure and fluid or drug administration, respectively. A left thoracotomy was performed at the fourth intercostal space. A prominent branch of the left anterior descending coronary artery (LAD) was identified, and a silk ligature was placed around this vessel approximately halfway between the base and the apex for the production of coronary artery occlusion and reperfusion. Each rabbit was anticoagulated with 500 U heparin immediately before LAD occlusion. Coronary artery occlusion was verified by the presence of epicardial cyanosis in the ischemic zone, and reperfusion was confirmed by observing an epicardial hyperemic response. Hemodynamic data were continuously recorded on a polygraph throughout experimentation.
Experimental Protocol
The experimental design is depicted in figure 1. Baseline systemic hemodynamic data were recorded 30 min after instrumentation was completed. All rabbits underwent 30 min of LAD occlusion followed by 3 h of reperfusion. In four separate groups, rabbits were randomly assigned to receive only coronary artery occlusion and reperfusion (control), the selective COX-2 inhibitor celecoxib (3 mg/kg in 20% dimethyl sulfoxide, intraperitoneal) five different times at 12 h intervals for 2 days before LAD occlusion (48, 36, 24, 12, and 2.5 h before coronary occlusion and reperfusion), or exposure to 1.0 MAC isoflurane 24 h before acute experimentation in the absence or presence of celecoxib pretreatment. In two additional groups, rabbits received a single dose of celecoxib either 30 min before administration of isoflurane or 2.5 h before LAD occlusion after exposure to isoflurane on the previous day. Rabbits that developed intractable ventricular fibrillation and those with a left ventricular (LV) area at risk (AAR) for infarction less than 15% of the total LV mass were excluded from subsequent analysis.
Determination of Myocardial Infarct Size
At the end of each experiment, myocardial infarct size was measured by means of a histochemical staining technique as previously described. 27Briefly, the LAD was reoccluded, and 3 ml intravenous patent blue dye was injected. The LV AAR for infarction was separated from surrounding normal areas (stained blue), and the two regions were incubated at 37°C for 30 min in 1% 2,3,5-triphenyltetrazolium chloride in 0.1 m phosphate buffer adjusted to a pH of 7.4. After overnight storage in 10% formaldehyde, infarcted (unstained) and noninfarcted (stained red) myocardium within the AAR were carefully separated and weighed. Infarct size was expressed as a percentage of the AAR.
Protein Isolation
Left ventricular tissue samples were also obtained from rabbits with (n = 3) and without (n = 3) previous exposure to isoflurane 24 h before anesthesia with intravenous sodium pentobarbital (30 mg/kg). After surgical instrumentation as described above, hearts were quickly excised, and the LV was isolated, quickly frozen in liquid nitrogen, and stored at −70°C. Tissue samples were homogenized in cold lysis buffer (pH 7.4) composed of 50 mm Tris, 1 mm EDTA, 1 mm EGTA, 1 mm sodium fluoride, 200 μm sodium orthovanadate, 1 mm phenylmethylsulfonyl fluoride, and 1 ml/20 g tissue protease inhibitor cocktail (Sigma P-8340; St. Louis, MO). The homogenate was centrifuged at 700 g for 10 min to remove cellular debris and isolate total protein. Total protein was used because previous studies have indicated that COX-1 and COX-2 are membrane proteins contained within both the endoplasmic reticulum and the nuclear envelope. 28,29All procedures were performed at 4°C, and protein concentration was determined by the Bradford method (BIO-RAD, Hercules, CA).
Western Immunoblotting
Equivalent amounts (50 μg) of protein samples were mixed with loading buffer (125 mm Tris, 4% sodium dodecyl sulfate, 5% glycerol, 0.03% bromophenol blue, and 1% 2-mercaptoethanol; pH 6.8) and heated at 95°C for 10 min. Samples were loaded onto a 12% polyacrylamide gel (BIO-RAD) and electrophoretically size separated at 180 V for approximately 1 h. After size separation, proteins were electrophoretically transferred to a nitrocellulose membrane (BIO-RAD) at 100 V for 1 h and then blocked with 3% bovine serum albumin in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) at room temperature for 2 h. Subsequently, nitrocellulose membranes were incubated overnight at 4°C in TBS-T containing 1% bovine serum albumin and a 1:1,000 dilution of goat primary antibodies for COX-1 or COX-2 (Santa Cruz Biotechnology, Santa Cruz, CA). Membranes were washed four times with TBS-T for 10 min before a 30-min incubation with a 1:5,000 dilution of horseradish peroxidase–labeled donkey anti-goat immunoglobulin G (Santa Cruz Biotechnology) in TBS-T containing 1% bovine serum albumin. Bound antibody was detected by enhanced chemiluminescence (Amersham Pharmacia, Piscataway, NJ) on radiograph film. Ponceau staining of nitrocellulose membranes was used to verify equal protein loading. Quantitative analysis of the band densities was performed using AlphaImager 2000 software (Alpha Innotech Corporation, San Leandro, CA).
Statistical Analysis
Statistical analysis of data within and between groups was performed with analysis of variance for repeated measures followed by the Student-Newman-Keuls test. Statistical significance was defined as P < 0.05. All data are expressed as mean ± SEM.
Results
Forty-nine rabbits were instrumented to obtain 43 successful experiments. One rabbit was excluded because the LV AAR was less than 15% of the LV mass (single dose of celecoxib 2.5 h before LAD occlusion + isoflurane). Five rabbits were excluded because of intractable ventricular fibrillation (1 control; 1 celecoxib alone; 1 isoflurane alone; 1 celecoxib + isoflurane; and 1 single dose of celecoxib 2.5 h before LAD occlusion + isoflurane).
Hemodynamics
No differences in baseline hemodynamics were observed between groups (table 1). Coronary artery occlusion and reperfusion produced similar decreases in mean arterial pressure and rate–pressure product in each group. No differences in hemodynamics were observed between groups. Arterial blood gas tensions and acid–base status were maintained within the normal range during pretreatment with isoflurane in all rabbits (data not shown).
Infarct Size
Body weight, LV weight, AAR weight, and the ratio of AAR to total LV mass were similar between groups (table 2). Isoflurane significantly (P < 0.05) reduced infarct size (22 ± 3% of the LV AAR) as compared with control experiments (39 ± 2%;fig. 2). Administration of celecoxib alone five times over 2 days before LAD occlusion had no effect on infarct size (36 ± 4%), but this celecoxib pretreatment protocol abolished the protective effect of exposure to isoflurane (36 ± 4%). Administration of a single dose of celecoxib 2.5 h before LAD occlusion also eliminated the delayed protective effects of isoflurane (36 ± 4%). In contrast, a single dose of celecoxib administered 30 min before exposure to isoflurane did not affect infarct size (22 ± 4%).
Protein Expression
Isoflurane did not alter COX-1 or COX-2 protein expression in rabbit myocardium 24 h after exposure to the volatile anesthetic as compared with control hearts (fig. 3).
Discussion
Delayed IPC was recently shown to be mediated by COX-2 in rabbits 22and mice. 23Activation of δ-opioid receptors also produced delayed preconditioning in rabbits 30and rats. 31These beneficial effects were abolished by selective COX-2 inhibition. IPC produced COX-2 transcription and translation and also enhanced the myocardial concentration of the prostaglandins PGE2and 6-keto-PGF1α24 h after the initial ischemic stimulus. 22Similar results were also observed during the second window of δ-opioid–induced protection in rabbits 30but not rats. 31For example, prostacyclin (PGI2) synthase expression and PGI2production but not COX-2 increased 24 h after administration of the selective δ-opioid agonist BW373U86 in rat hearts. 31The second window of preconditioning precipitated by heat stress was associated with an increase in COX-2 protein in rat myocardium, 32but adenosine A1or A3receptor–induced late preconditioning did not seem to involve a COX-2-dependent mechanism. 33Therefore, COX-2 has been previously implicated in ischemic and pharmacologic delayed preconditioning, but a role for this enzyme has yet to be uniformly established and may also be species dependent.
Our laboratory has previously demonstrated that isoflurane protects the myocardium against irreversible ischemic injury and, further, that this protection is characterized by an acute memory phase in rats, 11rabbits, 10and dogs. 8The results of the current investigation indicate that isoflurane reduces myocardial damage when administered 24 h before coronary artery occlusion in rabbit hearts in vivo , a protective effect that is equal in magnitude to acute preconditioning. 10The current findings also show that this isoflurane-induced delayed preconditioning is abolished by administration of celecoxib after but not before exposure to isoflurane. These data suggest that COX-2 serves as a mediator but not a trigger for isoflurane-induced delayed preconditioning. The current results also indicate that COX-1 and COX-2 protein expression is not affected by exposure to isoflurane 24 h before coronary artery occlusion and reperfusion. These data show that enhanced COX-2 expression is not the mechanism for the delayed protective effect of the volatile agent and indirectly suggest that an increase in COX-2 activity may instead play a role in this process. Whether the activity of preexisting COX-2 was enhanced or prostaglandin synthesis was increased was not ascertained in the current study. The cellular specificity of COX expression and activity within the heart were also not identified. However, recent evidence indicates that cardiac myocytes are the major source of increased inducible nitric oxide synthase (iNOS) expression in delayed IPC, 34a crucial signaling element that likely modulates COX-2 expression and activity. 35
The signaling mechanisms that regulate COX-2 expression and the functional involvement of COX-2–dependent prostaglandin synthesis during the second window of preconditioning remain to be fully elucidated. COX is the rate-limiting enzyme in the conversion of arachidonic acid to PGH2. 36In contrast with the constitutive COX-1 isoform, COX-2 is normally absent from cells and is induced in response to stress. iNOS and nitric oxide produced by this enzyme are also mediators of delayed preconditioning 21and are frequently coinduced with COX-2. 36IPC enhances iNOS expression in cardiac myocytes. 34NO may act as both a trigger and a mediator of delayed IPC 20through a COX-2–dependent mechanism. 37IPC-induced increases in PGE2and 6-keto-PGF1αare attenuated by iNOS inhibitors, but blockade of COX-2 does not affect iNOS activity. 35These data suggest that COX-2 is located downstream of iNOS in this signaling pathway. In addition, iNOS, reactive oxygen species, and soluble guanylate cyclase do not seem to be involved in regulating increases in COX-2 expression during the second window of preconditioning, 35despite observations indicating that these molecules play important signaling roles in delayed protection. 20,38,39These data strongly suggest that other autocoids are involved in mediating the second window of preconditioning through COX-2.
COX-2 up-regulation in delayed IPC is dependent on protein kinase C (PKC), protein tyrosine kinase, and nuclear factor κB. 35The nitric oxide donor nitroglycerin produced a second window of protection in rabbits that was abolished by the PKC inhibitor chelerythrine. 13In fact, the ε isoform of PKC has been shown to play a critical role in delayed IPC. 14A role for PKC activation and translocation in APC has also been strongly implied in several studies, but whether PKC mediates delayed protection associated with volatile agents remains to be identified. Trigger and mediator functions have been suggested for several families of protein tyrosine kinase during the second window of preconditioning. Protein tyrosine kinase subtypes also seem to be essential modulators of iNOS activity. 15Our laboratory has recently demonstrated that Src protein tyrosine kinase is involved in APC, 40but how this kinase species modulates delayed preconditioning by volatile anesthetics is also currently unknown. The mitogen-activated protein kinase family has been implicated as a critical regulator of protein expression in the delayed protection elicited by δ-opioid agonists 18and adenosine. 41In fact, the p38 mitogen-activated protein kinase subfamily activator anisomycin mimics late preconditioning by enhancing both nuclear factor κB and iNOS activity. 42The janus tyrosine kinases and the signal transducers and activators of transcription also have been shown to play essential roles in COX-2 up-regulation that occurs during the second window of preconditioning. 43Whether these other previously identified signal transduction molecules are also involved in delayed APC remains to be determined.
The results of the current investigation also show that celecoxib administered five times over 2 days abolishes the second window of protection produced by isoflurane. Although these initial experiments were conducted to determine whether COX-2 was involved in isoflurane-induced delayed preconditioning, this drug regimen was also chosen on the basis of the controversy that COX-2 inhibitors may increase cardiovascular events. 44The incidence of myocardial ischemic episodes is increased during the perioperative period. Therefore, many patients taking COX-2 inhibitors before and after surgery may be at a higher risk for development of cardiovascular complications. Our current data, although preliminary, support the contention that COX-2 inhibition may magnify the occurrence of cardiovascular events.
The current findings should be interpreted within the constraints of several potential limitations. The LV AAR for infarction and coronary collateral blood flow represent major determinants of myocardial infarct size. The LV AAR values were similar between experimental groups, and minimal coronary collateral blood flow has been previously reported in rabbits. 45Therefore, it seems unlikely that the current results were substantially affected by these variables. There were no differences in systemic hemodynamics between groups before and after coronary artery occlusion. Therefore, it seems highly likely that the observed reductions in myocardial infarct size produced by remote exposure to isoflurane occurred independent of many of the hemodynamic determinants of myocardial oxygen consumption. Values of the rate–pressure product, an indirect index of myocardial oxygen consumption, were also similar between experimental groups. Nevertheless, myocardial oxygen consumption was not directly calculated, nor was coronary collateral blood flow measured during the current investigation.
Our laboratory previously reported that isoflurane administered 24 h before coronary artery occlusion and reperfusion did not protect against myocardial infarction in dogs. 12Differences between the current and previous findings in rabbits and dogs may be due to some degree to variation in experimental design (i.e. , 2 vs. 6 h of isoflurane pretreatment, spontaneous vs. controlled ventilation, respectively) but more likely resulted because the late window of protection seems to occur at distinct times in different species. 46In addition, the current investigation did not determine myocardial infarct size in rabbits at any time interval other than 24 h after isoflurane administration. Future research is needed to characterize the time and duration of the early and second windows of APC.
In summary, the current results indicate that exposure to isoflurane 24 h before coronary artery occlusion and reperfusion reduces experimental myocardial infarct size in rabbits. This isoflurane-induced delayed preconditioning was abolished by celecoxib when the selective COX-2 inhibitor was administered after but not before exposure to the volatile agent. COX-2 protein expression in rabbit myocardium 24 h after administration of isoflurane was unchanged and thus not responsible for the delayed protective effects of isoflurane. However, eicosanoids directly produced by this enzyme complex or others synthesized from intermediate products produced by COX-2 seem to be key elements of the signal transduction in APC. The results suggest that COX-2 is an essential mediator but not a trigger of the intracellular signaling cascade responsible for isoflurane-induced delayed preconditioning in vivo .
The authors thank David A. Schwabe, B.S.E.E. (Senior Research Technologist, Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin), for technical assistance. The authors also thank Mary Lorence-Hanke, A.A. (Administrative Assistant, Department of Anesthesiology, Medical College of Wisconsin), for assistance in preparation of the manuscript.