Anesthetic Preconditioning Enhances Ca²⁺Handling and Mechanical and Metabolic Function Elicited by Na⁺–Ca²⁺Exchange Inhibition in Isolated Hearts

The investigation conformed to the Guide for the Care and Use of Laboratory Animals  (NIH publication 85-23, revised 1996). Previous approval was obtained from the Medical College of Wisconsin Biomedical Research Committee, Milwaukee, Wisconsin.

Our methods have been described in detail previously.3,4,11,14,20Ketamine (30 mg) and heparin (1,000 U) were injected intraperitoneally into 56 English short-haired guinea pigs (250–350 g) 15 min before the animals were killed. After thoracotomy, the aorta was cannulated distal to the aortic valve, and the inferior and superior venae cavae were cut from the heart. Each heart was immediately perfused via  the aortic root with a cold oxygenated modified Krebs-Ringer's solution (equilibrated with 95% O²–5% CO²) and rapidly excised. All hearts were perfused at an aortic root perfusion pressure of 55 mmHg. The perfusate (pH, 7.39 ± 0.1; partial pressure of oxygen [Po²]; 560 ± 10 mmHg) was in-line filtered (5-μm pore size) and had the calculated composition of (nonionized) 137 mm Na⁺, 5 mm K⁺, 1.2 mm Mg²⁺, 2.5 mm Ca²⁺, 134 mm Cl⁻, 15.5 mm HCO³⁻, 1.2 mm H²PO⁴⁻, 11.5 mm glucose, 2 mm pyruvate, 16 mm mannitol, 0.05 mm EDTA, and 0.1 mm probenecid, with 5 U/l insulin. Perfusate and bath temperatures were maintained at 37.2°± 0.1°C before and after ischemia by preset thermostatically controlled water circulatory systems operated in parallel.

Left ventricular pressure (LVP) was measured isovolumetrically with a transducer connected to a thin, saline-filled latex balloon inserted into the left ventricle through the mitral valve from a cut in the left atrium. Balloon volume was initially adjusted to a diastolic LVP of 0 mmHg so that any subsequent increase in diastolic LVP reflected an increase in left ventricle wall stiffness or diastolic contracture. Bipolar electrodes were placed in the right atrial appendage and in the right ventricular free wall to monitor spontaneous heart rate. Coronary flow (aortic inflow) was measured by an ultrasonic flowmeter (T106X; Transonic System, Ithaca, NY) placed directly into the aortic flow line. Coronary effluent Na⁺, K⁺, Ca²⁺, Po², partial pressure of carbon dioxide (Pco²), and pH were measured off-line. Coronary sinus effluent was collected from a small catheter placed into the right ventricle through the pulmonary artery after ligating both venae cavae. Coronary outflow (coronary sinus) oxygen tension was also measured continuously on-line with a Clark-type oxygen electrode. Myocardial oxygen consumption was calculated as (coronary flow/g heart weight) × (arterial Po²− venous Po²) × 24 μl O²/ml (37°C) at 760 mmHg. Cardiac work efficiency was calculated as systolic − diastolic LVP × heart rate/oxygen consumption. At the end of 60 min of reperfusion, each heart was fast frozen in liquid nitrogen for a later determination of specific protein content.

This laboratory has published these methods and calibrations in detail.4,11,14,20,21Briefly, intracellular [Ca²⁺] was measured by spectrophotofluorimetry at the left ventricular free wall using a trifurcated fiberoptic cable to direct the excitation and emission wavelengths. Hearts were loaded with the Ca²⁺-sensitive dye indo-1 AM. Indo-1 fluorescence (F) intensity was corrected for background autofluorescence, and changes in F due to ischemia and reperfusion were measured in six additional hearts, wherein the same protocols were used as for Ca²⁺determination except that only the vehicle was perfused. Calibration curves were derived using modifications of the standard equation for fluorescent indicators used by Brandes et al.  10In brief, total intracellular [Ca²⁺] ([Ca²⁺]^tot) was calculated from the total F³⁸⁵to total F⁴⁵⁶ratio (R^tot), R^minand R^max, and K^daccording to the equation [Ca²⁺]^tot= S⁴⁵⁶× K^d[(R^tot− R^min)/R^max− R^tot)].

Noncytosolic fluorescence (primarily mitochondrial) was measured at the end of each experiment after perfusing hearts with 100 μm MnCl²for 10 min to quench fluorescence derived from the cytosolic compartment. This procedure does not alter the LVP transients. Details of this subtraction method have been published.10,21[Ca²⁺] data are displayed as systolic and diastolic [Ca²⁺] and their first time derivatives. The loss of membrane integrity in damaged cells on reperfusion could result in a leakage of indo-1 and lower signal intensities. However, because this is a ratiometric determination of [Ca²⁺], both F signals degrade similarly, which results in the F ratio remaining relatively unchanged with slightly less than half of the postloaded signal strength and fivefold greater than the unloaded signal strength.

Frozen hearts were thawed and homogenized with a Brinkmann Polytron (Brinkmann Instruments, Inc., Westbury, NY) 15-mm probe, four times for 15 s, with 30-s intervals, and then again with five volumes of ice-cold isolation buffer containing 30 mm 3-(N-morpholino)propanesulfonic acid (MOPS; pH 7.0; Sigma, St. Louis, MO) and a protease inhibitor cocktail (Sigma). The homogenates were then centrifuged at 3,000g  for 10 min. The supernatant obtained was stored at −80°C until use. Protein assays were preformed by the Bradford method using a BioRad assay kit (BioRad Laboratories, Hercules, CA). Western blot assays were conducted as previously described.22Briefly, aliquots of samples (homogenates) were solubilized in Laemmli sample buffer and fractionated by SDS-polyacrylamide gel electrophoresis using approximately 4–20% slab gels. After the transfer to nitrocellulose membranes, they were blocked with 5% nonfat milk in phosphate-buffered saline and probed with primary antibodies against a ryanodine receptor (RyR2), sarcoplasmic reticular (SR) Ca²⁺-pump adenosine triphosphatase (ATPase SERCA2a), and phospholamban (Affinity Bioreagents Inc., Golden, CO). Secondary antibodies were conjugated to horseradish peroxidase (1:10,000 dilution in phosphate buffered saline), which was then detected by chemiluminescence with SuperSignal Chemiluminescent Substrate (Pierce Chemical Co., Rockford, IL) using Kodak X-Omat films (Eastman Kodak Co., Rochester, NY). The protein amount was determined by densitometry using Kodak 1D software and normalized to protein load. Common samples were included in each blot to serve as internal controls. Positive (purified proteins) and negative (blocking peptide or blot without primary antibodies) controls were used to establish the specificity of the protein signals.

Each experiment lasted 170 min and began after 30 min of equilibration, 30 min of dye loading, and a 20-min dye washout period. Hearts were randomly divided into seven groups of eight hearts each. Untreated time controls (no ischemia) were not subjected to ischemia or any treatment. The ischemia (ISC) group was subjected to 30 min of ischemia and 60 min of reperfusion. The same time protocols were used in the KBR (Tocris Cookson, Inc., Ellisville, MO; 10 μm), SEA (Taisho Pharmaceutical Co., Ltd, Tokyo, Japan; 1 μm), APC (sevoflurane, 2.2%, 0.41 ± 0.03 mm), and APC + KBR or SEA groups. KBR or SEA was given only during the initial 10 min of reperfusion. In the APC group, sevoflurane was given for 15 min followed by a 15-min washout before the onset of ischemia. In two other groups, KBR or SEA treatments were combined with APC.

All data are expressed as mean ± SEM rather than SD because the number of animals per group was identical. The analysis of variance for repeated measures (Super Anova 1.11 software for Macintosh; Abacus Concepts, Berkeley, CA) was used to assess within group differences over time. Data were compared among the grouped data at discrete time points before ischemia and during reperfusion. A two-way analysis of variance was used to assess the among group differences at these time points. Student-Newman-Keuls multiple-comparison post hoc  tests were used to differentiate within or among the group differences. Differences among the means were considered significant when P  was less than 0.05 (two-tailed).

Tables and graphs display changes in the control, ISC, KBR, SEA, APC, APC + KBR, and APC + SEA groups before ischemia and at 2, 10, 30, and 60 min of reperfusion. All noted changes are significant. Coronary flow was much lower than baseline throughout reperfusion in each group, but it was higher in each APC group than in the control, KRB, or SEA groups, which tended only to show increases (table 1). The heart rate was slower than baseline during early reperfusion in each group, and no differences were observed in heart rate among groups after 10 min of reperfusion (table 1). Oxygen consumption was lower than baseline throughout reperfusion in each group but was depressed less in each treated group than in the ISC group (table 1). Moreover, oxygen consumption in APC + KBR or SEA groups was higher than in KBR, SEA, and APC alone groups. Cardiac efficiency decreased during initial reperfusion and remained lower than baseline throughout reperfusion in each group, but was higher in each treated group than in the ISC group during reperfusion (table 1). Moreover, efficiency recovered better in APC + KBR or SEA groups than in KBR, SEA, and APC alone groups.

Systolic [Ca²⁺] decreased slowly over the first 10 min during ischemia and then increased abruptly during initial reperfusion in each group (fig. 1A). However, the increase in systolic [Ca²⁺] throughout reperfusion was less in the APC, APC + KBR, and APC + SEA groups wherein it returned to control levels after 5 min of reperfusion. Systolic [Ca²⁺] remained similarly higher in the ISC, KBR, and SEA groups. Systolic LVP decreased slightly in all APC treatment groups, approached 0 mmHg during early ischemia, and then increased with diastolic LVP during the last 10 min of ischemia (fig. 1B). During early reperfusion, systolic LVP tended (nonsignificant) to return to the control level in the KBR, SEA, APC, and APC + KBR or SEA groups when compared with the ISC group at 60 min of reperfusion.

Diastolic [Ca²⁺] increased incrementally during ischemia and increased abruptly and markedly during the initial reperfusion period in each group (fig. 2A). Diastolic [Ca²⁺] remained increased throughout reperfusion in each ischemia group but was lower in the KRB, SEA, and KRB or SEA + APC groups. In contrast, systolic [Ca²⁺] remained higher only in the ISC, KRB, and SEA groups. Diastolic LVP increased during the last 10 min of ischemia (fig. 2B). The time of onset of LVP diastolic contracture was earlier and its magnitude was greater in the ISC group than in all other groups. On reperfusion, diastolic LVP increased less in the KBR, SEA, and APC groups and even less in the APC + KBR and APC + SEA groups, but remained above baseline levels in each ischemia group throughout reperfusion.

The maximal time derivatives of Ca²⁺and LVP, d[Ca²⁺]/dt^max(cytosolic Ca²⁺influx; fig. 3A) and dLVP/dt^max(contractility; fig. 3B), approached zero during ischemia and increased abruptly during initial reperfusion in each group. The increases in d[Ca²⁺]/dt^maxand dLVP/dt^maxthroughout reperfusion were higher in all treated groups than in the ISC group after 60 min of reperfusion. The APC + KBR and APC + SEA groups exhibited better recovery of d[Ca²⁺]/dt^maxand dLVP/dt^maxcompared with the APC, KBR, and SEA alone groups.

The minimum time derivatives of d[Ca²⁺]/dt^min(cytosolic Ca²⁺efflux; fig. 4A) and dLVP/dt^min(relaxation; fig. 4B) were inversely but qualitatively similar to those of d[Ca²⁺]/dt^maxand dLVP/dt^max, i.e. , the return to baseline levels of d[Ca²⁺]/dt^minand dLVP/dt^min(relaxation) were better in the APC, KBR, and SEA groups but most complete in the APC + KBR or SEA groups.

Protein content of Ca²⁺release channel RyR2 was significantly reduced by IR injury (fig. 5). APC effectively prevented the attenuation of RyR2 protein loss, whereas KBR or SEA had no effect. In addition, the APC + KBR and APC + SEA groups gave results similar to those of the APC group alone. SERCA2a protein content was significantly reduced by IR injury. APC effectively reduced SERCA2a protein loss, whereas KBR or SEA had no effect. The APC + KBR or SEA groups exhibited results similar to those in the APC group alone. Phospholamban expression was unaltered by IR injury with or without APC, KBR, or SEA. Brief exposure to sevoflurane before ischemia exhibited no difference among the time control, ischemia, or APC groups on NCX protein content on reperfusion; ischemic preconditioning also had no effect (fig. 6).

This study demonstrates that the induction of APC (i.e. , treatment only briefly before ischemia), combined with inhibition of NCX during initial reperfusion, affords additive cardioprotection and reduced intracellular [Ca²⁺] loading in an isolated heart model. Moreover, the degradation of major SR Ca²⁺-cycling proteins with IR injury was attenuated by APC, but not by NCX inhibition. KBR and SEA are widely used as pharmacologic tools to study the roles of NCX at the cellular and organ levels. It was reported that KBR at 10 μm did not affect Na⁺–H⁺exchange, dihydropyridine-sensitive Ca²⁺uptake, passive Na⁺uptake, sarcolemmal Ca²⁺ATPase, SERCA2a, or Na⁺–K⁺ATPase.23SEA at concentration up to 3 μm did not affect other ion channels, receptors, and enzymes.17In the current study, we found that 10 μm KBR and 1 μm SEA exhibited similar cardiac functional recovery in our experiment model.

Reperfusion injury results in large part from cytosolic [Ca²⁺] overload. If the increase in intracellular [Ca²⁺] is prolonged, a cascade of events is initiated that ultimately results in lethal injury.24We have detailed the time course of change in contractility, and in relaxation, with cytosolic Na⁺, cytosolic and mitochondrial [Ca²⁺], and the reduced form of nicotinamide adenine dinucleotide during IR injury in the intact heart.20,25From these studies, it was apparent that contractile performance became dissociated from cytosolic [Ca²⁺], particularly during early reperfusion, because the higher Ca²⁺levels were associated with reduced contractile force and impaired relaxation. Our studies support the notion that altered Na⁺–H⁺and Na⁺–Ca²⁺exchange function are important factors in IR injury because Ca²⁺loading was temporally associated with Na⁺loading during early reperfusion.14,20 

Anesthetic preconditioning has been shown to be protective for several variables, such as contractile function, coronary flow, free radical generation during both ischemia and reperfusion, and infarct size.3,26Therefore, our understanding of the mechanisms underlying the myocardial protection by halogenated anesthetics remains incomplete. We reported that APC improved basal and nitric oxide–mediated coronary flow as well as cardiac rhythm, perfusion, mechanical, and metabolic function.3,27APC, like ischemic preconditioning, also reduced cytosolic [Ca²⁺] loading while it improved mechanical and metabolic function and contractile responsiveness to Ca²⁺on reperfusion.4In another study,11a high concentration of sevoflurane (3.5%) given for 10 min immediately before ischemia, for 10 min on initial reperfusion, or both before and after ischemia afforded cardioprotection by reducing dysrhythmias and improving mechanical and metabolic function associated with both reduced systolic and phasic Ca²⁺loading on reperfusion; sevoflurane before ischemia was slightly more protective than when given on reperfusion.

As in our previous studies,4,11the current results show that pretreatment with a halogenated anesthetic reduced systolic [Ca²⁺] loading and improved coronary flow during early reperfusion. Cardiac efficiency was also better restored on reperfusion in the APC alone and the NCX inhibitor groups compared with the ISC group. Moreover, APC plus treatment with KBR or SEA had a greater effect on improving cardiac efficiency than APC or NCX exchange inhibition alone. Because only viable cells consume oxygen and produce work, restored cardiac efficiency may indicate improved functioning of remaining viable cells over time during reperfusion. This work confirms and extends our previous results4,11,14,15on the individual cardiac protective effects of APC and Na⁺–H⁺exchange inhibition.

Unlike the Ca²⁺pump, the forward mode of NCX aids, but is not essential for, the extrusion of excess cell Ca²⁺, as shown in NCX knockout mice.28However, an increase in cell Na⁺can slow the forward mode operation of NCX or cause a reverse mode operation of NCX, which leads to an increase in intracellular [Ca²⁺].29It is believed that the activation of the reverse mode of NCX contributes to myocardial IR injury.13,30Use of a pharmaceutical agent to inhibit reverse mode NCX would be expected to protect against cardiac ischemic injury by reducing Ca²⁺overload during reperfusion. Previous reports have confirmed the beneficial effects of NCX inhibitors on myocardial IR injury.12,13,18In an investigation of the correlation between diastolic [Ca²⁺] and contractility, reperfusion after 30 min, low-flow ischemia was associated with a slower return in the rate–pressure product and a higher diastolic [Ca²⁺] than in hearts treated with 10 μm KBR 5 min before ischemia until 1 min of reperfusion.12Treatment with 10 μm KBR during initial reperfusion also improved mechanical work and energetics in cross-circulated rat hearts after 15 min of ischemia and 60 min of reperfusion.31The increase of diastolic [Ca²⁺] is likely an important determinant of contractile dysfunction after ischemia and reperfusion. In our study, administration of the inhibitors KBR or SEA, given only during the initial 10 min of reperfusion, also reduced diastolic [Ca²⁺].

The mechanism of cardioprotection by NCX inhibition is tied to the activity of other exchangers and ion pumps. Myocardial ischemia is characterized by ATP depletion and intracellular acidosis. These changes lead to inactivation of the Na⁺–K⁺ATPase pump and to activation of the Na⁺–H⁺exchanger, respectively, which resulted in intracellular Na⁺accumulation.32,33During early reperfusion, Na⁺accumulation is further accelerated by activation of Na⁺–H⁺exchange, which follows washout of extracellular H⁺.34Therefore, reverse mode NCX is most likely to occur during early reperfusion when Na⁺–H⁺exchange is high because of the large transcellular pH gradient. The reverse mode of NCX is then activated, and intracellular Ca²⁺overload occurs.35The increase in intracellular [Ca²⁺] leads to deleterious changes in cell function, including, importantly, the contractile apparatus.36 

In our study, a suppressive effect of NCX inhibitors KBR and SEA on the increases in diastolic [Ca²⁺] was clearly observed during the period of reperfusion. A previous study demonstrated that KBR administered before ischemia reduced the increase in cytosolic [Ca²⁺] during the late period of ischemia and early reperfusion.12Reduction of the peak increase in diastolic [Ca²⁺] during reperfusion is mediated by restoration of SR [Ca²⁺] uptake and mitochondrial Ca²⁺buffering [Ca²⁺], as well as Ca²⁺extrusion by the Ca²⁺pump and normal mode NCX (Ca²⁺efflux). Therefore, the more rapid decline in diastolic [Ca²⁺] on reperfusion in the KBR- or SEA-treated hearts than in nontreated hearts is likely a consequence of the inhibition of NCX.

Alterations in SR function have been observed after IR injury.37–39,40Measures of SR function have been estimated by SR Ca²⁺uptake, Ca²⁺release, and degradation of Ca²⁺-cycling proteins.39,41The loss of SR proteins with IR injury may be due to activation of proteases that are associated with intracellular [Ca²⁺] overload.42Our study suggests that APC diminishes the degradation of SR protein because APC prevented the loss of SERCA2a and RyR2 proteins. This is possibly mediated by one or more of the time-dependent kinase pathways known to be involved in APC. These downstream mechanisms require further study. KBR and SEA treatments, however, did not reduce the loss of the proteins observed after ischemia and reperfusion. This may not be surprising because the inhibition of NCX did not reduce the marked increases in systolic Ca²⁺after ischemia.

The restoration of Ca²⁺-cycling protein content is likely to be associated with improved SR function. The decrease in SERCA2a content after reperfusion was large, whereas that of its regulator was unchanged. Therefore, the ratio of phospholamban/SERCA2a protein content was increased on reperfusion. This suggests an enhanced inhibition of SERCA2a by phospholamban after ischemia. By not altering the protein content of phospholamban, while increasing SERCA2a content, APC improved this ratio. This may in part relieve the inhibition of SERCA2a by phospholamban and so improve SR Ca²⁺uptake during reperfusion. It is important to note that we found no difference among the time control, ischemia, and APC groups on NCX protein content after reperfusion; this suggests that neither ischemia nor previous exposure to sevoflurane induces a loss or a gain of this protein.

The proteolytic degradation or modification of Ca²⁺-cycling proteins by APC may be a downstream mechanism of Ca²⁺overload to mediate contractile dysfunction. The effects of APC on restoring SR protein content may be complimentary to the beneficial effects of APC on contractile proteins (such as troponin I) and cytoskeletal proteins (such as fodrin, desmin, and α-actinin) observed in other studies.43–45Indeed, in a preliminary study, we observed that APC and ischemic preconditioning promoted the binding of heat shock protein 25 and αΒ-crystallin to myofilaments during IR in isolated guinea pig hearts (M. T. Jiang, M.B., Ph.D., unpublished data, February 2002). Therefore, it is also possible that the phosphorylation and translocation of heat shock proteins to the cytoskeleton and or the myofilaments may play a cardioprotective role in APC.

Limitations of this study are that the experimental conditions do not necessarily represent the clinical setting of cardiac ischemia. The perfusate was cell free, and the animal model was one of global ischemia in nonejecting young hearts. Infarct size was not measured because the tissue was used for Western blotting. It is also possible that cardiac protection by NCX inhibition as shown here would not occur under clinical conditions. KBR and SEA likely inhibit the reverse mode (i.e. , Ca²⁺influx) of NCX more effectively than the forward mode.23,46,47Moreover, KBR may not be a specific selective inhibitor of NCX.17SEA has been reported to be a potent and highly selective inhibitor of reverse mode NCX; however, its selectivity is also controversial.48Therefore, a more highly selective reverse mode NCX inhibitor would be useful to study the role of NCX and the therapeutic potential of specific NCX inhibition. In a whole cell patch clamp study, sevoflurane and halothane both significantly reduced the outward and inward NCX current, which could limit the negative inotropic effects and help to maintain SR Ca²⁺content49; it was suggested that anesthetics may also reduce contractile failure by modulating NCX function during ischemia and reperfusion. In another study,50halothane and isoflurane was shown to interfere with neuronal cytosolic [Ca²⁺] regulation by inhibiting NCX. Therefore, an anesthetic itself may have some inhibitory effect on NCX.

Overall, our study demonstrates that APC, coupled with an NCX inhibitor, improves contractility and relaxation while reducing Ca²⁺loading by different means. Through a memory effect, APC may reduce systolic Ca²⁺loading and improve function in part by improving SR Ca²⁺pump and Ca²⁺release activities. The inhibition of NCX seems to acutely reduce diastolic Ca²⁺loading to improve function by inhibiting a reverse mode of NCX during initial reperfusion. Moreover, IR-induced degradation of major SR Ca²⁺-cycling proteins is attenuated by APC treatment before ischemia. Although APC and NCX inhibitors produce myocardial protective effects via  different signaling pathways, both interventions clearly lead to protection via  a common final mechanism that reduces cytosolic [Ca²⁺] loading. Inhalational anesthetics are often selected for patients with coronary artery disease at risk for ischemia and infarction during cardiac and noncardiac surgery. Previous administration of a inhalational anesthetic combined with NCX administered briefly during the initial reperfusion period may be a practical, effective method to preserve cardiac function after ischemia. Further research will determine how improved Ca²⁺homeostasis is a consequence of APC and NCX inhibition via  intracellular signaling pathways and transmembrane ion balance.