VOLATILE anesthetics are well known to have cardioprotective effects similar to those of ischemic preconditioning, 1–11but the mechanisms remain unclear. Recent evidence indicates that the protective effects of KATPchannel openers may be due to action on mitochondrial KATPchannels. 12–16We have previously reported that 15 min of volatile anesthetic exposure induces mitochondrial flavoprotein oxidation via mitochondrial KATPchannel activation, 17which may contribute to anesthetic-induced preconditioning. Volatile anesthetic exposure produces a memory period after exposure during which damage from ischemia is diminished, even though the anesthetic is no longer being administered. 8We designed the present study to examine time factors that may influence mitochondrial oxidation in anesthetic-induced preconditioning using mitochondrial flavoprotein fluorescence (MFF) as an endogenous reporter of mitochondrial redox state. 13,14,18
US National Institutes of Health Standards (NIH Publication #95–23, Revised 1996) and our Institutional Animal Care Committee policies were followed.
Cardiac myocytes were isolated from ventricles of guinea pigs (200–300 g) as previously described. 17Briefly, during pentobarbital anesthesia, the hearts were quickly excised, mounted on a Langendorff apparatus, and perfused via the aorta with an oxygenated enzyme solution containing Joklik minimum essential medium, 0.4 mg/ml collagenase (Gibco, Grand Island, NY) and 0.17 mg/ml protease (Sigma, St. Louis, MO). The digested ventricular tissue was chopped coarsely and shaken in a water bath for further dispersion. The cells were filtered, centrifuged, and washed. Rod-shaped cells with clear borders and striations were used within 8 h of isolation.
Flavoprotein Fluorescence Measurements
Cells were superfused with modified Tyrode solution containing (in mm) 140 NaCl, 5 KCl, 1 MgCl2, 10 HEPES, and 2 CaCl2(pH 7.4 with NaOH) at 21°C. MFF 12–14was excited every 30 s with light from a xenon laser bandpass filtered to 488 ± 20 nm, and emission was recorded through a 515-nm long pass filter. Relative fluorescence was averaged during excitation, and fluorescence intensity was expressed as arbitrary units (au; range, 0–256). At the end of each protocol, the maximum MFF was determined by administration of dinitrophenol (100 μm), and the minimum MFF was determined by administration of cyanide (4 mm).
From previously published data, 15 min of inhalational anesthetic exposure was shown to be effective for preconditioning in vitro 9and in vivo . 2,7Therefore, we used a 15-min period of isoflurane (Abbott Laboratories, Chicago, IL) exposure that was preceded by 10 min of equilibration with glucose-free Tyrode buffer (fig. 1). For a repeated exposure protocol, isoflurane was administered three separate times for 15 min interposed by 10-min washout intervals. For a sustained exposure protocol, isoflurane was administered for 45 min followed by 10 min of washout. To test the effect of a long delay interval between repeated exposures, a 15-min exposure of isoflurane was followed by 45 min of washout and then a second exposure lasting 15 min. In each protocol, two sets of isoflurane concentrations were used, which were measured by gas chromatography.
Data are presented as mean ± SD, and the number of cells is shown as n. Analysis of variance for repeated measurements and Scheffé tests were used to verify differences in fluorescence data.
With repeated isoflurane administration (figs. 2A and B), second and third exposures produced a peak MFF response at the beginning of each exposure rather than a sustained increase to a peak, as seen during the first exposure (fig. 2A). MFF concentrations at the end of each exposure were sequentially decreased both during high-concentration isoflurane exposure (32.1 ± 4.1, 30.5 ± 4.5, and 27.8 ± 2.6 au for the first, second, and third exposure, respectively;P < 0.0001) and low-concentration isoflurane exposure (26.6 ± 3.2, 23.0 ± 3.1, and 18.7 ± 1.9 au for the first, second, and third exposure, respectively;P < 0.0001;fig. 2B). During administration of low-concentration isoflurane, MFF at the end of the third exposure was less than the baseline concentration (22.6 ± 1.5 vs. 18.7 ± 1.9 au for baseline vs. third exposure, P < 0.01;fig. 2B). During this protocol, isoflurane concentrations were 1.5 ± 0.3 mm (high) and 0.8 ± 0.3 mm (low).
At the beginning of sustained (45-min) isoflurane exposure, high-concentration isoflurane showed stronger effects on MFF than low-concentration isoflurane (22.4 ± 2.1 vs. 31.6 ± 2.5 au for low-concentration vs. high-concentration isoflurane, P < 0.01;figs. 3A and B). By the end of the exposure, a decreasing MFF during high-concentration isoflurane exposure resulted in concentrations lower than those during low-concentration isoflurane exposure (29.2 ± 2.9 vs. 23.7 ± 3.9 au for low-concentration vs. high-concentration isoflurane exposure, P < 0.01). After washout, although MFF in high-concentration isoflurane returned to the same concentration as baseline (23.3 ± 2.2 vs. 23.7 ± 3.9 au for baseline vs. washout, P = 0.81), the MFF during low-concentration isoflurane exposure was still higher than baseline (22.4 ± 2.1 vs. 29.2 ± 2.9 au for baseline vs. washout, P < 0.01). In this protocol, measured isoflurane concentrations were 1.5 ± 0.1 mm (high) and 0.9 ± 0.1 mm (low).
In the third protocol, with delayed reexposure after 45 min of washout, MFF of cells previously exposed to low-concentration or high-concentration isoflurane were both below original baseline (low-concentration isoflurane: 21.1 ± 2.2 vs. 18.7 ± 2.1 au for baseline vs. after washout, P < 0.05; high-concentration isoflurane: 22.4 ± 1.4 vs. 20.6 ± 1.6 au for baseline vs. after washout, P < 0.05;figs. 4A and B). A second exposure of isoflurane did not increase MFF in either group (low-concentration isoflurane: 18.7 ± 2.1 vs. 19.2 ± 3.3 for after washout vs. second exposure, P = 0.73; high-concentration isoflurane: 20.6 ± 1.6 vs. 22.1 ± 2.2 for after washout vs. second exposure, P = 0.14). In this protocol, measured isoflurane concentrations were 1.7 ± 0.2 mm (high) and 0.8 ± 0.1 mm (low).
The principal findings of these experiments are that alterations in mitochondrial redox state induced by isoflurane are dependent on the concentration and temporal pattern of exposure. Whereas brief exposure to volatile anesthetic results in mitochondrial flavoprotein oxidation, sustained exposure, especially to high concentrations of isoflurane, leads to depression of oxidation and failure to respond to subsequent anesthetic exposure.
It is unknown whether mitochondrial oxidation or reduction is advantageous for myocardial protection during ischemia. Ischemic preconditioning reduces myocardial oxygen consumption during ischemia, 19which may be accounted for by decreased contractile activity and by depression of mitochondrial respiration by nitric oxide. It is therefore possible, despite our earlier findings of an initial volatile anesthetic oxidation response, 17that isoflurane ultimately reduces mitochondrial oxidation and myocardial oxygen consumption, thereby inducing myocardial protection during ischemia.
From the present data on repeated isoflurane exposure, it is evident that the first exposure activates processes that change the sensitivity for subsequent mitochondrial oxidation and reduces the oxidizing effect of subsequent isoflurane exposures. This negative effect persists after a 45-min washout interval. The depressed resting concentrations of oxidation even 45 min after isoflurane exposure provide further evidence supporting a prolonged decreased oxidation in cells previously exposed to isoflurane.
We observed that, unlike exposure to high-concentration isoflurane, 45 min of exposure to low-concentration isoflurane continues to increase the oxidation concentration. We speculate that there is a difference in threshold between oxidizing and suppressive pathways for isoflurane-induced changes. We are not sure why repeated exposure to the same low concentration of isoflurane shows a dominance of the suppressive effect on oxidation.
Inhibition of the respiratory chain or addition of uncouplers of oxidative phosphorylation can both limit the extent of enzyme release in the intact heart and prevent the onset of irreversible morphologic changes in isolated myocytes. 20Volatile anesthetics are known to have mitochondrial uncoupling effects that may reduce the damage to myocytes. Both of these facts are compatible with our hypothesis that the beneficial preconditioning effects of volatile anesthetic exposure are due to subsequent inhibition of myocyte oxidation. One possible intracellular pathway mediating these effects may be the generation of free radicals that trigger ischemic preconditioning memory and activation of kinases. 21
Because our experiments examined isolated, nonbeating myocytes at room temperature, we cannot be certain that the mitochondrial redox state in the beating and warm heart responds in the same fashion. Nonetheless, our observation of the contrasting response to a single exposure compared with repeated or prolonged exposure indicates unexpected complexity of isoflurane effects on mitochondrial respiration.