Attenuation of Mitochondrial Respiration by Sevoflurane in Isolated Cardiac Mitochondria Is Mediated in Part by Reactive Oxygen Species

All investigations conformed to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health No. 85–23, revised 1996), and were approved by the Institutional Animal Care and Use Committee (Medical College of Wisconsin, Milwaukee, Wisconsin). Thirty mg of ketamine and 1,000 U of heparin were injected intraperitoneally into 20 albino English shorthaired guinea pigs (weight, 250–300 g). The animals were decapitated 15 min later when unresponsive to noxious stimulation. Immediately after a thoracotomy, the heart was taken out and immersed in a 4°C cold isolation buffer of 200 mm mannitol, 50 mm sucrose, 5 mm KH²PO⁴, 1 mm EGTA, 5 mm 3-(n-morpholino)propanesulfonic acid, and 0.1% bovine serum albumin (pH 7.15, adjusted with potassium hydroxide). The atria were discarded, and the ventricles were minced into 1-mm pieces. The tissue was rinsed, transferred to a glass Potter-Elvehjem homogenizing vessel (Kontes Glass Co., Vineland, NJ) on ice, and gently homogenized with a Teflon (DuPont, Wilmington, DE) pestle for 30 s in the presence of 1 mg/ml protease. This was followed by another 30 s of homogenization after 10-fold dilution of the protease. The mitochondria were then isolated by differential centrifugation at 4°C. 19The tissue suspension was centrifuged at 8,000 g  for 10 min to remove the protease. The resulting pellet was then resuspended in 28 ml isolation buffer, and the suspension centrifuged at 700 g  for 10 min to remove cellular debris. The supernatant containing the mitochondrial fraction was further centrifuged at 8,000 g  for 10 min. The pellet was resuspended in 7 ml isolation buffer without EGTA and centrifuged at 8,000 g  for 10 min. The final mitochondrial pellet was resuspended in 500 μl cold isolation buffer without EGTA. The total protein concentration was determined 20with bovine serum albumin as a standard. Anatomic integrity of the isolated mitochondria was verified by electron microscopy.

The 500-μl mitochondrial suspension was kept at 4°C. Immediately before each experiment, an aliquot of the concentrated mitochondria was added to 27°C respiration buffer 21(110 mm KCl, 5 mm K²HPO⁴· 3 H²O, 10 mm 3-[n-morpholino]propanesulfonic acid, 10 mm Mg-acetate, 1 mm EDTA, 1 μm tetrasodium pyrophosphate, 0.1% BSA; pH 7.15, adjusted with potassium hydroxide) to yield 500 μl with a concentration of 500 μg protein/ml. Mitochondria from one heart were sufficient for approximately 15 experiments. Mitochondrial oxygen concentration was measured polargraphically with a Clark-type oxygen electrode (Model 1302; Strathkelvin Instruments, Glasgow, Scotland) in a water-jacketed 500-μl chamber (Model MT200A; Strathkelvin Instruments), equipped with a Teflon-coated magnetic stirring bar and monitored by an oxygen meter (Model 782; Strathkelvin Instruments). The oxygen electrode was calibrated with air-saturated water (Po²∼ 150 mmHg) and sodium sulfite (Na²SO³) solution (Sigma, St. Louis, MO; to achieve near zero Po²) at the same temperature as the buffer to be used. The rate of state 3 mitochondrial respiration was determined as the maximum rate of oxygen decrease after addition of substrate and adenosine-diphosphate, 21as shown in figures 1 and 2. Data were stored on-line using the manufacturer’s software (Strathkelvin Instruments). Microsoft Excel software (Microsoft Corp., Redmond, WA) was later used for analysis.

After sealing the chamber with a Plexiglas plug (time, t = 0 min), drugs, substrates, and adenosine-diphosphate (5 μl each) were subsequently injected into the chamber according to the protocol displayed in figure 1(all of the following concentrations are final). The superoxide dismutase mimetic manganese(III) tetrakis (4-benzoic acid) porphyrin chloride (TBAP; OxisResearch, Portland, OR; 10 μm), 22the negatively-charged spin trap N -tert-Butyl-a-(2-sulfophenyl)nitrone sodium (SPBN; Sigma; 10 μm), 23the mK^ATPchannel blocker 5-hydroxydecanoic acid (5-HD; Sigma; 300 μm), the nonspecific K^ATPchannel blocker glibenclamide (Sigma; 2 μm), 24or their respective vehicle (buffer) was added at t = 1 min to test for possible antagonism of a sevoflurane-induced effect on mitochondrial ETC. Pyruvate with malate (Sigma; 10 mm each dissolved in buffer) or succinate (Sigma; 10 mm dissolved in buffer) with the complex I blocker rotenone (Sigma; 10 μm dissolved in dimethyl sulfoxide) was added at t = 2 min as a substrate for complex I or complex II of the ETC, respectively. Sevoflurane (Abbott Laboratories, North Chicago, IL; 0.13, 0.39, 1.3, or 3.9 mm) or its vehicle dimethyl sulfoxide was added at t = 3 min to test for a sevoflurane-induced alteration of mitochondrial respiration. Adenosine-diphosphate (Sigma; 250 μm) in buffer was added at t = 4 min to start state 3 respiration. In some experiments, 2,4-dinitrophenol (Sigma; 100 μm in dimethyl sulfoxide) as an uncoupler, or antimycin A (Sigma; 100 μm in dimethyl sulfoxide) as a blocker of complex III, were given at t = 3 min to verify mitochondrial function using the chosen protocol (fig. 2A). Chamber oxygen concentration in μM was monitored for up to 12 min or until the oxygen concentration was 0. Representative tracings of 1.3 mm sevoflurane and sevoflurane plus TBAP are displayed in figure 2B. All experiments were performed at 27°C. Experiments with mitochondria from the same animal were randomized to one of the above treatment groups with at least three control experiments interspersed. All state 3 respiration rates from experiments of one heart were normalized and expressed as a percentage of control experiments from the same heart.

All data were expressed as mean ± standard error of the mean. Group data were compared by analysis of variance to determine significance (Super ANOVA 1.11 software for Macintosh; Abacus Concepts, Berkeley, CA). If F  values (P < 0.05) were significant, post hoc  comparisons of means tests (Student-Newman-Keuls) were used to compare the groups. Differences among means were considered statistically significant when P < 0.05 (two-tailed). Regression analysis was used to determine the relationships between sevoflurane concentration and the change in state 3 respiration for both substrate combinations. Statistical symbols used were * (vs.  Con) and † (vs.  1.3 mm sevoflurane).

Sevoflurane attenuated state 3 respiration in a dose-dependent manner when compared to vehicle controls for both substrate combinations (fig. 3). Regression analysis showed exponential relationships between sevoflurane concentration and attenuation of state 3 respiration:y = 77.0 ± 2.7 − (10.8 ± 5.2 · log x ) for pyruvate with malate (R² = 0.11; P < 0.05), and y = 86.2 ± 1.2 − (15.7 ± 2.0 · log x ) for succinate with rotenone (R² = 0.67; P < 0.05), with y  being the percentage of state 3 respiration compared to control experiments and x  being the sevoflurane concentration.

Neither the ROS scavengers TBAP or SPBN (fig. 4), nor the putative K^ATPchannel blockers glibenclamide or 5-HD (fig. 5), had an effect on state 3 respiration when given alone. Both TBAP and SPBN abolished the attenuation of state 3 respiration by 1.3 mm sevoflurane when pyruvate with malate was given (fig. 4A). This attenuation of state 3 respiration was not observed when succinate with rotenone was given (fig. 4B). Neither glibenclamide nor 5-HD reversed the sevoflurane-induced attenuation of mitochondrial respiration for either substrate combination (fig. 5).

The present study shows that (1) sevoflurane exposure causes a dose-dependent attenuation of mitochondrial state 3 respiration in isolated intact mitochondria; (2) the attenuated electron transport is not limited to complex I; (3) the results with 5-HD and glibenclamide as K^ATPchannel antagonists suggest that this effect is independent of mK^ATPchannel opening; and (4) the reversal by the ROS scavengers TBAP and SPBN for complex I substrate (pyruvate with malate), but not for complex II substrate (succinate with rotenone), suggests that attenuated electron transport at complex I, but not at other sites, is mediated by superoxide.

Usually, cardioprotection by APC in different models is assessed by a reduction in infarct size 2–6,8,11,25–28and improved return of mechanical or metabolic function 8,28,29after ischemia/reperfusion injury compared to nonpreconditioned hearts. Reversal of this cardioprotection by antagonists of suspected signaling pathways or receptors is used to identify some of the mechanisms involved. 2,5,25–27,30 

Another approach is to measure expected intracellular changes during the preconditioning stimulus, i.e. , exposure to the volatile anesthetic, in the presence or absence of known antagonists of APC to identify those cellular or subcellular effects of the anesthetic exposure that are relevant for triggering APC. For example, in addition to increased ROS fluorescence, 7we recently showed that temporary exposure to a preconditioning concentration of sevoflurane caused a reversible increase in NADH fluorescence in intact hearts, and that this was reversed by 5-HD. 28 

Without the on-line NADH measurement, our findings would have confirmed the findings of other studies that volatile anesthetics such as sevoflurane induce preconditioning by opening mK^ATPchannels. The observed increase in NADH fluorescence during the anesthetic exposure, however, suggested attenuation of mitochondrial respiration, i.e. , a more reduced mitochondrial state, 31rather than a more oxidized state as expected with mK^ATPchannel opening. 12,32,33Furthermore, the paradoxic reversal of this NADH increase by 5-HD casts some doubt on the specificity of 5-HD as a mK^ATPchannel blocker. Hanley et al. , using submitochondrial preparations, suggested that 5-HD, a fatty acid, could serve as a substrate to feed electrons into the ETC at the level of coenzyme Q. 34,35This could provide a bypass for ETC sites that are attenuated by lipophilic drugs, 36like the K^ATPchannel openers diazoxide and pinacidil 13,34or volatile anesthetics. 35 

Taken together, these findings offer not only the possibility that volatile anesthetics may attenuate ETC function to induce APC, but also that blockade by 5-HD does not necessarily furnish direct evidence of a mK^ATPchannel opening. 37,38This has already been acknowledged in the more recent literature. 14,39,40Moreover, there is an ongoing debate about whether volatile anesthetics open mK^ATPchannels under physiologic conditions (no shortage of substrate, oxygen, or adenosine-diphosphate to make enough ATP) and this opening triggers APC, 14or whether they open mK^ATPchannels faster and to a greater extent under nonphysiologic conditions, like ischemia, and mK^ATPchannels hence serve as an effector of APC. 12,41 

Increasing evidence implies that ROS play a central role in triggering cardiac APC. Not only was APC abolished when different ROS scavengers were given, 4–7but exposure to a preconditioning concentration of a volatile anesthetic also led to a measurable, albeit temporary, increase in ROS formation as assessed by ethidium-fluorescence. 7,14This transient increase in ROS during anesthetic exposure could subsequently lead to protein kinase C activation 11and mK^ATPchannel opening, 42and thus result in a memory effect that lasts beyond the discontinued anesthetic exposure.

How and where ROS are formed during anesthetic exposure remains unclear. As a consequence of opening of mK^ATPchannels, increased mitochondrial electron transport 32could result in increased constitutive formation of superoxide and thus lead to preconditioning. 43This would implicate mK^ATPchannel opening as a trigger of preconditioning. Attenuated electron transport, on the other hand, can also increase ROS formation. 44–46The present study clearly demonstrates a dose-dependent attenuation of mitochondrial respiration during anesthetic exposure in isolated cardiac mitochondria under state 3 conditions. Together with our previous reports of an increase in reduced NADH fluorescence 8,28and a decrease in oxidized flavoprotein fluorescence 47during anesthetic exposure in intact beating hearts, we suggest that anesthetic-induced attenuation of electron transport is the likely mechanism for ROS formation.

Mitochondrial swelling, as would be expected with K⁺influx during K^ATPchannel opening, might also indirectly attenuate ETC function. 13We chose the two chemically different K^ATPchannel blockers glibenclamide and 5-HD at otherwise effective concentrations (2 μm and 300 μm, respectively) 24,29to test this hypothesis. Both failed to prevent the sevoflurane-induced attenuation of ETC function, suggesting that mK^ATPchannel opening does not initiate this effect.

Altered ETC function may not only result in increased ROS generation but may also be caused by ROS. 15–18Previous studies on ROS in triggering APC 4–7,14did not examine a possible mediating role of ROS in altering ETC function during anesthetic exposure. We used two different ROS scavengers to test this possibility: TBAP as a mimetic of superoxide dismutase (O^●̶2→ H²O²) and SPBN as a free-radical spin-trap. Our findings in isolated cardiac mitochondria show a differential role for ROS in sevoflurane-induced ETC attenuation. When succinate (in the presence of rotenone) was used as the substrate to feed electrons to oxygen via  complexes II, III, and IV of the ETC, none of the ROS scavengers affected the sevoflurane-induced attenuation of state 3 respiration. When pyruvate with malate was used as the substrate to produce mainly NADH feeding electrons to oxygen via  complexes I, III, and IV of the ETC, both scavengers abolished the sevoflurane-induced attenuation of state 3 respiration. Taken together with our preliminary findings 48that scavenging ROS abolished the increase in NADH fluorescence during anesthetic exposure in intact beating hearts, we suggest that complex I function is attenuated by superoxide formed during sevoflurane exposure, whereas the effect of sevoflurane on other complexes is apparently not mediated by superoxide. This also offers the interesting possibility of a positive feedback mechanism of ROS likely formed at complex III 9to attenuate complex I function, and thus leading to more ROS formation.

In summary, the present results indicate that sevoflurane attenuates mitochondrial electron transport in isolated cardiac mitochondria. This effect appears to be partially mediated by ROS, but independent of mK^ATPchannel opening. Understanding altered mitochondrial function as a trigger for cardiac preconditioning by volatile anesthetics may provide an important component in deciphering the complexity of preconditioning as a means to attenuate ischemia/reperfusion injury.

The authors thank Ming Tao Jiang, Ph.D., Assistant Professor; James Heisner, B.S., Research Technologist; Samhita Rhodes, Ph.D., Postdoctoral Fellow, all of the Department of Anesthesiology; and Mary Kane, B.S., Research Technologist; Ellen Buchmann, B.S., Senior Research Technologist; and Harry T. Whelan, M.D., Professor, all of the Department of Neurology, Medical College of Wisconsin, Milwaukee, Wisconsin, for their valuable contributions to this study.