Anesthetic-induced Preconditioning Delays Opening of Mitochondrial Permeability Transition Pore via  Protein Kinase C-ϵ–mediated Pathway

Experimental procedures and protocols used in this study were in accordance with the Institutional Animal Care and Use Committee of the Medical College of Wisconsin (Milwaukee, Wisconsin). All conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 95-23, Revised 1996).26 

Ventricular cardiomyocytes were isolated from hearts of adult male Wistar rats (150–250 g) by enzymatic dissociation as previously described.5,6Cells were resuspended in HEPES-buffered Tyrode solution (in mm): 132 NaCl, 5 KCl, 1 CaCl², 1.2 MgCl², 5 d-glucose, 10 HEPES, pH 7.4. Cells were stored at room temperature. Myocytes were allowed to recover for 1 h and were used for experiments within 5 h. Only rod-shaped, quiescent myocytes with visible striations and no visible membrane damage were used for experiments.

To induce mPTP pore opening, cardiomyocytes were loaded with the fluorescent probe tetramethylrhodamine ethylester (TMRE, 100 nm; Invitrogen, Carlsbad, CA) for 25 min at room temperature. TMRE, a lipophilic cation, accumulates preferentially into mitochondria. On laser-illumination, TMRE generates ROS within mitochondria, which leads to opening of mPTP.12,27–30In some experiments, after incubation with TMRE, adult rat myocytes were loaded with calcein-AM (1.0 μm; Invitrogen) and cobalt chloride (2.0 mm; Sigma-Aldrich, St. Louis, MO) for 15 min at room temperature. Calcein-AM is de-esterified and distributed in mitochondria and cytosol, where cytosolic calcein fluorescence is quenched by cobalt chloride so that only the mitochondrial dye is seen. Selected regions of the myocyte (50 μm²) were subjected to laser-induced oxidative stress until mPTP opening had occurred, visualized as a collapse of mitochondrial membrane potential (ΔΨ^m),31,32and release of the fluorescent dye calcein (620 Da) from mitochondria.33Calcein release was used to verify the opening of mPTP independently from changes of ΔΨ^m.

Cells were imaged in a polycarbonate recording chamber (Warner Instruments, Hamden, CT) using a confocal microscope (Eclipse TE2000-U; Nikon, Tokyo, Japan) with a 60×/1.4 oil-immersion objective (Nikon). For TMRE fluorescence, cells were illuminated by use of the 543-nm emission line of a HeNe laser. The emitted fluorescence was collected at 590 nm. Calcein was excited with the 488-nm line of an Argon laser, and fluorescence intensity was recorded at 520 nm. Images were recorded with EZ-C1 2.10 software (Nikon). The zoom function was used to select the region of interest (50 μm²), and that region was scanned at 3.5-s intervals, typically between 90 and 120 image frames. The scanning speed was set to a pixel dwell time of 1.92 μs. The recorded image sequence (512 × 512 pixels) was used to yield changes in ΔΨ^msignal throughout the recording. A set of neutral-density (ND4 and ND8) filters was adjusted to minimize dye bleaching. To ensure comparability between experiments, all settings of the confocal microscope (pinhole size, detector sensitivity, pixel dwell time, and laser power) were identical in all experiments.

Images were analyzed using MetaMorph 6.2 software (Universal Imaging, West Chester, PA) and the NIH ImageJ software 1.41 (National Institutes of Health, Bethesda, MD). Intensity of a cell-free area was subtracted as background. After background subtraction, image series were corrected for photobleaching by normalization to a monoexponential decay that was calculated from the average intensities for the whole recording. Time required to induce mPTP opening (t^mPTP) was determined from ΔΨ^mrecordings. The peak signal value over recorded region (50 μm²) was normalized as 100%, and the lowest value as 0%. After normalization, the time required for a 50% decrease in signal was calculated and denoted as t^mPTP. In some experiments, the opening of mPTP was determined by constructing pseudolinescans or x, t plots vertically to wave of TMRE dissipation. With this analysis, each time frame is presented as a single pixel plotted on y axis (representing time), and the x axis represents the length of the selected region.

Isolated cardiomyocytes were placed in a recording chamber on the stage of the confocal microscope, and cells were allowed to settle for 10 min. APC was induced by exposing myocytes to isoflurane (Baxter, Deerfield, IL) dispersed in Tyrode solution by sonication and delivered to cardiomyocytes using the airtight glass syringe and polyethylene tubing system. Isoflurane was administered for 20 min before 5 min of washout. Control cells did not receive isoflurane. To investigate the involvement of the PKC, APC was performed in the presence of the isoform-nonselective PKC inhibitor chelerythrine (1 μm; Sigma-Aldrich). To determine the PKC subclasses contribution, isoform-specific PKC-δ inhibitor rottlerin (0.2 μm; Sigma-Aldrich) and PKC-ϵ inhibitor myristoylated PKC-ϵ V1–2 peptide (ϵV1–2, 1 μm; Biomol Research, Plymouth, PA) were applied during APC. To confirm contribution of mPTP, myocytes were also treated with the inhibitors of mPTP cyclosporin A (0.5 μm; Calbiochem, La Jolla, CA) and bongkrekic acid (50 μm; A.G. Scientific, San Diego, CA). To verify that ROS triggered mPTP opening, ROS scavenger 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox, 2 mm; Calbiochem) was used. All experiments were conducted at room temperature. The protocols for all experimental groups are illustrated in figure 1. At the end of each experiment, samples of buffer containing isoflurane were collected and analyzed by gas chromatography (Gas chromatograph GC-8A; Shimadzu, Kyoto, Japan). The average concentration of isoflurane used in this study was 0.63 ± 0.3 mm.

For measurement of mPTP opening in isolated mitochondria, rats were preconditioned in vivo  with isoflurane in the absence or presence of chelerythrine. In anesthetized animals (Inactin, 100–150 mg/kg) tracheotomy was performed, and trachea was cannulated. Animals then underwent mechanical ventilation with a rodent ventilator (Harvard Apparatus 683, South Natick, MA), using air-oxygen mixture. In the APC group, isoflurane was administered for 30 min and discontinued 15 min before isolation of mitochondria. In the APC-chelerythrine group, chelerythrine was administered as intravenous bolus (5 mg/kg). Ten minutes after chelerythrine injection, rats received isoflurane. In chelerythrine group, rats received only chelerythrine. Control rats did not receive isoflurane or chelerythrine (fig. 1B). Isoflurane was administered via  a vaporizer (Ohio Medical Products 100F, Madison, WI). Heart was excised after treatment, and mitochondria were isolated as described in the next paragraph. Isoflurane concentration was measured at the tip of the tracheotomy tube using an infrared gas analyzer that was calibrated with known standards. The concentration of isoflurane used for in vivo  preconditioning was 1.4%.34 

Cardiac mitochondria were isolated by homogenization and differential centrifugation as described previously.10The heart was quickly excised after thoracotomy, and the left ventricle was immersed in cold isolation buffer (in mM): 50 sucrose, 200 mannitol, 5 KH²PO⁴, 1 EGTA, 5 3-(N-morpholino)propanesulfonic acid, and 0.1% bovine serum albumin; pH 7.3 adjusted with KOH. The tissue was homogenized with Polytron homogenizer (IKA-Werke, Staufen, Germany), and mitochondria were then isolated by differential centrifugation. The final mitochondrial pellet was resuspended in cold isolation buffer without EGTA. Total protein concentration was assessed with detergent compatible protein kit (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard. Mitochondria were kept on ice and used within 4 h after isolation.

For isoflurane treatment of isolated mitochondria, the mitochondrial suspension was separated in two aliquots, which were subsequently diluted to 1 mg/ml mitochondria and incubated either with isoflurane (0.5 mm), which was dissolved in dimethyl sulfoxide, or with dimethyl sulfoxide alone for 15 min at room temperature. To remove isoflurane, both suspensions were then diluted in isolation buffer and centrifuged at 8000g  for 10 min (fig. 1C).

Opening of mPTP in isolated cardiac mitochondria was assessed by measuring ΔΨ^musing rhodamine 123 (50 nm; Invitrogen) in the presence of pyruvate (5 mm) and malate (5 mm). Excessive and/or prolonged mitochondrial Ca²⁺accumulation is associated with opening of the mPTP, which accelerates ion exchanges across the inner mitochondrial membrane and eventually leads to a sudden loss of ΔΨ^m, indicating a massive depolarization due to mPTP opening. Isolated mitochondria (0.5 mg/ml) were exposed to 5 μm pulses of Ca²⁺until ΔΨ^msuddenly decreased. Specificity for mPTP was confirmed with cyclosporin A (1 μm). The concentration of Ca²⁺(μM mg⁻¹of mitochondrial protein) necessary to trigger mPTP pore opening was measured.

Mitochondrial and cytosolic fractions from isoflurane-treated and untreated cardiomyocytes were prepared by differential centrifugation in 0.3 M mannitol, 0.1% bovine serum albumin, 2 mm EDTA, 10 mm HEPES, pH 7.4, as described previusly.35Protein samples (50 μg) were separated on a 7.5% sodium dodecyl sulfate gel, and transfered to immunoblot membrane (Bio-Rad). Western blotting was performed using a 1:200 dilution of a rabbit polyclonal anti- PKC-ϵ antibody (Santa Cruz Biotechnology, Santa Cruz, CA). The blots were stripped and reprobed with an antibody against subunit I of cytochrome c oxidase (Santa Cruz Biotechnology) as a marker for mitochondria. The blots were scanned and analyzed with NIH ImageJ 1.41.

Data were analyzed using Origin 7 software (OriginLab, Northampton, MA). Data are reported as means ± SD, and n refers to the number of experiments. In all experimental groups, cardiomyocytes or mitochondria were isolated from at least five different rats. Comparisons between groups were performed by one-way ANOVA and use of Tukey test for post hoc  analysis. Differences with P < 0.05 were considered significant.

Opening of mPTP can be detected in intact cells by measuring permeability of the inner mitochondrial membrane to the fluorescent dye calcein.33We tested whether photoexcitation induced dissipation of ΔΨ^mcoincides with calcein leakage from the mitochondria. Figure 2Ashows a typical recording of TMRE fluorescence obtained from a 50 × 50-μm region, as assessed by confocal microscopy. Fluorescence at 590 nm (TMRE) and between 515 and 525 nm (calcein) was recorded simultaneously from the same region. Photoexcitation of the selected region induced calcein to move from mitochondria to the cytosol concurrent with the loss of ΔΨ^m. In the cytosol, calcein fluorescence was quenched by cobalt chloride, resulting in a more diffuse staining that indicated an opening of mPTP. Figure 2Billustrates TMRE and calcein fluorescence from the same regions plotted as a function of time. The role for the mPTP was also confirmed using mPTP blockers cyclosporin A and bongkrekic acid. As shown in figure 2C, dissipation of ΔΨ^mwas significantly delayed in the presence of 0.5 μm cyclosporin A and 50 μm bongkrekic acid, respectively. To investigate whether ROS formation was involved in the loss of ΔΨ^m, we examined the effect of the free radical scavenger Trolox.28Treatment with 2 mm Trolox significantly slowed ΔΨ^mloss during laser-scan compared with control cell (fig. 2C). The data are summarized in figure 2D.

The effect of isoflurane on t^mPTPwas first examined in the presence of isoform-nonspecific PKC inhibitor, chelerythrine (1 μm). Figure 3Ashows representative recordings where APC by isoflurane produced a significant delay in t^mPTP. Chelerythrine prevented the isoflurane-induced effect. Chelerythrine at applied concentration had no significant effect on t^mPTPin control cells, as summarized in figure 3B.

The impact of in vivo  preconditioning on Ca²⁺-induced mPTP opening in isolated mitochondria was also assessed. Figure 3Cshows recordings where mitochondria were challenged with incremental Ca²⁺concentrations up to the point of dissipation of ΔΨ^m, indicating mPTP opening. In mitochondria from control animals, the Ca²⁺concentration required to open the mPTP was 45 ± 8 μm · mg⁻¹protein. After APC, the Ca²⁺concentration required to open the mPTP was significantly increased to 64 ± 8 μm · mg⁻¹protein (P < 0.05). This increase was attenuated when isoflurane was administered in the presence of PKC inhibitor chelerythrine (5 mg/kg), thus confirming the results obtained using isolated myocytes (fig. 3D).

To assess the effect of isoflurane in the absence of cytosolic components, isolated mitochondria were directly exposed to isoflurane before inducing mPTP with Ca²⁺. The time course in figure 4Ademonstrates that isoflurane was unable to delay mPTP opening. The Ca²⁺concentration required for mPTP opening was not significantly higher in isoflurane exposed compared to untreated mitochondria (45 ± 8 μm · mg⁻¹control vs.  47 ± 12 μm · mg⁻¹isoflurane). Figure 4Bsummarizes the effects of isoflurane on mPTP in isolated mitochondria.

To distinguish which PKC subclass is required for the APC-induced t^mPTPdelay, PKC-specific isoform inhibitors were applied. Cells were subjected to APC with isoflurane in the presence of isoform-specific blockers, rottlerin (0.2 μm), or ϵV1–2 (1 μm). As illustrated in figure 5A, the delay in t^mPTPwas still observed after APC in the presence of rottlerin, suggesting that PKC-δ is not involved for APC-induced delay of t^mPTP. However, delay of mPTP opening was completely blocked by 1 μm ϵV1–2. Both rottlerin and ϵV1–2, at the concentrations used, had no significant effect on t^mPTPin control cells, as shown in figure 5B. As illustrated by Western blot in figure 6A, PKC-ϵ concentration was increased significantly after isoflurane exposure in the mitochondrial fraction of cardiomyocytes, whereas it was slightly decreased in the cytosol (n = 3/group, data summary in fig. 6B). These results indicate that APC induces translocation of PKC-ϵ toward mitochondria and that PKC-ϵ is involved in the mechanism of isoflurane-induced delay of mPTP opening.

The current study investigated the role of PKC in isoflurane-induced delay in mPTP opening. Our results indicate that isoflurane treatment triggers PKC activation, which subsequently leads to inhibition of mPTP opening. More specifically, inhibition of PKC-ϵ translocation with ϵV1–2 abrogated the isoflurane-induced delay in mPTP opening, indicating a PKC-ϵ–dependent signal transduction pathway is involved. Accordingly, isoflurane caused translocation of PKC-ϵ toward mitochondria as evidenced by Western blotting. The fact that APC applied after chelerythrine in vivo  was not effective in delaying mPTP opening in isolated mitochondria underscored the importance of PKC signaling towards mitochondria as a target. In addition, isoflurane exposure of isolated mitochondria was not protective against mPTP opening, further suggesting that cytosolic proteins such as PKC play an important role in signal transduction to mitochondria in other to delay mPTP opening. The protective effect of isoflurane observed in the current study appears to be in part independent of direct, depolarizing effects of the anesthetic on mitochondria, as the cardioprotective benefits of isoflurane were demonstrated in cells with the same initial ΔΨ^m.10 

Mitochondria have been known as central mediators of cell survival in ischemia and reperfusion injury. Moreover, mPTP opening has been recognized as an important mediator of cell death.7Almost 30 yr ago, Hunter and Harworth described increased permeability of the inner mitochondrial membrane caused with high Ca²⁺.36Later, the role of mPTP opening in myocardial ischemia and reperfusion injury was also reported.37The molecular composition of mPTP is still under debate. Many models consider that mPTP is a multi-protein complex that spans inner and outer mitochondrial membranes under certain conditions. Proposed components forming the pore include the voltage-dependent anion channel, adenine nucleotide transporter, and cyclophilin-d.38However, knockout experiments questioned the involvement of voltage-dependent anion channel and adenine nucleotide transporter in mPTP formation. It is now thought that they may have a regulatory function on mPTP, whereas a critical role is still attributed to cyclophilin-d.39Pore opening results in the collapse of the ΔΨ^m, respiratory uncoupling, release of cytochrome c, and apoptosis-inducing factors.7Cardioprotection against ischemia and reperfusion injury by ischemic and pharmacologic preconditioning has been previously shown to involve prevention of mPTP opening.12Mitochondrial potassium channel opener diazoxide and PKC activation by phorbol-myristate acetate both had a protective effect against mPTP opening induced by high Ca²⁺loads in isolated mitochondria.13Modulation of mitochondrial function has also been considered a key component of cardioprotection by volatile agents.10Moreover, Piriou et al.  showed that APC by desflurane inhibits Ca²⁺-induced mPTP opening in isolated mitochondria.11 

The mechanisms by which anesthetics modulate mPTP opening are not completely elucidated. Previous studies demonstrated that the ischemic preconditioning-induced cardioprotection is PKC-dependent4,40and PKC isoform-specific, with δ and ϵ isoforms being the most important. Overexpression and association of PKC-ϵ with different mitochondrial proteins, including components of mPTP, leads to subsequent inhibition of mPTP.41It has also been reported that PKC delays the opening of mPTP through inhibition of glycogen synthase kinase-3β.12APC also causes activation of PKC42and, more specifically, translocation of PKC-ϵ and PKC-δ,14,19,24although controversy exists regarding the exact role of each isoform. APC with sevoflurane stimulates translocation of PKC-δ to mitochondria and PKC-ϵ to sarcolemma.14PKC-δ is translocated to mitochondria after preconditioning with opioids and adenosine.43,44In contrast, isoflurane caused PKC-ϵ translocation towards mitochondria and PKC-δ towards sarcolemmal membrane,14,24and increased PKC-ϵ phosphorylation.19In isolated hearts, pharmacologic inhibition of PKC-ϵ abolished sevoflurane-induced cardioprotection.18The importance of PKC-ϵ translocation toward mitochondria was also shown in human myocardial tissue samples of patients exposed to sevoflurane preconditioning undergoing on-pump coronary artery bypass graft surgery.45This is in agreement with our observation that PKC-ϵ is responsible for isoflurane-induced effect on mPTP. Interestingly, PKC-ϵ has been suggested to be constitutively present within cardiac mitochondria; as such, it directly confers protection without requiring translocation.46In our study, the observations that isolated mitochondria were protected from Ca²⁺-induced mPTP opening after in vivo  exposure to isoflurane and that protection was reversed by chelerythrine strongly support the role of cytosolic signaling pathways targeting mitochondria, in agreement with our PKC-ϵ translocation data. In fact, our results that mitochondria directly exposed to isoflurane did not exhibit delay in mPTP opening suggest that mitochondrial PKC-ϵ is not involved in the isoflurane effect on mPTP. Other groups have also found that translocation of PKC-ϵ is required for myocardial protection against ischemia and reperfusion injury.23,40,47 

Our study has a few limitations. We used a cellular model of oxidative stress to study the mechanism of APC-induced delay of mPTP opening.12,28,31,37,48This model simulates ROS production during the reperfusion of ischemic myocardium and may not include other contributors to mPTP opening in cardiomyocytes during reperfusion. On the other hand, our experiments on isolated mitochondria strongly support the protective effect of isoflurane on Ca²⁺-induced mPTP formation. Rottlerin is known to inhibit PKC-δ more potently than other PKC isoforms,14–16,24,49but other unspecific inhibitory effects may also exist. We did not investigate the effect of isoflurane preconditioning or inhibition of PKC on the outcome of ischemia and reoxygenation injury. However, other studies have confirmed involvement of PKC in anesthetic preconditioning-induced protection against ischemia and reoxygenation injury.18,19 

In conclusion, our study shows for the very first time that isoflurane activates PKC-ϵ–dependent signaling pathway targeted towards mitochondria, leading to a delay in mPTP opening under conditions of oxidative stress.

We thank Samantha J. Mueller, B.S., Research Technologist, for technical assistance in isolation of ventricular myocytes and David A. Schwabe, B.S., Research Technologist, for assistance in rat surgical procedures. We also thank Mary B. Ziebell, Research Technologist, for isoflurane measurements and Terri L. Misorski, A.A.S., Program Coordinator, for editorial assistance (all from Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin).