Mitochondrial depolarization underlies delay in permeability transition by preconditioning with isoflurane: roles of ROS and Ca²⁺

Abstract

During reperfusion, the interplay between excess reactive oxygen species (ROS) production, mitochondrial Ca²⁺ overload, and mitochondrial permeability transition pore (mPTP) opening, as the crucial mechanism of cardiomyocyte injury, remains intriguing. Here, we investigated whether an induction of a partial decrease in mitochondrial membrane potential (ΔΨ_m) is an underlying mechanism of protection by anesthetic-induced preconditioning (APC) with isoflurane, specifically addressing the interplay between ROS, Ca²⁺, and mPTP opening. The magnitude of APC-induced decrease in ΔΨ_m was mimicked with the protonophore 2,4-dinitrophenol (DNP), and the addition of pyruvate was used to reverse APC- and DNP-induced decrease in ΔΨ_m. In cardiomyocytes, ΔΨ_m, ROS, mPTP opening, and cytosolic and mitochondrial Ca²⁺ were measured using confocal microscope, and cardiomyocyte survival was assessed by Trypan blue exclusion. In isolated cardiac mitochondria, antimycin A-induced ROS production and Ca²⁺ uptake were determined spectrofluorometrically. In cells exposed to oxidative stress, APC and DNP increased cell survival, delayed mPTP opening, and attenuated ROS production, which was reversed by mitochondrial repolarization with pyruvate. In isolated mitochondria, depolarization by APC and DNP attenuated ROS production, but not Ca²⁺ uptake. However, in stressed cardiomyocytes, a similar decrease in ΔΨ_m attenuated both cytosolic and mitochondrial Ca²⁺ accumulation. In conclusion, a partial decrease in ΔΨ_m underlies cardioprotective effects of APC by attenuating excess ROS production, resulting in a delay in mPTP opening and an increase in cell survival. Such decrease in ΔΨ_m primarily attenuates mitochondrial ROS production, with consequential decrease in mitochondrial Ca²⁺ uptake.

damage during ischemia and reperfusion (I/R) of the heart involves complex processes where the cellular machinery itself becomes a source of deleterious mediators of injury. Such processes include excessive production of mitochondrial reactive oxygen species (ROS) and cellular Ca²⁺ overload (6, 37), which trigger opening of the mitochondrial permeability transition pore (mPTP) (16, 19), a critical event in the transition towards cell death (18). The mPTP opening instantly dissipates mitochondrial membrane potential (ΔΨ_m), ceases mitochondrial ATP production, and initiates cell death pathways (6).

Studies suggest that most of the cell death occurs during reperfusion (51, 53), which can be attenuated with antioxidants (52), indicating an important role of ROS. In cardiomyocytes, ROS are primarily generated at complexes I and III of the mitochondrial electron transport chain (50). During I/R, accumulation of cytosolic Ca²⁺, which drives accumulation of mitochondrial Ca²⁺ (45), is primarily attributed to insufficient ATP production and derangement of intracellular ion homeostasis (37). It is suggested that excess ROS production and mitochondrial Ca²⁺ overload are mutually dependent processes, where ROS can stimulate Ca²⁺ release into the cytosol from the sarcoplasmic reticulum (12, 55) and decrease cytosolic Ca²⁺ clearance by inhibiting Ca²⁺ pumps (25, 35). In turn, Ca²⁺ can enhance mitochondrial ROS production (6). The vicious cycle is closed when ROS and Ca²⁺ trigger mPTP opening in individual mitochondria that by itself causes a burst of ROS and mitochondrial Ca²⁺ release, resulting in a self-sustained progression of mPTP opening throughout the mitochondrial population. These processes have been described as ROS-induced ROS release (3, 56) and Ca²⁺-induced Ca²⁺ release from mitochondria (22). Thus, the disruption of these interdependent processes could have important implications in attenuation of injury and increase in cardiomyocyte survival.

Anesthetic-induced preconditioning (APC) elicits endogenous defense mechanisms that increase resistance of cardiomyocytes to I/R injury (8). Multiple protective pathways have been implicated in APC, including activation of mitochondrial ATP-sensitive K⁺ channels, as one of the effectors of protection (34). Interestingly, cardioprotection can also be mediated by mitochondrial uncoupling proteins and agents known to depolarize mitochondria (2, 14), indicating the importance of ΔΨ_m regulation for the resistance to I/R injury. Although cardioprotective effects of volatile anesthetics have been demonstrated in randomized clinical trials (30) and recognized by the American College of Cardiology/American Heart Association Guidelines (13), the mechanisms of protection are not completely understood.

To further delineate the mechanism underlying cardioprotection from oxidative stress by preconditioning with isoflurane, we designed our study to investigate the role of subtle changes in ΔΨ_m as an effector of protection. We hypothesized that APC induces partial decrease in ΔΨ_m that attenuates mitochondrial ROS generation in stressed cardiomyocytes, which leads to the delay in mPTP opening and subsequent increase in cell survival. The magnitude of APC-induced decrease in ΔΨ_m was mimicked with the protonophore 2,4-dinitrophenol (DNP) and reversed by the addition of pyruvate, which allows for the separation of the effects of changing ΔΨ_m from other effectors of protection by APC. We also investigated the interdependence between ROS production and mitochondrial Ca²⁺ accumulation in relation to mild changes in ΔΨ_m.

METHODS

The Institutional Animal Care and Use Committee of the Medical College of Wisconsin (Milwaukee, WI) approved the animal use and experimental protocols of this study.

Isolation of ventricular myocytes.

Cardiomyocytes were isolated from ventricles of adult male Wistar rats (180–250 g) using collagenase and protease for enzymatic dissociation, as previously reported (46). Rats were anesthetized with sodium thiobutabarbital (Inactin, Sigma-Aldrich, St. Louis, MO, 150 mg/kg, administered intraperitoneally). After isolation, myocytes were stored in Tyrode solution (in mM: 132 NaCl, 10 HEPES, 5 glucose, 5 KCl, 1 CaCl₂, 1.2 MgCl₂; adjusted to pH 7.4). Experiments were conducted at room temperature within 5 h after isolation using Tyrode solution.

Isolation of cardiac mitochondria.

Cardiac mitochondria were isolated from Wistar rats as reported previously (42). Hearts were placed 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] and homogenized with a T 25 disperser (IKA-Werke, Staufen, Germany). Mitochondria were isolated by differential centrifugation, and protein concentration was determined. Mitochondria were stored in isolation buffer on ice and used within 4 h after isolation. Experiments were conducted at room temperature. In the APC group, mitochondria were isolated from rats that were preconditioned in vivo with isoflurane.

Experimental procedures in cardiomyocytes and oxidative stress.

APC was induced by isoflurane (0.5 mM), which was dissolved in Tyrode solution by sonication and applied as depicted in Fig. 1. Isoflurane concentration was verified by gas chromatography (GC-8A, Shimadzu, Japan) and varied ± 10% of reported 0.5 mM. The protonophore DNP (200 nM, Sigma-Aldrich) was used to mimic the magnitude of APC-induced mitochondrial depolarization. Pyruvate, a respiration substrate that can increase ΔΨ_m, was titrated, and 2 μM was used to repolarize mitochondria following APC or DNP treatment. Hydrogen peroxide (H₂O₂, 250 μM) was used to induce oxidative stress in cell survival experiments, ROS production, and simultaneous cytosolic and mitochondrial Ca²⁺ measurements (33, 34). H₂O₂ was not used to trigger mPTP opening, since it caused cell hypercontracture, which hampered identification of pore opening. Instead, mPTP opening was instigated using a well-established approach, where oxidative stress was induced by laser-photoexcitation as described below and previously (42). H₂O₂ alone did not affect fluorescence probes [except 5-(and-6)-chloromethyl-2′,7′-dichlorofluorescein (CM-DCF)], as measured by spectrofluorometry.

Fig. 1.

Experimental procedures used for cardiomyocytes. A: protocols for mitochondrial membrane potential (ΔΨ_m) and mitochondrial permeability transition pore (mPTP) opening measurements. *Tetramethylrhodamine ethyl ester (TMRE) loading time in mPTP protocol was modified to ensure similar TMRE concentrations, despite differences in ΔΨ_m among experimental groups. B: protocols for cell survival experiments, reactive oxygen species (ROS) measurements, and simultaneous cytosolic and mitochondrial Ca²⁺ measurements. In ROS production measurements, H₂O₂ was discontinued after 20 min, followed by dye loading for 10 min to prevent oxidation of 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein (CM-H₂DCF) by exogenous H₂O₂. Detailed explanation of protocols is found in appropriate sections in methods. Ctrl, control; APC, anesthetic-induced preconditioning; Pyr, pyruvate; DNP, 2,4-dinitrophenol; Iso, isoflurane; Tyr, Tyrode solution.

Laser-scanning confocal microscopy.

In cardiomyocytes, measurements of ΔΨ_m mitochondrial redox state, ROS production, cytosolic and mitochondrial Ca²⁺, and mPTP opening were conducted using a laser-scanning confocal microscope (Leica TCS SP5, Mannheim, Germany) with a ×63/1.4 numerical aperture Apo oil objective lens. Fluorescent indicators were excited with 488 nm line of an argon laser and 543 nm green HeNe laser, and images were acquired using LAS AF software (Leica).

Measurements of ΔΨ_m in cardiomyocytes.

The ΔΨ_m-sensitive fluorescent dye tetramethylrhodamine ethyl ester (TMRE) (Invitrogen, Carlsbad, CA) and autofluorescence of mitochondrial flavoproteins (FPs) were used for the assessment of ΔΨ_m in cardiomyocytes (Fig. 1A) as previously described (32). Excitation and emission wavelengths for TMRE were λ_ex/λ_em = 543/560–610 nm and for FPs were λ_ex/λ_em = 488/500–550 nm. For a better representation of the effect of pyruvate on ΔΨ_m, pyruvate was used initially at the higher concentrations (250 μM) than in the rest of our experiments. To reverse the APC-induced mitochondrial depolarization, 2 μM pyruvate was used. Fluorescence values were normalized to baseline expressed as 100%.

Cardiomyocyte survival experiments.

Cardiomyocytes were exposed to H₂O₂-induced oxidative stress as depicted in Fig. 1B. The number of live cells was counted at the beginning and at the end of the experiment. Rod-shaped cells without membrane blebs that excluded Trypan blue were considered alive (46). Cell death was normalized to control (Ctrl), expressed as 100%.

Analysis of mPTP opening in cardiomyocytes.

Photoexcitation-generated oxidative stress was used to induce opening of mPTP that was observed as a rapid and complete mitochondrial depolarization using ΔΨ_m-sensitive dye TMRE (21, 24, 42, 56). The TMRE (100 nM) incubation time was adjusted as needed to ensure equal TMRE concentrations in all experimental groups, despite differences in ΔΨ_m. After TMRE loading, the dye was washed out, and 30 × 30 μm recording region of cardiomyocytes was exposed to narrowly focused laser scanning. Only cells with equal initial TMRE fluorescence intensity were included in the study, and the settings of the confocal microscope were consistent to ensure equal delivery of oxidative stress. As we noted previously, the rapid release of TMRE coincides with the release of calcein, a 620-Da molecule, confirming opening of the mPTP (42, 56). Cyclosporine A (1 μM) was used to inhibit mPTP opening. Arbitrary mPTP opening time was determined as the time of loss of average TMRE fluorescence intensity from the recorded region (excluding nucleus) by half between initial and residual fluorescence intensity (Fig. 4B). It corresponded to complete depolarization of ∼50% of mitochondria in the recorded region.

Fig. 4.

Stress-induced opening of mPTP is delayed by APC and DNP. A: photoexcitation-generated oxidative stress induces mPTP opening as observed by the rapid dissipation of TMRE fluorescence. B: the time by which TMRE fluorescence intensity decreased by half between values at the beginning and after global mPTP opening was considered as the “arbitrary mPTP opening time” (arrows) and was compared among experimental groups. AU, arbitrary units. C: cyclosporine A (CsA), an inhibitor of mPTP, extended arbitrary mPTP opening time compared with control, indicating a delay in mPTP opening. APC also increased mPTP opening time, which was abrogated by 2 μM pyruvate (APC + Pyr). D: DNP (200 nM) increased mPTP opening time, and 2 μM pyruvate (DNP + Pyr) attenuated this effect. Data are means ± SE; n = 6–12/group. *P < 0.05 vs. Ctrl; #P < 0.05 APC or DNP vs. APC + Pyr or DNP + Pyr, respectively.

Measurements of ROS production in cardiomyocytes.

ROS production was monitored using 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate acetyl ester (CM-H₂DCFDA) (Invitrogen) that, upon deesterification and oxidation by ROS, yields fluorescent CM-DCF. The fluorescence intensity of CM-DCF was acquired using 8,000-Hz resonant scanner of the confocal microscope with λ_ex/λ_em = 488/500–550 nm. After exposure of cardiomyocytes to 250 μM H₂O₂ for 20 min, H₂O₂ was washed out twice before cells were loaded with 4 μM CM-H₂DCFDA for 10 min. Data were normalized to initial value expressed as 100%.

Measurements of ROS production in isolated mitochondria.

Isolated mitochondria (0.5 mg/ml protein) were suspended in experimental buffer (in mM: 130 KCl, 5 K₂HPO₄, 20 MOPS, 2.5 EGTA, 5 pyruvate, 5 malate, 0.001 Na₄P₂O₇, and 0.1% BSA; pH adjusted to 7.3 with KOH) containing the fluorescence probe Amplex Red (12.5 μM, Invitrogen) and 0.1 U/ml horseradish peroxidase to monitor the rate of H₂O₂ production. The fluorescence intensity of resorufin that reflects H₂O₂ release was measured with a spectrofluorometer (QM-8, Photon Technology International) (λ_ex/λ_em = 530/583 nm). Excess ROS production was initiated by antimycin A (AA, 200 nM), a mitochondrial complex III inhibitor (9). In all experimental groups, the rates of increase in resorufin fluorescence intensity were analyzed after the addition of AA. To eliminate the contribution of changes in mitochondrial Ca²⁺ on ROS production, experiments were conducted in a Ca²⁺-free buffer and in the presence of EGTA.

Measurements of Ca²⁺ uptake in isolated mitochondria.

Mitochondrial matrix free Ca²⁺ was assessed after loading indo-1 AM (10 μM, Invitrogen) into mitochondria (10 mg/ml protein) (15). Residual dye was removed by centrifugation at 8,000 g, and the mitochondrial pellet was resuspended in experimental buffer without EGTA. Spectrofluorometry was used for the detection of indo-1 fluorescence intensity (λ_ex/λ_em = 350/405–460 nm). CaCl₂ (20 μM) pulses were added every 1 min. Mitochondrial autofluorescence was not affected by DNP. Changes in indo-1 fluorescence intensity are expressed as a percentage of baseline.

Simultaneous measurements of Ca²⁺ in cardiomyocyte cytosol and mitochondria.

Ca²⁺-sensitive dyes, rhod-2 AM (4 μM, Invitrogen) and fluo-4 AM (2 μM, Invitrogen), were used to simultaneously detect mitochondrial and cytosolic Ca²⁺, respectively, by confocal microscopy. The predominant localization of rhod-2 and fluo-4 in mitochondria and cytosol, respectively, was achieved following a two-step cold/warm loading protocol (28). Briefly, cells were incubated with rhod-2 AM at room temperature for 60 min, followed by 60-min incubation at 37°C, and 120 min of dye washout at 37°C. During the last 30 min of rhod-2 AM washout, cells were loaded with fluo-4 at 37°C, followed by 20-min dye washout at room temperature. The switch from room temperature to 37°C and back to room temperature was gradual. Excitation and emission wavelengths of rhod-2 and fluo-4 were λ_ex/λ_em = 543/560–610 and λ_ex/λ_em = 488/500–550 nm, respectively. Data were normalized to baseline expressed as 100%.

Statistical analysis.

Data are presented as means ± SE, with n indicating the number of independent experiments. Statistical comparisons were performed using one-way or repeated-measures analysis of variance with Tukey post hoc test where appropriate. Differences at P < 0.05 were considered significant.

RESULTS

APC-induced mitochondrial depolarization in cardiomyocytes was mimicked by DNP and reversed by pyruvate.

In isolated cardiomyocytes, ΔΨ_m was assessed using both ΔΨ_m-sensitive indicator TMRE and endogenous fluorescence of FPs. Figure 2A demonstrates the ability of pyruvate to induce sustained hyperpolarization of mitochondria (11), which will be used later to reverse drug-induced mitochondrial depolarization. Cardiomyocytes preconditioned with isoflurane (APC) exhibited partial mitochondrial depolarization, observed as increase in FP fluorescence intensity to 121 ± 6% of control (100%) (Fig. 2B) and decrease in TMRE fluorescence intensity to 78 ± 7% of control (Fig. 2C). At 200 nM, DNP had similar effects on FP and TMRE fluorescence intensities as APC, while 2 μM pyruvate reversed both APC- and DNP-induced changes in TMRE and FP fluorescence intensities (Fig. 2, B and C). This indicates that APC-induced mild mitochondrial depolarization can be mimicked with a low concentration of DNP and reversed by adding a low concentration of pyruvate.

Fig. 2.

APC- and DNP- induced mitochondrial depolarization in cardiomyocytes is reversed by pyruvate. Mitochondrial membrane potential was assessed by fluorescence of flavoproteins (FPs) and TMRE. A: the ability of pyruvate to increase the ΔΨ_m was demonstrated after the addition of 250 μM pyruvate. Following a transient FP oxidation and a decrease in ΔΨ_m, pyruvate induced a sustained FP reduction and mitochondrial hyperpolarization. B and C: APC and DNP (200 nM) increased FP and decreased TMRE fluorescence intensities compared with control. Addition of low concentration of pyruvate (2 μM) reversed the effects of APC and DNP, while pyruvate alone slightly hyperpolarized mitochondria. Data are means ± SE; n = 7–8/group. *P < 0.05 vs. Ctrl; #P < 0.05 APC or DNP vs. APC + Pyr or DNP + Pyr, respectively.

APC- and DNP-induced protection of cardiomyocytes is reversed by pyruvate.

Cell survival experiments were conducted to examine the cytoprotective effects of APC and the role of a partial decrease in ΔΨ_m. APC protected cardiomyocytes from oxidative stress, indicated by a decrease in cell death, 57 ± 5% of control (Fig. 3A). DNP, at the concentration that mimicked the magnitude of APC-induced decrease in ΔΨ_m, attenuated cell death, albeit less than APC (Fig. 3B). Addition of pyruvate at the concentration that repolarized mitochondria attenuated and abolished protection afforded by APC and DNP, respectively. This suggests that mitochondrial depolarization is, at least in part, a common mechanism of cytoprotection by APC and DNP.

Fig. 3.

Partial mitochondrial depolarization by APC and DNP increases survival of cardiomyocytes exposed to oxidative stress. Cell death after oxidative stress with H₂O₂ is expressed as percentage of dead cells in stress control (Ctrl). A: APC reduced cell death, and pyruvate (2 μM) (APC + Pyr) attenuated the cytoprotection. Pyruvate alone attenuated cell death. B: DNP (200 nM) also attenuated cell death, and coapplication of pyruvate and DNP attenuated DNP-induced cytoprotection (DNP + Pyr). Summary data from 8–9 experiments/group are means ± SE. *P < 0.05 vs. Ctrl; #P < 0.05 APC or DNP vs. APC + Pyr or DNP + Pyr, respectively.

Delay in mPTP opening by APC and DNP is reversed by pyruvate.

The mPTP opening, a trigger of cell death pathways, was assessed by following rapid decrease in TMRE fluorescence intensity, i.e., dissipation of ΔΨ_m (Fig. 4A). The arbitrary mPTP opening time was increased in the presence of cyclosporine A, an mPTP inhibitor, indicating a delay in mPTP opening (Fig. 4B). APC also increased arbitrary mPTP opening time (120 ± 4% of control), which was abolished in the presence of pyruvate (Fig. 4C). DNP also induced a delay in mPTP opening (136 ± 8% of control), an effect attenuated in the presence of pyruvate (Fig. 4D). The similar trend of a delay in mPTP opening by APC and DNP, which was in both groups reversed by pyruvate, indicated that a partial decrease in ΔΨ_m was an underlying protective mechanism in both groups.

Attenuation of ROS production by APC and DNP is reversed by pyruvate.

The H₂O₂ was used to induce oxidative stress. To assess endogenous ROS production, the CM-DCF fluorescence intensity was monitored after the washout of H₂O₂. Cells treated with H₂O₂ exhibited greater increase in CM-DCF fluorescence intensity than nontreated cells, indicating substantial increase in ROS production (Fig. 5, A and B). APC attenuated increase in CM-DCF fluorescence intensity in stressed cells (Ctrl vs. APC; 210.5 ± 5.0% vs. 181.0 ± 5.7%), indicating attenuation of ROS production (Fig. 5A). This effect of APC was abolished in the presence of pyruvate (APC + Pyr; 228.7 ± 14.0%). Consistent with the results in cell survival and mPTP opening experiments, DNP also attenuated ROS production, which was again reversed with pyruvate (Fig. 5B). These results suggest that a partial decrease in ΔΨ_m was also a common mechanism of attenuation of ROS production by APC and DNP.

Fig. 5.

APC and DNP attenuate ROS production in stressed cardiomyocytes. ROS production in cardiomyocytes preexposed to oxidative stress is measured by 5-(and-6)-chloromethyl-2′,7′-dichlorofluorescein (CM-DCF) fluorescence. A: compared with the nonstressed cells (time Ctrl), stressed cells (Ctrl) had a significantly greater increase in CM-DCF fluorescence, i.e., ROS production. APC decreased ROS production, and pyruvate (2 μM) reversed this effect (APC + Pyr). B: treatment of cells with DNP (200 nM) also attenuated ROS production compared with the stress control, which was again reversed by pyruvate (2 μM) (DNP + Pyr). Data are means ± SE; n = 7–12/group. *P < 0.05 vs. stress control; #P < 0.05 APC or DNP vs. APC + Pyr or DNP + Pyr, respectively.

In isolated mitochondria, partial decrease in ΔΨ_m attenuates excess ROS production, but not Ca²⁺ uptake.

Isolated mitochondria were used to specifically examine whether a partial decrease in ΔΨ_m directly affects ROS production and Ca²⁺ uptake. Excess ROS production was induced with AA added to isolated mitochondria in a Ca²⁺-free buffer (Fig. 6A). AA caused a substantial burst of ROS in the control group, which was attenuated to 80 ± 3% of control in mitochondria isolated from APC-treated animals, and to 66 ± 9% of control in mitochondria treated with DNP (200 nM) (Fig. 6B). Addition of DNP alone did not significantly alter ROS production under basal conditions before treatment with AA. Figure 6, C and D, demonstrates Ca²⁺ uptake into mitochondria after the addition of cumulative pulses of 20 μM Ca²⁺. Compared with the control, APC or DNP treatment did not attenuate mitochondrial Ca²⁺ uptake despite the mitochondrial depolarization. On the contrary, Ca²⁺ uptake increased in both groups. These results indicate that in isolated mitochondria, a subtle depolarization by DNP significantly attenuates excess ROS, but not Ca²⁺ uptake.

Fig. 6.

Low concentrations of DNP and APC attenuate ROS production but not Ca²⁺ uptake in isolated mitochondria. A: representative signal traces of resorufin fluorescence intensity measurements as indication of ROS production. Application of antimycin A (AA; 200 nM), a complex III inhibitor, substantially enhanced ROS production (solid line). AA-induced ROS production was decreased in the presence of 200 nM DNP or in mitochondria isolated from preconditioned animals (dashed lines). B: summary data of rates of increase in resorufin fluorescence intensity during application of AA. APC and DNP significantly attenuated antimycin A-induced ROS production. C and D: measurements of mitochondrial Ca²⁺ using indo-1 AM. Addition of 20 μM Ca²⁺ pulses to the buffer resulted in mitochondrial Ca²⁺ uptake, which was not attenuated by DNP (200 nM) or APC, but it was increased. Data are means ± SE; n = 7–8/group in B and 8–11/group in C and D. *P < 0.05 vs. Ctrl.

In cardiomyocytes, partial decrease in ΔΨ_m attenuates cytosolic and mitochondrial Ca²⁺ accumulation.

Our previous studies indicated that APC attenuates accumulation of cytosolic and mitochondrial Ca²⁺ (32, 48). Here we tested the effects of a partial decrease in ΔΨ_m on Ca²⁺ loading into cytosol (Fig. 7A) and mitochondria (Fig. 7B), measured simultaneously in cardiomyocytes exposed to oxidative stress. Application of H₂O₂ induced an increase in fluo-4 and rhod-2 fluorescence intensities, indicating increase in cytosolic and mitochondrial Ca²⁺, respectively, which is in agreement with previous studies showing ROS-induced Ca²⁺ release (55). A partial mitochondrial depolarization with DNP attenuated the increase in Ca²⁺-sensitive fluorescence in both compartments compared with control. This suggests that, unlike the results from the isolated mitochondria (Fig. 6D), a partial decrease in ΔΨ_m attenuates mitochondrial as well as cytosolic Ca²⁺ accumulation in cells exposed to oxidative stress.

Fig. 7.

Accumulation of cytosolic and mitochondrial Ca²⁺ in cardiomyocytes exposed to oxidative stress is attenuated by partial mitochondrial depolarization. Isolated cardiomyocytes were coloaded with cytosolic (A) and mitochondrial (B) Ca²⁺ indicators, fluo-4 AM and rhod-2 AM, respectively, and their fluorescence intensities were recorded simultaneously. Upon application of H₂O₂, fluo-4 and rhod-2 fluorescence intensities increased. In cells treated with DNP (200 nM), this effect was attenuated. Summarized data are means ± SE; n = 11–17/group. *P < 0.05 vs. Ctrl.

DISCUSSION

This study demonstrates the mechanisms by which a partial mitochondrial depolarization, elicited by preconditioning with isoflurane or by treatment with the uncoupling agent DNP, protected cardiomyocytes from oxidative stress. The similar trend in exerting cytoprotective effects when mitochondrial depolarization was induced by APC or DNP, and the reversal of these effects by repolarizing mitochondria with pyruvate, implies that a partial decrease in ΔΨ_m is a common mechanism of protection by both treatments. The ΔΨ_m-dependent attenuation of excess ROS production by mitochondria appears to play a central role in the adaptation to oxidative stress by APC, which induces a delay in oxidative stress-triggered mPTP opening, thereby increasing cell survival. Our results suggest that the attenuation of mitochondrial Ca²⁺ accumulation in cells with partly depolarized mitochondria following oxidative stress is not a direct result of mitochondrial depolarization, but an indirect effect of attenuated ROS production and oxidative stress-induced Ca²⁺ release into the cytosol and mitochondria.

Here, we describe a mechanism by which APC affords protection to cardiomyocytes against oxidative stress. Our data indicate that APC induced a mild decrease in ΔΨ_m and attenuated excess ROS production in cardiomyocytes exposed to oxidative stress. Comparable decrease in ΔΨ_m by the protonophore DNP also attenuated ROS production in stressed cells and in isolated mitochondria, as well. Moreover, a reversal of the ΔΨ_m by pyruvate attenuated these effects of APC and DNP, suggesting that a partial decrease in ΔΨ_m is a common effect of these treatments responsible for attenuation of ROS production. This is in agreement with previous studies that showed a strong correlation between ΔΨ_m and ROS production, where even a moderate decrease in ΔΨ_m of a few millivolts substantially decreased ROS production (29, 47). Oxidative stress, due to excess ROS production during early reperfusion, is suggested as one of the primary causes of cell injury, and its elimination by ROS scavengers can attenuate I/R injury (52). According to our results, it is possible that APC attenuates mitochondrial ROS production by slightly decreasing ΔΨ_m during reperfusion, when replenished oxygen allows ΔΨ_m buildup and the burst of ROS production. Indeed, it is suggested that transient inhibition of mitochondrial respiration during early reperfusion may exert cardioprotection (1, 7). However, it is unclear what mediates mitochondrial depolarization by APC. It is suggested that mitochondrial K⁺ channels, often identified as effectors of protection by preconditioning, could be these mediators (10, 27, 32, 34). However, their role in this mechanism is questionable since their activation has been correlated with both increase and decrease in ROS production (20, 49). These controversial findings may result from differential behavior of these channels depending on the metabolic state of the cell (44). Other interesting mediators of mitochondrial depolarization with strong cardioprotective effects are uncoupling proteins, which are in fact considered potential physiological regulators of mitochondrial ROS production (2, 38, 39). However, they have not been (yet) associated with APC.

APC and DNP also caused a delay in the opening of mPTP and improved cardiomyocyte survival during oxidative stress. Repolarization of mitochondria with pyruvate reversed these effects. This suggests that attenuation of ΔΨ_m-dependent endogenous excess ROS production, which is one of the major triggers of mPTP opening, delays pore opening and leads to increased survival of cardiomyocytes. Opening of mPTP is considered a critical step in the transition towards cell death during I/R, and many cardioprotective strategies, including APC, induce a delay in mPTP opening (16, 37, 42). While previous studies tested the role of mitochondrial K_ATP channels and protein kinase C as preconditioning signal mediators in APC-induced mPTP opening, here we describe the mechanism by which a partial decrease in ΔΨ_m, as an effector of protection by APC, leads to a delay of mPTP opening and subsequent increase in cardiomyocyte survival. The disparities between APC and DNP in the extent of cell protection may imply additional mechanism of protection by APC. For example, APC involves multiple protective pathways, such as activation of antiapoptotic proteins (43), sarcolemmal K⁺ channels (34), and protein kinase C (54), that may exert additional cardioprotective effects independent of changes in ΔΨ_m. Moreover, APC and the uncoupler DNP may not share the same mechanism to induce mitochondrial depolarization, as discussed above. However, the similar trend in attenuation of ROS production, delay in mPTP opening, and increase in cell survival between APC and DNP, and the reversibility of each by pyruvate, suggests that APC- and DNP-induced cytoprotection shares, at least in part, a common mechanism, a partial decrease in ΔΨ_m.

The mechanisms by which uncoupling agents induce cytoprotection are somewhat controversial. It has been suggested that another protonophore, carbonylcyanide p-trifluoromethoxyphenylhydrazone (FCCP), induces cardioprotection by generating a burst of ROS that triggers preconditioning, without detectable decrease in ΔΨ_m (4, 5). Contrary to this, we did not observe an increase in ROS production during DNP application, thus excluding the possibility of ROS-induced preconditioning by DNP in our model. It is possible that these differences are a result of variations in the approach to measure ΔΨ_m and ROS and/or the different nature of action and potency of DNP and FCCP, because it was shown that another uncoupling agent exhibited concentration-dependent, dual effect on mitochondrial ROS production (36). Consistent with our results, several groups have demonstrated attenuation of ROS production by DNP (40, 41).

A low concentration of pyruvate (2 μM) was used to reverse the mild mitochondrial depolarization by APC and DNP in cardiomyocytes. It was reported that the application of pyruvate or high concentrations of its metabolic precursor glucose to cells hyperpolarizes mitochondria (11). We observed a biphasic effect of pyruvate, a transient mitochondrial depolarization that is followed by sustained hyperpolarization, a so-called “pyruvate paradox” (11). That is, pyruvate initially depolarizes mitochondria due to pyruvate-proton symport into mitochondria (17, 23). As the pyruvate is being metabolized feeding electrons into the respiratory chain, a greater number of protons is extruded from mitochondria, increasing ΔΨ_m. In our cell survival experiments, application of pyruvate alone decreased cell death, which could be due to its effects independent of changes in ΔΨ_m. The ROS-scavenging properties of pyruvate have been reported, but at much higher concentrations (mM) than used in our study (2 μM) (33). Possibly, coapplication of pyruvate and H₂O₂ in cell survival experiments, even at this lower pyruvate concentration, scavenged H₂O₂ and increased cell survival. To verify that pyruvate antagonized the protective effects of APC by reversing ΔΨ_m, we also used oligomycin (25 ng/ml) to increase ΔΨ_m (data not shown). In this experiment, we observed, as we did for pyruvate, that oligomycin prevented APC-induced mitochondrial depolarization and abolished APC-induced attenuation of ROS production in stressed cells.

Attenuation of excess ROS production in cardiomyocytes and isolated mitochondria is caused by a decrease in ΔΨ_m since decrease in ΔΨ_m inhibits proton pumping by the respiratory chain and coupled electron transfer, which results in “electron leak” and ROS production (6). However, in isolated mitochondria, a similar mitochondrial depolarization was not sufficient to attenuate ΔΨ_m-driven Ca²⁺ uptake (Fig. 6, C and D), but it attenuated accumulation of cytosolic and mitochondrial Ca²⁺ in cardiomyocytes exposed to oxidative stress (Fig. 7). We showed that H₂O₂ increases cytosolic and mitochondrial Ca²⁺, which is in agreement with the well-established phenomenon of ROS-induced Ca²⁺ release (55), where oxidative stress/ROS can activate ryanodine receptor (12) and also inhibit Ca²⁺ pumps (26, 35). Therefore, it appears that attenuation of mitochondrial ROS by partial decrease in ΔΨ_m decreased ROS-induced Ca²⁺ release, which would explain attenuation of mitochondrial and cytosolic Ca²⁺ in stressed cells, but not in isolated mitochondria. This notion is supported by studies which show that Ca²⁺ overload occurs after ROS burst during reperfusion in cardiomyocytes (28). Interestingly, in experiments using isolated mitochondria, APC and DNP enhanced Ca²⁺ uptake, which may result from changes in the intramitochondrial H⁺ concentration that may also be affected by both treatments (31). Further detailed studies are needed to investigate the role of pH on mitochondrial Ca²⁺ handling.

The mechanisms of interplay between excess ROS production and accumulation of cytosolic and mitochondrial Ca²⁺ as mediators of cell injury during oxidative stress need further investigation. Nonetheless, there is the possibility of a self-sustained nature of mPTP opening caused either by bursts in mitochondrial ROS production or mitochondrial Ca²⁺ uptake (3, 22). It seems that excess ROS production, cytosolic, and mitochondrial Ca²⁺ accumulation, as mediators of I/R injury, are interdependent and part of a complex process where both ROS production and Ca²⁺ accumulation become amplified, leading to sustained mPTP opening and, eventually, cell death. Therefore, even a partial decrease in ΔΨ_m induced by APC or DNP may have dramatic effect on cell survival by attenuating excess ROS production and thereby interrupting the amplification process and the interactions between ROS production, Ca²⁺ accumulation, and mPTP opening induced during reperfusion.

A limitation of this study is that oxidative stress, such as H₂O₂, does not fully account for events that occur during I/R injury, but it has been used widely as an accepted method to mimic conditions during reperfusion injury. Another limitation of our study is that the experiments were performed at room temperature where reactions occur at a slower rate compared with living organisms. However, despite the difference in temperature, it is likely that similar (patho)physiological mechanisms would occur at 37°C as well. Experiments on isolated mitochondria in which APC was conducted in vivo should be interpreted with caution since potential stress that was inflicted during the isolation procedure may modify preconditioning-induced changes. Lastly, the study was conducted in isolated cardiomyocytes or mitochondria, which excludes other factors, for example blood flow, vascular elements, or paracrine and exocrine agents, that could modulate the anesthetic action during APC in vivo. Therefore, these findings should be also interpreted with caution with respect to the in vivo situation.

In conclusion, this study shows that cytoprotection by isoflurane-induced preconditioning could be in part mediated by a partial mitochondrial depolarization that primarily attenuates excess ROS production in cardiomyocytes exposed to oxidative stress. As a consequence, ROS-induced mPTP opening is delayed, which leads to an increase in cell survival. We also addressed the interplay between mild mitochondrial depolarization, ROS generation, and Ca²⁺ uptake and showed that attenuation of excess ROS production is the direct result of mild mitochondrial depolarization, while the attenuation of cytosolic and mitochondrial Ca²⁺ accumulation is likely a consequence of decreased ROS production.

GRANTS

This work was supported in part by National Institutes of Health Grants P01GM066730 and R01HL034708 (to Z. J. Bosnjak).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).