INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ISCHEMIC PRECONDITIONING (IPC) is a phenomenon in which brief episodes of conditioning ischemia can blunt subsequent lethal injury of the heart (23). This endogenous cardioprotective mechanism has been conceptually divided into triggers and mediators/effectors (3). Mitochondrial ATP-sensitive potassium (mitoK_ATP) channels have been proposed to be the viable end-effectors of IPC (8, 18, 29). A recent study by Pain et al. (27) further proposed that mitoK_ATP channels serve as triggers rather than end-effectors via the generation of reactive oxygen species (ROS). Thus the mitoK_ATP channel appears to play dual roles, both as a trigger and a mediator/effector of cardioprotection (11, 17, 24). Although the definition of IPC was initially used to describe a reduction in myocardial necrosis, it has been reported that IPC reduces ischemic injury by decreasing apoptosis (6, 9, 20, 28, 36). Recently, Akao et al. (1) demonstrated that diazoxide, a putative mitoK_ATP channel opener, inhibited apoptosis induced by exposure to H₂O₂ for 16 h. They used cultured neonatal rat ventricular myocytes and applied diazoxide simultaneously with H₂O₂, suggesting that the mitoK_ATP channel acts as a mediator/effector of cardioprotection against apoptosis. However, it remains unclear whether antiapoptotic effects can be triggered by opening of mitoK_ATP channels. The present study was designed to investigate the critical timing of mitoK_ATP channel opening required to confer antiapoptotic effects, and it demonstrates for the first time that opening of mitoK_ATP channels by diazoxide acts as a trigger of cardioprotection against apoptosis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Preparation of cultured neonatal rat cardiomyocytes. The experimental protocol was approved in advance by the Ethics Review Committee for Animal Experimentation of Oita Medical University. Neonatal cardiomyocytes were prepared from 3- to 5-day-old Wistar rats as described previously (35). Cardiomyocytes were then plated onto 30-mm culture dishes at a density of 5 × 10⁶ per dish and cultured in DMEM supplemented with 5% fetal bovine serum at 37°C under 5% CO₂. On day 4, cardiomyocytes beating synchronously were used for experiments.

Assessment of apoptosis and mitochondrial membrane potential by flow cytometry. Flow cytometric analysis was performed with an EPICS (Beckman Coulter Instruments) on a minimum of 1 × 10⁴ unfixed cells per sample. Cardiomyocytes were trypsinized with trypsin-EDTA, resuspended in PBS, and loaded with 10 µM propidium iodide (PI; Wako Pure Chemical Industries, Osaka, Japan) and 1 µM annexin V-FITC (Immunotech, Marseille, France) at 4°C for 10 min. In separate experiments, before being resuspended by trypsinization, cardiomyocytes were incubated with 10 µM rhodamine-123 (Rh-123; Molecular Probes, Eugene, OR) at 37°C for 10 min and then loaded with PI at room temperature for 5 min. Apoptosis was identified as cells with low forward scatter (FSC) on side scatter (SSC)/FSC dot plots, PI dim staining on FSC/PI dot plots, and annexin V-positive and PI-negative staining on annexin V/PI dot plots (5, 14, 19). The mean fluorescence intensity of Rh-123 on the histograms measured by flow cytometry was used to determine the loss of mitochondrial membrane potential (). Fluorescence probes were excited with an air-cooled 488-nm argon laser. The emission fluorescence was monitored at 525 nm for Rh-123 or annexin V-FITC and 620 nm for PI. Data were analyzed with the Coulter software package (Phoenix Flow).

Analysis of apoptotic nuclei by fluorescent microscopy. To detect the characteristic features of apoptotic nuclei, unfixed cardiomyocytes were stained with 0.12 µM Hoechst 33342 (Wako), a fluorescent DNA-binding dye, for 10 min. Fluorescence of Hoechst 33342 (excited at 365 nm and emitted at 400 nm) was captured with a charge-coupled device camera under a fluorescent microscope. Apoptotic cells were identified by their typical morphological appearance, with chromatin condensation and nuclear fragmentation. An average of >500 nuclei from random fields was analyzed for each data point.

Determination of lactate dehydrogenase in culture medium. Lactate dehydrogenase (LDH) released to the culture medium was determined with an LDH assay kit (Eiken Chemical, Tokyo, Japan). Approximately 5 × 10⁶ cardiomyocytes were placed into 35-mm culture dishes containing 1 ml of culture medium (DMEM supplemented with 5% fetal bovine serum). After 2-h exposure to H₂O₂, 500 µl of supernatant was carefully collected for LDH determination.

Experimental protocols. The experimental protocols are depicted in Fig. 1. Apoptosis was induced by exposing cardiomyocytes to 500 µM H₂O₂. The dose of H₂O₂ was chosen on the basis of previous reports that the induction of apoptosis in neonatal rat cardiomyocytes occurred via activation of the mitochondrial apoptotic pathway (4, 32). To investigate the role of the mitoK_ATP channel as a trigger of cardioprotection against H₂O₂-induced apoptosis, cardiomyocytes were pretreated with diazoxide for 30 min and then washed twice with PBS before incubation with H₂O₂ for 2 h. In the control group, cardiomyocytes obtained from sister cultures received vehicle only without exposure to H₂O₂. In the H₂O₂ group, cardiomyocytes were incubated with H₂O₂ alone. In the H₂O₂+DZ(P) group, cardiomyocytes were pretreated with 100 µM diazoxide (Sigma, St. Louis, MO) for 30 min before H₂O₂ incubation. In the H₂O₂+DZ(P)+5-HD(P) group, cardiomyocytes were pretreated with diazoxide and 500 µM 5-hydroxydecanoate (5-HD; Sigma) before H₂O₂ incubation. In the H₂O₂+DZ(P)+5-HD(C) group, cardiomyocytes were pretreated with diazoxide before H₂O₂ incubation and 5-HD was coadministered with H₂O₂. In the H₂O₂+DZ(P)+ HMR(P) group, cardiomyocytes were pretreated with diazoxide and 30 µM HMR-1098 (kindly provided by Aventis Pharma) before H₂O₂ incubation.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Experimental protocols. Analysis by flow cytometry and/or fluorescent microscopy was performed at the times indicated by arrowheads. In the names of study groups, P (pretreatment before exposure to H2O2) and C (cotreatment with H2O2) indicate the timing of each drug administration. See MATERIALS AND METHODS for details.

Statistical analysis. Data are expressed as means ± SE. Differences between groups were examined for statistical significance by ANOVA with Fisher's post hoc test. A P value <0.05 denoted the presence of a statistically significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

H₂O₂ induces apoptosis of cardiomyocytes. Figure 2A shows the representative flow cytometric analysis of cardiomyocytes stained with PI and annexin V-FITC. Side and forward scatter (SSC/FSC) dot plots identified two distinct cell subpopulations, and cells with high FSC signals (R1) were predominant in controls. Exposure to H₂O₂ for 2 h shifted the cells into the lower portion of the diagram (R2). When cell subpopulations were detected on FSC/PI dot plots, apoptotic cells in the dim PI fluorescence region (R4) increased after exposure to H₂O₂. Moreover, annexin V-positive and PI-negative apoptotic cells (R9) were prominent after exposure to H₂O₂. Figure 2A further shows the nuclear morphology of cardiomyocytes stained with Hoechst 33342. The apoptotic features of chromatin condensation and nuclear fragmentation were observed after 2-h exposure to H₂O₂. As summarized in Fig. 2B, the percentage of apoptotic cells in the low-FSC subpopulation (R2) was significantly increased from 18 ± 1% in the control sister culture to 45 ± 2% after exposure to H₂O₂ (n = 5; P < 0.01). Similarly, H₂O₂ significantly increased the degree of apoptosis assessed by PI staining (36 ± 2% vs. 16 ± 2%, n = 5; P < 0.01), annexin V-PI double staining (40 ± 2% vs. 19 ± 1%, n = 5; P < 0.01), and Hoechst 33342 staining (35 ± 4% vs. 7 ± 1%, n = 11; P < 0.01). These results indicate that H₂O₂ induced apoptosis of rat neonatal cardiomyocytes and that the degree of apoptosis determined by four different methods was comparable.

View larger version (52K): [in this window] [in a new window]   Fig. 2.   A: representative flow cytometric and morphological analyses of apoptotic cells. Cardiomyocytes were exposed to H2O2 for 2 h. Apoptotic cells were assessed by side scatter (SSC)/forward scatter (FSC) light scatter and propidium iodide (PI) staining and annexin V-PI double staining as a fraction of R2 (low FSC), R4 (dim staining of PI), and R9 (annexin V positive, PI negative) regions. Nuclear morphology of cardiomyocytes stained with Hoechst 33342 shows apoptotic features of chromatin condensation and nuclear fragmentation. B: summarized data for H2O2-induced apoptotic cell death. Data represent means ± SE of 5-11 experiments. *P < 0.01 vs. control sister culture.

Time course of H₂O₂-induced apoptosis and loss. Figure 3 shows the time courses of apoptotic cell death and loss during exposure to H₂O₂. Here, apoptotic cells were identified as the low-FSC subpopulation and loss was assessed by relative change in the mean fluorescence intensity of Rh-123. The percentage of apoptotic cell death increased progressively and peaked at 2 h (285 ± 78% of control sister culture, n = 8; P < 0.01). The time course of decrease in Rh-123 fluorescence paralleled that of apoptotic cell death, and after 2-h exposure to H₂O₂ decreased to 53 ± 10% (n = 8; P < 0.01) of the control sister culture. These results suggest that H₂O₂-induced apoptosis was associated with loss of .

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Time courses of change in H2O2-induced apoptotic cell death and mitochondrial membrane potential () loss in protocol. Apoptotic cell death was measured by SSC/FSC dot plot (top), and loss of was measured by rhodamine-123 (Rh-123) fluorescence intensity (bottom). Each data point is the mean ± SE of 4-12 experiments. *P < 0.01 vs. control sister culture.

Diazoxide prevents H₂O₂-induced apoptosis and loss. We then examined whether pretreatment with diazoxide attenuated H₂O₂-induced apoptosis and loss. As shown in Fig. 4, H₂O₂-induced apoptotic cells in the low-FSC subpopulation accompanied the dissipation of , which was evident by the leftward shift in Rh-123 fluorescence. The mean intensity of Rh-123 fluorescence measured on the histogram was 40.0 in the sister control culture and 12.4 after H₂O₂ incubation for 2 h. Pretreatment with diazoxide for 30 min attenuated both apoptotic cell death and loss induced by exposure to H₂O₂, and the mean intensity of Rh-123 fluorescence was restored to 24.7. Summarized data shown in Fig. 5 confirm that the results in Fig. 4 are indeed representative. H₂O₂ significantly increased apoptotic cell death to 338 ± 41% of the control sister culture (n = 15; P < 0.01) and decreased the assessed by mean intensity of Rh-123 fluorescence to 56 ± 8% of control (n = 15; P < 0.01). Diazoxide pretreatment [H₂O₂+DZ(P) group] significantly attenuated apoptotic cell death to 228 ± 19% of control (n = 15; P < 0.01 vs. H₂O₂ group) and restored the loss of to 78 ± 8% of control (n = 15; P < 0.01 vs. H₂O₂ group). The selective mitoK_ATP channel blocker 5-HD (30) applied together with diazoxide [H₂O₂+DZ(P)+5-HD(P) group] prevented the effects of diazoxide, and the extent of apoptotic cell death and loss of was similar to that induced by H₂O₂ alone. In contrast, when 5-HD was coadministered with H₂O₂ after the application of diazoxide [H₂O₂+DZ(P)+5-HD(C) group], the drug failed to inhibit the effect of diazoxide. These effects of 5-HD suggest that opening of the mitoK_ATP channel by diazoxide indeed acts as a trigger of cardioprotection against H₂O₂-induced apoptosis. Moreover, HMR-1098, a selective sarcolemmal K_ATP channel blocker (31), applied together with diazoxide before H₂O₂ [H₂O₂+DZ(P)+HMR(P) group], did not affect the protective effects of diazoxide, suggesting that the sarcolemmal K_ATP channel is not involved in the antiapoptotic effect of diazoxide.

View larger version (57K): [in this window] [in a new window]   Fig. 4.   Representative SSC/FSC dot plots (left panels), Rh-123/PI dot plots (center panels), and histograms of Rh-123 fluorescence (right panels) obtained from sister control (top panels), cardiomyocytes exposed to H2O2 for 2 h (middle panels), and cardiomyocytes pretreated with diazoxide before exposure to H2O2 (bottom panels). The vertical dashed line in each histogram indicates the position of the mean intensity of Rh-123 fluorescence.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Summarized effects of diazoxide, 5-HD, and HMR on apoptotic cell death and loss of induced by exposure to H2O2 for 2 h in protocol. Data represent means ± SE of 11-15 experiments and are expressed as a % of control sister culture. *P < 0.01 vs. sister control; #P < 0.01 vs. H2O2 group.

Diazoxide prevents H₂O₂-induced necrosis. H₂O₂ induced not only apoptosis but also necrosis. The level of LDH in the culture medium was significantly increased from 0.059 ± 0.013 IU/ml (n = 6) in the control sister culture to 0.278 ± 0.015 IU/ml after 2-h exposure to H₂O₂ (n = 6; P < 0.001). Pretreatment with diazoxide for 30 min significantly reduced the LDH level to 0.222 ± 0.014 IU/ml (n = 6; P < 0.05 vs. H₂O₂ group). When necrotic cell death was assessed as a percentage of the bright PI fluorescence region (R3), H₂O₂ increased necrosis from 4.7 ± 0.5% (n = 6) in control to 9.3 ± 1.3% (n = 6; P < 0.01) after H₂O₂ exposure. Pretreatment with diazoxide again attenuated necrotic cell death to 5.6 ± 0.3% (n = 6; P < 0.01 vs. H₂O₂ group).

Transient depolarization of triggers cardioprotection. We investigated the effect of diazoxide on during the triggering period (Fig. 6). Diazoxide significantly depolarized the and decreased the intensity of Rh-123 fluorescence to 74 ± 5% of the control sister culture (n = 5; P < 0.01) after a 15-min application. This depolarization of restored to 87 ± 4% of control (n = 5; P = not significant) after 30 min. Transient depolarization of observed after a 15-min application of diazoxide was abolished by 5-HD (88 ± 6%, n = 5). Thus the measured just before application of H₂O₂ was comparable among the groups. Nevertheless, diazoxide prevented the loss of induced by subsequent exposure to H₂O₂ for 2 h. These results suggest that the transient depolarization of during the triggering period might contribute to the mechanism of cardioprotection.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Time course of change in during application of diazoxide and/or 5-HD and 2 h after exposure to H2O2. Diazoxide and/or 5-HD was applied for 30 min before H2O2 exposure. Data represent means ± SE of 5 experiments and are expressed as % of control sister culture. *P < 0.05 vs. sister control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major findings of the present study were that 1) pretreatment of rat cardiomyocytes with diazoxide, a putative mitoK_ATP channel opener, elicited the transient depolarization and attenuated the subsequent apoptotic cell death and loss induced by exposure to H₂O₂; and 2) these effects of diazoxide were antagonized by the mitoK_ATP channel blocker 5-HD but not by the sarcolemmal K_ATP channel blocker HMR-1098. Our data therefore suggest that mitoK_ATP channels serve as a trigger of cardioprotection against apoptosis induced by oxidative stress.

H₂O₂ has been reported to induce apoptosis of cardiomyocytes through activation of the mitochondrial apoptotic pathway (4, 32), where loss of is a key step associated with cytochrome c release from the mitochondria (10). Our results confirmed that this was indeed the case, and H₂O₂ induced the apoptosis of cultured neonatal rat cardiomyocytes, determined by flow cytometric analysis (light scatter, PI, and annexin V-FITC staining) and by nuclear morphology (Hoechst 33342 staining). Furthermore, consistent with a previous report (4), H₂O₂-induced apoptosis was associated with the loss of as shown by Rh-123 fluorescence. Besides apoptosis, we also found that H₂O₂ increased necrotic cells, which is in agreement with a previous study in neonatal mouse cardiomyocytes (33). LDH release increased by approximately fivefold, whereas PI-positive cells increased by approximately twofold after exposure to H₂O₂. A probable reason for this is that necrotic cells become nonadherent and hence may be removed when the cell layer is washed and resuspended in PBS for flow cytometric analysis. The degree of necrotic cells assessed as a percentage of the bright PI fluorescence region (~9%) was substantially smaller than that of apoptotic cells (~40%). Accordingly, cultured neonatal rat cardiomyocytes predominantly showed apoptotic cell death after exposure to H₂O₂ under our experimental conditions.

Although the mitoK_ATP channel was initially proposed to be the end-effector of IPC (8, 18, 29), the triggering action of mitoK_ATP channels has also been proposed using infarct size as the end point. Pain et al. (27) showed that 5-HD administered early to bracket preconditioning ischemia could abolish the infarct size-limiting effect of IPC. Baines et al. (2) demonstrated that diazoxide administered before ischemia but not after the onset of index ischemia reduced infarct size. The results presented here confirm that pretreatment with diazoxide acts as a trigger, thereby attenuating the necrotic cell death induced by H₂O₂. Recently, in addition to reduction of infarct size, IPC has been reported to reduce apoptosis (6, 9, 20, 28, 36). Akao et al. (1) reported that diazoxide attenuated both apoptotic cell death and loss induced by exposure to H₂O₂ for 16 h, and these effects of diazoxide were antagonized by 5-HD. Because diazoxide and/or 5-HD were applied together with H₂O₂ in their study, the results support the idea that mitoK_ATP channel acts as a mediator/effector of cardioprotection against apoptosis. A novel and interesting finding in the present study is that apoptotic cell death and loss after 2-h exposure to H₂O₂ were significantly attenuated when diazoxide was applied for 30 min before exposure to H₂O₂. We confirmed that this indeed resulted from opening of mitoK_ATP channels: 5-HD, but not HMR-1098, completely abolished the antiapoptotic effects of diazoxide (Fig. 5). Moreover, once cardiomyocytes were pretreated with diazoxide, subsequent application of 5-HD together with H₂O₂ could not block the protection afforded by diazoxide. Together, these results indicate that the mitoK_ATP channel acts as a trigger of cardioprotection against apoptosis. We further found that the triggering action of diazoxide to reduce apoptotic cell death could not be observed after 16-h exposure to H₂O₂ (data not shown). Thus the mitoK_ATP channel apparently triggers an early phase of protection that lasts for ~2 h.

The precise mechanism by which diazoxide prevents apoptosis remains unclear. It has been reported that diazoxide-induced opening of mitoK_ATP channels causes mild depolarization of , thereby attenuating mitochondrial Ca²⁺ overload induced by ouabain and metabolic inhibition (13, 22). Mitochondrial Ca²⁺ overload results in the opening of permeability transition pore (PTP), which in turn causes the loss of and release of cytochrome c (10). Therefore, prevention of PTP by attenuating mitochondrial Ca²⁺ overload may be a critical mechanism of cardioprotection. Indeed, Akao et al. (1) demonstrated that diazoxide inhibited the cytochrome c release and loss of induced by H₂O₂. Recently, Minners et al. (21) reported that pharmacological preconditioning by diazoxide uncoupled mitochondria and decreased in Girardi cells and C₂C₁₂ myotubes. In the present study, we also found that diazoxide depolarized the during triggering period (Fig. 6). However, it should be noted that depolarization of induced by diazoxide was transient and that there was no significant depolarization of just before application of H₂O₂. This finding implies that prevention of mitochondrial Ca²⁺ overload in association with cannot account for the triggering mechanism of cardioprotection. Alternatively, ROS generation is thought to be a trigger of signaling pathways mediating IPC. Pain et al. (27) demonstrated that the triggering effect of diazoxide was lost when a scavenger of ROS was coadministered with diazoxide. Forbes et al. (7) provided direct evidence that diazoxide increases ROS production. Therefore, it has been hypothesized that opening of mitoK_ATP channels by diazoxide may lead to ROS generation. ROS may then activate downstream PKC. Okamura et al. (25) reported that the protective effect of IPC against apoptosis was blocked by a PKC inhibitor. Liu et al. (16) further demonstrated that PKC- is involved in inhibition of apoptosis by IPC. Moreover, Wang and Ashraf (34) demonstrated diazoxide-induced PKC translocation in Langendorff-perfused rat hearts and showed that these effects could be blocked by PKC inhibitors. Thus such a PKC-dependent mechanism mediated by ROS generation might contribute to the antiapoptotic effect of diazoxide. In a preliminary experiment, however, we observed that a ROS scavenger only partially inhibited the transient depolarization of during application of diazoxide and could not completely abolish the antiapoptotic effect of diazoxide. Regarding the infarct size-limiting effects, ROS seems to be involved in the triggering action of mitoK_ATP channels. However, it remains unclear whether the ROS-dependent mechanism may be involved in the antiapoptotic effect of mitoK_ATP channels.

In conclusion, the results of our study provide novel evidence that opening of mitoK_ATP channels by diazoxide acts as a trigger and reduces apoptotic cell death. Although we used H₂O₂ to induce apoptosis, it will be interesting to see whether mitoK_ATP channels trigger antiapoptotic effects in an in vivo setting. More recently, it has been proposed that diazoxide inhibits respiratory chain, which may serve as a predictive mechanism for ROS generation (12). However, this idea is contrary to a study demonstrating that diazoxide can suppress ROS generation and reduce cytochrome c release at reoxygenation in a potassium-independent manner (26). Furthermore, although diazoxide has been shown to depolarize , a modest depolarization of would be expected to decrease ROS generation rather than increase it (15). Obviously, how diazoxide might be linked to ROS generation remains unclear. Further studies are required to define the mechanism by which diazoxide sets the heart into a preconditioned state against apoptosis.