We show here that Ca2+ and reactive oxygen species (ROS) are involved in the up-regulation of c-Jun NH2-terminal kinase 1 (JNK1) activity during apoptosis induced by ginsenoside Rh2 (G-Rh2) in HeLa, MCF10A-ras, and MCF7 cells. Addition of antioxidants such as N-acetyl-l-cysteine or catalase attenuates G-Rh2-induced ROS generation, JNK1 activation, and apoptosis. The overexpression of catalase down-regulates caspase-3 and JNK1 activities. G-Rh2 treatment of cells results in mitochondrial depolarization, second mitochondrial activator of caspase release, and translocation of Bax into the mitochondria, and these events are inhibited by antioxidants. Ca2+ is also involved in mitochondrial depolarization during G-Rh2-induced apoptosis. These results suggest that ROS and Ca2+ are important signaling intermediates leading to stress-activated protein kinase/extracellular signal-regulated kinase kinase 1/JNK1 activation and depolarization of the mitochondrial membrane potential in G-Rh2-induced apoptosis.
Apoptosis is a highly regulated and organized death process that includes cellular morphological changes, oligonucleosomal DNA cleavage, and activation of caspases. Previous studies suggest that reactive oxygen species (ROS) play an important role in apoptosis (Carmody and Cotter, 2001; Djavaheri-Mergny et al., 2004), aging, and degenerative disease (Wallace, 2005). In addition, ROS, including superoxide anions (), hydrogen peroxide (H2O2), and hydroxyl radicals (·OH), act as signaling intermediates (Adler et al., 1999; Thannickal and Fanburg, 2000), including mediation of alterations in normal Ca2+ homeostasis. ROS also are link signaling pathways leading to cell death through the mitogen-activated protein kinase pathway (Yu et al., 2004), the nuclear factor κB pathway (Sakon et al., 2003; Pham et al., 2004), and apoptosis signal-regulating kinase 1 (Kadowaki et al., 2005). Under physiological conditions, generated ROS are eliminated by cellular antioxidant enzymes. For example, superoxide dismutases (SOD) remove by catalyzing a dismutation reaction, involving oxidation of to oxygen and reduction of another to H2O2 (Farr et al., 1986). H2O2 is subsequently eliminated by catalase, glutathione peroxidases, and peroxiredoxins (Thannickal and Fanburg, 2000). Catalase promotes the conversion of H2O2, a potentially harmful oxidizing agent, to water and molecular oxygen. When ROS generation exceeds the antioxidative capacity of cells, oxidative stress can occur and cause cell death (Benhar et al., 2001; Martindale and Holbrook, 2002). ROS are produced by various reagents, such as tumor necrosis factor α (Djavaheri-Mergny et al., 2004), paeoniflorin (Tsuboi et al., 2004), bortezomib (Yu et al., 2004), rhein (Lin et al., 2003), and so on.
Ginsenoside Rh2 (G-Rh2), a ginseng saponin isolated from the root of Panax ginseng C.A. Meyer, has been shown to induce apoptosis via ROS generation (Kim et al., 1999) and activation of several kinases, such as c-Jun NH2-terminal kinase (JNK), cyclin-dependent kinase 2, and protein kinase C delta, during apoptosis (Jin et al., 2000; Ham et al., 2003; Oh et al., 2005). However, the mechanisms by which G-Rh2 up-regulates these kinase activities in cells are unknown. Previously, we reported that JNK1 activity is up-regulated as a consequence of proteolytic cleavage of p21WAF1/CIP1 and that this activation is a functionally relevant event in apoptotic cells induced by G-Rh2 (Ham et al., 2003). The results of our previous studies show that the dominant-negative version of JNK1 or JNK inhibitor (JBD) suppresses the G-Rh2-induced apoptosis, suggesting that the G-Rh2-induced activation of JNK1 is involved in the induction of apoptosis (Ham et al., 2003).
JNK has been suggested to play an important role in regulating the Bcl-2 protein family that includes the antiapoptotic proteins Bcl-2 (Deng et al., 2001) and Bcl-XL (Fan et al., 2000) and the proapoptotic proteins Bax (Tsuruta et al., 2004), Bak (Lee et al., 2005), Bid (Deng et al., 2003), and Bim (Putcha et al., 2003). Some Bcl-2 proteins are translocated to mitochondria by the action of JNK and consequently are involved in apoptotic cell death by releasing cytochrome c and second mitochondrial activator of caspases (Smac)/DIA-BLO to the cytosol (Van Loo et al., 2002; Deng et al., 2003; Tsuruta et al., 2004). The release of apoptogenic factors is accompanied by mitochondrial dysfunction, including mitochondrial membrane potential changes or the induction of the mitochondrial permeability transition. The intrinsic pathway can be initiated by mitochondrial membrane depolarization that facilitates the release of cytochrome c, which then associates with apoptotic protease-activating factor 1 and caspase-9 to promote caspase activation. Although the mechanism underlying the release of cytochrome c from the mitochondrial membrane is not known, it is evident that the Bcl-2 family of proteins is intimately involved in the regulation of this process. In the present study, we examine whether G-Rh2-induced JNK1 activation is functionally linked to oxidative stress or Smac release via mitochondrial dysfunction during apoptosis. We show that Ca2+ and ROS participate in mitochondrial membrane depolarization, stress-activated protein kinase/extracellular signal-regulated kinase kinase 1 (SEK1)/JNK1 activation, and Bax translocalization during G-Rh2-induced apoptosis. Thus, Ca2+ and ROS generation play an important role in G-Rh2-induced apoptosis that involves the activation of the JNK pathway.
Materials and Methods
Reagents. G-Rh2 with purity 98% was dissolved in 70% ethanol at a concentration of 10 mg/ml and stored at –20°C. N-Acetyl-l-cysteine (NAC) and catalase were purchased from Sigma-Aldrich (St. Louis, MO). 2′,7′-Dichlorodihydrofluorescein diacetate (DCFH-DA), JC-1, and the MitoCapture kit were purchased from Calbiochem (San Diego, CA). Glutathione S-transferase (GST)-c-Jun and protein A agarose beads were purchased from Upstate Biotechnology (Lake Placid, NY). The caspase-3 substrate, Ac-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC), was purchased from BD PharMingen (San Diego, CA). 1,2-Bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid acetoxymethyl ester (BAPTA-AM) was purchased from Invitrogen Molecular Probes (Carlsbad, CA).
Cell Culture and Transfections. HeLa cells or MCF7 cells were maintained in Dulbecco's modified Eagle's medium or RPMI 1640 medium supplemented with 10% fetal bovine serum and antibiotics/antimycotics (Gibco-BRL, Gaithersburg, MD) at 37°C in 5% CO2 atmosphere. The MCF10A-ras cells were cultured in Dulbecco's modified Eagle's medium/Ham's F-12 medium supplemented with 5% heat-inactivated horse serum, 10 μg/ml insulin, 100 ng/ml cholera toxin, 0.5 μg/ml hydrocortisone, 20 ng/ml recombinant epidermal growth factor, 2 mM l-glutamine, and 100 μg/ml penicillin/streptomycin/fungi zone mixture at 37°C in 5% CO2 atmosphere. Transient transfections were performed in six-well plates using Polyfect (Qiagen, Valencia, CA). In general, 2 × 105 cells/well were seeded and transfected using 4 μg of total DNA. Twenty-four to 40 h after transfection, cells were treated with G-Rh2 and analyzed for cell viability and by Western blot.
Subcellular Fractionation. HeLa cells were washed with phosphate-buffered saline (PBS), and the pellet was suspended in 3 volumes of ice-cold buffer A (20 mM HEPES, pH 7.5, 1.5 mM MgCl2, 10 mM KCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, and protease inhibitor mixture) containing 250 nM sucrose. The cells were homogenized using a Dounce homogenizer, and cytosolic and mitochondrial extracts were isolated as described previously (Chauhan et al., 2003).
Western Blot Analysis. Western blotting was performed according to standard methods. Cell pellets were lysed in lysis buffer (0.5% Triton X-100, 20 mM Tris-HCl, pH 7.5, 2 mM MgCl2, 1 mM DTT, 1 mM EGTA, 50 mM β-glycerophosphate, 25 mM NaF, 1 mM Na3VO4, 100 μg/ml phenylmethylsulfonyl fluoride, and protease inhibitor mixture) for 1 h. Lysates were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA). The membrane was blocked with 5% nonfat dried milk (Carnation; Nestle, Vevey, Switzerland) in PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4-7H2O, 1.4 mM KH2PO4, pH 7.4) containing 0.05% Tween 20 and incubated in 1:1000 dilutions of mouse monoclonal anti-phospho-c-Jun, mouse monoclonal anti-phospho-JNK1, rabbit polyclonal anti-JNK1, rabbit polyclonal anti-poly(ADP-ribose) polymerase (PARP), mouse monoclonal anti-hemagglutinin (HA), mouse monoclonal anti-α-tubulin, rabbit polyclonal anti-Bax, goat polyclonal antivoltage-dependent anion channel (VDAC), rabbit polyclonal anti-actin (Santa Cruz Biochemicals, Santa Cruz, CA), mouse monoclonal anti-Smac, rabbit polyclonal anti-cleaved PARP (Cell Signaling Technology, Danvers, MA), or mouse monoclonal anti-FLAG (Sigma-Aldrich). Bands were visualized with horseradish peroxidase-conjugated antibodies and the enhanced chemiluminescence system (iNtRON Biotechnology, Kyungki-Do, Korea).
Immune Complex Kinase Assay. Cells were lysed with lysis buffer. The lysates were immunoprecipitated by shaking with rabbit polyclonal anti-JNK1 at 4°C. After an overnight incubation, immune complexes were recovered with protein A-agarose beads and washed three times with lysis buffer and three times with 20 mM HEPES (pH 7.4). Pellets were mixed with 1 μg of GST-c-Jun, 10 μCi of [γ-32P]ATP, 500 μM unlabeled ATP, and kinase buffer containing 20 mM DTT, 100 mM MgCl2, and 2 mM Na3VO4. The kinase reaction was performed at 30°C for 30 min, and the supernatants were resolved by 12% SDS-PAGE. After staining, the gel was dried, and the phosphorylation of the c-Jun substrate was determined by autoradiography.
Flow Cytometry Analysis. Apoptosis in cell populations was determined by flow cytometric analysis. After treatment with G-Rh2, with or without catalase, both floating and trypsinized adherent cells were pelleted and washed with PBS. Cells were fixed with ice-cold 70% ethanol and treated with 1 mg/ml RNase for 30 min at 37°C. Propidium iodide (PI) then was added to the solution to a final concentration of 50 μg/ml, and the DNA content was quantified by flow cytometry (FACSCalibur; Becton Dickinson Biosciences, San Jose, CA).
For the detection of ROS generation, cells were washed with PBS and incubated with 20 μM DCFH-DA for 60 min at 37°C in the dark. After washing the cells with PBS, the fluorescence intensity of the cell suspension was read using a FACSCalibur.
For measuring the mitochondrial membrane potential, cells were incubated for 30 min with 2 μM JC-1 in culture medium. The adherent cell layer was washed with PBS and dislodged with trypsin. Cells were collected by centrifugation and resuspended in 0.5 ml of PBS for analysis using a FACSCalibur. Cytometer settings were optimized for green (FL-1) fluorescence, and the data were analyzed with Cell Quest Software (BD Biosciences, San Jose, CA).
Confocal Microscopy Analysis. Cells treated with or without G-Rh2 were incubated with 1 μg/ml MitoCapture cation dye (Calbiochem) or DCFH-DA dye at 37°C in a 5% CO2 incubator for 30 min or 1 h, respectively. Cells stained with MitoCapture dye were photographed using a fluorescence microscope (Olympus, Center Valley, PA) with a fluorescein isothiocyanate channel for green monomers (Ex/Em = 488/530 ± nm) and a PI channel for red aggregates (Ex/Em = 488/590 ± nm). Cells stained with DCFH-DA were photographed using a fluorescence microscope (Olympus) with an excitation wavelength of 488 nm and an emitter wavelength of 525 nm.
Caspase-3 Activity Assay. The substrate Ac-DEVD-AMC for the caspase-3 activity assay was used in a procedure modified from the manufacturer's instructions. Cell lysates (20 μg) were added to the reaction buffer (20 mM HEPES, pH 7.5, 10% glycerol, 2 mM DTT) containing 25 μM Ac-DEVD-AMC in 96-well plates. Lysates were incubated at 37°C for 1 h. Fluorescence of the cleavage product was measured using a SpectraFluor F129003 (TECAN US, Durham, NC) at excitation and emission wavelengths of 405 and 465 nm, respectively.
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Dye Assay. Exponentially growing cells were seeded in triplicate in a 96-well plate at 3.75 × 103 cells/ml. After treatment with G-Rh2, with or without NAC, 30 μl of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium solution (5 mg/ml) was added to each well and incubated for 3 h. The formazan grains formed by viable cells were solubilized with dimethyl sulfoxide, and the color intensity was measured at 570 nm with an enzyme-linked immunosorbent assay reader.
DNA Fragmentation Assay. After harvesting the cells, the cell pellet was dissolved in DNA extraction buffer (50 mM Tris-HCl, pH 8.0, 10 mM EDTA, 0.5% SDS) containing 0.5 mg/ml proteinase K and incubated at 50°C for 6 h. The lower layer of the phenol/chloroform solution was added to the supernatant containing DNA. After centrifugation, chloroform/isoamyl alcohol (24:1) solution was added to the upper layer of supernatant. The DNA was precipitated by NaCl and ethanol, electrophoresed thorough a 1.8% agarose gel, and photographed under visualization with UV light.
G-Rh2-Induced Apoptosis Involves ROS. We have previously shown that G-Rh2 (Fig. 1) at the concentration of 7.5 μg/ml induced apoptosis in SK-HEP-1 hepatocellular carcinoma cells (Ham et al., 2003). This study showed that G-Rh2 also induced apoptosis within 2 h in HeLa cervical carcinoma cells, as determined by counting the cells with membrane blebbing (Fig. 2A, top), red positive-PI staining (Fig. 2A, bottom), and DNA fragmentation (Fig. 2B). To investigate the mechanism by which G-Rh2 induced apoptosis in HeLa cells, we first examined whether G-Rh2 might be able to induce generation of ROS in HeLa cells during apoptosis. Previous reports suggested that ROS, which are known to be an important component of cytotoxic action, were involved in apoptotic cell death (Kadowaki et al., 2005; Rahmani et al., 2005) and JNK activation (Ventura et al., 2004). In the cells that were labeled with the cell-permeable fluorescent dye, DCFH-DA, the fluorescent intensity of DCF, an oxidized product of DCFH-DA by ROS, gradually increased, peaking at 6 h after treatment of the cells with G-Rh2 and sharply decreasing thereafter (Fig. 2C). It is likely that decreasing of ROS at 8 h is caused by dying cells. We next examined whether G-Rh2-induced ROS generation could be prevented by treatment with the antioxidant NAC. Using confocal microscopy, the production of ROS, with or without NAC, was monitored by measuring the fluorescent intensity of DCF. The results showed that ROS generation was markedly induced in G-Rh2-treated HeLa cells (Fig. 2D, top right), whereas the G-Rh2-induced generation of ROS was significantly prevented in the cells treated with NAC and G-Rh2 (Fig. 2D, bottom right).
Next we assessed whether ROS generation was required for G-Rh2-induced cell death. The increased caspase-3 activity in the G-Rh2-treated cells was dose-dependently down-regulated by pretreatment with NAC (Fig. 3A). DNA ladders were observed in G-Rh2-only–treated cells, but this effect was blocked by the addition of NAC (data not shown). These results indicated that the generation of ROS was functionally associated with G-Rh2-induced apoptosis.
Antioxidant Enzymes Inhibit G-Rh2-Induced ROS Generation and Cell Death. To assess the types of radicals involved in G-Rh2-induced apoptosis, two different antioxidant enzymes, SOD and catalase, were added to G-Rh2-treated cells. The production of ROS was decreased by 30 and 70% by treatment with equal units of SOD and catalase, respectively (data not shown). Thus, catalase was used as the ROS scavenger in subsequent experiments. To determine whether H2O2 generation was required for G-Rh2-induced apoptosis, G-Rh2-treated cells were pretreated with catalase, and their caspase-3 activity was determined (Fig. 3B). Treatment of HeLa cells with increasing units of catalase gradually and significantly down-regulated caspase-3 activity by approximately 50% in G-Rh2-induced cells (Fig. 3B). Flow cytometry analysis showed that the percentages of sub-G1 fractions representing apoptotic cells increased from 2% to 35.13% by treatment with G-Rh2 and that this increase was attenuated by 25% in cells that were cotreated with catalase and G-Rh2 (Fig. 3C). Flow cytometry analysis of cells stained with DCFH-DA also showed that catalase inhibited the generation of ROS (Fig. 3D). The fluorescence peak of DCF in G-Rh2-treated cells was shifted to the right (Fig. 3D, solid line) compared with the control (Fig. 3D, shaded area), whereas the fluorescence peak in cells that were treated with both catalase and G-Rh2 was shifted to the left (Fig. 3D, dashed line) compared with that of cells treated with G-Rh2 alone (Fig. 3D, solid line), indicating that catalase inhibited the production of H2O2.
G-Rh2 Induces Disruption of the Mitochondrial Membrane Potential. It has been suggested ROS are involved in mitochondrial dysfunctions such as induction of the permeability transition (Sullivan et al., 2005) and uncoupling of oxidation and phosphorylation (Brand, 2000; Brookes, 2005). To test whether the G-Rh2-induced generation of ROS was associated with mitochondrial dysfunction, we assessed Δψm, the disruption of mitochondrial membrane potential in the cells, after staining with MitoCapture or JC-1, a mitochondrial-specific dye. As shown in Fig. 4A, the intensity of red fluorescence decreased significantly within 2 h after G-Rh2 treatment and diminished markedly thereafter (Fig. 4A, top). In contrast, the intensity of the green fluorescence increased after treatment with G-Rh2 (Fig. 4A, bottom), and this increase paralleled ROS generation for 6 h (Fig. 2C). Reduction in mitochondrial membrane potential was confirmed using flow cytometry. Treatment with G-Rh2 for 6 h induced the reduction of the mitochondrial membrane potential: the mean value (1041.29) was shifted to the right (Fig. 4B, middle) compared with that of the control mean value (655.20) (Fig. 4B, top). The reduction of the mitochondrial membrane potential by treatment with G-Rh2 was blocked significantly by pretreatment of the cells with catalase (Fig. 4B, bottom): the mean value (717.98) was shifted to the left compared with that of G-Rh2-only–treated cells (Fig. 4B, middle). These results indicated that ROS generation was functionally associated with mitochondrial depolarization.
G-Rh2-Induced Mitochondrial Dysfunction Involves Ca2+in HeLa Cells. Mitochondria, an important source of ROS production, accumulate Ca2+ that can cause the mitochondria to generate ROS (Duchen, 2000). To examine whether Ca2+ might be involved in the G-Rh2-induced ROS signaling pathway, an intracellular calcium chelator, BAPTA-AM, was used. Pretreatment of cells with increasing doses of the calcium chelator gradually reduced caspase-3 activity by up to 50% compared with cells treated with G-Rh2 alone (Fig. 4C), indicating that Ca2+ was involved in G-Rh2-induced apoptosis.
The reduction of the mitochondrial membrane potential in cells treated with G-Rh2 was attenuated significantly by pretreatment with 5 μM BAPTA-AM (Fig. 4D, dashed line), shifting the mean value (489.00) to the left compared with the mean value (527.10) (Fig. 4D, solid line) of cells treated with G-Rh2 alone. This result indicated that G-Rh2 treatment induced Ca2+-dependent mitochondrial depolarization of the cells.
G-Rh2-Induced ROS Mediates JNK1 and SEK1 Activities That Are Required for Apoptosis. Previously, we showed that JNK1 activity was involved in G-Rh2-induced apoptotic cell death of SK-HEP-1 cells (Ham et al., 2003). Therefore, we tested whether JNK1 activity also was involved in apoptosis of G-Rh2-induced HeLa cells. The intracellular levels of the phosphorylated forms of JNK1 (46 kDa) and c-Jun, which is phosphorylated by active JNK1, and phospho-SEK1, the active form of the upstream kinase of JNK1, all increased similarly within 1 h of G-Rh2 treatment (Fig. 5A). Thus, JNK1 and SEK1 activities were up-regulated during G-Rh2-induced apoptotic cell death, and the activation of these kinases preceded the PARP cleavage that occurred within 2 h (Fig. 5A). The JNK2 activity (55 kDa), as determined by immunoblotting the cellular levels of phospho-JNK2 using specific JNK2 antibody, was also detected to increase in G-Rh2-induced HeLa cells, although this activity was much weaker than that of JNK1 (data not shown). Thus, we did not rule out a possibility that JNK2 activity was also involved during G-Rh2-induced apoptosis. Next we tested whether ROS might be involved in G-Rh2-induced JNK1 activation in HeLa cells. The results indicated that the phosphorylation of c-Jun, representing JNK1 activity, increased markedly in G-Rh2-treated cells compared with the control cells, whereas the G-Rh2-induced JNK1 activation was suppressed significantly by treatment with NAC. This indicated that ROS were involved in JNK1 activation induced by treatment with G-Rh2 (Fig. 5B). We also examined whether H2O2 was involved in the JNK1 activation because H2O2 was the primary ROS involved in G-Rh2-induced apoptosis (Fig. 3). The JNK1 activity gradually decreased with increasing units of catalase compared with cells treated with G-Rh2 alone (Fig. 5C, 1st panel). Phosphorylation of c-Jun was similarly inhibited by the addition of catalase, whereas the protein level of JNK1 was unaffected (Fig. 5C, 2nd and 3rd panels). SEK1 activity also was suppressed significantly by treatment with catalase compared with that in cells treated with G-Rh2 alone (Fig. 5C, 4th panel). Thus, extracellular H2O2 was involved in both JNK1 and SEK1 activation in G-Rh2-treated cells.
Next we assessed whether Ca2+ also might be involved in the G-Rh2-induced JNK1 activation pathway. The JNK1 activity gradually decreased with increasing concentrations of BAPTA-AM compared with the activity in cells treated with G-Rh2 alone (Fig. 5D, top). In contrast, the protein levels of JNK1 and actin were unaltered (Fig. 5D, middle and bottom). This result indicated that Ca2+ was involved in the ROS-JNK pathway during G-Rh2-induced apoptosis of HeLa cells.
We further examined whether overexpression of catalase inhibited the G-Rh2-induced apoptotic effects. Cells were cotransfected with green fluorescent protein and catalase or JBD. The JNK1 activity, as represented by the levels of phospho-JNK1 or phospho-c-Jun, decreased with increasing expression of catalase (Fig. 6A). These results indicated that JNK1 activity was more significantly suppressed in the catalase-transfected cells than in the cells with externally added catalase (Figs. 5C and 6A). This indicates that intracellular ROS is mainly associated with JNK activation. Cell viability was measured by counting the transfected cells with condensed chromatin structures. The results indicated that expression of JBD or catalase significantly increased the viability of G-Rh2-treated cells by approximately 80 and 70%, respectively, compared with the cell viability of G-Rh2-treated mock transfectants (55%) (Fig. 6B). These results indicated that the intracellular H2O2 of the G-Rh2-treated cells was functionally linked to JNK1 activation during apoptosis.
G-Rh2 Induces the Mitochondrial Release of Smac That Is Associated with ROS-Mediated JNK1 Activation in HeLa Cells. Because the changes in mitochondrial membrane potential lead to the release of apoptogenic factors such as cytochrome c, Smac, or apoptosis-inducing factor, the mitochondrial release of Smac was investigated in cells after treatment with G-Rh2. Smac release into the cytosol was detected within 30 min after treatment with G-Rh2 (Fig. 7A). This G-Rh2-induced Smac release was inhibited in the cells by pretreatment with NAC or catalase or by overexpression of JBD (Fig. 7B).
Because active JNK promotes the mitochondrial translocation of Bax during apoptosis (Tsuruta et al., 2004), we investigated whether ROS-mediated JNK activation was required for Bax translocation to the mitochondria. During G-Rh2-induced apoptotic cell death, the translocation of Bax to mitochondria was increased compared with that of the control (Fig. 7C). When catalase was added, G-Rh2-induced translocation of Bax into the mitochondria was significantly suppressed (Fig. 7C), indicating that G-Rh2-induced ROS was involved in Bax mitochondrial translocation.
ROS Generation Induced by Treatment with G-Rh2 Is a General Event in Various Cancer Cell Lines. To know whether ROS generation by treatment with G-Rh2 is general phenomenon, we tested ROS production using other cancer cells. MCF10A-ras, which is transformed by H-ras oncogene, and MCF7, human breast carcinoma cells, were used to observe ROS generation. The G-Rh2 treatment in MCF10A-ras cells for 4 h induced the cellular levels of ROS compared with those in the control-untreated cells (Fig. 8A). The ROS levels in the G-Rh2-treated MCF7 cells for 3 h were 40% higher than those of the untreated cells (Fig. 8C). The JNK activation and PARP cleavage were also observed in both cancer cells (Fig. 8, B and D).
G-Rh2 Activates JNK1 via Ca2+and ROS Generation in Cancer Cells. The present study shows that G-Rh2-induced JNK activation is mediated by ROS and Ca2+. The early G-Rh2-induced ROS generation (Fig. 2C) and activation of JNK1 that are associated with apoptotic cell death are equally suppressed in HeLa cells by pretreatment with NAC (Figs. 3A and 5B), catalase (Figs. 3, B and C, and 5C), or BAPTA-AM (Figs. 4C and 5D), indicating that Ca2+ and ROS are the up-regulators of JNK1 activation during G-Rh2-induced apoptosis.
Several recent studies have suggested that the increased accumulation of ROS, primarily H2O2, is responsible for the prolonged JNK activation in tumor necrosis factor α-stimulated RelA–/– cells and arsenic-challenged ikkβ–/– cells (Chen et al., 2003; Sakon et al., 2003). Here we show that in G-Rh2-treated cancer cells, including HeLa, MCF10A-ras, and MCF7, ROS production is markedly increased and functionally linked to JNK1 activation during apoptosis. G-Rh2-induced ROS production during apoptosis in cancer cells appears to be a general phenomenon.
Our data also show that Ca2+ is involved in activation of JNK1 and caspase-3 during G-Rh2-induced apoptosis (Figs. 4C and 5D). Ca2+, an important second messenger, is known to be involved in many other cellular functions. Thus, both ROS and Ca2+ are important signaling intermediates for JNK1 activation in G-Rh2-induced apoptosis.
G-Rh2-Induced Generation of ROS and Ca2+AreFunctionally Linked to Mitochondrial Depolarization. Our previous results showed that panaxadiol, a diol-type of ginsenoside and an analog of G-Rh2, induces apoptosis through mitochondrial depolarization and cytochrome c release (Jin et al., 2003). As expected, G-Rh2 induced mitochondrial depolarization and the mitochondrial release of Smac in HeLa cells (Figs. 4A and 7A). These results suggest that G-Rh2-induced apoptosis is functionally associated with oxidative stress and mitochondrial dysfunction. Furthermore, catalase efficiently prevents the depolarization of the mitochondrial membrane potential (Fig. 4B) and Smac release (Fig. 7B). Interestingly, after treatment with BAPTA-AM, a Ca2+ chelator, the G-Rh2-induced depolarization of the mitochondrial membrane potential is prevented, indicating that Ca2+ is likely to be involved in the ROS signaling pathway that is linked to mitochondrial dysfunction.
There is a possibility that the intracellular elevation of Ca2+ in the G-Rh2-induced cells may have resulted from proteolytic degradation of protein that involves pumping cytosolic Ca2+ into the endoplasmic reticulum after treatment of G-Rh2 because the elevation of Ca2+ is related to endoplasmic reticulum and the degradation of protein in endoplasmic reticulum membrane is sensitive to oxidative stress (Ding et al., 2004). Elevation of Ca2+ in the cytosol is likely to be linked to the perturbed mitochondrial permeability transition resulting from depolarization of the mitochondrial redox potential that consequently induces the mitochondrial release of cytochrome c and Smac. In addition, alteration in their permeability transition disturbs the mitochondrial electron transport systems that eventually result in generation of ROS in the cells. Recent studies showed that ROS can act as a “second messenger” to activate ROS release from neighboring mitochondria (Zorov et al., 2006). Thus, G-Rh2-induced ROS release may constitute a positive feedback mechanism that can promote the neighboring mitochondrial release of ROS.
The Release of Mitochondrial Protein May Be Mediated via the ROS-Mediated JNK Activation and Bax Translocalization. Previously, we showed that G-Rh2-induced cell death is inhibited by JBD and thus proposed that JNK activation is required for apoptosis induced by G-Rh2 (Ham et al., 2003). Other studies have suggested that activated JNK translocates into the mitochondria during apoptosis and plays an important role in the release of Smac and cytochrome c (Tournier et al., 2000; Chauhan et al., 2003). Therefore, we have assessed whether G-Rh2-induced ROS generation is associated with the mitochondrial translocation of active JNK. Our results from Fig. 7B suggest that ROS-mediated JNK activation is functionally linked to the mitochondrial release of Smac.
Earlier studies have shown that Bax, a proapoptotic protein that is localized primarily in the cytoplasm (Wei et al., 2001), is redistributed to the outer mitochondrial membrane during apoptosis (Hsu et al., 1997), resulting in the release of apoptogenic factors from the mitochondria into the cytoplasm (Yamaguchi et al., 2003). Our results, which are consistent with earlier reports, indicate that the mitochondrial release of Smac protein into the cytosol is promoted in G-Rh2-treated cells (Fig. 7A), whereas Smac release is significantly attenuated in cells that are pretreated with NAC or catalase to scavenge ROS or that overexpress JBD to inhibit JNK activation (Fig. 7B). Importantly, the Bax protein, which is localized primarily in the cytosol of control cells, is significantly translocated into the mitochondria by treatment of the cells with G-Rh2, and this mitochondrial translocation of Bax protein is prevented by treatment of the cells with catalase (Fig. 7C). These results suggest that the mitochondrial translocation of Bax protein is functionally associated with ROS-linked JNK1 activation during G-Rh2-induced apoptosis.
Taken together, the results of the present study suggest that G-Rh2-induced apoptosis in HeLa, MCF10A-ras, and MCF7 cells is initiated by ROS and Ca2+ generation. The generated ROS leads to the activation of SEK1 and JNK1 and proceeds through the intrinsic pathway that includes Bax-dependent Smac release via mitochondrial depolarization.
G-Rh2 was provided by Dr. Dong-Hyun Kim (Kyung Hee University, Korea). HA-tagged catalase cDNA was provided by Dr. Yun-Sil Lee (Korea Cancer Center Hospital).
This study was supported by the National Research Laboratory Fund (M10104000129-02J000005910), the Ministry of Science and Technology, and Grant R01-2000-000-00113-0 from the Basic Research Program of the Korea Science and Engineering Foundation.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: ROS, reactive oxygen species; SOD, superoxide dismutase(s); G-Rh2, ginsenoside-Rh2; JNK, c-Jun NH2-terminal kinase; JBD, JNK binding domain; Smac, second mitochondrial activator of caspases; SEK1, stress-activated protein kinase/extracellular signal-regulated kinase kinase 1; NAC, N-acetyl-l-cysteine; DCFH-DA, 2′,7′-dichlorodihydrofluorescein diacetate; JC-1, 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide/chloride; GST, glutathione S-transferase; Ac-DEVD-AMC, Ac-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin; BAPTA-AM, 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid acetoxymethyl ester; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; PARP, poly(ADP-ribose) polymerase; HA, hemagglutinin; VDAC, voltage-dependent anion channel; PI, propidium iodide.
- Received June 26, 2006.
- Accepted September 12, 2006.
- The American Society for Pharmacology and Experimental Therapeutics