Abstract
Viscum album L. coloratum agglutinin (VCA), isolated from Korean mistletoe, is a strong inducer of apoptosis in a variety of tumor cells; however, the underlying molecular mechanisms responsible are not clear. Here, we show that VCA induces apoptotic killing, as demonstrated by DNA fragmentation, Hoechst 33258 staining, terminal deoxynucleotidyl transferase dUTP nick-end labeling assay, and flow cytometry analysis in hepatocarcinoma Hep3B cells. VCA treatment results in a significant increase in reactive oxygen species (ROS) and loss of mitochondrial membrane potential (Δψm). Furthermore, treatment with the antioxidant N-acetyl-l-cysteine reduces ROS induction by VCA, preventing apoptosis in Hep3B cells, indicating that oxidative stress is involved in VCA-mediated cell death. Our results also show rapid changes in mitochondrial transition permeability, Bax translocation, cytochrome c release, caspase-3 activity, and poly(ADP-ribose) polymerase degradation in Hep3B cells occurring in VCA-induced apoptosis. There is much evidence that implicates c-Jun NH2-terminal kinase (JNK) activation with apoptosis in a variety of cellular and animal models. In this study, we show that VCA induces JNK phosphorylation, which is abolished with pretreatment with a JNK inhibitor. Moreover, Hep3B cells overexpressing JNK1 or stress-activated protein kinase kinase (SEK1) seem to be more susceptible to cell death from ROS and loss of Δψm induced by VCA, whereas expression of dominant-negative JNK1 or SEK1 in Hep3B cells do not. These data suggest that JNK phosphorylation may be a major regulator involved in VCA-induced apoptosis. Together, these results suggest that VCA induces apoptosis by inducing ROS production and a loss of Δψm, in which JNK phosphorylation plays a critical role in these events.
Extracts of mistletoe (Viscum album L.) are therapeutically active ingredients in anticancer treatments (Bloksman et al., 1979; Büssing et al., 1996). The European mistletoe lectins (Viscum album L. agglutinin; VAAs), composed of A- and B-chains, have molecular masses between 55 and 63 kDa and bind either to d-galactose alone or with N-acetyl-galactosamine (Franz, 1986; Dietrich et al., 1992). In cell culture and in animal models, VAA elicits cellular responses, supporting the adjuvant effects of mistletoe extracts in cancer therapy. VAA-1 is a potent inducer of apoptosis in several tumor cell lines, inhibiting de novo protein synthesis and activating caspases (Bantel et al., 1999; Savoie et al., 2000). In addition, VAA-1 has been shown to possess immunomodulatory activity in vivo, increasing levels of various cytokines, including interleukin-1, interleukin-6, and TNF-α (Hajto et al., 1998a,b), to induce tumor cell death through natural killer cell activation.
Isolated from the Korean mistletoe (Viscum album L. coloratum), Viscum album L. agglutinin (VCA) is a d-galactose- and N-acetyl-galactosamine-specific lectin that has demonstrated anticancer activity as well (Park et al., 1998, 2000). The N-terminal amino acid and gene sequences of VCA differ from the VAA sequences (Khwaja et al., 1980), but the anticancer effect of VCA is believed to act similarly: the 34-kDa B-chain binds and anchors itself to the tumor cell surface, whereas the 31-kDa A-chain inhibits protein synthesis (Büssing et al., 1999; Lyu et al., 2001). However, recent studies report that liquid extracts of Korean mistletoe inhibit tumor metastasis caused by hematogenous and nonhematogenous tumor cells by suppressing tumor growth (Yoon et al., 1995) and inhibiting tumor-induced angiogenesis (Park et al., 2001). In addition, our previous studies have demonstrated that p53 regulation is associated with telomerase inhibition in VCA-induced apoptosis (Lyu et al., 2002). Despite the known biological and physiological functions of Korean mistletoe, the exact role and molecular mechanisms involved in VCA activity, specifically the apoptotic cytotoxicity of tumor cells, as an anticancer substance are still unclear.
In many hepatocellular models of apoptosis, oxidative stress induced by reactive oxygen species (ROS) is a frequent mediator of apoptosis (Lemasters, 1999). The ability of ROS-mediated oxidative stress during massive cellular damage has been associated with lipid peroxidation, mitochondrial release of the proapoptotic proteins AIF and cytochrome c, loss of mitochondrial membrane potential (Δψm), and depletion of cellular antioxidants such as glutathione (Leist et al., 1997; Crompton, 1999). In addition, the Jun-NH2-terminal kinase/stress-activated protein kinase (JNK/SAPK) subfamily of mitogen-activated protein kinases (MAPKs) is associated with stress responses and can mediate cellular growth and differentiation, as well as apoptosis (Kyriakis et al., 1994; Gupta et al., 1995). Recent studies have pointed toward the activation of JNK/SAPK as the signal transducer for oxidative stress, growth factor deprivation, and cytokine release in several tumor cell lines (Xia et al., 1995). The proapoptotic effects of JNK include induction of cytochrome c release from mitochondria through phosphorylation of Bcl-2 and Bcl-xL, as well as activation of FasL, TNF-α, c-myc, and p53 promoters.
It has been reported that VAA-1 alters mitochondrial transmembrane potential and increases intracellular levels of ROS in human neutrophils, and induces apoptosis through ROS-independent and Mcl-1-dependent mechanisms (Lavastre et al., 2002). However, earlier studies have reported that Korean mistletoe lectin- and European mistletoe-induced apoptosis were regulated by JNK/SAPK activation and caspase-3 and -7 activation in human leukemia cells (Park et al., 2000; Savoie et al., 2000). Although it has been demonstrated that mistletoe lectins induce apoptosis through several apoptotic-related mechanisms, the relationships between release of mitochondrial proteins, caspase activation, ROS production, and loss of mitochondrial permeability transition (MPT) remain unclear. Therefore, the possibility that VCA induces apoptosis through mitochondrial damage by ROS-dependent mechanisms in Hep3B cells has not been excluded.
In this study, intracellular ROS formation and loss of Δψm were observed to play important roles in mediating apoptotic processes induced by VCA. These processes may be regulated by activation of the SEK/JNK pathway. These studies provide a rational basis for new clinical perspectives in future mistletoe therapy.
Materials and Methods
Materials. Anti-ERK1/2, anti-phospho-ERK1/2, anti-p38, antiphospho-p38, anti-JNK, and anti-phospho-JNK antibodies were obtained from Cell Signaling Technology Inc. (Beverly, MA). Antipoly(ADP-ribose) polymerase (PARP), anti-Bcl-2, anti-Bax, anti-Bad, and anti-Bcl-XL antibodies were obtained from BD Biosciences PharMingen (San Diego, CA). Anti-cytochrome c antibody, anti-GCK, and anti-voltage-dependent anion channel (VDAC) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). GST-c-Jun was purchased from Cell Signaling Technology Inc. Dichlorodihydrofluorescein diacetate (DCFH-DA), DiOC6, tetramethylrhodamine methyl ester perchlorate (TMRM), Alexa-488 anti-rabbit IgG, and MitoTracker Red CMXRos were purchased from Molecular Probes (Eugene, OR). SP600125 and the caspase-3 inhibitor I (DEVD-CHO) was purchased from Calbiochem (San Diego, CA). N-acetylcysteine (NAC), cyclosporin A, phenylmethylsulfonyl fluoride, aprotinin, leupeptin, penicillin, streptomycin, rhodamine 123 (Rh123), mouse monoclonal anti-human β-actin (clone AC-15), propidium iodide (PI), and other common chemicals all came from Sigma-Aldrich (St. Louis, MO). The plasmids pcDNA3-HA-JNK1 wild type (wt), pEBG-SEK1 wt, and its dominant-negative mutant pcDNA-HA-JNK1-KR (Lys-Arg), or pEBG-SEK1-KR were from Dr. B. J. Song (National Institute of Alcohol abuse and Alcoholism, National Institutes of Health, Bethesda, MD), which have been described in a previous article (Soh et al., 2000).
Preparation of VCA, Mistletoe Extract, or Powder of Mistletoe. Korean mistletoe grown on oak trees was collected in the winter in South Korea. Leaves, berries, and 1- to 4-year-old plants stems were sorted and chopped into slices. Crude protein solutions were prepared by binding the protein onto the cation exchanger as was described previously (Na et al., 1996; Park et al., 2000). In brief, 3.5 g of SP Sephadex C-50 (Pfizer, Inc., Täby, Sweden) was added to 1 liter of aqueous extract and stirred at 4°C. The gel-filled chromatography column was washed with 0.1 M acetate buffer, pH 4.0, and the proteins were eluted with buffer solution (0.1 M Tris-HCl, pH 8.0, and 0.5 M NaCl). The solution was put on the column filled with asialofetuin-Sepharose 4B, and the column was washed with phosphate-buffered saline (PBS) and concentrated by ultrafiltration (mol. wt. = 10; Millipore Corporation, Bedford, MA). Purity and molecular mass were determined by SDS-PAGE as described previously (Büssing et al., 1999; Park et al., 2001). The hemagglutination and sugar specificity of lectin was measured as described previously (Park et al., 1998). To prepare mistletoe extract, the leaves, berries, and 1- to 4-year-old stems of the plants sorted and chopped in slices and then crushed with 10 volumes of saline between two rollers going in opposite directions in a vegetable juice miller (Angel Life Co., Seoul, Korea). The mixture was separated by filtration through cheesecloth and centrifuged at 12,000 rpm for 30 min. Thereafter, the supernatant was filtered in stages with 60-, 20-, 7.2-, and 0.45-μm pore sizes and lyophilized. To prepare powder of mistletoe, the leaves, berries, and 1- to 4-year-old plant stems were lyophilized and ground into powder.
Cell Culture and Determination of Cell Viability. Monolayers of Hep3B cells were maintained in culture at 37°C and 5% CO2 in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine serum (FBS), 100 IU/ml penicillin, and 100 μg/ml streptomycin. Hep3B cells grown for 1 day in 96-well microtiter plates (1 × 104/well) were incubated with varying concentrations of VCA for different times. Cell viability was measured using 3-[4, 5-dimethyl-thiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) as a substrate.
Stable Transfection. Hep3B cells were grown in six-well culture plates to about 50 to 60% confluence. Wild-type JNK1 and SEK1 or dominant-negative JNK1 (Lys→Arg) and SEK1 (Lys→Arg), and vector control DNA were transfected into Hep3B cells using a Lipofectin reagent (Invitrogen, Carlsbad, CA). After 16 h, the medium was replaced with normal growth medium, and cells were grown for an additional 48 h. Then, the cells were exposed to a selective concentration (as determined by “a killing curve”) of 400 μg/ml geneticin (G418-sulfate; Invitrogen) to isolate stably transfected cells. Positive colonies were passed three times before selecting unique clones. The stably transfected cell lines were maintained under constant selective pressure in the continued presence of 400 μg/ml geneticin. After transfection with the respective cDNA, the level of protein expressed was determined by Western blotting using the specific antibody toward each target protein.
Western Blotting. Cells were lysed in radioimmunoprecipitation assay buffer (50 mM Tris, pH 7.5, 1% Nonidet P-40, 150 mM sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, and 1 μg/ml pepstatin) and then vortexed and centrifuged at 16,000 rpm at 4°C for 10 min. The supernatants were mixed in Laemmli loading buffer, boiled for 4 min, and then subjected to SDS-PAGE. After electrophoresis, proteins were transferred onto nitrocellulose membranes and blotted against primary antibodies for 16 h. Membranes were washed with 0.05% (v/v) Tween 20 in PBS, pH 7.4, and incubated with a 1:4000 dilution of horseradish peroxidase-conjugated secondary antibodies for 45 min. Protein bands were visualized by an enhanced chemiluminescence reaction (Amersham Biosciences Inc., Piscataway, NJ).
Terminal Deoxynucleotidyl Transferase-Mediated dUTP Nick-End Labeling (TUNEL). Apoptotic cells were detected using Apop Tag (Oncor, Gaithersburg, MD), an in situ apoptosis detection kit. In brief, cells were fixed on slides and then incubated with terminal transferase and avidin-labeled-dUTP, followed by further incubation with FITC-labeled anti-avidin antibody. The slides were then examined under fluorescence microscopy.
Flow Cytometric Analysis of Apoptosis. After trypsinization, approximately 106 cells were collected by centrifugation at 1000g for 5 min. Cells were then washed in PBS followed by resuspension and fixation in 70% ethanol for approximately 2 h. Next, cells were washed with PBS and resuspended in 500 μl of PBS containing 100 μg/ml RNase, followed by a 30-min incubation period at room temperature. Cellular DNA was then stained by the addition of 50 μg/ml PI, and cells were analyzed on a FACScan using Cellquest software (BD Biosciences, San Jose, CA).
DNA Fragmentation. Cultured hepatocytes were washed twice with PBS and lysed in buffer (100 mM NaCl, 10 mM Tris-HCl, pH 8.0, 25 mM EDTA, 0.5% SDS, and 0.1 mg/ml proteinase K) at 37°C for 18 h. DNA was extracted with an equal volume of phenol/chloroform (1:1) and precipitated at -70°C. DNA pellets were resuspended in 10 μl of 10 mM Tris, pH 7.8, 1 mM EDTA buffer and incubated for 1 h at 37°C with 1 μg/ml RNase (Roche Molecular Biochemicals, Indianapolis, IN) to remove RNA. DNA pellets were electrophoresed for 2 to 3 h at 90 V on 1.8% agarose gels. The gel was stained with ethidium bromide, and the DNA fragments were visualized under ultraviolet light.
Immunocytochemistry. Hep3B cells were grown on poly-d-lysine-coated coverslips in six-well plates. Cells were pretreated with 100 nM MitoTracker CMXRos during the last 30 min of treatment, washed with PBS, and fixed with 3.7% formaldehyde in PBS for 10 min at room temperature. The coverslips were soaked in a blocking solution (PBS containing 5% bovine serum albumin and 0.2% Triton X-100) for 30 min and incubated with anti-Bax antibody (1:300) overnight at 4°C and then with Alexa-488 anti-rabbit IgG antibody (1: 400) for 30 min in the blocking solution before mounting. The fluorescence was analyzed by confocal microscope (Nikon, Tokyo, Japan). The cells exhibiting strong punctuate staining of Bax, which overlapped with the distribution of MitoTracker CMXRos, were counted as cells with mitochondrially localized Bax.
Measurement of ROS. Intracellular ROS generation was monitored by flow cytometry, using fluorescence produced from 2′,7′-dichlorofluorescein (DCF) after being oxidized from DCFH-DA (Lavastre et al., 2002) (Molecular Probes). In brief, after treatment with 10 ng/ml VCA, cells were incubated with 5 μM DCFH-DA during the last 30 min at 37°C and then detached with trypsin. Cells were washed twice with PBS and resuspended with PBS containing 1% FBS. Fluorescence intensity was measured using a FACSCalibur flow cytometer (BD Biosciences).
Measurement of Δψm. To measure Δψm, cells were incubated with DiOC6 (100 nM; Molecular Probes) during the last 30 min of treatment (Pastorino and Hoek, 2001). DiOC6 is a fluorescent dye that is incorporated into mitochondria in a Δψm-dependent manner. Accumulation of a membrane-permeable cationic fluorescent dye, Rh123 (Sigma-Aldrich), was used to further confirm Δψm (Bai et al., 1999). Cells were incubated with 5 μg/ml Rh123 for 40 min before treatment end. Cells were then harvested by trypsinization, washed with 1× PBS, and resuspended in PBS (for DiOC6) and PBS containing 25 μg/ml PI (for Rh123). Fluorescence intensity was determined by flow cytometry.
Analysis of MPT Using Laser Scanning Confocal Microscope. To monitor MPT changes, the fluorescent probe TMRM (red fluorescence) was used as described by Pastorino and Hoek (2001). TMRM is a cationic fluorophore that accumulates electrophoretically in normal mitochondria because of the negative Δψm. Decreases in TMRM red fluorescence in mitochondria indicate disruption of Δψm. In brief, Hep3B cells were cultured in four-well chamber slides. After washing with PBS, cells were first loaded with 0.5 mM TMRM in FBS-free culture medium at 37°C for 15 min. After staining, cells were washed twice with PBS and then mounted onto a laser scanning confocal microscope (LSM 410; Carl Zeiss GmbH, Jena, Germany) and incubated in FBS-free medium for treatment with VCA (10 ng/ml). The red fluorescence of TMRM was excited simultaneously with 568-nm lines from an argon-krypton laser.
Isolation of Cytosol and Mitochondrial Fractions. Cells were plated on 10-cm2 petri dishes at a density of 5 × 106 cells/dish. After treatment, cells were harvested by trypsinization, washed once in PBS, and resuspended in 3 volumes of isolation buffer (20 μM/ml HEPES, pH 7.4, 10 μM/ml KCl, 1.5 μM/ml MgCl2, 1 μM/ml sodium EDTA, 1 μM/ml dithiothreitol, 10 μM/ml phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, and 10 μg/ml aprotinin in 250 μM/ml sucrose). After chilling on ice for 5 min, the cells were disrupted by 50 strokes in a glass homogenizer. The homogenate was centrifuged twice at 2500g at 4°C to remove unbroken cells and nuclei. Then, the mitochondria were pelleted by centrifugation at 9000g at 4°C for 30 min to obtain the heavy membrane fraction enriched with mitochondria. The supernatant from the initial 9000g centrifugation was centrifuged again at 100,000g to give cytosolic supernatant and light membrane fractions. Mitochondrial and cytosolic fractions (60 μg) were separated on 15% SDS-PAGE gels and immunoblotted with antibody against cytochrome c.
Bax Oligomerization Assay. The Bax oligomerization assay was adapted as described previously (Majewski et al., 2004). In brief, the mitochondria-enriched fraction from isolated cells were subjected to protein cross-linking by incubation in freshly prepared 10 mM bismaleimidohexane (Pierce Chemical, Rockford, IL)/16.8% dimethyl sulfoxide/PBS for 30 min at room temperature with occasional mixing. Samples were boiled for 5 min in 1× Laemmli buffer, resolved by SDS-PAGE (12%), transferred onto nitrocellulose membranes, and subjected to Western blot analysis using an anti-Bax antibody. For Bax-VDAC binding assay, total cellular extracts were immunoprecipitated with anti-VDAC antibody, washed twice with lysis buffer, and then subjected to Western blot against anti-Bax antibody.
JNK Kinase. To assess JNK activity, cells were washed twice with PBS, pH 7.4, containing 1 mM sodium vanadate, and lysed in 0.5 ml of lysis buffer (30 mM Tris, pH 7.5, 150 mM sodium chloride, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, 1% Nonidet P-40, and 10% glycerol). Total cellular extracts were immunoprecipitated with anti-JNK antibody overnight, washed twice with lysis buffer, and then once with kinase buffer (50 mM Tris, pH 7.4, 5 mM MgCl2, 10 mM MnCl2, and 0.1 mM sodium orthovanadate). Pellets were resuspended in 50 ml of kinase buffer containing 2 μg of GST-c-Jun (Cell Signaling Technology Inc.) and 5 mCi of [γ-32P]ATP and incubated at 30°C for 10 min. The reaction samples were boiled in SDS sample buffer containing 2.5% 2-mercaptoethanol for 5 min. The solubilized proteins were resolved by SDS-PAGE and quantified by filmless autoradiographic analysis.
Enzymatic Assay of Caspase-3. Cleavage of the caspase-3 substrate I (N-acetyl-DEVD-p-nitroaniline) (Calbiochem) was used as a measure of caspase-3 activity. The p-nitroaniline was used as the standard. Cleavage of the substrate was monitored at 405 nm, and the specific activity was expressed in picomoles of the product (nitroaniline) per minute per milligram of protein.
Statistical Analysis. For comparing values obtained in three or more groups, one-factor analysis of variance was used, followed by Turkey's post hoc test, and P < 0.05 was taken to imply statistical significance.
Results
VCA Induces Apoptosis in Hepatocytes. First, to examine the cytotoxic effect of VCA on a human hepatocellular carcinoma cell line, Hep3B cells were treated with VCA at different times and doses. Cell viability was measured by the MTT assay or by trypan blue exclusion. VCA significantly inhibited cell viability in a time- and dose-dependent manner (Fig. 1A). To investigate whether VCA-induced cell death was caused by apoptosis, DNA fragmentation was assessed. Although no evidence of genomic DNA cleavage was detected in control cells, VCA-treated cells were markedly fragmented at nucleosomal intervals, ∼200 base pairs, in a time- and dose-dependent manner (Fig. 1B). For further confirmation, the cells were also treated with 10 ng/ml VCA for 12 h and morphological changes were observed under a fluorescent microscope after Hoechst 33258 staining. Compared with untreated controls, VCA-treated cells exhibited a number of morphological changes characteristic of programmed cell death. During the apoptotic course, blebbing of the cell membrane occurred, and the cells were fragmented into characteristic apoptotic bodies (Fig. 1C). In addition, we also examined TUNEL assay and DNA content by FACS analysis. VCA treatment caused a 4.8 ± 0.5-fold increase in TUNEL-positive Hep3B cells (Fig. 1D), and flow cytometric analyses showed that treatment with VCA markedly induced apoptosis (M1, sub-G1 phase, 2.70% in nontreated Hep3B cells versus 18.41% in VCA-treated Hep3B cells) (Fig. 1E). Together, these findings indicate that VCA treatment strongly induced apoptosis in Hep3B cells.
VCA Induces Apoptotic Killing through a Caspase-3-Dependent Mechanism in Hep3B Cells. The role of caspase-3 activation as an important mediator of apoptosis is well established (Allen et al., 1998). Therefore, the effects of VCA on caspase-3 activation in Hep3B cells were determined. After the cells were treated with VCA for various times and various doses, lysates were obtained to measure the catalytic activity of the caspase-3 protease. As shown in Fig. 2A, the activity of caspase-3-like proteases was significantly increased beginning 6 h after VCA treatment and reached a maximal level after 12 h (5.8-fold). VCA also significantly induced activation of caspase-3-like proteases in a dose-dependent manner (Fig. 2B), observed first as a 4-fold increase of caspase-3-like protease activity after 12 h of treatment with 10 ng of VCA. Pretreatment of the cells with a specific caspase-3 inhibitor, z-DEVD-CHO (100 μM), significantly reduced VCA-induced caspase-3 activation (Fig. 2, A and B) and apoptosis (Fig. 2C). To determine whether VCA-activated caspase-3 is functionally active in these cells, cleavage of PARP was examined by Western blot analysis. As shown in Fig. 2D, treatment with VCA rapidly caused cleavage of PARP in Hep3B cells; this cleavage was markedly inhibited by pretreatment with the caspase 3 inhibitor z-DEVD-CHO.
VCA Alters Antiapoptotic and Proapoptotic Protein Expression and Induces Cytochrome c Release from Mitochondria. Because the expression levels of the proapoptotic Bax and antiapoptotic Bcl-2 proteins and their ratios are considered critical factors for initiating apoptosis through mitochondrial damage (Gross et al., 1999), the expression of several antiapoptotic and proapoptotic proteins in VCA-treated Hep3B cells was examined. As shown in Fig. 3A, the expression of Bcl-xL and Bcl-2 was reduced about 3.2- and 2.3-fold, respectively, whereas expression of Bax significantly increased by 2.9-fold after treatment for 24 h. In contrast to changes in Bax and Bcl-xL protein expression, considerable changes in Bad and Apaf-1 expression was not detected. Next, we examined whether expression levels of Bcl-2-related proteins were modified by inhibition of apoptosis by caspase-3 inhibitor. Results show that changes in Bcl-2-related protein expression induced by VCA was not significantly affected by the addition of the caspase-3 inhibitor (Fig. 3B), suggesting that alterations in Bcl-2 family proteins after VCA treatment may not be a late event in apoptosis or reflect secondary apoptosis changes. Release of mitochondrial cytochrome c from cytosol into mitochondria is a key step in the mitochondrial pathway of apoptosis (Esposti et al., 1999). Therefore, we examined whether cytochrome c release was observed in VCA-treated Hep3B cells. As shown in Fig. 3C, VCA treatment caused cytochrome c release from the mitochondria into the cytosol at 6 h, continuing through 24 h. These results suggest that VCA-induced apoptosis may be regulated through mitochondrial damage.
VCA Induces the Translocation of Bax into Mitochondria through Conformational Change as an Early Event in Apoptosis. Recent studies have suggested that some apoptosis-promoting conditions induce Bax translocation into mitochondria and conformational changes in Bax (Putcha et al., 1999; Gilmore et al., 2000) that involve releasing cytochrome c. To investigate this hypothesis, we examined whether VCA induces Bax translocation into mitochondria during apoptosis by experiments using FITC-Bax and MitoTracker CMXRos, a mitochondrial marker. Our results show that FITC-Bax was localized mainly in the cytosol in normal Hep3B cells but that it was distributed in both the cytosol and mitochondria in VCA-treated Hep3B cells (Fig. 4A). The translocation of Bax into mitochondria was observed beginning 6 h after treatment of VCA and peaked at 12 h (Fig. 4B). The addition of 100 μM caspase-3 inhibitor had little effect on Bax translocation induced by VCA (data not shown), demonstrating that VCA-induced Bax translocation may be an early event necessary for apoptosis by caspase-3. The molecular mechanism(s) underlying Bax subcellular redistribution, however, remains a matter to be determined. Previous reports have suggested that conformational changes in Bax through dimerization or direct binding to VDACs, a major permeability pathway for metabolites in the mitochondrial outer membrane, is required for mitochondria-mediated apoptosis (Wei et al., 2001). To explore this hypothesis, we examined whether VCA induces the interaction of Bax with VDAC using isolated mitochondrial fractions. As shown in Fig. 4C, the interaction of Bax with VDAC was significantly increased in VCA-treated cells. In addition, oligomerization of Bax within the mitochondria is believed to play a pivotal role in the induction of cytochrome c release (Wei et al., 2001). Therefore, whether VCA induces Bax oligomerization leading to cytochrome c release from mitochondria was also examined. As shown in Fig. 4D, VCA accelerated and amplified Bax oligomerization in a time-dependent manner, consistent with Bax translocation into mitochondria. These results suggest that VCA-induced apoptosis may be regulated through mitochondrial damage, including alterations in Bcl-2-related proteins and Bax conformational changes.
VCA Induces ROS Production and Reduces Δψm. Mitochondrial damage is important in cell death, and in some models of apoptosis, it is an early occurring event (Lemasters, 1999). Elevated amounts of intracellular ROS are sufficient to trigger cell death, and it has been suggested that ROS are biochemical mediators of apoptosis (Pastorino and Hoek, 2001). Therefore, to determine whether ROS is involved in the regulation of apoptosis induced by VCA, the fluorescent DCFH-DA product was determined using flow cytometry. As shown in Fig. 5A (left) and B (open column), VCA induced greater production of ROS in Hep3B cells than did control cells (M1, 46.62% in control cells versus 73.38% in VCA-treated Hep3B cells) (ROS peak shifted to the right). Next, we examined whether VCA induced-ROS production in hepatocytes could be coincident with changes in Δψm during apoptosis in these cells. To this end, we used the fluorescent lipophilic cation DiOC6 as an indicator of the energy state of the mitochondria. As shown in Fig. 5A (right) and B (closed column), VCA treatment led to a rapid drop in mitochondrial energy, as reflected by a decrease in fluorescence from baseline after 12 h of treatment (M1, 47.55% in control cells versus 77.43% in VCA-treated Hep3B cells) (Δψm peak shifted to the left, indicating that fewer cells retained DiOC6 in their mitochondria). Moreover, the antioxidant NAC (Kowaltowski et al., 1998) suppressed ROS production and loss of Δψm induced by VCA (Fig. 5, A and B) and protected cells from apoptosis (Fig. 5C). These data demonstrate that VCA-induced apoptosis may be closely related to mitochondrial function and membrane permeability. To visualize the changes in MPT caused by VCA, cells were first loaded with TMRM and then treated with VCA in an FBS-free medium. The bright mitochondria were stained with red-fluorescing TMRM, which is typical of polarized mitochondria in Hep3B cells (Fig. 6A). After treatment with VCA for 50 min, the intensity of TMRM fluorescence in Hep3B cells decreased significantly. This fluorescence decreased further after 90 min of VCA exposure and was no longer detectable after 2 h. TMRM fluorescence was reduced to 38.4 ± 7.6% of control cells within 50 min (Fig. 6B). The loss of mitochondrial TMRM was prevented by the addition of the antioxidant NAC (Fig. 6A). Addition of NAC attenuated the decrease in mitochondrial TMRM fluorescence by 87.5% (Fig. 6B). Consistent with this inhibition, NAC also inhibits the release of cytochrome c from mitochondria into cytosol and the change of Bcl-2-related proteins induced by VCA (Fig. 6C). These results suggest that VCA-induced apoptosis is mediated in an oxidative stress-dependent manner, including ROS production, loss of Δψm, and MPT change.
JNK Phosphorylation Plays a Critical Role in ROS Production and Loss of Δψm Induced by VCA. To examine the mechanism involved in ROS production and the loss of Δψm in VCA-mediated apoptosis, we studied the possible involvement of several MAP kinases involved in early signal transduction: JNK/SAPK, ERK1/2, and p38 MAPK. This was accomplished by measuring the activities of JNK/SAPK, ERK1/2, and p38 by Western blot analysis using specific phospho-antibodies. As shown in Fig. 7A, as early as 10 min after VCA treatment, JNK phosphorylation was strongly induced and persisted for up to 30 min. In contrast, neither ERK1/2 nor p38 was strongly activated after VCA treatment. Expression of these proteins was confirmed using antibodies against JNK, ERK, and p38 MAPK. To further confirm induction of JNK activity by VCA, we used an immunocomplex kinase assay using GST-c-Jun as a substrate to measure JNK/SAPK activity. As shown in Fig. 7B, VCA treatment rapidly induced JNK activation with peak effect occurring at 20 min and returning to basal levels at 60 min, whereas expression of JNK protein remained unchanged. Consistent with JNK activation, VCA also activated SEK, directly upstream of JNK, at 10 min and remained activated for 20 min. These data suggest that JNK phosphorylation may be a major regulator involved in the induction of ROS production and loss of Δψm during VCA-induced apoptosis.
To determine this possibility, we studied the effects of the JNK inhibitor SP600125 on apoptosis, ROS production, and loss of Δψm. As shown in Fig. 8A, pretreatment with 30 μM SP600125 markedly inhibited JNK activation induced by VCA. From the results of DNA fragmentation and DNA content analysis, SP600125 pretreatment significantly inhibited the rate of VCA-induced apoptosis (Fig. 8, B and C). Consistent with this inhibition, the levels of VCA-induced ROS production and loss of Δψm were also significantly decreased from pretreatment with SP600125 compared with VCA-treated cells (M1, 83.03% in VCA-treated cells versus 60.65% in SP600125-pretreated Hep3B cells) (Fig. 8D). In contrast, loss of Δψm induced by VCA was also significantly inhibited by SP600125 treatment (M1, 84.04% in VCA-treated cells versus 64.31% in JNK inhibitor-pretreated Hep3B cells), indicating that JNK activation play an important role in ROS production, loss of Δψm, and apoptosis induced by VCA. However, this was not unexpected because JNK is generally a well described downstream target of ROS (Xia et al., 1995). Therefore, to define whether JNK is an upstream or downstream regulator of ROS production, we pretreated Hep3B cells with NAC, followed by treatment with VCA. In these experiments, JNK/SAPK phosphorylation and GST-c-Jun kinase activity induced by VCA were unaffected with pretreatment of NAC (Fig. 8E), suggesting that JNK/SAPK plays a role as an upstream regulator of ROS production, resulting in apoptosis.
A Critical Role for JNK1 Overexpression in Apoptosis Induced by VCA. To further confirm the essential role of JNK activation in cell death, ROS production, and loss of Δψm by VCA, Hep3B cells were stably transfected with JNK1 cDNA, followed by treatment with VCA. As shown in Fig. 9A, greater levels of JNK proteins were detected in three different JNK1-transfected clones compared with vector-transfected clones. GST-c-Jun phosphorylation induced by VCA was significantly increased in JNK1-transfected Hep3B cells and remained strong after 60 min (Fig. 9B). Next, we compared VCA-induced apoptosis in vector- and JNK1-transfected Hep3B cells by measuring DNA fragmentation and MTT assay. As shown in Fig. 9C, treatment with VCA induced significantly more apoptosis in JNK1-transfected Hep3B cells than in vector-transfected cells, which was also confirmed by FACS analysis (M1, sub-G1, 20.4 ± 0.5% in VCA-treated vector-transfected cells versus 38.9 ± 0.5% in VCA-treated JNK1-transfected cells) (Fig. 9D).
Hep3B cells were stably transfected with dominant-negative JNK1 cDNA, followed by treatment with VCA. As shown in Fig. 10, A-C, stable transfection of dominant-negative JNK1 abolished VCA-induced apoptosis. We then attempted to confirm the critical role of JNK1 in the potentiation of ROS production and loss of Δψm. As expected, VCA-induced ROS production was significantly increased in wild-type JNK1-transfected Hep3B cells (M1, 73.03% in VCA-treated control cells versus 89.05% in JNK1 overexpressing Hep3B cells), whereas ROS production was decreased in dominant-negative JNK-transfected cells (M1, 58.95% in dominant-negative JNK1-overexpressing Hep3B cells) (Fig. 10D, left). Similar to effects observed with ROS production, the loss of Δψm induced by VCA was also significantly increased in wild-type JNK1-transfected Hep3B cells (M1, 74.78% in VCA-treated control cells versus 89.04% in JNK overexpressing Hep3B cells), whereas loss of Δψm was decreased in dominant-negative JNK1-transfected cells (M1, 64.04% in dominant-negative JNK-overexpressing Hep3B cells) (Fig. 10D, right). Combining these results, Hep3B cells transfected with wild-type JNK1 are more susceptible to VCA-induced apoptosis, ROS production, and loss of Δψm, suggesting that JNK activation play an important role in these events.
The Critical Role of SEK, an Upstream Signal Mediator of JNK, in ROS Production and Loss of ΔψM Induced by VCA. To examine the possible involvement of signal mediators upstream of JNK activation, Hep3B cells were stably transfected with wild-type SEK or dominant-negative SEK cDNA. As shown in Fig. 11A, JNK phosphorylation by VCA significantly increased in wild-type SEK1-transfected Hep3B cells compared with vector- and dominant-negative SEK1-transfected cells. Next, to examine the functional role of JNK activation on apoptosis induced by VCA, we cotransfected phospho-enhanced green fluorescent protein and wild-type SEK1 or dominant-negative SEK1 and then performed the TUNEL assay and Hoechst 33258 staining. As shown in Fig. 11, B and C, apoptosis induced by VCA was increased in GFP-positive SEK1-transfected cells, observed as green fluorescence, but lower levels of apoptotic cells were detected in dominant-negative SEK1-transfected cells relative to vector-transfected cells, suggesting that VCA induces apoptosis through the SEK1-JNK1 signaling pathway in Hep3B cells. Measuring fluorescence intensity, VCA-induced ROS production and loss of Δψm in both vector- and wild-type SEK1 or dominant-negative SEK1-transfected Hep3B cells were compared. As shown in Fig. 11D, increased ROS production by VCA in wild-type SEK1-transfected Hep3B cells was strongly abolished in dominant-negative SEK1-transfected cells (M1, 84.32 ± 0.4 versus 62.55 ± 0.4). In addition, the loss of Δψm induced by VCA in SEK1-transfected cells was abolished in dominant-negative SEK1-transfected cells (91.3 ± 0.5 versus 69.47 ± 0.5). These results suggest that SEK1-JNK1 signaling pathway plays an essential role in ROS production, loss of Δψm, and apoptosis induced by VCA.
Discussion
Numerous studies have demonstrated that VCA is a potent inducer of apoptosis in several cell lines but that the underlying mechanisms are not clear. In this study, we determined that VCA induces ROS production and loss of Δψm after the release of cytochrome c into the cytosol and alters the ratio of Bax/Bcl-2 expression and Bax distribution. Furthermore, the SEK1-JNK1 signaling pathway may play an important role in the production of ROS, loss of Δψm, and apoptosis induced by VCA in Hep3B cells.
During apoptosis, the proto-oncogene Bcl-2 is thought to control mitochondrial permeability transition, allowing for the release of cytochrome c (Esposti et al., 1999). The balance between the expression of anti- and proapoptotic proteins is important for controlling cell death in several cell types (Gross et al., 1999). Our data show that VCA enhances proapoptotic Bax protein expression but decreases levels of the antiapoptotic proteins Bcl-2 and Bcl-xL compared with control cells. In addition, cytochrome c release from the mitochondria into the cytosol was detected in VCA-treated cells, but considerable change in the expression of Bad, and Apaf-1 by VCA was undetected (Fig. 3). Blocking caspase-3 activation with a specific inhibitor, z-DEVD-CHO, abolished VCA-induced apoptosis in Hep3B cells (Fig. 2), which is consistent with our previous conclusion that VCA-induced apoptosis is caspase-3-dependent in the human leukemia cell line HL-60 (Lyu et al., 2001) but that it had little effect on the change of Bcl-2-related proteins by VCA.
The activation of the proapoptotic molecule Bax are required to mediate cytochrome c release from mitochondria and regulate intrinsic apoptosis pathway involving the mitochondria. In viable cell, a substantial portion of Bax is monomeric and found either in the cytosol or loosely attached to membranes. After a death stimulus, cytosolic and monomeric Bax translocates to mitochondria where it becomes an integral membrane protein and cross-linkable as a homodimer, which results in release of cytochrome c (Putcha et al., 1999). Consistent with those reports, we also observed, after treatment of VCA, an increased Bax translocation from cytoplasm into outer mitochondrial membrane, where it is thought to play a role in the release of proapoptotic factors, including cytochrome c, supporting our immunocytochemistry data using FITC-Bax and MitoTracker CMXRos C (Fig. 4, A and B). Moreover, our data show that VCA induces the oligomerization of Bax in a time-dependent manner, which is consistent with Bax translocation into mitochondria (Fig. 4D). It has been also suggested that oligomerized Bax interacts with resident mitochondria protein such as VDAC, and affect the release of proapoptotic activator, including cytochrome c (Wei et al., 2001). As shown in Fig. 4C, the interaction of Bax with VDAC was significantly increased in VCA-treated cells consistent with previous reports. Collecting the data suggests that VCA-induced Bax conformational change could be an early event followed by cytochrome c release, caspase-3 activation, and then apoptosis.
Mitochondrial damage is important in cell death, and in some models of apoptosis, damage may be an early event in apoptosis (Crompton, 1999) and is consistent with intracellular ROS production and changes in Δψm during apoptosis (Lemasters, 1999). Several studies have suggested that increased ROS production, preceding the loss of Δψm, is correlated with the release of cytochrome c and caspase-3 activation in various cell types (Esposti et al., 1999). However, it is unclear whether the increase in ROS levels after exposure to VCA is caused by losses in Δψm and correlates with the release of cytochrome c and caspase-3 activation in our cell systems. Our study does show clear evidence, however, that VCA treatment significantly induces ROS production and loss of Δψm, which are both shown to occur before cell death (Fig. 5A), and is consistent with an alteration in Bax/Bcl-2 protein ratio, cytochrome c release, and activation of caspase-3 in Hep3B cells. Our results also show that oxidative stress plays a critical role in the regulation of ROS production and loss of Δψm induced by VCA, because anti-oxidants such as NAC and cyclosporine A (not shown) effectively prevented ROS production, loss of Δψm, and cell death (Fig. 5). However, other antioxidants such as superoxide dismutase, deferoxamine, and catalase failed to rescue Hep3B cells from VCA-induced apoptosis (not shown). In contrast to our results, Lavastre et al. (2002) recently reported that VAA-1 alters mitochondrial transmembrane potential and increases intracellular levels of ROS in human neutrophils through ROS-independent and Mcl-1-dependent mechanisms. The reasons for the apparent in cellular responses after treatment with mistletoe lectins are unknown. This apparent discrepancy may be a result of the difference in the cell types, experimental conditions, and methods of lectin extraction used.
The underlying mechanism by which VCA induces ROS production and loss of Δψm in Hep3B cells remains undefined. Several mitochondrial proteins, including Bcl-2 and Bax, may possible to play important roles in VCA-induced ROS and loss of Δψm, so the roles of these mitochondrial proteins in VCA-induced ROS production and loss of Δψm are further studying. The functions of the MAPKs are rather difficult to elucidate, but some acceptable differences have been delineated. For instance, ERK is believed to preferentially regulate cell growth and differentiation (Minden and Karin, 1997), and the JNK and p38 MAPK function mainly in stress responses (Kyriakis et al., 1994). However, JNK and p38 MAPK can also be activated by mitogenic factors, such as epidermal growth factor and phorbol esters (Davis, 1993) and by T cell activation (Brint et al., 2002). Thus, JNK and p38 seem to act not only in apoptosis but also in mitogenic signaling. Although ROS has been reported to activate ERK, JNK, and/or p38 MAPK signaling cascades (Yoshizumi et al., 2000), the role of these kinases in ROS signaling remains obscure. In particular, the direct relationship between MAPK activation and ROS production induced by VCA has not been extensively studied, although it was reported that cytotoxic lectin-II isolated from Korean mistletoe induced apoptotic cell death in the human leukemic cell line U-937, via activation of ERK1/2, p38 MAPK, and JNK/SAPK (Park et al., 2000; Pae et al., 2001). In the present study, VCA strongly activates JNK/SAPK, but weakly activates ERK1/2 and p38 MAPK in Hep3B cells (Fig. 7). In addition, VCA-induced DNA fragmentation is not significantly enhanced when ERK1/2 and p38 MAPK activation is selectively inhibited by PD098059, an ERK1/2-specific inhibitor, and SB203580, a p38-specific inhibitor (not shown). Furthermore, we obtained similar results at lower concentrations of SP600125 (10 or 20 μM), because SP600125 is not a specific inhibitor of JNK/SAPK, but a broad MAPK inhibitor, especially at the 30 μM concentration used in these experiments. From these results, we suggest that ERK1/2 and p38 MAPK do not have opposite effects on cell survival in cytotoxic responses by VCA, which may not contribute to the modulation of VCA-mediated cytotoxic activity. Therefore, we studied the possibility that JNK/SAPK activation is involved in regulating ROS production, loss of Δψm, and apoptosis induced by VCA. Our data show that VCA synergistically enhances induction of ROS production, loss of Δψm, and apoptosis in wild-type JNK1-(Fig. 9) or SEK1-transfected cells (Fig. 11), but weakly in SP600125-treated (Fig. 8) cells and in dominant-negative JNK1- or SEK1-transfected cells (Figs. 10 and 11). These results show that JNK/SAPK activation may mediate ROS production and Δψm loss during apoptosis induced by VCA in Hep3B cells. However, it is difficult presently to directly link the ability of VCA to induce apoptosis solely through JNK activation in these cells, because the specific JNK inhibitor did not completely inhibit VCA-induced apoptosis. Furthermore, our data show that VCA-induced apoptosis, ROS production, and loss of Δψm are only partially inhibited by blocking JNK activation by dominant-negative JNK1 or SEK1 transfection. This was not unexpected because JNK has been a well described downstream target of ROS (Xia et al., 1995; Yoshizumi et al., 2000). To define whether JNK is an upstream or downstream regulator of ROS production, we pretreated Hep3B cells with NAC, followed by treatment with VCA. Figure 8E show that JNK/SAPK phosphorylation and GST-c-Jun kinase activity induced by VCA were unchanged with pretreatment of NAC, suggesting that JNK/SAPK is an upstream regulator of ROS production during apoptosis. However, we can hypothesize that VCA-induced ROS production and loss of Δψm are partially mediated through JNK activation because the possibility of other mechanisms involved in ROS production, loss of Δψm, and apoptosis induced by VCA have not been ruled out.
Several studies have demonstrated that ROS production and apoptosis can be regulated through inhibition of nuclear factor-κB activation, and inactivation of the cell death effectors caspase-3, -8, and -9 (Kowaltowski et al., 1998). It is conceivable, although unproven in the present study, that VCA also activates T cells, and cell death may be mediated by several cytokines such as TNF-α or interferon-γ, which are produced by T cell activation. Studies regarding these potential candidate pathways are currently under investigation.
Finally, we conclude that VCA elicits apoptosis through ROS production and loss of Δψm partially mediated by JNK activation, followed by alteration of the Bax/Bcl-2 protein ratio, Bax translocation, cytochrome c release, and caspase-3 activation in Hep3B cells. Targeting this signaling pathway may offer therapeutic approaches in anticancer therapy in human hepatic cancers.
Acknowledgments
We thank Dr. B. J. Song and M. A. Bae for providing JNK and SEK cDNA. We also thank Dr. Van-Anh Nguyen for peer review of this article.
Footnotes
-
This work was supported by research grants from the Korean National Institutes of Health intramural research grant (347-6111-211-207). This work was supported by research grants from the Korean National Institutes of Health for the 2003 program year.
-
ABBREVIATIONS: VAA, Viscum album agglutinin; TNF, tumor necrosis factor; VCA, Viscum album L. coloratum agglutinin; ROS, reactive oxygen species; Δψm, mitochondrial membrane potential; JNK, c-Jun NH2-terminal kinase; SAPK, stress-activated protein kinase; MAPK, mitogen-activated protein kinase; MTP, mitochondria transition permeability; ERK, extracellular signal-regulated kinase; PARP, poly(ADP-ribose) polymerase; VDAC, voltage-dependent anion channel; GST, glutathione S-transferase; SEK1, stress-activated protein kinase kinase; DCFH-DA, dichlorodihydrofluorescein diacetate; DiOC6, 3,3-dihexyloxacarbocyanine; TMRM, tetramethylrhodamine methyl ester; NAC, N-acetyl-l-cysteine; Rh123, rhodamine 123; PI, propidium iodide; wt, wild type; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; FBS, fetal bovine serum; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling; FITC, fluorescein isothiocyanate; FACS, fluorescence-activated cell sorting; DCF, 2′,7′-dichlorofluorescein; GFP, green fluorescent protein; SP600125, anthra[1,9-cd]pyrazol-6(2H)-one 1,9-pyrazoloanthrone; z-, N-benzyloxycarbonyl; -CHO, aldehyde; PD098059, 2′-amino-3′-methoxyflavone.
- Received April 9, 2004.
- Accepted August 27, 2004.
- The American Society for Pharmacology and Experimental Therapeutics