Sustained Activation of JNK-p38 MAP Kinase Pathways in Response to Cisplatin Leads to Fas Ligand Induction and Cell Death in Ovarian Carcinoma Cells

gel S western blot analysis. B, Nuclear fragmentation in 2008C13 cells after infection. Cells were fixed, and nuclear condensation was analyzed with DAPI. C, Chemoresistant 2008C13 cells were treated as in ( B ), total RNA was extracted, and expression of FasL and beta-actin mRNAs was determined by semi-quantitative RT-PCR.


INTRODUCTION
4 transcription factors. The broad range of these substrates indicates that MAPKs have pivotal roles in cellular signal transduction and suggests that the extent and duration of MAPK activation play key roles in controlling cell functions.
The ERK pathway, which is induced in response to mitogenic stimuli such as peptide growth factors, cytokines, and phorbol esters, involves ERK1 and ERK2 (ERK1/2), the participation of Raf-1 and Ras oncoproteins, and the activation of MEK1/2 (12). Once activated, ERK phosphorylates several substrates, including Elk-1 (13). The ERK pathway plays a major role in regulating cell proliferation and differentiation (12) and provides a protective effect against apoptosis (14). The signaling cascades involving JNK and p38, on the other hand, are key mediators of stress signals and seem to be responsible mainly for protective responses, stress-dependent apoptosis, and inflammatory responses. These cascades can be stimulated by various stresses, such as UV-or gamma-irradiation, osmotic stress, heat shock, proinflammatory cytokines such as tumor necrosis factor-alpha or interleukin-1ß, and chemotherapeutic drugs (10,12).
To understand the molecular basis for the failure of CDDP-based chemotherapy, we compared the cellular response of the human ovarian carcinoma cell line 2008 and its resistant subclone 2008C13 (15) after treatment with a platinum-based anticancer agent. We found that differences in the duration of the activation of MAPK pathways correlated with CDDP-induced apoptosis. A strong, sustained activation of both pathways seemed to be a required priming step for CDDP-induced apoptosis; this activation of both (Life Technologies, Inc.), and 1% streptomycin-penicillin. The 3T3 and 3T3 c-jun -/cells were cultured in Dulbecco's modified Eagle's medium supplemented as described above. Cells were incubated at 37 °C in a humidified atmosphere containing 5% CO 2 . Recombinant adenovirus vectors expressing green fluorescent protein (Ad-GFP) and activated mutants of MKK7 and MKK3 (Ad-MKK7D and Ad-MKK3bE) were constructed as previously described (17). Cells were infected with adenoviruses at a multiplicity of infection of 50 plaque-forming units (pfu)/cell for 5 h and then incubated for another 30 h to allow expression of the protein of interest, as described elsewhere (17).
Drug uptake and Adduct Level assays-Cells were treated with 20µM CDDP, after which cells were washed with PBS and fresh medium was added. Cell pellets were made immediately after 1 min and 1 h of CDDP exposure (1 min, 1 h). Then, after 1 h drug exposure, cells were washed and cultured in a drug-free medium for an additional 4 h (5 h). For protein analysis, cells were digested overnight in 0.2 N NaOH at 55-60 °C. Intracellular platinum levels were determined by solubilizing the cell pellet in hyamine hydroxide and analyzed by flameless atomic absorption spectrophometry (FAAS) using conditions previously described (detection limit = 100 pg Pt) (18,19). For platinum adduct levels, cell pellets were lysed in extraction buffer (10 mM Tris, pH 8.0, 100 mM EDTA, 20 µg/ml Rnase, 0.5% SDS) overnight at 37 °C, then treated with proteinase-K (100 µg/ml, 50 °C, 3 h) and the DNA was extracted in phenolchloroform. The amount of platinum bound to DNA was determined by FAAS.
Immunocomplex Kinase Assays-Cells were serum-starved in 0.1% serum for 12-16 h before CDDP treatment. Whole cell extracts were prepared and treated as previously described (20). Briefly, endogenous JNK1 was immunoprecipitated from 300 µg of cell lysate with the anti-JNK1 monoclonal antibody clone 333.8 (PharMingen, San Diego, CA) and protein A-agarose beads at 4 °C for 4 h. The precipitates were washed twice in lysis buffer and twice in kinase buffer. JNK kinase activity was measured by using 2 µg of glutathione-S-transferase (GST)-c-Jun(1-79) as the substrate, and the reaction was initiated by the addition of 10 µM ATP and 10 µCi of [γ- 32 -ATT TCT GCC ACT GCA GCC CTC AGG-3') and Fas   backward (5'-TCC AGT TCG CTG GGC AGA CTT CTC-3') and FasL forward (5'-ATG TTT CAG CTC TTC CAC CTA CAG A-3') and FasL backward (5'-CCA GAG AGA GCT CAG ATA CGT TGA C 3').
These sequences span nucleotides 76-706 on Fas cDNA and 365-856 on FasL cDNA and yield PCR products of 630 bp and 492 bp, respectively (21). Each reverse-transcribed mRNA product was internally controlled with beta-actin PCR by using the primers forward (5'-TGA CGG GGT CAC CCA CAC TGT GCC CAT CTA-3') and backward (5'-CTA GAA TTT GCG GTC GAC GAT GGA GGG-3'), covering the 2199-3065 region of beta-actin cDNA and giving a 867-bp PCR product (21). The FasL and Fas RT-PCR products were subsequently confirmed by direct sequencing.
Propidium Iodide and DAPI Staining-To detect apoptosis, nuclear staining was performed using 5 µg/ml of 4',6-diamidino-2-phenylindole (DAPI), and cells were analyzed with a fluorescence microscope (magnification X400 for nuclear analysis and X100 for morphologic analysis). Apoptotic cells were identified by morphology and by condensation and fragmentation of their nuclei. The percentage of apoptotic cells was calculated as the ratio of apoptotic cells to total cells counted, multiplied by 100. Three separate experiments were conducted and at least 300 cells were counted for each experiment.

CDDP-Induced Apoptosis in Ovarian Carcinoma Cells-CDDP-sensitive cells 2008 and CDDP-
resistant 2008C13 cells were exposed to CDDP for 1 h, after which the drug was washed out to mimic in vivo chemotherapy. The time course for the induction of apoptosis was determined by microscopic examination of DAPI-stained cells (Fig. 1).
In the chemosensitive 2008 cells, exposure to 20 µM CDDP resulted in morphologic alterations characteristic of apoptosis, including membrane blebbing, nuclear condensation and fragmentation ( To study the mechanism behind this CDDP-resistance, we first studied drug uptake and DNA adduct formation in both the sensitive and resistant 2008 cells ( Table 1). As seen, drug uptake and the DNA adduct formation were similar in the two cell lines after 1 min of CDDP treatment whereas at 1 h and resistance to CDDP in 2008C13 cells is greater than two fold (15,16) this has prompted us to investigate additional mechanisms of resistance. Phosphorylation of ERK was detected at a 1 h time point in the 2008 cells, but an early, somewhat persistent phosphorylation (1 min to 1 h) was observed in the 2008C13 cells (Fig. 2B). This ERK phosphorylation pattern was not sustained and was the opposite of the phosphorylation pattern seen for the JNK and p38 MAPKs. In the resistant cells, ERK phosphorylation also seemed to be biphasic with a second increase occurring at 12-16 h. These data agree with previous findings that CDDP induced apoptosis correlates with an increase in JNK and c-Jun phosphorylation in other cell types (23 ,26,27). Our findings also suggest that the difference in duration of the activation of the JNK and p38 MAP kinase pathways in CDDP-sensitive and -resistant cells could contribute to directing the outcome to survival or apoptosis.

Absence of c-Jun or Inhibition of JNK Pathways Confers Resistance to Apoptosis and Increases
Cell Survival After Treatment with CDDP-The delay between JNK (or p38) activation and the onset of apoptosis, together with the requirement for prolonged JNK (or p38) phosphorylation, suggests that expression of new genes may be required to activate apoptosis. Because the c-Jun transcription factor is an important and specific target for JNK (10) and is possibly involved in the apoptosis triggered by DNAdamaging agents (26,27), we hypothesized that prolonged activation of the JNK and p38 pathways induced by CDDP is required to induce apoptosis in a manner dependent on c-Jun transcription. Conversely, we by guest on March 22, 2020 http://www.jbc.org/ Downloaded from also hypothesized that a transient activation or immediate inactivation of these kinase pathways is too brief to transactivate AP-1-responsive genes and thus would lead to resistance to CDDP-induced apoptosis.
To further investigate the involvement of c-Jun activation in CDDP-induced apoptosis, we compared immortalized 3T3 fibroblast cell lines that have a targeted disruption of the c-jun gene (28) with their parental 3T3 cells that express a wild-type (wt) c-jun. We first analyzed whether the absence of the cjun gene affected cell survival after CDDP exposure. As shown in Fig. 3A, as indicated by DAPI-staining, the CDDP-treated c-jun -/cells were significantly more resistant to apoptosis than were parental cells.
About 80% of the c-jun -/-3T3 cells survived CDDP treatment, but only 25% of the wt c-jun 3T3 cells were still viable after 4 -5 days, suggesting that activation of a downstream set of target genes through c-jun leads to CDDP-induced apoptosis. To determine the possible role of the specific members of the MAP kinase family in mediating this process, we tested the parental 3T3 cells and c-jun -/-3T3 cells with the pyridinylimidazole compound SB202190, a strong inhibitor of JNK and the p38/HOG kinase (29)(30)(31).
Survival measurements revealed that pretreatment with SB202190 led to a significant increase in cell survival in both the absence and the presence of c-jun (Fig. 3A). Inhibition of JNK and p38 decreased the effect of CDDP cisplatin in wild-type c-jun 3T3 cells (72% survived with SB202190 pretreatment versus 26% without SB202190). Results of an immune complex kinase assay, which showed a decrease in phosphorylation of GST-c-Jun by immunoprecipitated JNK (Fig. 3B), confirmed that SB202190 treatment inhibited CDDP-induced c-Jun N-terminal phosphorylation and that this inhibition of c-Jun phosphorylation correlated with the inhibition of apoptosis (Fig. 3A).
These data confirm that c-Jun and its activation by JNK were required for efficient induction of apoptosis by the alkylating agent CDDP. Blockade of JNK activation, whether by a small drug inhibitor or by a lack of c-jun, protected cells from cisplatin-induced apoptosis. Incubation with SB202190 resulted in a marked reduction in cell death (Fig. 4A). Cells treated with CDDP alone, however, displayed the typical features of apoptosis: condensation of the nuclei (Fig. 4A, left panel), shrinkage of the cytoplasm, and membrane blebbing (as seen by phase contrast microscopy; data not shown). Interestingly, pretreatment of the cells with SB202190 markedly suppressed the morphologic changes induced by CDDP (Fig. 4A, left panel). To confirm these findings with an independent assay, we measured apoptosis by propidium iodide staining and flow cytometry. At 24 h after treatment with CDDP, 51% of the 2008 cells showed a hypodiploid (sub-G 1 ) DNA content, reflecting apoptosis (Fig. 4B).
However, incubation with SB202190 before CDDP treatment reduced the extent of cell death considerably, from 51% to 9% (Fig. 4B) and inhibited CDDP-induced cell death in a dose-dependent manner (data not shown).
To determine whether phosphorylation of the c-Jun transactivation domain was necessary for the induction of apoptosis, we transfected 2008 cells with either wild-type (wt) c-Jun or an HA-c-Jun(A63/73) in which the serines at 63 and 73 that are normally phosphorylated by JNK were replaced by alanine, resulting in an inactive protein (20). Transient expression of the HA-c-Jun(A63/73) was associated with a marked decrease in CDDP-induced apoptosis compared with the level seen in cells transfected with wt c-Jun, (Fig. 4C). This protective effect was restricted to the c-Jun(A63/73)-expressing cells, whereas in the surrounding cells that did not express this protein, the induction of apoptosis was not inhibited. As seen in

Expression of FasL in Response to CDDP is Impaired in 2008C13 Resistant Variants But Not in
CDDP-Sensitive Ovarian Carcinoma Cells-Because CDDP's ability to induce apoptosis seems to depend on its ability to activate JNK-and c-Jun-dependent transcriptional events, we next sought to identify JNK and c-Jun targets that might mediate CDDP-induced apoptosis.
To determine whether the Fas ligand gene is a target of CDDP-dependent c-Jun activation, we examined whether the exposure of ovarian carcinoma cells to CDDP affected the expression of FasL. To determine whether the JNK and p38 MAK kinase pathway in ovarian cancer cells was involved in the FasL-mediated apoptosis induced by CDDP, we tested whether inhibiting JNK and p38 with the small drug inhibitor SB202190 would block FasL expression. We found that pretreatment of the 2008 cells with SB202190 significantly inhibited FasL mRNA induction in response to CDDP (Fig. 5D) but had no significant effect on Fas-mediated apoptosis (Fig. 5E). The effect of this inhibition on FasL transcriptional activity might have been due to the inhibition of AP-1 activity, because the SB202190 completely inhibited JNK activation and therefore c-Jun phosphorylation (29)(30)(31). Taken together, these findings strongly suggest that FasL induction was required for CDDP-induced ovarian carcinoma apoptosis and that this process depended on JNK and the phosphorylation of c-Jun at serines 63 and 73. However, incubation of cells with SB202190 did not prevent Fas/FasL from triggering apoptosis (Fig. 5E), suggesting that the JNK pathway might not be downstream or might not be required for Fas-mediated apoptosis.

Caspase Inhibition Does Not Block CDDP-Induced FasL Expression-Initiation of apoptotic cell
death and caspase activation in response to various stimuli, including CDDP, requires the release of cytochrome c from the mitochondrial intermembrane space into the cytosol (35). In the cytoplasm, cytochrome c promotes the assembly of a protein complex called the apoptosome, which includes caspase-9 bound to the CED-4 homologue Apaf-1 (36,37). Upon activation, caspase-9 instigates a proteolytic cascade involving multiple caspases, a process culminating in the cleavage of numerous substrate proteins and, ultimately, cell death (38). Activation of caspase-3 and cleavage of PARP cleavage was seen during 12h-48h after CDDP treatment, in the CDDP-sensitive cells undergoing apoptosis (Fig 1C). Caspases may function both upstream and downstream of Fas/FasL in the apoptosis signaling pathways. To examine further the effect of caspase activation on CDDP-induced apoptosis in the 2008 cells, we used the pancaspase inhibitor zVAD-fmk to determine whether activation of FasL was linked to the caspase activation leading to apoptosis after CDDP treatment. We treated the 2008 cells with CDDP with or without zVADfmk (50 µM) and examined them at 18 and 24 h after the treatment. As shown in Fig. 6, zVAD-fmk substantially attenuated CDDP-induced PARP cleavage and apoptosis, which was determined by visualization of DAPI-stained cells (data not shown); however, zVAD-fmk had no effect, on CDDPinduced FasL induction ( Fig. 6A-C). These findings indicated that in ovarian carcinoma cells, CDDP induced FasL up-regulation is upstream of caspase activation.  (Fig. 6D). This observation agreed with our other findings on the extent and time course of PARP cleavage and apoptosis in these cells (Figs. 1 and 6A and 6B). Whereas JNK activation in the 2008 cells was detected earlier (1 min to 5 h) after CDDP treatment (see Fig. 2), cytochrome c release was markedly delayed, starting only after 12 h (Fig. 6D). These findings are compatible with JNK acting upstream of the cytochrome c release. Notably, cytochrome c release was abrogated in the resistant 2008C13 cells, even at 24 h after the CDDP treatment (Fig. 6D). Thus, it seems that the cells selected for resistance to CDDP have defects in the release of cytochrome c and in the resulting activation of downstream caspases. Moreover, FasL was upregulated when MKK7D and MKK3bE were transduced, and FasL upregulation correlated with the rate of apoptosis (Fig. 7B, C). Control virus expressing GFP had no effect (Fig. 7B, C).

Constitutive Activation of JNK/p38 MAP Kinases through MKK7/3 Sensitizes Chemoresistant 2008C13 Cells to CDDP-Induced Apoptosis by Inducing FasL Expression-That
Taken together, these results suggest that activation of the JNK and p38 pathways through MKK7D and

DISCUSSION
Primary and secondary resistance to chemotherapy is a central problem in cancer treatment (1,4,9).
Resistance to chemotherapy may result from a failure of the apoptosis pathways that are activated in response to drug treatment. Recent evidence indicates that the MAPK family protein kinases JNK and p38 are important mediators of the apoptosis induced by stressful stimuli (25,31,39,44,45). The JNKs and p38 MAPKs are collectively termed stress-activated protein kinases because they are activated by a variety of stress-related stimuli [for review see (10,11)]. The stress kinases are also activated by chemotherapy drugs, including paclitaxel, doxorubicin, vinblastine, and etoposide (46), and by certain DNA-damaging agents, such as 1-D-arabinofuranosylcytosine, CDDP, and mitomycin C (24,46,47). The results described in this study map the early signaling events by which activation of JNK and p38 after CDDP treatment can lead to apoptosis in ovarian cancer cells. Previous studies have described the link between induction of JNK activation and apoptosis in response to cisplatin treatment in ovarian cells and other cell types (25)(26)(27). Those studies also suggested that the transcriptional activity of the c-Jun protein, which is increased by phosphorylation of c-Jun at serines 63 and 73 by JNK, is closely associated with apoptosis. However, none of those studies indicated a potential mechanism by which JNK activation, c-Jun phosphorylation, or both could trigger apoptosis.
Our findings demonstrate that prolonged phosphorylation (1 min to 12 h) of JNK and p38 MAP kinase, accompanied by c-Jun/ATF-2 phosphorylation, is an important step in the apoptosis-signaling cascade induced by CDDP. More importantly, this sustained JNK activation and c-Jun phosphorylation paralleled the phosphorylation of p38 and ATF-2 (Fig. 2). These effects preceded and triggered upregulation of FasL, which in turn contributed to the apoptotic response (Figs, 1, 2, 4, and 5). Thus, the duration of JNK and p38 pathway signaling is a critical factor in determining cell survival or apoptosis; transient activation was insufficient to induce death in CDDP-resistant cells, and prolonged JNK and p38 activation triggered cell death in CDDP-sensitive cells. Our findings indicate that resistance to CDDP in ovarian carcinoma cells is due in part to lack of prolonged activation of stress kinases and phosphorylation of c-Jun and ATF-2, a c-Jun dimerization partner. The transient activation observed in resistant cells seems to be insufficient to induce gene expression of a major initiator of apoptosis, FasL (Fig. 5). This differential response to CDDP between 2008 and 2008C13 cells may be due to differences in cellular uptake and induced DNA damage (Table 1) is unlikely since expression of FasL was uninduced in 2008C13 cells even after a two-fold increase in the drug concentration (Fig. 5A).
We also showed that inhibition of c-Jun activity, either by using a mutant defective in the JNK phosphoacceptor sites [c-Jun(A63/73)] (Fig. 4C) or by inhibiting JNK activation with the small drug inhibitor SB202190 (Fig. 4A, B) could block CDDP-induced apoptosis. Most importantly, this blockade correlated with c-Jun's inability to activate FasL gene expression (Fig 5D). In line with this evidence, others have reported that activation of JNK is required for an apoptotic response to alkylating agents (26,27,39,51). Several studies have suggested that c-Jun is involved in genotoxin-induced apoptosis (52).
One study demonstrated that a dominant-negative c-Jun mutant reduced apoptosis in human monoblastic leukemia cells after exposure to various DNA-damaging agents (53). Both c-jun -/fibroblasts and Jnk1 -/-Jnk2 -/double-knockout murine embryonic fibroblasts were found to be resistant to apoptosis induced by UV irradiation, anisomycin, and alkylating agents (39,51,52,54), all of which may be mediated by the induction of FasL (52). Our data are also consistent with a previous demonstration that long-lasting activation of JNK and p38 kinase after withdrawal of survival factors or induction of MEKK1∆ resulted in enhanced c-Jun phosphorylation and in induction of FasL, leading to neuronal cell death (31,50). Further, the direct inhibition of JNK or c-Jun can block neuronal apoptosis induced by survival-factor withdrawal (44,55,56). Most importantly, mice harboring a mutant allele of c-jun with serines 63 and 73 mutated to alanines are resistant to neuronal apoptosis induced by kainate (48). However, the normal physiological function of c-Jun or JNK, even in the context of a stress response, does not necessarily include induction of apoptosis (10,11).
Tumor cells can inactivate pro-apoptotic cytokines such as Fas/FasL in a way that confers resistance to chemotherapy (8). Furthermore, it is most likely that upstream MAK kinases are involved in this pathway, since selective reactivation of JNK or p38 kinase by MKK7 or MKK3 induced apoptosis in the chemoresistant cells through transcriptional up-regulation of FasL expression (Fig. 7). Interestingly, activation of both MKK6 and p38 is required for gamma-irradiation-induced G 2 arrest, and the expression of dominant-negative alleles of MKK6 or p38 allows cells to escape the DNA damage-induced G 2 delay to transcriptional regulation and that the FasL promoter therefore was directly activated by c-Jun through an AP-1 binding site in transient transfection experiments (33). Other studies have shown that withdrawal of survival factor or activation of the JNK pathway leads to an induction of apoptosis that is preceded by up-regulation of an immediate downstream target, the Fas ligand (FasL), in cerebellar granule neurons and PC12 cells (31,33,58). In addition, inhibition of the interaction of Fas ligand with its receptor Fas leads to a reduction in apoptosis in response to genotoxic stress and growth factor withdrawal (31,33,52,58).
Consistent with the idea that the Fas ligand is a target in the JNK-c-Jun signaling pathway that induces apoptosis is the presence of AP-1 binding sites in the human FasL promoter region, which presumably contribute to the dependence of FasL-Fas interactions on c-Jun phosphorylation (31,33,58,59). Indeed, several reports have identified AP-1 sites in the FasL promoter that are recognized by Jun-Fos or c-Jun-ATF-2 heterodimers. The presence of these sites is required for optimal responsiveness to such cellular stresses as exposure to UV gamma-irradiation and to alkylating agents (33,52,59,60).
In ovarian tumors, FasL may be a pro-apoptotic target of JNK/AP-1 signaling, since inhibition of JNK and p38 with SB202190 led to inhibition of the induction of FasL mRNA (Fig. 5D), and, conversely, reactivation of FasL induction by activated MKK7 or MKK3 (which induced persistent activation of JNK or p38, respectively) triggered cell death through Fas-L expression (Fig. 7). Interestingly, sustained suppression of Fas/FasL has been reported in CDDP-resistant cells, suggesting that the inability of these cells to up-regulate these receptors and ligands may be an important determinant of the ability of the cells to undergo apoptosis in response to chemotherapeutic agents (8). Clearly, c-Jun is required for FasL upregulation since c-Jun-deficient fibroblasts, in contrast to wild-type cells, exhibited a defect in CDDPinduced apoptosis, and since the inhibition of JNK by SB202190 significantly prevented CDDP-induced cell death in wild-type cells (Fig. 3A). In support of our findings is the fact that c-Jun-dependent FasL induction has been demonstrated in several systems in response to DNA-damaging agents, including the topoisomerase II inhibitors, UV irradiation, and the alkylating agent methyl methanesulfonate (21,33,52). Kolbus et al. also showed that the resistance of c-jun -/fibroblasts to apoptosis was accompanied by impaired expression of the FasL providing evidence that c-Jun-dependent expression of FasL represents a rate-limiting step in the apoptosis induced by methyl methanesulfonate (52). Thus, lack of c-Jun activity increases survival to genotoxic stresses (52,54), and reintroduction of the c-jun gene into c-jun -/fibroblast cells sensitizes cells to UV-induced apoptosis (54). The loss-of-function approach in fibroblasts allowed the identification and dissection of c-Jun-dependent and -independent processes upstream and downstream of Fas activation. Once activated, Fas-induced death signaling is not affected by the loss of c-Jun nor JNK activation, demonstrating that only the initiation and not the execution of stress-induced apoptosis depends on c-Jun (Fig. 5E) (52). These data strongly suggest that one mechanism underlying chemoresistance might be the inability to sustain activation of stress kinases and therefore to enable up-regulation of the downstream target death gene, FasL.
Downstream consequences of Fas/FasL interaction are complex and depend in part on the cell type being studied. Some studies implicate JNK in apoptosis and others describe a lack of correlation between JNK and cell death or even interference of JNK activation with apoptosis [for reviews, see (10,11)]. The present study demonstrated that JNK and p38 activation is not due to a downstream FasL signaling event, since JNK and p38 activation occurred during the first hour (i.e., before FasL was expressed 6-18 h later) ( Figs. 2 and 5). Moreover, the caspase inhibitor zVAD-fmk did not prevent CDDP-induced FasL (Fig. 6 Clearly, neither c-Jun nor JNK are required for the expression and activity of cellular components located downstream of Fas, because c-jun -/cells and jnk1 -/-jnk2 -/mouse embryonic fibroblasts were sensitive to FasL-induced apoptosis (39,52).
The molecular pathways triggered by anticancer drugs that lead to the activation of stress pathways are not well understood. Exactly how cisplatin triggers stress kinase pathways is not yet known, nor are the sequential events between CDDP-induced oxidative stress, DNA damage, JNK/p38 activation, and apoptosis. We have also reported that cisplatin-resistance in the ovarian carcinoma is associated with a defect in apoptosis through X-chromosome-linked inhibitor of apoptosis (XIAP) regulation (61). Based on our data and literature reports, several hypotheses can be proposed. It is clear that powerful pro-oxidants that cause generation of reactive oxygen species and free radicals are generated in response to nonredox-active initiators of apoptosis such as cisplatin. In addition, studies have shown that pretreatment of cells with the antioxidants glutathione or N-acetyl-cysteine effectively blocks CDDP-induced apoptosis and CDDP-induced activation of JNK and p38 (57,62,63). Finally, another possibility in the control of stress kinase activities in response to CDDP may relate to interference with phosphatases (64). MAP kinase phosphatases play an important role in selectively regulating the duration of JNK and p38 phosphorylation and dephosphorylation (65) and are activated by various stresses that activate JNK. All known phosphotyrosine and threonine phosphatases, including the dual-specificity phosphatases, contain an essential catalytic cysteinyl residue (66) that is sensitive to thio-(SH)-reactive agents such as CDDP.
Therefore, oxidative stress generated by CDDP not only may deplete reduced glutathione and other antioxidant molecules but also may cause the oxidation of the sulfhydryl groups on these phosphatases, leading to their inactivation (51,67). Our study shows that the sensitization of carcinoma cells to genotoxic stress is largely due to potentiation of the JNK and p38 pathways. Indeed, the CDDP-resistant 2008C13 cells exhibited a defect in the activation of JNK that may contribute to the resistance of advanced tumors to cancer therapy. This suppression of stress kinase activation supports the possible role of an alteration of phosphatase activities in CDDP-resistant cells. Thus, the specific induction of MAP kinase phosphatase by these agents seems to be responsible for protecting cells from apoptosis by preventing prolonged activation of JNK/p38 kinases (51,67). The duration of JNK/p38 activation could thus be regulated by MAP kinase phosphatases through a feedback mechanism. Whether CDDP inactivates JNK in resistant cell lines via stimulation of MAP kinase phosphatases is under investigation by our group.
In conclusion, we have demonstrated that an early key determinant of CDDP-induced apoptosis is the duration of JNK and p38 phosphorylation. Prolonged JNK and p38 activation results in phosphorylation of the target AP-1 transcription factors c-Jun and ATF-2, thus promoting the expression of FasL and binding of FasL to the Fas receptor, which leads to cell death (Fig. 8). Since in CDDP-sensitive cells stress kinase pathways remain potentiated, the cells remain sensitive to stress signals such as genotoxic agents. These findings raise the possibility that defects in this cascade may contribute to a failure of chemotherapy-induced apoptosis. Therefore, modulation of apoptotic pathways through the MAP kinase signaling cascade may become a therapeutic goal for the prevention and treatment of cancer.  treatment, whole-cell extracts were prepared, and protein extracts were resolved by SDS-PAGE and immunoblotted with antiphospho-p38 and antiphospho-ATF-2, antiphospho-JNK or antiphospho-c-Jun, and antiphospho-ERK1/2, which recognize the activated forms of p38, JNK and ERK, respectively. The total amount of p38, ATF-2, JNK1, and c-Jun proteins was assessed using antibodies that recognized these proteins independent of their phosphorylation status.    (Fig. 4A), total RNA was extracted, and expression of FasL and beta-actin mRNAs was determined by RT-PCR. E, same as Fig. 4A, except that cells were treated with Fas monoclonal antibodies after SB202190 treatment as indicated.