Mixed lineage kinase 3 (MLK3)-activated p38 MAP kinase mediates transforming growth factor-beta-induced apoptosis in hepatoma cells.

Although transforming growth factor beta1 (TGF-beta1) acts via the Smad signaling pathway to initiate de novo gene transcription, the TGF-beta1-induced MAPK kinase activation that is involved in the regulation of apoptosis is less well understood. Even though the p38 MAP kinase and c-Jun NH(2)-terminal kinases (JNKs) are involved in TGF-beta1-induced cell death in hepatoma cells, the upstream mediators of these kinases remain to be defined. We show here that the members of the mixed lineage kinase (MLK) family (including MLK1, MLK2, MLK3, and dual leucine zipper-bearing kinase (DLK)) are expressed in FaO rat hepatoma cells and are likely to act between p38 and TGF-beta receptor kinase in death signaling. TGF-beta1 treatment leads to an increase in MLK3 activity. Overexpression of MLK3 enhances TGF-beta1-induced apoptotic death in FaO cells and Hep3B human hepatoma cells, whereas expression of the dominant-negative forms of MLK3 suppresses cell death induced by TGF-beta1. The dominant-negative forms of MLK1 and -2 also suppress TGF-beta1-induced cell death. In MLK3-overexpressing cells, ERK, JNKs, and p38 MAP kinases were further activated in response to TGF-beta1 compared with the control cells. In contrast, overexpression of the dominant-negative MLK3 resulted in suppression of TGF-beta1-induced MAP kinase activation and TGF-beta1-induced caspase-3 activation. We also show that only the inhibition of the p38 pathway suppressed TGF-beta1-induced apoptosis. These observations support a role for MLKs in the TGF-beta1-induced cell death mechanism.

The mixed lineage kinases (MLKs) are a family of serine/ threonine protein kinases that function in a phosphorelay module to control the activity of specific mitogen-activated protein kinases (MAPKs) (8 -12). MLKs have been shown to function as MKK kinases and lead to the activation of JNKs via the activation of MKKs (13)(14)(15)(16). Members of the family include MLK1, MLK2 (also called MST), MLK3 (also called SPRK or PTK1), dual leucine zipper-bearing kinase (DLK, also called MUK or ZPK), and leucine zipper-bearing kinase (LZK) (9,1,(11)(12)(13)(14)(15)(16)(17)(18)(19). Studies show that members of the Rho family of GTPases, Rac and Cdc42, have been found to bind to and to modulate the activities of MLK2 and -3, and coexpression of MLK3 and activated Cdc42 leads to enhanced MLK3 activation (15,18,20,21). Recent observations demonstrate that the overexpression of constitutively active forms of Rac1/Cdc42 promotes apoptotic death and that the dominant-negative forms of the pathway members block death (23). Studies on neuronal and other cell types indicate that the MLK family members are likely to act between Rac1/Cdc42 and MKK4 and -7 in death signaling. Overexpression of MLKs effectively induces apoptotic death of cultured neuronal PC12 cells and sympathetic neurons, whereas the expression of dominant-negative forms of MLKs suppresses death evoked by nerve growth factor deprivation or expression of activated forms of Rac1 and Cdc42.
Mitogen-activated protein (MAP) kinases are a group of protein serine/threonine kinases that are activated in response to a wide variety of external stress signals such as UV irradiation, heat shock, and many chemotherapeutic drugs and mediate signal transduction cascades that play an important regulatory role in cell growth, differentiation, and apoptosis (24). During TGF-␤1-induced apoptosis, three MAP kinases (extracellular signal-regulated kinase (ERK), c-Jun NH 2 -terminal kinase (JNK), and p38 kinase) showed simultaneously sustained activation in FaO rat hepatoma cells (25). TGF-␤1-induced apoptosis was markedly enhanced when ERK activation was selectively inhibited by the mitogen-activated protein kinase/ extracellular signal-regulated kinase kinase inhibitor PD98059. In contrast, both interference with p38 activity by overexpression of the dominant-negative (DN) MKK6 mutant and inhibition of the JNK pathway by overexpression of the DN SEK1 mutant resulted in suppression of TGF-␤1-induced apoptosis. Recently, it has been shown that ectopic expression of GADD45b in AML12 murine hepatocytes activates p38 and triggers apoptotic cell death, suggesting that GADD45b participates in TGF-␤1-induced apoptosis by acting upstream of p38 activation.
Because of the widespread role of MAP kinase activation in TGF-␤1-induced cell death signaling in hepatoma cells, we carried out further studies to define the upstream activators of the MAP kinase pathway involved in TGF-␤1-induced cell death signaling. Here, we show that MLKs are mediators of TGF-␤1-induced cell death signaling in hepatoma cells.
Expression Constructs, Transfections, and Reporter Assays-Both DN and WT forms of MLK3 and DN forms of MLK1 and MLK2 were described previously (26). Transfection was performed using Lipofectin (Invitrogen) according to the manufacturer's instructions. For the reporter assay, FaO cells and HepG2 cells were transiently transfected with SBE4-luc (27) and the internal control pCMV-␤-gal in 6-well plates using Lipofectin. After 24-h transfection, cells were treated with 5 ng/ml TGF-␤1 for 24 h in medium. Luciferase activity was quantified by using an enhanced luciferase assay kit (BD Biosciences). Values were normalized with the ␤-galactosidase activity. All assays were performed in triplicate and represented as mean Ϯ S.E. of three independent transfections.
Cell Culture and Generation of Stable Cell Lines-FaO rat hepatoma cells and Hep3B cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin (PSG; Invitrogen). For stable expression of DN and WT MLK3, FaO and Hep3B cells were co-transfected with pCMS-EGFP-WT or DN MLK3 expression plasmid and pcDNA expression plasmid. One day after transfection, cells were selected for neomycin resistance. After 2 weeks of selection, neomycinresistant and fluorescence-positive colonies were isolated, expanded, and analyzed.
Assessment of Cell Survival-In Hep3B cells, the number of healthy, nonapoptotic enhanced green fluorescent protein (EGFP)-positive cells in the same field was assessed. The percentage of surviving cells was calculated relative to the numbers present in control (without TGF-␤1 treatment) wells. The numbers of transfected cells counted in control cultures were at least 200. In FaO cells, apoptotic cells were measured by trypan blue staining, and the percentage of surviving cells was calculated relative to the numbers of surviving cells versus apoptotic cells. Values are the means of counts on 3 wells Ϯ S.E. Similar results were obtained in three additional independent experiments.
Immunoblot Analysis-Whole-cell extracts were obtained in a radioimmune precipitation assay buffer (10 mM Tris, pH 7.5, 150 mM NaCl, 0.1% SDS, 1.0% Triton X-100, 1% deoxycholate, 5 mM EDTA, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and protease inhibitor mixture (Complete, Roche Applied Science)). Extracts were separated by SDS-PAGE followed by electrotransfer to polyvinylidene difluoride membranes, were probed with polyclonal or monoclonal antiserum, followed by horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG, respectively, and were visualized by chemiluminescence according to the manufacturer's instructions (Pierce). For immunoprecipitation the cell lysates were incubated with the appropriate antibody for 4 h, followed by incubation with GammaBind beads (Amersham Biosciences) for 1 h at 4°C. The beads were washed four times with the buffer used for cell solubilization. Immune complexes were then eluted by boiling for 3 min in 2ϫ Laemmli buffer, and then extracts were analyzed by immunoblotting as described above.
Immune Complex Protein Kinase Assay-Kinases MKK3/6 and p38 and kinases MKK4/7 and JNK were immunoprecipitated from control cell lysates using the appropriate antibody, respectively. MLK3 was obtained from cell lysates that were harvested after incubation with or without TGF-␤1 for the indicated times. Immunocomplexes were recovered with the aid of GammaBind beads and were washed twice with lysis buffer containing 500 mM NaCl, twice with lysis buffer, and twice again with kinase buffer (20 mM MOPS, pH 7.2, 2 mM EGTA, 20 mM MgCl 2 , 1 mM dithiothreitol, 1 mM sodium orthovanadate, and 25 mM ␣-glycerol phosphate). The MLK3 precipitates were mixed with MKK3/6 and p38 and with MKK4/7 and JNK, respectively. The kinase reaction was initiated by addition of 200 M ATP and 2 g of substrates.
The activities of p38 were examined by using purified ATF2 as substrate, and the kinase activities of JNK were assayed with c-Jun as substrate. Following 15-min incubation at 30°C, the reactions were terminated by the addition of SDS loading buffer. Immunoblots were performed using antiphospho-ATF2 and antiphospho-c-Jun antibodies, respectively.

MLK Family Members Are Expressed in FaO Rat Hepatoma
Cells-RT-PCR was used to determine the expression of members of the MLK family in FaO cells. As shown in Fig. 1, transcripts for MLK1, MLK2, MLK3, and DLK were readily detectable in FaO cells. We did not observe the effects of TGF-␤1 on expression of any of the analyzed genes.
Phosphorylation of MLK3 and Activation of MLK3 Substrates in Response to TGF-␤1 Treatment-We next addressed whether TGF-␤1 treatment affects endogenous MLKs. We only examined MLK3 because of the quality of the available antibody. TGF-␤1 treatment did not induce the level of endogenous MLK3 protein expression in FaO cells and Hep3B human hepatoma cells. Because there is evidence that MLK3 is activated by phosphorylation (12), we examined the phosphorylation of endogenous MLK3 induced by TGF-␤1-induced apoptotic stimuli. The increase in MLK3 phosphorylation level was gradual, persistent, and apparent within 30 min of TGF-␤1 treatment, with maximum activation occurring 8 h after TGF-␤1 treatment and slowly declining to basal levels over 24 h in FaO cells RNA was reverse transcribed, and expression levels of MLK1, MLK2, MLK3, and DLK were evaluated by PCR using specific primers as described under "Materials and Methods." Cyclophilin served as an internal control. For each primer pair, PCR assays were performed at the same cycle number. All assays were performed at least twice; representative results are shown. (Fig. 2a). Interestingly, changes in phosphorylation levels were observed only in the fast migrating form. The phosphorylation of the more slowly migrating form was not modulated by the TGF-␤1 stimulation. In Hep3B cells, we observed bimodal MLK3 phosphorylation in response to TGF-␤1 (Fig. 2b). The primary peak of MLK3 phosphorylation was rapid and tran- and Hep3B cells (b) were harvested after incubation with or without TGF-␤1 for the indicated times, and TGF-␤1-mediated MLK3 activation was assayed by phospho-MLK3 blotting following immunoprecipitation of MLK3. Immunoreactive MLK3 was probed by using anti-MLK3 antibody. The same blots were reprobed with anti-␤-actin antibody for ␤-actin to ensure equal loading. c and d, MLK3 activates p38 and JNK activities. MKK3/6 and p38 as well as MKK4/7 and JNK were immunoprecipitated from control cell lysates using the appropriate antibody, respectively. MLK3 was obtained from cell lysates that were harvested after incubation with or without TGF-␤1 for the indicated times. The MLK3 precipitates were mixed with MKK3/6 and p38 and with MKK4/7 and JNK, respectively. The activities of p38 were examined by in vitro kinase assay using purified ATF2 as substrate (c), and the kinase activities of JNK were assayed with c-Jun as substrate (d). sient, peaking at 30 min, followed by a decline toward base-line values by 4 h. The primary peak of MLK3 phosphorylation was followed by a gradually rising, sustained secondary peak, maximal between 12 and 24 h. We next determined the activation of the downstream substrates, c-Jun and ATF2, of both p38 and JNK in cell extracts by in vitro kinase assays in FaO cells (Fig.  2, c and d). The timing of the maximum activation of time-dependent phosphorylation of c-Jun and ATF2 coincided well with that of maximum MLK3 activation.
Expression of MLK3 Enhances TGF-␤1-induced Apoptosis-To investigate whether MLK3 is directly involved in TGF-␤1-induced apoptosis, we generated cell lines stably overexpressing MLK3 and the dominant-negative form of MLK3 (dnMLK3) in FaO cells (Fig. 3a) and Hep3B cells (Fig. 4a). Without TGF-␤1 treatment, apoptotic cell death was not observed in cells expressing MLK3. However, TGF-␤1 treatment significantly enhanced TGF-␤1-mediated apoptosis compared with the control cells in FaO cells (Fig. 3, c and d) and Hep3B cells (Fig. 4, b and c). FaO cells expressing MLK3 had ϳ7% survival, whereas the control FaO cells had 12% survival by 24 h after TGF-␤1 treatment. In Hep3B cells, the survival of MLK3-expressing cells was about 7%, whereas that of the control cells was about 20% by 72 h after TGF-␤1 treatment. Because MLK3 was cloned into the pCMS.EGFP vector, stable cells expressing either MLK3 or the dominant-negative form of MLK3 could be visualized by green fluorescence. Counts of intact GFP-positive cells revealed a rapid decline in numbers within 72 h after TGF-␤1 treatment in the Hep3B cells expressing MLK3 (Fig. 4b). Previously, we have demonstrated that caspase-3 is involved in TGF-␤1-induced apoptosis and that TGF-␤1 treatment induced the progressive proteolytic processing of caspase-3 (4). Cleavage of caspase-3 was markedly increased in FaO cells expressing MLK3 (Fig. 2d). To assess the effect of the dominant-negative form of MLK3 on TGF-␤1induced apoptosis, stable cell lines were generated using the dominant-negative MLK3 cloned into pCMS.EGFP as described previously (26). Expression of dnMLK3 blocked TGF-␤1-induced apoptosis and cleavage of caspase-3 (Figs. 3 and 4). Thus, our results indicate that activation of MLK3 by TGF-␤1 is an important step in the apoptotic pathway of TGF-␤1.
To test the possibility of whether the members of the MLK family play distinct or overlapping roles in mediating a TGF-␤

FIG. 5. Ectopic expression of the dominant-negative forms of MLK1, MLK2, and MLK3 suppresses TGF-␤1-induced apoptosis.
Dominant-negative forms of MLK1 (pCDNA3 His-dnMLK1 (K171A)) and MLK2 (pCMS-EGFP-HA-dnMLK2) were transfected into FaO cells. One day after transfection, cells were selected for neomycin resistance. After 2 weeks of selection, neomycin-resistant and fluorescence-positive colonies were pooled, expanded, and analyzed. Expression levels of dominant-negative forms of MLK1, MLK2, and MLK3 in FaO cell lines were examined using anti-His antibody for dnMLK1 and anti-GFP antibody for dnMLK2 and dnMLK3. The TGF-␤1-induced activation of caspase-3 was examined in these stable cell lines after TGF-␤1 treatment for 24 h. effect, we also generated cell lines stably overexpressing the dominant-negative forms of MLK1 and MLK2 in FaO cells. Their overexpression had the same effect as that of the MLK3 mutant on TGF-␤1-induced apoptosis (Fig. 5).
MLK3 Does Not Inhibit TGF-␤/Smad-mediated Transcriptional Responses-We next investigated whether MLK3 is primarily/solely involved in mediating the apoptotic activity of TGF-␤1 or if it is also an important mediator for other types of TGF-␤1 activities, such as the activation of Smad-mediated transcription of target genes, using FaO cells stably expressing wtMLK3 and dnMLK3. To analyze the effect of MLK3 on TGF-␤/Smad-dependent transcription, FaO cells stably expressing wtMLK3, dnMLK3, and control vector were co-transfected with SBE4-luc, which contains four tandem repeats of the Smad-binding element (SBE) (27). Ectopic expression of wtMLK3 or dnMLK3 had no effect on the TGF-␤1-induced transcriptional activity of this reporter gene construct (Fig. 6a). We also examined whether MLK3 mediates TGF-␤/Smad-mediated transcription of endogenous target genes. Vector control-, wtMLK3-, and dnMLK3-expressing FaO cells were analyzed for the expression of several TGF-␤/Smad-responsive genes by Western blot analysis. Cells were treated with TGF-␤1 for 24 h, and Western blot analysis of p21 and c-Myc was performed. TGF-␤1 treatment increased the levels of p21 and p15 proteins in all three cells (Fig. 6b), suggesting that MLK3 is primarily/solely involved in mediating the apoptotic activity of TGF-␤1 in FaO cells.
Activation of MAP Kinases in FaO Cells Expressing MLK3 during TGF-␤1-induced Apoptosis-As shown in Fig. 2a, the phosphorylation level of MLK3 increases with maximum acti- vation occurring 8 h after TGF-␤1 treatment. All three MAP kinases (ERK1/2, JNK1/2, and p38) show a gradual increase with maximum activation occurring 20 h after treatment with TGF-␤1 (Fig. 7a). These results suggest that the activations of all three MAP kinases follow the activation of MLK3.
We then investigated the activities of all three MAP kinases (ERK1/2, JNK1/2, and p38) in either MLK3-or dnMLK3-expressing FaO cells during TGF-␤1-induced apoptosis. In the MLK3-expressing cells (Fig. 7b, wtMLK3-#1), all three MAP kinases were not activated in the absence of TGF-␤1 treatment, whereas these MAP kinases were activated in the presence of TGF-␤1 treatment. The activation levels of all three MAP kinases were significantly enhanced in FaO cells expressing MLK3 compared with the vector control cells that had been treated with TGF-␤1 (Fig. 7b). In contrast, the dominant-negative form of MLK3 blocked TGF-␤1-induced activation of all three MAP kinases. Therefore, our results indicate that the activation of MLK3 by TGF-␤1 is an important step in the TGF-␤1-induced activation of these MAP kinases.
Role of MAP Kinases Activated in TGF-␤1-induced Cleavage of Caspase-3-To determine the role of activated MAP kinases in TGF-␤1-induced cleavage of caspase-3 in either MLK3-or dnMLK3-expressing cells, the effects of mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor PD9805923, a p38 kinase inhibitor, SB203580, and a JNK inhibitor, SP600125 (anthra [1,9-cd]pyrazol-6(2H)-one), were examined. Treatment with PD98059 accelerated TGF-␤1-induced apoptosis in all three cell lines. However, treatment with SB203580 completely abolished TGF-␤1-induced cleavage of caspase-3 in vector control cells, and TGF-␤1-induced cleavage of caspase-3 was markedly suppressed in both wtMLK3-and dnMLK3-expressing cells (Fig. 8a). In contrast, treatment with SP600125 slightly enhanced TGF-␤1-induced cleavage of caspase-3. To verify the specificity of the JNK inhibitor SP600125 (28), we examined whether the TGF-␤1 induction of JNK activation is suppressed by SP600125 in FaO cells. A recent study shows that TGF-␤ induction of JNK1 activity is blocked by the addition of the JNK inhibitor SP600125 (29). As shown in Fig. 8b, treatment with SP600125 at 10 mM markedly decreases TGF-␤1-induced JNK activation. DISCUSSION Although several elements in the TGF-␤-mediated apoptotic signaling pathway that lead to JNK and p38 kinase activation have been defined, a key question that remains unresolved is how the TGF-␤ signaling pathway activates JNK and p38 kinases.
JNK and p38 MAP kinases, stress-induced MAP kinases, have been shown to participate in the induction of apoptosis by TGF-␤1 (25,30). TGF-␤1-induced apoptosis in CH33 cells is mediated by Daxx, which facilitates JNK activation (31). Daxx, a protein associated with the Fas receptor that mediates activation of the Jun amino-terminal kinase and programmed cell death induced by Fas, interacts physically with the cytoplasmic domain of the type II TGF-␤ receptor. Recently, Yoo et al. (30) reported that the ectopic expression of GADD45b in AML12 murine hepatocytes is sufficient to activate p38 and to trigger apoptotic cell death, whereas antisense inhibition of GADD45b expression blocks TGF-␤-dependent p38 activation and apoptosis, suggesting that GADD45b participates in TGF-␤-induced apoptosis by acting upstream of p38 activation. We have shown previously that during the progression of TGF-␤1-induced apoptosis in FaO cells, all three MAP kinases exhibit concurrent activation; however, the strong and sustained proapoptotic signals from both p38 and JNK mediate TGF-␤1-induced apoptosis in FaO cells (25). The significant contribution of both p38 and JNK activated by TGF-␤1 to apoptosis of FaO cells was confirmed by studies using a specific inhibitor of p38, genetic constructs encoding dominant-negative forms interfering with the p38 or JNK pathway, and transient expression of WT p38␣, p38␤, JNK2, the dominant-negative form of JNK2, or the dominant-negative form of SEK1.
Mixed lineage kinases (8) were identified as dual specificity kinases, i.e. kinases with both serine/threonine and tyrosine kinase activity (8 -11). Although the tyrosine kinase activity of MLKs has not yet been reported, MLKs contain the tyrosine kinase signature motif in addition to the serine/threonine kinase motif (for review, see Ref. 10). All the known MLKs in mammalian cell lines act as MAPK kinase kinases (MKKKs) to activate the c-Jun amino-terminal kinase pathways. MLK3 and DLK have also been shown to activate the p38 MAPK pathway, but it is not known whether other MLKs can also activate p38. Therefore, in this study, we explored whether MLKs are involved in the TGF-␤1-induced apoptotic cascades by activating MAP kinases.
Our study indicates that the MLK family plays a role in TGF-␤1-induced apoptosis in hepatoma cells. As shown in Figs. 2 and 7a, the activation of all three MAP kinases following the activation of MLKs suggests that MLKs participate in TGF-␤1induced apoptosis by acting upstream of p38 activation. First, transcripts for members of the MLK family, MLK1, MLK2, MLK3, and DLK, are expressed in FaO hepatoma cells. Second, TGF-␤1 treatment induced phosphorylation of MLK3 in a timedependent manner. Eleven in vivo phosphorylation sites of MLK3 were identified, and most of the phosphorylation sites are clustered at the carboxyl terminus (32). However, the phosphorylation sites of MLK3 induced by TGF-␤ are not known. Third, overexpression of MLK3 in FaO and Hep3B cells enhanced TGF-␤1-induced apoptosis. Lastly, the dominant-negative form of MLK3 effectively protected both the FaO and Hep3B cells from apoptosis induced by TGF-␤1.
MLKs are known to activate JNK in vivo by directly phosphorylating and activating the JNK kinase SEK1 (MKK4 and MKK7) (13)(14)(15)(16). The ability of MLK3 to activate the ERK/ MAPK pathway has been reported (33); the study demonstrated that the overexpression of wild type MLK3 leads to a morphological transformation of NIH3T3 fibroblasts and that MLK3 preferentially activates the JNK/stress-activated protein kinase (SAPK) cascade, and to a lesser degree, the ERK pathway. Interestingly, MEK1 is highly phosphorylated in vivo in MLK3-transformed fibroblasts, whereas ERK phosphorylation was barely detectable. Recently, another report indicated that although MLK3 can activate MEK1, sustained JNK activation induced by MLK3 can result in the attenuation of the mitogen-activated ERK pathway (34). However, in this study, we demonstrated that in FaO cells expressing MLK3, all three MAP kinases, ERK, JNK, and p38, were markedly activated in response to TGF-␤1. The different patterns of activation of the stress kinases in different model systems may be caused not only by the cell types or stress stimuli but also by the timing, intensity, and duration of activation. ERK has been reported to be cytoprotective against apoptosis triggered by oxidative stress, tumor necrosis factor ␣, growth factor deprivation, and proapoptotic drugs (35)(36)(37)(38). In this study, we have observed that enhanced activation of ERK in FaO cells could not override the TGF-␤1-induced proapoptotic signals. These findings imply that the strong apoptotic signal induced by TGF-␤1 may be able to block cell responses to growth and survival factors acting through the ERK/MAPK pathway. Supporting evidence includes the observation that treatment with PD98059, a specific inhibitor of mitogen-activated protein kinase/extracellular signal-regulated kinase kinase, significantly enhanced TGF-␤1induced caspase-3 activation.
Cell death is controlled by the integration of multiple signals through a complex network including both positive and negative regulators of apoptosis. In this study, we also attempted to examine the specific role of each MAP kinase on TGF-␤1induced apoptosis. In a previous study, we showed that both interfering with p38 activity by overexpression of the dominant-negative MKK6 mutant and inhibition of the JNK pathway by overexpression of the DN SEK1 mutant resulted in the suppression of mitochondrial cytochrome c release, abrogating TGF-␤1-induced apoptosis (25). Our results demonstrated that interfering with the p38 pathway with a p38 kinase inhibitor, SB203580, markedly decreased TGF-␤1-induced caspase-3 activation, whereas inhibition of the JNK pathway with a JNK inhibitor, SP600125, slightly increased TGF-␤1-induced caspase-3 activation. In the previous study, we could not observe the specific role of JNK on TGF-␤1-induced apoptosis because DN SEK1 blocked not only the TGF-␤1-induced JNK activation but also the TGF-␤1-induced p38 activation. The role of SEK1 in the activation of both JNK and p38 is supported by the finding that SEKϪ/Ϫ fibroblasts exhibit defects in both JNK and p38 phosphorylation (22).
Taken together, our results provide the first evidence that MLKs serve as upstream mediators of TGF-␤1-induced activation of MAP kinases in TGF-␤1-induced apoptotic cascades and that TGF-␤1 induces phosphorylation of MLK3. TGF-␤1 treatment activates not only p38 but also JNK and ERK, but TGF-␤1 induces apoptosis in FaO cells through the p38 MAP kinase pathway. The involvement of p38 in TGF-␤1-induced apoptosis is further supported by the recent study in AML12 cells and primary hepatocytes showing that TGF-␤1-induced apoptosis in these cells is mediated through the activation of only p38 MAP kinase (30).