Negative Regulation of MEKK1-induced Signaling by Glutathione S-Transferase Mu*

Mitogen-activated protein kinase/extracellular signal-regulated kinase kinase kinase 1 (MEKK1) is an important component in the stress-activated protein kinase pathway. Glutathione S-transferase Mu 1-1 (GST M1-1) has now been shown to inhibit the stimulation of MEKK1 activity induced by cellular stresses such as UV and hydrogen peroxide. GST M1-1 inhibited MEKK1 activation in a manner independent of its glutathione-conjugating catalytic activity. In vitro binding and kinase assays revealed that GST M1-1 directly bound MEKK1 and inhibited its kinase activity. Co-immunoprecipitation analysis showed a physical association between endogenous GST M1-1 and endogenous MEKK1 in L929 cells. Overexpressed GST M1-1 interfered with the binding of MEKK1 to SEK1 in transfected HEK293 cells. Furthermore, GST M1-1 suppressed MEKK1-mediated apoptosis. Taken together, our results suggest that GST M1-1 functions as a negative regulator of MEKK1.

Glutathione S-transferases (GSTs) are a group of detoxification proteins that catalyze the conjugation of reduced GSH to various electrophilic molecules. GSTs also function as nonenzymatic binding ligands that interact with a variety of lipophilic compounds including steroid hormones (25). The mammalian cytosolic GSTs are classified into five classes based on their primary structure: Alpha; Mu; Pi; Sigma; and Theta (25). The Alpha, Mu, and Pi are the major ones among these classes. There are many lines of evidence that GSTs have a protective role against a variety of cytotoxic agents (26 -28). The Pi and Alpha classes of GST have been shown to suppress JNK (26,29). We previously reported that GST M1-1 functions as an endogenous inhibitor of apoptosis signal-regulating kinase 1 (ASK1), a MAP3K of the JNK/SAPK and p38 mitogen-activated protein kinase pathways, thereby suppressing ASK1-mediated cell death (28).
To further understand the mechanism underlying the modulation of the JNK/SAPK signaling pathway by GST M1-1, we have investigated a possible effect of GST M1-1 on the components of this pathway and report in this study that GST M1-1 physically associates with MEKK1 and inhibits its kinase activity. GST M1-1, by binding to MEKK1, interferes with an interaction between MEKK1 and its substrate SEK1. Thus, our findings suggest that MEKK1 is a new target of GST M1-1.

EXPERIMENTAL PROCEDURES
Cell Culture and DNA Transfection-Human embryonic kidney 293 (HEK293) cells and mouse fibrosarcoma L929 cells were cultured at 37°C in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Hyclone). For DNA transfection, HEK293 cells were plated with a density of 3 ϫ 10 6 cells/100-mm culture dish, incubated overnight, and then transfected with the indicated expression vectors by calcium phosphate method (30) or with the use of LipofectAMINE (Invitrogen).
Immune Complex Kinase Assay-Cells were lysed in buffer A that contained 40 mM Tris-HCl, pH 8.0, 120 mM NaCl, 1% Triton X-100, 2 mM EDTA, 2 mM EGTA, 25 mM ␤-glycerophosphate, 10% glycerol, 25 mM NaF, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 10 g/ml aprotinin. Cell lysates were subjected to microcentrifugation at 12,000 ϫ g, and the resulting supernatants were subjected to immunoprecipitation with the use of the indicated antibodies. The immunoprecipitates were rinsed with buffer  Where indicated, the cells were irradiated with 80 J/cm 2 UV light and incubated further for 30 min. Cells were lysed, and cell lysates were subjected to immunoprecipitation with anti-HA antibody. The immunoprecipitates were assayed for MEKK1 activity in the absence or presence of GST M1-1 (2 g protein/assay) or GST P1-1 (2 g protein/assay), respectively. B, HEK293 cells were transfected with expression vectors encoding FLAG-SEK1 or HA-JNK1. After 48 h of transfection, the cells were exposed to 80 J/cm 2 UV light where indicated, incubated further for 30 min, and subjected to immunoprecipitation with anti-FLAG or anti-HA antibody. The resulting precipitates were examined for SEK1 or JNK1 activity in the absence or presence of GST M1-1 (2 g protein/assay).
Co-immunoprecipitation-Cell lysates were lysed with buffer B that contained 50 mM Tris-HCl, pH 8.0, 120 mM NaCl, 0.5% Nonidet P-40, 1 mM EDTA, 10 mM NaF, 2 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 10 g/ml aprotinin, and the cell lysates were subjected to microcentrifugation at 12,000 ϫ g at 4°C. The soluble fraction was subjected to immunoprecipitation with the indicated antibodies. The resulting immunoprecipitates were subjected to SDS-PAGE and immunoblot analysis as described previously (28).
Cell Death Assay-HEK293 cells were transiently transfected with pEGFP (Clontech) and the indicated plasmids. After 48 h of transfection, the cells were unexposed or exposed to the indicated agents. Cells were fixed with 4% paraformaldehyde and stained with 4Ј,6-diamidino-2-phenolindole. The number of cells with apoptotic nuclei was counted among GFP-expressing cells by fluorescence microscopy. The apoptotic cell death was calculated as the percentage of apoptotic cells over total GFP-positive cells.

RESULTS AND DISCUSSION
GST M1-1 Inhibits MEKK1 Activity-To investigate whether GST M1-1 regulates MEKK1, HEK293 cells were transfected with a plasmid vector encoding HA-tagged MEKK1 alone or together with a plasmid encoding FLAG-tagged GST M1-1. The exposure of the transfected cells to UV light resulted in an increase in MEKK1 activity, and this activation was inhibited by GST M1-1 (Fig. 1A). GST M1-1 also inhibited H 2 O 2 -induced stimulation of MEKK1 activity (Fig. 1B). In comparison, GST M1-1 did not affect the UV-induced stimulation of MLK3 (Fig.  1D) or TAK1 (data not shown). MLK3 and TAK1 are other MAP3Ks in the JNK signaling pathway (3,33).
Cdc42, a member of the small GTP-binding protein Rho family, has been shown to associate with and stimulate MEKK1 (19). The overexpression of Cdc42V12, a constitutive active mutant of Cdc42, resulted in the activation of MEKK1 . Cell lysates were also subjected to immunoblot analysis with the indicated antibodies. D, L929 cell lysates were subjected to immunoprecipitation with rabbit preimmune IgG or rabbit anti-GST M1-1 antibody. The immunoprecipitates (IP) were analyzed by immunoblotting with rabbit anti-MEKK1 antibody. Immunoblotting (IB) of cell lysates (2% total) was also shown. (Fig. 1C). GST M1-1 inhibited the Cdc42V12-induced activation of MEKK1.
MEKK1 Is a Direct Target of GST M1-1-We next examined whether GST M1-1 inhibits MEKK1 by directly acting on the kinase. In vitro kinase study revealed that MEKK1 activity was inhibited by GST M1-1 protein but not by another isoform of GST, GST P1-1 ( Fig. 2A). In comparison, GST M1-1 did not inhibit either SEK1 or JNK1 activity in vitro (Fig. 2B). Moreover, in vitro binding data indicated that GST M1-1 directly bound to MEKK1 but not to either SEK1 or JNK1 (Fig. 3A). In separate in vitro binding experiments, GST M1-1 also bound a MEKK1 fragment, MEKKK1-s, which corresponds to the amino acids 861-1493 of full-length MEKK1 (data not shown).
We next carried out co-immunoprecipitation experiments to investigate a physical interaction between GST M1-1 and MEKK1 in intact cells. HEK293 cells were co-transfected with expression vectors encoding FLAG-tagged GST M1-1 and HAtagged MEKK1. Immunoblot analysis using anti-HA antibody of the FLAG immunoprecipitates indicated that GST M1-1 physically associated with MEKK1 in the co-transfected cells (Fig. 3B). The interaction between MEKK1 and GST M1-1 was not affected by the exposure of cells to UV light (data not shown). In a separate co-immunoprecipitation experiment, GST M1-1 physically interacted with MEKK1-s (Fig. 3C). Furthermore, GST M1-1 inhibited the UV-stimulated kinase activity of MEKK1-s in transfected HEK293 cells (data not shown). We next examined an interaction between two endogenous GST M1-1 and MEKK1 proteins in L929 fibrosarcoma cells. L929 cells have been shown to express high levels of Mu class GST proteins (34). Co-immunoprecipitation data indicate that endogenous GST M1-1 physically interacted with endogenous MEKK1 in L929 cells (Fig. 3D). Taken together, our results suggest that GST M1-1 physically associates with and inhibits MEKK1.
GST M1-1 Interferes with an Interaction between MEKK1 and SEK1-Given that GST M1-1 physically interacts with MEKK1, we investigated whether GST M1-1 could disrupt the binding of MEKK1 to its substrate SEK1. HEK293 cells were transfected with expression vectors encoding HA-MEKK1 and SEK1-FLAG in the absence or the presence of the GST M1-1 construct. A co-immunoprecipitation study showed that MEKK1 physically interacted with SEK1 in the transfected Where indicated, the cells were exposed to 80 J/cm 2 UV light and incubated further for 30 min (80 J/cm 2 ). Cell lysates were subjected to immunoprecipitation with anti-HA antibody, and the resulting precipitates were examined for the kinase activity of MEKK1 by immune complex kinase assay. cells and that this interaction was inhibited by GST M1-1 (Fig.  4A). In comparison, GST M1-1 did not inhibit an interaction between ASK1 and SEK1. GST M1-1 did not affect the binding of MEKK1 to JNK1 (Fig. 4B). MEKK1 has been shown to form the scaffold complex by physically associating with SEK1 and JNK (35).
GST M1-1 Inhibits MEKK1 Activity Independently of Its GSH-conjugating Activity-GST M1-1 is an enzyme that catalyzes a conjugation of GSH to a substrate compound. The tyrosine residue in the sixth amino acid position of murine GST M1-1 is thought to be critical for the enzymatic activity (36,37). To investigate whether the inhibitory action of GST M1-1 on MEKK1 requires the catalytic activity of GST M1-1, HEK293 cells were transfected with plasmids encoding HA-MEKK1 and wild-type GST M1-1 or GST M1-1(Y6F) and examined for UVinduced MEKK1 activation. GST M1-1(Y6F) is a catalytically inactive mutant of GST M1-1. GST M1-1(Y6F) inhibited the UV-stimulated kinase activity of MEKK1 as effectively as wildtype GST M1-1 did (Fig. 5). Thus, these results suggest that the GSH-conjugating activity of GST M1-1 is dispensable for the inhibitory action of it on MEKK1.
GST M1-1 Inhibits MEKK1-mediated Apoptosis-MEKK1 has been shown to be required for JNK activation and cell death induced by microtubule-disrupting agents such as nocodazole (38). Therefore, we examined whether GST M1-1 inhibits the nocodazole-induced stimulation of MEKK1 and JNK1 activities as well as apoptotic cell death (Fig. 6). Nocodazole treatment enhanced the kinase activities of MEKK1 (Fig.  6A) and JNK1 (Fig. 6B), and these activations were suppressed in the cells expressing ectopic GST M1-1. The exposure of HEK293 cells to nocodazole also resulted in a marked increase in apoptotic cell death. The nocodazole-induced apoptosis was inhibited by MEKK1(K1253M), a dominant-negative mutant of MEKK1, but not by a dominant-negative mutant of ASK1 (ASK1(K709R)) (Fig. 6C). These data suggest that MEKK1 mediates nocodazole-induced apoptosis. The ectopic expression of GST M1-1 suppressed nocodazole-induced apoptosis. Overexpressed GST M1-1 also inhibited UV-induced apoptosis. GST catalyzes the conjugation of reduced glutathione to electrophilic xenobiotics or products of oxidative stresses. It also non-enzymatically binds a variety of compounds that include thyroid hormones, heme, bilirubin, bile acids, steroids, and fatty acids (25). Aside from being involved in drug metabolism and detoxification processes, several isoforms of GSTs appear to modulate the intracellular signaling events induced by cellular stresses. The Pi class GST directly interacts with JNK as a monomer at the non-stressed state, thereby keeping a basal JNK activity low (29). GST A2-2 inhibits the hydrogen peroxide-stimulated activities of JNK and caspase-3 as well as hydrogen peroxide-induced apoptosis in human K562 cells (26). We previously showed that GST M1-1, by directly binding ASK1, suppresses ASK1 activation and ASK1-dependent apoptosis (28). Our findings in this study demonstrate that GST M1-1 physically associates with MEKK1 in vitro and in intact cells and that GST M1-1 inhibits the stress-stimulated MEKK1 activity. In comparison, GST M1-1 does not inhibit MLK3 and TAK1, both of which are other MAP3Ks of the JNK/SAPK pathway. Collectively, GST M1-1 appears to tightly regulate the stress-activated protein kinase pathway by directly targeting two MAP3Ks, MEKK1 and ASK1. Interestingly, GST M1-1 binds the carboxyl-terminal half-region of MEKK1 that contains the catalytic domain and it inhibits the interaction between MEKK1 and SEK1. In comparison, GST M1-1 binds the amino-terminal regulatory domain of ASK1 and it inhibits ASK1 oligomerization (28). Thus, GST M1-1 appears to inhibit MEKK1 and ASK1 in distinct manners, although more studies are needed for detailed mechanisms. Regarding a role of MEKK1 in apoptosis induced by microtubule-disrupting agents, our present study shows that MEKK1 mediates nocodazole-induced apoptosis. However, it is noteworthy that targeted gene-disruption studies have shown that MEKK1 can function as either a proapoptotic or an antiapoptotic mediator in different cell types (15,38). MEKK Ϫ/Ϫ embryonic stem cells from mice are more sensitive to nocodazole-induced apoptosis, and the ectopic expression of MEKK1 protects the transfected MEKK Ϫ/Ϫ embryonic stem cells from nocodazole-induced apoptosis (15). In contrast, MEKK Ϫ/Ϫ DT40 B-cells from chicken are more resistant to nocodazole-induced apoptosis compared with wild-type DT40 cells (38). More studies are needed to better understand these opposite effects of MEKK1. Taken together, our present findings suggest that MEKK1 is a new target of GST M1-1 and that the inhibition of MEKK1 activation is an integral part of the mechanism underlying the negative regulation of stress-induced signaling by GST M1-1.