C-TAK1 Regulates Ras Signaling by Phosphorylating the ... - Cell Press

Download PDF
18 downloads 0 Views 409KB Size Report
Comment
activity of the isolated kinase-like domain (Stewart et al.,. Introduction ...... Hurley, J.H., Newton, A.C., Parker, P.J., Blumberg, P.M., and Nishi- zuka, Y. (1997).
Molecular Cell, Vol. 8, 983–993, November, 2001, Copyright 2001 by Cell Press

C-TAK1 Regulates Ras Signaling by Phosphorylating the MAPK Scaffold, KSR1 Ju¨rgen Mu¨ller,1 Ste´phane Ory,1 Terry Copeland,1 Helen Piwnica-Worms,2 and Deborah K. Morrison1,3 1 Regulation of Cell Growth Laboratory Center for Cancer Research NCI-Frederick Frederick, Maryland 21702 2 Howard Hughes Medical Institute and Department of Cell Biology and Physiology Washington University School of Medicine St. Louis, Missouri 63110

Summary Kinase suppressor of Ras (KSR) is a conserved component of the Ras pathway that interacts directly with MEK and MAPK. Here we show that KSR1 translocates from the cytoplasm to the cell surface in response to growth factor treatment and that this process is regulated by Cdc25C-associated kinase 1 (C-TAK1). C-TAK1 constitutively associates with mammalian KSR1 and phosphorylates serine 392 to confer 14-3-3 binding and cytoplasmic sequestration of KSR1 in unstimulated cells. In response to signal activation, the phosphorylation state of S392 is reduced, allowing the KSR1 complex to colocalize with activated Ras and Raf-1 at the plasma membrane, thereby facilitating the phosphorylation reactions required for the activation of MEK and MAPK. Introduction The Ras GTPase is a critical regulator of cellular proliferation and differentiation in multicellular organisms. A major route by which Ras transmits many cellular signals is through the sequential activation of the cytoplasmic serine/threonine kinases Raf-1, MEK, and MAPK (Marshall, 1996; Wittinghofer, 1998; Shields et al., 2000). Despite the fact that the primary components of this pathway, their functional role, and their epistatic relationship to one another have been characterized in considerable detail, there are still several aspects of Ras signal transmission that are not fully understood. For example, it has been well established that localization of Ras and Raf-1 to the plasma membrane is essential for pathway activation under normal conditions (Leevers et al., 1994; Magee and Marshall, 1999; Bar-Sagi, 2001), but it is unclear whether MEK and MAPK are also recruited to the membrane and if so, what regulatory mechanisms might control these events. In this respect, several genetically identified components of the Ras pathway, including kinase suppressor of Ras (KSR), might participate in such events, although their exact roles in Ras signaling have not yet been fully elucidated. Determining the function of KSR1 has been particularly problematic, due largely to the fact that the C-terminal 3

Correspondence: [email protected]

half of KSR1 contains a putative kinase-like domain. Based on this finding together with genetic epistasis experiments, it was initially predicted that KSR1 might be a protein kinase that acts upstream of, or in parallel to, Raf-1 (Therrien et al., 1995). Although some reports have suggested that mammalian KSR1 can phosphorylate Raf-1 (Zhang et al., 1997b; Xing and Kolesnick, 2001), several observations are inconsistent with the idea that KSR1 has intrinsic kinase activity. First, all mammalian KSR1 proteins lack a critical lysine residue in the catalytic domain that is normally required for the phosphotransfer reaction (Therrien et al., 1995; Mu¨ller et al., 2000). Second, mutagenesis of residues predicted to be important for kinase activity does not impair the function of C. elegans KSR (Stewart et al., 1999). Third, expression of the isolated kinase-like domain of KSR1 results in dominant inhibition of Ras signaling rather than constitutive activation of the pathway, as was initially predicted based on the model that KSR1 is a kinase required for Ras signaling (Therrien et al., 1996; Yu et al., 1997; Joneson et al., 1998). Significantly, KSR1 has been found to interact with proteins that do possess kinase activity, including Raf-1, MEK, and MAPK (Therrien et al., 1996; Denouel-Galy et al., 1997; Xing et al., 1997; Yu et al., 1997). Both MEK and MAPK associate directly with KSR1, while the interaction with Raf-1 appears to be indirect. MEK binding is mediated by the kinase-like domain and is required for both the positive function of full-length KSR1 and the dominant inhibitory activity of the isolated kinase-like domain (Stewart et al., 1999; Mu¨ller et al., 2000). Taken together, these findings suggest that KSR1 itself lacks enzymatic activity and instead serves as a docking platform for the authentic kinase components of the Ras/MAPK cascade. To further elucidate the putative MAPK scaffolding function of KSR1, we sought to identify proteins that are associated with KSR1 and might therefore regulate the biological activity of KSR1 or its associated MAPK components. Here, we report that Cdc25C-associated kinase 1 (C-TAK1) is a new member of the KSR1 complex. C-TAK1 constitutively associates with the N terminus of KSR1 and is responsible for the phosphorylation of KSR1 at serine 392 (S392), a site that mediates binding to 14-3-3 proteins. We show that KSR1 is a cytoplasmic protein that rapidly translocates to the plasma membrane in response to growth factor treatment, thereby relocalizing associated MEK proteins to the membrane and promoting MEK activation by Raf-1. Strikingly, mutation of S392 and disruption of 14-3-3 binding to this site results in the constitutive localization of KSR1 at the plasma membrane and enhanced Ras/MAPK pathway activity, suggesting that phosphorylation of S392 by C-TAK1 functions to retain KSR1 in the cytosol of quiescent cells. These findings identify C-TAK1 as a regulator of the MAPK scaffolding function of KSR1 and underscore the emerging significance of regulated intracellular scaffolds that promote the assembly of multiprotein signaling complexes.

Molecular Cell 984

Figure 1. Detection of KSR1-Associated Kinases (A) Cos cells expressing either WT KSR1 or the various KSR1 deletion mutants were left untreated (⫺) or treated for 5 min with EGF (⫹) prior to lysis. The KSR1 proteins were immunoprecipitated from the lysates using Pyo antibody, and immune complex kinase assays were performed. Labeled proteins were separated by SDS-PAGE, transferred to nitrocellulose, and visualized by autoradiography (left panel). The membrane was then probed with Pyo antibody to verify the expression of the transfected proteins (right panel) and reprobed with antibodies against MAPK and MEK (lower left panels). WT KSR1 and the deletion mutants are schematically depicted. (B) KSR1 immunoprecipitates were prepared as in (A), and immune complex kinase assays were performed with the addition of purified Raf-1. The membrane was probed with a Raf-1 antibody to verify that equal amounts of Raf-1 were present in the reactions and reprobed with an antibody recognizing phosphorylated and activated MEK (P-MEK).

Results Detection of a Protein Kinase that Phosphorylates the N-Terminal Domain of KSR1 To investigate whether other protein kinases in addition to MEK and MAPK directly associate with KSR1, we performed immune complex kinase assays using fulllength wild-type (WT) KSR1 and various KSR1 deletion mutants. KSR1 proteins were immunoprecipitated from untreated or EGF-treated Cos cells that had been lysed in a buffer containing 1% NP-40. The KSR1 immune complexes were then washed extensively with NP-40 lysis buffer that contained 1 M NaCl, and incubated with [␥-32P]ATP to detect any associated kinases that might phosphorylate proteins present in the KSR1 complex. As shown in Figure 1A, the proteins detectably phosphorylated in these assays were WT KSR1 and the truncated N-terminal proteins, N⬘539 and N⬘424. Since the N-terminal mutants lack the putative kinase domain, the phosphorylation of these proteins is presumably due to the presence of associated protein kinases. Furthermore, the phosphorylation of N⬘424 was not increased by EGF treatment, in contrast to WT KSR1 and N⬘539, suggesting that KSR1 is a substrate for both inducible and constitutive kinase activities. Interestingly, no phosphorylated proteins were observed in the samples containing the isolated KSR1 kinase-like domain (C⬘542),

consistent with other studies that have failed to demonstrate any catalytic activity for this protein (Mu¨ller et al., 2000). Next, to determine if the phosphorylation of the KSR1 proteins could be attributed to MEK or MAPK, we reprobed the KSR1 immunoprecipitates for the presence of these known KSR1-interacting kinases (Figure 1A). In agreement with previous studies (Cacace et al., 1999; Mu¨ller et al., 2000), MEK constitutively associated with WT KSR1 and the C-terminal C⬘542 protein, while MAPK interacted with WT KSR1 and the N-terminal N⬘539 protein in an EGF-inducible manner. Neither MEK nor MAPK, however, was detected in the immunoprecipitates of the N⬘424 protein, implying that this mutant is phosphorylated by a previously unidentified KSR1associated kinase. Raf-1 Phosphorylates and Activates KSR1-Associated MEK A recent study has reported that KSR1 isolated under the conditions described above is capable of phosphorylating Raf-1 on threonine 269 (Xing and Kolesnick, 2001). Previously, however, we were unable to detect KSR1-dependent phosphorylation of Raf-1 when KSR1 was isolated under more stringent conditions using lysis buffers that contain SDS (Michaud et al., 1997; Mu¨ller et al., 2000). Furthermore, the results presented in Figure

Phosphorylation and Regulation of KSR1 by C-TAK1 985

Figure 2. The N⬘424-Associated Kinase Phosphorylates KSR1 at S392 KSR1 proteins labeled in immune complex kinase assays were digested with trypsin, and the tryptic phosphopeptides were separated by HPLC. The profile of radioactivity collected in the HPLC column fractions is shown for WT KSR1, N⬘424, and N⬘424 proteins containing serine to alanine mutations at amino acid residues 297 (S297A) and 392 (S392A). The number under the Edman degredation column indicates the cycle in which 32P counts were released. The asterisk indicates the residue phosphorylated.

1A indicate that KSR1 isolated according to Xing and Kolesnick (2001) is still complexed with several protein kinases. Therefore, to determine whether KSR1 isolated under these less stringent conditions or any of the KSR1associated kinases can phosphorylate Raf-1, we added purified Raf-1 to the immune complex kinase assays. As shown in Figure 1B, no increase in Raf-1 phosphorylation was observed in any of the KSR1-containing samples. In contrast, we found that MEK bound to either WT KSR1 or the C⬘542 mutant was efficiently phosphorylated on its activating serine residues (S217/S221) by purified Raf-1. Taken together, these results do not support the model that KSR1 is a Raf-1 kinase, and instead indicate that KSR1 facilitates signal transmission between Raf-1 and MEK. Identification of the KSR1 Residues Phosphorylated by the Nⴕ424 KSR1-Associated Kinase To provide insight as to the identity of the kinase that phosphorylates the N⬘424 protein in vitro, we initiated experiments to determine the residue(s) of KSR1 phosphorylated by this kinase. KSR1 proteins labeled as described in Figure 1A were isolated from the gel matrix and digested with trypsin. The resulting tryptic phosphopeptides were then separated using reversed phase high-pressure liquid chromatography (HPLC). As depicted in Figure 2, the HPLC profiles were identical for the N⬘424 protein isolated from either untreated or EGF-

treated cells and contained one major phosphopeptide eluting in fraction 23 and one minor phosphopeptide eluting in fraction 7. These phosphopeptides were also present in the full-length WT protein; however, WT KSR1 contained additional phosphopeptides eluting in fractions 41 and 53 that were not observed in the N⬘424 profiles. The phosphopeptides unique to WT KSR1 were only detected in EGF-treated cells and appear to represent sites phosphorylated by kinases whose activity or KSR1 association is induced by EGF, such as that observed for MAPK. To precisely map the residues of the N⬘424 protein that are phosphorylated, peptides isolated in fractions 23 and 7 were subjected to phosphoamino acid analysis and N-terminal sequencing by Edman degradation (Figure 2). From this analysis, we were able to determine that the peptide contained in fraction 23 was phosphorylated on serine 392 (S392), while the peptide found in fraction 7 was phosphorylated on serine 297 (S297). To verify the identification of these sites, N⬘424 proteins containing serine to alanine mutations at positions 297 and 392 were phosphorylated in immune complex kinase assays and analyzed as described above. The resulting HPLC profiles revealed that the N⬘424 protein mutated at S392 lacked the major peptide eluting in fraction 23, while the protein mutated at S297 lacked the minor peak eluting in fraction 7 (Figure 2). Thus, S392 and, to a much lesser extent, S297 are the KSR1 residues phosphorylated by the N⬘424-associated kinase. The C-TAK1 Protein Kinase Constitutively Associates with the N Terminus of KSR1 S392 and S297 have previously been identified as in vivo sites of KSR1 phosphorylation that mediate the interaction between KSR1 and 14-3-3 dimers (Cacace et al., 1999). Known protein kinases that have been reported to phosphorylate their substrates on residues generating 14-3-3 binding sites include Akt (Zimmermann and Moelling, 1999), C-TAK1 (Peng et al., 1998), and the checkpoint kinases Chk1 and Chk2/Cds1 (Peng et al., 1997; Sanchez et al., 1997; Zeng et al., 1998). In addition, members of the AGC kinase family such as protein kinase C (PKC), protein kinase A (PKA), and p90Rsk have been found to phosphorylate substrates at sites contained within an RxxS sequence (Pearson and Kemp, 1991), which is the motif surrounding both the S392 and S297 sites. When KSR1 immunoprecipitates were probed for the presence of these kinases, C-TAK1 was detected in samples containing N⬘424 but not in those containing C⬘542. Akt, Chk1, PKC, p90Rsk, and the catalytic subunit of PKA were not observed in any of the KSR1 samples (Figure 3A and data not shown). C-TAK1 also associated with WT KSR1, but not with the N⬘320 or N⬘249 proteins, and the binding of C-TAK1 to either the WT or N⬘424 proteins was not dependent on or induced by EGF treatment (Figures 3B and 3C). The biological relevance of the interaction was indicated by the coimmunoprecipitation of endogenous C-TAK1 and KSR1 in mouse brain (Figure 3C), a tissue expressing high levels of KSR1 protein and in which endogenous complex formation among KSR1, MEK, MAPK, and 14-3-3 has been detected (Mu¨ller et al.,

Molecular Cell 986

Figure 3. C-TAK1 Associates with KSR1 (A) N⬘424 and C⬘542 proteins were immunoprecipitated from cycling Cos cells, and the immune complexes were examined by immunoblot analysis using antibodies recognizing C-TAK1, Akt, PKC, Chk1, and the Pyo epitope. Total cell lysate was included as a control to identify the position of the respective protein (arrow). The migration of the IgG heavy chain is indicated by an asterisk. (B) KSR1 deletion mutants and FLAG-epitope-tagged proteins encoding the N-terminal and C-terminal domains of Raf-1 were immunoprecipitated from Cos cells, and the immune complexes were examined by immunoblot analysis using C-TAK1, Pyo, and FLAG antibodies. (C) KSR1 proteins were immunoprecipitated from untreated (⫺) and EGF-treated (⫹) Cos cells. KSR1, Raf-1, and rabbit IgG immunoprecipitates were also prepared from mouse brain lysates. The immune complexes were then examined by immunoblot analysis using C-TAK1, KSR1, Raf-1, and Pyo antibodies. (D) N⬘424 and N⬘424 proteins containing alanine substitutions at amino acid residues 397 and 401 (N⬘424 IV/AA) were immunoprecipitated from cycling Cos cells, and immune complex kinase assays were performed. The labeled proteins were examined by autoradiography and immunoblot analysis using C-TAK1 and Pyo antibodies.

2000). In addition, the specificity of the interaction was demonstrated by the finding that C-TAK1 did not associate with endogenous or transfected Raf-1 (Figures 3B and 3C), a protein that also contains 14-3-3 binding sites. Finally, we have identified mutations in the N⬘424 protein that severely reduce C-TAK1 binding (substitution of isoleucine 397 and valine 401 to alanine), and the phosphorylation of this mutant is greatly inhibited in immune complex kinase assays (Figure 3D), strongly implicating C-TAK1 as the N⬘424-associated kinase. KSR1 Is a Substrate of Purified C-TAK1 To further establish that C-TAK1 is the N⬘424-associated kinase, we next examined whether purified C-TAK1 could phosphorylate KSR1 in vitro. For these experiments, KSR1 proteins were isolated from Cos cells using lysis

buffers that contained SDS, NP-40, and deoxycholate. These conditions dissociate KSR1 from its interacting components (Cacace et al., 1999), and the KSR1 proteins purified in this manner did not exhibit any associated kinase activity. The KSR1 proteins were then incubated with purified C-TAK1 in the presence of [␥-32P]ATP. As a control for specificity, the KSR1 proteins were also incubated with purified Chk1, Chk2, and Akt. From this analysis, we found that N⬘424 was phosphorylated only by C-TAK1 and that C⬘542 was not phosphorylated by any of the kinases tested (Figure 4A). Purified C-TAK1 also phosphorylated WT KSR1, but it did not phosphorylate the KSR1 N⬘320 or N⬘249 proteins, nor did it significantly phosphorylate purified Raf-1 (data not shown). To determine which KSR1 residue(s) are phosphorylated by C-TAK1, the N⬘424 protein labeled in this assay

Phosphorylation and Regulation of KSR1 by C-TAK1 987

for N⬘424 phosphorylated in immune complex kinase assays (Figure 2) and contained one major peptide eluting in fraction 23 and one minor peptide eluting in fraction 7. Sequencing of these peptides together with subsequent mutant analysis (data not shown) revealed that C-TAK1 phosphorylates KSR1 primarily on S392, with weak phosphorylation being observed at S297. Based on these findings and the results presented in Figure 3, we conclude that C-TAK1 is the N⬘424-associated kinase. In Vivo Regulation of KSR1 Phosphorylation at S392 To investigate whether the phosphorylation state of the KSR1 S297 or S392 sites is regulated in vivo, WT KSR1 proteins were metabolically labeled with [32P]orthophosphate, digested with trypsin, and examined by HPLC analysis (Figure 4C). Consistent with previous reports (Cacace et al., 1999; Volle et al., 1999), we found that S297 and S392 (eluting in fractions 7 and 23, respectively) were major sites of KSR1 phosphorylated in quiescent cells. When cells were treated with EGF, the phosphorylation state of S297 remained unchanged, while the phosphorylation of S392 was significantly reduced (Figure 4C). Although EGF treatment did induce the phosphorylation of a peptide eluting in fraction 21, this fraction does not contain the S392 peptide, as this peak was still observed in the HPLC profile of KSR1 containing mutations at the S297 and S392 sites. Thus, these findings indicate that S392 is dephosphorylated following growth factor treatment. In the experiments described above, the in vitro kinase activity of C-TAK1 and its association with KSR1 were not significantly affected by EGF treatment. Therefore, the dephosphorylation of S392 in vivo does not appear to be the result of C-TAK1 inactivation or dissociation and is likely to be due to the activation of an unidentified phosphatase that does not stably associate with the KSR1 complex.

Figure 4. In Vitro and In Vivo Phosphorylation of KSR1 on S392 (A) N⬘424 and C⬘542 KSR1 proteins were incubated with recombinant C-TAK1, Akt, Chk1, and Chk2 in the presence of [␥-32P]ATP. Phosphorylated KSR1 proteins were separated by SDS-PAGE, transferred to nitrocellulose membranes, and quantitated using a phosphoimager. The activities of Chk1 and Chk2 were confirmed using the CHKtide substrate, and the activity of Akt was demonstrated using the Crosstide substrate (data not shown). (B) N⬘424 phosphorylated by C-TAK1 in vitro was digested with trypsin and examined by HPLC analysis. (C) Cos cells expressing WT or S297A/S392A KSR1 were labeled in vivo with [32P]orthophosphate and either left untreated or were treated with EGF prior to cell lysis. KSR1 proteins were immunoprecipitated, digested with trypsin, and examined by HPLC analysis. The values for the cpm incorporated into the S392 site are 12,461 cpm from untreated cells and 6,017 cpm from EGF-treated cells.

was subjected to trypsin digestion and HPLC analysis. As shown in Figure 4B, the HPLC profile of N⬘424 phosphorylated by C-TAK1 was identical to that observed

Mutation of S392 Localizes the KSR1 Complex to the Plasma Membrane As 14-3-3 binding has been shown to regulate the subcellular localization of the C-TAK1 substrate Cdc25C (Kumagai and Dunphy, 1999; Peng et al., 1997, 1998; Yang et al., 1999), we next examined the role of S392 phosphorylation and, consequently, 14-3-3 binding with respect to KSR1 subcellular localization. For these studies, full-length KSR1 proteins were generated in which S392 alone or S392 and S297 were mutated to alanine residues (termed S392A and S297A/S392A mutants, respectively). The intracellular localization of WT KSR1 and the mutant KSR1 proteins was then examined in Cos cells by indirect immunofluorescence. As shown in Figure 5A, WT KSR1 was localized in the cytoplasm of quiescent cells, but rapidly translocated to the cell periphery in response to EGF treatment. Strikingly, we found that significant amounts of the S392A proteins were constitutively localized to the plasma membrane even in the absence of growth factor stimulation. The S297A/S392A mutant exhibited the same staining pattern as the S392 protein, indicating that the dissociation of 14-3-3 from the S392 site had the same effect as completely eliminating 14-3-3 binding by mutation of both the S297 and S392 sites. Similar results were ob-

Molecular Cell 988

and S392 must be disrupted to reduce 14-3-3 dimer binding to KSR1. Equivalent levels of MEK were found in all the KSR1 immunoprecipitates; however, increased levels of MAPK were observed in the S392A and S297A/ S392A samples. In EGF-treated cells, the levels of MAPK associating with the S392A and S297A/S392A mutants were higher than those observed for WT KSR1. In addition, even though MAPK was not observed in KSR1 immunoprecipitates from untreated cells, low levels of MAPK were detected in S392A and S297A/S392A samples. Therefore, while the constitutive interaction between KSR1 and MEK is not affected by mutation of the serine sites, the association of KSR1 with MAPK is enhanced, indicating that 14-3-3 binding and/or membrane localization may influence the accessibility of the FxFP MAPK binding site on KSR1.

Figure 5. Effect of S392 Mutation on KSR1 Subcellular Localization, Complex Formation, and Biological Activity (A) Cos cells expressing WT, S392A, or S297A/S392A KSR1 were left untreated or were treated with EGF. The intracellular localization of the KSR1 proteins was then determined by indirect immunofluorescence using Pyo antibody. (B) KSR1 proteins were immunoprecipitated from untreated (⫺) or EGF treated (⫹) Cos cells, and the immune complexes were examined by immunoblot analysis using Pyo, MEK, MAPK, and 14-3-3 antibodies. (C) Xenopus oocytes were injected with RNA encoding either the WT, S392A, or S297A/S392A KSR1 proteins and activated RasV12. Germinal vesicle breakdown (GVBD) was scored 5 hr following Ras injection. Oocyte lysates were prepared and examined by immunoblot analysis using Pyo and phospho-MAPK antibodies (P-MAPK).

tained when the intracellular localization of the KSR1 proteins was examined in NIH/3T3 cells (data not shown). Thus, the dephosphorylation of the S392 site correlates with the translocation of KSR1, and mutation of S392 constitutively localizes KSR1 to the plasma membrane. To further investigate the effect of S392 phosphorylation on KSR1 function, we compared the ability of the WT, S392A, and S297A/S392A proteins to interact with components of the KSR1 complex (Figure 5B). As expected, 14-3-3 was present in the immunoprecipitates of WT KSR1 and S392A, but was not detected in those of the S297A/S392A mutant, further confirming that S297 and S392 mediate the binding of KSR1 to 14-3-3 dimers. The levels of 14-3-3 bound to S392A were not significantly different than those bound to WT KSR1, and the amount of 14-3-3 bound to WT KSR1 did not change in response to EGF treatment, consistent with previous findings that the interaction with both S297

KSR1 Proteins Mutated at S392 Have Enhanced Biological Activity To measure the biological activity of the S392A and S297A/S392A proteins, we used the meiotic maturation of Xenopus oocytes as an assay system (Therrien et al., 1996). In this system, expression of WT KSR1 cooperates with activated Ras to accelerate oocyte maturation, while expression of the isolated KSR1 kinase-like domain blocks Ras signaling. When we examined the effect of the S392A or S297A/S392A proteins in this assay, we found that expression of these proteins alone was insufficient to induce oocyte maturation; however, in comparison to WT KSR1, both mutant proteins demonstrated an enhanced ability to augment Ras-dependent maturation (Figure 5C). In addition, while equivalent amounts of the WT and mutant KSR1 proteins were expressed in the oocytes, increased levels of activated MAPK were detected in oocytes coexpressing Ras and either the S392A or S297A/S392A mutant (Figure 5C). These findings suggest that by accelerating the activation of MAPK, the KSR1 proteins defective in S392 phosphorylation have an increased ability to augment Ras signaling. Colocalization of KSR1, MEK, and Activated MAPK The findings presented above are consistent with C-TAK1 acting as a negative regulator of the KSR1 scaffolding function. To further investigate this idea, we examined the intracellular localization of MEK in Cos cells expressing either WT, S392A, or S297A/S392A KSR1 proteins (Figure 6A). Consistent with the constitutive association between MEK and KSR1, MEK appeared to colocalize with KSR1 in all cells examined. MEK was found in the cytoplasm of quiescent cells expressing WT KSR1, but could be detected at the cell surface following EGF treatment. Moreover, MEK was already localized to the plasma membrane prior to growth factor addition in cells expressing either the S392A or S297A/S392A proteins. Next, we examined the intracellular localization of activated MAPK using antibodies that specifically recognize the activated, phosphorylated form of MAPK (Figure 6B). Prior to EGF treatment, no specific staining could be detected in any of the KSR1-expressing cells (data not shown), indicating that although significant amounts of the S392A and S297A/S392A proteins are localized at the plasma membrane of resting cells, growth factor

Phosphorylation and Regulation of KSR1 by C-TAK1 989

Figure 6. Colocalization of KSR1, MEK, and Activated MAPK Cos cells expressing WT, S392A, or S297A/ S392A KSR1 were left untreated or were treated with EGF. The intracellular localization of the KSR1 proteins (A and B), MEK (A), and activated MAPK (B) was determined by indirect immunofluorescence using Pyo, MEK, and phospho-MAPK (P-MAPK) antibodies.

addition is still required for activation of the MAPK cascade. In response to EGF treatment, however, activated MAPK was found to colocalize with KSR1 in membrane ruffles, supporting the model that KSR1 coordinates the assembly of a membrane-associated complex that facilitates the activation of MAPK. Discussion Phosphorylation of KSR1 by the C-TAK1 Protein Kinase To further clarify the role of KSR1 in Ras signaling, we initiated experiments to isolate unknown components of the KSR1 complex that may regulate KSR1 function and/or Ras/MAPK signaling. In these studies, we identify C-TAK1 as a new member of the KSR1 complex. C-TAK1 is a serine/threonine protein kinase that belongs to the EMK/PAR/MARK kinase family (Peng et al., 1998). Other members of this family include MARK1, a kinase that modulates microtubule structure (Drewes et al., 1997), and EMK, an enzyme that has been implicated in the regulation of cell polarity (Kemphues, 2000), microtubule dynamics (Drewes et al., 1997), Wnt pathway signaling (Sun et al., 2001), and the maintenance of immune sys-

tem homeostasis (Hurov et al., 2001). C-TAK1 was originally described as p78, a protein marker lost during pancreatic carcinogenesis (Parsa, 1988) and was subsequently cloned by virtue of its ability to associate with and phosphorylate Cdc25C (Ogg et al., 1994; Peng et al., 1998). Here, we find that C-TAK1 constitutively associates with the N-terminal region of KSR1 and that KSR1, like Cdc25C, is a substrate of C-TAK1. The primary residue of KSR1 phosphorylated by C-TAK1 is S392. This site together with S297 have been previously identified as two in vivo phosphorylation sites of KSR1 that mediate the binding to 14-3-3 dimers (Cacace et al., 1999). In our assays, we found that the phosphorylation of S392 by C-TAK1 was highly specific. When other kinases known to phosphorylate substrates on sites generating 14-3-3 binding sites were examined, only C-TAK1 was able to phosphorylate KSR1. In addition, mutations in KSR1 that severely reduce C-TAK1 binding greatly inhibit the phosphorylation of KSR1 at the S392 site in vitro. As further demonstration of specificity, we found that the association between KSR1 and C-TAK1 could be detected under physiological conditions and that C-TAK1 did not interact with or significantly phosphorylate Raf-1, a protein that also contains

Molecular Cell 990

Figure 7. Model for KSR1 Function KSR1 is localized in the cytoplasm of quiescent cells with MEK bound to its kinase-like domain and a dimer of 14-3-3 bound to the S297 and S392 phosphorylation sites. C-TAK1 constitutively associates with the N-terminal region of KSR1 to maintain the phosphorylation of S392 and, consequently, the binding of 14-3-3 to this site (the regulated site, “Reg”). In response to input signals such as growth factor treatment, the S392 site of KSR1 becomes dephosphorylated (by an unknown mechanism), perhaps exposing the KSR1 C1 domain and the FxFP MAPK binding site. As a result, the KSR1 complex translocates to the plasma membrane, colocalizing MEK with its upstream activator Raf-1 and downstream target MAPK.

14-3-3 binding motifs. While we did observe very weak phosphorylation of S297 in vitro, it is not clear whether this is an authentic target site for C-TAK1 phosphorylation. The observation that purified C-TAK1 did not phosphorylate the KSR1 N⬘320 mutant, which contains the S297 site but cannot bind C-TAK1, suggests that if C-TAK1 is bound to KSR1 it may phosphorylate S297 at low levels in vitro simply because the sequence motif surrounding S297 is similar to that of the S392 site. Regulation of Intracellular Protein Localization by C-TAK1 Prior to this study, the only known substrates of C-TAK1 were Cdc25C and the tyrosine phosphatase PTPH1. C-TAK1 phosphorylates S216 of Cdc25C (Peng et al., 1998) and S359 of PTPH1 (Zhang et al., 1997a), sites on both proteins that confer binding to 14-3-3. For Cdc25C, it is known that the phosphorylation of S216 plays an important regulatory role (Peng et al., 1997; Graves et al., 2001). In interphase, 14-3-3 binding to the S216 site helps to retain Cdc25C in the cytoplasm. At the start of mitosis, S216 is dephosphorylated, resulting in the release of 14-3-3 and the exposure of a nuclear localization sequence that enables Cdc25C to translocate to the nucleus where it dephosphorylates and activates Cdc2. The activity of C-TAK1 has been found to be constant throughout the cell cycle, and it is thought that C-TAK1 is the cytoplasmic kinase that keeps Cdc25C phosphorylated at S216 when the cell is not undergoing mitosis (Peng et al., 1998). However, in response to DNA damage, Chk1 is a nuclear kinase that becomes activated and can phosphorylate Cdc25C on Ser216, enabling Cdc25C to be exported from the nucleus (Peng et al., 1997; Sanchez et al., 1997). Regardless of the kinase that phosphorylates Cdc25C, the functional consequence of S216 phosphorylation and 14-3-3 binding is to keep Cdc25C sequestered in the cytoplasm. Based on our findings, we propose that phosphorylation of the KSR1 S392 site by C-TAK1 plays an analogous regulatory role in modulating the intracellular localization of KSR1 (depicted in Figure 7). By indirect immunofluorescence, we found that KSR1 was localized in the cytoplasm of resting cells but rapidly translocated

to the cell periphery following growth factor treatment. Moreover, we found that the changes in the subcellular localization of KSR1 correlated with changes in the in vivo phosphorylation state of the S392 site. Both S297 and S392 were highly phosphorylated in quiescent cells; however, the phosphorylation of S392 was significantly reduced following growth factor treatment. It is important to note that, although we did not detect a change in the amount of 14-3-3 bound to KSR1 in response to growth factor treatment, this observation is consistent with previous findings that 14-3-3 binds as a dimer and that the interaction with both the S297 and S392 sites must be disrupted to reduce the overall association with 14-3-3 (Cacace et al., 1999). Strikingly, we found that proteins mutated at either the S392 site alone or mutated at both the S297 and S392 sites were constitutively located at the plasma membrane. Thus, the close correlation between S392 phosphorylation and the subcellular localization of the protein strongly suggests that the S392 site is the regulated 14-3-3 binding site of KSR1. As has been shown for the 14-3-3-mediated cytoplasmic retention of Cdc25C, we propose that 14-3-3 binding to the S392 site may mask an amino acid sequence or protein domain that is involved in the intracellular targeting of KSR1. The S297 and S392 sites that mediate the binding of 14-3-3 are located on either side of the conserved cysteine-rich C1 domain of KSR1. For other proteins such as PKC and Raf-1, C1 domains have been found to interact with membrane-bound lipids/ proteins and to be required for the stable membrane association of these molecules (Hurley et al., 1997). Recently, our laboratory has solved the solution structure of the KSR1 C1 domain and has found that it contains a likely binding site for membrane-bound lipids/proteins and that it is essential for the translocation of KSR1 to the plasma membrane (M. Zhou and D.K.M., unpublished data). Therefore, the release of 14-3-3 from the S392 site may result in the exposure of the C1 domain, which in turn may mediate the membrane targeting of KSR1. KSR1 Functions as a Scaffold for the Ras/MAPK Pathway The model that KSR1 functions as a MAPK scaffold first emerged when KSR1 was found to interact with the kinase

Phosphorylation and Regulation of KSR1 by C-TAK1 991

components of the MAPK cascade. MEK has been shown to constitutively associate with KSR1, and, consistent with this observation, we found that MEK colocalized with WT KSR1 in the cytoplasm of quiescent cells and at the plasma membrane of growth factor-treated cells. Since activated Ras induces the translocation of Raf-1 to the plasma membrane, this finding suggests that a primary function of KSR1 may be to transport MEK to the intracellular location where its upstream activator Raf-1 is found. Therefore, in unstimulated cells, C-TAK1mediated phosphorylation of the KSR1 S392 site may be required to keep the KSR1/MEK complex constrained to the cytoplasm in an inactive state, a mechanism that is deactivated upon signal reception. In support of this model, MEK was constitutively localized to the plasma membrane in cells expressing the S392A and S297A/ S392A KSR1 proteins. The constitutive membrane localization of the KSR1 complex, however, appears to be insufficient to activate the MAPK cascade and instead seems to prime the pathway for accelerated MAPK activation once input signals are received. In cells expressing either the S392A or S297A/S392A KSR1 proteins, activated MAPK was detected only in membrane ruffles after growth factor treatment. In addition, the KSR1 mutants were unable to promote maturation of Xenopus oocytes on their own, but in the presence of activated Ras, they exhibited an enhanced ability to accelerate MAPK activation and oocyte maturation. Thus, while an important function of KSR1 may be to colocalize MEK with its activator Raf-1, constitutive localization of the KSR1/MEK complex at the membrane would still require other signals to induce the membrane localization and activation of Raf-1. A more rapid activation of the MAPK cascade would be expected, however, if the KSR1/MEK complex was already preassembled at the membrane. Significantly, we found that MEK bound to KSR1 could be efficiently phosphorylated and activated by Raf-1, indicating that KSR1 does not interfere with the ability of Raf-1 to phosphorylate MEK and may instead present MEK in a conformation that is conducive for phosphorylation by Raf-1. These findings further support the idea that a primary function of KSR1 is to localize MEK with activated Raf-1 at the plasma membrane and then to provide a platform for the subsequent phosphorylation required for Ras/MAPK signal transmission. This model for KSR1 function would also explain why the kinase-like domain of KSR1 acts in a dominant inhibitory manner (Joneson et al., 1998; Therrien et al., 1996; Yu et al., 1997). Since these proteins contain the MEK binding region but lack sequences required for membrane localization, they would be expected to sequester MEK in the cytosol and thereby short-circuit the activation of MAPK. Conclusion In summary, we present evidence that KSR1 functions as a regulated MAPK scaffold for the Ras pathway. Functionally equivalent scaffolding proteins have also been identified for the mammalian JNK pathway and for MAPK modules in yeast (Whitmarsh and Davis, 1998), indicating that the coordinated assembly of MAPK proteins with their activators may be an evolutionarily con-

served feature of these signaling pathways. From this study, we find that C-TAK1 plays a critical role in regulating the scaffolding function of KSR1 by phosphorylating KSR1 at a site that confers 14-3-3 binding, thus sequestering the KSR1 complex in the cytoplasm in the absence of cell signaling. In response to growth factor treatment, the phosphorylation state of the KSR1 S392 site is reduced by an as yet unidentified mechanism, leading to the translocation of KSR1 and its associated proteins to the plasma membrane. Regulated intracellular localization is an emerging theme in signal transduction and cell cycle regulation, and proteins that control this process are likely to play pivotal roles in both normal and pathological conditions. In quiescent cells, key molecules involved in signal transduction and cell cycle regulation must be kept in an inactive state. Therefore, proteins such as C-TAK1 and 14-3-3, which contribute to the maintenance of inactive protein conformations, would be expected to be important cellular regulators. The regulation of Ras/ MAPK signaling by C-TAK1 further illustrates the use of complex regulatory mechanisms in fundamental cellular processes. Our results emphasize how phosphorylation can influence the cell biological parameters of key proteins rather than simply modulating the activity of individual components of a signaling cascade. Experimental Procedures Antibodies and Reagents The phospho-MAPK, phospho-MEK, and AKT antibodies were obtained from Cell Signaling (Beverly, MA). The MEK antibody was purchased from Transduction Laboratories (Lexington, KY), the FLAG eptitope antibody was from Sigma (St. Louis, MO), and antibodies recognizing the Pyo epitope and KSR1 have been previously described (Cacace et al., 1999; Therrien et al., 1996). Antibodies directed against MAPK, Chk1, PKC, PKA, and p90Rsk were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The C-TAK1 protein and antibody were previously described in Peng et al. (1998). The Chk1, Chk2, and Akt purified kinases, as well as the CHKtide and Crosstide substrates were purchased from Upstate Biotechnology (Lake Placid, NY). Generation of DNA Constructs Constructs encoding WT-KSR1 and the KSR1 deletion mutants have been previously described (Therrien et al., 1996). The full-length WT construct encodes amino acids 1–873, the C-terminal domain construct C⬘542 starts at amino acid 542, and the N-terminal domain constructs N⬘539, N⬘424, N⬘320, and N⬘249 encode amino acids 1–539, 1–424, 1–320, and 1–249, respectively. Point mutations were introduced into KSR1 by site-directed mutagenesis (QuickChange kit; Stratagene, La Jolla, CA). All mutations were confirmed by DNA sequencing. Cell Culture, Transfection, Metabolic Labeling, and Coimmunoprecipitation Assays Cos cells were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal calf serum at 37⬚C in 5% CO2. Plasmid DNAs were transfected using Fugene reagent (Roche, Indianapolis, IN) according to the manufacturer’s instructions. Transfected cells were either incubated for 48 hr in fully supplemented medium or serum starved for 18 hr prior to growth factor addition. Cells were then lysed in NP-40 lysis buffer (20 mM Tris [pH 8.0]; 137 mM NaCl; 10% glycerol; 1% NP-40; 0.15 U/ml aprotinin; 1 mM PMSF; 20 ␮M leupeptin; 5 mM sodium vanadate) and KSR1 proteins were immunoprecipitated. After extensive washing in NP40 lysis buffer containing 1 M NaCl, the samples were examined directly by SDS-PAGE and immunoblotting or were analyzed in immune complex kinase assays. For metabolic labeling experiments,

Molecular Cell 992

transfected Cos cells were incubated for 4–6 hr at 37⬚C in phosphate-free DMEM containing 2.5% dialyzed calf serum and 1 mCi of [32P]orthophosphate (Amersham, Piscataway, NJ) per ml of labeling medium. EGF stimulation was performed as described above, and cells were washed twice with ice cold Tris-buffered saline (TBS) (137 mM NaCl; 20 mM Tris [pH 7.4]) prior to lysis for 20 min at 4⬚C in radioimmunoprecipitation assay (RIPA) buffer (NP-40 lysis buffer that contains 0.5% sodium deoxycholate and 0.1% sodium dodecyl sulfate [SDS]). Immune Complex Kinase Assays Transfected Cos cells were lysed in NP-40 lysis buffer and KSR1 proteins were immunoprecipitated. The immunoprecipitates were washed five times with NP-40 lysis buffer containing 1 M NaCl. After a final wash in 30 mM Tris (pH 7.4), the immune complexes were incubated in kinase buffer (100 mM Tris [pH 7.4]; 25 mM ␤-Glycerolphosphate; 5 mM EGTA; 1 mM DTT; 1 mM sodium vanadate; 60 ␮M ATP; 10 mM MgCl2) containing 20 ␮Ci of [␥-32P]ATP. After incubation for 30 min at 30⬚C, the assays were terminated by the addition of gel sample buffer. The samples were resolved by SDS-PAGE, the phosphoproteins were visualized by autoradiography, and the amount of 32P incorporated into the proteins was quantitated using a phosphor-imager and ImageQuant software (Molecular Dynamics, Piscataway, NJ). In experiments where KSR1 was used as an in vitro substrate, the KSR1 proteins were isolated from RIPA lysates, washed as described above, and then incubated in kinase buffer containing the appropriate purified protein kinases. Phosphorylation Site Mapping After separation by SDS-PAGE, 32P-labeled proteins were transferred to nitrocellulose membranes and visualized by autoradiography. Membrane pieces containing the phosphoproteins were excised and blocked with 1.5% PVP-40 for 1 hr at 37⬚C. The bound KSR1 proteins were then subjected to enzymatic digestion with trypsin. The resulting tryptic phosphopeptides were separated and eluted by reversed-phase HPLC as previously described (Cacace et al., 1999). HPLC fractions containing peaks of radioactivity were subjected to phosphoamino acid analysis and semiautomated Edman degradation as previously described (Morrison et al., 1993). Immunofluorescence Cos cells seeded onto 18 mm glass coverslips were transfected with the appropriate KSR1 constructs. Forty-eight hours after transfection, serum starved cells were either left untreated or were treated for 7 min with EGF. The cells were then washed once with phosphate buffered saline (PBS) and fixed in freshly prepared 4% paraformaldehyde/PBS for 10 min at 25⬚C. Following two washes with PBS, the cells were permeabilized for 5 min with 0.1% Triton X-100 in PBS. The cells were washed again with PBS and blocked for 1 hr in 3% bovine serum albumin (BSA) in PBS. Following an incubation for 1 hr at 25⬚C in the appropriate antibody, the cells were washed four times with PBS and incubated with either anti-mouse or antirabbit Alexa dye secondary antibody (Molecular Probes, Eugene, OR) diluted 1:1000 in blocking buffer for 45 min at 25⬚C, protected from light. After four more washes in PBS, the coverslips were washed in distilled water and mounted in Prolong antifade medium (Molecular Probes). RNA Transcription, Oocyte Injection, and Analysis Capped RNA was transcribed using the Message Machine kit (Ambion, Austin, TX). Buffer or RNA (30 ng) encoding the various KSR1 constructs was injected into stage VI oocytes as described (Therrien et al., 1996). After approximately 12 hr, the oocytes were injected with RasV12 RNA and were subsequently scored for GVBD, as evidenced by the appearance of a white spot at the animal pole. For biochemical analysis, oocytes were lysed (10 ␮l of RIPA buffer per oocyte) by trituration with a pipette tip. Lysates were cleared by centrifugation at 14,000 ⫻ g for 5 min at 4⬚C. Acknowledgments We thank Dan Ritt for excellent technical assistance and Monica Murakami, Marion Lohrum, Mark Fortini, and Martina Pyrski for criti-

cal reading of the manuscript. We also acknowledge Neil Michaud for valuable input in the initial phase of this project. Received July 9, 2001; revised September 18, 2001. References Bar-Sagi, D. (2001). A Ras by any other name. Mol. Cell. Biol. 21, 1441–1443. Cacace, A.M., Michaud, N.R., Therrien, M., Mathes, K., Copeland, T., Rubin, G.M., and Morrison, D.K. (1999). Identification of constitutive and ras-inducible phosphorylation sites of KSR1: implications for 14-3-3 binding, mitogen-activated protein kinase binding, and KSR1 overexpression. Mol. Cell. Biol. 19, 229–240. Denouel-Galy, A., Douville, E.M., Warne, P.H., Papin, C., Laugier, D., Calothy, G., Downward, J., and Eychene, A. (1997). Murine Ksr interacts with MEK and inhibits Ras-induced transformation. Curr. Biol. 8, 46–55. Drewes, G., Ebneth, A., Preuss, U., Mandelkow, E.M., and Mandelkow, E. (1997). MARK, a novel family of protein kinases that phophorylate mictotuble-associated proteins and trigger microtubule disruption. Cell 89, 297–308. Graves, P.R., Lovly, C.M., Uy, G.L., and Piwnica-Worms, H. (2001). Localization of human Cdc25C is regulated both by nuclear export and 14-3-3 protein binding. Oncogene 20, 1839–1851. Hurley, J.H., Newton, A.C., Parker, P.J., Blumberg, P.M., and Nishizuka, Y. (1997). Taxonomy and function of C1 protein kinase C homology domains. Protein Sci. 6, 477–480. Hurov, J.B., Stappenbeck, T.S., Zmasek, C.M., Ranganath, S.H., White, L.S., Russell, J.H., Chan, A.C., Murphy, K.M., and PiwnicaWorms, H. (2001). Immune cell dysfunction and autoimmune disease in mice lacking the EMK/Par1 protein kinase. Mol. Cell. Biol. 21, 3206–3219. Joneson, T., Fulton, J.A., Volle, D.J., Chaika, O.V., Bar-Sagi, D., and Lewis, R.E. (1998). Kinase suppressor of Ras inhibits the activation of extracellular ligand-regulated (ERK) mitogen-activated protein (MAP) kinase by growth factors, activated Ras, and Ras effectors. J. Biol. Chem. 273, 7743–7748. Leevers, S.J., Paterson, H.F., and Marshall, C.J. (1994). Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature 369, 411–414. Kemphues, K. (2000). PARsing embryonic polarity. Cell 101, 345–348. Kumagai, A., and Dunphy, W.G. (1999). Binding of 14-3-3 proteins and nuclear export control the intracellular localization of the mitotic inducer Cdc25. Genes Dev. 13, 1067–1072. Magee, T., and Marshall, C. (1999). New insights into the interaction of Ras with the plasma membrane. Cell 98, 9–12. Marshall, C.J. (1996). Ras effectors. Curr. Opin. Cell Biol. 2, 197–204. Michaud, N.R., Therrien, M., Cacace, A., Edsall, L.C., Spiegel, S., Rubin, G.M., and Morrison, D.K. (1997). KSR stimulates Raf-1 activity in a kinase-independent manner. Proc. Natl. Acad. Sci. USA 94, 12792–12796. Morrison, D.K., Heidecker, G., Rapp, U.R., and Copeland, T.D. (1993). Identification of the major phosphorylation sites of the Raf-1 kinase. J. Biol. Chem. 268, 17309–17316. Mu¨ller, J., Cacace, A.M., Lyons, W.E., McGill, C.B., and Morrison, D.K. (2000). Identification of B-KSR1, a novel brain-specific isoform of KSR that functions in neuronal signaling. Mol. Cell. Biol. 20, 5529– 5539. Ogg, S., Gabrielli, B., Piwnica-Worms, H., Peng, C.Y., Graves, P.R., Thoma, R.S., Byrnes, M.J., Wu, Z., and Stephenson, M.T. (1994). Purification of a serine kinase that associates with and phosphorylates human Cdc25C on serine 216. J. Biol. Chem. 269, 30461–30469. Parsa, I. (1988). Loss of a Mr 78,000 marker in chemically induced transplantable carcinomas and primary carcinoma of human pancreas. Cancer Res. 48, 2265–2272. Pearson, R., and Kemp, B. (1991). Protein kinase phosphorylation

Phosphorylation and Regulation of KSR1 by C-TAK1 993

site sequences and consensus specificity motifs: tabulations. Methods Enzymol. 200, 62–81. Peng, C.-Y., Graves, P.R., Thomas, R.S., Wu, Z., Shaw, A.S., and Piwnica-Worms, H. (1997). Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25c on serine 216. Science 277, 1501–1505. Peng, C.Y., Graves, P.R., Ogg, S., Thoma, R.S., Byrnes, M.J., Wu, Z., Stephenson, M.T., and Piwnica-Worms, H. (1998). C-TAK1 protein kinase phosphorylates human Cdc25C on serine 216 and promotes 14-3-3 protein binding. Cell Growth Differ. 9, 197–208. Sanchez, Y., Wong, C., Thoma, R., Richman, R., Wu, Z., PiwnicaWorms, H., and Elledge, S. (1997). Conservation of the Chk1 checkpoint pathway in mammals: linkage of DNA damage to Cdk regulation through Cdc25. Science 277, 1497–1501. Shields, J., Pruitt, K., McFall, A., Shaub, A., and Der, C. (2000). Understanding Ras: ‘it ain’t over ‘til it’s over’. Trends Cell Biol. 10, 147–154. Stewart, S., Sundaram, M., Zhang, Y., Lee, J., Han, M., and Guan, K.-L. (1999). Kinase suppressor of Ras forms a multiprotein signaling complex and modulates MEK localization. Mol. Cell. Biol. 19, 5523– 5534. Sun, T.Q., Lu, B., Feng, J.J., Reinhard, C., Jan, Y.N., Fantl, W.J., and Williams, L.T. (2001). PAR-1 is a dishevelled-associated kinase and a positive regulator of Wnt signalling. Nat. Cell Biol. 3, 628–636. Therrien, M., Chang, H.C., Solomon, N.M., Karim, F.D., Wassarman, D.A., and Rubin, G.M. (1995). KSR, a novel protein kinase required for RAS signal transduction. Cell 83, 879–888. Therrien, M., Michaud, N.R., Rubin, G.M., and Morrison, D.K. (1996). KSR modulates signal propagation within the MAPK cascade. Genes Dev. 10, 2684–2695. Volle, D.J., Fulton, J.A., Chaika, O.V., McDermott, K., Huang, H., Steinke, L.A., and Lewis, R.E. (1999). Phosphorylation of the kinase suppressor of ras by associated kinases. Biochemistry 38, 5130– 5137. Whitmarsh, A., and Davis, R. (1998). Structural organization of MAPkinase signaling modules by scaffold protein in yeast and mammals. Trends Biochem. Sci. 23, 481–485. Wittinghofer, A. (1998). Signal transduction via Ras. Biol. Chem. 379, 933–937. Xing, H., Kornfeld, K., and Muslin, A.J. (1997). The protein kinase KSR interacts with 14-3-3 protein and Raf. Curr. Biol. 7, 294–300. Xing, H.R., and Kolesnick, R. (2001). Kinase suppressor of Ras signals through Thr269 of c-Raf-1. J. Biol. Chem. 276, 9733–9741. Yang, J., Winkler, K., Yoshida, M., and Kornbluth, S. (1999). Maintenance of G2 arrest in the Xenopus oocyte: a role for 14-3-3-mediated inhibition of Cdc25 nuclear import. EMBO J. 18, 2174–2183. Yu, W., Fantl, W.J., Harrowe, G., and Williams, L.T. (1997). Regulation of the MAP kinase pathway by mammalian Ksr through direct interaction with MEK and ERK. Curr. Biol. 8, 56–64. Zeng, Y., Forbes, K.C., Wu, Z., Moreno, S., Piwnica-Worms, H., and Enoch, T. (1998). Replication checkpoint requires phosphorylation of the phosphatase Cdc25 by Cds1 or Chk1. Nature 395, 507–510. Zhang, S.H., Kobayashi, R., Graves, P.R., Piwnica-Worms, H., and Tonks, N.K. (1997a). Serine phosphorylation-dependent association of the band 4.1-related protein-tyrosine phosphatase PTPH1 with 14-3-3beta protein. J. Biol. Chem. 272, 27281–27287. Zhang, Y., Yao, B., Delikat, S., Bayoumy, S., Lin, X.-H., Basu, S., McGinley, M., Chan-Hui, P.-Y., Lichenstein, H., and Kolesnick, R. (1997b). Kinase suppressor of Ras is ceramide-activated protein kinase. Cell 89, 63–72. Zimmermann, S., and Moelling, K. (1999). Phosphorylation and regulation of Raf by Akt (protein kinase B). Science 286, 1741–1744.