HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 32, 24421-24428, August 11, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/32/24421    most recent M003148200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sun, W. Articles by Johnson, G. L. Search for Related Content PubMed PubMed Citation Articles by Sun, W. Articles by Johnson, G. L. Social Bookmarking                      What's this?

MEK Kinase 2 Binds and Activates Protein Kinase C-related Kinase 2

BIFURCATION OF KINASE REGULATORY PATHWAYS AT THE LEVEL OF AN MAPK KINASE KINASE^*

Weiyong Sun, Sylvie Vincent^§, Jeffrey Settleman^§, and Gary L. Johnson^¶

From the  Department of Pharmacology, University of Colorado Health Sciences Center and University of Colorado Cancer Center, Denver, Colorado 80262 and the ^§ Massachusetts General Hospital Cancer Center and Harvard Medical School, Charlestown, Massachusetts 02129

Received for publication, April 12, 2000, and in revised form, May 17, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

MEK kinase 2 (MEKK2) is a 70-kDa protein serine/threonine kinase that has been shown to function as a mitogen-activated protein kinase (MAPK) kinase kinase. MEKK2 has its kinase domain in the COOH-terminal moiety of the protein. The NH₂-terminal moiety of MEKK2 has no signature motif that would suggest a defined regulatory function. Yeast two-hybrid screening was performed to identify proteins that bind MEKK2. Protein kinase C-related kinase 2 (PRK2) was found to bind MEKK2; PRK2 has been previously shown to bind RhoA and the Src homology 3 domain of Nck. PRK2 did not bind MEKK3, which is closely related to MEKK2. The MEKK2 binding site maps to amino acids 637-660 in PRK2, which is distinct from the binding sites for RhoA and Nck. This sequence is divergent in the closely related kinase PRK1, which did not bind MEKK2. In cells, MEKK2 and PRK2 are co-immunoprecipitated and PRK2 is activated by MEKK2. Similarly, purified recombinant MEKK2 activated PRK2 in vitro. MEKK2 activation of PRK2 is independent of MEKK2 regulation of the c-Jun NH₂-terminal kinase pathway. MEKK2 activation of PRK2 results in a bifurcation of signaling for the dual control of MAPK pathways and PRK2 regulated responses.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mitogen-activated protein kinases (MAPKs)¹ are components of a three kinase module that also includes a MAPK kinase and MAPK kinase kinase (1). MEKK2 is a MAPK kinase kinase that we have shown is activated in response to several extracellular stimuli including antigen receptors in T cells and mast cells and growth factors such as epidermal growth factor and Kit ligand (stem cell factor) (2, 3). We recently demonstrated that MEKK2 translocates to the cytoplasmic face of the contact site of T cells interacting with an antigen loaded presenting cell (3). MEKK2 but not MEKK1 or MEKK3 is translocated to the T cell interface with the antigen presenting cell. MEKK2 is translocated and activated within seconds of exposure of T cells to antigen presentation. MEKK2 activation was required for its translocation because kinase-inactive MEKK2 was not recruited to the contact between the T cell and antigen presenting cell. In this system, MEKK2 signaling was found to be required for maintenance of the conjugate formation between T cell and antigen presenting cell. Although poorly defined, the activation of MEKK2 seems to involve more than one pathway. For example antigen receptor activation of MEKK2 is inhibited by wortmannin, indicating the activation of phosphatidylinositol 3-kinase is required. In contrast, epidermal growth factor stimulation of MEKK2 in COS-7 cells is insensitive to wortmannin.

MEKK2 is a 70-kDa serine-threonine kinase that has its kinase domain in the COOH-terminal half of the protein (4). Analysis of the NH₂-terminal moiety of MEKK2 does not reveal an identifiable motif that has been defined in other proteins that are known to regulate protein-protein (i.e. SH3, proline-rich, etc.) or protein-lipid (i.e. pleckstrin homology domains) interactions. Thus, the sequence of MEKK2 does not readily allow predictions of its regulation and interactions with other molecules in the cell. In an attempt to define the regulation of MEKK2 in antigen and growth factor responses, we have performed two-hybrid analysis to identify proteins that bind MEKK2. Several binding partners were identified in this screen. As we detail in this report, one binding partner that was identified in this screen was the protein kinase C-related kinase 2 (PRK2).

Protein kinase C (PKC)-related kinases (PRKs) constitute a subclass of lipid and proteolysis-activated serine/threonine kinases that are highly homologous to PKCs in their catalytic domains (5-7). Human PRK1 (also known as PKN, for protein kinase N) and PRK2 share structurally very similar kinase domains (87% identity), but their regulatory NH₂ termini are less conserved (48% identity) (6). PKN and PRK2 have been demonstrated to be an effector of the small GTPase, Rho (8-12). Binding of Rho·GTP activates the kinase activity of PKN and PRK2 (12). PRK2 may also bind another small GTPase, Rac, bound to GTP (12). Rho and Rac are involved in the regulation of cytoskeletal organization as well as many other cellular processes including membrane trafficking, activation of JNK and p38 MAPK pathways, transcription, cell growth, and development (13-16). Consistent with the finding that PKN and PRK2 are effectors for Rho is the observations that PKN and PRK2 can enhance or mediate changes in the actin cytoskeleton and gene transcription (9, 12, 17-20). RhoB has also been reported to mediate PKN association with endosomes (21).

In addition to its binding to Rho and Rac, PRK2 binds the middle SH3 domain of the SH2-SH3 adaptor protein, Nck (9). PRK2 is, therefore, predicted to be recruited to tyrosine-phosphorylated proteins that bind the Nck SH2 domain. Nck could also bind to proteins having the proline-directed SH3 binding motif for the first SH3 domain of Nck. Thus, PRK2 activation may coordinate receptor protein-tyrosine kinase signaling with Rho-activated pathways (9, 22). Consistent with this hypothesis is reports that Nck also binds the Wiskott-Aldrich syndrome protein (WASP) and the p21-activated protein kinase (PAK) 1 and 3 (9, 23). WASP and PAK are effectors for Rac and Cdc42 (24, 25). Both WASP and PAKs are involved in regulating the cytoskeleton. This suggests that, like PRK2, Nck could localize Rho-related GTP-binding proteins in the cell for control of the actin cytoskeleton and tyrosine kinase signaling.

Finally, a recent report indicated that the COOH-terminal 77 residues of PRK2 (termed the PDK1 interacting fragment) bind the kinase domain of the 3-phosphoinositide-dependent protein kinase-1 (PDK1) (26). Thus, PRK2 appears to participate in a diverse set of inter-related signaling programs, suggesting it may have a scaffolding function to organize specific signal transduction responses in the cell. Consistent with this prediction is the observation that PRK2 was rapidly cleaved by caspase 3-like proteases during Fas- and staurosporine-induced apoptosis (27, 28). This cleavage event resulted in increased kinase activity but also released the kinase domain from tethering to Rho and Nck, which would effectively disrupt the signaling complex.

In the present study we show that MEKK2 and PRK2 are binding partners. MEKK2 interaction activates PRK2 kinase activity.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Antibodies-- Mouse monoclonal anti-HA (12CA5), anti-Flag (M5), and anti-V5 antibodies were purchased from Roche Molecular Biochemicals, Sigma, and Invitrogen, respectively. Rabbit polyclonal anti-MEKK2 antibody has been described elsewhere (2).

Yeast Two-hybrid Screening of cDNA Library and Interaction Analysis-- Full-length mouse MEKK2 was fused in-frame to the C terminus of the bacterial DNA-binding protein LexA in vector BTM116 (29). It was used to transform the yeast reporter strain L40 (30) together with mouse T-cell lymphoma library cDNA (CLONTECH) cloned COOH-terminal to the activation domain of Gal4 (GAD) in plasmid pACT (31). A total of 2.4 × 10⁶ transformants were plated on synthetic complete plates lacking trytophan, leucine, histidine, lysine, and uracil (SCHis) but supplemented with 10 mM 3-aminotriazole (3-AT). After 3.5 days of incubation at 30 °C, 57 of the fastest growing clones were picked and streaked on SCHis +15 mM 3-AT media. The largest colony from each of the original 57 clones was then streaked on SC+His plates and tested for -galactosidase (-gal) production using a filter-lift assay (32). Yeast total DNAs were isolated from LacZ⁺ clones, and the library plasmids were rescued into Escherichia coli strain HB101 (Promega). After re-transformation into L40 with pLexA-MEKK2, 24 clones were confirmed to be His⁺ LacZ⁺.

For two-hybrid interaction analysis, L40 cells were spread directly on SCHis + 3-AT media after transformation, or streaked on SCHis + 3-AT plates after first growth on plates with histidine. A second two-hybrid system was also used in the study; in this case yeast CG1945 (CLONTECH) was used as the host strain, and peptides were fused to the Gal4 DNA binding domain (GBD) in vector pAS2-1 (CLONTECH) as "baits." In both systems, "prey" peptides were fused either to the GAD in plasmid pACT (31) or its derivative pACT2 (33), or to the activation domain of VP16 (VAD) in plasmid pVP16 (30). Quantitation of -gal activity was assayed on liquid cultures using o-nitrophenyl--D-galactopyranoside (ONPG) as substrate (34).

Mammalian Cell Transfection and Co-immunoprecipitation-- HEK293 cells were grown to 50-80% confluence in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum and penicillin-streptomycin, and after washing with DMEM were transfected with various combinations of expression plasmids in the presence of LipofectAMINE (Life Technologies, Inc.). Transfected cells were grown in DMEM + 10% fetal bovine serum for 36-48 h and harvested in lysis buffer I (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 2 mM MgCl₂, 1 mM Na₃VO₄, 0.5% Triton X-100, 2 mM phenylmethylsulfonyl fluoride, 10 µg/ml pepstatin A, 50 µM leupeptin, 2 µg/ml aprotinin, and 1 mM dithiothreitol). Lysates were cleared by brief centrifugation, and 400-800 µg of lysates were incubated with M5 or 12CA5 antibody (as indicated in the figure legends) in a total volume of 400 µl for 2 h at 4 °C with rocking. Thirty µl of 1:1 rec-protein G-Sepharose 4B slurry (Zymed Laboratories Inc.) was added to the mixture and the incubation continued for 1 h. Beads were washed three times with 400 µl of lysis buffer, heat-denatured in 1× SDS-PAGE loading buffer, and resolved on a 10% SDS-PAGE gel. Proteins were transferred to a Protran nitrocellulose membrane (Schleicher & Schuell) and immunoblotted using horseradish peroxidase-coupled goat-anti-mouse secondary antibody and enhanced chemiluminescence (NEN Life Science Products).

In Vitro Binding Assays-- For analyzing binding of PRK2/PKN to endogenous MEKK2, 10 µg of bacterially expressed and purified glutathione S-transferase (GST) fusion proteins pre-bound to glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) were incubated with 400 µg of HEK293 cell lysates at 4 °C for 4 or 10 h with gentle rocking. After three washes, beads were subjected to SDS-PAGE and Western blotting with anti-MEKK2 antibody and horseradish peroxidase-conjugated protein A.

Binding of recombinant baculovirus-expressed MEKK2 with transfected PRK2 was performed by using 800 µg of lysate from HEK293 cells that were transfected with either pFlag-hPRK2 or empty vector. One µg of purified MEKK2 was added to the lysate for 2 h or overnight, followed by incubation with M5 antibody and protein G-Sepharose 4B beads for another 2 h. Binding of MEKK2 was detected as described above.

In Vitro Kinase Assays-- HEK293 cells were transfected with Flag-hPRK2, kinase-inactive Flag-hPRK2KE, or empty vector. Cells were lysed in a higher-strength lysis buffer II (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 50 mM NaF, 5 mM Na₄P₂O₇, 1 mM Na₃VO₄, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 2 mM phenylmethylsulfonyl fluoride, 10 µg/ml pepstatin A, 50 µM leupeptin, 2 µg/ml aprotinin, and 1 mM dithiothreitol), and lysates (1500 µg for PRK2KE- and vector-transfected cells) containing equal amounts of PRK2 or PRK2KE were precipitated with M5 antibody and rec-protein G-Sepharose 4B beads. Beads were washed twice with lysis buffer II and twice with kinase buffer (20 mM HEPES, pH 7.5, 50 mM NaCl, 5 mM MgCl₂, and 5 mM MnCl₂) and suspended with 80 µl of kinase buffer. Twenty µl of suspension was used in a kinase reaction with 0.5 µg of purified MEKK2 and [-³²P]ATP (90 nM, 4500 Ci/mmol) in a total of 50 µl; in some experiments bovine serum albumin was supplied in reactions without MEKK2 to equalize the amounts of total protein. Heat inactivation of MEKK2 was achieved by heating at 100 °C for 15 min and chilling on ice. Kinase reactions were incubated at 30 °C for 20 min and terminated by adding SDS-PAGE loading buffer.

For analysis using myelin basic protein (MBP) as substrate, equal amounts of epitope-tagged proteins were immunoprecipitated from transfected HEK293 cells (for cells transfected with empty vector, kinase-inactive PRK2, or MEKK2, 800 µg of lysate was used). Beads were washed and incubated with 20 µg of MBP (Upstate Biotechnology) in a 50-µl kinase reaction.

For analysis of endogenous JNK kinase activity, 200 µg of transfected HEK293 cell lysates were incubated with 20 µg of bacterially expressed and purified GST-c-Jun_1-79 or GST protein pre-bound to beads in 400 µl of lysis buffer I at 4 °C for 2 h with rocking. Fifty µl of kinase buffer containing [-³²P]ATP was then added to the washed beads, and the kinase reaction was performed as described.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

MEKK2 Interacts with PRK2-- In an effort to identify binding partners of MEKK2, the yeast two-hybrid system (35) was used to screen a mouse T-cell lymphoma cDNA library. Full-length MEKK2 was fused to the bacterial DNA-binding protein LexA and used as "bait" to transform the yeast reporter strain L40 (30) along with the library cDNA fused to the Gal4 activation domain (GAD) (36). A total of 2.4 × 10⁶ transformants were plated and 24 clones were strong positives for both His3 and LacZ reporter constructs. Among them, two clones were identical isolates encoding the sequence corresponding to residues 479-670 of human PRK2 (hPRK2). An L40 strain expressing both LexA-MEKK2 and GAD-mPRK2(aa 479-670) was prototrophic for histidine (Fig. 1A) and exhibited a -gal activity at least 20-fold higher than the control transformants (Fig. 1B). To further demonstrate specific interaction between MEKK2 and PRK2, the bait and prey constructs were switched and their binding was tested in a different two-hybrid system. The mPRK2(aa 479-670) was fused to the GBD and MEKK2 was fused to VAD (30). Yeast strain CG1945 (from CLONTECH) transformed with GBD-mPRK2(aa 479-670) and VAD-MEKK2 could grow into colonies on synthetic complete plates lacking histidine (SCHis) and supplemented with 15 mM 3-AT. In contrast, cells transformed with control plasmids plus either GBD-mPRK2(aa 479-670) or VAD-MEKK2 were not able to grow even at 1 mM 3-AT (data not shown). We also tested binding of MEKK2 to full-length PRK2 using the two-hybrid analysis using the full-length hPRK2. MEKK2 and hPRK2 exhibited a weaker but still positive interaction in the histidine prototrophy assay compared with PRK2(aa 479-670) (data not shown). The weaker interaction of the full-length protein relative to the binding domain for the partner protein is a common finding in yeast two-hybrid analysis. The findings suggested that PRK2 is capable of binding MEKK2.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   MEKK2 interacts with PRK2. A and B, MEKK2 binds PRK2 (residues 479-670) in the yeast two-hybrid system. A, yeast L40 cells were transformed with the indicated combinations of plasmids and plated on SC+His medium. Transformant colonies were then streaked on SCHis + 15 mM 3-AT plates and incubated at 30 °C for 2.5 days. B, liquid cultures of yeast transformant cells were analyzed for -gal activity using ONPG as substrate. Results were calculated in Miller units and represented as average ± standard deviation (S.D.). C and D, MEKK2 associates with full-length PRK2 in mammalian cells. C, HEK293 cells were transfected with HA-MEKK2 plus either Flag-hPRK2 or empty pCMV5 vector. After lysis cell lysates were immunoprecipitated with anti-Flag M5 antibody and blotted with anti-HA 12CA5 antibody. D, reciprocally, cells were transfected with Flag-hPRK2 plus HA-MEKK2 or pCMV5. Cell lysates were precipitated with 12CA5 and blotted with M5 antibody.

To substantiate MEKK2-PRK2 interaction in mammalian cells, HEK293 cells were transfected with HA-epitope-tagged MEKK2, together with Flag-tagged full-length hPRK2 or control empty vector plasmid. Cell lysates were immunoprecipitated with anti-Flag antibody and Western-blotted with anti-HA antibody after separation by SDS-PAGE. Fig. 1C shows that MEKK2 was co-immunoprecipitated with PRK2 but HA-MEKK2 was not detected in immunoprecipitates from control transfections. Reciprocally, association of PRK2 with MEKK2 was readily detected when cell lysates were precipitated with anti-HA antibody to immunoprecipitate MEKK2 and subsequently immunoblotted with the anti-FLAG antibody to detect PRK2 (Fig. 1D). These findings demonstrate that MEKK2 and PRK2 associate with each other by two independent experimental techniques: yeast two-hybrid analysis and co-immunoprecipitation.

Refinement of the PRK2 Binding Region for MEKK2-- The PRK2 region required for MEKK2 binding was mapped using the yeast two-hybrid method. Serial truncations of the mPRK2(aa 479-670) fragment were fused to the activation domain of Gal4 and tested for their ability to retain binding of MEKK2 in the yeast L40 strain (Fig. 2A). This approach identified a region of 55 amino acids (corresponding to residues 616-670 of hPRK2) that showed binding for MEKK2, as measured by growth of transformants, comparable to the 479-670 fragment of PRK2.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Mapping the MEKK2-PRK2 binding sites. A, various truncations of the mouse PRK2 (amino acids 479-670) were fused to the activation domain of Gal4 (GAD) in plasmid pACT or its derivative pACT2 and transformed into yeast L40 together with LexA-MEKK2 cloned in pBTM116. Interaction was determined by the ability of the transformant cells to grow on SCHis + 15 mM 3-AT. B, different fragments of MEKK2 was fused to LexA in pBTM116 and transformed along with GAD-mPRK2(aa 479-670) into L40. Binding was examined by the histidine prototrophy assay.

A similar approach was taken in an attempt to define the region of MEKK2 capable of binding mPRK2(aa 479-670) (Fig. 2B). However, neither the N-terminal regulatory region nor the COOH-terminal kinase domain of MEKK2 alone showed significant interaction with PRK2. One interpretation of these findings is that PRK2 may interact with more than one site in MEKK2, requiring both its NH₂ and COOH termini. Alternatively, the MEKK2 NH₂-terminal sequences expressed may not fold properly in the absence of the COOH-terminal kinase domain. Attempts at expressing NH₂-terminal MEKK2 and MEKK3 sequences in E. coli have proven extremely difficult with extreme sensitivity to proteases (data not shown). This finding is consistent with the notion that the MEKK2 NH₂-terminal constructs do not fold properly and therefore are not a reliable reagent for mapping the MEKK2 interaction domain for PRK2. Most likely, a more detailed mutagenesis strategy of full-length MEKK2 will probably have to be undertaken to map the PRK2 interaction sequences.

Specificity of the MEKK2-PRK2 Interaction-- MEKK2 and MEKK3 are closely related to each other. The kinase domains of MEKK2 and MEKK3 are 96% conserved in amino acid sequence. The NH₂-terminal moieties of MEKK2 and MEKK3 are approximately 55% conserved in primary sequence. Therefore, we tested the possible binding of MEKK3 to the PRK2(aa 479-670) sequence that binds MEKK2 using the yeast two-hybrid system. Yeast L40 cells were transformed with LexA-MEKK3 plus GAD-mPRK2(aa 479-670) and plated on SC+His media. Transformant colonies were then streaked on SCHis + 3 mM 3-AT plates to test for the transactivation of the His3 reporter gene (Fig. 3A), or grown in liquid media and analyzed for LacZ expression (Fig. 3B). As shown, L40 cells carrying plasmids LexA-MEKK3 and GAD-mPRK2(aa 479-670) were not able to grow on SCHis + 3 mM 3-AT plates, even after incubation for over 1 week (data not shown). In comparison, cells with LexA-MEKK3 plus VAD-14-3-3 showed robust growth on the same minimal plates and strong -gal activity, consistent with our previous report that MEKK3 binds 14-3-3 protein (37). This indicates that LexA-MEKK3 was expressed and nucleus-localized, and the failure of GAD-mPRK2(aa 479-670) to transactivate the reporter genes was due to its inability to bind MEKK3. In support of this conclusion, in a different two-hybrid system, we found that yeast CG1945 cells transformed with GBD-mPRK2(aa 479-670) plus VAD-MEKK3 were negative for histidine prototrophy (data not shown); in contrast, as mentioned above, GBD-mPRK2(aa 479-670) bound VAD-MEKK2 in CG1945 cells. These findings indicated that the PRK2 interaction was specific for MEKK2 because a closely related kinase, namely MEKK3, did not bind PRK2(aa 479-670). The results also argue that the binding of PRK2(aa 479-670) requires specific sequences in the NH₂ terminus of MEKK2 that are not conserved in MEKK3 because the kinase domains are nearly identical.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   MEKK3 does not bind PRK2. A and B, interaction of MEKK3 with PRK2 was examined in the yeast two-hybrid system. A, yeast L40 was transformed with the indicated combinations of plasmids and grown on SC+His medium. Yeast colonies were then streaked on SCHis + 3 mM 3-AT plates and incubated at 30 °C for 3 days. B, -gal assay of liquid cultures using ONPG as substrate. C, MEKK3 does not bind PRK2 in mammalian cells. Top panel, HEK293 cells were transfected with the indicated expression plasmids. Cell lysates were then immunoprecipitated with anti-Flag M5 antibody and blotted with anti-V5 antibody. Middle panel, 20 µg lysates were blotted with anti-V5 antibody. Bottom panel, 20 µg of lysates were blotted with M5 antibody.

To experimentally confirm that MEKK2 but not MEKK3 binds PRK2 in mammalian cells, we transfected HEK293 cells with Flag-hPRK2 and either COOH-terminally V5-epitope-tagged MEKK3 (MEKK3-V5) or MEKK2-V5. Cell lysates were immunoprecipitated with anti-Flag antibody and immunoblotted with anti-V5 antibody. As predicted, MEKK2-V5 co-immunoprecipitated with Flag-hPRK2 but MEKK3-V5 did not (Fig. 3C). MEKK3-V5 was expressed at similar protein levels as MEKK2-V5 and Flag-hPRK2 was expressed at similar levels in the presence of either MEKK2-V5 or MEKK3-V5 (Fig. 3C). Therefore, MEKK2 but not MEKK3 binds PRK2.

Obviously, it was also important to determine if MEKK2 could bind PKN (PRK1), a kinase sharing 87% homology in the kinase domain and 48% homology in the NH₂-terminal regulatory domain to PRK2. Thus, a fragment of human PKN (residues 477-628) that is homologous to PRK2(aa 479-670) was fused to the activation domain of Gal4 and tested for its ability to interact with MEKK2 in yeast L40. By the criteria of histidine prototrophy and -gal activity, no significant binding could be demonstrated. The corresponding regions from the human and the mouse PRK2 protein interacted strongly with MEKK2 (Fig. 4, A and B). The yeast two-hybrid data indicate MEKK2 binds PRK2 but not PKN.

View larger version (62K): [in this window] [in a new window]   Fig. 4.   PKN does not bind MEKK2. A and B, yeast two-hybrid analyses of PKN-MEKK2 interaction. A, the region of hPKN (residues 477-628) that is homologous to PRK2(aa 479-670) was tested for its interaction with MEKK2 in yeast L40. After growth on SC+His medium, cells were streaked on SCHis + 15 mM 3-AT plates and incubated at 30 °C for 3 days. B, -gal activities of liquid cultures using ONPG as substrate. C, PKN does not bind MEKK2 in vitro. Human PRK2 (residues 616-670) and the corresponding region from hPKN were expressed as GST fusions in bacteria and purified by binding to glutathione-Sepharose beads. Beads were incubated with HEK293 cell lysates, and binding of endogenous MEKK2 was detected by Western blotting with anti-MEKK2 antibody. D, alignment of the MEKK2-binding region of hPRK2 (residues 616-670) with the mouse sequence and the related region from human PKN. Residues identical to hPRK2 are highlighted in black, and conservative substitutions (to which the BLOSUM-62 matrix (Ref. 55) gives a positive score) are shaded in gray. The start of the kinase domains (7) is indicated by an arrow.

To gain direct evidence for the selectivity of MEKK2-PRK2 interaction, an in vitro binding assay was employed. Initially, we generated GST fusion constructs of hPRK2(aa 479-670) and hPKN(aa 477-628). Unfortunately, despite a variety of expression conditions we tried, neither protein was expressed in bacteria. As shown in Fig. 2A, we had mapped the binding site for MEKK2 to a 55-residue region (aa 616-670) of mPRK2. Therefore, we decided to fuse GST to hPRK2(aa 616-670) and the corresponding region (aa 585-628) of hPKN. Purified GST, GST-hPKN(aa 585-628), or GST-hPRK2(aa 616-670), pre-bound to glutathione-Sepharose 4B beads, was incubated with HEK293 cell lysates. After extensive washing, bead bound proteins were resolved by SDS-PAGE and immunoblotted with anti-MEKK2 antibody (Fig. 4C). The results definitively show hPRK2(aa 616-670) binds MEKK2 at endogenous expression levels of MEKK2 in HEK 293 cells. MEKK2 does not bind hPKN(aa 585-628).

Fig. 4D shows the alignment of the MEKK2 binding region for hPRK2 and mPRK2 with PKN. Within the start of the kinase domain, residues 619-628 of hPKN are identical to the corresponding sequences of PRK2(aa 661-670). We also demonstrated that PRK2(aa 479-622) does not bind MEKK2, indicating that amino acids 623-660 encode the PRK2 binding sequence for MEKK2. The sequence corresponding to hPRK2(aa 637-660) where the kinase domain begins is particularly conserved between the mouse and human PRK2 sequences and either absent or divergent in PKN consistent with this sequence contributing to the PRK2 domain that binds MEKK2.

MEKK2 Activates PRK2-- In vivo activation of PRK2 was characterized by analyzing the kinase activity of PRK2 immunoprecipitated from transfected HEK293 cells (Fig. 5). Since no physiological PRK2 substrates have been identified, we used MBP as a substrate in the PRK2 kinase assay (7, 12). In the absence of transfected MEKK2, expressed Flag-tagged hPRK2 had little measurable kinase activity (Fig. 5A, lane 2). Strikingly, when Flag-hPRK2 was co-transfected with MEKK2, a dramatic increase in PRK2 activity was observed in anti-Flag immunoprecipitations from cell lysates (lane 3). Surprisingly, cell expression of kinase inactive MEKK2 also activated PRK2 (lanes 4 and 5). This result suggested that an interaction independent of MEKK2 kinase activity resulted in the activation of PRK2. However, the level of PRK2 activation was less with kinase-inactive MEKK2 than that observed with the wild type kinase, indicating that wild type MEKK2 was more active than the kinase inactive mutant in stimulating PRK2 activity. Immunoblotting of cell lysates (panel B) revealed that even when the kinase-inactive form of MEKK2 was expressed at a higher level than the wild type MEKK2, it was less effective than the wild type kinase in activating PRK2 (compare lanes 3 and 5). PRK2 is clearly the kinase assayed in the immunoprecipitates because expression of Flag-tagged kinase inactive PRK2 (Flag-PRK2KE) in the presence of MEKK2 does not result in MBP phosphorylation in the in vitro kinase assay (lane 6). The immunoblots in panel B also show that the expressed wild type MEKK2 migrates as a slower band in SDS-PAGE than the kinase inactive MEKK2 due to autophosphorylation of expressed MEKK2. The kinase-inactive MEKK2 does not show this gel shift. The lower bands in panel B seen most clearly in lanes 1 and 2 are nonspecific bands recognized by the anti-HA antibody and are unrelated to MEKK2.

View larger version (32K): [in this window] [in a new window]   Fig. 5.   PRK2 is activated by MEKK2 in vivo. A, HEK293 cells were transfected with the indicated combinations of expression plasmids. Equal amount of Flag-hPRK2 or Flag-hPRK2KE was immunoprecipitated from cell lysates and incubated in a kinase reaction using MBP as substrate. B, equal amounts of cell lysates were Western-blotted with anti-MEKK2 antibody.

The ability of MEKK2 to activate PRK2 was verified using purified recombinant MEKK2 (Fig. 6). Immunoprecipitated PRK2 was incubated with MEKK2 purified from Sf9 cells infected with baculovirus encoding the MEKK2 cDNA. Immunoprecipitations demonstrated that purified recombinant MEKK2 bound to Flag-PRK2 similar to endogenous cellular MEKK2 (data not shown). Immunoprecipitated PRK2 by itself had little kinase activity (lane 2). Addition of MEKK2 to the Flag-PRK2 immunoprecipitate dramatically increased the phosphorylation of PRK2 (compare lanes 2 and 3). Interestingly, incubation of PRK2 with heat-inactivated MEKK2 to inhibit MEKK2 kinase activity also stimulated the phosphorylation of PRK2 (lane 1). The increased phosphorylation of PRK2 required the kinase activity of PRK2, demonstrating the MEKK2-dependent stimulation of PRK2 autophosphorylation (compare lanes 1, 3, and 5).

View larger version (69K): [in this window] [in a new window]   Fig. 6.   MEKK2 binding is sufficient for activation of PRK2 in vitro, but the kinase activity of PRK2 is required for phosphorylation of MEKK2. Flag-tagged human PRK2 or kinase inactive PRK2KE was immunoprecipitated from transfected HEK293 cells and incubated with recombinant MEKK2 purified from baculovirus-infected insect cells in a kinase reaction. The reactions were resolved on an SDS-PAGE gel and exposed to a Kodak X-Omat film.

Co-incubation of hPRK2 and MEKK2 led to an increased phosphorylation of MEKK2 (lane 3). Clearly, kinase inactive PRK2 (PRK2KE) does not enhance MEKK2 phosphorylation (lane 5) and heat-inactivated MEKK2 is not a substrate for activated PRK2 (lane 1). Thus, PRK2 may phosphorylate MEKK2 or increase the autophosphorylation activity of MEKK2 resulting from their interaction.

PRK2 Does Not Influence MEKK2 Activation of the JNK Pathway-- HEK293 cells were transfected with MEKK2, hPRK2, or a combination of both, and cell lysates assayed for JNK activity (Fig. 7). Expression of MEKK2 gave a pronounced stimulation of JNK activity (compare lanes 1, 2, and 3). Cells transfected with hPRK2 showed basal JNK activity (compare lanes 3 and 4). Furthermore, cells co-transfected with hPRK2 plus MEKK2 showed the same JNK activity as cells transfected with MEKK2 alone (compare lanes 1 and 2), indicating that PRK2 does not regulate the JNK pathway under conditions where MEKK2 activates both the JNK pathway and PRK2.

View larger version (51K): [in this window] [in a new window]   Fig. 7.   PRK2 does not contribute to JNK activation. HEK293 cells were transfected with the indicated plasmids and equal amounts of cell lysates were incubated with purified GST-c-Jun-(1-79) or GST bound to Sepharose beads. Endogenous JNK activity was subject to analysis in a kinase reaction followed by subsequent SDS-PAGE and autoradiography. A, 5-min exposure to a film. B, 10-s exposure.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

PRK2 apparently has a complex regulation responding to several different regulatory pathways in cells (6, 38). It has been shown to be activated by RhoA and to bind Rac (12, 39). In addition, PRK2 can be activated by cardiolipins and is a substrate for caspase-dependent cleavage and activation (7, 27, 28, 40). Activation of PRK2 by RhoA, cardiolipins and protease-catalyzed cleavage appears to involve the release of pseudo-substrate inhibition of PRK2 activity (7, 41). These findings indicate that the primary mechanism defined for PRK2 activation involves its interaction with regulatory molecules that release the internal inhibition of the kinase encoded within the PRK2 sequence.

The results we have presented demonstrate that MEKK2 and PRK2 are binding partners. The region of PRK2 that binds MEKK2 is divergent in the closely related PKN protein and PKN does not interact with MEKK2. This finding demonstrates a unique regulatory function for the control of PRK2 activity relative to PKN by MEKK2. Importantly, the interaction of MEKK2 and PRK2 activates PRK2 kinase activity. MEKK2 regulation of PRK2 was demonstrated in cells and in vitro with purified recombinant MEKK2. As mentioned above, previously described regulatory mechanisms for the activation of PRK2 have involved the release of a pseudo-substrate inhibition of its kinase activity (7, 41). Our findings with MEKK2 activation of PRK2 are consistent with a protein-protein interaction induced release of a pseudo-substrate inhibition of PRK2 kinase activity, with the kinase activity of MEKK2 not being required for its activation of PRK2.

The activation of one kinase by the interaction with a second kinase, independent of the second kinase's phosphorylation activity has been described previously. The kinase suppressor of Ras (KSR) has been proposed to regulate the activity of Raf independent of its kinase activity (42). Expression experiments indicate that modest expression of KSR can activate Raf and high expression inhibits Raf activation (42, 43). The function of KSR for Raf activation has been proposed to be that of a scaffold for organization of the Raf-MEK-ERK signaling module independent of its kinase function (44-48). PKC has been shown to interact with and activate phospholipase D independent of PKC kinase activity (49). Finally, the Ste20-like germinal center kinase was shown to activate the JNK pathway independent of its kinase activity (50). The mechanism for kinase-inactive germinal center kinase activation of the JNK pathway was not defined but was predicted to be due to protein-protein interactions reminiscent of KSR regulation of the Raf-MEK-ERK pathway (47). These examples indicate that scaffolding and possibly oligomerization of heterokinases can regulate the kinase activity of specific binding partners in the macromolecular complex. MEKK2 regulation of PRK2 is another example of such a regulatory mechanism.

Fig. 8 shows a model of the potential organization of the PRK2-MEKK2 signaling complex in cells. PRK2 has been shown to bind the middle SH3 domain of Nck (9), to bind PDK1 (26), and to be activated by RhoA (12). These findings would suggest that multiple mechanisms exist for the activation of PRK2 involving tyrosine kinase binding of Nck, phosphatidylinosotol 3,4,5-trisphosphate activation of PDK1, and stimulation of RhoA GTP binding, in addition to PRK2 interaction with MEKK2. Based on these findings, our hypothesis is that PRK2-MEKK2 interactions function to co-localize two protein kinases that regulate different and divergent signaling pathways in the cell. The requirement for the organization of signaling complexes for MAPK pathways in yeast and mammalian cells is becoming increasingly apparent. In yeast it is clear that Ste5 and PBS2 have scaffolding functions for regulating the mating response and activation of Hog-1 in response to high osmolarity, respectively. In mammalian cells, MP-1 has been shown to function as a scaffold for MEK1 and ERK1 (51) and KSR appears to be a scaffold for Raf, MEK, and ERK (47). Similarly, JIP-1 was shown to function as a scaffold for HPK1, MLK3, MKK7, and JNK1/2 (52). MEKK1 has also been proposed to be a scaffold for MKK4 and JNK1/2 (53). Our findings may be most like that for KSR and Raf in that the two kinases are not in the same pathway.

View larger version (12K): [in this window] [in a new window]   Fig. 8.   Model depicting the different association complexes possible for PRK2 and MEKK2 with Nck, RhoA, and PDK1 (see "Results" for details).

Both MEKK2 and PRK2 have been shown to re-distribute in the cell following specific stimulation of cells (3, 54). MEKK2 has been shown to localize to the T cell receptor signaling complex in response to antigen presentation. PRK2 has been shown to re-distribute from the cytoplasm to the germinal vesicle soon after hormone treatment of starfish oocytes. RhoA is widely distributed in the cytoplasmic compartment of the cell and signals from tyrosine kinases and other GTP-binding proteins can activate RhoA. Thus, it is likely that PRK2·MEKK2 signaling complexes could be formed by several different mechanisms in different locations in the cell. At present the function of PRK2 is unclear. Findings from different laboratories have suggested that PRK2 may be involved in the regulation of the cytoskeleton and specific gene expression (9, 12). MEKK2 is clearly involved in the regulation of the JNK (4) and ERK5² pathways. The association of PRK2 and MEKK2 would allow for the coordinate regulation of their respective functions. In this sense, PRK2 and MEKK2 would be components of a macromolecular signaling module. This module would regulate multiple pathways in response to specific stimulatory inputs. Such a signaling module need not be necessarily preformed but could be brought together, for example, by the activation of RhoA, phosphatidylinositol 3,4,5-kinase or a phosphotyrosine motif that binds the SH2 domain of Nck. It will be important to define the structure and organization of modules such as that of PRK2·MEKK2 to understand the modular nature of signal transduction.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants DK37871, GM30324, DK48845, and CA58187.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Pharmacology, C-236, University of Colorado Health Sciences Center, 4200 E. Ninth Ave., Denver, CO 80262. Tel.: 303-315-1009; Fax: 303-315-1022; E-mail: gary.johnson@uchsc.edu.

Published, JBC Papers in Press, May 18, 2000, DOI 10.1074/jbc.M003148200

² W. Sun and G. L. Johnson, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: MAPK, mitogen-activated protein kinase; DMEM, Dulbecco's modified Eagle's medium; ERK, extracellular signal-regulated kinase; -gal, -galactosidase; GAD, Gal4 activation domain; GBD, Gal4 DNA binding domain; GST, glutathione S-transferase; JNK, c-Jun NH₂-terminal kinase; MBP, myelin basic protein; MEK, mitogen-activated protein kinase/extracellular signal-regulated kinase kinase; MEKK, , mitogen-activated protein kinase/extracellular signal-regulated kinase kinase kinase; ONPG, o-nitrophenyl--D-galactopyranoside; PAGE, polyacrylamide gel electrophoresis; PDK, 3-phosphoinositide-dependent protein kinase; PKC, protein kinase C; PKN, protein kinase N; PRK, protein kinase C-related kinase; VAD, VP16 activation domain; HA, hemagglutinin; SH, Src homology; aa, amino acid(s); 3-AT, 3-aminotriazole; WASP, Wiskott-Aldrich syndrome protein; PAK, p21-activated protein kinase; KSR, kinase suppressor of Ras.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Betson and J. Settleman A Rho-Binding Protein Kinase C-Like Activity Is Required for the Function of Protein Kinase N in Drosophila Development Genetics, August 1, 2007; 176(4): 2201 - 2212. [Abstract] [Full Text] [PDF]

				M. Blonska, Y. You, R. Geleziunas, and X. Lin Restoration of NF-{kappa}B Activation by Tumor Necrosis Factor Alpha Receptor Complex-Targeted MEKK3 in Receptor-Interacting Protein-Deficient Cells Mol. Cell. Biol., December 15, 2004; 24(24): 10757 - 10765. [Abstract] [Full Text] [PDF]

				S.-J. Kim, J.-H. Kim, Y.-G. Kim, H.-S. Lim, and J.-W. Oh Protein Kinase C-related Kinase 2 Regulates Hepatitis C Virus RNA Polymerase Function by Phosphorylation J. Biol. Chem., November 26, 2004; 279(48): 50031 - 50041. [Abstract] [Full Text] [PDF]

				C. McDonald, P. O. Vacratsis, J. B. Bliska, and J. E. Dixon The Yersinia Virulence Factor YopM Forms a Novel Protein Complex with Two Cellular Kinases J. Biol. Chem., May 9, 2003; 278(20): 18514 - 18523. [Abstract] [Full Text] [PDF]

				W. Sun, X. Wei, K. Kesavan, T. P. Garrington, R. Fan, J. Mei, S. M. Anderson, E. W. Gelfand, and G. L. Johnson MEK Kinase 2 and the Adaptor Protein Lad Regulate Extracellular Signal-Regulated Kinase 5 Activation by Epidermal Growth Factor via Src Mol. Cell. Biol., April 1, 2003; 23(7): 2298 - 2308. [Abstract] [Full Text] [PDF]

				H. Mukai The Structure and Function of PKN, a Protein Kinase Having a Catalytic Domain Homologous to That of PKC J. Biochem., January 1, 2003; 133(1): 17 - 27. [Abstract] [Full Text] [PDF]

				W. Sun, K. Kesavan, B. C. Schaefer, T. P. Garrington, M. Ware, N. L. Johnson, E. W. Gelfand, and G. L. Johnson MEKK2 Associates with the Adapter Protein Lad/RIBP and Regulates the MEK5-BMK1/ERK5 Pathway J. Biol. Chem., February 9, 2001; 276(7): 5093 - 5100. [Abstract] [Full Text] [PDF]

				C. S. T. Hii, N. Moghadammi, A. Dunbar, and A. Ferrante Activation of the Phosphatidylinositol 3-Kinase-Akt/Protein Kinase B Signaling Pathway in Arachidonic Acid-stimulated Human Myeloid and Endothelial Cells. INVOLVEMENT OF THE ErbB RECEPTOR FAMILY J. Biol. Chem., July 13, 2001; 276(29): 27246 - 27255. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/32/24421    most recent M003148200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map