Portal de Periódicos da CAPES

The Yersinia Virulence Factor YopM Forms a Novel Protein Complex with Two Cellular Kinases

2003; Elsevier BV; Volume: 278; Issue: 20 Linguagem: Inglês

10.1074/jbc.m301226200

1083-351X

Christine McDonald, Panayiotis O. Vacratsis, James B. Bliska, Jack E. Dixon,

Pharmacological Effects of Natural Compounds

Pathogenic Yersinia contain a virulence plasmid that encodes genes for intracellular effectors, which neutralize the host immune response. One effector, YopM, is necessary for Yersinia virulence, but its function in host cells is unknown. To identify potential cellular pathways affected by YopM, proteins that co-immunoprecipitate with YopM in mammalian cells were isolated and identified by mass spectrometry. Results demonstrate that two kinases, protein kinase C-like 2 (PRK2) and ribosomal S6 protein kinase 1 (RSK1), interact directly with YopM. These two kinases associate only when YopM is present, and expression of YopM in cells stimulates the activity of both kinases. RSK1 is activated directly by interaction with YopM, and RSK1 kinase activity is required for YopM-stimulated PRK2 activity. YopM activation of RSK1 occurs independently of the actions of YopJ on the MAPK pathway. YopM is also required for Yersinia-induced changes in RSK1 mobility in infected macrophage cells. These results identify the first intracellular targets of YopM and suggest YopM acts to stimulate the activity of PRK2 and RSK1. Pathogenic Yersinia contain a virulence plasmid that encodes genes for intracellular effectors, which neutralize the host immune response. One effector, YopM, is necessary for Yersinia virulence, but its function in host cells is unknown. To identify potential cellular pathways affected by YopM, proteins that co-immunoprecipitate with YopM in mammalian cells were isolated and identified by mass spectrometry. Results demonstrate that two kinases, protein kinase C-like 2 (PRK2) and ribosomal S6 protein kinase 1 (RSK1), interact directly with YopM. These two kinases associate only when YopM is present, and expression of YopM in cells stimulates the activity of both kinases. RSK1 is activated directly by interaction with YopM, and RSK1 kinase activity is required for YopM-stimulated PRK2 activity. YopM activation of RSK1 occurs independently of the actions of YopJ on the MAPK pathway. YopM is also required for Yersinia-induced changes in RSK1 mobility in infected macrophage cells. These results identify the first intracellular targets of YopM and suggest YopM acts to stimulate the activity of PRK2 and RSK1. Yersinia outer protein mitogen-activated protein kinase leucine-rich repeat protein kinase C-like 2 p90 ribosomal protein S6 kinase 1 fetal bovine serum matrix-assisted laser desorption/ionization-time of flight post-source decay liquid chromatography mass spectrum myelin basic protein enhanced green fluorescent protein Dulbecco's modified Eagle's medium hemagglutinin isopropyl-1-thio-β-d-galactopyranoside glutathioneS-transferase cAMP-response element-binding protein The bacterium Yersinia has three species that are pathogenic for humans. Yersinia pestis causes plague and is transmitted by fleas from rodent reservoirs to humans, frequently resulting in fatal infections. This bacterium is responsible for three plague pandemics and has been involved in two plague epidemics in western India as recently as 1944 (1Perry R.D. Fetherston J.D. Clin. Microbiol. Rev. 1997; 10: 35-66Crossref PubMed Google Scholar). Two other species,Yersinia enterocolitica and Yersinia pseudotuberculosis, are also human pathogens, and infection with these species results in gastrointestinal disorders such as appendicitis, ileocolitis, and mesenteric adenitis. Contaminated food or water is the most common route of infection by these species, with swine serving as a major reservoir for these bacteria (2Bottone E.J. Clin. Microbiol. Rev. 1997; 10: 257-276Crossref PubMed Google Scholar). Although the severity of the disease and the mode of transmission of pathogenicYersinia species differs, they all target lymphoid tissues and are able to block innate immune responses to allow extracellular replication of the bacterium. Yersinia spp. harbor a virulence plasmid of ∼70 kb that encodes proteins termed Yops1(for Yersiniaouterproteins) that are either intracellular effectors or membrane proteins that create a delivery system for these effectors (3Juris S.J. Shao F. Dixon J.E. Cell Microbiol. 2002; 4: 201-211Crossref PubMed Scopus (68) Google Scholar,4Cornelis G.R. Boland A. Boyd A.P. Geuijen C. Iriarte M. Neyt C. Sory M.P. Stainier I. Microbiol. Mol. Biol. Rev. 1998; 62: 1315-1352Crossref PubMed Google Scholar). Six effector proteins from Yersinia (YopE, YopH, YopJ/P, YopM, YopT, and YpkA/YopO) are delivered into eukaryotic cells to inactivate the host immune response. YopE, YopH, YopT, and YpkA target the actin cytoskeleton through different mechanisms to prevent phagocytosis of the bacterium. YopJ inhibits the production of certain inflammatory cytokines by blocking the mitogen-activated protein kinase (MAPK) and NFκB signaling pathways. YopJ has also been implicated in apoptosis induction in macrophages through caspase-8 activation and inhibition of NFκB-induced transcription. The effects of YopM on host response to Yersinia infection are largely unknown. However, YopM is required for full virulence ofY. pestis and Y. enterocolitica, as demonstrated in mouse infection models (5Leung K.Y. Reisner B.S. Straley S.C. Infect. Immun. 1990; 58: 3262-3271Crossref PubMed Google Scholar, 6Mulder B. Michiels T. Simonet M. Sory M.P. Cornelis G. Infect. Immun. 1989; 57: 2534-2541Crossref PubMed Google Scholar). The YopM protein is an acidic, ∼42-kDa protein composed almost entirely of leucine-rich repeat (LRR) motifs. The LRR motif is thought to be a protein-protein interaction motif and is present in multiple signaling molecules found both intracellularly and extracellularly (7Kobe B. Kajava A.V. Curr. Opin. Struct. Biol. 2001; 11: 725-732Crossref PubMed Scopus (1249) Google Scholar). Initial studies suggested that YopM is secreted from the bacteria and interacts with thrombin to inhibit platelet aggregation (8Reisner B.S. Straley S.C. Infect. Immun. 1992; 60: 5242-5252Crossref PubMed Google Scholar). Subsequent studies demonstrated that YopM is translocated into target cells via the type III secretion apparatus (9Boland A. Sory M.P. Iriarte M. Kerbourch C. Wattiau P. Cornelis G.R. EMBO J. 1996; 15: 5191-5201Crossref PubMed Scopus (169) Google Scholar). Once inside the target cell, YopM moves from the cytoplasm to the nucleus via a vesicle-associated pathway (10Skrzypek E. Cowan C. Straley S.C. Mol. Microbiol. 1998; 30: 1051-1065Crossref PubMed Scopus (125) Google Scholar, 11Skrzypek E. Myers-Morales T. Whiteheart S.W. Straley S.C. Infect. Immun. 2003; 71: 937-947Crossref PubMed Scopus (36) Google Scholar). A gene chip microarray analysis of Y. enterocolitica-infected macrophage cell lines suggests that YopM affects the expression of genes involved in cell cycle and cell growth (12Sauvonnet N. Pradet-Balade B. Garcia-Sanz J.A. Cornelis G.R. J. Biol. Chem. 2002; 277: 25133-25142Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). However, the mechanism by which YopM produces these effects on gene expression is unknown. This paper describes the identification of two intracellular targets of YopM, protein kinase C-like 2 (PRK2) and ribosomal S6 protein kinase 1 (RSK1). Human embryonic kidney 293 cells were maintained in DMEM (Invitrogen) + 10% fetal bovine serum (FBS, Invitrogen) + penicillin/streptomycin (Invitrogen). Cells were transfected in 100-mm plates with 5 μg of DNA using FuGENE 6 (Roche Applied Science) at a ratio of 3:1 (FuGENE 6:DNA) according to the manufacturer's instructions. Mouse macrophage J774A.1 cells were maintained in DMEM + 10% FBS (Hyclone) + 1 mm sodium pyruvate (Invitrogen). The following antibodies were used in the experiments described: anti-FLAG M2 (Sigma), anti-PRK2 (Cell Signaling Technology and Transduction Laboratories), anti-RSK1 (C-21, Santa Cruz Biotechnology), anti-HA (Y-11, Santa Cruz Biotechnology and clone 3F10, Roche Applied Science), anti-GST (B-14, Santa Cruz Biotechnology), anti-GFP (FL, Santa Cruz Biotechnology), anti-CREB (Cell Signaling Technology), and anti-Akt (Cell Signaling Technology). YopM polyclonal rabbit antisera was a gift of O. Schneewind, University of Chicago (13Lee V.T. Anderson D.M. Schneewind O. Mol. Microbiol. 1998; 28: 593-601Crossref PubMed Scopus (119) Google Scholar). FLAG-YopM was constructed by amplifying by PCR nucleotides 7–1265 of YopM from the Y. enterocolitica virulence plasmid, pYVe8081, and cloning into pFLAG-CMV-2 (Eastman Kodak Co.), resulting in the incorporation of an amino-terminal FLAG tag. EGFP-YopM was constructed by PCR amplifying YopM (nucleotides 7–1265) from FLAG-YopM and cloned into pEGFP-C1 (Clontech) resulting in a fusion protein with an amino-terminal EGFP tag. The p67N-YopM plasmid was constructed by PCR amplifying nucleotides 1–1265 of Y. enterocolitica YopM and cloning into pMMB67HE (14Furste J.P. Pansegrau W. Frank R. Blocker H. Scholz P. Bagdasarian M. Lanka E. Gene (Amst.). 1986; 48: 119-131Crossref PubMed Scopus (810) Google Scholar). YopM-6xHis was a gift of Don Huddler (University of Michigan). FLAG-PRK2 wild type and FLAG-PRK2 kinase-deficient were gifts of Gary Johnson, University of Colorado Health Sciences Center (15Sun W.Y. Vincent S. Settleman J. Johnson G.L. J. Biol. Chem. 2000; 275: 24421-24428Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar). PRK2 deletion constructs were produced by PCR amplification of regions of human PRK2 and cloning into the mammalian expression vector pEBG-3X, resulting in the incorporation of a GST tag on the amino terminus. Mammalian expression constructs of avian RSK1 (HA-RSK1 and HA-RSK1 K112/464R) were gifts of John Blenis, Harvard Medical School (16Shimamura A. Ballif B.A. Richards S.A. Blenis J. Curr. Biol. 2000; 10: 127-135Abstract Full Text Full Text PDF PubMed Scopus (246) Google Scholar). The mammalian expression construct of rat RSK1 (pMT2-HA-RSK1) was a gift of Valerie Castle, University of Michigan (17Grove J.R. Price D.J. Banerjee P. Balasubramanyam A. Ahmad M.F. Avruch J. Biochemistry. 1993; 32: 7727-7738Crossref PubMed Scopus (55) Google Scholar). RSK1 deletion constructs were produced by PCR amplification of regions of rat RSK1 and cloning into pcDNA3-HA. Subcloning an EcoRI fragment of pMT2-HA-RSK1 into pcDNA3 created the RSK1 template for in vitrotranscription/translation. pSFFV-YopJ-FLAG was a gift of Kim Orth (18Orth K. Palmer L.E. Bao Z.Q. Stewart S. Rudolph A.E. Bliska J.B. Dixon J.E. Science. 1999; 285: 1920-1923Crossref PubMed Scopus (321) Google Scholar). 293 cells were transfected in 100-mm plates with 5 μg of expression plasmid for a total of 40–45 h. Prior to harvest, cells were starved in DMEM for 16–18 h. Cells were washed with PBS and lysed in RIPA− buffer (150 mmNaCl, 50 mm Tris, pH 8, 0.4 mm EDTA, 10% glycerol, 1% Nonidet P-40, 1 mm pefabloc (Roche Applied Science), 10 μm leupeptin (Roche Applied Science), 1 mm benzamidine (Sigma), 10 μm E64 (Sigma)) followed by centrifugation at 15,000 rpm for 30 min at 4 °C. For cell fractionation experiments, nuclear and cytoplasmic extracts were prepared as described previously (19Dignam J.D. Lebovitz R.M. Roeder R.G. Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9131) Google Scholar). Lysates were precleared with 10–20 μl of protein G-agarose (Invitrogen) for 30 min of rocking at 4 °C. Lysates were then incubated with 15–20 μl of anti-FLAG M2-Sepharose (Sigma) or 1–2 μg of antibody and 10–20 μl of protein G-agarose for 1–2 h at 4 °C. Beads were washed 4 times with 1 ml of 1 m RIPA− buffer (1 mNaCl, 50 mm Tris, pH 8, 0.4 mm EDTA, 10% glycerol, 1% Nonidet P-40), 2 times with 1 ml of 500 mmRIPA− buffer (500 mm NaCl, 50 mmTris, pH 8, 0.4 mm EDTA, 10% glycerol, 1% Nonidet P-40), and 2 times with 1 ml of RIPA− buffer. Immunoprecipitates used in kinase assays were washed an additional 2 times with 1 ml of 20 mm Hepes, pH 7.4, 10 mm MgAc. Silver staining of SDS-PAGE gels was performed as described previously (20Blum H. Gross H.J. Beier H. Virology. 1989; 169: 51-61Crossref PubMed Scopus (34) Google Scholar). Immunoprecipitates from 12 to 16 100-mm plates of transfected 293 cells were separated by SDS-PAGE followed by staining in 0.25% Coomassie Blue, 10% glacial acetic acid, 40% MeOH. The gel was destained in 10% glacial acetic acid, 40% MeOH and dried onto Whatman paper. Protein bands of interest were excised from the dried gel and digested with trypsin as described previously (21Vacratsis P.O. Phinney B.S. Gage D.A. Gallo K.A. Biochemistry. 2002; 41: 5613-5624Crossref PubMed Scopus (33) Google Scholar). MALDI-TOF mass spectrometry was performed on a Voyager DE-STR instrument (Applied Biosystems) in linear positive mode or reflector positive mode (for PSD analysis) using α-cyano-4-hydroxycinnamic acid as the UV absorbing matrix. Peptide values from the mass fingerprint were used to search the NCBI data base using the Profound and MS-Fit software tools. Nanospray LC/MS/MS was performed on a LCQDECA quadrupole ion trap mass spectrometer (Thermo Finnigan) equipped with a picoview nanospray source (New Objective). A 5-μl aliquot of the pool of tryptic peptides was injected onto a 75-μm inner diameter × 15-cm C18 picofrit column (New Objective) that terminated in a 15-μm tip spray needle. Peptides were eluted with 60% acetonitrile, 1% formic acid. Mass spectra was acquired using the Top3 triple play mode, and MS/MS spectra were used to search the NCBI using SEQUEST software. Immunoprecipitated proteins were incubated for 30 min at room temperature in 20 mm Hepes, pH 7.4, 10 mm MgAc, 1 mm dithiothreitol, 50 μm ATP, 20 μCi of [γ-32P]ATP (PerkinElmer Life Sciences), with 1 μg of MBP (Sigma) as substrate in a 20-μl reaction volume. One-third of the reaction was separated by SDS-PAGE, dried onto Whatman paper, and exposed to film at −80 °C with an intensifying screen. The PRK2 and RSK1 kinases werein vitro transcribed and translated in the presence of [35S]methionine (PerkinElmer Life Sciences) using the TnT T7 Quick-coupled Transcription/Translation System (Promega) from linearized PRK2-V5 (AgeI) or pcDNA3-HA-RSK1 (NcoI) plasmid templates. Radiolabeled proteins were incubated with bacterially produced recombinant YopM-6xHis protein bound to nickel-agarose beads (Qiagen) in 50 mm Tris, pH 8, 150 mm NaCl, 1% Nonidet P-40. Beads were subsequently washed with 50 mm Tris, pH 8, 150 mm NaCl, 1% Nonidet P-40, 10 mm imidazole and run out on SDS-PAGE. The gel was fixed in 40% MeOH, 10% glacial acetic acid, treated with Amplify (Amersham Biosciences), and dried onto Whatman paper. The gel was then exposed to film at −80 °C with a screen. The Y. pseudotuberculosis strains used in this study were derived from YP126 (serogroup O:3). YP22 has been described previously (22Palmer L.E. Hobbie S. Galan J.E. Bliska J.B. Mol. Microbiol. 1998; 27: 953-965Crossref PubMed Scopus (238) Google Scholar) and is deficient for the expression of YopE, YopH, and YopK proteins. The YP33 strain was derived from YP22 and contains a frameshift inyopM, disrupting the expression of the YopM protein. The YP33/yopM strain was constructed by introducing the IPTG-inducible YopM expression plasmid, p67N-YopM into YP33 through conjugation. Yersinia strains were grown in Luria broth at 26 °C, diluted 1:40 in Luria broth supplemented with 20 mm sodium oxalate and 20 mm MgCl2(A600 0.1), and grown for 1 h at 26 °C. These cultures were then shifted to 37 °C and grown for an additional 2 h. For infections using YP33/yopM, YopM expression was induced with 0.1 mm IPTG for the 2 h at 37 °C. Yersinia were washed in Hanks' balanced salt solution (Invitrogen) and resuspended atA600 0.5 in Hanks' balanced salt solution. J774A.1 cells were grown to 80–90% confluency in DMEM + 10% FBS + 1 mm sodium pyruvate and infected for 2 h at 37 °C at a multiplicity of infection of 50. The crystal structure of YopM has been solved and reveals that the YopM monomer has an amino-terminal α-helical hairpin followed by a curving repeat structure composed of LRR motifs (23Evdokimov A.G. Anderson D.E. Routzahn K.M. Waugh D.S. J. Mol. Biol. 2001; 312: 807-821Crossref PubMed Scopus (138) Google Scholar). The LRR motifs in YopM form a structure that has a concave face composed of β-sheets and a polyproline II helical conformation on the convex side. YopM is also shown to form a homo-tetramer that creates a tube-like structure with a pore of ∼35 Å. The structure does not indicate a catalytic function for YopM but instead suggests that it may act as a protein scaffold recruiting proteins into a complex. To identify potential cellular targets affected by thisYersinia effector, a biochemical approach was used to identify proteins that interact with YopM in mammalian cells. In this approach, proteins that associate with YopM in transiently transfected cells were isolated by immunoprecipitation and identified by mass spectrometry. Lysates from 293 cells transfected with a FLAG-YopM expression vector or the empty FLAG vector were immunoprecipitated with anti-FLAG antibody, and the associated proteins were visualized by silver stain of SDS-PAGE gels (Fig.1A). Three proteins with approximate molecular masses of 120, 80, and 78 kDa were found to specifically co-immunoprecipitate with YopM. The YopM-associated proteins were excised from a Coomassie-stained gel, trypsin-digested, and the resulting peptides analyzed by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry or nanospray tandem mass spectrometry (LC/MS/MS). The pool of p120 tryptic peptides was analyzed by MALDI-TOF (Fig.1B). The peptide values from the peptide mass fingerprint were used to search the NCBI non-redundant data base using two web site tools, ProFound (prowl.rockefeller.edu/cgi-bin/ProFound) (24Zhang W. Chait B.T. Anal. Chem. 2000; 72: 2482-2489Crossref PubMed Scopus (551) Google Scholar), and MS-Fit (prospector.ucsf.edu). Both programs predicted the identity of p120 to be protein kinase C-like 2 (PRK2), a novel serine/threonine protein kinase of the protein kinase C family (25Palmer R.H. Ridden J. Parker P.J. FEBS Lett. 1994; 356: 5-8Crossref PubMed Scopus (39) Google Scholar, 26Palmer R.H. Ridden J. Parker P.J. Eur. J. Biochem. 1995; 227: 344-351Crossref PubMed Scopus (121) Google Scholar). In addition, MALDI post-source decay (PSD) was employed to obtain partial sequence information on two prominent peptides in the mass fingerprint and were found to match the PRK2 sequence. The identity of the p80 and p78 proteins was determined by LC/MS/MS analysis (Fig. 1C). The MS/MS fragmentation spectra was used to search the NCBI non-redundant data base for protein identification. Three MS/MS spectra from the p80 and p78 samples unambiguously identified these proteins as p90 ribosomal protein S6 kinase 1 (RSK1/p90rsk/MAPKAP-K1), a member of the growth factor-regulated S6 serine/threonine kinase family (27Blenis J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5889-5892Crossref PubMed Scopus (1151) Google Scholar, 28Frodin M. Gammeltoft S. Mol. Cell. Endocrinol. 1999; 151: 65-77Crossref PubMed Scopus (611) Google Scholar). The mass spectral identification of p120, p80, and p78 was confirmed by Western blot analysis of YopM-associated proteins (Fig.2A). Lysates from 293 cells transfected with a FLAG-YopM plasmid or vector alone were immunoprecipitated with anti-FLAG antibody, and proteins were visualized by sequential Western blots using anti-FLAG, anti-PRK2, and anti-RSK1 antibodies. A single ∼45-kDa protein corresponding to FLAG-YopM was detected in anti-FLAG Western blots. Anti-PRK2 Western blots reacted with a single band of 120 kDa co-immunoprecipitating with FLAG-YopM. Anti-RSK1 Western blots detected a YopM co-immunoprecipitating doublet of ∼80 kDa. These results confirm the mass spectral identification of p120 as PRK2 and p80/p78 as RSK1. The ability of YopM to interact with these cellular proteins was examined in Yersinia-infected cells to determine whether the interactions detected in transfected cells are also seen in the context of an infection. The macrophage cell line J774A.1 was infected withyopM+ (YP22) or yopM−(YP33) Y. pseudotuberculosis strains, immunoprecipitated with YopM polyclonal antisera, and the co-immunoprecipitation of RSK1 and PRK2 determined by Western blot. YopM co-immunoprecipitated RSK1 (Fig. 2B), but the co-immunoprecipitation of PRK2 was not detectable in this assay (data not shown). The inability to detect PRK2 may be due to a number of factors but is most likely associated with the quality of the YopM antisera. For example, when FLAG-YopM-transfected 293 cells were immunoprecipitated with the YopM antiserum, the interaction of PRK2 and YopM was much weaker and difficult to detect (data not shown), indicating that the antiserum does not effectively immunoprecipitate the YopM-PRK2 complex. However, our result firmly establishes that YopM forms a complex with RSK1 inY. pseudotuberculosis-infected macrophage cells. The interaction of YopM with PRK2 and RSK1 was also assessed in vitro to determine whether these kinases bind to YopM directly. Recombinant YopM with a His6 tag was bound to nickel beads and incubated with radiolabeled in vitrotranscribed/translated kinases (Fig. 2C). Proteins bound to the YopM beads were separated by SDS-PAGE, and the association of the kinases was visualized by fluorography. Both PRK2 and RSK1 bound to recombinant YopM in vitro. Deletion of the carboxyl terminus of these kinases abolished interaction with YopM in this assay (data not shown). These results, as well as results described in experiments to follow (Fig. 3), indicate that a region in the carboxyl-terminal half of these kinases is required for interaction with YopM. The in vitro binding data suggest a direct interaction between YopM and PRK2, as well as YopM and RSK1. Further experiments were performed to determine whether YopM forms a single complex with these two kinases or two separate kinase complexes. Lysates from vector-transfected cells or cells transfected with FLAG-YopM were immunoprecipitated with anti-PRK2, anti-RSK1, or anti-FLAG antibodies, and proteins were visualized by Western blot (Fig. 2D). In vector-transfected cells, PRK2 and RSK1 do not co-immunoprecipitate, indicating that these two kinases normally do not associate in cells (lanes 2 and 3). However, in cells that express YopM, PRK2 co-immunoprecipitated both YopM and RSK1 (lane 4). RSK1 immunoprecipitates also showed interaction with both PRK2 and YopM (lane 5). These results demonstrate that YopM forms a single complex with PRK2 and RSK1 in transfected cells and that YopM induces the association of PRK2 and RSK1. The subcellular localization of YopM is distinct from the other Yop effectors. YopM has been demonstrated to be injected into the cytoplasm of cells and translocate to the nucleus via a vesicle-associated pathway (10Skrzypek E. Cowan C. Straley S.C. Mol. Microbiol. 1998; 30: 1051-1065Crossref PubMed Scopus (125) Google Scholar, 11Skrzypek E. Myers-Morales T. Whiteheart S.W. Straley S.C. Infect. Immun. 2003; 71: 937-947Crossref PubMed Scopus (36) Google Scholar). PRK2 has been described as a cytoplasmic protein, whereas RSK1 has been demonstrated to shuttle between the cytoplasm and the nucleus of cells. To determine whether the subcellular localization of PRK2 and RSK1 is altered through association with YopM, the localization of the kinases was examined and compared with their localization in cells expressing YopM (Fig. 2E). Nuclear and cytoplasmic extracts were made from cells transfected with vector or FLAG-YopM, immunoprecipitated with anti-PRK2, anti-RSK1, or anti-FLAG antibodies, and visualized by Western blot. As a control for proper cell fractionation, the localization of a nuclear transcription factor, CREB, was also assessed in these extracts by Western blot. In vector-transfected as well as FLAG-YopM-transfected cells, PRK2 and RSK1 were localized primarily to the cytoplasm, with some nuclear localization as well. YopM was also detected in both cellular compartments, with the majority of the YopM protein localized in the cytoplasm. These results demonstrate that PRK2, RSK1, and YopM localize to similar cellular locations, and association of the kinases with YopM does not redirect their localization. In order to characterize further the protein interactions in the YopM complex, truncated kinase constructs were tested for their ability to co-immunoprecipitate with YopM (Fig. 3). The PRK2 domain structure contains three homologous repeat (HR1) regions in the amino terminus, followed by a central repeat region similar to the pseudosubstrate site of protein kinase C kinases (HR2), an SH3 domain-binding PXXP motif, and a carboxyl-terminal serine/threonine kinase domain (26Palmer R.H. Ridden J. Parker P.J. Eur. J. Biochem. 1995; 227: 344-351Crossref PubMed Scopus (121) Google Scholar). PRK2 deletion constructs were constructed in-frame with an amino-terminal GST tag and transfected into cells with FLAG-YopM. YopM complexes were immunoprecipitated with anti-FLAG antibody and Western-blotted with anti-GST antibody (Fig.3A). YopM co-immunoprecipitated GST-PRK2 constructs containing the carboxyl-terminal amino acids 512–985 (Δ512N) and 648–985 (Δ648N). A weak interaction was also detected between YopM and GST-PRK2 amino acids 331–651 with long exposures. These results indicate that the major region of PRK2 required for interaction with YopM is between amino acids 648 and 985 which encompasses the kinase domain of PRK2. The RSK1 domain structure consists of two distinct kinase domains separated by a linker region. Deletion constructs of RSK1 were constructed in-frame with an amino-terminal HA tag. The HA-RSK1 expression constructs were co-transfected into cells with FLAG-YopM, followed by anti-FLAG immunoprecipitation and Western blot analysis with anti-HA antibody (Fig. 3B). FLAG-YopM was able to co-immunoprecipitate full-length HA-RSK1 (FL) and the HA-RSK1 construct composed of amino acids 1–423 (Δ423C). Further truncation of HA-RSK1 abolished co-immunoprecipitation with YopM. These results suggest that interaction of RSK1 and YopM requires the presence of the linker region between the dual kinase domains of RSK1. We have demonstrated that YopM forms a complex with two cellular kinases, PRK2 and RSK1. In vitro kinase assays were performed to determine the effect of complex formation on the kinase activity of PRK2 and RSK1 (Fig.4). Endogenous kinases were immunoprecipitated from vector-transfected or cells transfected with FLAG-YopM and used in an in vitro kinase assay with [γ-32P]ATP and myelin basic protein (MBP) as substrate. As both RSK1 and PRK2 kinase activity is increased in cells grown in serum, transfected cells were serum-starved for 16 h prior to immunoprecipitation with rabbit IgG (c), anti-PRK2, anti-RSK1, or anti-Akt antibodies (Fig. 4A). The ability of both PRK2 and RSK1 to phosphorylate MBP was increased when YopM was expressed in cells. The kinase activity of Akt, a kinase not found in the YopM complex, was not affected by the expression of YopM. These results suggest that expression of YopM specifically activates PRK2 and RSK1. The relative amounts of YopM expression required to activate PRK2 and RSK1 were investigated by in vitro kinase assays from cells transfected with increasing amounts of FLAG-YopM plasmid. Cells used for assessing PRK2 kinase activity were transfected with 0–5 μg of FLAG-YopM, whereas cells used in the RSK1 assays were transfected with 0–1 μg of FLAG-YopM, with the total amount of transfected plasmid maintained at 5 μg with the addition of vector plasmid. Cells were starved for 16 h and the endogenous PRK2 and RSK1 immunoprecipitated, and activity was assessed by in vitrokinase assay (Fig. 4, B and C). Results showed a dose-dependent increase in both PRK2 and RSK1 kinase activity with increasing amounts of YopM expressed in cells. The amounts required for maximal activity of the two kinases are different, with RSK1 activated by much lower amounts of YopM expressed. These results demonstrate that the endogenous kinase activity of PRK2 and RSK1 is increased in a dose-dependent manner by YopM expressed in cells and that this activation of kinase activity is specific to two kinases that interact with YopM. The previous experiments do not differentiate whether the increase of PRK2 and RSK1 kinase activity is due to direct actions of YopM on the kinases or through stimulation of cellular pathways that lead to the activation of these kinases. To address this, recombinant YopM protein was added toin vitro kinase assays with endogenous PRK2 or RSK1 immunoprecipitated from serum-starved cells (Fig.5A). Addition of recombinant YopM increased the kinase activity of RSK1 in vitro but did not significantly affect PRK2 kinase activity. The results suggest that YopM has a direct effect on RSK1 kinase activity and that the stimulation of PRK2 activity is due to an indirect mechanism. The role of PRK2 and RSK1 kinase activities in the stimulation of the kinases by YopM was assessed using kinase-deficient mutants of PRK2 and RSK1 that contain mutations in critical lysines of their kinase domains that render them inactive (Fig. 5, B–D). In these experiments, cells were co-transfected with an EGFP-tagged YopM (EGFP-YopM), an HA-tagged RSK1 (HA-RSK, either wild-type or kinase-deficient), and a FLAG-tagged PRK2 (FLAG-PRK2, either wild-type or kinase-deficient). These cells were then serum-starved and components of the YopM complex immunoprecipitated, and kinase activity was analyzed by in vitro kinase assays. The relative contribution of PRK2 and RSK1 kinase activity to the YopM complex was assessed by in vitro kinase assays with immunoprecipitated EGFP-YopM from transfected cells (Fig.5B). In this assay, no kinase activity over control levels (lane 1) was seen in immunoprecipitates from cells transfected with both kinase-deficient PRK2 and RSK1 (lane 5), indicating that these are the only two kinase