Both Normal and Transforming PCPH Proteins Have Guanosine

[CANCER RESEARCH 60, 1720–1728, March 15, 2000] Both Normal and Transforming PCPH Proteins Have Guanosine Diphosphatase Activity But Only the Oncoprotein Cooperates with Ras in Activating Extracellular Signal-regulated Kinase ERK11

Juan A. Recio, J. Guillermo Pa´ez, Baishali Maskeri, Mark Loveland, Juan A. Velasco, and Vicente Notario2 Laboratory of Experimental Carcinogenesis, Department of Radiation Medicine, Georgetown University Medical Center, Washington, DC 20007

ABSTRACT ORF3 of PCPH and the translation of the mutated protein to terminate 33 residues downstream in the new ORF. Consequently, the mt-PCPH Previous reports from our laboratory described the activation of the oncoprotein is a truncated form (246 versus 469 amino acids) of its PCPH gene into the PCPH oncogene (mt-PCPH, reported previously as Cph) by a single point mutational deletion. As a consequence, the mt- normal counterpart and contains an additional, rather hydrophobic PCPH oncoprotein is a truncated form of the normal PCPH protein. COOH-terminal tail (6). Although both proteins have ribonucleotide diphosphate-binding activity, Biochemical analyses of the in vitro translated products of mt- only mt-PCPH acted synergistically with a human H-Ras oncoprotein to PCPH and PCPH and functional analyses of cells transformed by transform murine NIH3T3 fibroblasts. We report here the expression of mt-PCPH oncogene or transfected with expression constructs of the PCPH and mt-PCPH proteins in Escherichia coli and the finding that PCPH allowed us to demonstrate (6) that: (a) the products encoded by the purified bacterial recombinant proteins have intrinsic guanosine both PCPH and mt-PCPH genes are ribonucleotide-binding proteins; diphosphatase (GDPase) activity. However, expression of the Syrian hamster PCPH and mt-PCPH proteins in haploid yeast strains engineered to (b) although they share partial homology with GTP/GDP exchange be GDPase deficient by targeted disruption of the single GDA1 allele did factors, they do not catalyze nucleotide exchange on the H-Ras protein not complement their glycosylation-disabled phenotype, suggesting the or any other small G proteins tested; (c) steady-state levels of PCPH existence of significant functional differences between the mammalian and and particularly mt-PCPH mRNA are up-regulated in serum-deprived yeast enzymes. Results from transient cotransfections into NIH3T3, cells; and (d) the mt-PCPH oncoprotein, and to a lesser extent the COS-7, or 293T cells indicated that, in mammalian cells, both PCPH and PCPH protein, provide the cells with enhanced stress-survival func- mt-PCPH cause an overall down-regulation of the stimulatory effect of tions against a variety of stress factors. Although these results suggest epidermal growth factor or the activated ras or raf oncogenes on the Ras/mitogen-activated protein kinase/extracellular signal-regulated ki- that PCPH participates in cellular mechanisms of response to stress, nase (ERK) signaling pathway. However, despite this overall negative its biochemical activity remains unknown. regulatory role on Ras signaling, mt-PCPH, but not PCPH, cooperated Most recently, we completed the cDNA cloning, sequencing, and with the Ras oncoprotein to produce a prolonged stimulation of the chromosomal mapping of the mouse (4) and human (5) PCPH proto- phosphorylation of ERK1 but had no effect on the phosphorylation levels oncogenes and determined that PCPH expression is frequently altered of ERK2. These results represent a clear difference between the mecha- in human neoplasms (5). Indeed, PCPH was not expressed in 16 of 16 nisms of action of PCPH and mt-PCPH and suggest that the ability to primary human renal carcinomas, although it was highly expressed in cause a sustained activation of ERK1 may be an important determinant of the transforming activity of mt-PCPH. matched normal kidney, and PCPH mRNA was also undetectable in the majority (67.4%) of the human tumor cell lines tested. In addition, recent data (7) showed that some human tumor cell lines expressed

INTRODUCTION both the normal Mr 47,000 PCPH protein and a smaller immunore- ϳ active polypeptide with the size (Mr 30,000) of the mt-PCPH The PCPH oncogene (initially reported as Cph and termed mt- protein. These data suggested that PCPH loss and perhaps truncation PCPH here) was originally isolated in our laboratory from chemically may be involved in the development of some human tumors. There- initiated, neoplastic Syrian hamster embryo fibroblasts on the basis of fore, it becomes essential to identify the biochemical activity of the its ability to transform NIH3T3 cells upon serial cycles of transfection normal and transforming PCPH proteins. (1, 2). We further demonstrated that mt-PCPH synergizes with the We report here the expression of the PCPH and mt-PCPH proteins human H-ras oncogene in transforming NIH3T3 cells (2), and that the in Escherichia coli and the finding that the recombinant proteins have PCPH proto-oncogene (termed PCPH here) is highly conserved in eukaryotic cells, from yeast to humans, and it is expressed in most GDPase activity. However, expression of the Syrian hamster PCPH normal adult tissues in Syrian hamsters, mice, and humans (3–5), and mt-PCPH proteins in yeast strains engineered to be GDPase suggesting that the normal PCPH protein may have an important deficient by targeted disruption of the single GDA1 allele did not cellular function. Recently, we isolated full-length cDNA clones of complement their glycosylation-disabled phenotype, suggesting the the Syrian hamster mt-PCPH and PCPH, determined their nucleotide existence of significant functional differences between the mamma- sequence, and demonstrated that mt-PCPH was activated by a point lian and yeast enzymes. Results indicate that, in mammalian cells, mutational deletion (6) within the coding region of the proto- both PCPH and mt-PCPH cause an overall down-regulation of the oncogene. This single bp change causes both a shift in the normal Ras/MEK signaling pathway. Despite this, mt-PCPH, but not PCPH, activated ERK1 but not ERK2 when cotransfected with an activated Received 9/22/99; accepted 1/18/00. human H-ras oncogene. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by USPHS Grant CA64472 from the National Cancer Institute and by 3 The abbreviations used are: ORF, open reading frame; ERK, extracellular signal- USPHS Grant P30-CA51008. regulated kinase; MEK, mitogen-activated protein kinase/ERK; EGF, epidermal growth 2 To whom requests for reprints should be addressed, at Department of Radiation factor; GDPase, guanosine diphosphatase; IPTG, isopropyl-␤-D-thiogalactopyranoside; Medicine, Georgetown University Medical Center, Research Building, Room E215, 3970 MBP, myelin basic protein; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HA, Reservoir Road, NW, Washington, DC 20007. Phone: (202) 687-2102; Fax: (202) 687- hemagglutinin; AP-1, activating protein-1; SRE, serum response element; PKC, protein 2221; E-mail: [email protected]. kinase C. 1720

MATERIALS AND METHODS the color of the lysate on the absorbance measurements. Reactions were performed in the presence of L-[35S]methionine (Amersham, Arlington Plasmids and General Methods. The cytomegalovirus-based vector Heights, IL), as described previously (14). Radiolabeled translation products pcDNA3 (Invitrogen, Carlsbad, CA) containing the cDNA inserts (6) of PCPH were resolved by SDS-PAGE (12.5%), and dried gels were exposed to X-ray (pcDNA3-PCPH) and mt-PCPH (pcDNA3-mt-PCPH) was used for the ex- films. pression of the PCPH and mt-PCPH proteins in mammalian cells, using Construction of Bacterial Expression Vectors. The inducible pET- insertless pcDNA-3 as control. Expression of oncogenic Ras and Raf was 30a(ϩ) bacterial expression vector (Novagen, Madison, WI) was used to Val12 accomplished using constitutive vectors with Ha-ras (pRasV12) and v-raf subclone the inserts from the Syrian hamster PCPH and mt-PCPH cDNA (pvRaf) under the transcriptional control of the Rous sarcoma virus promoter clones (6). The PCPH and mt-PCPH inserts were excised from the original (8). Luciferase reporter vectors pAP1-Luc and pSRE-Luc were obtained from cDNA cloning vector by cleavage with BspHI and XhoI, and pET-30a(ϩ) was Stratagene Cloning Systems (La Jolla, CA). Protein determinations were done linearized by cleavage with NcoI and XhoI. The unique BspHI in the PCPH using the BCA Protein Assay System (Pierce, Rockford, IL). Luciferase and mt-PCPH sequences is located at the ATG codon in position 13. Because determinations were carried out 48 h after cotransfections with pAP1-Luc and BspHI and NcoI generate compatible 5Ј overhangs, ligation of the PCPH pSRE-Luc using the Promega (Madison, WI) assay system, with the help of a inserts with the linearized pET-30a(ϩ) vector allowed directional cloning and Lumat LB9501 luminometer (Berthold Analytical Instruments, EG&G Wallac, regenerated the original ATG codon 13, keeping the PCPH or mt-PCPH Gaithersburg, MD). sequences in-frame with the 5Ј sequences of the pET-30a(ϩ) vector (Fig. 3A). Mammalian Cells, Culture, and Transfection Conditions. Normal Thus, both PCPH and mt-PCPH expression constructs lacked the first 12 (84-3) and neoplastic (81C39) Syrian hamster cells, murine NIH3T3 fibro- codons of the PCPH or mt-PCPH sequence and encoded PCPH and mt-PCPH blasts, PCPH-derived stable NIH3T3 transfectants, and mt-PCPH-derived recombinant proteins containing a poly-His tag at their NH2 termini. Addi- NIH3T3 transformants were cultured as described (6). Human kidney 293T tional expression constructs encoding full-length, poly-His tagged PCPH and cells were maintained in DMEM supplemented with 10% fetal bovine serum. mt-PCPH proteins were generated by cloning the PCR-isolated, entire The calcium-phosphate precipitation technique (9) was used for the generation ORFs of PCPH and mt-PCPH into the appropriate restriction sites of the of stable NIH3T3 transfectants. Transient transfections of NIH3T3 and COS-7 pET-30a(ϩ) vector. Restriction endonucleases were from New England cells were carried out with up to 5 ␮g of DNA/plate using SuperFect (Qiagen, Biolabs (Beverly, MA). Inc., Valencia, CA) as recommended by the manufacturers. Transient trans- Induction and Purification of Bacterial Recombinant Proteins. E. coli fections of 293T cells were carried out using Lipofectamine (Life Technolo- BL21 (DE3) cells were transformed with the pET-30(ϩ)-PCPH or pET- gies, Inc., Gaithersburg, MD) with up to 10 ␮g of DNA/plate. All transfections 30(ϩ)-mt-PCPH constructs, and expression of the recombinant PCPH and 5 were performed in six-well plates containing 3.5 ϫ 10 cells/well. To control mt-PCPH proteins was induced by addition of 1 mM IPTG for various periods for transfection efficiency, we used cotransfection with pSV-␤-galactosidase of time. For the extraction of the recombinant proteins, induced cells were (10), and data normalizations after all transient transfections were carried out processed following the instructions recommended in the His-Bind Buffer kit using triplicate cultures, repeating every cotransfection protocol at least three (Novagen), but 6 M urea was added during the extraction because both proteins times and keeping the total amount of DNA constant by the addition of fractionated with inclusion bodies. Recombinant proteins were purified by appropriate amounts of empty pcDNA-3 vector DNA. one-step Ni2ϩ chelation chromatography on His-Bind Resin (Novagen), and GDPase Activity Assays. Precautions were taken to avoid contamination the purified proteins were then step-wise dialyzed against a solution containing with inorganic phosphate, a nucleotide phosphatase inhibitor, in all manipu- TBS (pH 7.4), 0.1% Triton X-100, 5% glycerol, and progressively decreasing lations. Cells were washed three times in TBS [25 mM Tris-HCl (pH 7.5), 140 concentrations of urea (4, 2, 1, and 0.5 M) and without urea in the final step. mM NaCl], collected by trypsinization, and resuspended in TBS. Cell extracts After dialysis the protein concentrations were determined, and the preparations were prepared by sonication for 1 min (three 20-s treatments, with 1-min were aliquoted and frozen at Ϫ70°C. cooling intervals), followed by addition of Triton X-100 to a final concentra- Yeast Transformation, Northern, Western, and Glycosylation Analysis. tion of 1.5% and incubation at room temperature for 30 min. Three different The Syrian hamster PCPH and mt-PCPH cDNA inserts (6) were directionally methods were used to assay GDPase activity. The standard GDPase assays subcloned into the EcoRI and XhoI sites of the Saccharomyces cerevisiae were carried using a modification of the method described by Abeijon et al. shuttle expression vectors p416GPD and p426GPD (15). Transcription of (11). Briefly, incubation mixtures contained, in a final volume of 50 ␮l, 0.75 sequences inserted in these vectors is driven by the GAPDH promoter. The

M Tris-HCl (pH 7.5), 0.1 M CaCl2, 0.1 M MgCl2, 1% Triton X-100, 2 mM GDP presence of a CEN6/ARS4 replicon makes p416GPD behave as a low copy (Sigma Chemical Co., St. Louis, MO), and 20 ␮g of cell extracts or purified number plasmid in S. cerevisiae, whereas p426GPD carries a yeast 2-␮m recombinant proteins. Incubations were done for 20 min at 37°C and stopped replicon and behaves as a high copy number plasmid. The PCPH and mt- by the addition of SDS to a final concentration of 2%, and the inorganic PCPH recombinant constructs were transformed as described (16) into haploid phosphate released was determined from the absorbance at 820 nm, as de- S. cerevisiae strains containing wild-type or disrupted alleles of GDA1, the scribed (12). In-gel GDPase activity assays were performed by native gel GDPase encoding gene. Yeast strains and plasmids carrying the wild-type electrophoresis of the extracts, followed by an incubation of the gels for 20 min (p13H) or the disrupted (pGD⌬1) GDPase genes, generously provided by Dr. at 37°C in a solution of 0.2 M imidazole buffer (pH 7.6), containing 20 mM C. B. Hirschberg (Boston University, Boston, MA) (17), were used for positive

GDP, 0.1 M CaCl2, and 3.5 mM Pb(NO3)2, and staining for inorganic phosphate and negative transformation controls, respectively. Expression of PCPH and as described (13). Alternatively, GDPase was determined as described above mt-PCPH was ascertained by the level of mRNA and protein extracted from but using 20 ␮M GDP and 1 ␮Ci of [8,5Ј-3H]GDP (25–50 Ci/mmol; DuPont exponentially growing cells. Conditions for RNA extraction, Northern analy- NEN, Wilmington, DE). After incubation, one-tenth of the reaction mixtures sis, and protein extraction were essentially as described (18). The full-length was loaded onto Cellulose MN300 polyethyleneimine TLC plates impregnated Syrian hamster PCPH cDNA insert was used as the probe for the Northern with fluorescent indicator (Macherey-Nagel Gmbh & Co., Du¨ren, Germany), analyses. Conditions for Western analyses are given below. Antibodies used and the nucleotides were separated by ascending chromatography in 0.75 M were: a rabbit anti-Gda1p antiserum (72-515) provided by Dr. Hirschberg (19), 3 KH2PO4 (pH 3.5). Then the TLC plates were sprayed with EN HANCE and a polyclonal antiserum (#367–10W) raised in rabbits against bacterial- (DuPont NEN) and exposed to X-OMAT AR films (Eastman Kodak, Roch- recombinant Syrian hamster PCPH protein purified to near homogeneity (Fig. ester, NY). These three methods were also used to determine the phosphatase 3B). The glycosylation status was monitored by studying the electrophoretic activity on UDP and IDP. mobility of purified extracellular chitinase as described (17). In Vitro Transcription/Translation. A one-reaction system was used for Western Analysis. Mammalian cells were scraped into harvest buffer [20 the simultaneous transcription and translation of the PCPH and mt-PCPH mM Tris-HCl (pH 7.5), 20 mM p-nitrophenyl phosphate, 1 mM EGTA, 50 mM proteins from the original cDNA clones (6). The system uses T7 RNA sodium fluoride, 50 ␮M sodium orthovanadate, and 5 mM benzamidine] and polymerase for transcription and a rabbit reticulocyte extract for in vitro sonicated for 5 s. Yeast cells were resuspended in harvest buffer and broken translation (TNT; Promega). A wheat-germ extract was used instead of the with glass beads by vortexing as described above. Cell extracts (30–50 ␮gof rabbit reticulocyte extract whenever GDPase activity assays were based on the protein) were subjected to SDS-PAGE on 4–15% gradient gels (20) and colorimetric determination of inorganic phosphate to avoid interference from blotted onto polyvinylidene difluoride membranes by electrotransfer, and the 1721

Fig. 1. Increased GDPase activity in extracts of cells expressing PCPH or mt-PCPH. A, homology between PCPH and the yeast GDPase, Gda1p. B, in-gel detection of GDPase activity in neoplastic, mt-PCPH-expressing Syrian hamster 81C39 cells relative to normal 84-3 cells; S and SF, medium with serum or serum free, respectively; experimental details are given in the text. C, nucleotide diphosphatase (NPase) activity on UDP, GDP, and IDP in extracts of mouse NIH3T3 fibroblasts transformed by mt-PCPH (379) and transfected with PCPH (389) or with empty pCDNA3 vector DNA (391).

membranes were probed with primary antibodies: polyclonal anti-MEK1/2, brief spin (22). The detergent phase was diluted 10-fold with lysis buffer anti-phospho-MEK1/2, anti-phospho-ERK1/2 antibodies (New England Bio- without Triton X-114. The lysate was precleared for 5 min with protein labs), or anti-GAPDH (Trevigen, Inc., Gaithersburg, MD). Visualization of G-Sepharose beads and further incubation with agarose-conjugated anti-p21Ras immunoreactive polypeptides was accomplished by using a peroxidase-conju- monoclonal Y13–259 (Calbiochem-Novachem, La Jolla, CA). Immunopre- gated secondary antibody and development with chemiluminescence (ECL; cipitates were collected and washed eight times with 50 mM HEPES buffer (pH

Amersham Pharmacia Biotech, Inc., Piscataway, NJ). Linearity in the densi- 7.4), 50 mM NaCl, 5 mM MgCl2, 0.1% Triton X-100, and 0.5% SDS. Radio- tometric data analysis was controlled by scanning films corresponding to labeled GDP and GTP were eluted in 2 mM EDTA, 2 mM DTT, 0.2% SDS, 0.5 various developing times for each set of experimental data and matching mM cold GDP, and 0.5 mM cold GTP at 68°C for 20 min and separated on

GAPDH loading controls. Densitometry was performed using the NIH Image polyethyleneimine-cellulose plates developed in 0.75 M KH2PO4 (pH 3.5). 1.61 software. Plates were autoradiographed, and the GDP:GTP ratio was determined by Determination of GDP:GTP Bound to RAS. The ratio of GDP:GTP densitometry, as described above. bound to p21Ras was determined essentially as described (21). Cells were Mitogen-activated Protein Kinase Assays. 293T cells at 60–80% con- ␮ 32 ␮ ␮ serum starved for 18 h and subsequently labeled for 4 h with 300 Ci of [ P]Pi fluence were transfected with 1 g of pcDNA-HA-Erk1 and 3 g of pcDNA3- per 100-mm dish in phosphate-free medium (Life Technologies, Inc.). Cells PCPH or pcDNA3-mt-PCPH. At 12 h after transfection, the cells were washed were then stimulated with EGF (50 ng/ml) for 5 min, after which they were put with PBS and incubated for another 12 h in serum-free medium. Then the cells on ice, washed rapidly with ice-cold TBS, and lysed in 50 mM HEPES (pH were lysed in 20 mM HEPES, 10 mM EGTA, 40 mM ␤-glycerophosphate, 1%

7.4), 1% Triton X-114, 100 mM NaCl, 5 mM MgCl2,1mM BSA, 10 mM NP40, 2.5 mM MgCl2,2mM sodium orthovanadate, 1 mM DTT, 1 mM benzamidine, 10 ␮g/ml leupeptin, 10 ␮g/ml aprotinin, 10 ␮M GTP, 10 ␮M phenylmethylsulfonyl fluoride, 20 ␮g/ml aprotinin, and 20 ␮g/ml leupeptin. GDP, 1 mM ATP, and 1 mM sodium phosphate, included to prevent postlysis Lysates were centrifuged, and supernatants were incubated with 2 ␮gof labeling of p21ras. Nuclei were removed by centrifugation, and the Triton anti-HA monoclonal antibody 12CA5 (Babco, Richmond, CA) for1hat4°C. X-114 and aqueous phases were separated at 37°C for 2 min, followed by a Protein G-Sepharose (20 ␮l; Amersham Pharmacia Biotech) was used to 1722

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 2000 American Association for Cancer Research. PCPH ONCOPROTEIN AND ERK1 ACTIVATION recover the immunoprecipitates by centrifugation. Pellets were washed three times with PBS containing 1% NP40 and 2 mM sodium orthovanadate, once with 100 mM Tris (pH 7.5) and 0.5 M LiCl, and once with kinase reaction buffer [12.5 mM MOPS (pH 7.5), 12.5 mM ␤-glycerophosphate, 7.5 mM

MgCl2, 0.5 mM EGTA, 0.5 mM sodium orthovanadate, and 0.5 mM NaF]. Beads were resuspended in a 30-␮l final volume containing, per reaction, 1 ␮Ci of [␥-32P]ATP, 20 ␮M cold ATP, 3.3 mM DTT, and 1.5 mg/ml of MBP (Sigma Chemical Co.) and incubated for 30 min at 30°C. Samples were run in 4–15% SDS-PAGE, blotted, and exposed. As a loading control, blots were incubated with the anti-HA 12CA5 monoclonal antibody and developed.

RESULTS Increased GDPase Activity in Extracts of Cells Expressing mt-PCPH. In addition to their partial homology to GTP/GDP exchange factors (6), the PCPH and mt-PCPH proteins share some homology with members of the family of nucleotide phosphohydro- lases, a rather heterogeneous group of proteins with diverse substrate specificities and cleavage patterns (23, 24). On the basis of this homology and of our previous finding that the PCPH and mt-PCPH proteins showed particularly high affinity for GDP in nucleotide binding assays (6), it seemed possible that the PCPH and mt-PCPH proteins may have GDPase activity. This notion was strengthened by the existence of a 54.7% homology (32.4% identity) between the PCPH protein and Gda1p, the S. cerevisiae GDPase (Fig. 1A). As a first approximation to explore whether the PCPH and mt-PCPH proteins had GDPase activity, we used in-gel activity assays (13) to determine the levels of GDPase activity in total cell extracts of normal Syrian hamster embryo 84-3 cells and immortal mouse NIH3T3 fibroblasts and compared them with Syrian hamster and murine cells that expressed a mutated PCPH allele, either as a result of chemical carcinogenic treatment (81C39 cells) or by gene transfer. Results (Fig. 1B) showed that: (a) the GDPase activity detected in total extracts of 81C39 cells was greater than that detectable in extracts of normal 84-3 cells; (b) comigrated with phosphohydrolase activities on UDP and IDP, also known as possible GDPase (EC 3.6.1.42) substrates; and (c) could be further increased when both 84-3, and especially 81C39 cells, were incubated in serum-free medium (Fig. 1B), conditions described previously (6) to cause an increase of mt-PCPH mRNA in these cells. Similar results were obtained on determinations of the relative hydrolytic activity on GDP, UDP, and IDP (Fig. 1C)in extracts from stable mt-PCPH-transformed NIH3T3 cells (379) and from NIH3T3 cells stably transfected with PCPH (389), relative to Fig. 2. Increased GDPase activity after in vitro translation of PCPH and mt-PCPH. A, relative GDPase activity in rabbit reticulocyte (RR) and wheat-germ (WG) lysates used for cells transfected with empty vector DNA (391). These data repre- the coupled transcription and translation of PCPH and mt-PCPH. GDPase was measured sented an excellent correlation between the expression of the mt- by the colorimetric determination of inorganic phosphate released from GDP. B, autora- PCPH and PCPH proteins and the detection of GDPase activity in diographic confirmation of the synthesis of PCPH and mt-PCPH in vitro. The data shown here correspond to the wheat-germ lysate presented in A. Reactions were carried out with extracts of two different cell types, further strengthening our notion linearized DNA of either empty pcDNA3 vector (Lane 1), the PCPH construct (Lane 2), 35 that the PCPH and mt-PCPH proteins may have GDPase activity. or the mt-PCPH construct (Lane 3) in the presence of L-[ S]methionine. Increased GDPase Activity after in Vitro Transcription/Trans- lation of PCPH and mt-PCPH. To rule out the possibility that the increased GDPase activity detected in mt-PCPH-orPCPH-express- translate mt-PCPH was greater than the GDPase activity in lysates ing cells was not attributable to the PCPH or mt-PCPH intrinsic used to synthesize the PCPH protein. Similar results were observed GDPase activity but resulted as a secondary effect of the transforma- with rabbit reticulocyte and wheat-germ lysates (Fig. 2A). No GDPase tion process or the overexpression of the PCPH protein on a distinct activity could be detected when the extracts containing the in vitro endogenous gene product, we next synthesized the PCPH and mt- translated proteins were boiled prior to the GDPase assays. These PCPH proteins in vitro using coupled transcription/translation assays results strongly suggested that the increased GDPase activity in rabbit and determined the levels of GDPase activity in the extracts before reticulocyte and wheat-germ cell lysates was attributable to the trans- and after translation. Plasmids containing the full-length coding re- lation of the PCPH and mt-PCPH proteins. gions of PCPH or mt-PCPH under the transcriptional control of the Expression of the PCPH and mt-PCPH Proteins in Bacteria. To bacteriophage T7 RNA polymerase promoter and empty vector DNAs rule out the possibility that the PCPH proteins had no intrinsic were used as templates. There was a significant increase in GDPase GDPase activity but were activating a latent GDPase preexisting in activity (Fig. 2A) after translation of both PCPH and mt-PCPH pro- the rabbit reticulocyte and wheat-germ lysates and to generate high teins (Fig. 2B). Although these assays are not quantitative, we con- yields of the PCPH and mt-PCPH proteins for further characterization, sistently observed that the GDPase activity in the lysates used to we expressed them in E. coli. Inducible bacterial expression vectors 1723

Fig. 3. Expression of PCPH and mt-PCPH in E. coli. A, scheme of the construction of the bacterial expression vectors. B, induction and purification of bacterial recombinant PCPH and mt-PCPH: SDS-PAGE analysis of extracts from bacteria transformed with empty vector DNA (control), the PCPH construct or the mt-PCPH construct, uninduced (U) or induced (I) by addition of IPTG for1htothemedium. Arrows, the mobility of recombinant PCPH and mt-PCPH before (left panel) and after (right panel) purification by Ni2ϩ chelation chromatography and refolding. C, TLC determination of the GDPase activity of purified recombinant (RP) PCPH and mt-PCPH. Controls included reactions without enzyme (C), with total extracts (TCE) of IPTG-induced bacteria expressing PCPH (shown here) or mt-PCPH (data not shown), and with heat-denatured (HD) recombinant proteins. The migration of the substrate (GDP) and the reaction product (GMP) is indicated. contained most of the coding regions of the PCPH and mt-PCPH PCPH and mt-PCPH Proteins Do Not Complement S. cerevisiae cDNAs (from amino acid 13 to the end of the ORFs), fused for ⌬gda1 Stains. Given the existence of significant sequence identity purification purposes to a poly-His tag at their NH2 termini under the between the PCPH protein and the yeast GDPase, the product encoded control of the bacteriophage T7 promoter and lac operator (Fig. 3A). by the GDA1 gene of S. cerevisiae (Fig. 1A), and our observation that The addition of IPTG to the culture media resulted in the accumula- the PCPH and mt-PCPH proteins have GDPase activity (Fig. 3), we tion in the cells of polypeptides with sizes corresponding to those examined whether the PCPH proteins shared functional mechanisms expected for the proteins encoded by the ORFs in the expression with the yeast Gda1p by expressing the mammalian proteins in ⌬gda1 vectors (Fig. 3B). Subcellular fractionation of bacteria containing disruptant strains and testing whether they could complement their PCPH or mt-PCPH revealed that both proteins were associated with reported glycosylation defects (17). As a glycosylation-dependent end inclusion bodies (data not shown). Recombinant His-tagged proteins point, we studied the relative electrophoretic mobility of chitinase in were solubilized with 6 M urea, purified by metal chelation chroma- ⌬gda1 yeast cells transformed with the PCPH or mt-PCPH cDNAs in tography, and refolded by slow, step-wise dialysis in solutions of comparison with untransformed ⌬gda1 cells. Results (data not shown) progressively decreasing urea concentrations. The purified recombi- indicated that the expression of the PCPH or mt-PCPH proteins did nant proteins showed the expected mobility in SDS-PAGE (Fig. 3B). not recover the mobility shift of chitinase observed in ⌬gda1 cells Aliquots (20 ␮g) of the purified recombinant proteins were used for (G2-7) when compared with wild-type cells (G2-5). However, expres- GDPase assays in which the GDP phosphohydrolase activity was sion of the wild-type GDA1 gene resulted in an almost complete monitored by the conversion of [3H]GDP into GMP. Results (Fig. 3C) recovery of the mobility shift. This lack of complementation by the demonstrated that both PCPH and mt-PCPH proteins have intrinsic PCPH and mt-PCPH proteins was observed regardless of whether high or GDPase activity, and that the activity is lost when the purified proteins low copy number plasmid vectors (15) were used for their expression. were subject to heat denaturation prior to incubation with radiolabeled These results strongly suggested that the primary function of the PCPH GDP. No difference was observed between the specific activities of and mt-PCPH proteins in mammalian cells is likely to be unrelated to purified recombinant PCPH and mt-PCPH. glycosylation, the well-established function of the yeast GDPase (17). 1724

cules remained in the inactive state in response to EGF. Thus, mt- PCPH appeared to act as a negative regulator of Ras activity. PCPH and mt-PCPH Behave as Negative Regulators of the Ras Signaling Pathway. To assess whether the negative effect on Ras activation had any consequences on the Ras signaling pathway as a whole, we transiently cotransfected the PCPH or mt-PCPH cDNAs into NIH3T3 (Fig. 5, A and B) and COS-7 cells (Fig. 5, C and D) with reporter plasmids in which luciferase expression is regulated by the presence of the AP-1 or SRE transcriptional enhancer elements. Transactivation of the AP-1 and SRE binding sites in the reporter vectors was induced by including vectors for the expression of oncogenic Ras (RasV12; Fig. 5, A and C) or Raf (v-raf; Fig. 5, B and D) in some of the cotransfections. Results (Fig. 5) demonstrated that neither PCPH nor mt-PCPH by themselves had any significant effect on the transactivation of AP-1 or SRE elements in either cell type. As expected, expression of either RasV12 or Raf by themselves resulted in considerable increases in AP-1 and SRE transactivation in NIH3T3 and COS-7 cells. However, when the cells were cotransfected with RasV12 or Raf and PCPH or mt-PCPH, the levels of transactivation of both AP-1 and SRE elements were clearly reduced (between 27.4 and 62.5% for AP-1, and between 34.3 and 86.5% for SRE) in both cell types. These data suggested that both PCPH and mt-PCPH were acting as general negative regulators of the Ras/Raf signaling pathway. To further characterize the effect of PCPH and mt-PCPH on Ras signaling, we next examined the level and phosphorylation status of other downstream components of the pathway in 293T cells transiently cotransfected to express RasV12 and PCPH or mt-PCPH. As shown in Fig. 6A, expression of RasV12 by itself resulted in a dramatic increase in the level of phosphorylation of MEK1/2, but this increase was reduced by the coexpression of either PCPH (nearly 20% inhibition) or mt-PCPH (ϳ37% inhibition). However, neither PCPH nor mt-PCPH altered the total level of MEK1/2 protein significantly, indicating that the changes observed in MEK1/2 phosphorylation were not attributable to protein down-regulation. These results supported a negative regulatory role of the PCPH and mt-PCPH proteins on the Ras/MEK signaling pathway. Differential Effect of PCPH and mt-PCPH on ERK1 Activa- Fig. 4. Effect of mt-PCPH on the GDP:GTP bound to Ras. Untransfected and tion. Because MEK1/2 phosphorylation was down-regulated in the 32 mt-PCPH transfected human 293T cells were radiolabeled with [ P]Pi for 4 h, stimulated presence of PCPH or mt-PCPH, we examined the phosphorylation ras with EGF for 5 min, and the GDP and GTP bound to p21 were extracted and resolved status of ERK1 and ERK2, which are MEK1/2 substrates. To this end, by TLC (A). Untransfected, unstimulated cells were used as control. Densitometric analysis (B) was used to determine the GDP:GTP ratios shown at the bottom. 293T cells were transiently cotransfected with pRasV12 and either pcDNA3-PCPH or pcDNA3-mt-PCPH, and the results were compared with those from transfections with pRasV12, pcDNA3-PCPH, Transient Expression of mt-PCPH Modulates Ras GDP/GTP or pcDNA3-mt-PCPH alone. Western analyses with anti-phospho- Levels in Mammalian cells. Previous results from our laboratory ERK1/2 showed (Fig. 6B) that neither PCPH nor mt-PCPH alone had demonstrated that, in stable cotransfection experiments, mt-PCPH any significant effect on ERK1/2 phosphorylation, whereas RasV12 synergized with the human H-ras oncogene to transform NIH3T3 increased ERK1 phosphorylation by ϳ2-fold. However, when fibroblasts (2). More recently, we also showed that the PCPH and RasV12 was coexpressed with PCPH, the level of phosphorylated mt-PCPH products are nucleotide-binding proteins with special affin- ERK1 decreased by almost 20%, relative to cells expressing RasV12 ity for GDP in vitro (6). These data strongly suggested that PCPH and alone. Most surprisingly, when RasV12 was coexpressed with mt- mt-PCPH proteins may interact directly or indirectly with Ras signal- PCPH, the level of phosphorylated ERK1 was almost doubled relative ing when expressed in mammalian cells. To test this possibility, we to the induction caused by RasV12 alone, whereas the phosphoryla- examined whether the expression of mt-PCPH would alter the ratio of tion levels of ERK2 did not change under our experimental condi- GDP:GTP bound to the endogenous Ras protein in NIH3T3 and 293T tions. These results suggested that there was a synergism between Ras cells after EGF stimulation. Results were similar for the two cell lines and mt-PCPH converging specifically on ERK1 and represented a and showed (Fig. 4A) that, as expected, EGF treatment caused an clear difference between the mechanisms of action of PCPH and increase in the amount of GTP bound to Ras relative to unstimulated mt-PCPH. controls, whereas the relative amount of Ras-bound GDP did not To confirm that the increased phosphorylation of ERK1 was indeed change significantly. Accordingly, the GDP:GTP ratio cells went reflected on its kinase activity, we transiently cotransfected 293T cells from 7.8 to 1.8 (Fig. 4B). However, in EGF-stimulated cells express- to express either HA-ERK1 and RasV12, or HA-ERK1 and RasV12 ing mt-PCPH, the GDP:GTP ratio was 2.7. These results indicated plus either PCPH or mt-PCPH. Transfected cells were lysed 48 h after that, in the presence of mt-PCPH, a greater proportion of Ras mole- transfection, the ERK1 protein was immunoprecipitated with an an- 1725

Fig. 5. Effect of PCPH and mt-PCPH on AP-1 and SRE transactivation by Ras (A and C) or Raf (B and D). NIH3T3 (A and B) and COS-7 (C and D) cells were cotransfected with RasV12 or v-raf plus either PCPH or mt-PCPH, and their effect on AP-1 and SRE transactivation was assessed by comparison with cells transfected with Ras, Raf, PCPH, or mt-PCPH individually. Background AP-1 and SRE activities were established from cells transfected with empty pcDNA3 vector DNA alone. The extent of transactivation was determined by the expression of luciferase activity from AP-1- and SRE-luciferase reporter constructs introduced in the various cellular backgrounds. Bars, SD. ti-HA monoclonal antibody, and immune-complex kinase assays were sylation-deficient phenotype of S. cerevisiae strains lacking the performed using MBP as the substrate. Results (Fig. 6C) showed that, GDA1-encoded GDPase does not detract from our definition of PCPH when normalized by the amount of immunoprecipitable ERK1 in each and mt-PCPH as GDPases. There are well-documented cases of lack lysate, the maximum ERK1 kinase activity was detected in cells of complementation in S. cerevisiae by the expression of highly coexpressing RasV12 and the mt-PCPH protein (ϳ33% greater than conserved mammalian proteins (25). In some cases, proteins that are in the case of RasV12 alone). On the contrary, coexpression of PCPH extremely well conserved from a structural point of view between resulted in an almost 20% decrease of the ERK1 kinase activity yeasts and humans do not have identical functions in S. cerevisiae and stimulation caused by RasV12 alone. These results further supported mammalian cells. For instance, the Ras proteins function in yeasts to the notion that the ras and mt-PCPH oncogenes cooperate for the integrate metabolism and control of the cell cycle by modulating activation of ERK1, despite the general down-regulatory role of adenylate cyclase (26, 27), whereas in mammalian cells Ras activity mt-PCPH on the Ras-signaling pathway. In addition, these data sug- seems to be largely independent of adenylate cyclase (28). Thus, it is gest that mt-PCPH may induce ERK1 activation by a mechanism very likely that the primary function of PCPH and mt-PCPH in involving a kinase different from MEK1/2. mammalian cells may not be related to the control of glycosylation. Although our preliminary results seem to confirm this notion, the DISCUSSION possibility that PCPH and mt-PCPH may be minor contributors to the This paper reports that the PCPH and mt-PCPH proteins are GD- glycosylation processes cannot be ruled out. In this regard, PCPH and Pases. Four lines of evidence support this conclusion: (a) the amino mt-PCPH might still be functionally comparable in mammalian cells acid sequences of PCPH and mt-PCPH are significantly homologous to the yeast Ynd1p, a second protein with GDPase activity described to Gda1p, the yeast GDPase (Ref. 17; Fig. 1A); (b) there was a clear recently (29) as being responsible for the residual GDPase activity correlation between the expression of mt-PCPH and GDPase activity detectable in ⌬Gda1 strains, which is required for O-glycosylation. in carcinogen-initiated cells and cells transformed by mt-PCPH trans- Our finding of the involvement of PCPH and mt-PCPH in the fection (Fig. 1, B and C); (c) GDPase activity was detected in rabbit regulation of the Ras-signaling pathway provides additional evidence reticulocyte and wheat-germ extracts after in vitro translation of indicating that glycosylation is probably not their primary function. PCPH or mt-PCPH (Fig. 2); and (d) highly purified bacterial recom- Our results conclusively indicate that both PCPH and mt-PCPH act as binant PCPH and mt-PCPH proteins had GDPase activity (Fig. 3). negative regulators of Ras signaling by interfering with the activation The lack of complementation by PCPH or mt-PCPH of the glyco- of several key early and late components of the pathway. The gener- 1726

ality of this effect is evidenced by the fact that similar results were obtained with different cell types. Down-regulation of the Ras signaling pathway (Figs. 5 and 6) may be an important mechanism by which PCPH contributes to the regulation of normal cell proliferation. Our laboratory has reported (5, 7) and continues to accumulate data indicating that PCPH is frequently lost in various types of human tumor cells. In agreement with these observations, the loss of a negative regulator of Ras signaling, such as normal PCPH, would result in an up-regulation of the Ras pathway and, in turn, contribute to the development of the neoplastic phenotype. The overall negative regulatory effect of mt-PCPH on Ras signaling (Figs. 4, 5, and 6A) may seem to contradict our previous finding that mt-PCPH and the human H-ras oncogene had a synergistic effect in the transformation of NIH3T3 cells (2). However, the cooperation between mt-PCPH and Ras may be explained by their convergence on ERK1 activation to a level greater than the addition of the increases induced by Ras and mt-PCPH individually (Fig. 6, B and C). Never- theless, the simultaneous down-regulation of upstream components of the Ras-signaling pathway and the activation of ERK1 by mt-PCPH raise the question of what is the MEK1/2-independent mechanism of ERK1 activation in cells coexpressing Ras and mt-PCPH. There are numerous reports in the literature describing various degrees of involvement of Ras, Raf, or MEK on ERK1/2 activation in different experimental systems: from Ras dependent (30, 31) to Ras independent (31–33); or Ras and Raf independent (34–36); and MEK dependent (32) or independent (37). The conclusion from all of these data is that EKR1/2 activation can be accomplished via mechanisms not involving components of the Ras-signaling pathway. Several mechanisms have been proposed as alternatives to the Ras/MEK pathway for ERK1/2 activation, including the inhibition of the VHR or MKP-1 phosphatases for which ERK1/2 are substrates (38, 39), tyrosine kinases, phosphatidylinositol 3-kinase, PKC iso- forms, phospholipase C (40), or Ca2ϩ store depletion (41). But in the majority of the cases, ERK1/2 activation has been reported to be mediated by PKC (31, 32, 42–44). This is important, because PKC has been found to be responsible for a slow and sustained activation of ERK1/2 that lasts for days, whereas Ras has been implicated in an immediate, transient activation of ERK that lasts only minutes (44). In our case, we detected the sustained type of ERK1 activation, at least up to 72 h after transfection, in cells coexpressing Ras and mt-PCPH. Therefore, these results would suggest that, under our experimental conditions, ERK1 activation may have been mediated by a PKC- dependent mechanism. However, the possible involvement of PKC in the activation of ERK1 in cells coexpressing RasV12 and mt-PCPH remains to be elucidated. The ability to cause a sustained activation of ERK1 may be an important determinant of the transforming activity of mt-PCPH. This suggestion is based on two lines of evidence: (a) the nontransforming PCPH lacks the ability to induce ERK1 activation; and (b) it has been shown in several systems that sustained ERK1/2 activation is neces- Fig. 6. Effect of PCPH and mt-PCPH on the activation of MEK1/2 and ERK1/2 by sary for transformation (35, 45, 46) through alterations in cell cycle oncogenic Ras. 293T cells (A and B) or NIH3T3 (C) cells were cotransfected with RasV12 (45, 47, 48), cytoskeleton (49), and the transcriptional machinery of plus PCPH or mt-PCPH, and their effect on MEK1/2 (A) or ERK1/2 (B and C) activation was assessed by Western blot analysis in comparison with cells transfected with Ras, the cells (47, 50). Therefore, ongoing studies are designed to establish PCPH, or mt-PCPH individually and relative to background control transfections with the mechanism by which mt-PCPH expression results in ERK1 acti- pcDNA3 vector DNA alone. PCPH and mt-PCPH expression was confirmed using an vation and its contribution to the mt-PCPH transforming activity. antibacterial recombinant PCPH rabbit antiserum. MEK1/2 and ERK1/2 activation was assessed with antibodies specific for their phosphorylated forms. Loading was controlled with anti-MEK1/2 (A) or anti-GAPDH (B) antibodies. Changes in activated ERK1 are shown at the bottom and were determined by densitometry. ERK1 kinase activity in NIH3T3 cells (C) was determined on MBP using immunoprecipitates of HA-tagged ACKNOWLEDGMENTS ERK1 protein, relative to the total content of ERK1 as determined by Western analysis with an anti-HA antibody. Quantitative data indicated below each lane were normalized Densitometry was performed at the Lombardi Cancer Center’s Macromo- for loading and transfection efficiency. lecular Synthesis and Sequencing Shared Resource. We thank Dr. Carlos B. Hirschberg for the S. cerevisiae strains, recombinant plasmids, and anti-Gda1p antiserum. 1727

REFERENCES 27. Gibbs, J. B., and Marshall, M. S. The ras oncogene–an important regulatory element in lower eucaryotic organisms. Microbiol. Rev., 53: 171–185, 1989. 1. Notario, V., Castro, R., Flessate, D. M., Doniger, J., and DiPaolo, J. A. Frequent 28. Barbacid, M. Ras genes. Annu. Rev. Biochem., 56: 779–827, 1987. activation of non-ras transforming sequences in neoplastic Syrian hamster cells 29. Gao, X-D., Kaigorodov, V., and Jigami, Y. YND1, a homologue of GDA1, encodes initiated with chemical carcinogens. Oncogene, 5: 1425–1430, 1990. membrane-bound apyrase required for Golgi N- and O-glycosylation in Saccharomy- 2. Velasco, J. A., Castro, R., Avila, M. A., Laborda, J., DiPaolo, J. A., Cansado, J., and ces cerevisiae. J. Biol. Chem., 274: 21450–21456, 1999. Notario, V. Cph, a novel oncogene which cooperates with H-ras in the transformation 30. Yue, J., Frey, R. S., and Mulder, K. M. Cross-talk between Smad1 and Ras/MEK of NIH/3T3 fibroblasts. Oncogene, 9: 2065–2069, 1994. signaling pathways for TGF␤. Oncogene, 18: 2033–2037, 1999. 3. Velasco, J. A., Zimonjic, D. B., Popescu, N. C., Cansado, J., DiPaolo, J. A., Albor, 31. Li, X., Lee, J. W., Graves, L. M., and Earp, H. S. Angiotensin II stimulates ERK via A., and Notario, V. Tissue-specific expression, evolutionary conservation and localization of the cph proto-oncogene on Syrian hamster chromosome X. Oncogene, 12: two pathways in epithelial cells: protein kinase C suppresses a G-coupled receptor- 2713–2717, 1996. EGF receptor transactivation pathway. EMBO J., 17: 2574–2583, 1998. 4. Recio, J. A., Zambrano, N., de la Penã, L., Powers, C., Siwarski, D., Huppi, K., and 32. Barrie, A. P., Clohessy, A. M., Buensuceso, C. S., Rogers, M. V., and Allen, J. M. Notario, V. cDNA isolation, expression and chromosomal localization of the mouse Pituitary adenylyl cyclase-activating peptide stimulates extracellular signal-regulated Pcph proto-oncogene. Mol. Carcinog., 26: 130–136, 1999. kinase 1 or 2 (ERK1/2) activity in a Ras-independent, mitogen-activated protein 5. Recio, J. A., Zambrano, N., de la Penã, L., Reig, J. A., Rhoads, A., Rouzaut, A., and kinase/ERK kinase 1 or 2-dependent manner in PC12 cells. J. Biol. Chem., 272: Notario, V. The human PCPH proto-oncogene: cDNA identification, primary struc- 19666–19671, 1997. ture, chromosomal mapping, and expression in normal and tumor cells. Mol. Car- 33. Masuelli, L., and Cutler, M. L. Increased expression of the Ras suppressor RSU-1 cinog., in press, 2000. enhances Erk-2 activation and inhibits Jun kinase activation. Mol. Cell. Biol., 16: 6. Velasco, J. A., Avila, M. A., and Notario, V. The product of the cph oncogene is a 5466–5476, 1996. truncated, nucleotide binding protein that enhances cellular survival to stress. Onco- 34. Graham, T. E., Pfeiffer, J. R., Lee, R. J., Kusewitt, D. F., Martinez, A. M., Foutz, T., gene, 18: 689–701, 1999. Wilson, B. S., and Oliver, J. M. MEK and ERK activation in ras-disabled RBL-2H3 7. Rouzaut, A., Recio, J. A., and Notario, V. Expression of the protein product of the mast cells and novel roles for geranylgeranylated and farnesylated protein in Fc PCPH proto-oncogene in human tumor cell lines. Radiat. Res., in press, 2000. epsilon RI-mediated signaling. J. Immunol., 161: 6733–6744, 1998. ␬ 8. Devary, Y., Rosette, C., DiDonato, J. A., and Karin, M. NF- B activation by 35. Gopalbhai, K., and Meloche, S. Repression of mitogen-activated kinases ERK1/ ultraviolet light not dependent on a nuclear signal. Science (Washington DC), 261: ERK2 activity by a protein tyrosine phosphatase in rat fibroblasts transformed by 1442–1445, 1993. upstream oncoproteins. J. Cell. Physiol., 174: 35–47, 1998. 9. Wigler, M., Pellicer, A., Silverstein, S., Axel, R., Urlaub, G., and Chasin, L. Transfer 36. Csar, X. F., Ward, A. C., Hoffmann, B. W., Guy, G. G., and Hamilton, J. A. cAMP of purified herpes virus thymidine kinase gene to cultured mouse cells. Proc. Natl. suppresses p21 ras and Raf-1 responses but not the Erk-1 response to granulocyte- Acad. Sci. USA, 76: 1373–1376, 1979. colony-stimulating factor: possible Raf-1-independent activation of Erk-1. Biochem. 10. Soldatenkov, V. A., Albor, A., Patel, B. K. R., Dreszer, R., Dritschilo, A., and Notario, V. Regulation of the human poly(ADP-ribose) polymerase promoter by the J., 322: 79–87, 1997. ETS transcription factor. Oncogene, 18: 3954–3962, 1999. 37. Grammar, T. C., and Blenis, J. Evidence for MEK-independent pathways regulating 11. Abeijon, C., Orlean, P., Robbins, P. W., and Hirschberg, C. Topography of glyco- the prolonged activation of ERK-MAP kinases. Oncogene, 14: 1635–1642, 1997. sylation in yeast: characterization of GDP-mannose transport and lumenal guanosine 38. Todd, J. L., Tanner, K. G., and Denu, J. M. Extracellular regulated kinases (ERK) 1 diphosphatase activities in Golgi-like vesicles. Proc. Natl. Acad. Sci. USA, 86: and ERK2 are authentic substrates for the dual-specificity protein-tyrosine phospha- 6935–6939, 1989. tase VHR. A novel role in regulating the ERK pathway. J. Biol. Chem., 274: 12. Ames, B. N. Assay of inorganic phosphate, total phosphate and phosphatases. 13271–13280, 1999. Methods Enzymol., 8: 115–118, 1966. 39. Velledor, A. F., Xaus, J., Marque`s, L., and Celada, A. Macrophage colony-stimulating 13. Yanagisawa, K., Resnick, D., Abeijon, C., Robbins, P. W., and Hirschberg, C. B. A factor induces the expression of mitogen-activated protein kinase phosphatase-1 guanosine diphosphatase enriched in Golgi vesicles of Saccharomyces cerevisiae. through a protein kinase C-dependent pathway. J. Immunol., 163: 2452–2462, 1999. Purification and characterization. J. Biol. Chem., 265: 19351–19355, 1990. 40. Chang, L. C., and Wang, J. P. Examination of the signal transduction pathways 14. Velasco, J. A., Cansado, J., Penã, M. C., Kawakami, T., Laborda, J., and Notario, V. leading to activation of extracellular signal-regulated kinase by formyl-methionyl- Cloning of the dihydroxyacid dehydratase-encoding gene (ILV3) from Saccharomy- leucyl-phenyalanine in rat neutrophils. FEBS Lett., 454: 165–168, 1999. ces cerevisiae. Gene (Amst.), 137: 179–185, 1993. 41. Maloney, J. A., Tsygankova, O. M., Yang, L., Li, Q., Szot, A., Baysal, K., and 15. Mumberg, D., Mu¨ller, R., and Funk, M. Yeast vectors for the controlled expression Williamson, J. R. Activation of ERK by Ca2ϩ store depletion in rat liver epithelial of heterologous proteins in different genetic backgrounds. Gene (Amst.), 156: 119– cells. Am. J. Physiol., 276: C221–C230, 1999. 122, 1995. 42. Okumura, K., Shirasawa, S., Nishioka, M., and Sasazuki, T. Activated Ki-Ras 16. Gietz, D., St. Jean, A., Woods, R. A., and Schiestl, R. H. Improved method for suppresses 12-O-tetradecanoylphorbol-13-acetate-induced activation of the c-Jun high-efficiency transformation of intact yeast cells. Nucleic Acids Res., 20: 1425, NH -terminal kinase pathway in human colon cancer cells. Cancer Res., 59: 2445– 1992. 2 2450, 1999. 17. Abeijon, C., Yanasigawa, K., Mandon, E. C., Haüsler, A., Moremen, K., Hirschberg, 43. Axmann, A., Seidel, D., Reimann, T., Hempel, U., and Wenzel, K-W. Transforming C. B., and Robins, P. W. Guanosine diphosphatasae is required for protein and ␤ sphingolipid glycosylation in the Golgi lumen of Saccharomyces cerevisiae. J. Cell growth factor- 1-induced activation of the Raf-MEK-MAPK signaling pathway in rat Biol., 122: 307–323, 1993. lung fibroblasts via a PKC-dependent mechanism. Biochem. Biophys. Res. Commun., 18. Avila, M. A., Velasco, J. A., Smulson, M. E., Dritschilo, A., Castro, R., and Notario, 249: 456–460, 1998. V. Functional expression of poly(ADP-ribose) polymerase in Schizosaccharomyces 44. Sellers, L. A. Prolonged activation of extracellular signal-regulated kinase by a protein kinase C-dependent and N17Ras-insensitive mechanism mediates the prolif- pombe results in mitotic delay at G1, increased mutation rate, and sensitization to radiation. Yeast, 10: 1003–1017, 1994. erative response of Gi/o-coupled somatostatin sst4 receptors. J. Biol. Chem., 274: 19. Berninsone, P., Lin, Z-Y., Kempner, E., and Hirschberg, C. B. Regulation of yeast 24280–24288, 1999. Golgi glycosylation. Guanosine diphosphatase functions as a homodimer in the 45. Weber, J. D., Raben, D. M., Phillips, P. J., and Baldassare, J. J. Sustained activation membrane. J. Biol. Chem., 270: 14564–14567, 1995. of extracellular-signal-regulated kinase 1 (ERK1) is required for continued expression 20. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of of cyclin D1 in G1 phase. Biochem. J., 326: 61–68, 1997. bacteriophage T4. Nature (Lond.), 227: 680–685, 1970. 46. Racke, F. K., Lewandowska, K., Goueli, S., and Goldfarb, A. N. Sustained activation 21. Downward, J., Riehl, R., Wu, L., and Weinberg, R. A. Identification of a nucleotide of the extracellular signal-regulated kinase/mitogen-activated protein kinase pathway exchange-promoting activity for p21ras. Proc. Natl. Acad. Sci. USA, 87: 5998–6002, is required for megakaryocytic differentiation of K562 cells. J. Biol. Chem., 272: 1990. 23366–23370, 1997. 22. Bordier, C. Phase separation of integral membrane proteins in Triton X-114 solution. 47. Shapiro, P. S., Whalen, A. M., Tolwinski, N. S., Wilsbacher, J., Froelich-Ammon, J. Biol. Chem., 256: 1604–1607, 1981. S. J., Garcia, M., Osheroff, N., and Ahn, N. G. Extracellular signal-regulated kinase 23. Maliszewski, C. R., Delespesse, G. J., Schoenborn, M. A., Armitage, R. J., Fanslow, activates topoisomerase II␣ through a mechanism independent of phosphorylation. W. C., Nakajima, T., Baker, E., Sutherland, G. R., Poindexter, K., Birks, C., Alpert, Mol. Cell. Biol., 19: 3551–3560, 1999. A., Friend, D., Gimpel, S. D., and Gayle, R. B. The CD39 lymphoid cell activation 48. Dieckgraefe, B. K., and Weems, D. M. Epithelial injury induces egr-1 and fos antigen. Molecular cloning and structural characterization. J. Immunol., 153: 3574– expression by a pathway involving protein kinase C and ERK. Am. J. Physiol., 276: 3583, 1994. 24. Hsieh, H. L., Tong, C. G., Thomas, C., and Roux, S. J. Light-modulated abundance G322–G330, 1999. of an mRNA encoding a calmodulin-regulated, chromatin-associated NTPase in pea. 49. Ezumi, Y., Uchiyama, T., and Takayama., H. Thrombopoietin potentiates the protein Plant Mol. Biol., 30: 135–147, 1996. kinase C-mediated activation of mitogen-activated protein kinase/ERK kinases and 25. Ron, D., Zannini, M., Lewis, M., Wickner, R. B., Hunt, L. T., Graziani, G., Tronick, extracellular signal-regulated kinases in human platelets. Eur. J. Biochem., 258: S. R., Aaronson, S. A., and Eva, A. A region of proto-dbl essential for its transforming 976–985, 1998. activity shows sequence similarity to a yeast cell cycle gene, CDC24, and the human 50. Impey, S., Obrietan, K., Wong, S. T., Poser, S., Yano, S., Wayman, G., Deloulme, breakpoint cluster gene, bcr. New Biol., 3: 372–379, 1991. J. C., Chan, G., and Storm, D. R. Cross talk between ERK and PKA is required for 26. Broach, J. Ras-regulated signaling processes in Saccharomyces cerevisiae. Curr. Ca2ϩ stimulation of CREB-dependent transcription and ERK nuclear translocation. Opin. Genet. Dev., 1: 370–377, 1991. Neuron, 21: 869–883, 1998.

1728

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 2000 American Association for Cancer Research. Both Normal and Transforming PCPH Proteins Have Guanosine Diphosphatase Activity But Only the Oncoprotein Cooperates with Ras in Activating Extracellular Signal-regulated Kinase ERK1

Juan A. Recio, J. Guillermo Páez, Baishali Maskeri, et al.

Cancer Res 2000;60:1720-1728.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/60/6/1720

Cited articles This article cites 45 articles, 21 of which you can access for free at: http://cancerres.aacrjournals.org/content/60/6/1720.full#ref-list-1

Citing articles This article has been cited by 7 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/60/6/1720.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/60/6/1720. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.