THE JOURNAL 0 1993 by The American Society for Biochemistry

OF BIOLOGICAL CHEM!STRY and Molecular Biology, Inc.

Vol. 268. No. 32, Issue of November 15, P P . 24427-24431, 1993 Prrnted in U. S. A.

Murine Cortactin 1s P ~ o s p ~ o r y ~ a t e ~ in Response to ~ i ~ r ~ b ~ a ~ t Growth Factor-1 on Tyrosine Residues Late in the G1 Phase of the BALB/c 3T3 Cell Cycle*

(Received for publication, April 27, 1993)

Xi Zhan, Xiaoguo Hu, Brian Hampton, Wilson H. Burgess, Robert Friesel, and Thomas MaciagS From the Department of Molecular Siorogy, Holland Laboratov, American Red Cross, Rockville, Maryland 20855

We have previously reported that BALBlc 3T3 cells require a prolonged exposure to fibroblast growth fac- tor (FGF)-1 for the stimulation of maximal DNA syn- thesis, and this event correlates with the tyrosine phos- phorylation of novel proteins late in GI including a protein termed pSO/p85 (Zhan, X., Hu, X., Friesel, R., and Maciag, T. (1993) J. Biol. Chem. 268, 9611- 9620). We have purified, sequenced, and cloned the cDNA encoding pSO/p85 and report that it is the mu- rine homolog of the chicken cortactin gene and a mem- ber of the human hematopoietic specific- 1 gene family. Immunochemical analysis of m-cortactin-tyrosine phosphorylation in response to FGF-1 demonstrates a biphasic phosphorylation pattern both as a weak im- mediate-early and strong mid to late GI response pro- tein. Because the chicken cortactin gene was o r i ~ n ~ ~ y isolated as a substrate for v-Src, FGF-1 may influence the enzymatic activity of other cell-associated tyrosine kinases which utilize p80/p85 (cortactin) as a polypep- tide substrate.

The regulation of mammalian cell proliferation and differ- entiation by growth factors involves their ability to induce and regulate intracellular signaling pathways by the stimula- tion of receptor-mediated tyrosine phosphorylation (1). Intra- cellular polypeptides that either associate with growth factor receptors or are phosphorylated in a growth factor-dependent manner are considered to be components of the signal trans- duction cascade, and these include phospholipase C-y (2, 3), GTPase-activating protein for ra.s (4)) the p85 subunit of phosphotidylinositol-3-kinase (5, 6)) src-iike tyrosine kinases (7), mitogen-activated protein kinases (8), and the rufproto- oncogene (9). However, the mechanisms utilized by growth factors and their receptors to modify the cytoskeleton and specific gene expression, as well as regulate DNA synthesis, are not well defined.

We reported that the commitment of BALB/c 3T3 cells to DNA synthesis in response to fibroblast growth factor (FGF)’-

* This work was supported by Grants HL32348 and HL44336 (to T. M.), Grant HL35762 (to W. H. B.), and Grant HD29561 (to R. F.1 from the National Institutes of Health. The costs of pu~lication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “adoertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed: Dept. of Molecular Biology, Holland Laboratory, American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855. Tel.: 301-738-0653; Fax: 301-738- 0465.

The abbreviations used are: FGF, fibroblast growth factor; FGFR, fibroblast growth factor receptor; HS, hematopoietic specific; PAGE, polyacrylamide gel electrophoresis; SH, src-homology; CAPS, 3-(cy- clohexylamino)propanesulfonic acid.

1 requires continual exposure of the target cell to the growth factor €or 12 h (10). The prolonged exposure to FGF-1 corre- lates with an afteration in the tyrosine phospho~Iation pat- tern seen in whole cell lysates during the mid to late GI phase of the BALB/c 3T3 cell cycle and the FGF-1-dependent appearance of p6O and p80/p85 in late GI as phosphotyrosine- containing proteins (10). The phosphorylation of these pro- teins is also evident in FGF-1-treated NIH 3T3 cells trans- fected with an FGF receptor (FGFR)-l cDNA (10). We report here a partial protein sequence of a phosphotyrosine-contain- ing protein with an apparent molecular weight of 80,000 that is phosphorylated late in G, and the cloning of its cDNA. The murine p80/p85 cDNA encodes a translation product that is the murine equivalent of chicken cortactin i l l) , a member of the human HS-1 gene family (12).

EXPERIMENTAL PROCEDURES

The Purification of m-Cortactin-Approximately 3 X lo9 NIH 3T3 cells transfected with FGFR-I (10) that were grown in Dulbecco’s modified Eagle’s medium supplemented with 0.5% (v/v) calf serum (Life Technologies, Inc.) and treated with 10 ng/ml human recombi- nant FGF-1 (13) and 10 units/mI heparin for 30 min were used for the preparation of cell extracts as described previously (10). Briefly, cells were lysed in 50 mM Tris, pH 8.0, containing 150 mM NaCl, 1% (v/v) Nonidet P-40,10% (v/v) glycerol, 0.5% (w/v) sodium deoxycho- late, 50 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 2 ~ g / m l aprotinin, and 1 mM sodium vanadate (lysis buffer), and further disrupted by sonication for 1 min. The extract was clarified by centrikgation at 20,000 X g for 30 min at 4 “C, and the supernatant passed through an aRti-ph~photyrosine Sepharose column (10 ml) preequilibrated with lysis buffer prepared as previously described (14). After washing with 10 volumes of lysis buffer, the phosphoty- rosine-containing proteins were eluted with lysis buffer containing 40 mM phenyl phosphate. The fractions containing the phosphotyrosine- containing proteins were identified by immunoblot analysis using an anti-phosphotyrosine antibody as described (14). Fractions were com- bined and concentrated by ultra~ltration (Amicon) according to procedures recommended by the manufacturer. The mixture of phos- photyrosine-containing proteins was subjected to poIyacryIamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS- PAGE) using a 7.5% (w/v) acrylamide gel. The proteins were trans- ferred to a Problot (Applied Biosystems) membrane in 10 mM CAPS, pH 11, 10% (v/v) methanol. The proteins were visualized by blotting with anti-phosphotyrosine antibodies or by staining with Coomassie Britiiant Blue (Fig. 1). The C~massie-s t~ined band corresponding to p80 was excised from the blot and subjected to automated Edman degradation using an Applied Biosystems model 473A protein sequen- ator. Alternatively, the phosphotyrosine-containing proteins were resolved by SDS-PAGE and transferred to nitrocellulose (Millipore). Proteins were visualized by staining with 0.2% (w/v) Ponceau S in 1% (v/v) acetic acid and subjected bo in situ proteolytic digestion using lysyl endopeptidase C (Boehringer Mannheim) as described 115). Peptides were isolated by reversedphase high performance liquid chromatography using an Applied Biosystems model 130. Isolated peptides were subjected to automated Edman degradation using an Applied Biosystems model 473A or 477A protein sequenator.

The Cloning of the m-Cortactin cDNA-Two degenerated deoxy-

24427

24428 FGF- 1 Regulates Cortactin Phosphorylation

nucleotide oligomers (T/CXGCXGTXGGXTTT/CGAT/CTA and T/CTCXACAGTAAGTTXGCXGGA/GAAXAA/GXCCA/GTA, where X is any base) were designed according to amino acid sequences derived from peptide-1 and peptide-3 (Fig. 2 A ) . The pool of T, polynucleotide kinase 32P-labeled oligomers was used to screen a BALB/c 3T3-derived XglO cDNA library (kindly provided by Lester Lau) as previously described (16). Briefly, nitrocellulose filters were prehybridized in duplicate for 90 min a t 42 “C in 6 X SSC containing 5 X Denhardt’s solution, 20 mM NaH,PO,, 0.5% (w/v) SDS, and 100 pug/ml carrier DNA and rehybridized in 6 X SSC containing 20 mM NaH,P04, 1 X Denhardt’s solution, and 100 pg/ml tRNA at 42 “C for 16 h. The filters were washed 3 times with 6 X SSC containing 0.1% (w/v) SDS for 10 min a t room temperature followed by three washes for 30 min a t 42 “C and 30 min a t 50 “C. After a final wash with 6 X SSC containing 0.1% (w/v) SDS for 20 min at 65 “C, the filters were exposed to x-ray film for 2 days. The manipulation of phage DNA preparation, restriction map analysis, and subcloning to a plasmid vector (pBluescript SKII) were based on standard procedures as described (16). Sequence analysis was performed using either Sequen- ase 2.0 (U. s. Biochemical Corp.) or by automated double-stranded DNA sequencing (Applied Biosystems).

Immunological Analysis-Antibodies were produced in rabbits in- jected with a peptide corresponding to residues 343-362 in the m- cortactin sequence that was synthesized on a multiple antigen pre- senting resin (Applied Biosystems). Synthesis was performed using an Applied Biosystems model 431A peptide synthesizer. Immunopre- cipitation reactions were performed using FGFR-1-transfected NIH 3T3 cells (10) grown to 60% confluence. The cells were incubated in methionine- and cysteine-free Dulbecco’s modified Eagle’s medium supplemented with 5% (v/v) dialyzed calf serum (Life Technologies, Inc.) for 20 min and grown in the same medium containing 160 pCi/ ml of a [”S]cysteine/methionine mixture (Du Pont-New England Nuclear) for 3 h. The cells were harvested, incubated in lysis buffer, and clarified by centrifugation a t 13,000 X g for 10 min a t 42 “C. The supernatant was divided into two aliquots. Proteins were precipitated with 30 pl of 50% (v/v) protein A-Sepharose (Pharmacia LKB Bio- technology Inc.) which was complexed with either a rabbit antiserum raised against the m-cortactin peptide or preimmune serum from the same rabbit. Immunoprecipitates were washed 4 times with lysis buffer and dissolved in 50 pl of 2 X SDS sample buffer (17), and proteins were resolved by SDS-PAGE using 7.5% (w/v) acrylamide gels. The radiolabeled proteins were detected by autoradiography.

T o analyze the phosphotyrosine content of m-cortactin, quiescent BALB/c 3T3 cells were stimulated with 10 ng/ml FGF-1 and 10 units/ml heparin for the indicated time and subjected to immunopre- cipitation using the m-cortactin antibody raised against a recombi- nant glutathione sulfate transferase-cortactin fusion protein. After electrophoresis, the gel was transferred to a nitrocellulose membrane and was probed with an anti-phosphotyrosine antibody as previously described (14).

RESULTS AND DISCUSSION

Extracts were prepared from NIH 3T3/FGFR-1 cells (14) that had been exposed to FGF-1 for 30 min, and the phospho- tyrosine-containing protein p80/p85 (Fig. 1) was purified as described under “Experimental Procedures.” Degenerate deoxynucleotide probes based on the amino acid sequence of three p80/p85 derived peptides were used to screen a murine cDNA library to obtain the p80/p85 cDNA. The nucleotide sequence of the murine p80/p85 cDNA contains an apparent open reading frame of 1638 base pairs. One putative transla- tion initiation start site with a corresponding Kozak sequence (18) is proceeded by an in-frame TAA stop codon a t nucleotide -93. Although the open reading frame encodes a polypeptide of 546 amino acids with a calculated molecular weight of 61,000 (Fig. 2 A ) , in uitro translation analysis of the p80/p85 cDNA yields a polypeptide that migrates with an apparent molecular weight of 80,000 (data not shown). Thus, the mo- lecular weight difference between predicted and observed values may reflect unique structural properties of the protein. The predicted protein sequence displays several interesting features including 6.5 tandem repeated sequence domains from residues 81 to 326 with each repeat unit comprised of 37

FIG. 1. SDS-PAGE analysis of phosphotyrosyl-containing proteins purified from FGFR-1-transfected NIH 3T3 cells. The phosphotyrosine-containing proteins from extracts of FGFR-1 NIH 3T3 cell transfectants were purified using anti-phosphotyrosine affinity as described under “Experimental Procedures.” The purified proteins were resolved by 7.5% (w/v) SDS-PAGE, and their positions were revealed by staining with Coomassie Brilliant Blue. The bands with the position of p90, p80, and p60 (arrows) co-migrated with those independently identified by immunoblot analysis using an anti- phosphotyrosine antibody.

amino acid residues. A src-homology (SH)-3 domain is also positioned at the carboxyl terminus. Several other sequences are also noteworthy: an acidic domain containing 10 glutamic or aspartic acid residues from a total of 22 amino acids near the amino terminus, a proline-rich region in the area between the tandem repeats and the SH-3 domain, 14 potential rec- ognition sites for protein kinase C, one potential recognition site for protein kinase A at residue 401, one potential N - linked glycosylation site, and three ATP-binding (P-loop) sites located within the tandem repeats. Although there are a total of 27 tyrosine residues in the predicted protein product, none of them fit the consensus for phosphorylation by tyro- sine kinases (19). However, 12 tyrosine residues are flanked by acidic amino acids within the amino-terminal half of the protein, and these sites may represent potential tyrosine phosphorylation residues (19).

The primary sequence of p80/p85 is 53% identical to human HS-1, a gene expressed specifically in the hematopoietic lin- eage and cloned by low stringency hybridization in an attempt to identify the human analog of the viral E1A gene (12). In addition, p80/p85 is also closely related to the chicken cortac- tin gene (11) with an overall 84% homology (Fig. 2B). Cor- tactin, also termed p80/p85, was recently isolated as a sub- strate for the v-Src oncogene and shown to be a cytoskeleton- associated protein (20). Both cortactin and p80/p85 contain 6.5 tandem repeats while HS-1 has only 3.5 repeats. The SH- 3 domain is precisely conserved between p80/p85 and cortac- tin (Fig. 2B); however, the primary sequence within the SH- 3 domain is only 70% similar between p80/p85 and HS-1. Although the region between tandem repeats and the SH-3 domain is rich in proline and hydrophilic residues in all three proteins, the sequence homology within the region is quite limited (Fig. 2B). The chicken cortactin cDNA contains an additional ATG site and Kozak sequence upstream from and in frame with the putative translational start site in p80/p85 and HS-1 (11). The translation of chicken cortactin from this alternative ATG start site would result in a protein product with a putative signal sequence at the NH2 terminus. How- ever, analysis of the cellular distribution of chicken cortactin has revealed exclusive intracellular staining (11). This obser-

FGF-1 Regulates Cortactin Phosphorylation

A

24429

FIG. 2. The structure of p8O/p85/ murine cortactin. A, deduced amino acid sequence of m-cortactin is shown, and the amino acid sequences derived from the three p8O/p85 lysyl endopepti- dase C-derived digestion fragments are boxed. The repeated regions are under- lined with gray boxes, and the boldface letters near the COOH terminus refer to the SH-3 domain. An acidic domain near the NH, terminus is underlined with a hatched box, and the proline-rich domain is represented by a solid line. R, sche- matic representation of the comparison of murine-cortactin with chicken-cortac- tin and human HS-1. The hatched box at the NH2 terminus of c-cortactin cor- responds to extended sequence rich in hydrophobic amino acids, and the total number of amino acid residues for the three proteins are indicated. Potential N-linked glycosylation sites are indi- cated by triangles. Other domains are as indicated with percent sequence identity for the individual domains listed.

B

m-cortactin

c-cortactin

HS1

sequence homology

vation indicates that the upstream start site may not be utilized or the putative signal sequence is not functional. Despite the difference in the presence of an alternative trans- lational start site, the overall sequence of murine p80/p85 is very similar to chicken cortactin. Indeed, recombinant p80/ p85 is recognized by monoclonal antibody 4F11,2 which was used for the isolation of the chicken cortactin cDNA (11). Thus, it is likely that p80/p85 represents the murine homolog of chicken cortactin, whereas HS-1 is a distant human cor- tactin-related gene. Therefore, murine p80/p85 will be re- ferred to as m-cortactin.

To analyze the expression of the m-cortactin protein, anti- serum was raised against a synthetic peptide corresponding to amino acid residues 343-362 which shares 70% sequence identity with the same region in chicken cortactin and shares no identity to HS-1. The antiserum precipitates specifically a protein with an apparent molecular mass of 80 kDa and co- precipitates two additional bands which migrate a t 65 and 50 kDa (Fig. 3A). While the identity of p65 and p50 is currently unknown, it was possible to compete p80, p65, and p50 by the addition of excess peptide to the immunoprecipitation mix- ture.* To verify that m-cortactin represents the phosphoty- rosyl-containing protein which was previously characterized as p80/p85 (lo), extracts prepared from FGFR-1-transfected NIH 3T3 cells were immunoprecipitated with anti-m-cortac- tin, resolved by SDS-PAGE, and probed with anti-phospho- tyrosine antibodies. As shown in Fig. 3B, lane 4, m-cortactin appears as a tyrosine-phosphorylated protein. In addition, extracts from FGFR-1-transfected NIH 3T3 cells were preab- sorbed with either m-cortactin antiserum or preimmune serum, resolved by SDS-PAGE, and subjected to immunoblot

X. Zhan and T. Maciag, unpublished observations.

Acidic Domain Repeats SH3

60

120

180

240

300

360

420

480

540

546

0 ~ 486 / I

; \>

65-88% 70-95% ,'

'37-70% 70-100%

A 6 MW (kDa) 1 z 3 4

95 - .)

65 - - ".

200- r?'?Yfl

116-

I 80- "-

NIH 3T3 cells transfected with FGFR-1 were labeled with [:"S]Met FIG. 3. Immunoprecipitation analysis of m-cortactin. A,

and [%]Cys and lysed as described under "Experimental Proce- dures." The cell extracts were divided into two equal aliquots. One aliquot was subjected to precipitation with a rabbit polyclonal anti- serum against a peptide deduced from the m-cortactin amino acid sequence (lane I ), and the other aliquot was subjected to precipitation with preimmune antiserum from the same rabbit (lane 2). The precipitates were resolved by 7.5% (w/v) SDS-PAGE, and the radio- labeled proteins were visualized by autoradiography. R, extracts de- rived from FGFR-1 NIH 3T3 cell transfectants were incubated with either polyclonal anti-m-cortactin antiserum (lane I ), preimmune serum (lane 2) , or without antibody (lane 3 ) . After centrifugation, the clarified supernatants were resolved by 7.5% (w/v) SDS-PAGE, and the gel was transferred to a nitrocellulose filter and blotted with an anti-phosphotyrosine antibody. In lane 4, a separate aliquot of NIH 3T3 cell extract was incubated with anti-m-cortactin antibody, and the immunoprecipitate was analyzed by 7.5% (w/v) SDS-PAGE. The arrow positions the migration of the p80 protein.

24430 FGF-1 Regulates Cortactin Phosphorylation

analysis using a phosphotyrosine antibody. Whereas the in- tensity of the p80 band remained unchanged in the preim- mune serum-absorbed extract (Fig. 3B, lune 2), it was signif- icantly diminished in the extracts absorbed with anti-m- cortactin antiserum (Fig. 3B, lane 1).

To confirm that m-cortactin is one of the proteins whose phosphotyrosine level is elevated in response to FGF-1 during the late GI phase of the BALB/c 3T3 cell cycle, the tyrosine phosphorylation of m-cortactin was analyzed in BALB/c 3T3 cells that were stimulated with FGF-1 for various periods of time. As shown in Fig. 4A, there is a basal level of tyrosine phosphorylation of m-cortactin in unstimulated cells, and the level of tyrosine phosphorylation is slightly enhanced when the cells were exposed to FGF-1 for 5 min. However, the presence of m-cortactin as a phosphotyrosine-containing pro- tein is quite evident during the mid to late GI phase (Fig. 44). Interestingly, the level of m-cortactin-tyrosine phosphoryla- tion was diminished in the period between 1 and 4 h after stimulation with FGF-1 (Fig. 4A). Because the level of m- cortactin protein did not vary during the time period studied (Fig. 4B), these data suggest that differential mechanisms may be involved in the regulation of m-cortactin-tyrosine phosphorylation during the early and late GI phase of the BALB/c 3T3 cell cycle in response to FGF-1.

It has been shown that chicken cortactin displays a punc- tate cytosolic staining pattern in vitro and co-localizes with actin in the peripheral extensions of cells. Therefore, it has been proposed that cortactin may play a role in focal adhesion

116 -

80 - anti-p-Tyr

49.5 -

anti-cortactin

FIG. 4. Phosphotyrosine and m-cortactin immunoblot analysis of BALB/c 3T3 cells treated with FGF-1. Upper panel, quiescent BALR/c 3T3 cells were stimulated with recombinant hu- man FGF-1 (10 ng/ml) and heparin (10 units/ml) for various time as indicated. Cell extracts were then prepared as described (14) and incubated with a rabbit GST-m-cortactin fusion protein antibody. The immunoprecipitates were then analyzed for their phosphotyro- sine content by immunoblot analysis using a monoclonal antibody against phosphotyrosine as described under "Experimental Proce- dures." Lower panel, cell extracts were prepared as described above and resolved by 7.5% (w/v) SDS-PAGE, and immunoblot analysis was performed using a rabbit GST-m-cortactin fusion protein anti- body as described under "Experimental Procedures." The upper band is immunoreactive m-cortactin, and the lower band is IgG heavy- chain from the immunoprecipitation.

plaque function in vitro (11,20). However, the increase in the level of tyrosine phosphorylation of m-cortactin induced by FGF-1 during the late GI phase of the BALB/c 3T3 cell cycle also implies that cortactin may have additional biological functions especially during the commitment of FGF-l-in- duced cells to migrate and/or proliferate in vitro. Indeed, cortactin-like proteins have also been localized as perinuclear proteins in murine 10T1/2 3T3 cells (21). We have also observed that m-cortactin protein in BALB/c 3T3 cells ex- hibits perinuclear and focal adhesion site-staining patterns: but additional analysis is required to define the intracellular distribution of m-cortactin during the FGF-1-induced cell cycle.

More recent studies have shown that tyrosine phosphoryl- ation of a cortactin-related protein in a src-transformed cell line is also elevated in response to epidermal growth factor, platelet-derived growth factor, and colony-stimulating growth factor-1 indicating roles for other growth factors in the regu- lation of cortactin function (12). Interestingly, the biphasic induction of tyrosine phosphorylation of cortactin during the GI phase in response to FGF-1 is also observed in cells treated with epidermal growth factor (21). Thus, differential signaling pathways may be utilized by different polypeptide growth factors to regulate the tyrosine phosphorylation of cortactin. However, a direct role for FGFR-1 as the enzyme responsible for the tyrosine phosphorylation of m-cortactin is unlikely. Indeed, FGFR-1 antibodies were not able to efficiently co- precipitate m-cortactin under standard immunoprecipitation conditions, and a recombinant GST-m-cortactin fusion pro- tein was not able to serve as an efficient substrate for recom- binant FGFR-1 in an in vitro kinase assay.' Furthermore, unlike other tyrosine kinase receptor substrates, m-cortactin does not contain an apparent SH-2 domain. Alternatively, additional cellular factor(s) may be required for FGFR-1 to associate with m-cortactin in vitro. Wong et al. (22) have recently reported that cortactin may be a substrate for c-src since a cortactin-related protein in platelets is phosphorylated by c-src in response to thrombin. Whether src or a src-related kinase(s) is responsible for the tyrosine phosphorylation of m-cortactin induced by FGF-1 has yet to be determined. Interestingly, Schuuring et al. (23) have recently cloned the human cortactin gene as a product of the human llq13 chromosome region which is amplified in many human tu- mors. Although the human llq13 chromosome region also encodes the FGF-3 (int-2) and FGF-4 (hstlKS3) genes, these FGF family members are not frequently expressed in tumors with an l lq13 amplification (23). Thus, the identification of m-cortactin as an FGF-1-induced phosphotyrosyl-containing protein during the late GI phase of the BALB/c 3T3 cell cycle may facilitate the elucidation of the interrelationship between FGF receptors and other intracellular signaling macromole- cules which may facilitate tumor cell growth.

Acknowledgments-We thank Tom Parsons (University of Vir- ginia) for the 4Fl l antibody and Lester Lau (University of Illinois) for the BALB/c 3T3 cell cDNA library. We also thank K. Wawzinski and A. Lott for expert secretarial assistance.

REFERENCES

2. Burgess, W. H., Dionne, C. A,, Kaplow, J., Mudd, R., Friesel, R., Zilberstein, 1. Ullrich, A., and Schlessinger, J. (1990) Cell 61, 203-212

A,. Schlessinaer. J.. and Jave. M. (1990) Mol. Cell. Biol. 10.4770-4777 3. Mohammadi, M., Honegger,'A.' M., Rotin, D., Fischer, R., Bellot, F., Li,

W.. Dionne. C. A.. Jave. M.. Rubinstein. M.. and Schlessineer. d . (1991) ~ d l . Cell. ~ ' i o l . 1 i, ~ k d - 5 0 7 8

and Aaronson. S. A. (1989) Nature 342. 711-714

, . ~, I . .

4. Molloy, C. J., Bottaro, D. P., Fleming, T. P., Marshall, M. S., Gibbs, J. B.,

5. Escohedo, J. A., Navankasattusas, S., Kavanaugh, W. M., Milfay, D., Fried, V. A,, and Williams, L. T. (1991) Cell 65,75-82

6. Otsu, M., Hiles, I., Gout, I., Fry, M. J., Ruiz-Larrea, F., Panayotou, G., Thompson, A,, Dhand, R., Hsuan, J., Totty, N., Smith, A. D., Morgan,

7.

8.

10. 9.

11.

12.

13.

FGF-1 Regulates COJ S. J. , Courtneidge, S. A,, Parker, P. J., and Waterfield, M. D. (1991) Cell

Kypta, R. M., Goldberg, Y., Ulug, E. T., and Courtneidge, S. A. (1990) Cell

Stureill. T. W.. and Wu. J. (1991) Biochim. B ~ O D ~ V S . Acta 1092.350-357

65, 91-104

62,481-492

Rap;, U. R. (1991) Oncogene 6,495-500 Zhan, X., Hu, X., Friesel, R., and Maciag, T. (1993) J. Bid. Chem. 268,

1 -

961 1-9620 Wu, H., Reynolds, A. B., Kanner, S. B., Vines, R. R., and Parsons, J. T.

Kitamura, D., Kaneko, H., Miyagoe, Y., Ariyasu, T., and Watanabe, T. (1991) Mol. Cell. Biol. 11,5113-5124

Imamura, T., Engleka, K., Zhan, X., Tokita, Y., Forough, R., Roeder, D., (1989) Nucleic Acids Res. 17,9367-9379

Jackson, A., Maier, J. A. M., Hla, T., and Maciag, T. (1990) Science 249, 1567-1570

. . - - . . - .

:ctil 14.

15.

16.

17. 18. 19. 20. 21.

23. 22.

’1 Phaspha~lat~on 24431 Friesel, R., Burgess, W. H., and Maciag, T. (1989) Mol. Cell. Bioi. 9, 1857-

1865 Egerton, M., Burgess, W. H., Chen, D., Druker, A. J., Bretscher, A,, and

Samelson, L. E. (1992) J. Immunol. 149, 1847-1852 Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Moleculur Clonin A

&@oratory Manual, Cold Spring Harbor Laboratory, Cold Spring Hartor,

Laemmli, U. K. (1970) Nature 227,680-685 Kozak. M. (1986) Cell 44.283-292

IU 1

Pearson, R.‘B., and Kem’ B. E. (1991) Methods Enzymol. 200,62-81

Ma:, M. C., Wilson, L. K., Moyers, J. S., Vines, R. R., Parsons, J. T., and Wu H., and Parsons, J. $ (1993) J , Cell. Biol, 120, 1417-1426

Wong, S., keynolds, A. B., and Papkoff, J. (1992) Oncogene 7,2407-2415 Schuuring, E., Verhoeven, E., Litvinov, S., and Michalides, R. (1993) Mol.

Parsons S. J. (1992) Oncogene 7,2429-2438

Cell. Bwl. 13, 2891-2898