Integrin-mediated Tyrosine Phosphorylation of SHPS-1 and Its Association with SHP-2

SHPS-1 is a receptor-like glycoprotein that undergoes tyrosine phosphorylation and binds SHP-2, an Src homology 2 domain containing protein tyrosine phosphatase, in response to various mitogens. Cell adhesion to extracellular matrix proteins such as fibronectin and laminin also induced the tyrosine phosphorylation of SHPS-1 and its association with SHP-2. These responses were markedly reduced in cells overexpressing the Csk kinase or in cells that lack focal adhesion kinase or the Src family kinases Src or Fyn. However, unlike Src, focal adhesion kinase did not catalyze phosphorylation of the cytoplasmic domain of SHPS-1 in vitro. Overexpression of a catalytically inactive SHP-2 markedly inhibited activation of mitogen-activated protein (MAP) kinase in response to fibronectin stimulation without affecting the extent of tyrosine phosphorylation of focal adhesion kinase or its interaction with the docking protein Grb2. Overexpression of wild-type SHPS-1 did not enhance fibronectin-induced activation of MAP kinase. These results indicate that the binding of integrins to the extracellular matrix induces tyrosine phosphorylation of SHPS-1 and its association with SHP-2, and that such phosphorylation of SHPS-1 requires both focal adhesion kinase and an Src family kinase. In addition to its role in receptor tyrosine kinase-mediated MAP kinase activation, SHP-2 may play an important role, partly through its interaction with SHPS-1, in the activation of MAP kinase in response to the engagement of integrins by the extracellular matrix.

The binding of integrins to extracellular matrix (ECM) 1 pro-teins such as fibronectin and laminin contributes to cell adhesion to the ECM and evokes a variety of cellular responses (1)(2)(3). Activation of integrins induces cell cycle progression (4), enhances growth factor-mediated cell proliferation and DNA synthesis (5), and promotes cell migration (6). These responses are thought to be mediated by integrin-induced tyrosine phosphorylation of various proteins (1-3) that may couple the integrin signal to activation of Ras and MAP kinase. Fak has been suggested to mediate integrin-induced activation of the Ras-MAP kinase signaling cascade (7,8). Fak is a non-receptor-type protein tyrosine kinase that colocalizes with integrins at focal contacts and is autophosphorylated in response to cell adhesion to the ECM (9,10). The integrin-induced autophosphorylation of Fak results in its interaction with the SH2 domain of Src kinase (7,11), after which Src may catalyze the tyrosine phosphorylation of both the kinase domain and the COOH-terminal Tyr 925 residue of Fak (7,8). Phosphorylation of Tyr 925 of Fak creates a binding site for the SH2 domain of Grb2, which constitutively binds to the guanine nucleotide exchange protein SOS (12) and may link integrin engagement to the activation of the Ras-MAP kinase cascade (7,8). However, the existence of an additional pathway for integrin-induced activation of MAP kinase that is independent of Fak or Grb2 binding to Fak has recently been suggested (13,14). In contrast to the apparent participation of protein tyrosine kinases in integrin activation of Ras and MAP kinase, it remains unknown whether a PTPase also is important in this process.
SHP-2 is a non-transmembrane PTPase that contains two SH2 domains (15)(16)(17). A dominant negative mutant of SHP-2 inhibited insulin-or EGF-stimulated activation of Ras and MAP kinase, DNA synthesis, or cell proliferation (18 -23). Furthermore, microinjection of RNA encoding a dominant negative SHP-2 into Xenopus oocytes resulted in severe posterior truncation and inhibited FGF-and activin-mediated mesoderm induction (24). Corkscrew, a putative Drosophila homolog of SHP-2, is thought to be required for Ras1 activation or to function in conjunction with Ras1 during signaling by the Sevenless receptor tyrosine kinase (25,26). Most recently, disruption of the mouse Shp-2 gene was shown to result in failure of normal gastrulation and subsequent death in the early stages of embryogenesis, as well as in a marked impairment in FGFinduced MAP kinase activation (27). Thus, SHP-2 may play a * This work was supported by a grant-in-aid for cancer research and a grant-in-aid for scientific research from the Ministry of Education, Science, Sports, and Culture of Japan; Grant 96-22809 from the Princess Takamatsu Cancer Research Fund; a grant from the Yamanouchi Foundation for Research on Metabolic Disorders; a grant from the Ciba-Geigy Foundation (Japan) for the Promotion of Science; National Institutes of Health Grants HD 24875 and HD 25326; and a grant from Kirin Brewery Co. 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  crucial role in intracellular signaling elicited by various growth factors and hormones, probably by contributing to activation of the Ras-MAP kinase pathway.
SHP-2 binds directly to growth factor receptors, including the platelet-derived growth factor, EGF receptor, c-KIT, and the erythropoietin receptor, in response to receptor stimulation with the corresponding ligand and undergoes tyrosine phosphorylation (28 -33). Furthermore, although SHP-2 does not interact directly with the insulin receptor, it binds through its SH2 domains to tyrosine-phosphorylated docking proteins such as IRS1, IRS2, and GAB1 in response to insulin (34 -36). Corkscrew also binds to another docking protein Drosophila DOS (37). These docking proteins possess multiple tyrosine phosphorylation sites that serve as binding sites for the SH2 domains of various signaling molecules, but they lack any catalytic activity. Thus, in response to ligand stimulation of various receptors, SHP-2 appears to be recruited either to the receptors themselves or to receptor docking proteins.
We have recently identified a membrane glycoprotein, SHPS-1 (38,39) (also known as SIRP (40) and BIT (41)), that possesses three immunoglobulin (Ig)-like domains in its putative extracellular region and four YXX(L/V/I) motifs, which are potential tyrosine phosphorylation and SH2 domain binding sites, in its cytoplasmic region. Various mitogens, including serum, insulin, lysophosphatidic acid, and EGF, induce tyrosine phosphorylation of SHPS-1 and its subsequent association with SHP-2 in cultured cells (38,40,42). Thus, in addition to IRS and GAB1 proteins, SHPS-1 also may recruit SHP-2 from the cytosol to a site near the plasma membrane in response to growth factor stimulation. We have also detected both tyrosine phosphorylation of SHPS-1 and its association with SHP-2 in unstimulated, serum-deprived cells, possibly reflecting the effects of cell adhesion (38). However, the precise mechanism by which cell adhesion induces the tyrosine phosphorylation of SHPS-1 and the physiological roles of the interaction of SHPS-1 with SHP-2 in cell adhesion-triggered signal transduction remain unclear.
We have now shown that the binding of integrins to ECM components such as fibronectin and laminin induces tyrosine phosphorylation of SHPS-1 and its subsequent association with SHP-2. This cell adhesion-induced phosphorylation of SHPS-1 appears to require both Fak and an Src family kinase. Furthermore, the formation of the SHPS-1⅐SHP-2 complex may contribute to MAP kinase activation in response to fibronectin stimulation.

EXPERIMENTAL PROCEDURES
Cells and Antibodies-Rat-1 fibroblasts were cultured in DMEM supplemented with 10% FBS. CHO cells that overexpress human insulin receptors (CHO-IR cells) were maintained in Ham's F-12 medium supplemented with 10% FBS. Before transfection experiments, COS-7 cells were also cultured in DMEM supplemented with 10% FBS.
Rat-1 cells that overexpress a catalytically inactive SHP-2 mutant were generated as described (21). Briefly, a point mutation that changed Cys 459 to Ser was introduced into human SHP-2 cDNA by site-directed mutagenesis, and the mutant cDNA was cloned into the EcoRI site of the mammalian expression vector pSR␣. Rat-1 cells (ϳ3 ϫ 10 5 per 6-cm dish) were transfected with both 10 g of pSR␣ containing mutant SHP-2 cDNA and 1 g of pHyg, which contains the hygromycin B phosphotransferase gene, with the use of 18 l of LipofectAMINE (Life Technologies, Inc.). The cells were cultured in DMEM containing hygromycin B (400 g/ml) (Wako, Osaka, Japan) and 10% FBS, and colonies were isolated 14 -21 days after transfection. Several cell lines expressing the mutant SHP-2 protein were identified by immunoblot analysis of cell lysates with polyclonal antibodies to SHP-2 as described below.
Generation of Rat-1 cells that overexpress rat wild-type SHPS-1 was also performed by transfection with pSR␣ containing wild-type SHPS-1 cDNA (38) and 1 g of pHyg as described above. Cell lines expressing the recombinant SHPS-1 protein were identified by immunoblot anal-ysis with polyclonal antibodies to SHPS-1. CHO-IR cells that overexpress wild-type Csk (Csk-WT) or a kinase-negative Csk (Csk-K/R) were generated by transfection with pSR␣ containing rat wild-type Csk cDNA or a mutant Csk cDNA that contained a point mutation that changed Lys 222 to Arg (kindly provided by M. Okada), as described above.
Src family kinase-deficient (Src Ϫ , Fyn Ϫ ) cells were isolated from the corresponding "knock-out" mouse embryos and immortalized with simian virus 40 large T antigen as described previously (43). Fak-deficient mouse cells (Fak Ϫ ) were established by introduction of a p53 mutation into Fak-deficient mouse embryos as described (44). All knock-out mouse cells and the corresponding wild-type cells were cultured in DMEM supplemented with 10% FBS. Semi-confluent wild-type and Src Ϫ , Fyn Ϫ , or Fak Ϫ cells were transiently transfected with 10 g of pSR␣ containing the wild-type rat SHPS-1 cDNA with the use of Lipo-fectAMINE (Life Technologies, Inc.) as described previously (38). Cells were harvested 2 days after transfection and lysed as described below for immunoprecipitation with mAb 2F34 to rat SHPS-1.
To generate a mAb specific to rat SHPS-1, we partially purified SHPS-1 from 100 10-cm plates of confluent SR-3Y1 cells as described previously (38). The resulting SHPS-1 preparation was injected three times into the hind foot pads of two BALB/c mice at 1-week intervals, after which lymphocytes were isolated from the draining lymph nodes and fused with P3U1 myeloma cells as described (42). Antibodies in culture supernatants of the resulting hybridomas were screened on the basis of their ability to immunoprecipitate SHPS-1 from the membrane fraction of SR-3Y1 cells, as detected by immunoblot analysis of the precipitates with antibodies to phosphotyrosine (PY20). Positive hybridomas were rescreened by the same procedure. Among several positive clones, clone 2F34 was selected, and mAb 2F34 was subsequently purified from ascites fluid of mice with a MAPS II kit (Bio-Rad). The 2F34 mAb reacts well with rat SHPS-1 but poorly with the corresponding protein from hamster or mouse. The detailed properties of this mAb will be described elsewhere. In contrast to mAb 2F34, mAb 4C6, which we prepared previously (42), reacts well with CHO cell SHPS-1 but poorly with the corresponding protein of rat or mouse. Rabbit polyclonal antibodies to SHPS-1 (38) or to SHP-2 (21) were generated with GST fusion proteins containing the COOH-terminal region of either rat SHPS-1 or human SHP-2 as antigen, as described previously. Rabbit polyclonal antibodies (␣91) to both p44 and p42 MAP kinase were prepared with a synthetic peptide corresponding to residues 307-327 of rat MAP kinase. The rabbit polyclonal antibodies to MAP kinase, ␣92, were prepared against synthetic peptide corresponding to residues 350 -367 of rat MAP kinase. Antibodies that react specifically with the tyrosine-phosphorylated form of MAP kinase were obtained from New England Biolabs. HRP-conjugated mAb PY20 to phosphotyrosine, polyclonal antibodies to Fak, and polyclonal antibodies to Csk were obtained from Santa Cruz Biotechnology. The 12CA5 mAb to the HA epitope tag was purified from the ascites of mice injected with 12CA5 hybridoma cells. The mAb to paxillin and the mAb to Grb2 were obtained from Transduction Laboratories; the mAb to v-Src (Ab-1) was from Oncogene Science; and the mAb to Fak was from Upstate Biotechnology.
Cell Adhesion, Immunoprecipitation, and Immunoblot Analysis-Before all experiments, cells were incubated for 12 h with DMEM in the absence of serum. Cells were detached by incubation for 15 min at 37°C in phosphate-buffered saline without Ca 2ϩ and Mg 2ϩ , collected by centrifugation, and washed twice with serum-free DMEM. They were then suspended in DMEM, replated on polystyrene dishes (ϳ2.5 ϫ 10 7 per 10-cm dish) coated with fibronectin or other ECM proteins (Falcon), and incubated at 37°C for various times in serum-free DMEM. In some experiments, cytochalasin D (Sigma) was added 10 min before replating of the cells.
The attached cells were then lysed on ice in 0.5 to 1 ml of ice-cold lysis buffer (20 mM Tris-HCl (pH 7.6), 140 mM NaCl, 2.6 mM CaCl 2 , 1 mM MgCl 2 , 1% Nonidet P-40, 10% (v/v) glycerol) containing 1 mM phenylmethylsulfonyl fluoride and 1 mM sodium vanadate. The lysates were centrifuged at 10,000 ϫ g for 15 min at 4°C, and the resulting supernatants were subjected to immunoprecipitation and immunoblot analysis. Total cell protein in lysates was standardized with the use of a protein assay kit (Bio-Rad) before immunoprecipitation. Immunoprecipitation was performed by incubation of cell lysates for 4 h at 4°C with various antibodies bound to protein G-Sepharose beads (2 g of antibody per 20 l of beads) (Amersham Pharmacia Biotech). The beads were then washed twice with 1 ml of WG buffer (50 mM Hepes-NaOH (pH 7.6), 150 mM NaCl, 0.1% Triton X-100) and resuspended in SDS sample buffer. Gel electrophoresis and immunoblot analysis, with PY20 or other antibodies and an ECL detection kit (Amersham Pharmacia Biotech) were performed as described previously (21,45).
Immune Complex Kinase Assays-A GST fusion protein (GST-SHPS-1-cyto) containing the cytoplasmic portion of SHPS-1 (amino acids 373-481) was generated and purified as described previously (38). Src kinase was immunoprecipitated from Rat-1 cell lysates with a mAb to v-Src as described above. The immune complex was washed twice with kinase assay buffer (50 mM Hepes-NaOH (pH 7.6), 3 mM MnCl 2 , 10 mM MgCl 2 , 1 mM dithiothreitol, 10 M ATP) and incubated with GST-SHPS-1-cyto (ϳ1 g) for 30 min at 24°C in kinase assay buffer (50 l) in the absence or presence of [␥-32 P]ATP (0.1 mCi/ml, 6000 Ci/mmol) (ICN). For examination of the phosphorylation of SHPS-1 by Fak, COS-7 cells (in 10-cm dishes) were transiently transfected with 10 g of the pRC CMV expression vector containing cDNA that encoded HA epitopetagged mouse Fak (kindly provided by S. K. Hanks) (46) with the use of 18 l of LipofectAMINE. After 48 h, the transfected cells were harvested, lysed, and subjected to immunoprecipitation with mAb 12CA5. The precipitates were then subjected to the phosphorylation reaction with GST-SHPS-1-cyto or poly(Glu, Tyr) (1 mg/ml) (Sigma) as described above. After each phosphorylation reaction, the samples were centrifuged at 10,000 ϫ g for 5 min at 4°C, and the resulting supernatants were mixed with SDS sample buffer, boiled, and subjected to gel electrophoresis and autoradiography. The immunoprecipitated Src kinase and HA-tagged Fak were also subjected to immunoblot analysis with antibodies to the corresponding protein tyrosine kinase to determine the amount of each protein in the immune complex, or to autoradiography for determination of autophosphorylation.
Determination of MAP Kinase Activation-MAP kinase activation was monitored by immunoblot analysis of cell lysates with antibodies to p44 and p42 MAP kinase (New England BioLabs) that recognize the enzymes only when they are activated by phosphorylation of Tyr 204 .
MAP kinase activation was also monitored by using the direct in vitro kinase assay as described previously (21). Briefly, after attached or suspension, cells were lysed in 500 l of lysis buffer (25 mM Tris-HCl (pH 7.4), 25 mM NaCl, 1 mM sodium orthovanadate, 10 mM NaF, 10 mM sodium pyrophosphate, 0.5 mM EGTA, 10 nM okadaic acid (Wako Chemicals), and 1 mM phenylmethylsulfonyl fluoride). The lysates were incubated for 2 h at 4°C with ␣92 anti-MAP kinase antibodies that had been bound to protein G-Sepharose beads. The immunoprecipitates were washed twice with lysis buffer and suspended in 35 l of assay buffer (25 mM Tris-HCl (pH 7.4), 10 mM MgCl 2 , 1 mM DTT, 40 M ATP, 1 Ci of [␥-32 P]ATP, 2 mM protein kinase inhibitor (Sigma), and 0.5 mM EGTA and myelin basic protein (1 lg/ml)(Sigma) as substrate). After incubation for 15 min at 20°C, the reaction was stopped by adding 10 l of stop solution, containing 0.6% HCl, 1 mM ATP, and 1% bovine serum albumin. Portions (30 l) of reaction mixtures were spotted on P-81 paper (Whatman), which was then washed three times with 0.5% phosphoric acid and once with acetone, and the associated radioactivity was determined by liquid scintillation counter.

Effects of Interaction of Integrins with ECM Proteins on the Tyrosine Phosphorylation of SHPS-1 and Its Association with SHP-2-
The effects of cell adhesion on tyrosine phosphorylation of SHPS-1 and its interaction with SHP-2 were investigated by detaching serum-deprived Rat-1 cells and then replating them on fibronectin-coated dishes. After incubation for 0 -120 min, cells were lysed and subjected to immunoprecipitation with mAb 2F34 to rat SHPS-1 followed by immunoblot analysis with either mAb PY20 to phosphotyrosine or antibodies to SHP-2. Cell detachment resulted in the loss of tyrosine phosphorylation of SHPS-1 and its association with SHP-2 normally apparent in adherent cells (Fig. 1). After replating of the cells onto fibronectin-coated dishes, both the tyrosine phosphorylation of SHPS-1 and its association with SHP-2 were evident within 15 min, as the cells started to spread, and were maximal at 60 -120 min. Similar time courses of tyrosine phosphorylation of SHPS-1 and its association with SHP-2 were also observed when CHO-IR cells were plated on fibronectincoated dishes (data not shown).
Serum-deprived Rat-1 cells were also detached and either maintained in suspension or replated on dishes coated with various ECM components and kept for 60 min (Fig. 2A). Replating of cells onto dishes coated with fibronectin or laminin markedly increased tyrosine phosphorylation of SHPS-1 and its association with SHP-2, compared with replating cells onto uncoated dishes. Furthermore, replating cells onto dishes coated with poly-L-lysine, a nonspecific substrate, or with collagen type I did not substantially increase tyrosine phosphorylation of SHPS-1 relative to that apparent in cells replated onto uncoated dishes. The extent of tyrosine phosphorylation of SHPS-1 in cells replated onto dishes coated with collagen type IV was slightly greater than that apparent in cells replated onto uncoated dishes. Because fibronectin and laminin are specific ligands for integrins (47), these results suggest that the engagement of integrins by components of the ECM may induce tyrosine phosphorylation of SHPS-1 and its association with SHP-2. Pretreatment of Rat-1 cells with cytochalasin D, which disrupts actin polymerization, inhibited fibronectin-induced tyrosine phosphorylation of SHPS-1 and its association with SHP-2 (Fig. 2B), suggesting that these effects of fibronectin require normal cytoskeletal organization.

FIG. 2. Effects of cell adhesion to various ECM components (A) and of cytochalasin D (B) on tyrosine phosphorylation of SHPS-1 and its association with SHP-2.
A, detached Rat-1 cells were maintained in suspension (Susp.) or were replated on either normal plastic dishes (ECM(Ϫ)) or dishes coated with various ECM components or poly-L-lysine and incubated for 1 h. Cells were then harvested and subjected to immunoprecipitation with mAb 2F34, followed by immunoblot analysis with HRP-conjugated PY20 or polyclonal antibodies to SHP-2. B, Rat-1 cells in suspension were incubated for 10 min in the absence or presence of 3 M cytochalasin D before replating on fibronectin-coated dishes. After further incubation for 30 min in the continued absence or presence of cytochalasin D, cell lysates were prepared and subjected to immunoprecipitation with mAb 2F34, followed by immunoblot analysis with HRP-conjugated mAb PY20 or polyclonal antibodies to SHP-2. kinase, and the subsequent tyrosine phosphorylation of various proteins localized at focal contacts (9 -11). Given that SHPS-1 may serve as a substrate for v-Src kinase (19,38), Src or another Src family kinase such as Fyn might be responsible for the tyrosine phosphorylation of SHPS-1 in response to fibronectin stimulation. To investigate this hypothesis, we examined the effects of overexpression of wild-type or kinase-inactive mutant Csk proteins on fibronectin-induced tyrosine phosphorylation of SHPS-1. Csk is an Src-like tyrosine kinase that inhibits the activity of Src family kinases by phosphorylating a COOH-terminal tyrosine residue (48,49). The wild-type Csk cDNA and the mutant Csk cDNA, which contained a point mutation that changed Lys 222 to Arg, were transfected separately into CHO-IR cells, and cell lines that overexpress the wild-type (Csk-WT cells) or catalytically inactive (Csk-K/R cells) Csk were established. The Lys 222 to Arg mutation completely abolished the kinase activity of recombinant Csk (data not shown). The amount of total Csk protein in Csk-WT and Csk-K/R cells was ϳ15 and 18 times that in the parental CHO-IR cells, respectively (Fig. 3A). Replating of parental CHO-IR cells on fibronectin-coated dishes induced the tyrosine phosphorylation of SHPS-1 (Fig. 3B) and its association with SHP-2 (data not shown). Overexpression of Csk-WT markedly reduced the extent of tyrosine phosphorylation of SHPS-1 induced by fibronectin stimulation, whereas overexpression of Csk-K/R enhanced the effect of fibronectin on the tyrosine phosphorylation of SHPS-1.
To examine further the role of Src family kinases in the tyrosine phosphorylation of SHPS-1 induced by fibronectin stimulation, we studied cells derived from Src or Fyn knock-out mice (43). Because antibodies that immunoprecipitate mouse SHPS-1 bound to SHP-2 were not available, we determined the extent of tyrosine phosphorylation of SHPS-1 by immunoprecipitation with antibodies to SHP-2. Fibronectin-induced tyrosine phosphorylation of ϳ120-kDa protein, which presumably was SHPS-1 and bound to SHP-2, was markedly reduced in adherent Src Ϫ which lack Src, relative to that in adherent wild-type cells (Fig. 4A). The extent of tyrosine phosphorylation of 120-kDa SHPS-1 was also reduced in adherent Fyn Ϫ cells compared with that in adherent wild-type cells. We also studied cells derived from Fak knock-out mice (44) to investigate the role of Fak in fibronectin-stimulated tyrosine phosphorylation of SHPS-1. Whereas adherence to a fibronectin-coated dish increased the amount of tyrosine-phosphorylated 120-kDa SHPS-1 associated with SHP-2 in wild-type cells, tyrosinephosphorylated 120-kDa SHPS-1 was not detected in SHP-2 immunoprecipitates from adherent Fak Ϫ cells. In contrast to SHPS-1, the extent of tyrosine phosphorylation of paxillin was markedly increased in adherent Fak Ϫ cells relative to that apparent in adherent wild-type cells (data not shown), consistent with previous observations (50). To confirm Src-, Fyn-, or Fak-deficient effect on fibronectin-induced tyrosine phosphorylation of SHPS-1, we transiently expressed rat SHPS-1 in Src Ϫ , Fyn Ϫ , or Fak Ϫ cells (Fig. 4B). The extent of tyrosine phosphorylation of SHPS-1, which was exogenously expressed and immunoprecipitated with mAb 2F34 specific to rat SHPS-1, in adherent Src Ϫ , Fyn Ϫ , and Fak Ϫ cells, was markedly reduced relative to that in adherent wild-type cells. These data suggest that Src family kinases such as Src and Fyn may contribute to fibronectin-induced tyrosine phosphorylation of SHPS-1 and that Fak may be essential for tyrosine phosphorylation of SHPS-1 in response to fibronectin stimulation.
We next examined whether Src or Fak catalyze the tyrosine phosphorylation of SHPS-1 in vitro. Incubation of a GST fusion protein containing the cytoplasmic domain of SHPS-1 (GST-SHPS-1-cyto) with Src kinase immunoprecipitated from Rat-1 cells resulted in the phosphorylation of both GST-SHPS-1-cyto (Fig. 5A) and Src (Fig. 5C). Recombinant c-Src also catalyzed the tyrosine phosphorylation of GST-SHPS-1-cyto in vitro (data not shown). In contrast, HA-tagged Fak immunoprecipitated from transfected COS-7 cells with antibodies to HA catalyzed the phosphorylation of poly(Glu, Tyr) (Fig. 5A) and Fak (Fig.  5B), but not that of GST-SHPS-1-cyto (Fig. 5A). A, cell lines that lack Src, Fyn, Fak, or various combinations thereof and corresponding wild-type cells were replated on fibronectin-coated dishes and incubated for 1 h, after which cell lysates were prepared and subjected to immunoprecipitation (IP) with polyclonal antibodies to SHP-2. The immunoprecipitates were then subjected to immunoblot analysis with HRP-conjugated PY20 (upper panel) or polyclonal antibodies to SHP-2 (lower panel). B, wild-type cells and cells lacking Src, Fyn, or Fak were transiently transfected with a pSR␣ vector containing the full-length rat SHPS-1 cDNA. The transfected cells were then replated on fibronectin-coated dishes and incubated for 1 h, after which cell lysates were prepared and subjected to immunoprecipitation with mAb 2F34 to rat SHPS-1. The immunoprecipitates were then subjected to immunoblot analysis with HRP-conjugated PY20 (upper panel) and polyclonal antibodies to SHPS-1 (lower panel).

Effects of Fibronectin Stimulation on MAP Kinase Activation in Rat-1 Cells That Overexpress a Catalytically
have found that Rat-1 cells are more suitable than CHO cells for studying fibronectin activation of MAP kinase, we generated Rat-1 cell lines that overexpress a catalytically inactive SHP-2 (Rat-1-SHP-2-C/S cells) or wild-type SHPS-1 (Rat-1-SHPS-1-WT cells). The levels of expression of SHP-2 and SHPS-1 in parental cells, Rat-1-SHP-2-C/S cells, and Rat-1-SHPS-1-WT cells are shown in Fig. 6A. Immunoblot analysis with mAb PY20 of cell lysates (Fig. 6A) or of immunoprecipitates prepared with mAb 2F34 (data not shown) revealed that the extent of tyrosine phosphorylation of the ϳ120-kDa SHPS-1 was markedly increased in Rat-1-SHP-2-C/S cells compared with that in parental cells, consistent with previous observations with CHO cells (38,42). Replating of parental Rat-1 cells on fibronectin-coated dishes resulted in activation of MAP kinase within 15 min, with the maximal response apparent at 30 min (Fig. 6, B and C). In contrast, the extent of fibronectin-induced MAP kinase activation was markedly reduced in Rat-1-SHP-2-C/S cells. Furthermore, the overexpression of wild-type SHPS-1 did not markedly enhance MAP kinase activation in response to fibronectin.
The binding of integrins to fibronectin induces autophosphorylation of Fak, the tyrosine phosphorylation of Tyr 925 in the COOH-terminal region of Fak by Src kinase, and the subse-quent association of Grb2 with the phosphorylated Tyr 925 residue (7,8). The extent of tyrosine phosphorylation of Fak in response to 30-min fibronectin stimulation was similar in Rat-1 and Rat-1-SHPS-1-WT cells but slightly increased in Rat-1-SHP-2-C/S cells (Fig. 6D). The identical result was obtained in 15-min fibronectin-stimulated cells (data not shown). Furthermore, the fibronectin-induced formation of a Fak⅐Grb2 complex was not affected by overexpression of SHPS-1-WT and slightly enhanced by the overexpression of SHP-2-C/S. DISCUSSION In the present study, we have characterized the cell adhesion-induced tyrosine phosphorylation of SHPS-1 and its subsequent association with SHP-2. Cell adhesion to ECM components such as fibronectin and laminin, but not to poly-L-lysine, a nonspecific substrate, markedly increased the extent of tyrosine phosphorylation of SHPS-1. Because fibronectin and laminin are specific ligands for integrins (47), our results indicate that the engagement of integrins by ECM proteins may induce this response. The binding of integrins to the ECM has previously been shown to increase the extent of tyrosine phosphorylation of various proteins, including Fak, paxillin, tensin, and p130 cas (9,(51)(52)(53), that are localized at focal adhesion contacts. Thus, SHPS-1 might be a member of the group of proteins that are localized at focal adhesion contacts and undergo tyrosine phosphorylation in response to the interaction of integrins with the ECM. The fibronectin-stimulated tyrosine phosphorylation of SHPS-1 was inhibited by cytochalasin D. Integrin aggregation has been previously shown to induce localized membrane accumulation of SHP-2, and this effect is inhibited by cytochalasin D or tyrosine kinase inhibitors such as herbimycin A or genistein (54). Thus, our data suggest that tyrosine phosphorylation of SHPS-1 contributes, at least in part, to integrininduced membrane accumulation of SHP-2, as postulated by Miyamoto et al. (54).
We have also explored the mechanism by which cell adhesion to fibronectin stimulates the tyrosine phosphorylation of SHPS-1. Overexpression of Csk markedly inhibited this effect of fibronectin, whereas the expression of a kinase-inactive Csk enhanced this response. Csk inhibits the activity of Src family kinases by phosphorylating a COOH-terminal tyrosine residue in these enzymes (48,49). Thus, the overexpression of Csk may reduce the extent of fibronectin-induced tyrosine phosphorylation of SHPS-1 by inhibiting the activity of an Src family kinase. The overexpression of the kinase-negative Csk may enhance this effect of fibronectin by competitively inhibiting the action of endogenous Csk and thereby increasing the activity of an Src family kinase. This notion is also supported by our observation that the extent of fibronectin-stimulated tyrosine phosphorylation of SHPS-1 was reduced in Src-or Fyn-deficient cells, suggesting that Src and Fyn may contribute to the tyrosine phosphorylation of SHPS-1 in response to cell adhesion.
Fibronectin stimulation also failed to induce tyrosine phosphorylation of SHPS-1 in Fak-deficient cells. Engagement of integrins by the ECM induces autophosphorylation of Fak and the consequent interaction of Fak with Src kinase, mediated by binding of the SH2 domain of Src to phosphorylated Tyr 397 of Fak (11,55). This interaction of Fak with Src is thought to recruit Src to focal adhesion contacts and to enhance its kinase activity (7,8). Src may then promote the tyrosine phosphorylation of the kinase domain of Fak and thereby increase Fak protein tyrosine kinase activity (46). It is therefore possible that Fak activated as a result of fibronectin stimulation could directly catalyze the tyrosine phosphorylation of SHPS-1. However, Fak, unlike Src, did not phosphorylate the cytoplasmic portion of SHPS-1 in vitro. Together, these observations sug- 1 and 3) and of COS-7 cells that had been transfected with the pRC CMV vector containing cDNA for HA epitope-tagged mouse Fak (lanes 2 and 4) were subjected to immunoprecipitation (IP) with a mAb to the HA epitope. Lysates prepared from Rat-1 cells were also subjected to immunoprecipitation with normal mouse IgG (lane 5) or a mAb to Src (lane 6). The immunoprecipitates from both cell types were then incubated with poly(Glu, Tyr) (PGT) (lanes 1 and 2) or a GST fusion protein containing the cytoplasmic portion of SHPS-1 (SHPS-1-cyto) (lanes 3-6) in kinase assay buffer as described under "Experimental Procedures." The supernatant of each phosphorylation reaction mixture was then subjected to SDS-polyacrylamide gel electrophoresis and autoradiography. B and C, the immunoprecipitates from the transfected COS-7 cells or Rat-1 cells in A were also subjected to immunoblot analysis with antibodies to Fak (␣-Fak) or to Src (␣-Src), respectively (upper panels), or to gel electrophoresis and autoradiography for detection of incorporated 32 P (lower panels). ␣-HA, mAb 12CA5 to the HA epitope tag; NMG, normal mouse IgG.

FIG. 5. Effects of Src and Fak on the phosphorylation of SHPS-1 in vitro. A, lysates of control COS-7 cells (lanes
gest that Fak is essential for tyrosine phosphorylation of SHPS-1 in response to fibronectin but that Fak does not directly catalyze this reaction. The importance of Fak may lie in either its enhancement of the catalytic activity of Src or its recruitment of Src to focal contacts. Because SHPS-1 appears to be an effective substrate for Src kinase in vitro, activated Src family kinases may catalyze the tyrosine phosphorylation of SHPS-1 in vivo. Regardless, both Src family kinases and Fak appear to be required for fibronectin-stimulated tyrosine phosphorylation of SHPS-1 and its association with SHP-2. In contrast to the phosphorylation of SHPS-1, loss of Fak did not inhibit, but rather enhanced, fibronectin-stimulated tyrosine phosphorylation of paxillin, as described previously (50). This observation suggests that fibronectin-induced tyrosine phosphorylation of SHPS-1 and that of paxillin are mediated by different mechanisms, even though tyrosine phosphorylation of paxillin is induced by integrin stimulation and catalyzed by a transforming variant of Src kinase (56).
SHP-2 has been suggested to mediate activation of the Ras-MAP kinase cascade in response to various growth factors, including insulin, EGF, platelet-derived growth factor, and FGF (16,17). Thus, our observation that cell adhesion to fibronectin induces tyrosine phosphorylation of SHPS-1 and its association with SHP-2 prompted us to examine the role of the SHPS-1⅐SHP-2 complex in fibronectin-induced activation of MAP kinase. We have now shown that expression of a catalytically inactive SHP-2 markedly inhibited MAP kinase activation in response to fibronectin stimulation, suggesting that SHP-2 may mediate the integrin-induced activation of MAP kinase as well as that induced by receptor protein tyrosine kinases. Fibronectin activation of MAP kinase is thought to be triggered by autophosphorylation of Fak-and Src-mediated phosphorylation of Tyr 925 of Fak, which creates a binding site for the SH2 domain of Grb2 (7,8). However, the expression of catalytically inactive SHP-2 did not substantially inhibit either the tyrosine phosphorylation of Fak or its association with Grb2 in response to fibronectin stimulation. Thus, SHP-2 may mediate fibronectin activation of MAP kinase through an as yet unidentified mechanism that is independent of the interaction of Fak with Grb2.
The overexpression of wild-type SHPS-1 did not markedly enhance MAP kinase activation in response to fibronectin stimulation, although it did increase the fibronectin-induced association of SHPS-1 with SHP-2 compared with that apparent in the parental cells (data not shown). However, it remains possible that endogenous SHPS-1 present in Rat-1 cells is sufficient for maximal MAP kinase activation in response to fibronectin, so that overexpression of exogenous SHPS-1 cannot further enhance this effect. The expression of a membranetargeted form of Corkscrew, a Drosophila homolog of SHP-2, can bypass Sevenless tyrosine kinase function to mediate development of the R7 photoreceptor (25), suggesting that recruitment of Corkscrew or SHP-2 to the plasma membrane is FIG. 6. Effects of overexpression of a catalytically inactive SHP-2 or of wild-type SHPS-1 on activation of MAP kinase, tyrosine phosphorylation of Fak, and the association of Fak with Grb2 in response to cell adhesion to fibronectin. A, lysates of parental Rat-1 cells or of Rat-1 cells overexpressing SHP-2-C/S or SHPS-1-WT were subjected to immunoblot analysis with polyclonal antibodies to SHPS-1, polyclonal antibodies to SHP-2, or HRP-conjugated PY20, as indicated. B, Rat-1 cells (top panels), Rat-1-SHP-2-C/S cells (middle panels), or Rat-1-SHPS-1-WT cells (bottom panels) were replated on fibronectin-coated dishes and incubated for the indicated times, after which cell lysates were prepared and subjected to immunoblot analysis with antibodies specific for tyrosine-phosphorylated MAP kinase (␣-pMAPK) (left panels). The same blots were also probed with ␣91 polyclonal antibodies to p44 and p42 MAP kinase to ensure that equal amounts of MAP kinase were present in each lane (right panels). The positions of p44 and p42 MAP kinase are indicated by arrowheads. C, Rat-1 cells, Rat-1-SHP-2-C/S cells, or Rat-1-SHPS-1-WT cells were maintained in suspension (Susp.) (open column) or replated on fibronectin (FN)-coated dishes (closed column) for 15 min, after which cell lysates were prepared and subjected to immunoprecipitation with ␣92 antibodies to MAP kinase, and MAP kinase activity in immunoprecipitates was assayed with myelin basic protein as the substrate. Data are means of duplicate determinations and are a representative of three experiments. D, Rat-1 cells, Rat-1-SHP-2-C/S cells, or Rat-1-SHPS-1-WT cells were maintained in suspension (Susp.) or replated on fibronectin (FN)-coated dishes. After 30 min, cell lysates were prepared and subjected to immunoprecipitation with polyclonal antibodies to Fak, followed by immunoblot analysis with HRP-conjugated PY20 or a mAb to Grb2 (␣-Grb2). The Fak region of the blot was also reprobed with a mAb to Fak. crucial for its function. Therefore, SHPS-1 likely plays an important role in recruitment of SHP-2 from the cytosol to a site near the plasma membrane in response to fibronectin stimulation. A phosphotyrosyl peptide containing the sequence surrounding the binding sites for the SH2 domains of SHP-2 in SHPS-1 stimulates the PTPase activity of recombinant SHP-2 in vitro (57), suggesting that binding of SHP-2 to tyrosinephosphorylated SHPS-1 in response to fibronectin stimulation may increase the catalytic activity of SHP-2 in vivo. SHP-2 may subsequently dephosphorylate SHPS-1 and activate an unknown factor, which also associates with SHPS-1 or is located near the plasma membrane, by catalyzing its tyrosine dephosphorylation, resulting in activation of the Ras-MAP kinase cascade. Identification of this putative factor should facilitate our understanding of the mechanism of SHP-2-mediated activation of the Ras-MAP kinase cascade.