Shp2, an SH2-containing protein-tyrosine phosphatase, positively regulates receptor tyrosine kinase signaling by dephosphorylating and inactivating the inhibitor Sprouty.

Src homology 2-containing phosphotyrosine phosphatase (Shp2) functions as a positive effector in receptor tyrosine kinase (RTK) signaling immediately proximal to activated receptors. However, neither its physiological substrate(s) nor its mechanism of action in RTK signaling has been defined. In this study, we demonstrate that Sprouty (Spry) is a possible target of Shp2. Spry acts as a conserved inhibitor of RTK signaling, and tyrosine phosphorylation of Spry is indispensable for its inhibitory activity. Shp2 was able to dephosphorylate fibroblast growth factor receptor-induced phosphotyrosines on Spry both in vivo and in vitro. Shp2-mediated dephosphorylation of Spry resulted in dissociation of Spry from Grb2. Furthermore, Shp2 could reverse the inhibitory effect of Spry on FGF-induced neurite outgrowth and MAP kinase activation. These findings suggest that Shp2 acts as a positive regulator in RTK signaling by dephosphorylating and inactivating Spry.

Receptor tyrosine kinase signaling regulates a wide variety of biological processes in response to extracellular signals, including cellular growth, differentiation and metabolism (1)(2)(3)(4). For example, fibroblast growth factor (FGF) 1 stimulates the receptor tyrosine kinase activity of the FGF receptor (FGFR), leading to tyrosine phosphorylation of the docking protein FRS2, consequent recruitment of multiple Grb2-Sos complexes, and activation of the Ras-mitogen-activated protein (MAP) kinase signaling pathway (5)(6)(7). Precise control of MAP kinase activation is critical for such cellular responses, and several molecules have been identified that serve to regulate the activation of MAP kinase either positively or negatively.
Shp2 is a widely expressed protein-tyrosine phosphatase (PTPase) that seems to play a positive role in the activation of MAP kinase in response to growth factors (8 -14). In response to growth factor stimulation, Shp2 becomes tyrosine-phosphorylated and binds to Grb2, where it seems to act as an adaptor protein to recruit the Grb2-Sos complex to the plasma membrane. This, in turn, leads to Ras activation (6,9,(15)(16)(17). However, several studies have shown that mutation of Shp2 in the putative Grb2 binding site did not interfere with the function of Shp2 in mediating MAP kinase activation (12,18). In contrast, several lines of evidence indicate that the phosphatase catalytic activity of Shp2 is required for MAP kinase activation by FGF and other growth factors (12). Thus, two critical tasks are identifying the substrate for Shp2 and determining how dephosphorylation contributes to the activation of MAP kinase. Recently, it has been reported that a major binding protein of Shp2 is the multi-site docking protein Gab1 (Grb2-associated binder 1) and that this association is required for MAP kinase activation by several growth factors (19 -22). Interestingly, genetic analysis of Corkscrew (Csw), the Drosophila homologue of Shp2, has led to the identification of the Daughter of Sevenless (Dos) protein as a putative Csw substrate (23)(24)(25). The Dos protein is structurally similar to Gab1. More recently, it has been reported that the primary function of Gab1 association with Shp2 is to target the PTPase activity of the latter to the membrane (26). However, the substrate of Shp2 PTPase as yet remains unclear.
Recently, we have found that Sprouty (Spry), which is a conserved inhibitor of the Ras-MAP kinase signaling pathway, becomes tyrosine-phosphorylated in response to growth factors. This tyrosine phosphorylation is required for Spry inhibitor activity (27,28). Spry binds to the SH2 domain of Grb2 in response to FGF stimulation and thereby inhibits the recruitment of the Grb2-Sos complex to FRS2 or Shp2, effectively suppressing Ras-MAP kinase signaling (27). Because Shp2 acts immediately proximal to activated receptors, and its PTPase activity is required for MAP kinase activation by FGF, we hypothesized that Spry is a possible target of Shp2. In this study, we show that Shp2 can inactivate Spry by dephosphorylation and thereby prolong the activation of MAP kinase in response to FGF. This result suggests that Spry may be a target of Shp2 in Ras-MAP kinase signaling. Furthermore, our results suggest that the inhibitor activity of Spry is rapidly and reversibly controlled at the post-translational level, thus representing a novel type of negative feedback mechanism in receptor tyrosine kinase signal transduction.

Construction of the Constitutively Active and Catalytically Inactive
Mutants of Shp2-In the constitutively active and catalytically inactive mutants of Shp2 (E76A and C459S), Glu-76 and Cys-459 were replaced by Ala and Ser, respectively. These mutants were constructed by PCRbased mutagenesis.
Bacterial Expression and Purification of Recombinant Shp2-For bacterial expression, the open reading frame of Shp2 (WT, E76A, or C459S, respectively) was amplified by PCR and subcloned into pGEX6P3 (Amersham Biosciences). Escherichia coli cells transformed with pGEX-Shp2 WT, E76A, or C459S were grown overnight to saturation in 10 ml of LB medium containing 50 g/ml ampicillin. The cells were grown in 3 liters of LB medium containing 50 g/ml ampicillin at 37°C to reach A 600 ϭ 0.6. One hour after the temperature shift to 25°C, isopropyl-␤-D-thiogalactopyranoside was added to a final concentration of 0.5 mM, and cells were cultured for 12 h. Purification of glutathione S-transferase-Shp2 WT, E76A, or C459S was performed by the method described previously (29).
Luciferase Assay-Cells were treated for 15-24 h with or without 25 ng/ml bFGF, and the luciferase activity in cell lysates was measured by using a luciferase assay system (Promega) in a Berthold Lumat LB 9507 luminometer. Relative luciferase activities were normalized by co-expressed ␤-galactosidase activities.
Neurite Outgrowth-PC12 cells possessing one or more neurites of a length of more than 1.5-fold the diameter of the cell body were scored as positive. More than 200 cells were scored in each point and the averages from three independent experiments are shown.

Shp2 Promotes Tyrosine Dephosphorylation of Spry in
Vivo-To test the hypothesis that Spry is a possible target of Shp2, we first determined the effect of Shp2 on tyrosine phosphorylation of mouse Spry2 (mSpry2). As Shp2 is normally inactive because of auto-inhibition by its N-terminal SH2 domain (30), we used a constitutively active mutant of Shp2 (Shp2 E76A), in which Glu-76 was replaced by Ala. C2C12 cells were co-transfected with FGFR, Myc-mSpry2, and increasing amounts of wild-type HA-Shp2 (Shp2 WT) or HA-Shp2 E76A. After 24h, the cell extracts were subjected to immunoprecipitation with anti-Myc antibody followed by immunoblotting with anti-pTyr antibody to detect tyrosine phosphorylation levels in the mSpry2 protein. As reported previously (27), mSpry2 was tyrosine phosphorylated in an FGFR-dependent manner (Fig. 1A, lane 2). Co-expression of Shp2 WT resulted in reduced FGFR-dependent tyrosine phosphorylation of mSpry2 in a dosedependent manner (Fig. 1, lanes 3 and 4). Co-expression of Shp2 E76A also decreased the levels of tyrosine phosphorylation on mSpry2 drastically (Fig. 1, lanes 5 and 6).
Next, we tested the effect of Shp2 on Xenopus Spry1 (xSpry1). Similarly to mSpry2, expression of FGFR induced tyrosine phosphorylation on xSpry1 (Fig. 1B, lane 2). Co-expression of Shp2 WT resulted in reduced FGFR-induced tyrosine phosphorylation of xSpry1 in a dose-dependent manner (lanes 3 and 4). Furthermore, co-expression of Shp2 E76A decreased the levels of tyrosine phosphorylation on xSpry1 efficiently (lanes 5 and 6). Taken together, these results sug- gest that Shp2 promotes the dephosphorylation of phosphotyrosine in Spry1/2 in vivo.
Shp2 Dephosphorylates Spry in Vitro-We examined whether Shp2 dephosphorylates Spry directly in vitro. C2C12 cells were co-transfected with Myc-xSpry1 and FGFR. Myc-xSpry1 was immunoprecipitated with anti-Myc antibody and incubated with purified recombinant Shp2 WT, Shp2 E76A, or catalytically inactive Shp2 C459S as glutathione S-transferase fusion proteins. As shown in Fig. 2, Shp2 E76A, but not Shp2 WT or Shp2 C459S, dephosphorylated xSpry1 tyrosine in a time-dependent manner. This tyrosine dephosphorylation could be inhibited in the presence of sodium vanadate, a potent inhibitor of tyrosine phosphatases. Essentially identical results were obtained with mSpry2 (data not shown). These results suggest that Spry is a possible direct target for Shp2.
Shp2 Induces Dissociation of Spry from Grb2-We have previously shown that Spry binds to Grb2 in response to FGF stimulation, resulting in the inhibition of Grb2-Sos complex recruitment to FRS2 or Shp2, upstream signaling molecules in the FGF signaling pathway (27). As the binding of tyrosinephosphorylated Spry to Grb2 is a critical step in the inhibition of the Ras-MAP kinase pathway by Spry, we examined the effect of Shp2 on the association between Spry and Grb2. C2C12 cells were co-transfected with the indicated combinations of Myc-mSpry2, Shp2 WT, Shp2 E76A, and FGFR and then immunoprecipitated with anti-Grb2 antibody. The immunoprecipitates were subjected to the immunoblotting with anti-Myc antibody to detect associated mSpry2. As shown in Fig. 3A, whereas co-expression of Shp2 WT had little effect on the binding of mSpry2 to Grb2 in response to FGFR, Shp2 E76A caused a dose-dependent decrease in the FGFR-induced association between mSpry2 and Grb2. The association between xSpry1 and Grb2 was also decreased by expression of Shp2 E76A (Fig. 3B). These results suggest that Shp2 induces dissociation of Spry from Grb2 following their association in response to FGFR signaling.
Shp2 Suppresses the Inhibition of FGF Signaling by Spry-We examined the ability of Shp2 to suppress Spry-mediated inhibition of FGF signaling, using a reporter assay that measures the transcription activity of Elk1, a nuclear target of ERK MAP kinase (31,32). PC12 cells were transfected with Gal4-Elk1 and Gal4-E1b-luciferase reporter. After the activation of ERK, the chimeric Gal4-Elk1 transcription factor is phosphorylated and thereby activated. This factor binds to the Gal4 upstream activating sequence, resulting in the induction of luciferase expression (32). ERK activation can occur downstream of the FGF signaling pathway. Treatment of cells with FGF resulted in induction of the reporter (Fig. 4). However, expression of xSpry1 or mSpry2 inhibited induction by FGF. Spry-mediated inhibition of FGF signaling was partially relieved by co-expression of Shp2 E76A but not by Shp2 WT. Taken together, these results suggest that Shp2 acts as a positive regulator in the FGF signaling pathway by dephosphorylating and inactivating the inhibitor Spry.
Shp2 and Spry Function Together to Control the Duration of MAP Kinase Activation-Because Spry controls the duration of MAP kinase activation in a tyrosine phosphorylation-dependent manner (27), we examined the effect of Shp2 on the kinetics of MAP kinase activation in response to FGF stimulation. In C2C12 cells, activation of MAP kinase was detected at 5 min and was maximal at 10 min after FGF stimulation. After this, the activation of MAP kinase decreased gradually (Fig. 5, A and  B). When xSpry1 was expressed in cells, tyrosine phosphorylation of xSpry1 peaked at 15 min after FGF treatment, and the activity of MAP kinase consequently decreased at 15-30 min after treatment (Fig. 5, A and B). Co-expression of Shp2 E76A caused a drastic decrease in FGF-induced tyrosine phosphorylation levels in xSpry1. As a result, activation of MAP kinase persisted for as long as that seen in the absence of xSpry1 expression (Fig. 5, A and B). Expression of catalytically inactive Shp2 C459S failed to block the inhibitory activity of xSpry1 on MAP kinase activation, and indeed was seen to potentiate it slightly (Fig. 5, A and B). Thus, Shp2 is able to counteract the inhibitory effect of xSpry1 on MAP kinase activation.
In PC12 cells, FGF treatment induces neurite outgrowth by means of the activation of MAP kinase (6). Therefore, we examined the effects of Shp2 and xSpry1 on neurite outgrowth in response to FGF stimulation (Fig. 6). As observed previously, expression of xSpry1 blocked neurite outgrowth induced by FGF stimulation in PC12 cells. Co-expression of Shp2 E76A, but not of Shp2 C459S, reversed the inhibitory effect of xSpry1 on neurite outgrowth in response to FGF stimulation. These results are consistent with those obtained above with respect to the effects of Shp2 and xSpry1 on FGF-mediated MAP kinase activation.

DISCUSSION
The present studies suggest a mechanism underlying the requirement of Shp2 PTPase activity in MAP kinase activation. Recently, it was reported that the recruitment of Shp2 PTPase to the membrane is required for its stimulation of MAP kinase activation (26). In response to growth factor stimulation, Spry translocates to the plasma membrane, where it binds to Grb2 in a tyrosine phosphorylation-dependent manner, resulting in the inhibition of the recruitment of the Grb2-Sos complex to FRS2 or Shp2. Thus, Shp2 has a positive role in MAP kinase activation by dephosphorylating Spry at the plasma membrane and inactivating the inhibitory activity of Spry. These findings also suggest that the co-operation between Shp2 and Spry controls the duration of MAP kinase activation, which is crucial to determining cell fate.