Protein-tyrosine Phosphatase 1D Modulates Its Own State of Tyrosine Phosphorylation*

The insulin receptor-mediated signal transduction pathway involves insulin receptor substrate 1 and a variety of proteins containing Src homology-2 (SH2) domains, such as phosphatidylinositol 3-kinase, Grb2, and protein-tyrosine phosphatase 1D (PTP1D). Upon insulin stimulation of baby hamster kidney cells overexpressing the IR, the catalytically inactive mutant of PTP1D, C463A, becomes tyrosine-phosphorylated and coprecipitates with Grb2. Tyrosine phosphorylation of this mutant is significantly reduced when wild type PTP1D is coexpressed. Substitution of tyrosine residues 546 and 584 with phenylalanine abrogates tyrosine phosphorylation of the catalytically inactive mutant and abolishes its interaction with Grb2. The insulin receptor (IR) 1 belongs to a family of structurally related transmembrane growth factor receptors that exhibit intrinsic tyrosine kinase activity (1, 2). Ligand stimulation induces rapid activation of the kinase activity, resulting in autophosphorylation of the (cid:98) -chain of the receptor itself and phosphorylation of several other cytoplasmic proteins (1, 3), including insulin receptor substrate 1 (IRS1) (4), the p85 reg-ulatory subunit of phosphatidylinositol 3-kinase (5, 6), Grb2 (7, 8), Nck (9), and protein-tyrosine phosphatase 1D (PTP1D) (10). PTP1D

We show here that the catalytically inactive mutant of PTP1D, C463A (32), becomes phosphorylated on tyrosine and coprecipitates with Grb2 after insulin treatment in BHK-IR cells. Furthermore, we demonstrate that tyrosines at positions 546 and 584 are responsible for this phosphorylation and that substitution of these residues with phenylalanine abolishes Grb2 association. Coexpression of wild type PTP1D and the catalytically inactive mutant leads to significantly reduced tyrosine phosphorylation of the latter, suggesting that PTP1D may modulate its own tyrosine phosphorylation state. Our findings provide new insights into mechanisms of PTP1D action within the insulin receptor signal.

EXPERIMENTAL PROCEDURES
Reagents, Antibodies, and Plasmid Constructs-Protein A-and glutathione-Sepharose were purchased from Pharmacia Biotech Inc. All other reagents were obtained from Sigma.
The construction of cytomegalovirus promoter-based expression plasmids for PTP1D has been described previously (11,33).
Immunoprecipitation and Western Blotting-For immunoprecipitation, cell lysates were incubated for 4 h at 4°C with polyclonal anti-PTP1D antibody or with monoclonal anti-hemagglutinin antibody that had been bound to protein A-Sepharose beads (30 l of beads). The beads were then washed three times with 1.0 ml of buffer containing 50 mM HEPES, pH 7.5, 150 mM NaCl, 0.1% Triton X-100, 10% glycerol, 1 mM EDTA, 2 mM sodium orthovanadate, 10 mM sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin (HNTG buffer), suspended in SDS sample buffer, boiled for 5 min, and subjected to gel electrophoresis.
Following SDS-polyacrylamide gel electrophoresis, proteins were transferred to a nitrocellulose membrane (Schleicher & Schuell) and immunoblotted.
For binding assays, purified Grb2-GST (glutathione S-transferase) fusion protein was coupled to glutathione-Sepharose. Approximately 2 g of Grb2-GST was incubated with 10 l of glutathione-Sepharose beads and rotated for 2 h at 4°C. Beads were then washed three times with 1 ml of HNTG buffer, and Grb2-coupled beads were incubated with BHK-IR lysates for 4 h at 4°C. * This work was supported by a grant from Sugen, Inc. 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. ‡ These two authors made an equal contribution to this work. § To whom all correspondence should be addressed. Tel.: 49-89-8578-2513 (ext. 2778); Fax: 49-89-857-7866. 1 The abbreviations used are: IR, insulin receptor; IRS1, insulin receptor substrate 1; PTP1D, protein-tyrosine phosphatase 1D; BHK, baby hamster kidney; BHK-IR, BHK cells overexpressing the IR; HA, hemagglutinin.

RESULTS AND DISCUSSION
After stimulation with growth factors, the SH2-containing protein-tyrosine phosphatase 1D associates with various tyrosine kinases and becomes tyrosine-phosphorylated (11,15,18). PTP1D association with the IR has been observed in vitro (19 -22) and in vivo, 2 but tyrosine phosphorylation of the phosphatase has not yet been detected in intact cells (10,28). To investigate the role of PTP1D in the insulin signal, we transiently expressed the wild type phosphatase and the catalytically inactive mutant C463A in BHK-IR cells. We found that overexpressed PTP1D was weakly phosphorylated on tyrosine after insulin stimulation (Fig. 1, lanes 6 -10). Surprisingly, the PTP1D phosphorylation state was considerably higher when the C463A mutant was used (Fig. 1, lanes 11-15). Reprobing with anti-PTP1D antibody demonstrated that comparable amounts of protein were present in all relevant lanes (Fig. 1,  lower panel).
To determine the identity of the phosphorylated tyrosine residues, we generated tyrosine mutations in the inactive form of PTP1D using standard in vitro mutagenesis techniques. These Tyr/Phe mutants were then transiently expressed in BHK-IR cells. A single Tyr/Phe substitution at position 546 of the catalytically inactive phosphatase resulted in a significant decrease in phosphotyrosine content, as shown in Fig. 2 (upper  panel, lane 10). The combination of two tyrosine substitutions in the C-terminal tail (Y546/584F, Fig. 2, lower panel) abolished beyond detection tyrosine phosphorylation of the inactive form of PTP1D (Fig. 2, upper panel, lane 8). Reprobing with anti-PTP1D antibody demonstrated again that the observed differences were not due to different levels of protein expression (Fig. 2, middle panel).
To investigate the functional significance of the identified tyrosine phosphorylation sites, we incubated aliquots of the transfected BHK-IR lysates with a Grb2-GST fusion protein and determined the interaction with the different PTP1D forms by coprecipitation. As indicated in Fig. 3, the amount of associated catalytically inactive PTP1D C463A was markedly increased in comparison with wild type PTP1D. Substitution of tyrosine 546 with phenylalanine (Y546F) reduced this association dramatically and mutation of both tyrosine residues in 1D (Y546/584F) abolished the interaction between Grb2 and PTP1D.
The observed high level of tyrosine phosphorylation in the C463A mutant in contrast to wild type PTP1D suggested the possibility that the phosphatase activity of PTP1D may modulate its own state of tyrosine phosphorylation. To investigate this possibility further, we coexpressed the catalytically active and inactive forms of PTP1D in BHK-IR cells. In order to be able to distinguish between the two forms, the C463A mutant was modified with a hemagglutinin (HA) tag at the C terminus, permitting selective immunoprecipitation of this form of PTP1D with an anti-HA antibody. The coexpression of tagged and untagged inactive mutants served as a control. To verify the specificity of the hemagglutinin antibody, the untagged C463A variant was first expressed alone. As shown in Fig. 4  (lanes 6 and 8), tyrosine phosphorylation of the HA-tagged, inactive C463A-HA was clearly lower when coexpressed with wild type PTP1D rather than with the untagged inactive PTP1D mutant. These results demonstrate dephosphorylation of the tagged C463A-HA mutant by active PTP1D and therefore suggest an intermolecular transdephosphorylation mechanism. We cannot, however, rule out the possibilities of at least a partial contribution of an intramolecular mechanism or the involvement of other protein-tyrosine phosphatases of un-  Taken together, our data demonstrate that catalytically inactive PTP1D becomes markedly tyrosine-phosphorylated at positions 546 and 584 upon stimulation of BHK-IR cells with insulin. These tyrosine phosphorylation targets match those utilized after ␤-platelet-derived growth factor receptor activation (23). Moreover, we demonstrate that the inactive C463A mutant coprecipitates more efficiently with Grb2 after insulin stimulation relative to wild type PTP1D and that tyrosine residues 546 and 584 are required for the PTP/Grb2 interaction. For tyrosine 546, such an involvement in the interaction with Grb2 has been shown previously in the context of other signal transduction pathways (23,24). Furthermore, we demonstrate that inactive PTP1D, tyrosine-phosphorylated in response to BHK-IR cell stimulation with insulin, is efficiently dephosphorylated when coexpressed with its catalytically active wild type homolog, supporting a mechanistic model in which PTP1D is able to modulate its own phosphorylation and activation state as part of the insulin-triggered intracellular signal. On the basis of in vitro data, such a possibility had been suggested previously (37).
After our earlier proposal of an important role for PTP1D in growth factor signal transmission and regulation (11), which was supported recently by the demonstration of its involvement in insulin-induced Ras and MAP kinase activation (28,29,31), the observations presented here contribute important insights toward the understanding of molecular mechanisms governing the action of biological signal transmitters such as growth factors and insulin.  1, 3, 5, and 7). After cell lysis, the tagged PTP1D protein was precipitated with anti-hemagglutinin antibody, and following gel electrophoresis, the tyrosine phosphorylation pattern was detected by immunoblotting with an anti-phosphotyrosine antibody (upper panel). The same blot was reprobed with monoclonal anti-PTP1D antibody (lower panel). IgG, immunoglobulin heavy chain.