Communication Vol. 268, No. 29, Issue of October 15, pp. 21478-21481, 1993 THE JOURNAL OF BlomlcAL CHEMISTRY

0 1993 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

Activation of the SH2-containing Phosphotyrosine Phosphatase SH-PTP2 by Its Binding Site, Phosphotyrosine 1009, on the Human Platelet-derived Growth Factor Receptor p*

(Received for publication, July 23, 1993) Robert J. Lechleider*$n 11, Seiji Sugimoto(l**, Anton M. Bennett*, Adam S. KashishianH, Jonathan A. Cooper$*, Steve E. ShoelsonsQ llll, Christopher T. Walsh**, and Benjamin G. NeelS 11 1 ) From the $.Molecular Medicine Unit, Beth Israel Hospital, the §Division of Pulmonary Medicine, Brigham and Women’s Hospital, the **Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, and the @Joslin Diabetes Center, Harvard Medical School, Boston, Massachusetts 02115 and the $.$.Fred Hutchinson Cancer Center, Seattle, Washington 98104

Much progress has been made in elucidating early events in signal transduction by growth factor receptors with intrinsic tyrosine kinase activity. Upon ligand ad- dition, these receptors dimerize and activate, becoming phosphorylated at a number of tyrosyl residues. These phosphorylation sites serve as docking points for pro- teins containing SFC homology-2 ( S H 2 ) domains. How- ever, little is known about how phosphotyrosine phos- phatases (PTPs), participate in these events. Recently, we and others molecularly cloned a ubiquitously ex- pressed SH2 domain-containing PTP, SH-FTP2 (Syp, PTPlD, PTPgC), and found that it interacts directly with several activated growth factor receptors via its SH2 domains. Using a peptide competition assay, we now demonstrate that the major binding site for SH- PTP2 on the platelet-derived growth factor receptor is phosphotyrosine 1009. Immunoprecipitation studies in- dicate that SH-PTP2 is the previously unidentified “64- kDa” protein known to bind at this site. Addition of a phosphotyrosyl peptide comprising the region around Tyr-1009 stimulates SH-PTP2 activity 5-10-fold, whereas other phosphotyrosyl peptides from the platelet-derived growth factor receptor have no stimulatory effect. Our data suggest that binding of SH-PTP2 to the activated

* This work was supported in part by National Institutes of Health Grants CA49152 (to B. G. N.), GM20011 (to C. T. W.), and CA54786 (J.A.C.), by funds from F. Hoffmann-LaRoche Ltd. (to B. G. N. and C. T. W.) and the National Science Foundation (to S. E. S.), and by Joslin

nology Research Center Grant 36836. The costs of publication of this Diabetes Centerrnational Institutes of Health Diabetes and Endocri-

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.

Fund Postdoctoral Fellowship DRG-062. 1 Supported by Damon Runyon-Walter Winchell Cancer Research

11 Recipient of a career development award from the Juvenile Dia- /I These authors are equal contributors to this work.

betes Foundation. (1 11 Recipient of a junior faculty research award from the American

Cancer Society. To whom correspondence should be addressed. “el.: 617-735-2823; Fax: 617-735-2913.

receptor in vivo should result in stimulation of SH-PTP2 activity.

Many polypeptide growth factors signal via receptors with intrinsic protein tyrosine kinase (PTK)’ activity (1). Much pro- gress has been made in defining early events in RPTK signal- ing (2, 3). Upon ligand addition, RPTKs dimerize, are enzy- matically activated, and “autophosphorylate” at multiple sites (1). These tyrosyl phosphorylation sites serve as docking points for the recruitment of secondary signaling molecules contain- ing SH2 domains, most of which are also RPTK substrates. SH2 domains are regions of approximately 100 amino acids, first identified in src family PTKs, which direct high affinity interactions with specific Tyr(P) peptides (3-5). The specificity of SH2ITyr(P) peptide interactions has been demonstrated in vitro by competition of SH2 domain/Tyr(P) protein binding with specific Tyr(P) peptides and in vivo by the finding that different phosphotyrosine residues on RPTKs bind to different down- stream SH2-containing proteins (3-5). For example, independ- ent binding sites have been identified on the PDGFR for phos- phatidylinositol3-kinase, rasGAP, phospholipase C-y, and “64- kDa” and “120-kDa” proteins (6-12). Other RPTKs bind to different combinations of SH2-containing proteins (2-5). The strength and spectrum of these S H m ( P ) peptide interac- tions may, at least in part, determine the specificity of the cellular response to growth factors.

Although a general paradigm is emerging for early RPTK signaling events, the role of specific PTPs in modulating these pathways has remained obscure. Recently, we (13) and others (14-16) molecularly cloned a ubiquitously expressed PTP, SH- PTP2, also known as Syp (14), PTPlD (151, and PTP2C (161, containing two SH2 domains and a single phosphatase domain. SH-PTP2 becomes tyrosyl (14, 15, 17) and threonyl (17) phos- phorylated in response to growth factor stimulation, and di- rectly binds, via its N-SH2 (171, to the activated PDGFR and epidermal growth factor receptor (14,15, 17). In addition, upon insulin stimulation, SH-PTP2 binds to tyrosyl-phosphorylated insulin receptor substrate-1 in vitro (17, 18) and i n vivo (18h2 SH-PTP2 is highly similar to and is the likely mammalian homologue of Drosophila csw (19). csw acts to potentiate the actions of the Drosophila homologue of mammalian c-raf to positively transmit signals downstream of the torso RPTK (19, 20). Taken together, these data suggest an important role for SH-PTP2 in RPTK signal transduction.

Previous studies demonstrated that SH-PTP2 binds directly to activated RPTKs, but the precise site(s) of binding and the effect of binding on SH-PTP2 activity were not established. Earlier, we reported the purification of full length recombinant SH-PTP2 produced in E. coli and preliminary characterization of its enzymological properties (21). We noticed that SH-PTP2

The abbreviations used are: FTK, protein tyrosine kinase; RPTK, receptor protein tyrosine kinase; FTP, phosphotyrosine phosphatase; SH2, src homology-2; PDGF, platelet-derived growth factor; PDGFR, human PDGF receptor p; GAP, GTPase-activating protein; W T , wild type; RCM-lysozyme, reduced, carboxyamidomethylated and maley- lated lysozyme; GST, glutathione S-transferase; N-SH2, N-terminal SH2 domain; GST-N-SH2, GSTN-terminal SH2 domain fusion protein; PAGE, polyacrylamide gel electrophoresis.

R. J. Lechleider, M. Myers, M. White, and B. G. Neel, unpublished observations.

Activation of SH-PTP2 by Its PDGFR Binding Site 21479

exhibited non-Michaelis-Menten kinetics toward a Tyr(P) pep- tide encompassing Tyr-1009 of the PDGFR (PDGFR(Tyr(P1- 1009)), although it displayed typical pseudo-first order kinetics against para-nitrophenyl phosphate and the artificial protein substrate RCM-lysozyme (21). Songyang et al. (221, using a Tyr(P) peptide library screening technique, reported that the preferred binding sequence for the SH-PTP2 N-SH2 domain is: Tyr(P)-(hydrophobic or neutral amino acid)-Xaa-Val. The se- quence surrounding PDGFR Tyr(P)-1009 (pYTAV, with pY in- dicating the phosphorylated tyrosine) fits this consensus se- quence.

In the current study, we establish that the binding site for SH-PTP2 on the PDGFR is Tyr(P)-1009 and that SH-PTPP is the heretofore unidentified 64-kDa PDGFR-associated protein (11). Addition of a synthetic Tyr(P) peptide containing this site, but not other sites on the PDGFR, results in strong activation of SH-PTP2 enzymatic activity toward RCM-lysozyme. These data suggest that binding to the PDGFR in uiuo leads to enzy- matic activation of SH-PTP2.

MATERIALS AND METHODS Cell Culture-The canine epithelial cell line TRMP, which does not

express the PDGFR, and its derivatives ATWT, which overexpresses the PDGFR, and AT1009, which overexpresses a Tyr-1009 - Phe PDGFR mutant (6,101, were maintained in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum, 100 unitdml penicillin, and 100 unitdml streptomycin.

Binding Assays and Immunoprecipitations-ATWT cells were ren- dered quiescent by serum starvation for 24-48 h in Dulbecco’s modified Eagle’s medium containing 0.1% fetal bovine serum and antibiotics. Starved cells were stimulated with 50 ng/ml PDGF (Oncogene Science) for 10 min. Lysis in non-ionic detergent and clarification were carried out as described previously (171, and protein concentration was deter- mined by the bicinchoninic acid method, using a commercially available kit (Pierce Chemical Co.). Binding assays were performed with gluta- thione-agarose beads containing GST-N-SH2 (at 500 nM) and ATWT lysates (250 pg/ml) in the absence or presence of Tyr(P) peptides (at 40 PM) comprising the major tyrosyl phosphorylation sites of the PDGFR, or an unphosphorylated peptide corresponding to the sequence sur- rounding Tyr-1009 of the PDGFR (see below), in a final volume of 200 pl, as described (17). Peptides were preincubated with GST-N-SH2 beads for 30 min prior to the addition of ATWT lysates. Bound proteins were subjected to SDS-PAGE on 8% gels, and transferred to polyvinyl- idene difluoride membranes (Immobilon, Millipore) using standard pro- cedures. Immunoblotting was carried out using the monoclonal anti- phosphotyrosine antibody 4G10, a generous gift of Dr. Brian Druker (Dana Farber Cancer Institute), as described (17), except that immu- noblots were developed using enhanced chemiluminescence (ECL, Am- ersham Corp.) according to the manufacturer’s specifications.

For identification of SH-PTP2 as the 64-kDa PDGFR-associated pro- tein, PDGFR immunoprecipitates were prepared from starved and PDGF-stimulated TRMP, ATWT, and AT1009 cells using a polyclonal anti-PDGFR C-terminal antibody (Upstate Biotechnology, Inc., catalog no. 06-132), and subjected to in vitro kinase assay using radiolabeled [y3’P1ATP, as described previously (6, 23). A portion of the in vitro kinase reaction from ATWT immunoprecipitates was boiled in 1% SDS to dissociate bound proteins, diluted 10-fold with lysis buffer (17), and immunoprecipitated with affinity-purified anti-SH-PTP2 polyclonal an- tibodies, essentially as described (17). Immunoprecipitates were resus- pended in sample buffer and subjected to SDS-PAGE and autoradiog- raphy.

Protein Expression and Purification-The expression and purifica- tion of recombinant SH-PTP2 produced in E. coli is described in detail elsewhere (21). Preparations used in the current experiments were Over 90% pure, as assessed by Coomassie Blue staining.

Phosphatase Assays-RCM-lysozyme (Life Technologies, Inc.) was phosphorylated on its unique tyrosine (24) using recombinant v-Abl (Oncogene Science) according to the manufacturer’s instructions, except that ATP was present at 4 mM instead of 0.1 mM. Typical specific radio- activity obtained was 1500 cpdpmol. Phosphorylated RCM-lysozyme (2 PM) was incubated with 1.6 pg/ml purified SH-PTP2 with or without the indicated concentration of peptide at 30 “C for 5 min in 25 pl of reaction buffer (25 mM HEPES, pH 7.4, 150 mM NaC1, 100 pg/ml bovine serum albumin, 5 mM EDTA, and 10 mM dithiothreitol). Phosphate

release was quantitated using a charcoal binding assay (21), and de- phosphorylation velocity expressed as countdminute released into the supernatant. Under these conditions, phosphate release is linear with time of incubation (Ref. 21 and data not shown).

Peptide Synthesis-Peptide and Tyr(P) peptides were synthesized as described in Piccione et al. (25) using the methodology of Kitas et al. (26). The sequences of the phosphopeptides used were as follows (with pY indicating the phosphorylated tyrosine): PDGFR(Tyr(P)-740), DGGpYMDMSKDE; PDGFR(Tyr(P)-751), SVDpWPMLDMK, PDGFR- Tyr(P)-771), SSNpYMApYDNY, PDGFR(Tyr(P)-1009), SVLpYTAV- QPNE; PDGFR(Tyr(P)-1021), DNDpYIIPLPDPK. A single non-phos- phorylated peptide, PDGFR(Tyr-1009), with the sequence SVLW- AVQPNE was also synthesized, using standard techniques.

RESULTS AND DISCUSSION

As we had previously noted binding of the N-SH2 of SH- PTP2 to the PDGFR (171, we sought to determine the exact site of binding. Initial studies were carried out by direct binding of GST-N-SH2 beads to lysates from PDGF-stimulated canine epi- thelial (TRMP) cell lines expressing wild type PDGFR (ATWT) and various PDGFR phosphorylation site mutants (6,7,10,11). These studies indicated that GST-N-SH2 failed to bind to the activated PDGFR when Tyr-1009 was mutated to Phe, but mu- tations at the four other known major phosphorylation sites did not appreciably affect binding, suggesting that Tyr-1009 was the major binding site for SH-PTP2 on the PDGFR (data not shown). However, owing to experimental variations in the ex- tent of PDGF stimulation in the different cell lines and the binding capability of different preparations of GST-N-SH2 beads, it was difficult to obtain reproducible results with the direct binding assay.

To further explore the specificity of the interactions between SH-PTP2 and the PDGFR, we employed a peptide competition assay. Tyr(P) peptides corresponding to each of the known PDGFR tyrosyl phosphorylation sites (6-12) were tested for their ability to block WT PDGFR binding to GST-N-SH2. As shown in Fig. lA, and as demonstrated previously (17), GST- N-SH2, but not GST alone, binds to the PDGFR. Only the PDGFR(Tyr(P)-1009) peptide competed for binding (at 40 p~). Competition required tyrosyl phosphorylation of the peptide, since unphosphorylated PDGFR(Tyr-1009) peptide did not compete (Fig. lA). Competition by PDGFR(mP)-1009) was observed at concentrations as low as 10 PM, whereas no signifi- cant competition was observed with any other PDGFR(MP)) peptide at concentrations as high as 150 PM (data not shown). These data suggested that the major binding site for SH-PTP2 on the PDGFR is Tyr(P)-1009.

Previous workers have reported that a “64-kDa” protein binds to the PDGFR at Tyr(P)-1009 (11, 12). We prepared PDGFR immunoprecipitates from TRMP cells, ATWT cells, and AT1009 cells and subjected them to in vitro kinase reactions as described previously (6, 23). As previously reported (6, 7, 10, 111, PDGF stimulation results in the association of several proteins (arrows) with the WT PDGFR (Fig. l g , WT). p64 binding to the Phe-1009 mutant receptor is virtually undetect- able (Fig. 1 B , F10091, whereas all of the other receptor-associ- ated proteins appear to associate to comparable extents with the WT PDGFR and the Phe-1009 PDGFR mutant. There was no effect of mutation of the other known PDGFR tyrosyl phos- phorylation sites (Tyr(P)-740, Tyr(Pb751, Tyr(P)-771, or Tyr(P)- 1021) on p64 association with the PDGFR (data not shown, and see Refs. 11 and 12). The mobility of p64 under our electropho- resis conditions suggested that it was significantly larger than 64 kDa. Thus, although SH-PTP2 immunoprecipitated from mammalian cells (171, or expressed in E. coli (21), migrates on SDS-PAGE with an apparent molecular mass of 70 kDa, we asked whether p64 was identical to SH-FTP2. WT PDGFR immunoprecipitates were subjected to in vitro kinase assay and

21480 Activation of SH-PTPZ by Its PDGFR Binding Site

F1009 W 0 ” + - * - +

‘p120’ - a p + a # -200

-1 10

-76

-40

c

FIG. 1. SH-PTP2 binds to the activated PDGFR at tyrosine 1009. A, peptide competition of wild type PDGFR binding to GST-N- SH2. Lysates from PDGF-stimulated canine epithelial cells expressing the wild type PDGFR (ATWT cells) were purified on glutathione-aga- rose beads containing either glutathione S-transferase alone (GEX) or GST-N-SH2 in the absence (WT) or presence of the indicated TydP) peptides corresponding to known PDGFR phosphorylation sites. For the

Tyr-1009 is an unphosphorylated version of Tyr(P)-1009. B, a “64-kDa” exact sequence of the Tyr(P) peptides, see “Materials and Methods.”

protein binds to TydP)-1009 of the PDGFR. PDGFR immunoprecipi- tates were prepared from quiescent TRMP cells, which fail to express the PDGFR (0). from ATWT cells, which overexpress the WT PDGFR (WT), or from AT1009 cells, which overexpress a Phe-1009 PDGFR mutant (FI009), which were either unstimulated (-) or stimulated (+) with PDGF (50 ng/ml), and subjected to immune complex kinase assay (6, 23). The positions of known PDGFR binding proteins (6, 10, 11) and prestained markers (Bio-Rad) are indicated. C, SH-PTP2 is the “64- kDa” PDGFR-associated protein. Lunes a and b, PDGFR immunopre- cipitates were prepared from quiescent ATWT cells either unstimulated (-) or stimulated (+) with PDGF (50 ng/ml) and subjected to immune complex kinase assay (6, 23). The positions of known PDGFR-binding proteins (6,10,11) are indicated. Lanes c and d, reactions carried out as in lanes a and b were boiled in 1% SDS to disrupt protein-protein interactions, diluted 10-fold with lysis buffer, and re-precipitated with affinity-purified anti-SH-PTP2 polyclonal antibodies (17). The positions of prestained molecular weight markers (Life Technologies, Inc.) are indicated; note that these markers are different from those used in panel B. The arrow indicates SH-PTP2 p64, which on this gel system runs at a higher effective molecular weight than previously reported. Lanes a and b were autoradiographed for 1 h, whereas lanes c and d represent a 24-h exposure.

equal portions of the reaction were either electrophoresed di- rectly or boiled in 1% SDS, diluted 10-fold with lysis buffer, and re-immunoprecipitated with affinity-purified anti-SH-FTP2 antibodies (17). Re-immunoprecipitation with anti-SH-PTP2 antibodies identified p64 as SH-PTP2 (Fig. 1C). Taken to- gether, these in vitro and in vivo data strongly suggest that the major binding site for SH-FTP2 on the PDGFR is Tyr(P)-1009. Similar data have also been obtained by Kazlauskas et al. (27).

To model the potential effect of PDGFR binding on SH-FTP2,

B

0 .OI . I 1 10 100 1000 10000

peptide concentration (vM)

FIG. 2. SH-Pl”2 activity is stimulated by the PDGFR(Qr(P)- 1009) peptide. A, comparison of activation of SH-PTP2 by PDGFR- (Qr(P)) peptides. SH-PTP2 activity as assayed with 32P-labeled Tyr- RCM-lysozyme in the presence or absence of the indicated Tyr(P) peptide a t 40 p ~ . Shown are the means * S.E. of three independent determinations from one experiment. Similar results were obtained in four separate experiments. B, concentration dependence of activation of SH-PTP2 by Tyr(P) peptides. SH-PTP2 activity as assayed with 32P- labeled Tyr-RCM-lysozyme in the presence of increasing amounts of PDGFR(TydP)-1009) peptide (solid circles), PDGFR(Tyr(P)-740) pep- tide (open squares), or PDGFR(Tyr-1009) peptide, an unphosphorylated version of PDGFR(TydP)-1009) (open circles). Values shown are the mean * S.E. of triplicate determinations. For details see “Materials and Methods.”

we asked whether addition of a Tyr(P) peptide containing the Tyr(P)-1009 sequence (see “Materials and Methods”) affected SH-PTP2 activity. SH-FTP2 activity was assayed in the ab- sence or presence of Tyr(P) peptides corresponding to the major phosphorylation sites of the PDGFR. Addition of the PDGFR- (Tyr(P)-1009) peptide (at 40 PM) (Fig. 2 A ) strongly stimulated SH-PTP2 activity (>5-fold), measured as release of [32Plphos- phate. No other PDGFR(Tyr(P)) peptide had any stimulatory effect (Fig. 2 A ) , nor was there any effect of unphosphorylated PDGFR(Tyr-1009) (Fig. 2B ).

The concentration dependence for activation of SH-FTP2 by two Tyr(P) peptides is shown in Fig. 2 B . Half-maximal activa- tion by PDGFR(Tyr(P)-1009) occurred at about 50 PM. Consist- ent with activation of SH-PTP2 by PDGFR(Tyr(P)-1009) being mediated through the SH2 domain(s) of SH-PTP2, unphos- phorylated PDGFR(Tyr-1009) did not activate, even at the highest concentrations tested (>1 mM). The PDGFR(Tyr(P)-740) peptide also activated SH-PTP2, but only at 100-fold higher concentration than that required for PDGFR(Tyr(P)-1009)-in- duced activation (Fig. 2 B ) . Most likely, at these high concen- trations, nonspecific binding of the PDGFR(Tyr(P1-740) to the SH2 domain(s) of SH-PTP2 occurs. This suggestion is sup- ported by the results of peptide competition assays, in which the relative affinity of PDGFR(Tyr(P)-1009) for GST-N-SH2 was found to be at least 100-fold higher than that of PDGFR- (T~r(P)-740).~

Our data indicate that binding of PDGFR(Tyr(P)-1009) pep- tide to recombinant SH-PTP2 stimulates SH-FTP2 activity in vitro in a phosphotyrosyl dependent manner (Fig. 21, and that

S. E. Shoelson, data not shown.

Activation of SH-PTP2 by

Tyr(P)-1009 is the binding site for the N-SH2 of SH-PTP2 on the PDGFR in vitro (Fig. L4) and in vivo (Fig. 1, B and C). These data suggest an explanation for our earlier finding that SH-PTP2 displays non-Michaelis-Menten kinetics when meas- ured with the PDGFR(Tyr(P)-1009) peptide. Under these con- ditions, it is likely that the peptide serves as both substrate (binding at the active site) and allosteric modulator (binding the N-SH2).

Our data constitute the first example of regulation of a non- transmembrane PTP by a physiologically relevant stimulus. Previous workers have found that CD45 is inhibited following calcium ionophore treatment (28). The observed inhibition of CD45 activity correlated with serine dephosphorylation; how- ever, no clear causal relationship between increased intracel- lular calcium, CD45 phosphorylation, and CD45 activity was established, nor was inhibition of CD45 demonstrated under physiological conditions. Flint et al. (29) have reported a small decrease (approximately 30%) in the activity of PTPlB which correlates with its phosphorylation at mitosis, although this decrease in activity has not been observed by other workers (30L4

It is tempting to conclude that the PDGFR('&r(P)-1009) pep- tide activates SH-PTP2 in vitro by engaging its N-SH2. Con- sistent with this hypothesis, both binding of PDGFR(Tyr(P)- 1009) to the N-SH2 (Fig. 1) and PDGFR(Tyr(P)-1009)-mediated stimulation of SH-PTP2 activity (Fig. 2) require phosphoryla- tion of Tyr-1009. I t is not clear whether engagement of one or both SH2 domains is sufficient for SH-PTP2 activation. Fur- ther studies with different Tyr(P) peptides and with SH-PTP2 proteins that have mutations in one or both SH2 domains will be required to evaluate the specificity and participation of each SH2 domain in phosphatase regulation. Likewise, the detailed molecular mechanism by which peptide interactions within an SH2 domain could activate PTP activity remains to be eluci- dated.

SH2Pryr(P) peptide interactions can have several conse- quences. For normally cytosolic molecules such as rasGAP, SH2-mediated binding to RPTKs may serve to relocate en- zymes to the vicinity of their substrates (3-5). Alternatively, receptor binding can increase the activity of SH2-containing proteins, either indirectly or directly. Phospholipase C-y is ac- tivated by tyrosyl phosphorylation (31). The binding of phos- pholipase C-y via its SH2 domains to activated RPTKs im- proves its ability to serve as an RPTK substrate (32). Conversely, phosphatidylinositol 3-kinase has been shown to be activated directly by binding to insulin receptor substrate-1 (33) or upon addition of several different Tyr(P) peptides, each known to be specific for the phosphatidylinositol 3-kinase SH2 domain (33, 34).

Our in vitro data predict that binding of SH-PTP2 to the PDGFR in vivo should also lead to SH-PTP2 activation. Vogel et al. (15) reported slight activation (less than 2-fold) of SH- PTPB(PTP1D) by tyrosyl phosphorylation, whereas Feng et al. (14) found no difference in the activity of SH-PTP2 (Syp) fol- lowing growth factor stimulation. These experiments, however, could not differentiate between the effects of receptor binding and tyrosyl phosphorylation on SH-PTP2 activity. Our data clearly predict that receptor binding alone should be strongly activating. It may be that tyrosyl phosphorylation of SH-PTP2 is inhibitory, with the net effect in vivo being slight activation,

V. Shifrin, J. V. Frangioni, and B. G. Neel, unpublished data.

Its PDGFR Binding Site 2 148 1

as reported by Vogel et a1 (15). Alternatively, the immune com- plex PTP assay used by these workers may not faithfully re- produce the effects of receptor binding and/or tyrosyl phos- phorylation on enzymatic activity. Direct determination of the in vitro effects of tyrosyl phosphorylation on SH-PTP2 activity is complicated, since SH-PTP2 auto-dephosphorylates during the assay (14L5 In addition to its predicted direct effect on SH-PTP2 enzymatic activity, PDGFR binding also affects SH- PTP2's ability to be tyrosyl-phosphorylated; PDGF-induced SH-PTP2 tyrosyl phosphorylation is reduced approximately 2-3-fold in AT1009 cells compared with ATWT cells6 Ulti- mately, a complete description of the regulation of SH-PTP2 in vivo will require in vitro reconstitution of the combined effects of receptor binding and tyrosyl phosphorylation.

Acknowledgment-We thank M. Chaudhuri for phosphopeptide preparation.

2. 1.

3.

4. 5. 6.

7. 8.

9.

10. 11. 12. 13.

14. 15.

16.

17.

18.

20. 19.

21.

22.

23. 24.

25.

26.

27.

29. 28.

30.

31.

32.

33.

34.

REFERENCES Ullrich, A,, and Schlessinger, J. (1990) Cell 61, 203-212 Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth, B., Graziani, A., Kapel-

Koch, C., Anderson, D., Moran, M., Ellis, C., and Pawson, T. (19911 Science

Margolis, B. (19921 Cell Growth Differ 1992, 73-80 Pawson, T., and Gish, G. 11992) Cell 71,359-362 Kazlauskas, A., Kashishian, A., Cooper, J. A., and Valius, M. (1992) Mol. Cell.

Kashishian, A., Kazlauskas, A,, and Cooper, J. (19921 EMBO J. 11,1373-1382 Fantl, W., Escobedo, J., Martin, G., Turck, C., del Rosario, M., McCormick, F.,

Ronnstrand, L., Mori, S., Amdsson, A., Eriksson, A,, Wernstedt. C., Hellman,

Valius, M., Bazenet, C., and Kazlauskas,A. (1993) Mol. Cell. Biol. 13,133-143 Kashishian, A,, and Cooper, J. (1993) Mol. Biol. Cell 4, 49-57

Valius, M., and Kazlauskas, A. (1993) Cell 73, 321-334 Freeman Jr., R. M., Plutzky, J., and Neel, B. G. (1992) Proc. Nail. Acad. Sci.

Feng, G.-S., Hui, C.-C., and Pawson, T. (1993) Science 259, 1607-1611 Vogel, W., Lammers, R., Huang, J., and Ullrich, A. (19931 Science 259, 1611-

Ahmad. S., Banville, D.. Zhao, Z., Fischer, E. H., and Shen, S.-H. (19931 Proc.

Lechleider, R. J., Freeman, R. M., and Neel, B. G. (1993) J. Biol. Chem. 268,

Kuhne, M. R., Pawson, T., Lienhard. G. E., and Feng, G .3 . (19931 J. Biol.

Perkins, L. A,, Larsen, I., and Perrimon, N. 11992) Cell 70, 225-236 Lu, X., Chou, T.-B., Williams, N. G., Roberts, T., and Perrimon, N. (1993) Genes

& Dew 7,621-632 Sugimoto, S., Lechleider, R. J., Shoelson, S. E., Neel, B. G., and Walsh, C. T.

11993) J. Biol. Chem. 268, 22771-22776 Songyang, Z.. Shoelson, S. E., Chaudhuri, M., Gish, G., Pawson, T., Haser, W.,

Kmg, F., Roberts, T., Ratnofsky, S., Lechleider, R. J., Neel, B. G.. Birge. R.

ley, L. C. (19931 Cell 72, 767-778 B., Fajardo, J. E., Chou, M. M., Hanafusa. H., Schaffhausen, B., and Cant-

ler, R., and Soltoff, S. (1991) Cell 64, 281-302

252,668-674

Biol. 12,2534-2544

and Williams, L. (1992) Cell 69,413-423

U., Claesson, L, and Heldin, C. 11992) EMBO J. 11, 3911-3919

U. S. A. 89, 11239-11243

1614

Natl. Acad. Sci. U. S. A. 90, 2197-2201

13434-13438

Chem. 268, 11479-11481

Kazlauskas, A., Durdern, D., and Cooper, J. (1991) Cell Regul. 2,413-441 Tonks, N. K., Diltz, C. D., and Fischer, E. H. (1988) J. Biol. Chem. 263,

Piccione, E., Case, R. S., Domchek, S. M., Hu, P., Chaudhuri, M., Backer, J. M., 6731-6737

Kitas, E., Knorr, R., Trzeciak, A,, and Bannwarth, W. I 1991) Helu. Chim. Acta Schlessinger, J., and Shoelson. S. E. (19931 Biochemistry 32, 3197-3203

Kazlauskas, A., Feng, G.-S., Pawson, T., and Valius, M. I19931 Proc. Natl. 74, 1314-1328

Ostergaard, H. L., and Trowbridge, I . S. (1991) Science 253, 1423-1425 Acad. Sci. U. S. A,, in press

Flint, A., Gehbink. M., Franza, B. J., Hill, D., and Tonks, N. (19931 EMBO J. 12, 1937-1946

Schievella, A., Paige, L.. Johnson, K., Hill, D., and Erikson, R. (1993) Cell Growth Diffefer. 4, 239-246

Nishibe, S., Wahl, M., Hernandez-Sotomayor, S., Tonks, N., Rhee, S., and Carpenter, G. (1990) Science 250, 1253-1256

Rotin, D., Henegger, A., Margolis, B., Ullrich, A., and Schlessinger, J. (19921 J . Biol. Chem. 267,9678-9683

Backer, J. M., Myers, M. G., Jr., Shoelson, S. E., Chin, D. J.. Sun, X.-J., Miralpeix, M., Hu, P., Margolis, B., Skolnik, E. Y., Schlessinger, J., and

Carpenter, C., Auger, K., Chaudhuri, M., Yoakim. M.. Schaffhausen, B., White, M. F. (19921 EMBO J. 11,3469-3479

Shoelson, S., and Cantley, L. (1993) J. B i d Chem. 268, 9478-9483

R. J. Lechleider and B. G . Neel, unpublished data. fi R. J. Lechleider, A. M. Bennett, and B. G. Neel, unpublished data).