Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Identification of c‐erbB‐3 binding sites for phosphatidylinositol 3′‐kinase and SHC using an EGF receptor/c‐erbB‐3 chimera.

Identification of c‐erbB‐3 binding sites for phosphatidylinositol 3′‐kinase and SHC using an EGF... The EMBO Journal vol.13 no.12 pp.2831 -2841, 1994 Identification of c-erbB-3 binding sites for phosphatidylinositol 3'-kinase and SHC using an EGF receptor/c-erbB-3 chimera threonine sites which are present in potential regulatory A.Prigent1 and William J.Gullick2 Sally and c-erbB-4, suggesting that c-erbB-3 may EGFR, c-erbB-2 Molecular Oncology Laboratory, ICRF Oncology Group, to negative regulation by protein kinase C not be subject Hospital, Du Cane Road, London W12 OHS, UK Hammersmith et al., 1990). These data suggest that c-erbB-3 (Plowman 'Present address: Departments of Medicine and Pharmacology, different properties to other family may possess slightly of California, San Diego, La Jolla, CA 92093-0636, USA University members. author 2Corresponding advances have been made recently in understanding Great J.Brockes Communicated by which kinase receptors interact the mechanism by tyrosine signalling components. Following ligand c-erbB-3 is a member of the I receptor-related) with intracellular type (EGF on specific tyrosine residues of growth factor receptors for which no ligand has binding, autophosphorylation family of to the activated To facilitate ligand stimulation we have facilitates the recruitment proteins been identified. is achieved by the high-affinity a chimeric receptor which possesses an receptor. This association constructed elements called src homology kinase and promotes the growth of NIH 3T3 interaction of characteristic activatable residues (Pawson and fibroblasts. In this study we have shown that SHC and 2 (SH2) domains with phosphotyrosine A large number of proteins containing 3'-kinase bind to the activated EGF Schlessinger, 1993). phosphatidylinositol now been identified, including those with receptor/c-erbB-3 chimera. Whereas p85 is not SH2 domains have as the p21ras GTPase-activating to a significant extent, SHC appears to enzymatic activity such phosphorylated et al., 1988), phosphotyrosine phosphatases for phosphorylation on tyrosine. In protein (Trahey be a major substrate Freeman et al., 1992; Feng et al., 1993; to EGF receptor and c-erbB-2, we were unable (Shen et al., 1991; contrast and phospholipase C'y (Stahl et al., of activated c-erbB-3 to GRB2. Using Vogel et al., 1993) to detect binding with no obvious enzymatic activity corresponding to each of 13 potential 1988), and those synthetic peptides et al., 1992), which is the sites on c-erbB-3, we have shown that including GRB2 (Lowenstein phosphorylation Caenorhabditis elegans for SHC binding. Peptides mammalian homologue of the tyrosine 1309 is responsible sem-5 (Clark et al., 1992) and SHC (Pelicci et al., containing the motif YXXM inhibit p85 association. By protein Additional motifs called src homology 3 (SH3) comparison with recently reported SHC binding sites on 1992). T and Trk we have identified a SHC domains have also been identified in a number of these Middle antigen NPXY. which direct cellular localization of signalling binding motif, proteins 1993). Those SH2-containing factor receptors/ molecules (Bar-Sagi et al., Key words: c-erbB-3/growth which do not possess enzymatic activity act as phosphorylation proteins functional elements with the molecules linking other adaptor An example of this is the p85 protein activated receptors. 110 subunit of phosphatidylinositol which interacts with the p Introduction the enzymatic activity 3'-kinase which possesses (PI) Other examples include GRB2 c-erbB-3 was cloned almost simul- (Escobedo et al., 1991). The gene encoding couple the EGF receptor with the et al., 1989; Plowman which has been shown to taneously by two groups (Kraus binding both activated EGF kDa glycoprotein which signalling pathway by et al., 1990). It encodes a 160-180 p21rcLI nucleotide exchange protein sos and the guanine has all the structural features of the EGF family of tyrosine receptor Chardin et al., 1993; Egan also includes EGF receptor (EGFR), and Downward, 1993; kinase receptors [which (Buday Li et al., 1993; Olivier et al., et al., 1993)]. The highest et 1993; Gale et al., 1993; c-erbB-2 and c-erbB-4 (Plowman al., et al., 1993; Simon et al., 1993). identity with c-erbB-3 is within the 1993; Rozakis-Adcock proportion of sequence and associates with EGFR (60%), c-erbB-2 (62%) and SHC is phosphorylated by physically catalytic domains of and c-erbB-2 (Segatto et al., 1993). EGFR, c-erbB-2 and c-erbB-4 are activated EGF receptor c-erbB-4 (63%), although it appears to then interact with similar to each other than to c-erbB-3 Following phosphorylation somewhat more et The precise role of GRB2 (Rozakis-Adcock al., 1992). 80% sequence identity between receptor pairs). More- (-- is less well defined than that for GRB2, but it is the family member to contain unusual this protein over, c-erbB-3 only at least in the also be involved in p2lras signalling, within the kinase domain, including His740 and may residues kinases such as v-Src which of which are as glutamate and aspartate, case non-receptor tyrosine Asn815 present to GRB2. cannot couple directly in other kinases (Plowman et al., 1990). respectively, residues mapping the specific phosphotyrosine of the corresponding aspartate residue to asparagine By Mutation domain-containing proteins, it for binding SH2 kinase completely abolished kinase responsible in the tyrosine v-fps to the murine white has been identify phosphotyrosine-containing et 1988). Similarly, possible activity (Moran al., for different binding results from the same amino acid consensus sequences responsible spotting (W42) phenotype is now to to a and it predict c-kit which also encodes molecules, possible in the signalling replacement proto-oncogene molecules are to be which effector likely et c-erbB-3 lacks certain extent kinase al., 1990). Finally, a defective (Tan Press Oxford University S.A.Prigent and W.J.Gullick were constructed and analysed similarly. The double mutant involved in the signalling of a novel receptor (Songyang showed, if anything, a reduced activity relative to the wild- et al., 1993). Analysis of the c-erbB-3 amino acid sequence type protein (Figure 1). reveals several potential binding sites for PI3-kinase, a c-erbB-3 site The autocatalytic activity of and mutant phosphotyrosine phosphatase (SH-PTP2) binding wild-type which et was also compared in vivo (Figure 2). COS-1 cells, (Freeman et al., 1992; Feng et al., 1993; Vogel al., 1993) et sites for GRB2. express -50 000 EGF receptors per cell (Livneh al., and two possible binding and mutant c-erbB-3. function of the c-erbB-3 1986), were transfected with In this work, we investigated the wild-type of of indicator cells. In The transfected cells expressed a similar quantity protein and its effect on the growth second endogenous EGF receptors and c-erbB-3 protein, as particular, its interaction with specific messenger determined by immunoprecipitation from [35S]methionine- pathways was explored. labelled cells (data not shown). c-erbB-3 immunoprecipitates were analysed by immuno- Results blotting with an anti-phosphotyrosine antibody. The extent of phosphorylation of the mutant c-erbB-3 protein was Comparison of wild-type and mutant c-erbB-3 in vitro and in vivo Ni UT WT The kinase activity of c-erbB-3 was assessed by immune- transiently transfected COS-1 cells as 49.3 PY complex assay, using 49.3 PY of c-erbB-3 and an antibody a source protein anti-peptide + - + + - + k Da for c-erbB-3 for immunoprecipitation (Figure (49.3) specific - 200 '00 ....i.:.. antibody reacts with the intracellular domain 1). Since this :1: of and could potentially inhibit kinase activity, the .~ c-erbB-3 ..gv`..... c-erbB-3 protein was eluted with the corresponding peptide T, to the kinase assay. This peptide contains no tyrosine prior - 116 - 116 residues and therefore could not act as a substrate. c-erbB-3 showed a low level of autophosphorylation in immune- very -80 -80 which was on the of2-4 complex assays dependent presence mM Mn2+ ions and of Mg2+ ion concentration independent not acid revealed that (data shown). Phosphoamino analysis -49.5 - 49.5 was on and serine residues (data phosphorylation tyrosine To determine whether basal autocatalytic activity not shown). be enhanced His740 and Asn815 the could by replacing by of and mutant c-erbB-3 kinase Fig. 2. Comparison wild-type activity more favourable residues Glu and Asp, mutants potentially in vivo. Wild-type (WT) and mutant (MUT) c-erbB-3 proteins were in COS-1 cells. c-erbB-3 was precipitated transiently expressed protein from unstimulated cells (-) or cells which had been stimulated with antibody 49.3. Bound proteins were eluted EGF (+) using polyclonal with 49.3 and analysed by immunoblotting with peptide (2.5 mg/ml), A B WT MUT V 49.3 to the levels of the constructs antibody compare expression WT MUT V and with an antibody (PY). (49.3), anti-phosphotyrosine T E T E T E EL EL CL kDa XL_ kDa + + G Cy) c ) Cv) IL cD 200- C D ,t eti Lr) kDa 200- N."'. 0 0 97 - 97- 69 - Fig. 1. Comparison of wild-type and mutant c-erbB-3 kinase activity chimera in NIH 3T3 Fig. 3. Expression of EGF receptor/c-erbB-3 in vitro. The wild-type (WT) and mutant (MUT) c-erbB-3 proteins cells. To test for correct of the EGFR/c-erbB-3 chimera, expression were at similar levels as determined by immunoprecipitation cm expressed was protein precipitated from a 9 plate of [35S]methionine-labelled from metabolically labelled COS-1 cells using serum 49.3 (A). T and EGFR/c-erbB-3-3T3 cells (clone 18) using an antibody against the E refer to the total precipitated protein released by SDS, and that EGFR extracellular domain (EGFR1), an anti-peptide antibody reacting which may be eluted by addition of 49.3 peptide, respectively. Kinase with the juxtamembrane domain of c-erbB-3 (61.3) and an anti-peptide on assays performed immune-complexes from COS-1 cells transfected antibody reacting downstream of the kinase domain of c-erbB-3 (49.3). with c-erbB-3 (WT), mutant c-erbB-3 (MUT) and vector (V) wild-type Reaction with anti-peptide antibodies was inhibited by prior incubation are shown (B). with the corresponding peptide (PEP). 2832 chimera EGF receptor/c-erbB-3 + EGF N- r_ 0 + .SE _ _ z <: kDa + + kDa ~ ~ L -(JC,~ 200- kDa 200- p0 97- *4 is 69 - 97 - 0 w w 4_ * 69- *WI 30- 46- Fig. 4. EGF stimulation of phosphorylation. (A) Whole cell lysates of stimulated EGFR/c-erbB-3-3T3 cells were analysed by SDS-PAGE and immunoblotting with anti-phosphotyrosine antibodies. (+) and (-) refer to EGF-stimulated and unstimulated cells, respectively. In the case of NEN7 and A431 cells, one tenth of the amount of protein was used. (B) EGFR/c-erbB-3-3T3 cells were incubated with EGF for the indicated times. Phosphorylation was terminated rapidly by the addition of boiling sample buffer to cells which were analysed by SDS-PAGE and immunoblotting with anti-phosphotyrosine antibody. reduced relative to the wild-type. Phosphorylation of both of EGFR/c-erbB-3-3T3 cells Table I. Anchorage-independent growth mutant and wild-type c-erbB-3 was increased by treatment of the cells with EGF prior to lysis. Colonies per 30 000 cells 0 nM EGF 3 nM EGF 30 nM EGF Construction and of EGFR/c-erbB-3 1 1 86 16 78 I 1 expression Chimera clone 18 f + + 69 chimera in COS-1 and NIH 3T3 cells Chimera clone 42 50 + 16 205 + 2 1011 NIH mock 3T3 a system, the EGFR/c-erbB-3 To generate ligand-activatable transfected 1 1 3 1 2 1 chimera was fused inside the cell membrane at a common just Narl site. When in COS-1 cells, the protein was expressed 42 and mock- Clones 18 and expressing the EGFR/c-erbB-3 chimera as deduced by correctly located at the cell surface transfected NIH 3T3 cells were suspended in 1.2% Methocel in the protein from intact and precipitating the EGFR/c-erbB-3 >0.1 mm presence and absence of EGF (3 and 30 nM). Colonies for the intracellular and are the results of one lysed cells with antibodies specific diameter were scored after 2 weeks. Data shown representative experiment. The of extracellular domains (Kumar et al., 1991). ability cross- the chimeric protein to bind EGF was confirmed by linking of [125I]EGF to intact cells followed by immuno- NIH 3T3 cell lines EGF stimulation of phosphorylation in NIH 3T3 cells precipitation (data not shown). Stable created and analysed by EGF chimera expressing EGFR/c-erbB-3 were expressing the receptor/c-erbB-3 of EGF stimulated the of the chimeric protein binding assays using [1251]EGF to determine the number phosphorylation cell et 1978). Clones 18 and in 18 and 42 as deduced by immunoblotting of whole receptors per (Aharonov al., clones - 130 000 and 70 000 cell, cell with anti-phosphotyrosine antibodies (Figure 4A). 42, possessing receptors per lysates for functional studies. These The identity of the stimulated 180 kDa chimeric protein was respectively, were selected were verified FACS confirmed of the phosphoprotein receptor numbers by comparative by immunoprecipitation EGFR1 mAb to the common EGFR antibodies specific for EGFR and c-erbB-3 (data not using the with analysis to label the and NEN7 cells For NEN7 cells expressing the extracellular domain cells, shown). comparison, 2 x 2 x 106 chimeric molecules chimera and A431 cells expressing 106 EGFR/c-erbB-2 per EGFRIc-erbB-2 expressing When 10-fold less total et To check that EGF per cell were used. cell as a standard (Lehvaslaiho al., 1989). receptors in of NEN7 and A431 contained the c-erbB-3 protein was loaded the case cells, the expressed protein complete antibodies two distinct sites a similar signal on the anti-phosphotyrosine immunoblot was cytoplasmic domain, recognizing that the of stimulation the c-erbB-3 kinase domain et al., 1992) obtained, suggesting degree receptor within (Prigent three members is used to the EGFR/c-erbB-3 protein, and the obtained for the family comparable. were precipitate were raised were used of c-erbB-3 was maximal which the antibodies Stimulation autophosphorylation peptides against The s and decreased to almost basal level after 45 to for c-erbB-3 3). after 15 min, specifically compete binding (Figure and/or at - 180 kDa on due to by phosphatases EGFR/c-erbB-3 chimera migrated probably dephosphorylation an of the Moreover, SDS down-regulation receptor (Figure 4B). -polyacrylamide gels. 2833 and S.A.Prigent W.J.Gullick CONCENTRATION EGF (nM) 0 3 Clone Clone 3T3 NIH mock transfected in 18 42 Fig. 5. Growth of colonies 1.2% Methocel. Clones and the EGFR/c-erbB-3 chimera and mock-transfected NIH 3T3 cells were expressing in mm suspended and grown 1.2% Methocel. Colonies were after 3 weeks. The to 0.5x0.6 fields. photographed panels correspond activated additional band of -54 kDa was seen in the Association of EGFR/c-erbB-3 with SHC consistently EGFR/c-erbB-3-3T3 cells on Since the EGF receptor/c-erbB-3 chimera was able to repeated experiments (Figure in 4B). This was absent in EGF-stimulated NIH promote a growth response NIH 3T3 cells, we were phosphoprotein 3T3 cells and unstimulated EGFR/c-erbB-3-3T3 and interested to determine which elements of established SH2 cells, in to the activated c-erbB-3 signalling pathways were involved. To determine whether appeared parallel protein, that it be a substrate for SHC associated with activated NIH 3T3 cells suggesting may major c-erbB-3. c-erbB-3, expressing an EGFR/c-erbB-3 chimera 18 as (clone were described above) stimulated with EGF. To specifically - x and of precipitate EGFR/c-erbB-3 protein, lysates from 2 106 Anchorage-dependent -independent growth were an NIH 3T3 cell lines the EGF cells incubated with antibody raised the expressing receptor! against c-erbB-3 chimera intracellular portion of c-erbB-3 (49.3). This eliminated the EGF addition to EGFR/c-erbB-3-3T3 cells enabled them to possibility of cross-reaction with the small amount of EGF in NIH grow in an as determined receptor present the 3T3 cells. Since the p52 and anchorage-independent manner, in Methocel. Clone 42 showed forms of SHC by their ability to grow 1.2% p46 migrate at the same position as of than clone which total a higher basal level transformation immunoglobulin heavy chain, precipitates could not 18, formed colonies in the of EGF readily be only presence analysed by immunoblotting due to the cross- (Table I). Clones 18 and 42 formed colonies in the reaction of larger presence immunoglobulin heavy chain with the second of 30 nM EGF as with 3 nM of the immunoblot. To compared EGF (Figure 5). layer avoid this cross-reaction, For NEN7 and mock-transfected NIH 3T3 cells EGFR/c-erbB-3 chimera comparison, protein was eluted from the were also cultured in 1.2% Methocel. some small antibody -protein A Although -Sepharose complex using a solution colonies were observed for mock-transfected NIH 3T3 of peptide which cells, against antibody 49.3 was raised. Eluates these did not increase in number or size in to EGF. were analysed by SDS -PAGE and immunoblotting using response In the case of NEN7 cells, colonies were only observed in a rabbit polyclonal antibody recognizing the three forms of the of EGF as described presence previously (Lehvaslaiho SHC (Figure 6A). For comparison, total lysate from - et The effect of EGF cells was al., 1989). growth-promoting on clones analysed similarly (Figure 6B). The three forms 18 and 42 was confirmed of SHC and by [3H]thymidine incorporation (p66, p52 p46) associated specifically with studies. Using cells growing attached as EGF activated c-erbB-3, monolayers, although the molecular weights of p46 stimulation resulted in a 3.9-fold increase in and p52 appeared to be slightly higher on thymidine this gel system. at 0.5-5.0 nM incorporation EGF relative to the The predominant form present was p52, which also appeared unstimulated cells. NIH 3T3 maximal cells showed a 2.7-fold to be the most abundant form present in total cell lysates. stimulation in to EGF at the response same concentrations No SHC protein was present in precipitates from the parental not (data shown). NIH 3T3 cell line, even when using the EGFR1 antibody 2834 chimera receptor/c-erbB-3 EGF LU U X Lu CO _+ HI HV cccc a LU C) LU LU L IL + + * C + I C) COo cO co - IT" X . cn _ CO) CO kDa kDa -30 -69 -^ Fig. 7. Association of tyrosine-phosphorylated proteins with activated -_ a. c-erbB-3. Immunoprecipitations were performed on lysates from EGF- *Mw * 46 stimulated cells exactly as described in Figure 6A. Eluted proteins were analysed by SDS-PAGE and immunoblotting with an anti- phosphotyrosine antibody. -30 Fig. 6. Association of activated c-erbB-3 with SHC. (A) NIH 3T3 cells and transfected NIH 3T3 cells expressing I05 EGF receptor/c- erbB-3 chimeric receptors per cell (clone 18) were incubated for 2 min Cells were lysed and EGFR/c-erbB-3 protein was with 10-6 M EGF. with an antibody specific for c-erbB-3 (49.3). Precipitated precipitated proteins were eluted with the corresponding 49.3 peptide against which the was raised, and eluates were analysed by SDS-PAGE antibody and immunoblotting using an antibody recognizing SHC. (B) For 4-.-PIp total cell lysate from 10' cells was also analysed. comparison, LU0 LLU for the extracellular domain of EGF receptor (data specific * + not shown), and only a very small amount of the p52 form was found associated with EGFR/c-erbB-3 in the basal state. This represents association with a small proportion probably kDa of which is phosphorylated in the EGFR/c-erbB-3 unstimulated state. Association of EGFR/c-erbB-3 with tyrosine proteins phosphorylated Immunoblots which had been analysed for the presence of as described above were stripped and reprobed with -46 SHC an antibody (Figure 7). The major anti-phosphotyrosine comigrated exactly with the p52 form phosphorylated protein -30 and bands corresponding to p46 and p66 were also of SHC, 0 LUI 0 LUJ _ L. LUI Some other proteins of apparent molecular + detectable. ILI weights 85, 110 and 125 kDa were detectable. co OD Cl) Cl) 8. Association of PI3-kinase with activated c-erbB-3. Fig. Association of EGFR/c-erbB-3 with P13-kinase. were on lysates from EGF- Immunoprecipitations performed (A) from EGF-stimulated cells Immunoprecipitates prepared as described in 6A. Eluted proteins stimulated cells exactly Figure were for the association of antibody 49.3 analysed using SDS-PAGE and immunoblotting with a rabbit were analysed by with an antibody specifically the p85 subunit of P13-kinase. (B) P13-kinase by immunoblotting polyclonal antibody recognizing were on lysates from EGF-stimulated Immunoprecipitations performed the subunit of the enzyme. As was found reacting with p85 as described in 5A. Precipitated proteins bound to cells exactly Figure interacts with the activated EGFR/c- in the case of SHC, p85 with and as were incubated PI [-y-32P]ATP A-Sepharose protein and to a reduced extent with the erbB-3 receptor, greatly and TLC. A PI phosphate standard (PIP) described analysed by from unstimulated cells (Figure 8A). p85 P13-kinase at chimeric receptor PI with purified migrated prepared by phosphorylating the indicated. associated with EGF receptor precipitated position was also found 2835 EGF receptor/c-erbB-3 chimera from EGF-stimulated EGFR-3T3 which an PI4-kinase activity, inhibited the formation of this cells, express phospholipid (data not shown). equivalent number of EGF receptors, using antibody the was several-fold less than that EGFR1; however, signal obtained with the activated EGF Inhibition of p85 and SHC association receptor/c-erbB-3 chimera, using tyrosine and was comparable with that seen in unstimulated cells phosphopeptides the EGF chimera not To determine the c-erbB-3 binding sites for SHC and expressing receptor/c-erbB-3 (data p85, shown). This result was confirmed for PI3-kinase we looked at the ability of tyrosine phosphopeptides by assay in the A corresponding to all potential sites within activity immunoprecipitates. phosphorylated lipid phosphorylation (Rf: a 32P-labelled the c-erbB-3 C-terminal tail to inhibit formation 0.57), which comigrated with complex standard Dr in permeabilized NIH 3T3 cells the EGFR/c- phosphatidylinositol 3-phosphate (provided by expressing was when erbB-3 chimera All R.Woschowlski, ICRF, London), produced (Table II). tyrosine phosphopeptides EGFRIc-erbB-3-3T3 the motif precipitates from EGF-stimulated cells containing p85 binding pYXXM (la, 2a, 3a, 4, 6 and inhibited association with whereas were incubated with phosphatidylinositol and 7) p85 c-erbB-3, ['y-32P]ATP in the reaction buffer of Triton other had no effect (Figure 8B). Inclusion 1% peptides inhibitory (Figure 9B). Complete inhibition was only 4 X-100, which inhibits PI3-kinase activity and favours obtained with peptides and 7. Peptide 5 to was which corresponding Y1309 the only peptide inhibited SHC association The lane (Figure 9C). Table H. Sequences of c-erbB-3 phosphopeptides to 8 showed a reduced corresponding peptide slightly signal for however this is attributable to reduced of SHC; loading la (Y1241, Y1243) TTPDEDpYEpYMNRQR EGFR/c-erbB-3 in this case (Figure 9A). lb (Y1243) TTPDEDYEpYMNRQR To further assess of in the importance residues peptide 2a (Y1178, Y1180) DEDEEpYEpYMNRRRR 5 for SHC binding, two additional peptides were synthesized 2b (Y1180) DEDEEYEpYMNRRRR 11 and in which either the 3a LEELGpYEpYMDVGSD (peptides 12) asparagine residue (Y1203, Y1205) 3b (Y1205) LEELGYEpYMDVGSD (position -3 relative to the phosphotyrosine) or proline 4 (Y1035) SPSSGpYMPMNQGNL -2 residue (position relative to the phosphotyrosine) was 5 (Y1309) DSAFDNPDpYWHSRLF alanine changed to (Table II). Neither of these peptides was 6 (Y1257) GPGGDpYAAMGACPA to inhibit of able association SHC with c-erbB-3 under the 7 (Y1270) ASEQGpYEEMRAFQG conditions tested (data not shown). 8 EDVNGpYVMPDTHLK (Y1140) 9 (Y1288) APHVHpYARLKTLRS Comparison between EGF receptor, EGFR/c-erbB-2 10 RGDSApYHSQRHSLL (Y1113) and EGFR/c-erbB-3 with to to a respect binding lla (Y1309) AFDNADpYWHS GRB2 - GST fusion protein 12a (Y1309) AFDAPDpYWHS EGFR/c-erbB-3 from protein EGF-stimulated EGFRIc- aData are not shown for these X erbB-3-3T3 cells failed to associate with GRB2-GST fusion peptides. kDa PEP - - la lb 2a 2b 3o 3b 4 5 6 7 8 9 10 Lys EGF - + + + + + + + + 1 ~~~~~-20 c-erbB-3 -116 6 7 8 9 10 PEP - - la lb 2a 2b 3a 3b 4 5 Lys EGF - + +- + + + + + -116 )85 M+ + + + + + - -80 PEP -- 1a lb 2a 2b 3a 3b 4 5 6 7 8 9 10 Lys EGF - + + + + + + + C -80 SIIC 49.5 Fig. 9. Inhibition of p85 and SHC association with c-erbB-3 using synthetic phosphopeptides. NIH 3T3 cells expressing the EGF receptor/c-erbB-3 chimera were permeabilized in the presence of 70 x phosphopeptides as indicated. Cells were stimulated for 4 with EGF (5 M), protein min 10-7 j&M from cell was lysates precipitated with the 49.3 antibody, eluted with peptide 49.3 and analysed by SDS-PAGE and immunoblotting with antibodies for c-erbB-3 and SHC specific (A), p85 (B) (C). Peptide sequences are as indicated in Table II. 2836 chimera EGF receptor/c-erbB-3 A B 21 N 12E 1 2E 49.:3 49.3 coo rri tN EGFR-3T3 3T3 clone 18 3T3 I LL LLw - -J _N _ z z L z U) 2 + _ + _ + _ + _ + -+ - protein protein gg 49g Ewd, 97.4 Fig. 10. Association of EGFR/c-erbB-3 with a GRB2-GST fusion protein. (A) Cell lysates from EGF-stimulated (+) and unstimulated (-) cells (NEN7, EGFR-3T3 and EGFR/c-erbB-3-3T3; clone 18) were incubated with a GRB2-GST fusion protein immobilized on glutathione-agarose. Bound proteins were analysed by SDS-PAGE and immunoblotting with antibodies 21N, 12E and 49.3, which specifically recognize c-erbB-2, EGF receptor and c-erbB-3, respectively. (B) Cell lysates were prepared from EGFR-3T3, clone 18 and NIH 3T3 cells. Antibodies 12 E and 49.3 were used for immunoblotting of varying quantities of total cellular protein, as indicated. in c-erbB-3 et Plowman et protein immobilized on glutathione-agarose as determined different (Kraus al., 1989; al., structure of cAMP- by immunoblotting with polyclonal antibody 49.3 (Figure 1990). Evidence from the crystal kinase that the substituted lOA). For comparison, EGF receptor protein and EGFR/c- dependent protein (cAPK) suggests in c-erbB-3 to Aspl66 of cAPK) erbB-2 chimera protein were precipitated from EGF-treated aspartate (corresponding and in fact EGFR-3T3 and NEN7 cells, respectively, with the normally forms part of the catalytic loop may et The immobilized GRB2-GST fusion protein, and in these cases function as the catalytic base (Knighton al., 1991). residue which is absent in a prominent band was observed by immunoblotting with carbonyl group of the glutamate in in to the antibodies 12E and 21N, respectively. No immunoreactive c-erbB-3 (Glu9l cAPK) is close proximity In in these protein could be precipitated from EGF-stimulated NIH 3T3 MgATP binding site. the light of the differences of interest to determine cells, or from unstimulated EGFR-3T3 cells, although a apparently critical residues, it was activity. small amount of EGFR/c-erbB-2 protein was precipitated whether or not c-erbB-3 possesses catalytic indicate that c-erbB-3 a low from unstimulated NEN7 cells suggesting a somewhat Our findings possesses very in the unstimulated state, and elevated basal level of phosphorylation of this protein in these level of autocatalytic activity with and with cells. The sensitivity of antibodies 12E and 49.3 was that by replacing Asn815 aspartate, His740 no increase in is observed. By compared by immunoblotting of lysates prepared from glutamate, catalytic activity with a EGFR-3T3 and EGFR/c-erbB-3-3T3 cells (clone 18). Both replacing the extracellular domain ligand-activatable one detect cell lines express - 100 000 receptors per cell, and when EGF receptor domain, can, however, clearly of the chimeric receptor upon EGF equivalent amounts of total cell protein were analysed, phosphorylation This has been used to similar signals were obtained by immunoblotting (Figure stimulation. approach previously for which no were available LOB). characterize receptors ligands et Lehvaslaiho et Seedorf et al., (Riedel al., 1984; al., 1989; It is that this phosphorylation could be due 1991). unlikely as NIH 3T3 to EGF receptor cross-phosphorylation by Discussion number of EGF fibroblasts express a very low (-3000) Fiore et and EGF- has been for that all tyrosine receptors per cell (Di al., 1987a), It recognized many years of EGF certain conserved the most evident stimulated autophosphorylation endogenous receptor kinases possess features, motif in subdomain cells could not be detected. It of these a in NIH 3T3 is, however, being Gly-X-Gly-X-X-Gly II et that a could be attributed to I and a residue in subdomain (Hanks al., 1988). possible proportion trans-phos- lysine These observations and kinases EGF In both serine/threonine tyrosine phorylation by receptor. addition, residue found to be essential invariant or almost invariant that the contain a number of residues, demonstrate aspartate kinases et Tan residue in subdomain III and an for the function of other which include a glutamate (Moran al., 1988; to an in c-erbB-3 both of which are et be altered residue in subdomain al., 1990) may asparagine aspartate VI, 2837 S.A.Prigent and W.J.Gullick without inhibiting autocatalytic activity. It would be of some c-erbB-3 DSAFDNPDYWHSRLF interest to investigate the effect of the His/Glu and Trk IENPQYFSDA Asn/Asp mutations on the of the activatable EGFR/c- Middle T PSLLSNPTYSVMRSH catalytic activity erbB-3 chimera. Since the site of c-erbB-3 be catalytic may Fig. 11. Comparison of for on Trk binding sites SHC c-erbB-3, and different from other it be structurally kinases, may possible Middle T antigen. The SHC binding site on Trk was determined using kinase inhibitors more selective for this to design tyrosine the peptide indicated (Obermeier et al., 1993). Tyrosine 250 of Middle than for other kinases. receptor receptor tyrosine T was identified using site-directed mutagenesis and is contained within the sequence indicated (Dilworth et al., 1994). are in the Tyrosine kinase receptors frequently implicated cancers Both EGF growth of human (Aaronson, 1991). and c-erbB-2 are which can transform SHC are when receptor oncogenes gene products transforming overexpressed Fiore et NIH 3T3 fibroblasts when expressed at high levels (Di in fibroblasts (Pelicci al., 1992). et Both of these Whatever the exact the fact that SHC is the et al., 1987b; Hudziak al., 1987). growth mechanism, in a of human substrate for a number of both factor receptors are overexpressed variety major physiological receptor and in some cases are associated with and kinases to an cancers, poor non-receptor tyrosine points important role Lofts and We have in transduction. In this have shown prognosis (Gullick, 1991; Gullick, 1991). signal study we that SHC that c-erbB-3 is in some associates with and is c-erbB-3. shown, likewise, overexpressed phosphorylated by activated et cancers (Lemoine et al., 1992a,b; Rajkumar al., 1993). Moreover, we have identified a binding site for SHC In of to NIH 3T3 cells on this study, addition EGF expressing (Y1309) the c-erbB-3 protein. It would be of interest to to EGF receptor/c-erbB-3 chimeric receptors enabled them examine the mitogenic activity of mutant c-erbB-3 lacking which grow in an anchorage-independent manner is a this site. When the sequence surrounding this tyrosine residue property characteristic of, but not restricted to, transformed is compared with sequences reported recently to bind SHC of Kraus cells. These findings are in agreement with those on the Trk receptor tyrosine kinase and Middle T antigen, et al. (1993), who used a similar approach to demonstrate it is apparent that they all contain the motif NPXY which of the growth-promoting properties c-erbB-3. is likely to represent the consensus binding site for SHC In Since c-erbB-3 was able to with the (Figure 11). this study, peptides lacking either Asn(-3) couple signalling elements in NIH 3T3 cells to promote a growth response, or Pro(-2) were unable to inhibit the association between in some of the we were interested identifying proteins SHC and c-erbB-3, suggesting that both residues are identification of src domains has involved. The homology important for SHC binding. This is the first SH2-containing in of the facilitated rapid advances our understanding protein which appears to require the strict conservation of which kinases and residues mechanism by tyrosine signal (Pawson N-terminal to the phosphotyrosine residue for are now Schlessinger, 1993). Signalling pathways being specificity. In defined by interactions of SH2 domains with distinct contrast to EGF receptor and c-erbB-2, we could not of In phosphotyrosine residues within defined consensus detect binding c-erbB-3 to GRB2. this respect, c-erbB-3 at of the shows Trk sequences. It now seems possible that least some similarity to receptors which phosphorylate part SHC but do bind network of events following receptor activation might be not GRB2 (Suen et al., 1993). We conclude amino acid of the that the two predicted from the primary sequence potential binding sites predicted by Songyang to A et al. are not receptor. Most receptors appear signal through p21ra(. (1993) (i) autophosphorylated, (ii) do not bind that GRB2 links the activated or are number of reports have shown GRB2, (iii) they occupied by another protein. Clearly factor of it is possible that c-erbB-3 employs the EGF receptor to the nucleotide exchange p21ras, pathway thought to it close to the membrane. This is be used in that of sos, thereby localizing by v-Src, phosphorylation SHC by c- achieved interaction of the domain of GRB2 with erbB-3 of by SH2 promotes binding SHC to the GRB2 -sos complex of and the domain with a and of Y1068 the EGF receptor, SH3 subsequent activation p21ras. Indeed, we could motif of and precipitate a small amount of SHC from proline-rich sos (Buday Downward, 1993; EGF-stimulated 3T3 Chardin et et Gale et cells expressing the EGF al., 1993; Egan al., 1993; al., 1993; receptor/c-erbB-3 chimera using Li et al., 1993; Olivier et al., 1993; Rozakis-Adcock et the GRB2-GST fusion bound to al., protein glutathione-agarose 1993; Simon et not The al., 1993). (data shown). amount of SHC precipitated from An additional has been shown to be these cells with GRB2-GST was less than that observed protein, SHC, phosphorylated in response to EGF and to associate with for EGF-stimulated EGFR-3T3 cells, but significantly more GRB2. However, it has been shown that SHC does not form than that precipitated from NIH 3T3 cells. We have not yet of part the complex of sos, EGF receptor and GRB2 investigated whether Ras is activated in these cells in (Buday and the Downward, 1993), suggesting that formed response to EGF. complex between GRB2 and in the case of EGF SHC, receptor, forms In addition to SHC association with activated c-erbB-3, of a different which or part signalling pathway may may we also observed association of the p85 subunit of not converge with the p2lras signalling pathway as it is PI3-kinase. Like SHC and GRB2, p85 has no catalytic currently perceived. Some evidence that SHC may be domain and acts as an adaptor molecule which links the involved in p21ral signalling is provided by the observation activated tyrosine kinase with the catalytic p110 subunit of that the differentiation effect of SHC on PC12 cells is P13-kinase. p85 possesses two SH2 domains and one SH3 prevented by expression in these cells of a dominant negative domain. The presence of the p1O subunit in the immune Ras mutant (Pelicci et al., 1993). SHC is phosphorylated complexes from EGF-stimulated cells was confirmed by an in v-Src transformed cells (McGlade et al., 1992), resulting assay for P13-kinase activity. A protein comigrating with p85 in its binding to the GRB2 -sos complex. This association was phosphorylated in response to EGF, but to a lesser extent is proposed to stimulate guanine-nucleotide exchange on than SHC. Phosphorylation of p85 has only been observed Ras through an unknown mechanism. In contrast to GRB2, under conditions where the tyrosine kinase is overexpressed, 2838 receptor/c-erbB-3 chimera EGF and it seems that it is not essential for function (Hu et al., in that c-erbB-3 fails to interact with GRB2 in contrast to EGF receptor and c-erbB-2, whereas P13-kinase appears to 1992). The role of P13-kinase in cellular transformation was first suggested by Kaplan et al. (1985). PI3-kinase has been interact more strongly with c-erbB-3 than EGF receptor. This subsequently to interact with a number of growth suggests a mechanism for achieving signalling diversity since shown heterodimerization between related family members, which factor receptors, including PDGF receptor, CSF-1 receptor has been shown to occur between EGF receptor and c-erbB-2 and the insulin receptor substrate IRS1. PI3-kinase also (Goldman et al., 1990), could potentially activate a distinct associates to some extent with EGF receptor, although it has network of events. It will be interesting to investigate the been shown that p85 alone binds more avidly to EGF role of cross-talk between c-erbB-3 and other I growth receptor than the p85 -pl 0 active complex (Hu et al., type factor receptors. 1992). Relatively low levels of PI3-kinase activity have been found associated with activated EGF receptor as compared with PDGF receptor, suggesting that P13-kinase plays a Materials and methods minor role in EGF receptor signal transduction. The DNA constructs and vectors consensus binding site for p85 to phosphorylated tyrosine The cDNA clone for c-erbB-3 contained in the vector pCDM8 (Invitrogen) kinases is now well defined as the sequence YMXM or was a gift from Dr G.Plowman, Bristol-Myers Squibb Pharmaceutical possibly YXXM. In contrast to EGF receptor, c-erbB-3 Research Institute, Seattle, WA. This was used for transient expression in contains six potential PI3-kinase binding sites in its C- COS-1 cells. The EGFR cDNA was a gift from Dr G.Gill, UCSD, CA. terminal domain, and one in its kinase domain. Of these, For construction of the EGFR/c-erbB-3 chimera, two restriction fragments encoding the extracellular and transmembrane domains of EGF receptor only one is conserved in EGF receptor, c-erbB-2, and c- encoding amino acids 1-154; AflIL-NarI, encoding amino (NotI -Aflu, erbB-4 which is within the kinase domain and is not thought acids 155-647) and a restriction fragment encoding the intracellular portion to be phosphorylated in EGFR. We have shown that synthetic of c-erbB-3 (NarI-XbaI, encoding amino acids 651-1323) were inserted peptides corresponding to these sites (Y1035, Y1 178, into the vector pRc/CMV for evaluation by transient transfection in COS-1 Y1203, Y1241, Y1257 and Y1270) inhibited association of cells. The chimera construct was then subcloned into an LTR-based vector pLTRpoly (Lehvaslaiho et al., 1990) (provided by Dr K.Alitalo, Helsinki, p85 with c-erbB-3, suggesting that they could indeed act as Finland) for transfection into NIH 3T3 fibroblasts. docking sites for p85 if phosphorylated in vivo. These data suggest that PI3-kinase may play a more important role in Cell culture and transfections the direct signalling of c-erbB-3 than other family members. COS-1 (Gluzman, 1981), NIH 3T3 (Jainchill et al., 1969) and A431 (Giard et 1973) cells were obtained from the American Tissue Culture Collection al., The possibility clearly exists, however, for cross-talk via and NIH 3T3 cells transfected with an EGFR/c-erbB-2 chimera (NEN-7 heterodimerization with other type I family members, since cells; Lehvaslaiho et al., 1989) were a gift from Dr K.Alitalo, University we have shown that EGF receptor is able to phosphorylate of Helsinki, Finland. NIH 3T3 cells expressing - 105 EGF receptors per c-erbB-3 in vivo, at least under conditions where they are cell (EGFR-3T3) were kindly provided by Dr D.Hills, ICRF Oncology co-expressed at similar high levels. Other differences in the Group, Hammersmith Hospital, UK. Cells were grown in Dulbecco's modified Eagle medium (DMEM) (Life Technologies, UK) containing 10% signalling pathways of the type I family of growth factor fetal calf serum. 10 of plasmid with 1 yg of selection plasmid pRc/CMV yig receptors may exist, resulting in signalling diversity. (Invitrogen) were used to transfect 9 cm plates of NIH 3T3 fibroblasts using Whereas both EGF receptor and c-erbB-2 use phospholipase the calcium phosphate method (Graham and van der Eb, 1973). For selection as a substrate, c-erbB-3 contains no potential of transfected cells, 2.5 mg/ml G418 (Gibco) was included in the growth C'y medium. Nine centimetre plates of COS-1 cells were transfected with 10 phospholipase C^y binding site (Songyang et al., 1993). of DNA by the DEAE/Dextran method (Selden, 1992). jug Indeed, we were unable to detect binding of to c- PLCy erbB-3 by immunoblotting (data not shown). Selection of NIH 3T3 chimera clones It has been shown for PDGF receptor that if the tyrosine Transfected NIH 3T3 fibroblasts were labelled with the EGFR1 antibody (10 (Waterfield et al., 1982) and fluorescein-conjugated F(ab')2 residues within the p85 binding motif are mutated to prevent Ag/ml) rabbit anti-mouse immunoglobulins (1:20 dilution) (DAKOPATTS) and P13-kinase interaction, the ability of the PDGF receptor to subjected to two rounds of FACS sorting using a Becton Dickinson stimulate Ras activity is lost, as is its ability to activate raf FACStarPLUS sorter. At each sort, 5% of the most strongly fluorescent cells and MAP kinase and to stimulate DNA synthesis (Fantl were selected. Cells were then single-cell cloned and analysed for their ability et al., 1992). Others have shown that if the tyrosine residues 125I-labelled EGF (Amersham, UK) as described previously to bind (Aharonov et al., 1978). which bind PI3-kinase (740/751), PLCy (1021), GAP (771) and SH-PTP2 (1009) are all mutated, PDGF is unable to Site-directed mutagenesis DNA synthesis. When binding of activate Ras or stimulate A SacI fragment of c-erbB-3 (site in vector to base 3522) was subcloned was restored, Ras activation and either PI3-kinase or PLCGy mutagenesis was performed using the into M13mpl9. Site-directed mutagenesis kit using the oligonucleotides and Kazlauskas, 1993). Amersham oligonucleotide-directed mitogenesis were normal (Valius the Asn to Asp mutation) and (for CGGGCAGCCAGGTCTCTATGCAC that the GRB2 - sos complex is not These results suggest (for the His to Glu mutation) CTGCCAATGGCCAGCATTTCATCTG for Ras. the only mechanism stimulating The mutant clones were sequenced and the together in the same reaction. of transduction cannot simply Clearly the processes signal fragment (bases 2382-3000) was replaced in the mutated BamHI-Bgll the interaction of receptors with GRB2. full-length c-erbB-3 cDNA. be explained by an important role in cell signalling, although P13-kinase plays Immune-complex kinase assays for the products the precise targets phosphatidylinositol transfected with c-erbB-3 or mutant 9 cm plates of COS-1 cells transiently elusive. The role on the D3 remain phosphorylated position His/Glu740 and Asn/Asp815 mutations were lysed c-erbB-3 containing the in transduction is even less well defined, but with polyclonal antibody 49.3 of SHC signal and c-erbB-3 protein was immunoprecipitated et 1992). Immunoprecipitates were for kinases as described previously (Prigent al., use as a primary target tyrosine its extensive twice with phosphorylation buffer (50 mM HEPES pH 7.4, 5% washed for this We have shown a vital function protein. suggests Triton 150 mM 2 mM 12 mM 0.2% NaCl, X-100, MnCl2, glycerol, are for c-erbB-3. both SHC ard P13-kinase targets that sodium and 49.3 et al., 100 orthovanadate) (Prigent peptide MgCl2, 1tM distinct differences in the we have identified of 2.5 and incubated at Moreover, to a concentration was added mg/ml 37°C 1992) 10 tCi related c-erbB-3 from the which interact with closely tyrosine kinases, for 10 to elute [_y-32P]ATP proteins min antibody. 2839 and W.J.Gullick S.A.Prigent (Amersham, UK) was added for 10 min at room temperature (100 final as described above. Precipitates were then washed twice with 20mM HEPES ltM concentration, 1 Ci/mmol). Precipitates were boiled in sample buffer and pH 7.2, 5 mM MnCl2, 0.13 mM Na orthovanadate and 12.5 mM NaF. analysed by SDS-PAGE (Schagger and von Jagow, 1987). Proteins were Assays were performed in 100 of the same buffer containing 0.2 mg/ml A1 transferred to Immobilon-P PVDF membranes (Millipore) and subjected phosphatidylinositol, 10 and 50 ATP for 30 min at [-y-32P]ATP ACi ItM Bands of interest were excised and analysed for 37°C. Lipids were extracted as described previously (Morgan et al., to autoradiography. 1990), phosphoamino acid content as described previously (Kamps and Sefton, and analysed by TLC using the borate system (Walsh et For al., 1991). To compare the level of expression of the wild-type and mutant inhibition of P13-kinase, 1% Triton X-100 was included in the reaction buffer. 1989). proteins, transfected cells were metabolically labelled in parallel with proteins were immunoprecipitated with antibody Inhibition of p85 and SHC binding to using [35S]methionine, 49.3, EGFRIc-erbB-3 and electrophoresed on SDS-polyacrylamide gels which phosphopeptides eluted with peptide were treated with Amplify (Amersham) for fluorography, as described Synthetic peptides, as indicated in Table II, were synthesized by Nicola et al., 1992). O'Reilly and Elizabeth Li, Peptide Synthesis Unit, ICRF, London, UK. previously (Prigent EGFR/c-erbB-3-3T3 cells were permeabilized in the presence of 70 AM EGF stimulation of whole cells peptide as described previously (Buday and Downward, 1993), and treated in six-well dishes were washed and routinely incubated with EGF (5 x 10-7 M) for 4 min at room temperature. EGFR/c-erbB-3 Cell monolayers DMEM in the presence or absence of synthetic protein was precipitated with 49.3 as described above, but in the presence for 2 min at 37°C with M), supplied by Dr R.Goodlad, ICRF, London, UK. The of 70 peptide. Bound proteins were eluted with 49.3 peptide (2.5 mg/ml) EGF 1-48 (10-6 AM and cells were lysed in lysis buffer containing and analysed by SDS-PAGE and immunoblotting for c-erbB-3, p85 and medium was removed pH 7.4, 1 Triton X-100, 5mM EGTA, SHC, as described above. phosphatase inhibitors (50mM Tris % 150 mM NaCl, 25 mM benzamidine, 2 mM PMSF, 1 mM sodium Formation of receptor - GRB2 complexes in vitro orthovanadate, 100mM NaF, 10mM sodium pyrophosphate). The protein 6 of GRB2-GST fusion protein prepared as described concentration of cell lysates was determined by the Bradford protein assay previously (Egan Ag were bound to 35 of a of (Bio-Rad Laboratories Ltd, Defined amounts of protein were analysed et al., 1993) suspension glutathione agarose UK). A1 (Pharmacia, UK) for 30 min at room temperature. Lysate was by SDS-PAGE and immunoblotting with a rabbit polyclonal anti- prepared from 175 cm2 monolayers of cells [NIH 3T3, NEN7, EGFR-3T3 and phosphotyrosine antibody provided by Dr T.S.Pillay, Royal Postgraduate EGFR/c-erbB-3-3T3 (clone 18 as described in Results)], which had been Medical School, London, UK (Pillay and Makgoba, 1992). To determine treated with EGF (10-6 M) for 4 min, and from untreated cells. Cell the time course for activation of c-erbB-3, cells were incubated in six-well lysate was incubated for 2 h at 4°C with GRB2-GST-glutathione agarose over a period of up to 45 min with EGF (10-6 M). Cells were lysed plates complexes, precip-tates were washed extensively with TBS/0.1 in boiling sample buffer at the indicated times to terminate phosphorylation % Triton X-100 and bound proteins were eluted in reactions rapidly. Samples were subjected to SDS-PAGE and sample buffer and analysed by SDS-PAGE and immunoblotting. Rabbit polyclonal antibodies 12 E immunoblotting with anti-phosphotyrosine antibody. (Gullick et al., 1985), 21 N (Gullick et al., 1987) and 49.3 (Prigent et al., 1992) were used for detection of EGF receptor, EGF receptor/c-erbB-2 Determination of anchorage-independent growth chimera protein, and EGF receptor/c-erbB-3 chimera protein, respectively. Cells were seeded in methyl cellulose (Sigma) suspension (Methocel) as The sensitivity of 49.3 and 12 E were also compared by probing described previously (Lemoine et al., 1989). 35 mm-diameter wells of six- antibodies immunoblots of whole cell lysate from EGFR-3T3 and EGFR/c-erbB-3-3T3 well plates were coated with DMEM containing 0.9% agarose and duplicate cells (clone 18). wells were seeded with 30 000 cells suspended in DMEM containing 1.2% Methocel and 10% fetal calf serum, in the presence or absence of EGF or 30 nM). Colonies >0.1 mm were scored after 2 weeks. (3 Acknowledgements incorporation [3H]Thymidine We are grateful to Dr G.Plowman for supplying the c-erbB-3 clone, to Dr 96-well plates were seeded with 1000 cells/well and grown to 80% G.Gill for supplying the EGFR clone, to Dr K.Alitalo for supplying the confluence. The medium was replaced with DMEM containing 0.5% FCS NEN-7 cells, to Dr D.Hills for supplying the EGFR-3T3 cells, to Drs nM and cells were incubated for 48 h. EGF was added (0.05-50.0 final L.Cantley and S.Soltoff for providing the and to Dr anti-p85 antibody and cells were incubated for a further 16 h before addition concentration) T.S.Pillay for providing the antibody. We wish to thank anti-phosphotyrosine repeated of [3H]thymidine (0.2 PCi/well) for 4 h. Cells were ruptured by Dr S.Egan for providing the pGEX vector containing the GRB2-GST firezing and thawing and analysed on an LKB Wallac 1295-001 cell harvester sequence. We are also grateful to Dereck Davis and Simon Monard for and 1205 Betaplate liquid scintillation counter (LKB Pharmacia, UK). performiing the FACS sorting of NIH 3T3 transfectants, and to Nicola O'Reilly and Elizabeth Li for synthesizing the tyrosine phosphopeptides. Immunoprecipitation and immunoblotting We also wish to dtank Dr J.Feramisco for allowing some of the experiments Following EGF stimulation, cells were lysed in 50 mM Tris pH 7.4, 1% to be performed in his laboratory. Triton X-100, 5 mM EGTA, 150mM NaCl, 25 mM benzamidine, 2 mM mM 1 mM Na orthovanadate. EGF receptor/c-erbB-3 PMSF, 10 NaF and was for 2 h with 5 ug of purified rabbit chimeric protein precipitated References antibody 49.3 (Prigent et al., 1992) which reacts specifically polyclonal with the intracellular domain of c-erbB-3, or EGFR1 which reacts with the Aaronson,S.A. (1991) Science, 254, 1146-1152. extracellular domain of EGF receptor (Waterfield et al., 1982) and 2.5 mg Aharonov,A., Pruss,R.M. and Herschman,H.R. (1978) J. Bio. Chem., of Immunoprecipitates were washed extensively with protein A-Sepharose. 253, 3970-3977. TBS (50mM TrispH 7.4 and 150mM NaCl) containing 0.1% Triton X-100. Bar-Sagi,D., Rotin,D., Batzer,A., Mandiyan,V. and Schlessinger,J. (1993) EGFR/c-erbB-3 protein was then eluted from the protein A-Sepharose at Cell, 74, 83-91. 37°C for 15 min with 40 sl of a 5 mg/ml solution of the 49.3 peptide. Buday,L. and Downward,J. (1993) Cell, 73, 611-620. Eluates were analysed on 7.5% SDS-polyacrylamide gels as above. Proteins Chardin,P., Camonis,J.H., Gale,N.W., Van Aelst,L., Schlessinger,J., were transferred to nitrocellulose and probed with rabbit antibodies specific Wigler,M.H. and Bar-Sagi,D. (1993) Science, 260, 1338-1343. for c-erbB-3 (49.3), for the p85 subunit of PI3-kinase (provided by Drs Clark,S.G., Stern,M.J. and Horvitz,H.R. (1992) Nature, 356, 340-344. L.Cantley and S.Soltoff, Tufts University School of Medicine, Boston, MA, Di Fiore,P.P., Pierce,J.H., Fleming,T.P., Hazan,R., Ullrich,A., King,C.R., and commercially available from UBI, New York), antibodies specific for Schlessinger,J. and Aaronson,S.A. (1987a) Cell, 51, 1063-1070. SHC (UBI, New York) or antibodies reacting with phosphotyrosine (provided Di Fiore,P.P., Pierce,J.H., Kraus,M.H., Segatto,O., King,C.R. and by Dr T.S.Pillay, RPMS, UK). Immunoblots were developed using the Aaronson,S.A. Science, 237, 178-182. (1987b) ECL system, (Amersham, UK). When probing with more than one antibody Brewster,C.E.P., Jones,M.D., Pelicci,G. Dilworth,S.M., Lanfirancone,L., was required, blots were routinely stripped for 1 h at 60°C with 2% SDS, and Pelicci,P.G. (1994) Nature, 367, 87-90. 100mM in 62.5 mM Tris, pH 6.8, blocked and reprobed. S-mercapoethanol Egan,S.E., Giddings,B.W., Brooks,M.W., Buday,L., Sizeland,A.M. and Weinberg,R.A. (1993) Nature, 363, 45-51. P13-kinase assays Escobedo,J.A., Navankasattusas,S., Kavanaugh,W.M., Milfay,D., P13-kinase were as Fried,V.A. and Williams,L.T. (1991) Cell, 65, 75-82. described previously assays performed essentially et EGFR/c-erbB-3 was Fantl,W.J., Escobedo,J.A., Martin,G.A., Turck,C.W., del Rosario,M., al., 1990). protein precipitated (Morgan Briefly, from EGF-treated or untreated NIH 3T3 cells or EGFR/c-erbB-3-3T3 cells McCormick,F. and Williams,L.T. (1992) Cell, 69, 413-423. 2840 chimera EGF receptor/c-erbB-3 Feng,G.-S., Hui,C.-C. and Pawson,T. (1993) Science, 259, 1607-1611. Rozakis-Adcock,M., Femnley,R., Wade,J., Pawson,T. and Bowtell,D. Freeman,R.M.,Jr, Pluzky,J. and Neel,B.G. (1992) Proc. NatlAcad. Sci. (1993) Nature, 363, 83-85. USA, 89, 11239-11243. Schagger,H. and von Jagow,G. (1987) Anal. Biochem., 166, 368-379. Gale,N.W., Kaplan,S., Lowenstein,E.J., Schlessinger,J. and Bar-Sagi,D. Seedorf,K., Felder,S., Millauer,B., Schlessinger,J. and Ullrich,A. (1991) (1993) Nature, 363, 88-92. J. Biol. Chem., 266, 12424-12431. Giard,D.J., Aaronson,S.A., Todaro,G.J., Arnstein,P., Kersey,J.H., Segatto,O., Pelicci,G., Giuli,S., Digiesi,G., Di Fiore,P.P., McGlade,J., Dosik,H. and Parks,W.P. (1973) J. Natl Cancer Inst., 51, 1417-1423. Pawson,T. and Pelicci,P.G. (1993) Oncogene, 8, 2105-2112. Gluzman,Y. (1981) Cell, 23, 175-182. Selden,R.F. (1992) In Ausubell,F.M., Brent,R., Kingston,R.E., Goldman,R., Ben Levy,R., Peles,E. and Yarden,Y. (1990) Biochemistry, Moore,D.D., Seidman.J.G., Smith,J.A. and Struhl,K. (eds), Current 29, 11024-11028. Protocols in Molecular Biology. Vol. 2, John Wiley and Sons, New York, Graham,F.L. and van der Eb,A.J. (1973) Virology, 52, 456-467. NY, pp. 9.2.1-9.2.6. Gullick,W.J. (1991) Br. Med. Bull., 47, 87-98. Shen,S.-H., Bastien,L., Posner,B.I. and Chretien,P. (1991) Nature, 352, Gullick,W.J., Downward,J. and Waterfield,M.D. (1985) EMBO J., 4, 736-739. 2869-2877. Simon,M.A., Dodson,G.S. and Rubin,G.M. (1993) Cell, 73, 169-177. Gullick,W.J., Berger,M.S., Bennett,P.L.P., Rothbard,J.B. and Songyang,Z. et al. (1993) Cell, 72, 767-778. Waterfield,M.D. (1987) Int. J. Cancer, 40, 246-254. Stahl,M.L., Ferenz,C.R., Kelleher,K.L., Kriz,R.W. and Knopf,J.L. (1988) Hanks,S.K., Quinn,A.M. and Hunter,T. (1988) Science, 241, 42-52. Nature, 332, 269-272. Hu,P., Margolis,B., Skolnik,E.Y., Lammers,R., Ullrich,A. and Suen,K.-L., Bustelo,X.R., Pawson,T. and Barbacid,M. (1993) Mol. Cell. Schlessinger,J. (1992). Mol. Cell. Biol., 12, 981-990. Biol., 13, 5500-5512. Hudziak,R.M., Schlessinger,J. and Ullrich,A. (1987) Proc. NatlAcad. Sci. Tan,J.C., Nocka,K., Ray,P., Traktman,P. and Besmer,P. (1990) Science, USA, 84, 7159-7163. 247, 209-212. Jainchill,J.L., Aaronson,S.A. and Todaro,G.J. (1969) J. Virol., 4, 549-553. Trahey,M. et al. (1988) Science, 242, 1697-1700. Kamps,M.P. and Sefton,B.M. (1989) Anal. Biochem., 176, 22-27. Valius,M. and Kazlauskas,A. (1993) Cell, 73, 321-334. Kaplan,D., Schaffthausen,B., Cantley,L. and Roberts,T. (1985) Nature, 315, Vogel,W., Lammers,R., Huang,J. and Ullrich,A. (1993) Science, 259, 239-242. 1611- 1614. Knighton,D.R., Zheng,J., Ten Eyck,L.F., Ashford,V.A., Xuong,N.-H., Walsh,J.P., Caldwell,K.K. and Majerus,P.W. (1991) Proc. NatlAcad. Sci. and Sowadski,J.M. (1991) Science, 253, 407-414. USA, 88, 9184-9187. Taylor,S.S. Kraus,M.H., Issing,W., Miki,T., Popescu,N.C. and Aaronson,S.A. (1989) Waterfield,M.D., Mayes,E.L.V., Stroobant,P., Bennett,P.L.P., Young,S., Proc. Natl Acad. Sci. USA, 86, 9193-9197. Goodfellow,P.N., Bantng,G.S. and Ozanne,B. (1982) J. Cell. Biochem., Fedi,P., Starks,V., Muraro,R. and Aaronson,S.A. (1993) 20, 149-161. Kraus,M.H., Proc. Natl Acad. Sci. USA, 90, 2900-2904. Received on October 1, 1993; revised on April 15, 1994 Shepard,H.M. and Mendelsohn,J. (1991) Mol. Cell. Biol., 11, Kumar,R., 979-986. Lehtola,L., Sistonen,L. and Alitalo,K. (1989) EMBO J., Lehvaslaiho,H., 8, 159-166. Lehvaslaiho,H., Sistonen,L., diRenzo,F., Partanen,J., Comoglio,P., and Alitalo,K. (1990) J. Cell Biochem., 42, 123-133. Holtti,E. Lemoine,N.R., Mayall,E.S., Jones,T., Sheer,D., McDermid,S., Kendall- (1989) Br. J. Cancer, 60, 897-903. Taylor,P. and Wynford-Thomas,D. etal. (1992a) Br. J. Cancer, 66, 1116-1121. Lemoine,N.R. Lobresco,M., Leung,H., Barton,C., Hughes,C.M., Lemoine,N.R., and Kloppel,G. (1992b) J. Pathol., 168, Prigent,S.A., Gullick,W.J. 269-273. Nature, 363, 85-88. Li,N. et al. (1993) Livneh,E., Benveniste,M., Prywes,R., Felder,S., Kam,Z. and J. Cell Biol., 103, 327-31. Schlessinger,J. (1986) and Gullick,W.J. (1991) In Dickson,R.B. and Lippman,M.E. Lofts,F.J. Hormones: Advances in Cellular and (eds), Genes, Oncogenes and of Breast Cancer. Kluwer Academic Publishers, Molecular Biology pp. 161-179. Boston, USA, Vol. HI, et al., (1992) Cell, 70, 431-442. Lowenstein,E.J. Pelicci,G., Pelicci,P.G. and Pawson,T. (1992) Proc. McGlade,J., Cheng,A., Natl Acad. Sci. USA, 89, 8869-8873. Koch,C.A., Sadowski,I. and Pawson,T. (1988) Oncogene, Moran,M.F., 3, 665-672. and Parker,P.J. (1990) Eur. J. Biochem., 191, Morgan,S.J., Smith,A.D. 761 -767. Weismuller,K.-H., Jungs,G., Schlessinger,J. Obermeier,A., Lammers,R., J. Biol. Chem., 268, 22963 -22966. and Ullrich,A. (1993) et al. 73, 179-191. Olivier,J.P. (1993) Cell, and (1993) Curr. Biol., 3, 434-442. Pawson,T. Schlessinger,J. et al. Cell, 70, 93-104. Pelicci,G. (1992) et al. (1993) J. Cell. Biochem., Suppi. 17A, 238. Pelicci,G. FEBS Lett., 308, 38-42. Pillay,T.S. and Makgoba,M.W. (1992) Whitney,G.S., Neubauer,M.G., Green,J.M., Plowman,G.D., Todaro,G.J. and Shoyab,M. (1990) Proc. NatlAcad. McDonald,V.L., 87, 4905-4909. Sci. USA, Green,J.M., Culouscou,J.-M., Whitney,G.S., Plowman,G.D., Proc. Neubauer,M.G. and Shoyab,M. (1993) Carlton,G.W., Foy,L., 1746-1750. Natl Acad. Sci. USA, 90, Selden,C. and Lemoine,N.R., Hughes,C.M., Plowman,G.D., Prigent,S.A., 1273-1278. Oncogene, 7, Gullick,W.J. (1992) Lemoine,N.R. and Gullick,W.J. J. Gooden,C.S.R., (1993) Rajkumar,T., 271-278. Pathol., 170, and Schlessinger,J. Ullrich,A. (1984) Nature, 324, Riedel,H., Dull,T.J., 68-70. et al. 689-692. Rozakis-Adcock,M. (1992) Nature, 360, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Identification of c‐erbB‐3 binding sites for phosphatidylinositol 3′‐kinase and SHC using an EGF receptor/c‐erbB‐3 chimera.

Prigent, S.A.; Gullick, W.J.
The EMBO Journal , Volume 13 (12) – Jun 1, 1994
Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/identification-of-c-erbb-3-binding-sites-for-phosphatidylinositol-3-RQLgaxbVH2

References (0)

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1994
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1002/j.1460-2075.1994.tb06577.x
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal vol.13 no.12 pp.2831 -2841, 1994 Identification of c-erbB-3 binding sites for phosphatidylinositol 3'-kinase and SHC using an EGF receptor/c-erbB-3 chimera threonine sites which are present in potential regulatory A.Prigent1 and William J.Gullick2 Sally and c-erbB-4, suggesting that c-erbB-3 may EGFR, c-erbB-2 Molecular Oncology Laboratory, ICRF Oncology Group, to negative regulation by protein kinase C not be subject Hospital, Du Cane Road, London W12 OHS, UK Hammersmith et al., 1990). These data suggest that c-erbB-3 (Plowman 'Present address: Departments of Medicine and Pharmacology, different properties to other family may possess slightly of California, San Diego, La Jolla, CA 92093-0636, USA University members. author 2Corresponding advances have been made recently in understanding Great J.Brockes Communicated by which kinase receptors interact the mechanism by tyrosine signalling components. Following ligand c-erbB-3 is a member of the I receptor-related) with intracellular type (EGF on specific tyrosine residues of growth factor receptors for which no ligand has binding, autophosphorylation family of to the activated To facilitate ligand stimulation we have facilitates the recruitment proteins been identified. is achieved by the high-affinity a chimeric receptor which possesses an receptor. This association constructed elements called src homology kinase and promotes the growth of NIH 3T3 interaction of characteristic activatable residues (Pawson and fibroblasts. In this study we have shown that SHC and 2 (SH2) domains with phosphotyrosine A large number of proteins containing 3'-kinase bind to the activated EGF Schlessinger, 1993). phosphatidylinositol now been identified, including those with receptor/c-erbB-3 chimera. Whereas p85 is not SH2 domains have as the p21ras GTPase-activating to a significant extent, SHC appears to enzymatic activity such phosphorylated et al., 1988), phosphotyrosine phosphatases for phosphorylation on tyrosine. In protein (Trahey be a major substrate Freeman et al., 1992; Feng et al., 1993; to EGF receptor and c-erbB-2, we were unable (Shen et al., 1991; contrast and phospholipase C'y (Stahl et al., of activated c-erbB-3 to GRB2. Using Vogel et al., 1993) to detect binding with no obvious enzymatic activity corresponding to each of 13 potential 1988), and those synthetic peptides et al., 1992), which is the sites on c-erbB-3, we have shown that including GRB2 (Lowenstein phosphorylation Caenorhabditis elegans for SHC binding. Peptides mammalian homologue of the tyrosine 1309 is responsible sem-5 (Clark et al., 1992) and SHC (Pelicci et al., containing the motif YXXM inhibit p85 association. By protein Additional motifs called src homology 3 (SH3) comparison with recently reported SHC binding sites on 1992). T and Trk we have identified a SHC domains have also been identified in a number of these Middle antigen NPXY. which direct cellular localization of signalling binding motif, proteins 1993). Those SH2-containing factor receptors/ molecules (Bar-Sagi et al., Key words: c-erbB-3/growth which do not possess enzymatic activity act as phosphorylation proteins functional elements with the molecules linking other adaptor An example of this is the p85 protein activated receptors. 110 subunit of phosphatidylinositol which interacts with the p Introduction the enzymatic activity 3'-kinase which possesses (PI) Other examples include GRB2 c-erbB-3 was cloned almost simul- (Escobedo et al., 1991). The gene encoding couple the EGF receptor with the et al., 1989; Plowman which has been shown to taneously by two groups (Kraus binding both activated EGF kDa glycoprotein which signalling pathway by et al., 1990). It encodes a 160-180 p21rcLI nucleotide exchange protein sos and the guanine has all the structural features of the EGF family of tyrosine receptor Chardin et al., 1993; Egan also includes EGF receptor (EGFR), and Downward, 1993; kinase receptors [which (Buday Li et al., 1993; Olivier et al., et al., 1993)]. The highest et 1993; Gale et al., 1993; c-erbB-2 and c-erbB-4 (Plowman al., et al., 1993; Simon et al., 1993). identity with c-erbB-3 is within the 1993; Rozakis-Adcock proportion of sequence and associates with EGFR (60%), c-erbB-2 (62%) and SHC is phosphorylated by physically catalytic domains of and c-erbB-2 (Segatto et al., 1993). EGFR, c-erbB-2 and c-erbB-4 are activated EGF receptor c-erbB-4 (63%), although it appears to then interact with similar to each other than to c-erbB-3 Following phosphorylation somewhat more et The precise role of GRB2 (Rozakis-Adcock al., 1992). 80% sequence identity between receptor pairs). More- (-- is less well defined than that for GRB2, but it is the family member to contain unusual this protein over, c-erbB-3 only at least in the also be involved in p2lras signalling, within the kinase domain, including His740 and may residues kinases such as v-Src which of which are as glutamate and aspartate, case non-receptor tyrosine Asn815 present to GRB2. cannot couple directly in other kinases (Plowman et al., 1990). respectively, residues mapping the specific phosphotyrosine of the corresponding aspartate residue to asparagine By Mutation domain-containing proteins, it for binding SH2 kinase completely abolished kinase responsible in the tyrosine v-fps to the murine white has been identify phosphotyrosine-containing et 1988). Similarly, possible activity (Moran al., for different binding results from the same amino acid consensus sequences responsible spotting (W42) phenotype is now to to a and it predict c-kit which also encodes molecules, possible in the signalling replacement proto-oncogene molecules are to be which effector likely et c-erbB-3 lacks certain extent kinase al., 1990). Finally, a defective (Tan Press Oxford University S.A.Prigent and W.J.Gullick were constructed and analysed similarly. The double mutant involved in the signalling of a novel receptor (Songyang showed, if anything, a reduced activity relative to the wild- et al., 1993). Analysis of the c-erbB-3 amino acid sequence type protein (Figure 1). reveals several potential binding sites for PI3-kinase, a c-erbB-3 site The autocatalytic activity of and mutant phosphotyrosine phosphatase (SH-PTP2) binding wild-type which et was also compared in vivo (Figure 2). COS-1 cells, (Freeman et al., 1992; Feng et al., 1993; Vogel al., 1993) et sites for GRB2. express -50 000 EGF receptors per cell (Livneh al., and two possible binding and mutant c-erbB-3. function of the c-erbB-3 1986), were transfected with In this work, we investigated the wild-type of of indicator cells. In The transfected cells expressed a similar quantity protein and its effect on the growth second endogenous EGF receptors and c-erbB-3 protein, as particular, its interaction with specific messenger determined by immunoprecipitation from [35S]methionine- pathways was explored. labelled cells (data not shown). c-erbB-3 immunoprecipitates were analysed by immuno- Results blotting with an anti-phosphotyrosine antibody. The extent of phosphorylation of the mutant c-erbB-3 protein was Comparison of wild-type and mutant c-erbB-3 in vitro and in vivo Ni UT WT The kinase activity of c-erbB-3 was assessed by immune- transiently transfected COS-1 cells as 49.3 PY complex assay, using 49.3 PY of c-erbB-3 and an antibody a source protein anti-peptide + - + + - + k Da for c-erbB-3 for immunoprecipitation (Figure (49.3) specific - 200 '00 ....i.:.. antibody reacts with the intracellular domain 1). Since this :1: of and could potentially inhibit kinase activity, the .~ c-erbB-3 ..gv`..... c-erbB-3 protein was eluted with the corresponding peptide T, to the kinase assay. This peptide contains no tyrosine prior - 116 - 116 residues and therefore could not act as a substrate. c-erbB-3 showed a low level of autophosphorylation in immune- very -80 -80 which was on the of2-4 complex assays dependent presence mM Mn2+ ions and of Mg2+ ion concentration independent not acid revealed that (data shown). Phosphoamino analysis -49.5 - 49.5 was on and serine residues (data phosphorylation tyrosine To determine whether basal autocatalytic activity not shown). be enhanced His740 and Asn815 the could by replacing by of and mutant c-erbB-3 kinase Fig. 2. Comparison wild-type activity more favourable residues Glu and Asp, mutants potentially in vivo. Wild-type (WT) and mutant (MUT) c-erbB-3 proteins were in COS-1 cells. c-erbB-3 was precipitated transiently expressed protein from unstimulated cells (-) or cells which had been stimulated with antibody 49.3. Bound proteins were eluted EGF (+) using polyclonal with 49.3 and analysed by immunoblotting with peptide (2.5 mg/ml), A B WT MUT V 49.3 to the levels of the constructs antibody compare expression WT MUT V and with an antibody (PY). (49.3), anti-phosphotyrosine T E T E T E EL EL CL kDa XL_ kDa + + G Cy) c ) Cv) IL cD 200- C D ,t eti Lr) kDa 200- N."'. 0 0 97 - 97- 69 - Fig. 1. Comparison of wild-type and mutant c-erbB-3 kinase activity chimera in NIH 3T3 Fig. 3. Expression of EGF receptor/c-erbB-3 in vitro. The wild-type (WT) and mutant (MUT) c-erbB-3 proteins cells. To test for correct of the EGFR/c-erbB-3 chimera, expression were at similar levels as determined by immunoprecipitation cm expressed was protein precipitated from a 9 plate of [35S]methionine-labelled from metabolically labelled COS-1 cells using serum 49.3 (A). T and EGFR/c-erbB-3-3T3 cells (clone 18) using an antibody against the E refer to the total precipitated protein released by SDS, and that EGFR extracellular domain (EGFR1), an anti-peptide antibody reacting which may be eluted by addition of 49.3 peptide, respectively. Kinase with the juxtamembrane domain of c-erbB-3 (61.3) and an anti-peptide on assays performed immune-complexes from COS-1 cells transfected antibody reacting downstream of the kinase domain of c-erbB-3 (49.3). with c-erbB-3 (WT), mutant c-erbB-3 (MUT) and vector (V) wild-type Reaction with anti-peptide antibodies was inhibited by prior incubation are shown (B). with the corresponding peptide (PEP). 2832 chimera EGF receptor/c-erbB-3 + EGF N- r_ 0 + .SE _ _ z <: kDa + + kDa ~ ~ L -(JC,~ 200- kDa 200- p0 97- *4 is 69 - 97 - 0 w w 4_ * 69- *WI 30- 46- Fig. 4. EGF stimulation of phosphorylation. (A) Whole cell lysates of stimulated EGFR/c-erbB-3-3T3 cells were analysed by SDS-PAGE and immunoblotting with anti-phosphotyrosine antibodies. (+) and (-) refer to EGF-stimulated and unstimulated cells, respectively. In the case of NEN7 and A431 cells, one tenth of the amount of protein was used. (B) EGFR/c-erbB-3-3T3 cells were incubated with EGF for the indicated times. Phosphorylation was terminated rapidly by the addition of boiling sample buffer to cells which were analysed by SDS-PAGE and immunoblotting with anti-phosphotyrosine antibody. reduced relative to the wild-type. Phosphorylation of both of EGFR/c-erbB-3-3T3 cells Table I. Anchorage-independent growth mutant and wild-type c-erbB-3 was increased by treatment of the cells with EGF prior to lysis. Colonies per 30 000 cells 0 nM EGF 3 nM EGF 30 nM EGF Construction and of EGFR/c-erbB-3 1 1 86 16 78 I 1 expression Chimera clone 18 f + + 69 chimera in COS-1 and NIH 3T3 cells Chimera clone 42 50 + 16 205 + 2 1011 NIH mock 3T3 a system, the EGFR/c-erbB-3 To generate ligand-activatable transfected 1 1 3 1 2 1 chimera was fused inside the cell membrane at a common just Narl site. When in COS-1 cells, the protein was expressed 42 and mock- Clones 18 and expressing the EGFR/c-erbB-3 chimera as deduced by correctly located at the cell surface transfected NIH 3T3 cells were suspended in 1.2% Methocel in the protein from intact and precipitating the EGFR/c-erbB-3 >0.1 mm presence and absence of EGF (3 and 30 nM). Colonies for the intracellular and are the results of one lysed cells with antibodies specific diameter were scored after 2 weeks. Data shown representative experiment. The of extracellular domains (Kumar et al., 1991). ability cross- the chimeric protein to bind EGF was confirmed by linking of [125I]EGF to intact cells followed by immuno- NIH 3T3 cell lines EGF stimulation of phosphorylation in NIH 3T3 cells precipitation (data not shown). Stable created and analysed by EGF chimera expressing EGFR/c-erbB-3 were expressing the receptor/c-erbB-3 of EGF stimulated the of the chimeric protein binding assays using [1251]EGF to determine the number phosphorylation cell et 1978). Clones 18 and in 18 and 42 as deduced by immunoblotting of whole receptors per (Aharonov al., clones - 130 000 and 70 000 cell, cell with anti-phosphotyrosine antibodies (Figure 4A). 42, possessing receptors per lysates for functional studies. These The identity of the stimulated 180 kDa chimeric protein was respectively, were selected were verified FACS confirmed of the phosphoprotein receptor numbers by comparative by immunoprecipitation EGFR1 mAb to the common EGFR antibodies specific for EGFR and c-erbB-3 (data not using the with analysis to label the and NEN7 cells For NEN7 cells expressing the extracellular domain cells, shown). comparison, 2 x 2 x 106 chimeric molecules chimera and A431 cells expressing 106 EGFR/c-erbB-2 per EGFRIc-erbB-2 expressing When 10-fold less total et To check that EGF per cell were used. cell as a standard (Lehvaslaiho al., 1989). receptors in of NEN7 and A431 contained the c-erbB-3 protein was loaded the case cells, the expressed protein complete antibodies two distinct sites a similar signal on the anti-phosphotyrosine immunoblot was cytoplasmic domain, recognizing that the of stimulation the c-erbB-3 kinase domain et al., 1992) obtained, suggesting degree receptor within (Prigent three members is used to the EGFR/c-erbB-3 protein, and the obtained for the family comparable. were precipitate were raised were used of c-erbB-3 was maximal which the antibodies Stimulation autophosphorylation peptides against The s and decreased to almost basal level after 45 to for c-erbB-3 3). after 15 min, specifically compete binding (Figure and/or at - 180 kDa on due to by phosphatases EGFR/c-erbB-3 chimera migrated probably dephosphorylation an of the Moreover, SDS down-regulation receptor (Figure 4B). -polyacrylamide gels. 2833 and S.A.Prigent W.J.Gullick CONCENTRATION EGF (nM) 0 3 Clone Clone 3T3 NIH mock transfected in 18 42 Fig. 5. Growth of colonies 1.2% Methocel. Clones and the EGFR/c-erbB-3 chimera and mock-transfected NIH 3T3 cells were expressing in mm suspended and grown 1.2% Methocel. Colonies were after 3 weeks. The to 0.5x0.6 fields. photographed panels correspond activated additional band of -54 kDa was seen in the Association of EGFR/c-erbB-3 with SHC consistently EGFR/c-erbB-3-3T3 cells on Since the EGF receptor/c-erbB-3 chimera was able to repeated experiments (Figure in 4B). This was absent in EGF-stimulated NIH promote a growth response NIH 3T3 cells, we were phosphoprotein 3T3 cells and unstimulated EGFR/c-erbB-3-3T3 and interested to determine which elements of established SH2 cells, in to the activated c-erbB-3 signalling pathways were involved. To determine whether appeared parallel protein, that it be a substrate for SHC associated with activated NIH 3T3 cells suggesting may major c-erbB-3. c-erbB-3, expressing an EGFR/c-erbB-3 chimera 18 as (clone were described above) stimulated with EGF. To specifically - x and of precipitate EGFR/c-erbB-3 protein, lysates from 2 106 Anchorage-dependent -independent growth were an NIH 3T3 cell lines the EGF cells incubated with antibody raised the expressing receptor! against c-erbB-3 chimera intracellular portion of c-erbB-3 (49.3). This eliminated the EGF addition to EGFR/c-erbB-3-3T3 cells enabled them to possibility of cross-reaction with the small amount of EGF in NIH grow in an as determined receptor present the 3T3 cells. Since the p52 and anchorage-independent manner, in Methocel. Clone 42 showed forms of SHC by their ability to grow 1.2% p46 migrate at the same position as of than clone which total a higher basal level transformation immunoglobulin heavy chain, precipitates could not 18, formed colonies in the of EGF readily be only presence analysed by immunoblotting due to the cross- (Table I). Clones 18 and 42 formed colonies in the reaction of larger presence immunoglobulin heavy chain with the second of 30 nM EGF as with 3 nM of the immunoblot. To compared EGF (Figure 5). layer avoid this cross-reaction, For NEN7 and mock-transfected NIH 3T3 cells EGFR/c-erbB-3 chimera comparison, protein was eluted from the were also cultured in 1.2% Methocel. some small antibody -protein A Although -Sepharose complex using a solution colonies were observed for mock-transfected NIH 3T3 of peptide which cells, against antibody 49.3 was raised. Eluates these did not increase in number or size in to EGF. were analysed by SDS -PAGE and immunoblotting using response In the case of NEN7 cells, colonies were only observed in a rabbit polyclonal antibody recognizing the three forms of the of EGF as described presence previously (Lehvaslaiho SHC (Figure 6A). For comparison, total lysate from - et The effect of EGF cells was al., 1989). growth-promoting on clones analysed similarly (Figure 6B). The three forms 18 and 42 was confirmed of SHC and by [3H]thymidine incorporation (p66, p52 p46) associated specifically with studies. Using cells growing attached as EGF activated c-erbB-3, monolayers, although the molecular weights of p46 stimulation resulted in a 3.9-fold increase in and p52 appeared to be slightly higher on thymidine this gel system. at 0.5-5.0 nM incorporation EGF relative to the The predominant form present was p52, which also appeared unstimulated cells. NIH 3T3 maximal cells showed a 2.7-fold to be the most abundant form present in total cell lysates. stimulation in to EGF at the response same concentrations No SHC protein was present in precipitates from the parental not (data shown). NIH 3T3 cell line, even when using the EGFR1 antibody 2834 chimera receptor/c-erbB-3 EGF LU U X Lu CO _+ HI HV cccc a LU C) LU LU L IL + + * C + I C) COo cO co - IT" X . cn _ CO) CO kDa kDa -30 -69 -^ Fig. 7. Association of tyrosine-phosphorylated proteins with activated -_ a. c-erbB-3. Immunoprecipitations were performed on lysates from EGF- *Mw * 46 stimulated cells exactly as described in Figure 6A. Eluted proteins were analysed by SDS-PAGE and immunoblotting with an anti- phosphotyrosine antibody. -30 Fig. 6. Association of activated c-erbB-3 with SHC. (A) NIH 3T3 cells and transfected NIH 3T3 cells expressing I05 EGF receptor/c- erbB-3 chimeric receptors per cell (clone 18) were incubated for 2 min Cells were lysed and EGFR/c-erbB-3 protein was with 10-6 M EGF. with an antibody specific for c-erbB-3 (49.3). Precipitated precipitated proteins were eluted with the corresponding 49.3 peptide against which the was raised, and eluates were analysed by SDS-PAGE antibody and immunoblotting using an antibody recognizing SHC. (B) For 4-.-PIp total cell lysate from 10' cells was also analysed. comparison, LU0 LLU for the extracellular domain of EGF receptor (data specific * + not shown), and only a very small amount of the p52 form was found associated with EGFR/c-erbB-3 in the basal state. This represents association with a small proportion probably kDa of which is phosphorylated in the EGFR/c-erbB-3 unstimulated state. Association of EGFR/c-erbB-3 with tyrosine proteins phosphorylated Immunoblots which had been analysed for the presence of as described above were stripped and reprobed with -46 SHC an antibody (Figure 7). The major anti-phosphotyrosine comigrated exactly with the p52 form phosphorylated protein -30 and bands corresponding to p46 and p66 were also of SHC, 0 LUI 0 LUJ _ L. LUI Some other proteins of apparent molecular + detectable. ILI weights 85, 110 and 125 kDa were detectable. co OD Cl) Cl) 8. Association of PI3-kinase with activated c-erbB-3. Fig. Association of EGFR/c-erbB-3 with P13-kinase. were on lysates from EGF- Immunoprecipitations performed (A) from EGF-stimulated cells Immunoprecipitates prepared as described in 6A. Eluted proteins stimulated cells exactly Figure were for the association of antibody 49.3 analysed using SDS-PAGE and immunoblotting with a rabbit were analysed by with an antibody specifically the p85 subunit of P13-kinase. (B) P13-kinase by immunoblotting polyclonal antibody recognizing were on lysates from EGF-stimulated Immunoprecipitations performed the subunit of the enzyme. As was found reacting with p85 as described in 5A. Precipitated proteins bound to cells exactly Figure interacts with the activated EGFR/c- in the case of SHC, p85 with and as were incubated PI [-y-32P]ATP A-Sepharose protein and to a reduced extent with the erbB-3 receptor, greatly and TLC. A PI phosphate standard (PIP) described analysed by from unstimulated cells (Figure 8A). p85 P13-kinase at chimeric receptor PI with purified migrated prepared by phosphorylating the indicated. associated with EGF receptor precipitated position was also found 2835 EGF receptor/c-erbB-3 chimera from EGF-stimulated EGFR-3T3 which an PI4-kinase activity, inhibited the formation of this cells, express phospholipid (data not shown). equivalent number of EGF receptors, using antibody the was several-fold less than that EGFR1; however, signal obtained with the activated EGF Inhibition of p85 and SHC association receptor/c-erbB-3 chimera, using tyrosine and was comparable with that seen in unstimulated cells phosphopeptides the EGF chimera not To determine the c-erbB-3 binding sites for SHC and expressing receptor/c-erbB-3 (data p85, shown). This result was confirmed for PI3-kinase we looked at the ability of tyrosine phosphopeptides by assay in the A corresponding to all potential sites within activity immunoprecipitates. phosphorylated lipid phosphorylation (Rf: a 32P-labelled the c-erbB-3 C-terminal tail to inhibit formation 0.57), which comigrated with complex standard Dr in permeabilized NIH 3T3 cells the EGFR/c- phosphatidylinositol 3-phosphate (provided by expressing was when erbB-3 chimera All R.Woschowlski, ICRF, London), produced (Table II). tyrosine phosphopeptides EGFRIc-erbB-3-3T3 the motif precipitates from EGF-stimulated cells containing p85 binding pYXXM (la, 2a, 3a, 4, 6 and inhibited association with whereas were incubated with phosphatidylinositol and 7) p85 c-erbB-3, ['y-32P]ATP in the reaction buffer of Triton other had no effect (Figure 8B). Inclusion 1% peptides inhibitory (Figure 9B). Complete inhibition was only 4 X-100, which inhibits PI3-kinase activity and favours obtained with peptides and 7. Peptide 5 to was which corresponding Y1309 the only peptide inhibited SHC association The lane (Figure 9C). Table H. Sequences of c-erbB-3 phosphopeptides to 8 showed a reduced corresponding peptide slightly signal for however this is attributable to reduced of SHC; loading la (Y1241, Y1243) TTPDEDpYEpYMNRQR EGFR/c-erbB-3 in this case (Figure 9A). lb (Y1243) TTPDEDYEpYMNRQR To further assess of in the importance residues peptide 2a (Y1178, Y1180) DEDEEpYEpYMNRRRR 5 for SHC binding, two additional peptides were synthesized 2b (Y1180) DEDEEYEpYMNRRRR 11 and in which either the 3a LEELGpYEpYMDVGSD (peptides 12) asparagine residue (Y1203, Y1205) 3b (Y1205) LEELGYEpYMDVGSD (position -3 relative to the phosphotyrosine) or proline 4 (Y1035) SPSSGpYMPMNQGNL -2 residue (position relative to the phosphotyrosine) was 5 (Y1309) DSAFDNPDpYWHSRLF alanine changed to (Table II). Neither of these peptides was 6 (Y1257) GPGGDpYAAMGACPA to inhibit of able association SHC with c-erbB-3 under the 7 (Y1270) ASEQGpYEEMRAFQG conditions tested (data not shown). 8 EDVNGpYVMPDTHLK (Y1140) 9 (Y1288) APHVHpYARLKTLRS Comparison between EGF receptor, EGFR/c-erbB-2 10 RGDSApYHSQRHSLL (Y1113) and EGFR/c-erbB-3 with to to a respect binding lla (Y1309) AFDNADpYWHS GRB2 - GST fusion protein 12a (Y1309) AFDAPDpYWHS EGFR/c-erbB-3 from protein EGF-stimulated EGFRIc- aData are not shown for these X erbB-3-3T3 cells failed to associate with GRB2-GST fusion peptides. kDa PEP - - la lb 2a 2b 3o 3b 4 5 6 7 8 9 10 Lys EGF - + + + + + + + + 1 ~~~~~-20 c-erbB-3 -116 6 7 8 9 10 PEP - - la lb 2a 2b 3a 3b 4 5 Lys EGF - + +- + + + + + -116 )85 M+ + + + + + - -80 PEP -- 1a lb 2a 2b 3a 3b 4 5 6 7 8 9 10 Lys EGF - + + + + + + + C -80 SIIC 49.5 Fig. 9. Inhibition of p85 and SHC association with c-erbB-3 using synthetic phosphopeptides. NIH 3T3 cells expressing the EGF receptor/c-erbB-3 chimera were permeabilized in the presence of 70 x phosphopeptides as indicated. Cells were stimulated for 4 with EGF (5 M), protein min 10-7 j&M from cell was lysates precipitated with the 49.3 antibody, eluted with peptide 49.3 and analysed by SDS-PAGE and immunoblotting with antibodies for c-erbB-3 and SHC specific (A), p85 (B) (C). Peptide sequences are as indicated in Table II. 2836 chimera EGF receptor/c-erbB-3 A B 21 N 12E 1 2E 49.:3 49.3 coo rri tN EGFR-3T3 3T3 clone 18 3T3 I LL LLw - -J _N _ z z L z U) 2 + _ + _ + _ + _ + -+ - protein protein gg 49g Ewd, 97.4 Fig. 10. Association of EGFR/c-erbB-3 with a GRB2-GST fusion protein. (A) Cell lysates from EGF-stimulated (+) and unstimulated (-) cells (NEN7, EGFR-3T3 and EGFR/c-erbB-3-3T3; clone 18) were incubated with a GRB2-GST fusion protein immobilized on glutathione-agarose. Bound proteins were analysed by SDS-PAGE and immunoblotting with antibodies 21N, 12E and 49.3, which specifically recognize c-erbB-2, EGF receptor and c-erbB-3, respectively. (B) Cell lysates were prepared from EGFR-3T3, clone 18 and NIH 3T3 cells. Antibodies 12 E and 49.3 were used for immunoblotting of varying quantities of total cellular protein, as indicated. in c-erbB-3 et Plowman et protein immobilized on glutathione-agarose as determined different (Kraus al., 1989; al., structure of cAMP- by immunoblotting with polyclonal antibody 49.3 (Figure 1990). Evidence from the crystal kinase that the substituted lOA). For comparison, EGF receptor protein and EGFR/c- dependent protein (cAPK) suggests in c-erbB-3 to Aspl66 of cAPK) erbB-2 chimera protein were precipitated from EGF-treated aspartate (corresponding and in fact EGFR-3T3 and NEN7 cells, respectively, with the normally forms part of the catalytic loop may et The immobilized GRB2-GST fusion protein, and in these cases function as the catalytic base (Knighton al., 1991). residue which is absent in a prominent band was observed by immunoblotting with carbonyl group of the glutamate in in to the antibodies 12E and 21N, respectively. No immunoreactive c-erbB-3 (Glu9l cAPK) is close proximity In in these protein could be precipitated from EGF-stimulated NIH 3T3 MgATP binding site. the light of the differences of interest to determine cells, or from unstimulated EGFR-3T3 cells, although a apparently critical residues, it was activity. small amount of EGFR/c-erbB-2 protein was precipitated whether or not c-erbB-3 possesses catalytic indicate that c-erbB-3 a low from unstimulated NEN7 cells suggesting a somewhat Our findings possesses very in the unstimulated state, and elevated basal level of phosphorylation of this protein in these level of autocatalytic activity with and with cells. The sensitivity of antibodies 12E and 49.3 was that by replacing Asn815 aspartate, His740 no increase in is observed. By compared by immunoblotting of lysates prepared from glutamate, catalytic activity with a EGFR-3T3 and EGFR/c-erbB-3-3T3 cells (clone 18). Both replacing the extracellular domain ligand-activatable one detect cell lines express - 100 000 receptors per cell, and when EGF receptor domain, can, however, clearly of the chimeric receptor upon EGF equivalent amounts of total cell protein were analysed, phosphorylation This has been used to similar signals were obtained by immunoblotting (Figure stimulation. approach previously for which no were available LOB). characterize receptors ligands et Lehvaslaiho et Seedorf et al., (Riedel al., 1984; al., 1989; It is that this phosphorylation could be due 1991). unlikely as NIH 3T3 to EGF receptor cross-phosphorylation by Discussion number of EGF fibroblasts express a very low (-3000) Fiore et and EGF- has been for that all tyrosine receptors per cell (Di al., 1987a), It recognized many years of EGF certain conserved the most evident stimulated autophosphorylation endogenous receptor kinases possess features, motif in subdomain cells could not be detected. It of these a in NIH 3T3 is, however, being Gly-X-Gly-X-X-Gly II et that a could be attributed to I and a residue in subdomain (Hanks al., 1988). possible proportion trans-phos- lysine These observations and kinases EGF In both serine/threonine tyrosine phorylation by receptor. addition, residue found to be essential invariant or almost invariant that the contain a number of residues, demonstrate aspartate kinases et Tan residue in subdomain III and an for the function of other which include a glutamate (Moran al., 1988; to an in c-erbB-3 both of which are et be altered residue in subdomain al., 1990) may asparagine aspartate VI, 2837 S.A.Prigent and W.J.Gullick without inhibiting autocatalytic activity. It would be of some c-erbB-3 DSAFDNPDYWHSRLF interest to investigate the effect of the His/Glu and Trk IENPQYFSDA Asn/Asp mutations on the of the activatable EGFR/c- Middle T PSLLSNPTYSVMRSH catalytic activity erbB-3 chimera. Since the site of c-erbB-3 be catalytic may Fig. 11. Comparison of for on Trk binding sites SHC c-erbB-3, and different from other it be structurally kinases, may possible Middle T antigen. The SHC binding site on Trk was determined using kinase inhibitors more selective for this to design tyrosine the peptide indicated (Obermeier et al., 1993). Tyrosine 250 of Middle than for other kinases. receptor receptor tyrosine T was identified using site-directed mutagenesis and is contained within the sequence indicated (Dilworth et al., 1994). are in the Tyrosine kinase receptors frequently implicated cancers Both EGF growth of human (Aaronson, 1991). and c-erbB-2 are which can transform SHC are when receptor oncogenes gene products transforming overexpressed Fiore et NIH 3T3 fibroblasts when expressed at high levels (Di in fibroblasts (Pelicci al., 1992). et Both of these Whatever the exact the fact that SHC is the et al., 1987b; Hudziak al., 1987). growth mechanism, in a of human substrate for a number of both factor receptors are overexpressed variety major physiological receptor and in some cases are associated with and kinases to an cancers, poor non-receptor tyrosine points important role Lofts and We have in transduction. In this have shown prognosis (Gullick, 1991; Gullick, 1991). signal study we that SHC that c-erbB-3 is in some associates with and is c-erbB-3. shown, likewise, overexpressed phosphorylated by activated et cancers (Lemoine et al., 1992a,b; Rajkumar al., 1993). Moreover, we have identified a binding site for SHC In of to NIH 3T3 cells on this study, addition EGF expressing (Y1309) the c-erbB-3 protein. It would be of interest to to EGF receptor/c-erbB-3 chimeric receptors enabled them examine the mitogenic activity of mutant c-erbB-3 lacking which grow in an anchorage-independent manner is a this site. When the sequence surrounding this tyrosine residue property characteristic of, but not restricted to, transformed is compared with sequences reported recently to bind SHC of Kraus cells. These findings are in agreement with those on the Trk receptor tyrosine kinase and Middle T antigen, et al. (1993), who used a similar approach to demonstrate it is apparent that they all contain the motif NPXY which of the growth-promoting properties c-erbB-3. is likely to represent the consensus binding site for SHC In Since c-erbB-3 was able to with the (Figure 11). this study, peptides lacking either Asn(-3) couple signalling elements in NIH 3T3 cells to promote a growth response, or Pro(-2) were unable to inhibit the association between in some of the we were interested identifying proteins SHC and c-erbB-3, suggesting that both residues are identification of src domains has involved. The homology important for SHC binding. This is the first SH2-containing in of the facilitated rapid advances our understanding protein which appears to require the strict conservation of which kinases and residues mechanism by tyrosine signal (Pawson N-terminal to the phosphotyrosine residue for are now Schlessinger, 1993). Signalling pathways being specificity. In defined by interactions of SH2 domains with distinct contrast to EGF receptor and c-erbB-2, we could not of In phosphotyrosine residues within defined consensus detect binding c-erbB-3 to GRB2. this respect, c-erbB-3 at of the shows Trk sequences. It now seems possible that least some similarity to receptors which phosphorylate part SHC but do bind network of events following receptor activation might be not GRB2 (Suen et al., 1993). We conclude amino acid of the that the two predicted from the primary sequence potential binding sites predicted by Songyang to A et al. are not receptor. Most receptors appear signal through p21ra(. (1993) (i) autophosphorylated, (ii) do not bind that GRB2 links the activated or are number of reports have shown GRB2, (iii) they occupied by another protein. Clearly factor of it is possible that c-erbB-3 employs the EGF receptor to the nucleotide exchange p21ras, pathway thought to it close to the membrane. This is be used in that of sos, thereby localizing by v-Src, phosphorylation SHC by c- achieved interaction of the domain of GRB2 with erbB-3 of by SH2 promotes binding SHC to the GRB2 -sos complex of and the domain with a and of Y1068 the EGF receptor, SH3 subsequent activation p21ras. Indeed, we could motif of and precipitate a small amount of SHC from proline-rich sos (Buday Downward, 1993; EGF-stimulated 3T3 Chardin et et Gale et cells expressing the EGF al., 1993; Egan al., 1993; al., 1993; receptor/c-erbB-3 chimera using Li et al., 1993; Olivier et al., 1993; Rozakis-Adcock et the GRB2-GST fusion bound to al., protein glutathione-agarose 1993; Simon et not The al., 1993). (data shown). amount of SHC precipitated from An additional has been shown to be these cells with GRB2-GST was less than that observed protein, SHC, phosphorylated in response to EGF and to associate with for EGF-stimulated EGFR-3T3 cells, but significantly more GRB2. However, it has been shown that SHC does not form than that precipitated from NIH 3T3 cells. We have not yet of part the complex of sos, EGF receptor and GRB2 investigated whether Ras is activated in these cells in (Buday and the Downward, 1993), suggesting that formed response to EGF. complex between GRB2 and in the case of EGF SHC, receptor, forms In addition to SHC association with activated c-erbB-3, of a different which or part signalling pathway may may we also observed association of the p85 subunit of not converge with the p2lras signalling pathway as it is PI3-kinase. Like SHC and GRB2, p85 has no catalytic currently perceived. Some evidence that SHC may be domain and acts as an adaptor molecule which links the involved in p21ral signalling is provided by the observation activated tyrosine kinase with the catalytic p110 subunit of that the differentiation effect of SHC on PC12 cells is P13-kinase. p85 possesses two SH2 domains and one SH3 prevented by expression in these cells of a dominant negative domain. The presence of the p1O subunit in the immune Ras mutant (Pelicci et al., 1993). SHC is phosphorylated complexes from EGF-stimulated cells was confirmed by an in v-Src transformed cells (McGlade et al., 1992), resulting assay for P13-kinase activity. A protein comigrating with p85 in its binding to the GRB2 -sos complex. This association was phosphorylated in response to EGF, but to a lesser extent is proposed to stimulate guanine-nucleotide exchange on than SHC. Phosphorylation of p85 has only been observed Ras through an unknown mechanism. In contrast to GRB2, under conditions where the tyrosine kinase is overexpressed, 2838 receptor/c-erbB-3 chimera EGF and it seems that it is not essential for function (Hu et al., in that c-erbB-3 fails to interact with GRB2 in contrast to EGF receptor and c-erbB-2, whereas P13-kinase appears to 1992). The role of P13-kinase in cellular transformation was first suggested by Kaplan et al. (1985). PI3-kinase has been interact more strongly with c-erbB-3 than EGF receptor. This subsequently to interact with a number of growth suggests a mechanism for achieving signalling diversity since shown heterodimerization between related family members, which factor receptors, including PDGF receptor, CSF-1 receptor has been shown to occur between EGF receptor and c-erbB-2 and the insulin receptor substrate IRS1. PI3-kinase also (Goldman et al., 1990), could potentially activate a distinct associates to some extent with EGF receptor, although it has network of events. It will be interesting to investigate the been shown that p85 alone binds more avidly to EGF role of cross-talk between c-erbB-3 and other I growth receptor than the p85 -pl 0 active complex (Hu et al., type factor receptors. 1992). Relatively low levels of PI3-kinase activity have been found associated with activated EGF receptor as compared with PDGF receptor, suggesting that P13-kinase plays a Materials and methods minor role in EGF receptor signal transduction. The DNA constructs and vectors consensus binding site for p85 to phosphorylated tyrosine The cDNA clone for c-erbB-3 contained in the vector pCDM8 (Invitrogen) kinases is now well defined as the sequence YMXM or was a gift from Dr G.Plowman, Bristol-Myers Squibb Pharmaceutical possibly YXXM. In contrast to EGF receptor, c-erbB-3 Research Institute, Seattle, WA. This was used for transient expression in contains six potential PI3-kinase binding sites in its C- COS-1 cells. The EGFR cDNA was a gift from Dr G.Gill, UCSD, CA. terminal domain, and one in its kinase domain. Of these, For construction of the EGFR/c-erbB-3 chimera, two restriction fragments encoding the extracellular and transmembrane domains of EGF receptor only one is conserved in EGF receptor, c-erbB-2, and c- encoding amino acids 1-154; AflIL-NarI, encoding amino (NotI -Aflu, erbB-4 which is within the kinase domain and is not thought acids 155-647) and a restriction fragment encoding the intracellular portion to be phosphorylated in EGFR. We have shown that synthetic of c-erbB-3 (NarI-XbaI, encoding amino acids 651-1323) were inserted peptides corresponding to these sites (Y1035, Y1 178, into the vector pRc/CMV for evaluation by transient transfection in COS-1 Y1203, Y1241, Y1257 and Y1270) inhibited association of cells. The chimera construct was then subcloned into an LTR-based vector pLTRpoly (Lehvaslaiho et al., 1990) (provided by Dr K.Alitalo, Helsinki, p85 with c-erbB-3, suggesting that they could indeed act as Finland) for transfection into NIH 3T3 fibroblasts. docking sites for p85 if phosphorylated in vivo. These data suggest that PI3-kinase may play a more important role in Cell culture and transfections the direct signalling of c-erbB-3 than other family members. COS-1 (Gluzman, 1981), NIH 3T3 (Jainchill et al., 1969) and A431 (Giard et 1973) cells were obtained from the American Tissue Culture Collection al., The possibility clearly exists, however, for cross-talk via and NIH 3T3 cells transfected with an EGFR/c-erbB-2 chimera (NEN-7 heterodimerization with other type I family members, since cells; Lehvaslaiho et al., 1989) were a gift from Dr K.Alitalo, University we have shown that EGF receptor is able to phosphorylate of Helsinki, Finland. NIH 3T3 cells expressing - 105 EGF receptors per c-erbB-3 in vivo, at least under conditions where they are cell (EGFR-3T3) were kindly provided by Dr D.Hills, ICRF Oncology co-expressed at similar high levels. Other differences in the Group, Hammersmith Hospital, UK. Cells were grown in Dulbecco's modified Eagle medium (DMEM) (Life Technologies, UK) containing 10% signalling pathways of the type I family of growth factor fetal calf serum. 10 of plasmid with 1 yg of selection plasmid pRc/CMV yig receptors may exist, resulting in signalling diversity. (Invitrogen) were used to transfect 9 cm plates of NIH 3T3 fibroblasts using Whereas both EGF receptor and c-erbB-2 use phospholipase the calcium phosphate method (Graham and van der Eb, 1973). For selection as a substrate, c-erbB-3 contains no potential of transfected cells, 2.5 mg/ml G418 (Gibco) was included in the growth C'y medium. Nine centimetre plates of COS-1 cells were transfected with 10 phospholipase C^y binding site (Songyang et al., 1993). of DNA by the DEAE/Dextran method (Selden, 1992). jug Indeed, we were unable to detect binding of to c- PLCy erbB-3 by immunoblotting (data not shown). Selection of NIH 3T3 chimera clones It has been shown for PDGF receptor that if the tyrosine Transfected NIH 3T3 fibroblasts were labelled with the EGFR1 antibody (10 (Waterfield et al., 1982) and fluorescein-conjugated F(ab')2 residues within the p85 binding motif are mutated to prevent Ag/ml) rabbit anti-mouse immunoglobulins (1:20 dilution) (DAKOPATTS) and P13-kinase interaction, the ability of the PDGF receptor to subjected to two rounds of FACS sorting using a Becton Dickinson stimulate Ras activity is lost, as is its ability to activate raf FACStarPLUS sorter. At each sort, 5% of the most strongly fluorescent cells and MAP kinase and to stimulate DNA synthesis (Fantl were selected. Cells were then single-cell cloned and analysed for their ability et al., 1992). Others have shown that if the tyrosine residues 125I-labelled EGF (Amersham, UK) as described previously to bind (Aharonov et al., 1978). which bind PI3-kinase (740/751), PLCy (1021), GAP (771) and SH-PTP2 (1009) are all mutated, PDGF is unable to Site-directed mutagenesis DNA synthesis. When binding of activate Ras or stimulate A SacI fragment of c-erbB-3 (site in vector to base 3522) was subcloned was restored, Ras activation and either PI3-kinase or PLCGy mutagenesis was performed using the into M13mpl9. Site-directed mutagenesis kit using the oligonucleotides and Kazlauskas, 1993). Amersham oligonucleotide-directed mitogenesis were normal (Valius the Asn to Asp mutation) and (for CGGGCAGCCAGGTCTCTATGCAC that the GRB2 - sos complex is not These results suggest (for the His to Glu mutation) CTGCCAATGGCCAGCATTTCATCTG for Ras. the only mechanism stimulating The mutant clones were sequenced and the together in the same reaction. of transduction cannot simply Clearly the processes signal fragment (bases 2382-3000) was replaced in the mutated BamHI-Bgll the interaction of receptors with GRB2. full-length c-erbB-3 cDNA. be explained by an important role in cell signalling, although P13-kinase plays Immune-complex kinase assays for the products the precise targets phosphatidylinositol transfected with c-erbB-3 or mutant 9 cm plates of COS-1 cells transiently elusive. The role on the D3 remain phosphorylated position His/Glu740 and Asn/Asp815 mutations were lysed c-erbB-3 containing the in transduction is even less well defined, but with polyclonal antibody 49.3 of SHC signal and c-erbB-3 protein was immunoprecipitated et 1992). Immunoprecipitates were for kinases as described previously (Prigent al., use as a primary target tyrosine its extensive twice with phosphorylation buffer (50 mM HEPES pH 7.4, 5% washed for this We have shown a vital function protein. suggests Triton 150 mM 2 mM 12 mM 0.2% NaCl, X-100, MnCl2, glycerol, are for c-erbB-3. both SHC ard P13-kinase targets that sodium and 49.3 et al., 100 orthovanadate) (Prigent peptide MgCl2, 1tM distinct differences in the we have identified of 2.5 and incubated at Moreover, to a concentration was added mg/ml 37°C 1992) 10 tCi related c-erbB-3 from the which interact with closely tyrosine kinases, for 10 to elute [_y-32P]ATP proteins min antibody. 2839 and W.J.Gullick S.A.Prigent (Amersham, UK) was added for 10 min at room temperature (100 final as described above. Precipitates were then washed twice with 20mM HEPES ltM concentration, 1 Ci/mmol). Precipitates were boiled in sample buffer and pH 7.2, 5 mM MnCl2, 0.13 mM Na orthovanadate and 12.5 mM NaF. analysed by SDS-PAGE (Schagger and von Jagow, 1987). Proteins were Assays were performed in 100 of the same buffer containing 0.2 mg/ml A1 transferred to Immobilon-P PVDF membranes (Millipore) and subjected phosphatidylinositol, 10 and 50 ATP for 30 min at [-y-32P]ATP ACi ItM Bands of interest were excised and analysed for 37°C. Lipids were extracted as described previously (Morgan et al., to autoradiography. 1990), phosphoamino acid content as described previously (Kamps and Sefton, and analysed by TLC using the borate system (Walsh et For al., 1991). To compare the level of expression of the wild-type and mutant inhibition of P13-kinase, 1% Triton X-100 was included in the reaction buffer. 1989). proteins, transfected cells were metabolically labelled in parallel with proteins were immunoprecipitated with antibody Inhibition of p85 and SHC binding to using [35S]methionine, 49.3, EGFRIc-erbB-3 and electrophoresed on SDS-polyacrylamide gels which phosphopeptides eluted with peptide were treated with Amplify (Amersham) for fluorography, as described Synthetic peptides, as indicated in Table II, were synthesized by Nicola et al., 1992). O'Reilly and Elizabeth Li, Peptide Synthesis Unit, ICRF, London, UK. previously (Prigent EGFR/c-erbB-3-3T3 cells were permeabilized in the presence of 70 AM EGF stimulation of whole cells peptide as described previously (Buday and Downward, 1993), and treated in six-well dishes were washed and routinely incubated with EGF (5 x 10-7 M) for 4 min at room temperature. EGFR/c-erbB-3 Cell monolayers DMEM in the presence or absence of synthetic protein was precipitated with 49.3 as described above, but in the presence for 2 min at 37°C with M), supplied by Dr R.Goodlad, ICRF, London, UK. The of 70 peptide. Bound proteins were eluted with 49.3 peptide (2.5 mg/ml) EGF 1-48 (10-6 AM and cells were lysed in lysis buffer containing and analysed by SDS-PAGE and immunoblotting for c-erbB-3, p85 and medium was removed pH 7.4, 1 Triton X-100, 5mM EGTA, SHC, as described above. phosphatase inhibitors (50mM Tris % 150 mM NaCl, 25 mM benzamidine, 2 mM PMSF, 1 mM sodium Formation of receptor - GRB2 complexes in vitro orthovanadate, 100mM NaF, 10mM sodium pyrophosphate). The protein 6 of GRB2-GST fusion protein prepared as described concentration of cell lysates was determined by the Bradford protein assay previously (Egan Ag were bound to 35 of a of (Bio-Rad Laboratories Ltd, Defined amounts of protein were analysed et al., 1993) suspension glutathione agarose UK). A1 (Pharmacia, UK) for 30 min at room temperature. Lysate was by SDS-PAGE and immunoblotting with a rabbit polyclonal anti- prepared from 175 cm2 monolayers of cells [NIH 3T3, NEN7, EGFR-3T3 and phosphotyrosine antibody provided by Dr T.S.Pillay, Royal Postgraduate EGFR/c-erbB-3-3T3 (clone 18 as described in Results)], which had been Medical School, London, UK (Pillay and Makgoba, 1992). To determine treated with EGF (10-6 M) for 4 min, and from untreated cells. Cell the time course for activation of c-erbB-3, cells were incubated in six-well lysate was incubated for 2 h at 4°C with GRB2-GST-glutathione agarose over a period of up to 45 min with EGF (10-6 M). Cells were lysed plates complexes, precip-tates were washed extensively with TBS/0.1 in boiling sample buffer at the indicated times to terminate phosphorylation % Triton X-100 and bound proteins were eluted in reactions rapidly. Samples were subjected to SDS-PAGE and sample buffer and analysed by SDS-PAGE and immunoblotting. Rabbit polyclonal antibodies 12 E immunoblotting with anti-phosphotyrosine antibody. (Gullick et al., 1985), 21 N (Gullick et al., 1987) and 49.3 (Prigent et al., 1992) were used for detection of EGF receptor, EGF receptor/c-erbB-2 Determination of anchorage-independent growth chimera protein, and EGF receptor/c-erbB-3 chimera protein, respectively. Cells were seeded in methyl cellulose (Sigma) suspension (Methocel) as The sensitivity of 49.3 and 12 E were also compared by probing described previously (Lemoine et al., 1989). 35 mm-diameter wells of six- antibodies immunoblots of whole cell lysate from EGFR-3T3 and EGFR/c-erbB-3-3T3 well plates were coated with DMEM containing 0.9% agarose and duplicate cells (clone 18). wells were seeded with 30 000 cells suspended in DMEM containing 1.2% Methocel and 10% fetal calf serum, in the presence or absence of EGF or 30 nM). Colonies >0.1 mm were scored after 2 weeks. (3 Acknowledgements incorporation [3H]Thymidine We are grateful to Dr G.Plowman for supplying the c-erbB-3 clone, to Dr 96-well plates were seeded with 1000 cells/well and grown to 80% G.Gill for supplying the EGFR clone, to Dr K.Alitalo for supplying the confluence. The medium was replaced with DMEM containing 0.5% FCS NEN-7 cells, to Dr D.Hills for supplying the EGFR-3T3 cells, to Drs nM and cells were incubated for 48 h. EGF was added (0.05-50.0 final L.Cantley and S.Soltoff for providing the and to Dr anti-p85 antibody and cells were incubated for a further 16 h before addition concentration) T.S.Pillay for providing the antibody. We wish to thank anti-phosphotyrosine repeated of [3H]thymidine (0.2 PCi/well) for 4 h. Cells were ruptured by Dr S.Egan for providing the pGEX vector containing the GRB2-GST firezing and thawing and analysed on an LKB Wallac 1295-001 cell harvester sequence. We are also grateful to Dereck Davis and Simon Monard for and 1205 Betaplate liquid scintillation counter (LKB Pharmacia, UK). performiing the FACS sorting of NIH 3T3 transfectants, and to Nicola O'Reilly and Elizabeth Li for synthesizing the tyrosine phosphopeptides. Immunoprecipitation and immunoblotting We also wish to dtank Dr J.Feramisco for allowing some of the experiments Following EGF stimulation, cells were lysed in 50 mM Tris pH 7.4, 1% to be performed in his laboratory. Triton X-100, 5 mM EGTA, 150mM NaCl, 25 mM benzamidine, 2 mM mM 1 mM Na orthovanadate. EGF receptor/c-erbB-3 PMSF, 10 NaF and was for 2 h with 5 ug of purified rabbit chimeric protein precipitated References antibody 49.3 (Prigent et al., 1992) which reacts specifically polyclonal with the intracellular domain of c-erbB-3, or EGFR1 which reacts with the Aaronson,S.A. (1991) Science, 254, 1146-1152. extracellular domain of EGF receptor (Waterfield et al., 1982) and 2.5 mg Aharonov,A., Pruss,R.M. and Herschman,H.R. (1978) J. Bio. Chem., of Immunoprecipitates were washed extensively with protein A-Sepharose. 253, 3970-3977. TBS (50mM TrispH 7.4 and 150mM NaCl) containing 0.1% Triton X-100. Bar-Sagi,D., Rotin,D., Batzer,A., Mandiyan,V. and Schlessinger,J. (1993) EGFR/c-erbB-3 protein was then eluted from the protein A-Sepharose at Cell, 74, 83-91. 37°C for 15 min with 40 sl of a 5 mg/ml solution of the 49.3 peptide. Buday,L. and Downward,J. (1993) Cell, 73, 611-620. Eluates were analysed on 7.5% SDS-polyacrylamide gels as above. Proteins Chardin,P., Camonis,J.H., Gale,N.W., Van Aelst,L., Schlessinger,J., were transferred to nitrocellulose and probed with rabbit antibodies specific Wigler,M.H. and Bar-Sagi,D. (1993) Science, 260, 1338-1343. for c-erbB-3 (49.3), for the p85 subunit of PI3-kinase (provided by Drs Clark,S.G., Stern,M.J. and Horvitz,H.R. (1992) Nature, 356, 340-344. L.Cantley and S.Soltoff, Tufts University School of Medicine, Boston, MA, Di Fiore,P.P., Pierce,J.H., Fleming,T.P., Hazan,R., Ullrich,A., King,C.R., and commercially available from UBI, New York), antibodies specific for Schlessinger,J. and Aaronson,S.A. (1987a) Cell, 51, 1063-1070. SHC (UBI, New York) or antibodies reacting with phosphotyrosine (provided Di Fiore,P.P., Pierce,J.H., Kraus,M.H., Segatto,O., King,C.R. and by Dr T.S.Pillay, RPMS, UK). Immunoblots were developed using the Aaronson,S.A. Science, 237, 178-182. (1987b) ECL system, (Amersham, UK). When probing with more than one antibody Brewster,C.E.P., Jones,M.D., Pelicci,G. Dilworth,S.M., Lanfirancone,L., was required, blots were routinely stripped for 1 h at 60°C with 2% SDS, and Pelicci,P.G. (1994) Nature, 367, 87-90. 100mM in 62.5 mM Tris, pH 6.8, blocked and reprobed. S-mercapoethanol Egan,S.E., Giddings,B.W., Brooks,M.W., Buday,L., Sizeland,A.M. and Weinberg,R.A. (1993) Nature, 363, 45-51. P13-kinase assays Escobedo,J.A., Navankasattusas,S., Kavanaugh,W.M., Milfay,D., P13-kinase were as Fried,V.A. and Williams,L.T. (1991) Cell, 65, 75-82. described previously assays performed essentially et EGFR/c-erbB-3 was Fantl,W.J., Escobedo,J.A., Martin,G.A., Turck,C.W., del Rosario,M., al., 1990). protein precipitated (Morgan Briefly, from EGF-treated or untreated NIH 3T3 cells or EGFR/c-erbB-3-3T3 cells McCormick,F. and Williams,L.T. (1992) Cell, 69, 413-423. 2840 chimera EGF receptor/c-erbB-3 Feng,G.-S., Hui,C.-C. and Pawson,T. (1993) Science, 259, 1607-1611. Rozakis-Adcock,M., Femnley,R., Wade,J., Pawson,T. and Bowtell,D. Freeman,R.M.,Jr, Pluzky,J. and Neel,B.G. (1992) Proc. NatlAcad. Sci. (1993) Nature, 363, 83-85. USA, 89, 11239-11243. Schagger,H. and von Jagow,G. (1987) Anal. Biochem., 166, 368-379. Gale,N.W., Kaplan,S., Lowenstein,E.J., Schlessinger,J. and Bar-Sagi,D. Seedorf,K., Felder,S., Millauer,B., Schlessinger,J. and Ullrich,A. (1991) (1993) Nature, 363, 88-92. J. Biol. Chem., 266, 12424-12431. Giard,D.J., Aaronson,S.A., Todaro,G.J., Arnstein,P., Kersey,J.H., Segatto,O., Pelicci,G., Giuli,S., Digiesi,G., Di Fiore,P.P., McGlade,J., Dosik,H. and Parks,W.P. (1973) J. Natl Cancer Inst., 51, 1417-1423. Pawson,T. and Pelicci,P.G. (1993) Oncogene, 8, 2105-2112. Gluzman,Y. (1981) Cell, 23, 175-182. Selden,R.F. (1992) In Ausubell,F.M., Brent,R., Kingston,R.E., Goldman,R., Ben Levy,R., Peles,E. and Yarden,Y. (1990) Biochemistry, Moore,D.D., Seidman.J.G., Smith,J.A. and Struhl,K. (eds), Current 29, 11024-11028. Protocols in Molecular Biology. Vol. 2, John Wiley and Sons, New York, Graham,F.L. and van der Eb,A.J. (1973) Virology, 52, 456-467. NY, pp. 9.2.1-9.2.6. Gullick,W.J. (1991) Br. Med. Bull., 47, 87-98. Shen,S.-H., Bastien,L., Posner,B.I. and Chretien,P. (1991) Nature, 352, Gullick,W.J., Downward,J. and Waterfield,M.D. (1985) EMBO J., 4, 736-739. 2869-2877. Simon,M.A., Dodson,G.S. and Rubin,G.M. (1993) Cell, 73, 169-177. Gullick,W.J., Berger,M.S., Bennett,P.L.P., Rothbard,J.B. and Songyang,Z. et al. (1993) Cell, 72, 767-778. Waterfield,M.D. (1987) Int. J. Cancer, 40, 246-254. Stahl,M.L., Ferenz,C.R., Kelleher,K.L., Kriz,R.W. and Knopf,J.L. (1988) Hanks,S.K., Quinn,A.M. and Hunter,T. (1988) Science, 241, 42-52. Nature, 332, 269-272. Hu,P., Margolis,B., Skolnik,E.Y., Lammers,R., Ullrich,A. and Suen,K.-L., Bustelo,X.R., Pawson,T. and Barbacid,M. (1993) Mol. Cell. Schlessinger,J. (1992). Mol. Cell. Biol., 12, 981-990. Biol., 13, 5500-5512. Hudziak,R.M., Schlessinger,J. and Ullrich,A. (1987) Proc. NatlAcad. Sci. Tan,J.C., Nocka,K., Ray,P., Traktman,P. and Besmer,P. (1990) Science, USA, 84, 7159-7163. 247, 209-212. Jainchill,J.L., Aaronson,S.A. and Todaro,G.J. (1969) J. Virol., 4, 549-553. Trahey,M. et al. (1988) Science, 242, 1697-1700. Kamps,M.P. and Sefton,B.M. (1989) Anal. Biochem., 176, 22-27. Valius,M. and Kazlauskas,A. (1993) Cell, 73, 321-334. Kaplan,D., Schaffthausen,B., Cantley,L. and Roberts,T. (1985) Nature, 315, Vogel,W., Lammers,R., Huang,J. and Ullrich,A. (1993) Science, 259, 239-242. 1611- 1614. Knighton,D.R., Zheng,J., Ten Eyck,L.F., Ashford,V.A., Xuong,N.-H., Walsh,J.P., Caldwell,K.K. and Majerus,P.W. (1991) Proc. NatlAcad. Sci. and Sowadski,J.M. (1991) Science, 253, 407-414. USA, 88, 9184-9187. Taylor,S.S. Kraus,M.H., Issing,W., Miki,T., Popescu,N.C. and Aaronson,S.A. (1989) Waterfield,M.D., Mayes,E.L.V., Stroobant,P., Bennett,P.L.P., Young,S., Proc. Natl Acad. Sci. USA, 86, 9193-9197. Goodfellow,P.N., Bantng,G.S. and Ozanne,B. (1982) J. Cell. Biochem., Fedi,P., Starks,V., Muraro,R. and Aaronson,S.A. (1993) 20, 149-161. Kraus,M.H., Proc. Natl Acad. Sci. USA, 90, 2900-2904. Received on October 1, 1993; revised on April 15, 1994 Shepard,H.M. and Mendelsohn,J. (1991) Mol. Cell. Biol., 11, Kumar,R., 979-986. Lehtola,L., Sistonen,L. and Alitalo,K. (1989) EMBO J., Lehvaslaiho,H., 8, 159-166. Lehvaslaiho,H., Sistonen,L., diRenzo,F., Partanen,J., Comoglio,P., and Alitalo,K. (1990) J. Cell Biochem., 42, 123-133. Holtti,E. Lemoine,N.R., Mayall,E.S., Jones,T., Sheer,D., McDermid,S., Kendall- (1989) Br. J. Cancer, 60, 897-903. Taylor,P. and Wynford-Thomas,D. etal. (1992a) Br. J. Cancer, 66, 1116-1121. Lemoine,N.R. Lobresco,M., Leung,H., Barton,C., Hughes,C.M., Lemoine,N.R., and Kloppel,G. (1992b) J. Pathol., 168, Prigent,S.A., Gullick,W.J. 269-273. Nature, 363, 85-88. Li,N. et al. (1993) Livneh,E., Benveniste,M., Prywes,R., Felder,S., Kam,Z. and J. Cell Biol., 103, 327-31. Schlessinger,J. (1986) and Gullick,W.J. (1991) In Dickson,R.B. and Lippman,M.E. Lofts,F.J. Hormones: Advances in Cellular and (eds), Genes, Oncogenes and of Breast Cancer. Kluwer Academic Publishers, Molecular Biology pp. 161-179. Boston, USA, Vol. HI, et al., (1992) Cell, 70, 431-442. Lowenstein,E.J. Pelicci,G., Pelicci,P.G. and Pawson,T. (1992) Proc. McGlade,J., Cheng,A., Natl Acad. Sci. USA, 89, 8869-8873. Koch,C.A., Sadowski,I. and Pawson,T. (1988) Oncogene, Moran,M.F., 3, 665-672. and Parker,P.J. (1990) Eur. J. Biochem., 191, Morgan,S.J., Smith,A.D. 761 -767. Weismuller,K.-H., Jungs,G., Schlessinger,J. Obermeier,A., Lammers,R., J. Biol. Chem., 268, 22963 -22966. and Ullrich,A. (1993) et al. 73, 179-191. Olivier,J.P. (1993) Cell, and (1993) Curr. Biol., 3, 434-442. Pawson,T. Schlessinger,J. et al. Cell, 70, 93-104. Pelicci,G. (1992) et al. (1993) J. Cell. Biochem., Suppi. 17A, 238. Pelicci,G. FEBS Lett., 308, 38-42. Pillay,T.S. and Makgoba,M.W. (1992) Whitney,G.S., Neubauer,M.G., Green,J.M., Plowman,G.D., Todaro,G.J. and Shoyab,M. (1990) Proc. NatlAcad. McDonald,V.L., 87, 4905-4909. Sci. USA, Green,J.M., Culouscou,J.-M., Whitney,G.S., Plowman,G.D., Proc. Neubauer,M.G. and Shoyab,M. (1993) Carlton,G.W., Foy,L., 1746-1750. Natl Acad. Sci. USA, 90, Selden,C. and Lemoine,N.R., Hughes,C.M., Plowman,G.D., Prigent,S.A., 1273-1278. Oncogene, 7, Gullick,W.J. (1992) Lemoine,N.R. and Gullick,W.J. J. Gooden,C.S.R., (1993) Rajkumar,T., 271-278. Pathol., 170, and Schlessinger,J. Ullrich,A. (1984) Nature, 324, Riedel,H., Dull,T.J., 68-70. et al. 689-692. Rozakis-Adcock,M. (1992) Nature, 360,

Journal

The EMBO Journal – Springer Journals

Published: Jun 1, 1994

There are no references for this article.