Ligand-induced transphosphorylation between different - NCBI - NIH

3 downloads 348 Views 1MB Size Report
'Department of Pharmacology, New York University, 550 First. Avenue, New York ...... Itoh,N., Terachi,T., Ohta,M. and Seo,M.K. (1990) Biochem. Biophks. Res.
The EMBO Journal vol.10 no.10 pp.2849-2854, 1991

Ligand-induced transphosphorylation between different FGF receptors

F.Bellot, G.Crumley, J.M.Kaplow, J.Schlessinger', M.Jaye and C.A.Dionne Molecular Biology Department, Rh6ne-Poulenc Rorer Central Research, 680 Allendale Road, King of Prussia, PA 19406 and 'Department of Pharmacology, New York University, 550 First Avenue, New York, NY 10016, USA Communicated by J.Schlessinger

Recent evidence shows that different fibroblast growth factors (FGF) bind with similar high affinities to two FGF receptors (FGFR) called flg and bek. In order to explore the mechanism of FGFR tyrosine autophosphorylation, we have generated cell lines which co-express a kinasenegative mutant of FGFR and an active form of FGFR. The following transfected NIH 3T3 cells were generated: (i) cells which express a shorter truncated form of bek (two Ig domains) together with a kinase-negative mutant of full length bek (bek K517A), (ii) cells which express wild-type bek together with kinase-negative flg (flg K514A) and (iii) cells co-expressing wild-type flg together with bek K517A. Immunoprecipitations with either bekor flg-specific antisera followed by immunoblotting indicated that the double transfectants express the desired receptor species. The addition of acidic FGF (aFGF) to the various cell lines followed by immunoprecipitation with anti-FGFR antibodies and immunoblotting with anti-phosphotyrosine specific antibodies indicated that aFGF induces tyrosine phosphorylation of the kinasenegative FGFR mutants. These results show that tyrosine autophosphorylation of the kinase-negative FGFR is mediated by a transphosphorylation mechanism and that both homologous (bek - bek) and heterologous (bek - fig and flg - bek) transphosphorylation occurs in living cells. Recent evidence shows that tyrosine autophosphorylation of receptors with tyrosine kinase activities is essential for mediating interactions with signaling molecules. Therefore, heterologous transphosphorylation could amplify the response of cells to various forms of FGFs and their cognate receptors. Key words: FGF receptor/signal transduction/tyrosine phosphorylation

Introduction The fibroblast growth factor (FGF) family consists of seven closely related heparin-binding polypeptides which affect a wide variety of biological events including cell growth and differentiation, neurite outgrowth, embryogenesis and angiogenesis (reviewed by Burgess and Maciag, 1989). The FGFs exert their pleiotropic effects by binding to specific tyrosine kinase-linked cell surface receptors. Four distinct © Oxford University Press

FGF receptors (flg, bek, Cek-2 and FGFR4) have now been cloned from various species (Kornbluth et al., 1988; Ruta et al., 1988; Lee et al., 1989; Pasquale and Singer, 1989; Dionne et al., 1990; Hattori et al., 1990; Houssaint et al., 1990, Itoh et al., 1990; Johnson et al., 1990; Mansukhani et al., 1990; Musci et al., 1990; Pasquale, 1990; Hou et al., 1991; Keegan etal., 1991; Miki etal., 1991; Partanen et al., 1991). These tyrosine kinase-linked FGF receptors are composed of a glycosylated extracellular ligand binding domain formed by either two or three Ig-like domains (Hunkapiller and Hood, 1989), a single transmembrane sequence and a cytoplasmic tyrosine kinase domain divided by an insertion sequence of 14 amino acids. We have cloned cDNAs for the three Ig-like domain forms of human fig and bek and a short form of human bek lacking the first Ig-like domain and the acidic segment between the first and second Ig-like domains (Dionne et al., 1990; Crumley et al., 1991). When overexpressed in NIH 3T3 cells both the full-length forms and the shorter bek form bind with high affinity to aFGF, bFGF and hstl/kFGF (Dionne et al., 1990; Houssaint et al., 1990; Mansukhani et al., 1990; Crumley et al., 1991; F.Bellot, G.Crumley, J.M.Kaplow, C.Basilico, M.Jaye, J.Schlessinger and C.A.Dionne, in preparation). Previous work with other receptor-linked tyrosine kinases, for example EGFR, PDGFR, insulin R and CSF-1R, has helped to elucidate several general mechanisms for this class of receptors. The kinase activity has been shown to be critical for signal transduction, proper ligand-induced intracellular trafficking of the receptor, mitogenesis and cellular transformation. Furthermore, the tyrosine kinase receptors undergo ligand-induced dimerization which facilitates activation and intermolecular tyrosine phosphorylation of the cytoplasmic domains (reviewed by Schlessinger, 1988; Ullrich and Schlessinger, 1990). To determine whether FGF receptor autophosphorylation is mediated by an intermolecular mechanism, we coexpressed a kinase-negative full-length (140 kDa) mutant of bek together with a truncated (1 16 kDa) kinase competent form of bek in NIH 3T3 cells, which are devoid of endogenous bek. Stimulation of these cells with aFGF resulted in phosphorylation of the kinase-negative bek mutant, providing evidence for an intermolecular autophosphorylation mechanism. Transphosphorylation occurs also between two different FGF receptors as a kinasenegative bek mutant was phosphorylated by flg and a kinasenegative flg mutant was phosphorylated by bek. In both cases, the kinase-negative FGF receptors were phosphorylated on tyrosine in response to aFGF. The data show that autophosphorylation of the FGF receptors flg and bek occurs via an intermolecular transphosphorylation mechanism and transphosphorylation between heterologous FGF receptor tyrosine kinases is an available mechanism for signal transduction. 2849

F.Bellot et al.

Results Expression of kinase-negative fig and bek proteins Kinase-negative point mutants of flg (flg K514A) and bek (bek K517A) were constructed by site-directed mutagenesis replacing the conserved lysine residues in the ATP-binding site (Hanks et al., 1988) by alanine. The mutated cDNAs were inserted, under the control of the adenovirus major late promoter and a cytomegalovirus enhancer, into the eukaryotic expression vector, pMJ30 (Jaye et al., 1988). The expression plasmids were co-transfected with a hygromycin B resistance vector into the flg and bek overexpressing cell lines, NFlg26 and NBek8 respectively (Dionne et al., 1990), and into the cell line NBeklO3 which overexpresses a shorter bek protein (Crumley et al., 1991). Control cells were Table I. aFGF binding to cell lines Cell lines

Apparent Kd (nM)

Receptors/cell

NIH 3T3 NBeklO3-2 NBeklO3-2/BekK-3 NBek8 NFlg26 NFIgK-18 NBekK- 19 NBek8/FlgK-3 NFlg26/bekK-34

0.053 0.024 0.089 0.074 0.070 0.086 0.043 0.100 0.115

10 000 47 000 422 000 304 000 241 000 177 000 95 000 433 000 383 000

co-transfected with pSV2neo (Southern and Berg, 1982) and flg K514A or bek K517A expression vectors. Expression of the kinase-negative receptors in the transfected clones was monitored by immunoprecipitation and immunoblotting with rabbit antipeptide antibodies Flg lB and BekIB, which are specific for flg or bek, respectively (Dionne et al., 1990). The following clonal cell lines were selected: NFlgK -18 and NBekK -19, expressing flg K514A or bek K517A proteins, respectively. NFlg26/BekK-34 and NBek8/FlgK-3, which are cells previously transfected with flg or bek wild-type cDNA (Dionne et al., 1990) and now re-transfected with bek orflg kinase-negative cDNAs, respectively; and NBeklO3-2/BekK-3 which are cells previously transfected with a shorter bek cDNA (Crumley et al., 1991) and now re-transfected with the full-length bek K517A cDNA. In order to verify that the kinase-negative receptors retain the binding characteristics of their normal counterpart and to determine the number of receptors expressed by these transfected cells, we performed saturation binding experiments with [125I]aFGF on the different cell lines and analyzed the binding data according to the method of Scatchard (1949) (Table I). The flg K514A and bek K517A mutants exhibited an apparent Kd for aFGF similar to that of their respective wild-type form (Kd 0.1 nM). These ,'*

P[

! Tr

w 4Y

t.ell lili-n n s

IV

-; w4

Fig. 1. In vitro autophosphorylation of FGF-R. NNeo5 (lane 1), NFlgK -18 (lane 2) and NFlg26(lane 3) cells were immunoprecipitated with anti-FlglB and Neo5 (lane 4), NBekK -19 (lane 5) and NBek8 (lane 6) with anti-BekIB antibodies. Half of each immunoprecipitate was incubated with 10 isCi of [-y-32P]ATP and 5 mM MnCI2 for 20 min at room temperature and then analyzed by SDS-PAGE and autoradiography as described in Materials and methods. A 10 min exposure of the gel to X-ray film is presented here (panel A). The other part of the immunoprecipitate was used for immunoblotting with a mixture of anti-FlglB and anti-BeklB antibodies. A 3 h exposure of the gel to X-ray film is presented here (panel B). The positions of molecular weight standards are indicated.

2850

I,

-"

Fig. 2. Intermolecular tyrosine phosphorylation of bek. NBeklO3-2 (lanes 1 and 2) and NBeklO3-2/BekK-3 (lanes 3 and 4) cells were stimulated (+) with aFGF for 5 min at 37°C and lysed. The lysates were immunoprecipitated with anti-BekIB antibody, then subjected to SDS-PAGE and immunoblotting with anti-phosphotyrosine antibodies as described in Materials and methods (panel A). After autoradiography, the bound antibody-iodinated protein A complexes were stripped from the membrane which was then reprobed with antiBeklB antibody followed by [1251]protein A (panel B). Autoradiograms of the resulting blots are shown. The position of molecular weight markers are indicated.

Heterologous transphosphorylation of FGF receptors

results confirmed that a point mutation at the ATP-binding site in the kinase domain does not affect the ligand-binding site in the extracellular domain. Similar conclusions were reached for the EGF, PDGF and insulin receptors (Chou et al., 1987; Honegger et al., 1987; Sorkin et al., 1991). NFlgK- 18, NBekK- 19, NBek8/FlgK -3 and NFlg26/BekK-34 cells expressed between 90 000 and 150 000 kinase-negative receptors on their surface whereas NBeklO3-2/BekK-3 cells expressed -350 000 bek K517A receptors per cell. In vitro autophosphorylation of FGF receptors We assayed flg and bek immunoprecipitates from the transfected cell lines in in vitro autophosphorylation reactions to verify that the mutated FGF receptor cDNAs encoded

kinase-deficient proteins. Cell lysates from NNeoS, NFlgK- 18 and NFlg26 cells were immunoprecipitated with anti-Flg1B antibodies and each immunoprecipitate was divided into two portions. One portion was allowed to undergo autophosphorylation in the presence of ['y-32P]ATP and 5 mM MnCl2 for 20 min at room temperature and then analyzed by SDS -PAGE and autoradiography (Figure 1, lanes 1-3). The other portion was subjected to SDS -PAGE and immunoblotting with anti-FGFR antibodies (Figure IB). Similar analyses were performed with anti-BekIB immunoprecipitates of NNeoS, NBekK- 19 and NBek8 cell lysates (Figure 1, lanes 4-6). No endogenous murine flg and bek proteins were detectable either by autophosphorylation or by immunoblotting (Figure 1, lanes 1 and 4). Autophosphorylation was easily detected for wild-type forms of flg and bek (Figure lA, lanes 3 and 6) whereas the mutated proteins were clearly not autophosphorylated (Figure lA, lanes 2 and 5) despite a high level of expression in the cell lines (Figure 1B). These observations support the prediction that the lysines K514 in flg and K517 in bek are critical for tyrosine kinase activity of flg and bek, respectively.

Intermolecular tyrosine phosphorylation of FGF receptors The FGF receptors flg and bek possess intrinsic ligand activated protein tyrosine kinase activity, which is essential for receptor autophosphorylation and phosphorylation of intracellular substances on tyrosine residues. It has been shown that the EGF receptors undergo ligand-dependent tyrosine phosphorylation through an intermolecular mechanism (Honegger et al., 1989, 1990). In order to determine whether tyrosine phosphorylation of FGF receptors occurs by an intermolecular process, confluent monolayers of the NBek1O3-2/BekK-3 cells were stimulated with aFGF for 5 min at 37°C, lysed and subjected to immunoprecipitation with anti-BeklB antibodies followed by SDS-PAGE. After electrophoresis, the immunoprecipitated proteins were transferred to Immobilon P membrane and were probed with anti-phosphotyrosine antibodies followed by [1251] protein A (Figure 2A). After washing out the antibody - iodinated protein A complexes, the blot was reprobed with anti-BekiB (Figure 2B). In this experiment, the ligand-induced autophosphorylation of beklO3 is barely detectable (Figure 2A, lanes 1 -2), due to the fact that the level of expression of this receptor in this particular cell line is relatively low (Table I). However, in this experiment, the appearance of a Mr 90 000 tyrosinephosphorylated protein which coimmunoprecipitates with the bek protein serves as a marker that ligand-induced activation of the receptor has occurred. Ligand-dependent tyrosine phosphorylation of the bek K517A protein in NBeklO3-2/BekK-3 cells is easily observed (Figure 2A, lanes 3 and 4). These results indicate that tyrosine phosphorylation of the bek K5 17A protein occurs via transphosphorylation by the activated shorter bek 103 protein. The control cell line, NBekK -19, which expresses a similar level of bek K517A protein to NBeklO3-2/BekK-3, does not exhibit ligand-dependent phosphorylation, either by the mutated protein itself or by endogenous FGF receptors -

A

S

S

a II

__0 *,

66a

B .

0o Os

S_

Fig. 3. Cross-phosphorylation between heterologous FGF receptors. The FGF receptor overexpressing cell lines were incubated in the presence (+) or absence (-) of aFGF for 5 min at 37°C and then lysed. The lysates were separately immunoprecipitated with anti-FlgIB(F) and anti-BeklB(B) and subjected to SDS-PAGE and immunoblotting as described in Materials and methods. Panel A: autoradiograms of blots probed with antiphosphotyrosine antibodies. Panel B: autoradiogram of the same blots shown in A which were washed and reprobed with a mixture of anti-FlglB and anti-BeklB antibodies. The positions of molecular size standards are indicated. 2851

F.Bellot et al.

.'r4:

M.^i,:s.M,:|.1 ...

.,*....ii^.g...

', .i.,. !

t.: b.eX

=.,.m

t

z

_ || |

s

i

a i

|

|1

's3

'@

fS

.A

NBek8 and NBek8/FlgK-3 cell lines (Figure 3). The experimental protocol was the same as described in the previous section except that, after aFGF stimulation, the cell lysates were divided and immunoprecipitated with either antiFlglB or anti-BekIB. After SDS-PAGE and immunoblotting with anti-phosphotyrosine antibodies, the blots were washed and reprobed with a mixture of anti-Flg lB and antiBekIB antibodies to reveal the amount of FGF receptor protein contained in each immunoprecipitate.

-..

0XI7:. W:S

'0

.:

.-:

The cell lines transfected with the kinase-negative FGF receptor expression vectors produce high levels of protein as demonstrated by receptor binding assays (Table I) and immunoprecipitation/immunoblotting assays (Figure 3B, lanes 3, 4, 13 and 14). Despite the high level of protein, no in vivo kinase activity was observed (Figure 3A, lanes 3, 4, 13 and 14) consistent with the results obtained by in vitro autophosphorylation experiments (Figure 1). NFlg26/BekK-34 cells exhibit aFGF-dependent tyrosine phosphorylation of the 130 kDa bek K517A protein which is specifically immunoprecipitated with the anti-BekIB sera (lanes 11 and 12). We conclude that this protein is the transfected bek K517A since it reacts with BekIB antiserum and there is no detectable endogenous bek protein in the NFlg26 parental cells (Figure 3B, lanes 7 and 8). Since the NBekK- 19 cells do not exhibit ligand-induced phosphorylation of bek proteins, whereas the NFlg26/BekK-34 cells do, we also conclude that the bek K517A protein is transphosphorylated by flg in the doubly transfected cells. Similarly, the 145 kDa flg K514A protein, which immunoprecipitates with the FlgIB antisera, is phosphorylated upon aFGF stimulation of NBek8/FlgK-3 cells (Figure 3A, lanes 21, 22). The observed tyrosine phosphorylation of the flg K514A protein was due to heterologous phosphorylation by bek because tyrosinephosphorylated protein in anti-Flg immunoprecipitates of NFlgK-18 cells (lanes 13 and 14) was not detected even though these cells express an equivalent amount of transfected flg K514A mutants. Wild-type flg protein in NFlg26 cells (lanes 5 and 6) and in the doubly transfected NFlg26/BekK-34 cells (lanes 9 and 10) exhibits a high level of ligand-dependent tyrosine phosphorylation of the major form at 140 kDa and of a less abundant 180 kDa form. The 180 kDa flg protein is a hyperglycosylated form which has been previously described (Dionne et al., 1990). The 120 kDa and 115 kDa forms revealed by anti-FGF receptor antibodies in flg- and bek-expressing cell lines, respectively (Figure 3B), probably correspond to precursor forms not yet expressed at the cell surface since they do not bind FGF in crosslinking experiments (data not shown) and are not in vivo tyrosine-phosphorylated following FGF stimulation (Figure 3A). The wild-type bek protein in NBek8 cells and in the doubly transfected NBek8/FlgK-3 cells exhibited a high background of basal tyrosine phosphorylation (Figure 3A, lanes 19, 20, 23 and 24) which we have observed in other bek overexpressing cell lines. Although we observed only a small increase in bek autophosphorylation upon ligand stimulation, we do detect dramatic ligand-dependent phosphorylation of receptor substrates such as PLC--y (Bellot et al., in preparation) and flg kinase-negative forms (discussed below). These data clearly show that although the kinase negative -

.-

-

-0

Fig. 4. A model for homologous and heterologous transphosphorylation between FGF receptors. A. Intermolecular tyrosine phosphorylation between homologous FGF receptors was demonstrated with cells expressing a short bek103 protein and a kinase-deficient full-length bekK517A protein. Stimulation of NbeklO3-2/bekK-3 cells with aFGF results in tyrosine phosphorylation of the bekK517A protein. The two bek forms were distinguished by size differences. B. Transphosphorylation between heterologous FGFRs was demonstrated with cells expressing the wild-type bek protein and the kinase deficient flgK514A protein. Stimulation of NBek8/FlgK-3 cells with aFGF results in tyrosine phosphorylation of the flgK514A protein. Analogous results were obtained in cells expressing wild type flg protein and the bekK517A protein. The flg and bek proteins were distinguished with specific antisera.

-

-

-

-

(see below). The ease of detection of the tyrosinephosphorylated bek K517A protein in the NBeklO3-2/ BekK-3 cells is probably due to its higher expression relative to the previously transfected bek 1O3 protein (Table I). As discussed below, the greater intensity of the Mr 115 000 protein band from NBeklO3-2/BekK-3 cells compared with the same size protein from NBeklO3-2 cells most likely results from the comigration of an incompletely processed precursor of full-length bek K517A protein with the beklO3 protein (Figure 2B, lanes 3 and 4). -

Cross-phosphorylation occurs between heterologous FGF receptors In order to determine whether cross-phosphorylation could occur between flg and bek proteins, we examined the liganddependent FGF receptor tyrosine phosphorylation in NBekK - 19, NFlg26, NFlg26/BekK - 34, NFlgK - 18, 2852

-

Heterologous transphosphorylation of FGF receptors

bek and flg proteins are incapable of autophosphorylation, they are readily tyrosine phosphorylated in a liganddependent fashion in cell lines which overexpress the heterologous FGF receptors.

Discussion We have generated kinase-deficient point mutations in the FGF receptors flg and bek and then isolated transfected cell lines which overexpress the mutated forms. The kinasedeficient FGF receptors are expressed at the cell surface and bind aFGF with an affinity equal to that of the wild-type receptors, indicating that tyrosine kinase activity is not necessary for proper synthesis, processing, glycosylation and transport to the cell surface of FGF receptors. Similar results were obtained previously with kinase-deficient mutant forms of the EGF (Chen et al., 1987; Honegger et al., 1987), PDGF (Williams, 1989; Sorkin et al., 1991), CSF-1 (Downing et al., 1989) and insulin (Chou et al., 1987; McClain et al., 1987) receptors. The design and results of our experiments to examine FGFR transphosphorylation are illustrated in Figure 4. We took advantage of the difference in size of the bek K517A and shorter bek103 forms to demonstrate ligand-dependent intermolecular phosphorylation of bek. In this regard, the FGF receptors resemble other receptor-linked tyrosine kinases which can undergo intermolecular tyrosine phosphorylation, (Honegger et al., 1989, 1990). Activation of the FGF receptors flg and bek can clearly lead to heterologous cross-phosphorylation on tyrosine residues in a ligand-dependent fashion. The phosphorylation between flg and bek is probably due to direct interaction of the kinase domains, since phosphorylation of the kinasedeficient forms is only observed in cells which overexpress both receptor species. In addition, a recombinant fragment containing the C-terminal autophosphorylation sites of either flg or bek can be phosphorylated on tyrosine by recombinant flg cytoplasmic kinase domain (Mohammadi et al., 1991). Cross-phosphorylation of heterologous tyrosine kinase-linked receptors has previously been shown for EGFR/pl85 neu (Wada et al., 1990) but these receptors utilize different ligands for each monomeric subunit of the receptor dimer. PDGFR-ca/PDGFR-3 exhibit heterologous dimerization in response to a heterodimeric PDGF-AB or homodimeric PDGFAA (Hammacher et al., 1989) and the insulin and insulin-like growth factor receptors can cross-phosphorylate as heterodimeric receptors in response to a single ligand, insulin (Beguinot et al., 1988; Soos and Siddle, 1989). The FGF receptors on the other hand constitute the first system described which exhibits heterologous transphosphorylation and heterodimer formation (Bellot et al., in preparation) in response to a monomeric ligand which binds both monomeric receptors. Heterologous transphosphorylation of FGF receptors may have significant biological implications since many cell types express more than one form of FGF receptor and the ligand specificities of the various FGF receptors is just beginning to be unveiled. The FGF receptors flg, bek, Cek-2 (FGFR-3) and FGFR-4 all exhibit high affinity binding to aFGF (Dionne et al., 1990; Mansukhani et al., 1990; Keegan et al., 1991; Partanen et al., 1991). The bek forms, which we previously described, binds to aFGF and also to bFGF and hst/KFGF with similar high affinities (Dionne et al., 1990; Houssaint et al., 1990; Crumley et al., 1991; Bellot

et al., in preparation). A different splicing variant of bek (Hattori et al., 1990; Miki et al., 1991), however, binds aFGF and keratinocyte growth factor (FGF-7) but not bFGF. Recent studies provide evidence that tyrosine autophosphorylated regions in growth factor receptors represent binding sites of SH2 (src homology 2) domains contained in signaling molecules (Anderson et al., 1990; Margolis et al., 1990). We have recently identified a tyrosine autophosphorylation site on flg and shown that it represents a binding site for the SH2 domain of PLC--y (Mohammadi et al., in preparation). The fact that aFGF causes heterologous autophosphorylation of flg and bek may provide a mechanism for signal amplification as these two receptors are able to bind to PLC-'y and other SH2 containing signaling molecules. Moreover, since some of the FGFR forms described to date do not bind with high affinity to all FGFs, the heterologous transphosphorylation may amplify the response of cells to a particular FGF.

Materials and methods Materials Human recombinant aFGF (Jaye et al., 1987) was iodinated using a modified chloramine T procedure (Kan et al., 1988). The specific activity for different preparations was between 0.7 and 1 x 106 c.p.m./ng. Protein A was iodinated using the chloramine T method. Rabbit polyclonal antibodies directed against peptides specific for flg and bek proteins were obtained by immunizing animals with peptides coupled to keyhole limpet hemocyanin using the crosslinker I-ethyl-3-(3-dimethylamino propyl) carbodiimide (EDC) in complete Freund's adjuvant. Rabbits were boosted every three weeks using the peptide conjugate in incomplete Freund's adjuvant according to standard protocols. Rabbit antisera against the following peptides were generated: anti-Fig 1 (residues 809-823; PRHPAQLANGGLKRR) and anti Bekl (residues 805 -821; YGPCLPQYPHINGSVKT). Monoclonal anti-phosphotyrosine antibody was obtained from UBI (Lake Placid, NY). The NIH 3T3 cell lines were grown in DME medium (GIBCO) containing 10% heat-inactivated calf serum and penicillin/streptomycin antibiotics (Gibco BRL, Gaithersburg, MD). Geneticin 418 (Gibco, BRL) at 500 jig/ml or hygromycin B (Boehringer Mannheim, Indianapolis, IN) at 400 jig/mn were used for selection of resistant clones.

Construction of the kinase-negative mutants The full-lengthflg cDNA was excised from pflgFL16 (Dionne et al., 1990) with Neo and ApaI, blunted with T4 DNA polymerase and cloned into the Hindll site of M13mpl9 with the 5' end near the EcoRl site in the polylinker. The Lys514 residue was mutated to Ala5 14 using the oligonucleotide 5'-CCAAAGTGGCTGTGGCGATGTTGAAGTCGG-3' under standard conditions of site directed mutagenesis (Amersham). The mutated cDNA was excised with SmaI and HindIIl blunted with T4 DNA polymerase and cloned into the EcoRV site of the eukaryotic expression vector, pMJ30 (Jaye et al.. 1988). The full-length bek cDNA was excised from pGC36 (Dionne et al., 1990) with XbaI and SspI, blunted with T4 DNA polymerase and cloned into the SnaI site of M13mpl9 with the 5' end near the EcoRI site of the polylinker. Mutagenesis of the Lys517 residue to AlaS17 was achieved with the oligonucleotide 5'-CACCGTGGCCGTGGCGATGTTGAAAGATG-3'. The mutant cDNA was excised with Sacl and HindlIl, blunted with T4 DNA polymerase and cloned into the EcoRV site of pMJ30. Construction of hygromycin B resistance vector and transfection The hygromycin resistance cDNA (Sugden et al., 1985) flanked by thymidine kinase promoter and polyadenylation sequences, was removed as a 2.0 kb cassette from an Epstein-Barr virus expression vector (Young et al., 1988) using the restriction endonucleases FspI and NarI. The expression cassette was made blunt-ended with T4 DNA polymerase and inserted into the SmaI site of pGem-1 vector (Stratagene). Hygromycin B resistance was used as a second selectable marker in double transfectants already selected by neomycin resistance. In each case the plasmid bearing the FGF receptor mutant was co-transfected at 20:1 mass ratio with a plasmid encoding the gene for hygromycin B resistance. Other co-transfections using the neomycin resistance vector, pSV2neo, were also performed at a 1:20 mass ratio (Southern and Berg, 1982).

2853

F.Bellot et al. Binding of aFGF Equilibrium binding at 4°C of ['25I]aFGF to cells in 24-well tissue culture dishes was performed as previously described (Dionne et al., 1990) and the resulting binding data were analyzed by the method of Scatchard (1949). In vitro autophosphorylation of the FGF receptors Cells were grown to confluency in 15 cm dishes coated with human fibronectin (10 Ag/cm2) and solubilized in lysis buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 1 % Triton X-100, 10% glycerol, 1.5 mM MgCl2, 1 mM EGTA, 100 AM phenylmethylsulfonyl fluoride (PMSF), 1 Ag/ml aprotinin and 1 jig/ml leupeptin) at 4°C. After 15 min on ice, the lysates were centrifuged at 14 000 g for 20 min at 4°C. Proteins were immunoprecipitated for 1 h at 4°C with anti-FigIB or anti-BekIB rabbit polyclonal antibodies prebound to 3 mg of protein A-Sepharose. After washing with HNTG buffer (20 mM HEPES pH 7.5, 150 mM NaCl, 0.1 % Triton X-100, 10% glycerol), the immunoprecipitates were incubated for 20 min at room temperature with 1 MiCi [,y-32p]ATP and 5 mM MnCl2. Reactions were stopped by addition of 3 x Laemmli sample buffer (Laemmli, 1970) and proteins were separated by SDS-PAGE (6% gel) using prestained molecular weight markers (Sigma) as standards. The gel was dried and exposed to Kodak XAR5 film. In vivo tyrosine phosphorylation of the FGF receptors Cells were grown to confluency in fibronectin coated 15 cm dishes, washed briefly with 20 mM sodium acetate pH 4.0, 2 M NaCI in order to remove any bound growth factors from the surface of the cells and incubated for 30 min in binding buffer (DME medium containing 20 mM HEPES pH 7.5, 0.2% BSA and 50 MM sodium orthovanadate). Cells were then incubated for 5 min at 37°C with or without aFGF, washed with cold PBS at 4°C, solubilized in lysis buffer containing phosphatase inhibitors (200 AM sodium orthovanadate, 10 mM sodium pyrophosphate and 100 mM NaF) and the FGF receptors were immunoprecipitated from the lysates using the FlglB and BekIB rabbit antipeptide antibodies as described above. Immunoprecipitates were separated by SDS-PAGE (6% gel) and proteins were electroblotted onto an Immobilon P membrane (Millipore). After saturation in TBS-BSA (10 mM Tris pH 7.4, 150 mM NaCl, 5% BSA) the blots were incubated for 4 h at room temperature with a solution of monoclonal anti-phosphotyrosine antibody in TBS-BSA, washed with PBS and further incubated with [1251]protein A (500 000 c.p.m./ml) in TBS-BSA for 45 min at room temperature. Autoradiograms of the blots were obtained with Kodak XAR5 films. The antibody-iodinated protein A complexes were washed from the blots by soaking them in 200 mM glycine pH 2.5, 0.05 % Tween 20 at 80°C for 2 h. Blots were reprobed using a mixture of anti-FIgiB and anti-BekiB antibody, incubated with iodinated protein A and autoradiographed as described above.

Acknowledgements The authors acknowledge Dr Irit Lax for fruitful discussions and thank Mark Ravera for preparing the recombinant aFGF, Dave Nash for radioiodinations of raFGF and protein A, and Robin McCormick and Patricia Gallagher for excellent secretarial assistance.

References Anderson,D., Koch,C.A., Grey,L., Ellis,C., Moran,M.F. and Pawson,T. (1990) Science, 250, 979-981. Beguinot,F., Smith,R.J., Kahn,C.R., Maron,R., Moses,A.C. and White,M.F. (1988) Biochemistry, 27, 3222-3228. Burgess,W.H. and Maciag,T. (1989) Annu. Rev. Biochem., 58, 575 -606. Chen,W.S., Lazar,C.S., Poenie,M., Tsien,R.Y., Gill,G.N. and Rosenfeld,M.G. (1987) Nature, 328, 820-823. Chou,C.K., Dull,T.J., Russell,D.S., Gherzi,R., Lebwohl,R., Ullrich,A. and Rosen,O.M. (1987) J. Biol. Chem., 262, 1862-1867. Crumley,G., Bellot,F., Kaplow,J.M., Schlessinger,J., Jaye,M. and Dionne,C.A. (1991) Oncogene, in press. Dionne,C.A., Crumley,G., Bellot,F., Kaplow,J.M., Searfoss,G., Ruta,M., Burgess,W.H., Jaye,M. and Schlessinger,J. (1990) EMBO J., 9, 2685-2692. Downing,J.R., Roussel,M.F. and Sherr,C.J. (1989) Mol. Cell. Biol., 9, 2890-2896. Hammacher,A., Mellstrom,K., Heldin,C.H. and Westermark,B. (1989) EMBO J., 8, 2489-2495. Hanks,S.K., Quinn,A.M. and Hunter,T. (1988) Science, 241, 42-52. Hattori,Y., Odagiri,H., Nakatani,H., Miyagawa,K., Naito,K.,

2854

Sakamoto,H., Katoh,O., Yoshida,T., Sugimura,T. and Terada,M. (1990) Proc. Natl. Acad. Sci. USA, 87, 5983-5987. Honegger,A.M., Dull,T.J., Felder,S., Van Obberghen,E., Bellot,F., Szapary,D., Schmidt,A., Ullrich,A. and Schlessinger, J. (1987) Cell, 51, 199-209. Honegger,A.M., Kris,R.M., Ullrich,A. and Schlessinger,J. (1989) Proc. Natl. Acad. Sci. USA, 86, 925 -929. Honegger,A.M., Schmidt,A., Ullrich,A. and Schlessinger,J. (1990) Mol. Cell. Biol., 10, 4035 -4044. Hou,J., Kan,M., McKeehan,K., McBride,G., Adams,P. and Mckeehan,W.L. (1991) Science, 251, 665-668. Houssaint,E., Blanquet,P.R., Champion-Arnaud,P., Gesnel,M.C., Torriglia,A., Courtois,Y. and Breathnach,R. (1990) Proc. Natl. Acad. Sci. USA, 87, 8180-8184. Hunkapiller,T. and Hood,L. (1989) Adv. Immunol., 44, 1-63. Itoh,N., Terachi,T., Ohta,M. and Seo,M.K. (1990) Biochem. Biophks. Res. Commun., 169, 660-685. Jaye,M., Burgess,W.H., Shaw,A.B. and Drohan,W.N. (1987) J. Biol. Chem., 262, 16612-16617. Jaye,M., Lyall,R.M., Mudd,R., Schlessinger,J. and Sarver,N. (1988) EMBO J., 7, 963-969. Johnson,D.E., Lee,P.L., Lu,J. and Williams,L.T. (1990) Mol. Cell. Biol., 10, 4728-4736. Kan,M., DiSorbo,D., Hou,J., Hoshi,H., Mansson,P.-E. and McKeehan,W.L. (1988) J. Biol. Chem., 263, 11306-11313. Keegan,J., Johnson,D.E., Williams,L.T. and Hayman,M.J. (1991) Proc. Natl. Acad. Sci. USA, 88, 1095-1099. Kornbluth,S., Paulson,K.E. and Hanafusa,H. (1988) Mol. Cell. Biol., 8, 5541 -5544. Laemmli,U.K. (1970) Nature, 227, 680-685. Lee,P.L., Johnson,D.E., Cousens,L.S., Fried,V.A. and Williams,L.T. (1989) Science, 245, 57-60. McClain,D.A., Macgawa,H., Lee,J., Dull,T.J., Ullrich,A. and Olefsky,J.M. (1987) J. Biol. Chem., 262, 14663-14671. Mansukhani,A., Moscatelli,D., Talarico,D., Levytska,V. and Basilico,C. (1990) Proc. Natl. Acad. Sci. USA, 87, 4378-4382. Margolis,B., Li,N., Koch,A., Mohammadi,M., Hurwitz,D.R., Zilberstein,A., Ullrich,A., Pawson,T. and Schlessinger,J. (1990) EMBO J., 9, 4375-4380. Miki,T., Fleming,T.P., Bottaro,D.P., Rubin,J.S., Ron,D. and Aaronson,S.A. (1991) Science, 251, 72-75. Mohammadi,M., Honegger,A.M., Rotin,D., Fischer,R., Bellot,F., Li,W., Dionne,C.A., Jaye,M., Rubinstein,M. and Schlessinger,J. (1991) Mol. Cell. Biol., in press. Musci,T.J., Amaya,E. and Kirschner,M.W. (1990) Proc. Natl. Acad. Sci. USA, 87, 8365-8369. Partanen,J., Makela,T.P., Ecrola,E., Korhonen,J., Hirvonen,H., ClaessonWelsh,L. and Alitalo,K. (1991) EMBO J,10, 1347-1354. Pasquale,E.B. (1990) Proc. Natl. Acad. Sci. USA, 87, 5812-5816. Pasquale,E.B. and Singer,S.J. (1989) Proc. Natl. Acad. Sci. USA, 86, 5449-5453. Ruta,M., Howk,R., Ricca,G., Drohan,W., Zabelshansky,M., Laureys,D., Barton,E., Francke,U., Schlessinger,J. and Givol,D. (1988) Oncogene, 3, 9-15. Scatchard,G. (1949) Ann. NY Acad. Sci., 51, 660-672. Schlessinger,J. (1988) Trends Biochem. Sci., 13, 443-447. Soos,M.A. and Siddle,K. (1989) Biochem. J., 263, 553-563. Sorkin,A., Westermark,B., Heldin,C.-H. and Claesson-Welsh,L. (1991) J. Cell. Biol., 112, 469-478. Southern,P.J. and Berg,P. (1982) J. Mol. Appl. Genet., 4, 327-341. Sugden,B., Marsh,K. and Yates,J. (1985) Mol. Cell. Biol., 5, 410-413. Ullrich,A. and Schlessinger,J. (1990) Cell, 61, 203-212. Wada,T., Qian,X. and Greene,M.I. (1990) Cell, 61, 1339-1347. Williams,L.T. (1989) Science, 243, 1564-1570. Young,J.M., Cheadle,C., Foulke,S.F., Jr, Drohan,W.N. and Sarver,N. (1988) Gene, 62, 171-185. Received on June 12, 1991