Characterization of the Kinase Activity Essential for Tyrosine Phosphorylation of P130cas in ®Broblasts

Characterization of the kinase activity essential for tyrosine phosphorylation of p130Cas in ®broblasts

Ryuichi Sakai1,4, Tetsuya Nakamoto2, Keiya Ozawa1, Shin-ichi Aizawa3 and Hisamaru Hirai2

1Molecular Biology Division, Jichi Medical School, 3311-1 Yakushiji, Minamikawachi-machi, Kawachi-gun, Tochigi 329-04; 2The Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113; 3Department of Morphogenesis, Institute of Molecular Embryology and Genetics, Kumamoto University, School of Medicine, 2-2-1 Honjo, Kumamoto 860, Japan

The cellular transformation by v-Src or v-Crk induces and Gish, 1992). Each SH2 region binds to speci®c sets tyrosine phosphorylation of a common substrate mole- of phosphotyrosine-containing proteins by recognizing cule, p130Cas (Cas), which tightly binds these oncopro- a phosphotyrosine in the context of several adjacent teins in vivo. From its structure, Cas is deduced to be an amino acids (Moran et al., 1990; Muller et al., 1992; ideal substrate for Src family kinases and Abl kinase. Songyang et al., 1993). In non-receptor type tyrosine The tyrosine kinase activity associated with Cas was kinases, such as Src, Fps and Abl, the SH2 regions are analysed using mouse variant ®broblasts lacking at least located immediately at N-terminal to kinase domains. one of tyrosine kinases. In normal ®broblasts, Cas is Deletion or substitution of SH2 regions of activated associated with a signi®cant level of tyrosine kinase variants of Src and Fps often impairs the catalytic and activity which eciently phosphorylates Cas in vitro. The transforming activities of these kinases (Kitamura and Cas-associated tyrosine kinase activity was remarkably Yoshida, 1983; Sadowski et al., 1986; Raymond and elevated in Csk7/7 cells, which resulted in hyperphos- Parsons, 1987) and particular mutations within the phorylation of cellular Cas. The associated kinase SH2 region of Src induce host-dependent transforming activity was slightly increased in Src7/7 cells whereas phenotypes (DeCue et al., 1987; Hirai and Varmus, not signi®cantly changed in Abl7/7 nor Fak7/7 cells. On 1990). Furthermore, there is a signi®cant similarity the contrary, the Cas-associated kinase activity was between the substrate speci®city of these tyrosine remarkably decreased in Fyn7/7 cells. At the same time, kinases and the binding speci®city of their SH2 association of Cas with Fyn kinase in vitro was most regions, which was demonstrated by the experiment obviously detected in normal ®broblasts as well as using degenerate synthetic peptides (Pawson, 1995; Csk7/7 cells. Transient expression of v-Crk induced Songyang et al., 1995). elevation of the Cas-associated kinase activity in all of Thus speci®c protein tyrosine phosphorylation these cell lines except the primary culture of Fyn7/7 induced by each of tyrosine kinases is believed to be ®broblasts. These results indicate that Fyn kinase plays a critical machinery of cellular signal transduction. A an essential role in v-Crk-mediated phosphorylation of substantial amount of recent information is obtained Cas. regarding main target substrates of phosphorylation directly attacked by tyrosine kinases. For example, Keywords: p130Cas; tyrosine phosphorylation; Src phosphorylation of PLC-g (Wahl et al., 1988), GAP homology 2; signal transduction (Ellis et al., 1990), Nck (Park and Rhee, 1992; Li et al., 1992; Meisenhelder and Hunter, 1992), Vav (Bustelo et al., 1992; Margolis et al., 1992), Shc (Pelicci et al., 1992), Syp/SH-PTP2 (Feng et al., 1993), and cortactin Introduction (p80/p85) (Wu and Parsons, 1993) is observed during cellular transformation or by various growth factor Numbers of signaling molecules have been identi®ed in stimuli. Each of these signaling molecules may receive the signal transduction pathway from the membrane to several kinds of upstream signals and transduce a the nucleus. As a manner of the signal transfer between common set of signals which regulate cellular growth these molecules, phosphorylation of cellular proteins and dierentiation in a phosphorylation-dependent has been highlighted. Especially, the critical roles of manner. It is, though, still technically dicult to tyrosine phosphorylation in the signal transduction are identify the tyrosine kinase responsible for phosphor- certi®ed by the fact that many receptors for growth ylation of a particular substrate. factors and oncoproteins responsible for malignant We have recently cloned a novel kinase substrate, transformation are proved as tyrosine kinases by the p130Cas (Cas) which contains a cluster of multiple molecular cloning techniques. Recent studies suggest putative SH2-binding motifs and another signaling that a domain called Src homology 2 (SH2) domain region called Src homology 3 (SH3) domain (Sakai et has a signi®cant function to transduce the signal in a al., 1994a). The cellular transformation by v-Src or v- tyrosine phosphorylation-dependent manner (Pawson Crk induces tyrosine phosphorylation of Cas and stable association of Cas with these oncoproteins in vivo, suggesting a critical role of Cas in these oncogenic Correspondence: H Hirai signal transduction and cellular transformation 4Present address: Program in Molecular Biology and Cancer, Samuel (Matsuda et al., 1990; Birge et al., 1992; Sakai et al., Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario M5G 1X5, Canada 1994a,b). v-Crk was ®rst identi®ed as a regulator of Received 22 July 1996; revised 13 November 1996; accepted 13 tyrosine kinases and causes elevation of tyrosine kinase November 1996 activity which results in transformation of ®broblasts Tyrosine kinase associated with p130Cas RSakaiet al 1420 (Mayer et al., 1988). It is still unclear, however, what a 1 2 3 b kind of tyrosine kinases is regulated by v-Crk and how 3Y1 3Y1-Crk SR-3Y1 the regulation takes place. Because tyrosine phosphorylation of Cas is the most obvious change found during 205 — the transformation of ®broblasts by v-Crk, it is quite 117 — important to identify the tyrosine kinase(s) responsible for the phosphorylation of Cas to elucidate the 80 — mechanism of cellular transformation and kinase regulation by v-Crk. 49 — Tyrosine phosphorylation of Cas is also observed in some biological events such as integrin-mediated cell Figure 1 Phosphorylation of p130Cas in vivo.(a) 3Y1-Crk (lane adhesion (Nojima et al., 1995; Petch et al., 1995). Since 2) and SR-3Y1 cells (lane 3) as well as normal 3Y1 cells (lane 1) Cas is predicted to be an ideal target molecule for are labeled 3 h in vivo by orthophosphate, immunoprecipitated by anti-Cas antibody and electrophoresed. Elevated tyrosine several SH2-containing tyrosine kinases including Src phosphorylation of Cas in these cells can be detected as shifts family kinases and Abl judging from their substrate of bands towards apparently higher molecular weight. The speci®cities and SH2-binding speci®cities (Pawson, position of phosphorylated p130 is indicated by a square bracket 1995; Mayer et al., 1995), it might act as a cellular at both ends. Molecular weights are given in kilodaltons. (b) The binding partner of these tyrosine kinases. To know the bands corresponding to phosphorylated Cas were cut out from the gel of (a), digested by TPCK trypsin and mapped two- biological role of Cas in cellular signal transduction, it dimensionally. The origins of electrophoresis were marked by is critical to identify the tyrosine kinase associated with close circles Cas. That is also important for understanding how tyrosine kinases share their roles in signal transduction network. Detection of the Cas-associated kinase (CASK) activity In this work, we characterized and analysed the in normal NIH3T3 cells kinase activity associated with Cas. We utilized variants of mouse ®broblasts from embryos lacking various To identify the cellular kinases responsible for the tyrosine kinases as well as normal mouse ®broblasts for phosphorylation of Cas, we analysed tyrosine kinase analysis of kinase(s) responsible for tyrosine phosphor- activity associated with Cas in the variants of mouse ylation of Cas. The eects of transient expression of v- ®broblasts. We used mutant ®broblasts lacking various Crk on the tyrosine kinase activity were also analysed in tyrosine kinases derived from embryos of gene- these cells for identi®cation of the tyrosine kinase disrupted mice. The complexes associated with Cas involved in transformation by v-Crk. were precipitated from these cell lysates using anti-Cas antiserum and the in vitro kinase assays of the complexes were performed with or without an exogenous substrate, poly[Glu-Tyr]. The kinase activ- Results ities involved in the complexes were analysed by the level of phosphorylation of Cas or of poly[Glu-Tyr]. Phosphopeptide mapping of phosphorylated Cas In normal NIH3T3 cells, a distinct tyrosine kinase To analyse the substrate speci®cities of the kinases activity associated with Cas was observed (Figure 2a, responsible for phosphorylation of p130Cas (Cas) in lanes 1 and 5). The kinase activity eciently transformation of ®broblasts, tyrosine-phosphorylation phosphorylated Cas and poly[Glu-Tyr] in vitro.We patterns of trypsin-digested fragments of Cas were tentatively designate this kinase activity as the CASK compared between cells transformed by v-Crk and v- (Cas-associated Kinase) activity. Cas itself was the Src using two-dimensional phosphopeptide mapping. most ecient substrate of the kinase activity involved For these analyses, we analysed rat 3Y1 cells in the complex just as in ®broblasts transformed by v- expressing v-Crk (3Y1-Crk) or v-Src (SR-3Y1) as well Src or v-Crk (Sakai et al., 1994a). as normal 3Y1 cells. These cells were labeled in vivo by Since tyrosine phosphorylation of Cas is not obvious orthophosphate, immunoprecipitated by antiserum in normal ®broblasts judging from the immunoblotting against Cas, and at ®rst analysed by standard SDS with the anti-phosphotyrosine antibody (Sakai et al., polyacrylamide gel electrophoresis (PAGE). Elevated 1994a) or from phosphoamino acid analysis (data not tyrosine phosphorylation of Cas in 3Y1-Crk cells and shown), it is speculated that the CASK activity could SR-3Y1 cells could be detected as shifts of bands be masked by protein tyrosine phosphatase (PTPase) towards apparently higher molecular weight (Figure activity which might regulate phosphorylation of Cas. 1a). These broad bands were cut out, digested by In fact, we have also reported that orthovanadate, a trypsin and mapped two-dimensionally. Both in 3Y1- PTPase inhibitor enhances the phosphorylation level of Crk cells and in SR-3Y1 cells, more than ten spots Cas (Sakai et al., 1994b). were newly observed indicating multiple tyrosine phosphorylation sites (Figure 1b). Although positions The CASK activity in variant ®broblasts homozygously of most of the spots derived from trypsin-digested lacking tyrosine kinase genes fragments of Cas can be overlapped by each other, there is signi®cant dierence in the phosphorylation There are mainly three bands of Cas (Form A, B and patterns of each tryptic fragment between 3Y1-Crk and C) detected by anti-Cas antibodies as reported (Sakai SR-3Y1 cells. This result suggests that the substrate et al., 1994a). Form A and B are sharp bands detected speci®city of the tyrosine kinase responsible for the at 115 kD and 125 kD, respectively, p130-C is a broad phosphorylation of Cas in ®broblasts expressing v-Crk smear band at 125 ± 135 kD and anti-phosphotyrosine might be slightly dierent from that of v-Src kinase. antibodies mostly recognize this p130-C. Marked Tyrosine kinase associated with p130Cas RSakaiet al 1421 tyrosine phosphorylation of Cas was observed in the a ®broblasts lacking Csk as shown by decreased intensity Csk Csk of the lower band (Form A and B) and appearance of Csk Src Csk Src the upper smear band (Form C) in immunoblotting by 3T3 –/– 3T3 –/– anti-Cas antibody (Figure 2b, lanes 3 and 6). Although –/– –/– –/– –/– the total amount of Cas is comparable between Csk7/7 +Csk +Csk cells and normal NIH3T3 cells, the CASK activity was elevated by about 10-folds in Csk7/7 cells (Figure 2a, lanes 2 and 6). The elevation is suppressed by the 200 — expression of exogenous Csk (Figure 2a, lanes 3 and 7). In this experiment, several Cas-associated substrates for tyrosine kinases were visualized in Csk7/7 116 — ®broblasts. The main phosphorylated substrate except 97 — phosphorylated Cas was a 60 kD protein (Figure 2a, lane 2). Since it is reported that kinase activities of Src- family kinases such as Src, Fyn and Lyn are elevated in 66 — the Csk-de®cient cells, this 60 kD phosphoprotein is supposed to be one of the autophosphorylated Src- family kinases from its molecular weight. It is speculated that the CASK activity in CSK7/7 46 — ®broblasts might be derived from these Src family kinases. In the Csk-de®cient ®broblasts, physical association of Cas with tyrosine kinases was analysed by the in vitro kinase assay. As shown in Figure 3a, the association of Cas with Fyn was obviously detected in Csk-de®cient mouse ®broblasts. Surprisingly, the band of Cas associated with Fyn was also faintly 1 2 3 4 5 6 7 8 detectable in normal NIH3T3 ®broblasts. On the other hand, association of Cas with Src was not obvious in both Csk7/7 ®broblasts and NIH3T3 cells (Figure 3a, lanes 1 and 2), although a smear band around 110 ± b 130 was observed in Csk7/7 cell lysate immunopreci- 1 2 3 4 5 6 pitated by the anti-Src antibody (Figure 3a, lane 2). No physical association of Cas with Lyn, Abl nor Fak was 200 — observed in these cells using the same procedures (data not shown). We further analysed the association between Cas and Fyn kinase using GST (glutathione S-transferase)-fusion proteins. We used lysates of rat ®broblasts 3Y1, 3Y1-Crk and SR-3Y1 to know the 116 — anity with bacterially expressed SH2 and SH3 regions of Fyn fused with GST. As shown in ®gure 3b, every 97 — form of multiple bands of Cas (Form A, B and C) bind the GST-FynSH3 fusion protein, while the phosphotyrosine-containing forms of the Cas (Form C) bind the GST-FynSH2 fusion protein. GST-FynSH2 fusion 66 — protein was so ecient to precipitate the tyrosine phosphorylated Cas that a slight amount of the tyrosine phosphorylated form of Cas was also detected in the normal 3Y1 cells (Figure 3b lane 4). GST itself had no signi®cant anity to any forms of Cas (Figure 3b, lanes 1 ± 3). Elevation of the CASK activity was observed in the 45 — ®broblast lacking Src kinase (Figure 4a, lane 2 and Figure 4b and c, lane 5) without detectable elevation of Figure 2 Distinct CASK activity in normal and Csk-lacking tyrosine phosphorylation of Cas (Figure 2b lanes 2 and ®broblasts. (a) Results of in vitro kinase assay of the complex 5), although the CASK activity was much less than immunoprecipitated by anti-Cas antibody were shown. Normal that in the Csk-negative ®broblasts (Figure 2a, lanes 2, 7/7 7/ NIH3T3 cells (lane 1 and 5), Csk cells (lanes 2 and 6), Csk 4, 6 and 8). As shown in Figure 4a, lane 3, elevation of 7 cells transfected with cDNA of Csk (lanes 3 and 7) and Src7/7 cells (lanes 4 and 8) were analysed. Kinase reaction was the CASK activity was also seen in the ®broblast performed without (lanes 1 ± 4) or with poly-Glu/Tyr (lanes 5 ± lacking both Src and Yes kinases. It might be because 8) as exogenous substrates. The position of phosphorylated Cas is the expression of cellular Cas in these Src7/7 or Src7/7 indicated by a square bracket on the left. Molecular weights are Yes7/7 cells is elevated as shown by the Western given in kilodaltons. (b) Immunoblotting by anti-Cas (lanes 1 ± 3) and 4G10 (lanes 4 ± 6) or normal NIH3T3 calls (lanes 1,4), Src7/ blotting (Figure 4a) although the mechanism is not 7 cells (lanes 2,5) and Csk7/7 cells (lanes 3,6). Molecular weights known. In the ®broblasts lacking Abl or Fak, no are given in kilodaltons signi®cant dierence in the CASK activity was Tyrosine kinase associated with p130Cas RSakaiet al 1422

a a b c 1 2 3 4 5 1 2 3 4 1 2 3 4 5 1 2 3 4 5

205 —

116 —

97 —

Figure 4 CASK activity in mouse variant ®broblasts. Tyrosine 66 — kinase activity associated with Cas was measured by in vitro kinase assay. In vitro kinase assay of immunoprecipitants by anti- Cas antibody (upper panel) and immunoblotting of Cas (lower panel) are shown. (a) Normal NIH3T3 cell (lane 1), Src7/7 cells (lanes 2), Src7/7 Yes7/7 (lane 3) and Abl7/7 (lane 4) cells (b) 45 — Normal NIH3T3 cells (lane 1), primary culture of Fyn7/7 ®broblasts (lane 2), Fyn7/7 ®broblast cell line (lane 3), Fak7/7 cells (lane 4) and Src7/7 cells (lane 5) were analysed. (c) Kinase assay with poly-Glu/Tyr of the samples of (b) (lanes 1 ± 5) was also performed. The position of phosphorylated Cas is indicated b by a square bracket on the left of upper panels and the position of Cas is indicated by a square bracket on the left of lower panels GST Fyn SH2 FynSH3 in (a) and (b). – Src Crk – Src Crk – Src Crk

in the cell line of Fyn7/7 ®broblasts, which was extremely low but apparently detectable using 116 — poly[Glu-Tyr] as a substrate as shown Figure 4c. It 97 — might be suggested that some tyrosine kinase(s), which is less ecient to phosphorylate Cas when compared 1 2 3 4 5 6 7 8 9 with Fyn kinase, could become activated during the Figure 3 Association of p130Cas with Fyn. (a) Association of immortalization procedure of ®broblasts to compensate Cas with other tyrosine kinase was analysed by in vitro kinase the signi®cant biological role of Fyn kinase and should assay. NIH3T3 cells (lanes 1 and 4), Csk7/7 ®broblasts (lanes 2 thus be found as a detectable CASK activity. and 3) and Fyn7/7 ®broblasts (lane 5) are immunoprecipitated by Src (lanes 1, 2) or Fyn (lanes 3 ± 5) and immune complex kinase assay was performed. The position of phosphorylated Cas Induction of tyrosine kinase activity by v-Crk is indicated by a square bracket on the right. Molecular weights are given in kilodaltons. (b) Cell lysates of 3Y1 (lanes 1, 4 and 7), To analyse the eect of v-Crk expression in these SR-3Y1 (lanes 2, 5 and 8) and 3Y1-Crk (lanes 3, 6 and 9) were variant ®broblasts, sense and antisense v-Crk cDNA precipitated by GST (lanes 1,2 and 3), GST-FynSH2 (lanes 4,5 were inserted into the expression vector possessing the and 6) and GST-FynSH3 (lanes 7, 8, 9) and subjected to immunoblotting by anti-Cas antibody. Molecular weights are SRa promoter and transiently expressed in various given in kilodaltons ®broblasts by the calcium phosphate coprecipitation method. The eect of v-Crk on the CASK activity was analysed 48 h after the transfection by the in vitro observed when compared with that of NIH3T3 cells kinase activity comparing with that of the control (Figure 4a, lane 4 and Figure 4b, lane 4). These results vector possessing v-Crk cDNA in antisense direction. indicate that the CASK activity does not derive only The results of NIH3T3, Src-negative and Fyn-negative from one of Src, Yes, Abl and Fak, leaving the cells are shown in Figure 5. Both in NIH3T3 cells and possibility of the compensation mechanism by many in Src negative cells, the CASK activity was tyrosine kinases. prominently increased by the expression of v-Crk. On the contrary, the CASK activity was severely Obvious elevation of the CASK activity by v-Crk was decreased in the ®broblasts lacking Fyn while the also observed in other ®broblasts lacking Yes, Abl, amount of expressed Cas protein in the Fyn7/7 Fak or Csk (data not shown). ®broblasts was the same as that in normal ®broblasts In the primary culture of Fyn-negative ®broblasts, (Figure 4b, lane 3). In order to exclude the possibility no change in the CASK activity was induced by the of clonal selection of the cells with particular expression of v-Crk. When Fyn kinase was also characteristics during the establishment of the cell introduced into the ®broblasts by the transient line, primary culture ®broblasts which were directly expression method, induction of the CASK activity cultured from the embryo of Fyn-negative mice were by v-Crk reappeared as shown in Figure 5a. The also examined. Notably, the CASK activity was most amount of Cas was not aected by the transient extremely suppressed in the primary culture of embryo expression of v-Crk (Figure 5b). These results suggest ®broblasts from Fyn7/7 mice (Figure 4b, lane 2). The that Fyn is a critical kinase for the phosphorylation of signi®cantly lower CASK activity was observed in the Cas by expression of v-Crk at least in the primary Fyn7/7 primary culture cells when compared with that culture ®broblasts. In the cell line of Fyn-negative Tyrosine kinase associated with p130Cas RSakaiet al

1423 ummry of the nlyses of the geu tivity in vrint

a le I firolsts Fyn–/–

3T3 Fyn–/– Src–/– pirolsts geu tivity iffet of vEgrk gs expression

xorml xsrQQ C

+Fyn C ±G±

xorml el C

v-Crk – + – + – + – + C

±G±

r righ C CC

±G± ±G±

r es righ C CC

±G±

pyn @primryA ery vow ± C

Cas ±G±

@ell lineA pyn vow C C

±G±

pk xorml C C

±G±

gsk ery righ C C

poly 1 2 3 4 5 6 7 8 1 2 3 4 5 6 [Glu-Tyr] v-Crk – + – + – + – + – + – + – +

205 —

116 — 1 2 3 4 5 6 7 8 97 —

b 66 — 1 2 3 4 5 6 7 8

45 — 116 —

97 — Figure 6 Change in activity of tyrosine kinases by v-Crk. (a) Figure 5 Eects of transient expression of v-Crk on the CASK Changes in tyrosine kinase activity of the complexes immunopre- activity. v-Crk was transiently expressed in various kinase- cipitated with Src (lanes 1, 2), Fyn (lanes 3, 4), Abl (lanes 5, 6) de®cient mouse ®broblasts as well as NIH3T3 cells, and change and Cas (lanes 7,8) were examined by in vitro kinase assay. (b) in tyrosine kinase activity associated with Cas was analysed by in Kinase assay with poly-Glu/Tyr of the samples in a (lanes 1 ± 6) vitro kinase assay. In vitro kinase activity of immunoprecipitants was also performed. v-Crk is transiently expressed in lanes by anti-Cas antibody without exogenous substrates (upper panel) marked (+) while control plasmid is transfected in lanes marked and with poly-Glu/Tyr as exogenous substrates (lower panel) was (7). The position of phosphorylated Cas is indicated by a square examined (a). Immunoblotting of the corresponding ®broblasts by bracket on the left anti-Cas antibody was also performed (b). Results of NIH3T3 cells (lanes 1, 2), Fyn7/7 ®broblasts (lanes 3, 4), Fyn7/7 ®broblasts transfected with Fyn cDNA (lanes 5, 6) and Src7/7 ®broblasts (lanes 7, 8) are shown. v-Crk is transiently expressed in lanes marked (+) while control plasmid is transfected in lanes immunoprecipitated by anti-paxillin antibody and the marked (7) eect of v-Crk expression on the kinase activity involved in the complex was analysed. As shown in Figure 7, the signi®cant kinase activity was induced by the expression of v-Crk and that kinase activity ®broblasts, however, the elevation of CASK activity by phosphorylates Cas quite eciently in vitro while v-Crk was slightly detected (data not shown). It is paxillin itself was not obviously phosphorylated by indicated that the tyrosine kinase activated in the Fyn- this kinase activity. In this case, the kinase activity was negative cell line, also responds to the stimulation by v- not induced in Fyn7/7 primary ®broblasts by the Crk and partially compensates the function of Fyn. expression of v-Crk, suggesting this kinase activity is The results obtained from the analysis of Cas- identical to the CASK activity. associated kinase activity using mouse variant fibro- blasts are summarized in Table 1. By the expression of v-Crk in ®broblasts, the kinase Discussion activities of both c-Src and Fyn kinases are induced as shown in Figure 6 whereas the kinase activity of Abl is v-Crk was ®rst identi®ed as a regulator of tyrosine just slightly detectable. The enhancement of the kinases (Mayer et al., 1988) although the mechanism of tyrosine kinase activities of Fyn and Src is, in both the regulation is still unclear. It induces an elevated level cases, two- to three-folds, while the change in the of tyrosine-phosphorylation of several cellular proteins amount of phosphotyrosines of Cas during the in v-Crk-transformed cells while it lacks a kinase transformation by v-Crk is quite drastic. Although it domain (Matsuda et al., 1990; Mayer and Hanafusa, is not clear how v-Crk regulates the kinase activity of 1990). The main phosphoproteins have been identi®ed these tyrosine kinases, there seems to be a mechanism as p130Cas (Cas) and paxillin (Matsuda et al., 1991; in Cas to amplify the signal transduced by a small Sakai et al., 1994a; Birge et al., 1993). As for the change in the kinase activity into a remarkable change tyrosine kinases responsible for the phosphorylation of in the tyrosine phosphorylation level. these proteins, quite limited information is available. It Finally, the kinase activity associated with another is indicated that c-Src kinase has a stimulatory eect on v-Crk-associating phosphoprotein, paxillin, was exam- the transformation by v-Crk (Sabe et al., 1992). c-Abl is ined in the same manner. The cell lysates were found to be a tyrosine kinase which is responsible for Tyrosine kinase associated with p130Cas RSakaiet al 1424 forms tight complex with Src family kinases including 1 2 3 4 5 6 7 8 9 10 11 12 v-Src. According to our recent mutagenesis experi- v-Crk – + – + – + – + – + – + ments, this substrate domain binds the SH2 domain of v-Crk while the C-terminal region, called Src-binding 205 — domain, binds Src kinase through both SH2 and SH3 regions (Nakamoto et al., 1996). The Src-binding domain is also associated with the tyrosine kinase activity induced by v-Crk in vivo. The binding manner 116 — of Cas indicates the possibility that it might act as a 97 — regulatory partner of speci®c tyrosine kinases. As for the regulation of Src-family kinases, presented was a 66 — model that SH2-regions of Src family kinases bind intra-molecularly to the C-terminal phosphorylated tyrosines forming close forms and thus suppress the kinase activity of themselves. According to this model, the phosphorylation-dependent binding of Cas with the 45 — SH2 region of Src family kinases might convert the tyrosine kinase from a closed form into an open form which act as an activated kinase. Because the change in the phosphorylation of Cas is Figure 7 Paxillin is involved in the kinase-Cas complex. In vitro so drastic by the expression of v-Crk, a processive kinase assay of variant ®broblasts immunoprecipitated by anti- phosphorylation mechanism is suggested to explain the paxillin antibody was examined. Normal NIH3T3 cells (lanes 1, 2, phosphorylation of Cas (Mayer et al., 1995). An SH2- 7 and 8), primary cultured Fyn7/7 cells (lanes 3, 4, 9 and 10) and Src7/7 cells (lanes 5, 6, 11 and 12) were analysed. v-Crk is containing tyrosine kinase might approach to a transiently expressed in lanes marked (+) while control plasmid is phosphorylated tyrosine in the substrate region of Cas transfected in lanes marked (7). Kinase reaction was performed using the anity of SH2 region and phosphorylate the without (lanes 1 ± 6) or with poly-Glu/Tyr (lanes 7 ± 12) as tyrosine motif adjacent to the original phosphorylated exogenous substrates. The position of phosphorylated Cas is indicated by a square bracket on the left. Molecular weights are tyrosine. Then SH2 shifts to the newly phosphorylated given in kilodaltons tyrosine and in that way all the tyrosine motifs become phosphorylated in a processing manner. This mechanism is supported by in vitro experiments in the case of c- Abl kinase. There is the possibility that Fyn kinase the phosphorylation of c-Crk (Feller et al., 1994; Ren et might play a key role in starting the phosphorylation of al., 1994), and to phosphorylate Cas eciently in vitro the ®rst tyrosine and another kinase such as c-Abl might (Mayer et al., 1995). Our results that show the elevated processively phosphorylate Cas. CASK activity in Csk-de®cient cells indicate that a Our experiments showed that Cas associates with the member(s) of Src family kinases which is known to be distinct kinase activity even when it has no obvious suppressed by the activity of Csk kinase is essential for phosphorylation. There are experimental data to show the regulation of the CASK activity. Furthermore, the that some PTPase activity is regulating the phosphor- experiments using kinase-de®cient ®broblasts suggest ylation states of Cas (Sakai et al., 1994b). This PTPase that among the Src family kinases, especially the Fyn activity might possibly work to negate the basal kinase kinase might be mainly responsible for the CASK activity associating with Cas. Thus several fold activity of normal ®broblasts and elevation of the induction of the kinase activity induced by v-Crk CASK activity by v-Crk at least in a critical step for the may trigger the massive change in the phosphorylation process of Cas phosphorylation. It is also supported by of Cas along with the processive mechanism of other experiments showing the elevated kinase activity phosphorylation. of Fyn and Src in v-Crk-transformed cells and the A recent study indicates that dierent Src family association between Fyn and Cas in Csk-negative cells kinases contribute to the phosphorylation of various as well as in normal ®broblasts. It is quite interesting to kinds of substrate molecules (Thomas et al., 1995). For test whether v-Crk can transform the cells lacking Fyn example, cortactin and tensin are mainly phosphory- or Src kinase. However, it has been reported that v-Crk lated by Src kinase, while paxillin and focal adhesion can transform some established rat ®broblasts but not kinase (Fak) are phosphorylated by both Src and Fyn mouse ®broblasts (Sabe et al., 1992; Ogawa et al., 1994). kinase. Although there are no substrates known to be The substrate domain of Cas contains a total of 15 Fyn-speci®c so far, our data suggest that p130 is a repeats of six variants of putative SH2 binding motifs candidate molecule of the Fyn-speci®c substrate in in about 250 amino-acid stretch. This region may thus some conditions such as v-Crk-mediated cell transfor- recognize a set of SH2-containing molecules and bind mation. The role of Fyn kinase as the CASK activity some of them simultaneously or others of them might not be an absolute one, because, as shown in the competitively. There are also a set of SH2-binding Fyn-negative cell line, some compensation mechanism regions and an SH3-binding region at the C-terminal could be prepared. Considering our recent data to region of Cas. From its primary structure deduced show the involvement of Cas phosphorylation during from the amino acid sequence, Cas is supposed to be a the integrin-mediated cell adhesion (Nojima et al., novel type of adapter molecule which can accept 1995; Petch et al., 1995), it is quite interesting to signals from a variety of SH2 and SH3 containing examine whether any of Src-family kinases are essential molecules and output signals from its own SH3 region. for the phosphorylation of Cas in the integrin-mediated One of the notable characteristics of Cas is that it signaling. Recently, using the same set of mouse Tyrosine kinase associated with p130Cas RSakaiet al 1425 ®broblasts derived from kinase-negative mice, we antibody, 4G10, was obtained from UBI. GST-FynSH2 and found that c-Src is essential for the phosphorylation GST-FynSH3 constructs are kindly given from T Yamamoto of Cas in the signaling pathway via integrin (Hamasaki (Umemori et al., 1994). GST fusion proteins were expressed et al., 1996; Vuori et al., 1996), suggesting a responsible in E. coli and puri®ed by anity chromatography using kinase for the phosphorylation of Cas is dierent immobilized glutathione-Sepharose 4B beads (Phamacia). among signaling pathways. Further study is required for understanding the SDS-polyacrylamide gel electrophoresis and Western blotting activation mechanism of Src family kinases during the For protein analysis, cells were lysed in 1% Triton Buer transformation by v-Crk. It is also an interesting concern (10 mM Tris HCl, pH 7.4, 5 mM EDTA, 150 mM NaCl, 1% how the speci®city of each Src family kinase is regulated. Triton X-100, 10% glycerol, 10 u/ml aprotinin, 1 mM

Cas might be an ideal model molecule for the elucidation phenylmethylsulfonyl ¯uoride (PMSF), 1 mM Na3VO4). of the recognition mechanism of both substrate- SDS-polyacrylamide gel electrophoresis (PAGE) was done speci®city and binding-speci®city between kinase mole- as described by Laemmli (1970) using 7.5% polyacrylamide cules and substrate molecules. The physiological gel unless specially indicated. For Western blotting of total condition which causes phosphorylation of Cas and the cell lysates, samples containing 50 mgproteinperlanewere applied to gels. Western blotting were done as described signal originating from Cas should be analysed further. (Towbin et al., 1979) using aCas2 (1 : 2500), 4G10 (5 mg IgG/ml) or aHcrk antibody (1 : 2500) as ®rst antibodies and using ProtoBlot Western AP System (Promega) for second Materials and methods antibodies and staining following manufacturer's instruc- tions. Cells 3Y1-Crk is an isolated clone of rat 3Y1 cells (Kimura et Immune complex kinase assay and phosphopeptide mapping al., 1975) transfected with v-crk cDNA of an avian sarcoma virus, CT10 (Mayer et al., 1988). SR-3Y1 is a For immune complex kinase assay, cell lysates containing 3Y1 cell line transformed by v-Src of Rous sarcoma virus 500 mg of proteins were mixed with 1 mlofaCas2, 2.5 mlof (Zaitsu et al., 1988). Abl-negative ®broblasts are kindly anti-Fyn antibody, 1 ml of anti-Src antibody (2 ± 17) or 2 ml given from Dr BJ Mayer. Src-negative ®broblasts and Src/ of anti-paxillin antibody and incubated for 1 h on ice. Yes negative ®broblasts are kindly given from Dr HE Samples were rotated with Protein-A Sepharose (Sigma) for Varmus. Csk-negative, Fyn-negative and Fak-negative 1hat48C, then beads were washed three times using 1% mouse ®broblasts are established as described (Nada et Triton Buer and three times using Kinase Buer (50 mM al., 1994; Yagi et al., 1994; Ilic et al., 1995). Fibroblast Tris HCl, pH 7.4, 50 mM NaCl, 10 mM MgCl2,10mM cells are cultured in Dulbecco's modi®ed Eagle medium MnCl2).Kinasereactionwasperformedin30mlofKinase (DMEM) supplemented with 10% bovine serum. For Buer with 5 mCi of [g32P]ATP (Amersham) at room preparation of primary culture of embryonic ®broblasts, temperature for 15 min. Samples were boiled in Sample an E12 day mouse embryo was dissected in 1% trypsin by Buer (2% SDS, 0.1 M Tris-HCl, pH 6.8, 10% glycerol, pipetting, and plated onto tissue culture dishes in DMEM 0.01% bromophenol blue, 0.1 M dithiothreitol) and analysed with 10% fetal calf serum (FCS; Irvine Science Co. Ltd.). by 7.5% polyacrylamide gel. Gels were further treated with After they grew to con¯uence, they were passaged to use 1 N KOH for 1 h at 558C before exposure to an X-ray ®lm in for the transient transfection. order to detect phosphotyrosine-containing proteins. For phosphopeptide mapping, cells were labeled by orthophosphate for 3 h as described previously (Sakai et Transfection of v-Crk-expressing plasmid al., 1994b). After autoradiography, the bands corresponding For the transient expression of v-Crk, v-crk cDNA was to phosphorylated p130 were cut out, washed 2 h in 25% inserted in the expression vector with SRa promoter methanol, 2 h in 10% methanol, and 2 h in distilled water. (pSSRa) and transfected into ®broblasts by calcium Gels homogenized in 500 mlof50mMNaHCO3 were digested phosphate precipitation method as described (Wigler et by tolylsulfonyl phenylalanyl chrolomethyl ketone (TPCK) al., 1977). Cells were harvested at 48 h after transfection treated-trypsin (Worthington) at 378C overnight. Following and expression of v-Crk and kinase activity associated with digestion, supernatants were lyophilized, oxidized with 50 ml p130Cas (Cas) was measured. of performic acid, washed by 400 ml of water and dried. Samples were analysed by electrophoresis in 1% NaHCO3 (®rst dimension) and by ascending chromatography in chro- Antibodies and GST-fusion proteins matography buer (n-Butanol/pyridine/acetic acid/water; A rabbit polyclonal antibody (aHcrk) against v-Crk protein 15 : 10 : 3 : 12) on thin layer cellulose (TLC) plates. expressed by the baculovirus vector system and anity puri®ed using anti-gag antibody (1A1) was produced. A rabbit polyclonal antibody against p130 (aCas2) was Acknowledgements produced as described (Sakai et al., 1994a). 2 ± 17 We thank Hidesaburo Hanafusa and Tadashi Yamamoto (Microbiological Associates) was used as a monoclonal for providing the v-crk cDNA and GST-Fyn constructs, antibody against Src. Anti-Fyn polyclonal antibody respectively. We also thank HE Varmus and BJ Mayer for (WAKO, Osaka), anti-paxillin polyclonal antibody providing Src-de®cient and Abl-de®cient mouse ®broblasts, (Zymed) was also used. Anti-phosphotyrosine monoclonal respectively.

References

Birge RB, Fajardo JE, Mayer BJ and Hanafusa H. (1992). J. Bustelo XR, Ledbetter JA and Barbacid M. (1992). Nature, Biol. Chem., 267, 10588 ± 10595. 356, 68 ± 71. Birge RB, Fajardo JE, Reichman C, Shoelson SE, Songyang DeCue, JE, Sadowski I, Martin GS and Pawson T. (1987). Z, Cantley LC and Hanafusa H. (1993). Mol. Cell. Biol., Proc.Natl.Acad.Sci.USA,84, 9064 ± 9068. 13, 4648 ± 4856. Tyrosine kinase associated with p130Cas RSakaiet al 1426 Ellis C, Moran M, McCormick F and Pawson T. (1990). Park D and Rhee SG. (1992). Mol. Cell. Biol., 12, 5816 ± Nature, 343, 377 ± 381. 5823. Feller SM, Knudsen B and Hanafusa H. (1994) EMBO J. 13, Pawson T and Gish GD. (1992). Cell, 71, 359 ± 362. 2341 ± 2351. Pawson T. (1995). Nature, 373, 477. Feng GS, Hui CC and Pawson T. (1993). Science, 259, 1607 ± Pelicci G, Lanfrancone L, Grignani F, McGlade J, Cavallo 1611. F, Forni G, Nicoletti I, Grignani F, Pawson T and Pelicci Hamasaki K, Mimura T, Morino N, Furuya H, Nakamoto PG. (1992). Cell, 70, 93 ± 104. T, Aizawa S, Morimoto C, Yazaki Y, Hirai H and Nojima Petch LA, Bockholt SM, Bouton A, Parsons JT and Burridge Y. (1996). Biochem. Biophys. Res. Comm., 222, 338 ± 343. K. (1995). JCellSci.,108, 1371 ± 1379. Ilic D, Furuta Y, Kanazawa S, Takeda N, Sobue K, Raymond VW and Parsons JT. (1987). Virology, 160, 400 ± Nakatsuji N, Nomura S, Fujimoto J, Okada M, 410. Yamamoto T and Aizawa S. (1995). Nature, 377, 539 ± Ren R, Te ZS and Baltimore D. (1994). Genes Dev., 8, 783 ± 541. 795. Hirai H and Varmus HE. (1990). Proc. Natl. Acad. Sci. USA, Sabe H, Okada M, Nakagawa H and Hanafusa H. (1992). 87, 8592 ± 8596. Mol. Cell. Biol., 12, 4706 ± 4713. Kimura G, Itagaki A and Summers J. (1975). Int. J. Cancer, Sadowski I, Stone JC and Pawson T. (1986). Mol. Cell. Biol., 15, 694 ± 706. 6, 4396 ± 4408. Kitamura N and Yoshida M. (1983). J. Virol., 46, 985 ± 992. Sakai R, Iwamatsu A, Hirano N, Ogawa S, Tanaka T, Mano Laemmli UK. (1970). Nature, 227, 680 ± 685. H, Yazaki Y and Hirai H. (1994a). EMBO J., 13, 3748 ± Li W, Hu P, Skolnik EY, Ullrich A and Schlessinger J. 3756. (1992). Mol. Cell. Biol., 12, 5824 ± 5833. Sakai R, Iwamatsu A, Hirano N, Ogawa S, Tanaka T, Margolis B, Hu P, Katzav S, Li W, Oliver JM, Ullrich A, Nishida J, Yazaki Y and Hirai H. (1994b). J. Biol. Chem., Weiss A and Schlessinger J. (1992). Nature, 356, 71 ± 74. 269, 32740 ± 32746. Matsuda M, Mayer BJ, Fukui Y and Hanafusa H. (1990). SongyangZ,ShoelsonS,ChanudhuriM,GishG,PawsonT, Science, 248, 1537 ± 1539. Haser WG, King F, Roberts T, Ratnofsky S, Lechleider Matsuda M, Mayer BJ and Hanafusa H. (1991). Mol. Cell. RJ, Neel BG, Birge RB, Fajardo JE, Chou MM, Hanafusa Biol., 11, 1607 ± 1613. H, Schahausen B and Cantley LC. (1993). Cell, 72, 767 ± Mayer BJ, Hamaguchi M and Hanafusa H. (1988). Nature, 778. 332, 272 ± 275. Songyang Z, Carraway III, KL, Eck MJ, Harrison SC, Mayer BJ and Hanafusa H. (1990). Proc. Natl. Acad. Sci. Feldman RA, Mohammadi M, Schlessinger J, Hubbard USA, 87, 2638 ± 2642. SR, Smith DP, Eng C, Lorenzo MJ, Ponder BAJ, Mayer Mayer BJ, Hirai H and Sakai R. (1995). Current Biol., 5, BJ and Cantley LC. (1995). Nature, 373, 536 ± 539. 296 ± 305. Thomas SM, Soriano P and Imamoto A. (1995). Nature, 376, Meisenhelder J and Hunter T. (1992). Mol. Cell. Biol., 12, 267 ± 271. 5843 ± 5856. Towbin H, Staehelin T and Gordon J. (1979). Proc. Natl. Moran MF, Koch CA, Anderson D, Ellis C, England L, Acad. Sci. USA, 76, 4350 ± 4354. Martin GS and Pawson T. (1990). Proc. Natl. Acad. Sci. Umemori H, Sato S, Yagi T, Aizawa S and Yamamoto T. USA, 87, 8622 ± 8626. (1994). Nature, 367, 572 ± 576. Muller AJ, Pendergast AM, Havlik MH, Puil L, Pawson T Vuori K, Hirai H, Aizawa S and Ruoslahti E. (1996). Mol. and Witte ON. (1992). Mol. Cell. Biol., 12, 5087 ± 5793. Cell. Biol., 16, 2606 ± 2613. Nada S, Okada M, Aizawa S and Nakagawa H. (1994). Wahl MD, Daniel TO and Carpenter G. (1988). Science, 241, Oncogene, 9, 3571 ± 3578. 968 ± 971. NakamotoT,SakaiR,OzawaK,YazakiYandHiraiH. Wigler M, Silverstein S, Lee LS, Pellicer A, Cheng YC and (1996). J. Biol. Chem., 271, 8959 ± 8965. Axel R. (1977). Cell, 11, 223 ± 232. NojimaY,MorinoN,MimuraT,HamasakiK,FuruyaH, Wu H and Parsons JT. (1993). J. Cell. Biol., 120, 1417 ± 1426. Sakai R, Sato T, Tachibana K, Morimoto C, Yazaki Y Yagi T, Shigetani Y, Furuta Y, Nada S, Okado N, Ikawa Y and Hirai H. (1995). J. Biol. Chem., 270, 15398 ± 15402. and Aizawa S. (1994). Oncogene, 9, 2433 ± 2440. Ogawa S, Toyoshima H, Kozutsumi H, Hagiwara K, Sakai Zaitsu H, Tanaka H, Mitsudomi T, Matsuzaki A, Ohtsu M R, Tanaka T, Hirano N, Mano H, Yazaki Y and Hirai H. and Kimura G. (1988). Biomed. Res., 9, 181 ± 197. (1994). Oncogene, 9, 1669 ± 1678.