- Email: [email protected]
Sam68 Is a Ras-GAP-Associated Protein in Mitosis
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
245, 562–566 (1998)
RC988374
Sam68 Is a Ras-GAP-Associated Protein in Mitosis Estelle Guitard,1 Isabelle Barlat, Florence Maurier,1 Fabien Schweighoffer,1 and Bruno Tocque1 Gene Medicine Department, Rhoˆne-Poulenc Rorer, 13 quai Jules Guesde, 94403 Vitry sur Seine Cedex, France
Received February 25, 1998
Sam68 is the major tyrosine-phosphorylated and Srcassociated protein in mitotic cells. Sam68 stimulates G1/S transition and this effect is dependent on the integrity of its KH domain (hnRNPK Homology) which confers nucleic acid binding properties. During mitosis, Sam68 undergoes tyrosine phosphorylation, which negatively regulates its nucleic acid binding properties and mediates the interaction of Sam68 with critical SH2containing signaling proteins such as Grb2, PLCg1 and Ras-GAP. However, the interaction of Ras-GAP with Sam68 has been brought into question, based on the lack of co-immunoprecipitation between Sam68 and Ras-GAP in interphase cells. Here we show that the choice of anti-Ras-GAP antibodies is critical for the detection of Ras-GAP/Sam68 complex formation, and that this interaction is specific for G2/M transition in both NIH3T3 and Src-transformed cells. Such data reinforce the importance of the interaction of Ras-GAP with RNA binding proteins during cell proliferation through its SH2 and SH3 domains. q 1998 Academic Press
Sam68 is the main tyrosine-phosphorylated and Srcassociated protein in mitotic cells (1-5). Sam68 is a KHcontaining RNA binding protein (6,7) and has been shown to bind single strand nucleic acid (8,9). Cloning of an isoform of Sam68 deleted in the KH domain, with decreased RNA binding properties in vitro, brought the first evidence of an impact of this member of the hnRNPK protein family in the regulation of G1/S transition (10). During mitosis, Sam68 undergoes tyrosine phosphorylation which negatively regulates its nucleic acid binding properties (9,11) and interacts through both SH2- and SH3 mediated interactions with signaling molecules such as Grb2, PLCg1 and the Src tyrosine kinase family (12,13). We document in this report 1 Present address: BioScreen Therapeutics, 58, Boulevard Saint Denis, 92400 Courbevoie, France. The abbreviations used are: FCS, foetal calf serum; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; FACS, fluorescence-activated cell sorter; BrdUrd, Bromodeoxyuridine.
the interaction of Sam68 with Ras-GAP which may reflect an impact of this hnRNP on the Ras-GAP transduction pathway, or, alternatively, may represent an additional role for Ras-GAP in G2/M transition. MATERIALS AND METHODS Antibodies. The anti-Sam68 monoclonal antibody, directed against the KH domain (anti-KH antibody) was purchased from Transduction Laboratories. The anti-phosphotyrosine antibody was purchased from Upstate Biotechnology Inc. The anti-Ras-GAP antibodies are both monoclonal antibodies (MAb); the B4F8 MAb is directed against the SH3 domain of Ras-GAP and was obtained from Upstate Biotechnology Inc.; the 15F8 MAb was obtained in mice by immunization with the COOH-Ras-GAP terminal domain (residues 702 to 1044) (Biocytex, Marseille, France) and purified from ascitic fluids by protein A affinity chromatography. The antibody against myc-tag is the 9E10 anti-myc antibody. Monoclonal anti-Src antibody is directed against SH3 domain and was purchased from Calbiochem (clone 327). Cell culture. NIH 3T3 and Src-transformed NIH 3T3 (NIH 3T3Src, gift from B. Wasylyk) cells were routinely grown in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO-BRL) supplemented with 10% FCS1 and 2mM L-Glutamine (GIBCO-BRL), and were incubated at 377C in 5% CO2. For some experiments, cells were synchronised at different phases of the cell cycle. NIH 3T3 cells were synchronised in G1 by serum starvation during 24 hrs (in medium containing 0,5% FCS) and growth was then stimulated by 20% serum for 2hrs (5). In NIH 3T3-Src cells, G1 synchronisation was performed by mitotic shake-off after 16hrs nocodazole treatment and replating for 3hrs. Mitotic arrest was obtained in both types of cells with 16hrs nocodazole treatment (40ng/ml, SIGMA). Correct synchronisation of the cells was checked by FACS analysis. BrdUrd (30mM) was added 30min before collection. Cells were stained for DNA content with propidium iodide and for incorporation of BrdUrd with a fluoresceinconjugated anti-BrdUrd antibody. Transient transfections. NIH 3T3 and NIH 3T3-Src cells were transfected with 5mg of myc-tagged Sam68 SV40 expression vector (pSV2myc hump62). Transfection experiments were performed using lipofectamine (Life Technologies, Inc.) and peptide H1, which increases the transfection efficacy by complexing the cDNA (Patent submitted in France 95 01865, WO 9625508-A1), during 4 hrs at 377C. Cells were then incubated during 48 hrs with DMEM containing 10% FCS to ensure optimal expression of the protein. Preparation of cell lysates and immunoprecipitations. Cells were washed three times with PBS containing 1mM sodium orthovanadate (Na3VO4), then lysed for 30min at 47C in 500ml to 1ml of HNTG buffer, consisting of 50mM HEPES, pH7,5, 150mM NaCl, 1mM EGTA, 1mM MgCl2 , 10% glycerol, 1% Triton X-100, phosphatase inhibitors (1mM Na3VO4 and 10mM NaF), and protease inhibitors
562
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(1mg of leupeptin per ml, 1mg of trypsin inhibitor per ml, 1mg of pepstatin A per ml, 2mg of Aprotinin per ml, 10mg of benzamidine per ml, 1mM phenylmethylsulfonyl fluoride, 1mg of antipain per ml, and 1mg of chymostatin per ml). Lysates were cleared by centrifugation at 4000 RPM for 15min. Cell lysates were precleared with 50ml of Protein G-PLUS Agarose (Santa Cruz Biotechnology) for 1hr at 47C. They were then incubated with the appropriate antibodies for 2hrs at 47C. 50ml of Protein G-PLUS Agarose were then added to the immune complexes and incubation was continued overnight at 47C. Western blotting. The immunoprecipitates were washed three times with HNTG buffer, solubilised in Laemmli buffer and denaturated during 5min at 957C. The proteins were resolved on 4-20% SDSPAGE and transferred onto nitrocellulose membranes. The blots were blocked in PBS containing 0,05% Tween and 2% non-fat milk and then incubated overnight with the appropriate antibody in the same buffer. The blots were further incubated with secondary antibodies conjugated to peroxidase, and revealed by Enhanced ChemiLuminescence (Amersham Corp.).
RESULTS AND DISCUSSION In order to identify in vivo partners of Sam68, extracts from mitotic NIH 3T3 (Fig. 1, odd lanes) or NIH 3T3-Src cells (Fig. 1, even lanes) were immunoprecipitated with an anti-Sam68 monoclonal antibody directed against the KH domain (anti-KH antibody) and analysed by immunoblotting. As previously described (2,4), Src was detected by immunoblotting with an antiSrc antibody (Fig.1B, lanes 3 and 4) in these immunoprecipitates. In addition to Src, tyrosine-phosphorylated Sam68-associated proteins, including a 120 kDa protein, were detected (Fig.1A, lanes 1 and 2). Probing of the anti-KH immunoprecipitates with an anti-RasGAP monoclonal antibody (B4F8) revealed a 120 kDa band (Fig. 1B, lanes 1 and 2), indicating that Ras-GAP was present in the immunoprecipitates. To further confirm the Ras-GAP/Sam68 interaction, extracts from NIH 3T3 and NIH 3T3-Src cells synchronised in G2/M (Fig. 2A and Fig. 2B, respectively) were immunoprecipitated with two different antibodies directed against Ras-GAP: the 15F8 antibody which is directed against the catalytic domain of Ras-GAP and the B4F8 monoclonal antibody which is specific for the SH3 domain of Ras-GAP. Immunoprecipitation with the 15F8 antibody and resulting detection by immunoblotting with the KH antibody revealed one endogenous band of 68kDa (Fig. 2A, lane 1). In contrast, such an interaction was not observed when the immunoprecipitation was performed with the B4F8 antibody (Fig. 2A, lane 2). Similar results were obtained using G2/M cellular extracts from Src-transformed NIH 3T3 cells (Fig. 2B) with both the 15F8 (lane 1) and the B4F8 antibodies (lane 2). To further ensure that the 68kDa protein interacting with Ras-GAP in mitotic cells was the product of the hump62 gene, immunoprecipitation with either the 15F8 or B4F8 anti-Ras-GAP antibodies, followed by immunoblotting with the anti-myc 9E10 monoclonal antibody, was performed using mitotic extracts from NIH 3T3 and NIH 3T3-Src cells expressing myc-tagged
FIG. 1. Sam68 interacts with Ras-GAP in both mitotic NIH 3T3 and Src-transformed NIH 3T3 cells. (A) NIH 3T3 (lanes 1 and 3) and NIH 3T3-Src cells (lanes 2 and 4) were arrested in mitosis with 16 hrs nocodazole treatment. After immunoprecipitation (see Materials and Methods) of the cell lysates with an anti-KH antibody, proteins were resolved by 4-20% SDS-PAGE, and the figure represents the Western blot analysis using an anti-phosphotyrosine antibody (lanes 1 and 2). Expression of endogenous Sam68 was checked simultaneously by direct immunoblotting with the anti-KH antibody (lanes 3 and 4). (B) Nature of the phosphoproteins previously detected was determined in both mitotic NIH 3T3 (lanes 1 and 3) and NIH 3T3Src (lanes 2 and 4) cells by immunoprecipitation with the anti-KH antibody. Sam68 and associated proteins were resolved as previously described. The figure represents the immunoblotting with the B4F8 anti-Ras-GAP antibody (lanes 1 and 2), or the anti-src antibody (lanes 3 and 4). Arrows indicate the positions of Ras-GAP and Src proteins.
Sam68 after transient transfections with pSV2myc hump62 expression vector (Fig. 2A, lanes 3 and 4 and Fig. 2B, lanes 3 and 4). These results confirmed that Sam68 can be detected specifically in 15F8 immunoprecipitates and therefore interacts with Ras-GAP in mitotic cells. Ras-GAP immunoprecipitation with both the 15F8 and B4F8 antibodies was checked by probing of the same blots with the B4F8 antibody (Fig. 2A, lanes 5 and 6 ; Fig. 2B, lane 5 and 6). The lack of co-immunoprecipitation with the B4F8 anti-Ras-GAP antibody suggests that conformational environment is critical for complex formation between Ras-GAP and Sam68. Further the epitope of the B4F8 antibody maps to the SH3 region of Ras-GAP. The association between RasGAP and Sam68 may be dependent on tyrosine phosphorylation of Sam68 which could sustain the interaction with the SH2 domains of Ras-GAP. Therefore, it is possible that either the SH3 recognition by the B4F8
563
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ern-blotting, as previously described. In both cases, association of Ras-GAP with Sam68 was observed in G2/ M but not in early G1 cells (Figure 3). Direct implications of our results are that Sam68 and Ras associate together in proliferating cells and that this interaction is restricted to cells undergoing mitosis. Cell fractionation experiments revealed that expression of Sam68 is mainly localized within the nucleus compartment on NIH3T3 cells, whereas Ras-GAP is detectible only within the cytoplasm. It is therefore likely that the formation of the Ras-GAP/Sam68 complex requires the disruption of the nucleus membrane as occurs during mitosis. Sam68 becomes hyperphosphorylated mainly at tyrosine residues during mitosis (1, 2); previous evidence
FIG. 2. A monoclonal antibody directed against the Carboxy-terminal portion of Ras-GAP immunoprecipitates the product of hump62 cDNA. Total lysates of mitotic NIH 3T3 (A) or NIH 3T3-Src (B) cells were immunoprecipitated with either the 15F8 anti-RasGAP antibody, whose antigen binding site lies within the Carboxyterminus of Ras-GAP (lanes 1, 3 and 5), or the B4F8 anti-Ras-GAP antibody, which maps to the SH3 domain of Ras-GAP (lanes 2, 4 and 6). Ras-GAP and associated proteins were resolved as previously described, and Sam68 was detected by immunoblotting with the antiKH antibody (lanes 1 and 2). A similar experiment was performed in mitotic NIH 3T3 and NIH 3T3-src cells transiently transfected with pSV2myc hump62 expression vector. Exogenous myc-tagged Sam68 was detected with the 9E10 anti-myc antibody (lanes 3 and 4). The same immunoprecipitations were immunoblotted with the B4F8 anti-Ras-GAP antibody (lanes 5 and 6). Expression of endogenous Sam68 and exogenous mycSam68 was checked by direct Western blotting with both the anti-KH (lane 7) and 9E10 anti-myc (lane 8) antibodies.
antibody induces a conformational modification which inhibits the interaction of the SH2 domains of Ras-GAP with the phosphotyrosine residues of Sam68 or that both the SH2 and SH3 domains are involved in RasGAP/Sam68 interaction, as has already been described for the Src-tyrosine kinase family. To further confirm that the Ras-GAP/Sam68 complexes are specifically detected in G2/M, co-immunoprecipitations were carried out on synchronised cells, transfected or not with myc-tagged expression vectors. NIH 3T3 or NIH 3T3-Src transformed cells were synchronised in early G1 (by mitotic shake-off) or in G2/M phase (by nocodazole treatment). Cell synchronisation was checked by FACS analysis (data not shown). Proteins (from separate cells extracts) were immunoprecipitated with the 15F8 antibody and detected by west-
FIG. 3. Sam68 interacts with Ras-GAP exclusively in mitosis. NIH 3T3 cells (A) and NIH 3T3-Src cells (B) were either maintained in a proliferating state (P lanes) or synchronised at different stages of the cell cycle. G1 cells (G1 lanes) and mitotic cells (M lanes) were obtained as described in Materials and Methods. Cells were lysed and whole cells extracts were either immunoprecipitated with the 15F8 anti-Ras-GAP antibody (large frames) or directly immunoblotted (small frames). Endogenous Sam68 was detected with the antiKH antibody, while exogenous mycSam68, expressed after transient transfections, was detected with the 9E10 anti-myc antibody. Detection of Sam68 and mycSam68 were realised on separate cells extracts. Arrows indicate the positions of Sam68 and mycSam68.
564
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Model for the regulation of Sam68 function during the cell cycle. The figure illustrates diagrammatically progression through the cell cycle and proposes two mechanisms for the involvement of Sam68 in cell cycle progression. During G1/S transition, Sam68 and Sam68DKH, which are both expressed, have antagonistic effects on cyclin D1 expression, and therefore on cell cycle progression (10). During mitosis, while Sam68DKH is not expressed, Sam68 interacts with Ras-GAP, possibly through a third partner (X) yet to be identified. This interaction may modulate Sam68 function and suggests a role for the Ras-GAP/Sam68 complex in RNA metabolism.
demonstrate that Ras-GAP binds phosphorylated tyrosines through SH2 domains (12). It is reasonable to suggest that the SH2 domains of Ras-GAP may shield tyrosine-phosphorylated Sam68 from the enzymatic attack of regulatory phosphatases. Maintenance of the tyrosine-phosphorylated form of Sam68 may affect the activity of the protein in a number of ways. The ability of Sam68 to bind to specific mRNA molecules may also be affected and it has been previously demonstrated that phosphorylation of Sam68 inhibits its nucleic acid binding properties (9). It is possible that Ras-GAP interaction with Sam68, through stabilization of the tyrosine-phosphorylated form of Sam68, may lead to the release of specific RNA targets, which have to be translated or degraded during mitosis. Alternatively, the degradation of Sam68 may be influenced by maintaining the protein in a highly tyrosine-phosphorylated form. Apart from these suggestions, we cannot exclude the possibility that Ras-GAP modulates some other, as yet unknown, function of Sam68. Interactions between protein partners may be forged or lost depending on the state of Sam68 phosphorylation. Conversely, Sam68 may affect different unidentified functions of Ras-GAP during mitosis. The sole function of Sam68 documented so far is in the promotion of S phase entry through its nucleic acid binding properties (10). On the basis of our new results,
we propose the following scheme to illustrate the impact of Sam68 on cell cycle progression (Figure 4). The positive effect of Sam68 on G1/S transition is inhibited by an isoform of Sam68, Sam68DKH, which exhibits impaired nucleic acid binding properties (10). During S phase, Sam68DKH disappears and Sam68 promotes S phase entry possibly by interacting with its RNA target which has still to be discovered. It has been reported that cyclin D1 expression is regulated by the balance of expression between Sam68 and Sam68DKH (10), implicating cyclin D1 on a pathway elicited by Sam68 during G1/S transition. An alternative pathway involving the regulatory interactions between Sam68 and different and as yet unknown protein partners is also portrayed. There is no evidence at present to suggest a functional role for Sam68 during interphase. During mitosis, tyrosine phosphorylation of Sam68 may occur, due to the disruption of the nucleus membrane, which allows interaction with protein partners such as Ras-GAP. Although we cannot exclude this interaction takes place through a third protein partner. Our data documents the interaction of Ras-GAP with Sam68 during mitosis and indicates a possible role of Ras-GAP in the regulation of RNA binding proteins. It is of interest to note that, in response to growth factor stimulation and thus during cell cycle progression, RasGAP recruits two different nucleic acid binding pro-
565
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
teins: G3BP, which interacts with the SH3 domain, and Sam68, which appears to be recruited in a SH2dependent manner. Future experiments will aim to decipher the contribution of Ras-GAP to cell cycle progression through its regulation of RNA metabolism. REFERENCES 1. Courtneige, S. A., and Fumagalli, S. (1994) Trends Cell Biol. 4, 345–347. 2. Taylor, S. J., and Shalloway, D. (1994) Nature 368, 867–871. 3. Weng, Z., Thomas, S. M., Rickles, R. J., Taylor, J. A., Brauer, A. W., Seidel-Dugan, C., Michael, W. M., Dreyfuss, G., and Brugge, J. S. (1994) Mol. Cell. Biol. 14, 4509–4521. 4. Fumagalli, S., Totty, N. F., Hsuan, J. J., and Courtneidge, S. A. (1994) Nature 368, 871–874. 5. Lock, P., Fumagalli, S., Polakis, P., McCormick, F., and Courtneidge, S. A. (1997) Cell 84, 23–24.
6. Gibson, T. J., Thompson, J. D., Heringa, J. (1993) FEBS Lett. 324, 361–366. 7. Kiledjian, M., and Dreyfuss, G. (1992) EMBO. J. 11, 2655–2664. 8.Wong, G., Muller, O., Clark, R., Conroy, L., Moran, M. F., Polakis, P., and McCormick, F. (1992) Cell 69, 551–558. 9. Wang, L. L., Richard, S., and Shaw, A. S. (1995) J. Biol. Chem. 270, 2010–2013. 10. Barlat, I., Maurier, F., Duchesne, M., Guitard, E., Tocque, B., and Schweighoffer, F. (1997) J. Biol. Chem. 272, 3129–3132. 11. Taylor, S. J., Anafi, M., Pawson, T., and Shalloway, D. (1995) J. Biol. Chem. 270, 10120–10124. 12. Richard, S., Yu, D., Blumer, K. J., Hausladen, D., Olszowy, M. W., Connely, P. A., and Shaw, A. S., (1995) Mol. Cell. Biol. 15, 186–197. 13. Neet, K., and Hunter, T. (1995) Mol. Cell. Biol. 15, 4908–4920. 14. Parker, F., Maurier, F., Delumeau, I., Duchesne, M., Faucher, D., Debussche, L., Dugue´, A., Schweighoffer, F., and Tocque´, B. (1996) Mol. Cell. Biol. 16, 2561–2569. 15. Birge, R. B., Fajardo, J. E., Mayer, B. J., Hanafusa, H. (1992) J. Biol. Chem. 267, 10588–10595.
566
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
245, 562–566 (1998)
RC988374
Sam68 Is a Ras-GAP-Associated Protein in Mitosis Estelle Guitard,1 Isabelle Barlat, Florence Maurier,1 Fabien Schweighoffer,1 and Bruno Tocque1 Gene Medicine Department, Rhoˆne-Poulenc Rorer, 13 quai Jules Guesde, 94403 Vitry sur Seine Cedex, France
Received February 25, 1998
Sam68 is the major tyrosine-phosphorylated and Srcassociated protein in mitotic cells. Sam68 stimulates G1/S transition and this effect is dependent on the integrity of its KH domain (hnRNPK Homology) which confers nucleic acid binding properties. During mitosis, Sam68 undergoes tyrosine phosphorylation, which negatively regulates its nucleic acid binding properties and mediates the interaction of Sam68 with critical SH2containing signaling proteins such as Grb2, PLCg1 and Ras-GAP. However, the interaction of Ras-GAP with Sam68 has been brought into question, based on the lack of co-immunoprecipitation between Sam68 and Ras-GAP in interphase cells. Here we show that the choice of anti-Ras-GAP antibodies is critical for the detection of Ras-GAP/Sam68 complex formation, and that this interaction is specific for G2/M transition in both NIH3T3 and Src-transformed cells. Such data reinforce the importance of the interaction of Ras-GAP with RNA binding proteins during cell proliferation through its SH2 and SH3 domains. q 1998 Academic Press
Sam68 is the main tyrosine-phosphorylated and Srcassociated protein in mitotic cells (1-5). Sam68 is a KHcontaining RNA binding protein (6,7) and has been shown to bind single strand nucleic acid (8,9). Cloning of an isoform of Sam68 deleted in the KH domain, with decreased RNA binding properties in vitro, brought the first evidence of an impact of this member of the hnRNPK protein family in the regulation of G1/S transition (10). During mitosis, Sam68 undergoes tyrosine phosphorylation which negatively regulates its nucleic acid binding properties (9,11) and interacts through both SH2- and SH3 mediated interactions with signaling molecules such as Grb2, PLCg1 and the Src tyrosine kinase family (12,13). We document in this report 1 Present address: BioScreen Therapeutics, 58, Boulevard Saint Denis, 92400 Courbevoie, France. The abbreviations used are: FCS, foetal calf serum; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; FACS, fluorescence-activated cell sorter; BrdUrd, Bromodeoxyuridine.
the interaction of Sam68 with Ras-GAP which may reflect an impact of this hnRNP on the Ras-GAP transduction pathway, or, alternatively, may represent an additional role for Ras-GAP in G2/M transition. MATERIALS AND METHODS Antibodies. The anti-Sam68 monoclonal antibody, directed against the KH domain (anti-KH antibody) was purchased from Transduction Laboratories. The anti-phosphotyrosine antibody was purchased from Upstate Biotechnology Inc. The anti-Ras-GAP antibodies are both monoclonal antibodies (MAb); the B4F8 MAb is directed against the SH3 domain of Ras-GAP and was obtained from Upstate Biotechnology Inc.; the 15F8 MAb was obtained in mice by immunization with the COOH-Ras-GAP terminal domain (residues 702 to 1044) (Biocytex, Marseille, France) and purified from ascitic fluids by protein A affinity chromatography. The antibody against myc-tag is the 9E10 anti-myc antibody. Monoclonal anti-Src antibody is directed against SH3 domain and was purchased from Calbiochem (clone 327). Cell culture. NIH 3T3 and Src-transformed NIH 3T3 (NIH 3T3Src, gift from B. Wasylyk) cells were routinely grown in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO-BRL) supplemented with 10% FCS1 and 2mM L-Glutamine (GIBCO-BRL), and were incubated at 377C in 5% CO2. For some experiments, cells were synchronised at different phases of the cell cycle. NIH 3T3 cells were synchronised in G1 by serum starvation during 24 hrs (in medium containing 0,5% FCS) and growth was then stimulated by 20% serum for 2hrs (5). In NIH 3T3-Src cells, G1 synchronisation was performed by mitotic shake-off after 16hrs nocodazole treatment and replating for 3hrs. Mitotic arrest was obtained in both types of cells with 16hrs nocodazole treatment (40ng/ml, SIGMA). Correct synchronisation of the cells was checked by FACS analysis. BrdUrd (30mM) was added 30min before collection. Cells were stained for DNA content with propidium iodide and for incorporation of BrdUrd with a fluoresceinconjugated anti-BrdUrd antibody. Transient transfections. NIH 3T3 and NIH 3T3-Src cells were transfected with 5mg of myc-tagged Sam68 SV40 expression vector (pSV2myc hump62). Transfection experiments were performed using lipofectamine (Life Technologies, Inc.) and peptide H1, which increases the transfection efficacy by complexing the cDNA (Patent submitted in France 95 01865, WO 9625508-A1), during 4 hrs at 377C. Cells were then incubated during 48 hrs with DMEM containing 10% FCS to ensure optimal expression of the protein. Preparation of cell lysates and immunoprecipitations. Cells were washed three times with PBS containing 1mM sodium orthovanadate (Na3VO4), then lysed for 30min at 47C in 500ml to 1ml of HNTG buffer, consisting of 50mM HEPES, pH7,5, 150mM NaCl, 1mM EGTA, 1mM MgCl2 , 10% glycerol, 1% Triton X-100, phosphatase inhibitors (1mM Na3VO4 and 10mM NaF), and protease inhibitors
562
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(1mg of leupeptin per ml, 1mg of trypsin inhibitor per ml, 1mg of pepstatin A per ml, 2mg of Aprotinin per ml, 10mg of benzamidine per ml, 1mM phenylmethylsulfonyl fluoride, 1mg of antipain per ml, and 1mg of chymostatin per ml). Lysates were cleared by centrifugation at 4000 RPM for 15min. Cell lysates were precleared with 50ml of Protein G-PLUS Agarose (Santa Cruz Biotechnology) for 1hr at 47C. They were then incubated with the appropriate antibodies for 2hrs at 47C. 50ml of Protein G-PLUS Agarose were then added to the immune complexes and incubation was continued overnight at 47C. Western blotting. The immunoprecipitates were washed three times with HNTG buffer, solubilised in Laemmli buffer and denaturated during 5min at 957C. The proteins were resolved on 4-20% SDSPAGE and transferred onto nitrocellulose membranes. The blots were blocked in PBS containing 0,05% Tween and 2% non-fat milk and then incubated overnight with the appropriate antibody in the same buffer. The blots were further incubated with secondary antibodies conjugated to peroxidase, and revealed by Enhanced ChemiLuminescence (Amersham Corp.).
RESULTS AND DISCUSSION In order to identify in vivo partners of Sam68, extracts from mitotic NIH 3T3 (Fig. 1, odd lanes) or NIH 3T3-Src cells (Fig. 1, even lanes) were immunoprecipitated with an anti-Sam68 monoclonal antibody directed against the KH domain (anti-KH antibody) and analysed by immunoblotting. As previously described (2,4), Src was detected by immunoblotting with an antiSrc antibody (Fig.1B, lanes 3 and 4) in these immunoprecipitates. In addition to Src, tyrosine-phosphorylated Sam68-associated proteins, including a 120 kDa protein, were detected (Fig.1A, lanes 1 and 2). Probing of the anti-KH immunoprecipitates with an anti-RasGAP monoclonal antibody (B4F8) revealed a 120 kDa band (Fig. 1B, lanes 1 and 2), indicating that Ras-GAP was present in the immunoprecipitates. To further confirm the Ras-GAP/Sam68 interaction, extracts from NIH 3T3 and NIH 3T3-Src cells synchronised in G2/M (Fig. 2A and Fig. 2B, respectively) were immunoprecipitated with two different antibodies directed against Ras-GAP: the 15F8 antibody which is directed against the catalytic domain of Ras-GAP and the B4F8 monoclonal antibody which is specific for the SH3 domain of Ras-GAP. Immunoprecipitation with the 15F8 antibody and resulting detection by immunoblotting with the KH antibody revealed one endogenous band of 68kDa (Fig. 2A, lane 1). In contrast, such an interaction was not observed when the immunoprecipitation was performed with the B4F8 antibody (Fig. 2A, lane 2). Similar results were obtained using G2/M cellular extracts from Src-transformed NIH 3T3 cells (Fig. 2B) with both the 15F8 (lane 1) and the B4F8 antibodies (lane 2). To further ensure that the 68kDa protein interacting with Ras-GAP in mitotic cells was the product of the hump62 gene, immunoprecipitation with either the 15F8 or B4F8 anti-Ras-GAP antibodies, followed by immunoblotting with the anti-myc 9E10 monoclonal antibody, was performed using mitotic extracts from NIH 3T3 and NIH 3T3-Src cells expressing myc-tagged
FIG. 1. Sam68 interacts with Ras-GAP in both mitotic NIH 3T3 and Src-transformed NIH 3T3 cells. (A) NIH 3T3 (lanes 1 and 3) and NIH 3T3-Src cells (lanes 2 and 4) were arrested in mitosis with 16 hrs nocodazole treatment. After immunoprecipitation (see Materials and Methods) of the cell lysates with an anti-KH antibody, proteins were resolved by 4-20% SDS-PAGE, and the figure represents the Western blot analysis using an anti-phosphotyrosine antibody (lanes 1 and 2). Expression of endogenous Sam68 was checked simultaneously by direct immunoblotting with the anti-KH antibody (lanes 3 and 4). (B) Nature of the phosphoproteins previously detected was determined in both mitotic NIH 3T3 (lanes 1 and 3) and NIH 3T3Src (lanes 2 and 4) cells by immunoprecipitation with the anti-KH antibody. Sam68 and associated proteins were resolved as previously described. The figure represents the immunoblotting with the B4F8 anti-Ras-GAP antibody (lanes 1 and 2), or the anti-src antibody (lanes 3 and 4). Arrows indicate the positions of Ras-GAP and Src proteins.
Sam68 after transient transfections with pSV2myc hump62 expression vector (Fig. 2A, lanes 3 and 4 and Fig. 2B, lanes 3 and 4). These results confirmed that Sam68 can be detected specifically in 15F8 immunoprecipitates and therefore interacts with Ras-GAP in mitotic cells. Ras-GAP immunoprecipitation with both the 15F8 and B4F8 antibodies was checked by probing of the same blots with the B4F8 antibody (Fig. 2A, lanes 5 and 6 ; Fig. 2B, lane 5 and 6). The lack of co-immunoprecipitation with the B4F8 anti-Ras-GAP antibody suggests that conformational environment is critical for complex formation between Ras-GAP and Sam68. Further the epitope of the B4F8 antibody maps to the SH3 region of Ras-GAP. The association between RasGAP and Sam68 may be dependent on tyrosine phosphorylation of Sam68 which could sustain the interaction with the SH2 domains of Ras-GAP. Therefore, it is possible that either the SH3 recognition by the B4F8
563
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ern-blotting, as previously described. In both cases, association of Ras-GAP with Sam68 was observed in G2/ M but not in early G1 cells (Figure 3). Direct implications of our results are that Sam68 and Ras associate together in proliferating cells and that this interaction is restricted to cells undergoing mitosis. Cell fractionation experiments revealed that expression of Sam68 is mainly localized within the nucleus compartment on NIH3T3 cells, whereas Ras-GAP is detectible only within the cytoplasm. It is therefore likely that the formation of the Ras-GAP/Sam68 complex requires the disruption of the nucleus membrane as occurs during mitosis. Sam68 becomes hyperphosphorylated mainly at tyrosine residues during mitosis (1, 2); previous evidence
FIG. 2. A monoclonal antibody directed against the Carboxy-terminal portion of Ras-GAP immunoprecipitates the product of hump62 cDNA. Total lysates of mitotic NIH 3T3 (A) or NIH 3T3-Src (B) cells were immunoprecipitated with either the 15F8 anti-RasGAP antibody, whose antigen binding site lies within the Carboxyterminus of Ras-GAP (lanes 1, 3 and 5), or the B4F8 anti-Ras-GAP antibody, which maps to the SH3 domain of Ras-GAP (lanes 2, 4 and 6). Ras-GAP and associated proteins were resolved as previously described, and Sam68 was detected by immunoblotting with the antiKH antibody (lanes 1 and 2). A similar experiment was performed in mitotic NIH 3T3 and NIH 3T3-src cells transiently transfected with pSV2myc hump62 expression vector. Exogenous myc-tagged Sam68 was detected with the 9E10 anti-myc antibody (lanes 3 and 4). The same immunoprecipitations were immunoblotted with the B4F8 anti-Ras-GAP antibody (lanes 5 and 6). Expression of endogenous Sam68 and exogenous mycSam68 was checked by direct Western blotting with both the anti-KH (lane 7) and 9E10 anti-myc (lane 8) antibodies.
antibody induces a conformational modification which inhibits the interaction of the SH2 domains of Ras-GAP with the phosphotyrosine residues of Sam68 or that both the SH2 and SH3 domains are involved in RasGAP/Sam68 interaction, as has already been described for the Src-tyrosine kinase family. To further confirm that the Ras-GAP/Sam68 complexes are specifically detected in G2/M, co-immunoprecipitations were carried out on synchronised cells, transfected or not with myc-tagged expression vectors. NIH 3T3 or NIH 3T3-Src transformed cells were synchronised in early G1 (by mitotic shake-off) or in G2/M phase (by nocodazole treatment). Cell synchronisation was checked by FACS analysis (data not shown). Proteins (from separate cells extracts) were immunoprecipitated with the 15F8 antibody and detected by west-
FIG. 3. Sam68 interacts with Ras-GAP exclusively in mitosis. NIH 3T3 cells (A) and NIH 3T3-Src cells (B) were either maintained in a proliferating state (P lanes) or synchronised at different stages of the cell cycle. G1 cells (G1 lanes) and mitotic cells (M lanes) were obtained as described in Materials and Methods. Cells were lysed and whole cells extracts were either immunoprecipitated with the 15F8 anti-Ras-GAP antibody (large frames) or directly immunoblotted (small frames). Endogenous Sam68 was detected with the antiKH antibody, while exogenous mycSam68, expressed after transient transfections, was detected with the 9E10 anti-myc antibody. Detection of Sam68 and mycSam68 were realised on separate cells extracts. Arrows indicate the positions of Sam68 and mycSam68.
564
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Model for the regulation of Sam68 function during the cell cycle. The figure illustrates diagrammatically progression through the cell cycle and proposes two mechanisms for the involvement of Sam68 in cell cycle progression. During G1/S transition, Sam68 and Sam68DKH, which are both expressed, have antagonistic effects on cyclin D1 expression, and therefore on cell cycle progression (10). During mitosis, while Sam68DKH is not expressed, Sam68 interacts with Ras-GAP, possibly through a third partner (X) yet to be identified. This interaction may modulate Sam68 function and suggests a role for the Ras-GAP/Sam68 complex in RNA metabolism.
demonstrate that Ras-GAP binds phosphorylated tyrosines through SH2 domains (12). It is reasonable to suggest that the SH2 domains of Ras-GAP may shield tyrosine-phosphorylated Sam68 from the enzymatic attack of regulatory phosphatases. Maintenance of the tyrosine-phosphorylated form of Sam68 may affect the activity of the protein in a number of ways. The ability of Sam68 to bind to specific mRNA molecules may also be affected and it has been previously demonstrated that phosphorylation of Sam68 inhibits its nucleic acid binding properties (9). It is possible that Ras-GAP interaction with Sam68, through stabilization of the tyrosine-phosphorylated form of Sam68, may lead to the release of specific RNA targets, which have to be translated or degraded during mitosis. Alternatively, the degradation of Sam68 may be influenced by maintaining the protein in a highly tyrosine-phosphorylated form. Apart from these suggestions, we cannot exclude the possibility that Ras-GAP modulates some other, as yet unknown, function of Sam68. Interactions between protein partners may be forged or lost depending on the state of Sam68 phosphorylation. Conversely, Sam68 may affect different unidentified functions of Ras-GAP during mitosis. The sole function of Sam68 documented so far is in the promotion of S phase entry through its nucleic acid binding properties (10). On the basis of our new results,
we propose the following scheme to illustrate the impact of Sam68 on cell cycle progression (Figure 4). The positive effect of Sam68 on G1/S transition is inhibited by an isoform of Sam68, Sam68DKH, which exhibits impaired nucleic acid binding properties (10). During S phase, Sam68DKH disappears and Sam68 promotes S phase entry possibly by interacting with its RNA target which has still to be discovered. It has been reported that cyclin D1 expression is regulated by the balance of expression between Sam68 and Sam68DKH (10), implicating cyclin D1 on a pathway elicited by Sam68 during G1/S transition. An alternative pathway involving the regulatory interactions between Sam68 and different and as yet unknown protein partners is also portrayed. There is no evidence at present to suggest a functional role for Sam68 during interphase. During mitosis, tyrosine phosphorylation of Sam68 may occur, due to the disruption of the nucleus membrane, which allows interaction with protein partners such as Ras-GAP. Although we cannot exclude this interaction takes place through a third protein partner. Our data documents the interaction of Ras-GAP with Sam68 during mitosis and indicates a possible role of Ras-GAP in the regulation of RNA binding proteins. It is of interest to note that, in response to growth factor stimulation and thus during cell cycle progression, RasGAP recruits two different nucleic acid binding pro-
565
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC
Vol. 245, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
teins: G3BP, which interacts with the SH3 domain, and Sam68, which appears to be recruited in a SH2dependent manner. Future experiments will aim to decipher the contribution of Ras-GAP to cell cycle progression through its regulation of RNA metabolism. REFERENCES 1. Courtneige, S. A., and Fumagalli, S. (1994) Trends Cell Biol. 4, 345–347. 2. Taylor, S. J., and Shalloway, D. (1994) Nature 368, 867–871. 3. Weng, Z., Thomas, S. M., Rickles, R. J., Taylor, J. A., Brauer, A. W., Seidel-Dugan, C., Michael, W. M., Dreyfuss, G., and Brugge, J. S. (1994) Mol. Cell. Biol. 14, 4509–4521. 4. Fumagalli, S., Totty, N. F., Hsuan, J. J., and Courtneidge, S. A. (1994) Nature 368, 871–874. 5. Lock, P., Fumagalli, S., Polakis, P., McCormick, F., and Courtneidge, S. A. (1997) Cell 84, 23–24.
6. Gibson, T. J., Thompson, J. D., Heringa, J. (1993) FEBS Lett. 324, 361–366. 7. Kiledjian, M., and Dreyfuss, G. (1992) EMBO. J. 11, 2655–2664. 8.Wong, G., Muller, O., Clark, R., Conroy, L., Moran, M. F., Polakis, P., and McCormick, F. (1992) Cell 69, 551–558. 9. Wang, L. L., Richard, S., and Shaw, A. S. (1995) J. Biol. Chem. 270, 2010–2013. 10. Barlat, I., Maurier, F., Duchesne, M., Guitard, E., Tocque, B., and Schweighoffer, F. (1997) J. Biol. Chem. 272, 3129–3132. 11. Taylor, S. J., Anafi, M., Pawson, T., and Shalloway, D. (1995) J. Biol. Chem. 270, 10120–10124. 12. Richard, S., Yu, D., Blumer, K. J., Hausladen, D., Olszowy, M. W., Connely, P. A., and Shaw, A. S., (1995) Mol. Cell. Biol. 15, 186–197. 13. Neet, K., and Hunter, T. (1995) Mol. Cell. Biol. 15, 4908–4920. 14. Parker, F., Maurier, F., Delumeau, I., Duchesne, M., Faucher, D., Debussche, L., Dugue´, A., Schweighoffer, F., and Tocque´, B. (1996) Mol. Cell. Biol. 16, 2561–2569. 15. Birge, R. B., Fajardo, J. E., Mayer, B. J., Hanafusa, H. (1992) J. Biol. Chem. 267, 10588–10595.
566
AID
BBRC 8374
/
694f$$$941
04-02-98 17:54:13
bbrcg
AP: BBRC