- Email: [email protected]
Cell-Cycle Dependent Tyrosine Phosphorylation on Mortalin Regulates Its Interaction with Fibroblast Growth Factor-1
Biochemical and Biophysical Research Communications 280, 1203–1209 (2001) doi:10.1006/bbrc.2001.4225, available online at http://www.idealibrary.com on
Cell-Cycle Dependent Tyrosine Phosphorylation on Mortalin Regulates Its Interaction with Fibroblast Growth Factor-1 Eiichi Mizukoshi,* ,† Masashi Suzuki,* Tomoko Misono,* Alexei Loupatov,* Eisuke Munekata,† Sunil C. Kaul,* Renu Wadhwa,‡ and Toru Imamura* ,1 *National Institute of Bioscience and Human Technology, AIST, Tsukuba, Ibaraki 305-8566, Japan; †Institute of Applied Biochemistry, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan; and ‡Chugai Research Institute for Molecular Medicine, Inc., Niihari, Ibaraki 300-4101, Japan
Received November 24, 2000
We previously reported that endogenously expressed, intracellularly localized fibroblast growth factor (FGF)-1 interacts with mortalin. Here we report that FGF-1 added to the culture medium of quiescent BALB/c3T3 cells is taken up by the cells and interacts with mortalin in the cells in a regulated manner. Although both the internalized FGF-1 and mortalin were present at high levels throughout the FGF-1-initiated cell cycle, their interaction became apparent only in late G1 phase. Interestingly, mortalin was preferentially tyrosine phosphorylated at the same time, and when its normally weak phosphorylation in early G1 phase was augmented by treating the cells with vanadate, a strong interaction between mortalin and FGF-1 was established. Conversely, when phosphorylated mortalin was treated with tyrosine phosphatase, its interaction with FGF-1 was abrogated. These results indicate that FGF-1 taken up by cells preferentially interacts with mortalin in late G1 phase of the cell cycle, and that tyrosine phosphorylation of mortalin regulates this interaction. © 2001 Academic Press Key Words: fibroblast growth factor; FGF; mortalin; GRP75; internalization; interaction; tyrosine phosphorylation; cell-cycle.
The fibroblast growth factor (FGF) family of proteins is composed of over 20 members, including FGF homologous factors, that control such cellular processes as growth, differentiation, migration, and survival (1– 6). FGFs stimulate these biological responses by binding to surface receptor tyrosine kinases (FGFRs) and heparan sulfate sugar chains, thereby activating intracel1 To whom correspondence should be addressed at Biosignaling Department, National Institute of Bioscience and Human Technology, AIST, 1-1 Higashi, Ibaraki 305-8566, Japan. Fax: ⫹81-298-616149. E-mail: [email protected].
lular signaling cascades (7–9). The intracellular molecules involved in these signaling pathways include phospholipase gamma, FRS2/Snt, Grb-2, Shc, and MAPK (10 –13); however, it is now clear that the immediate-early events occurring downstream of the FGFR are not sufficient to exert the mitogenic activity of FGF-1 (14 –15). Instead, after binding to cell-surface FGFRs, receptor-bound FGF-1 is transported across the plasma membrane via an as yet unidentified mechanism, and is deposited in the cytosol where it is then able to enter the nucleus (14 –20); the mitogenic activity of FGF-1 correlates closely with its appearance in the cytosol and with the presence of a nuclear localization sequence in its primary structure (14 –15). We and others have reported that delivery of the FGF-1 nuclear localization signal into living cells stimulates DNA synthesis (21–22), while the ability of FGF-1 to induce DNA synthesis requires its continuous presence in the cytosol and nucleus during the entire G1 phase (17, 20). Thus, internalized FGF-1 likely acts in a cooperative and synergetic fashion with receptortriggered intracellular processes to exert its full mitogenic activity (20). Increasing evidence indicates that intracellular FGF-1 binds to several proteins. FGF-1 intracellular binding protein (FIBP) interacts with FGF-1 and may serve to shuttle FGF-1 between the cytosol and nucleus (23), while synaptotagmin-1 and S100A13 also aggregate with intracellular FGF-1 (24 –25). Other FGF family ligands interact with other intracellular proteins. For example, FGF-2 bound to Casein kinase II and might regulate nuclear protein and exert its activity in nucleus (26). Intracellular proteins, including p79 and Cystein-rich FGF receptor (ESL), form complex with FGF-2 (27). FGF-3 was coimmunoprecipitated with karyopherin alpha, a nuclear localization receptor (28).
1203
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
We previously identified mortalin/glucose-regulated protein (GRP)75/mitochondrial hsp (mthsp)70/PBP-74 as an intracellular FGF-1-binding protein (29). As its various names suggest, this molecule appears to be involved in diverse regulatory processes including cellular senescence, glucose regulation, mitochondrial transport, nephrotoxicity, and antigen processing (30 – 34). While heat shock proteins (HSPs) have been generally considered to function in the folding, assembly and degradation of newly synthesized proteins (35), increasing evidence indicates that they also associate with regulatory proteins such as receptor kinases and transcription factors (36). Hsp27 and Hsp90, moreover, are phosphorylated on Ser/Thr residues (36 –37), while Hsp75 and Heat shock cognate protein 70 undergo acute tyrosine phosphorylation in response to vanadate and hydrogen peroxide (38 –39). These results suggest that HSP proteins may be involved in signaling pathway in regulated manners. In the present study, we examined the interaction between mortalin and FGF-1 taken up from culture medium by BALB/ c3T3 cells with the aim of better understanding the regulation of mortalin-FGF-1 complex formation. MATERIALS AND METHODS Reagents. The expression and purification of human recombinant FGF-1 and the production of an anti-mortalin polyclonal antibody and anti-FGF-1 monoclonal antibodies (mAb1, mAb15) have all been described previously (15, 17, 30). Anti-mthsp70 monoclonal antibody was purchased from Affinity Bioreagents (Golden, CA). Anti-FGF-1 polyclonal antibody, anti-phosphotyrosine monoclonal antibody (PT66), and rabbit kidney-derived protein tyrosine phosphatase were from Sigma Chemical (St. Louis, MO). Anti-phosphotyrosine monoclonal antibody (4G10) was from Upstate Biotechnology (Lake Placid, NY). Anti-cyclin D1 monoclonal antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). Alexa Fluor 488 goat antimouse IgG (H⫹L) conjugate was from Molecular Probes (Eugene, OR). Texas Red-conjugated donkey anti-rabbit IgG antibody was from Amersham Pharmacia Biotechnology (Uppsala, Sweden). Cell culture. BALB/c3T3 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in Dulbecco’s modified essential medium (DMEM, Sigma, St. Louis, MO) supplemented with 10% calf serum (CS, Hyclone Laboratories, Logan, UT). DNA synthesis assay. To assay the kinetics of DNA synthesis, BALB/c3T3 cells were plated to a density of 7.9 ⫻ 10 4 cells/cm 2 in 48-well plates (Sumilon, Tokyo) and cultured for 24 h in DMEM with 10% CS. The medium was then replaced with DMEM supplemented with 100 nM insulin, and the cells incubated for an additional 48 h before stimulation with FGF-1. DNA synthesis was initiated by addition of recombinant human FGF-1 (100 ng/ml) and heparin (50 g/ml). The cultures were then pulse-labeled for 2 h with [ 3H]thymidine (2 Ci/ml; Moravek Biochemicals, CA), and at the end of labeling period DNA synthesis was determined as described previously (21). Immunoblot analysis. Samples resolved by SDS–polyacrylamide gel electrophoresis (SDS–PAGE) were electrotransferred onto nitrocellulose membranes. The membranes were then blocked overnight at 4°C in buffer containing 5% skim milk, 150 mM NaCl, 10 mM Tris–HCl pH 7.4 (TBS), incubated for 60 min at room temperature with the appropriate antibody, washed for 20 min with 5% skim milk
in TBS containing 0.1% Tween 20 (TBS-T), and incubated with peroxidase-labeled goat anti-mouse IgG (Chemicon, Temecula, CA) appropriately diluted with TBS-T buffer. The membranes were then washed with TBS-T four times, and the signals were visualized using enhanced chemiluminescence (ECL, Amersham life science, Buckinghamshire, England) and Kodak X-O mat AR film (Kodak, Rochester, NY). Immunoprecipitation. Quiescent monolayers of BALB/c3T3 cells were treated with recombinant FGF-1 (20 ng/ml) and heparin (10 g/ml). At the times indicated, the cells were washed three times with phosphate-buffered saline (PBS), harvested using a cell scraper and lysed for 20 min on ice in 1 ml of lysis buffer (0.5% Triton X-100, 0.1% sodium deoxycolate, 150 mM NaCl, 10 mM Tris–HCl, pH 7.4, 1 mM EDTA, 50 mM NaF, Protease Inhibitor Cocktail (Boehringer Manheim, Manheim, Germany), 1 mM sodium orthovanadate and phosphatase inhibitor cocktail (Sigma)). The lysates were clarified by centrifugation for 10 min at 12,000g and incubated for 1 h at 4°C with the appropriate antibodies. Immunocomplexes were precipitated with protein A⫹G Sepharose (CALBIOCHEM, La Jolla, CA) and then washed three times with lysis buffer. The immunoprecipitated proteins were resolved by SDS–PAGE and analyzed by immunoblotting. Immunofluorescence analysis. BALB/c3T3 cells plated on coverslips and maintained in DMEM supplemented with 100 nM insulin were stimulated with FGF-1 (100 ng/ml) and heparin (50 g/ml) for the indicated times, after which they were fixed for 20 min at room temperature with 4% paraformaldehyde in PBS and permeabilized for 10 min with 0.2% Triton X-100 in PBS. The coverslips were then blocked for 30 min at 37°C in PBS containing 6% BSA, incubated with polyclonal anti-mortalin antibody, and either monoclonal antiFGF-1 or anti-phosphotyrosine antibody, washed and double-labeled with Alexa Fluor-conjugated anti-mouse IgG and Texas Redconjugated anti-rabbit antibodies. The coverslips were then mounted in FA mounting fluid (DIFCO LABORATORIES, Detroit, MI), and confocal microscopy was carried out using Carl Zeiss (Jena, Germany) equipment as described previously (29). In vitro binding assay. Serum-starved BALB/c3T3 cells were stimulated with FGF-1 (20 ng/ml) and heparin (10 g/ml) for 12 h and then lysed. Mortalin in the cell lysate was then exposed to anti-mortalin antibody, and the resultant immunocomplexes were washed first with high salt buffer (50 mM Tris–HCl, pH 7.4, 0.5 M NaCl) and then with phosphatase buffer (50 mM Tris–HCl pH 7.4), after which they were incubated for 30 min at 37°C with tyrosine phosphatase (0.1 units) in phosphatase buffer. After washing extensively with binding buffer (50 mM Tris–HCl pH 7.4, 150 mM NaCl), the immunocomplexes were incubated for 1 h in binding buffer containing FGF-1 (200 ng/ml) and BSA (200 g/ml), and washed again with binding buffer. The bound proteins were resolved by SDS–PAGE and then subjected to immunoblotting using antiFGF-1, anti-phosphotyrosine and anti-mortalin antibodies as probes.
RESULTS FGF-1 Protein Taken Up by BALB/c3T3 Cells Interacts with Endogenous Mortalin in a Cell Cycle-Dependent Manner In an earlier study we identified mouse mortalin and GRP75 (a rat homolog of mouse mortalin) as intracellular proteins that bind to endogenously-expressed, intracellularly-localized FGF-1 (29). And as FGF-1 added to culture medium is known to be taken up by the cells (14 –20), in the present study we investigated whether exogenously-applied FGF-1 would also bind mortalin. For this purpose, we used BALB/c3T3 cells,
1204
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Cell-cycle dependent interaction of FGF-1 with mortalin in BALB/c3T3 cells. (A) Time course of the interaction of FGF-1 with mortalin in BALB/c3T3 cells. Quiescent BALB/c3T3 cells were stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml), then harvested and lysed at the indicated times. Cell lysates were immunoprecipitated with anti-FGF-1 polyclonal antibody. Immunocomplexes were separated on 7.5% SDS–PAGE and analyzed by immunoblotting with an anti-mthsp70 antibody that recognizes both mortalin variants (mot-1 and mot-2). (B) Time course of internalization of FGF-1. Immunocomplexes obtained in (A) were analyzed by immunoblotting with anti-FGF-1 monoclonal antibody (mAb1). (C) Time course of mortalin expression in BALB/c3T3 cells. The amount of mortalin in the cell lysates obtained in (A) were examined by immunoblotting with anti-mthsp70 antibody. (D) Colocalization of internalized FGF-1 and mortalin. Quiescent BALB/c3T3 cells were stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml). After 12 h of stimulation, cells were fixed and labeled with polyclonal anti-mortalin and anti-FGF-1 (mAb15) antibodies and corresponding Texas red- and Alexa Fluor-conjugated secondary antibodies. Shown are confocal images of immunofluorescence depicting the distributions of mortalin (left panel), FGF-1 (middle panel), and the superimposition of the two (right panel). (E) Time course of cyclin D1 expression. Serum-starved cells were stimulated at the indicated times with FGF-1 (100 ng/ml) and harvested. Cyclin D1 expression was determined by immunoblotting with anti-cyclin D1 antibody. (F) Kinetics of DNA synthesis in BALB/c3T3 cells. Cells were serum-starved for 48 h, stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml) for 2–32 h, and labeled with [ 3H]thymidine during the final 2 h of the stimulation period. The incorporated radioactivity was determined with the mean ⫹/⫺ SE of triplicate cultures.
which express moderate levels of mortalin but not FGF-1. Once the cells were made quiescent by serum starvation, FGF-1 (100 ng/ml) was added to the culture medium for 12 h, after which the cells were solubilized, and the cell-associated FGF-1 was immunoprecipitated using a specific antibody. As shown in Fig. 1A, FGF-1 was indeed immunoprecipitated as part of a complex
with mortalin. Interestingly, however, though levels of cell-associated FGF-1 were maximal within 30 min of its addition (Fig. 1B) in good agreement with the results in our earlier study (17), and levels of mortalin remained constant throughout the FGF-1-initiated cell cycle (Fig. 1C), formation of mortalin-FGF-1 complexes developed gradually over the 12-h stimulation period
1205
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Double-immunolabeling with anti-mortalin and anti-phosphotyrosine antibodies of the cells after 12 h of FGF-1 stimulation revealed that there was substantial overlap between the distributions of mortalin and phosphotyrosine-containing proteins in the juxtanuclear region (results not shown). Inhibition of Phosphatase Augments the Interaction between Mortalin and FGF-1 FIG. 2. Cell-cycle dependent tyrosine-phosphorylation of mortalin in BALB/c3T3 cells. (A) Time course of tyrosinephosphorylation of mortalin in BALB/c3T3 cells. Quiescent BALB/ c3T3 cells were treated with FGF-1 (20 ng/ml) and heparin (10 g/ml) for the indicated times. Cell lysates were then prepared and incubated with anti-phosphotyrosine antibody (PT66). The immunoprecipitates were analyzed by immunoblotting with anti-mthsp70 antibody. (B) Time course of tyrosine-phosphorylation of mortalin in BALB/c3T3 cells. Cell lysates prepared in (A) were precipitated with anti-mthsp70 antibody. The immunocomplexes were then analyzed by immunoblotting with anti-phosphotyrosine antibody (4G10). Mortalin is indicated by an arrow. (C) Mortalin levels in BALB/c3T3 cells treated with FGF-1 for various times. The immunoprecipitates in (B) were probed with anti-mthsp70 antibody to determine mortalin levels.
(Fig. 1A). Moreover, when FGF-1 and mortalin were localized immunohistochemically by labeling FGF-1stimulated (12 h) cells with anti-FGF-1 and antimortalin antibodies, considerable overlap between the FGF-1 and mortalin signals was observed in the juxtanuclear region (Fig. 1D). Because the interaction profile suggested a possible correlation between the interaction of FGF-1 and mortalin and cell cycle progression, we monitored the cell cycle by analyzing cyclin D1 expression and DNA synthesis. As shown in Fig. 1E, cyclin D1 expression peaked after about 16 h FGF-1 stimulation, and S phase, as determined by DNA synthesis, occurred after 20 –24 h of stimulation (Fig. 1F). The interaction of internalized FGF-1 and endogenous mortalin thus appears to correlate with the late G1 and early S phases of the cell cycle. Mortalin Is Tyrosine Phosphorylated during Late G1-S Phase of the FGF-1-Initiated Cell Cycle The cell cycle-dependent interaction of FGF-1 and mortalin suggested the occurrence of cell cycledependent, post-translational modification of these proteins. And since tyrosine-phosphorylation of cellular proteins is reported to increase in late G1 phase, we analyzed the proteins and found that murine mortalin was indeed tyrosine phosphorylated (Figs. 2A and 2B), though FGF-1 was not (data not shown). The level of mortalin phosphorylation increased with progression of the cell cycle, peaking 12 h after the onset of G1 (Figs. 2A and 2B), while expression of mortalin remained constant (Fig. 2C).
The results described so far suggest that the interaction between mortalin and internalized FGF-1 is regulated by tyrosine phosphorylation of the former. We therefore examined the extent to which modulation of mortalin tyrosine phosphorylation affected formation of the mortalin-FGF-1 complex. We first tested the effect of sodium orthovanadate, a phosphatase inhibitor, and found that inhibition of phosphatase activity significantly augmented tyrosine phosphorylation of mortalin (Fig. 3A). Moreover, treating cells with vanadate significantly increased the interaction of mortalin with FGF-1 in early to middle G1 phase—i.e., after 1, 2 and 4 h of FGF-1 stimulation (Fig. 3B). For example, mortalin-FGF-1 binding after 1 h of stimulation was increased approximately 4.5-fold in the presence of vanadate (Fig. 3B). By contrast, vanadate had no effect after 8 h of stimulation, when substantial tyrosine phosphorylation of mortalin was observed even in its absence (Figs. 3A and 3B). Uptake of FGF-1 by the cells was unaffected by vanadate, confirming that the increased interaction between mortalin and FGF-1 was not the result of a vanadate-induced increase in the availability of FGF-1 (Fig. 3B). Treatment of Phosphorylated Mortalin with Tyrosine Phosphatase Abrogates Complex Formation Our finding that an inhibitor of tyrosine phosphatase augmented the interaction between FGF-1 and mortalin in early G1 phase strongly suggests the involvement of phosphotyrosine in the formation of the mortalin-FGF-1 complex. This was further confirmed by examining the effect of tyrosine phosphatase. After stimulating cells with FGF-1 for 12 h, the tyrosinephosphorylated mortalin was extracted, digested with tyrosine phosphatase, and analyzed for its interaction with FGF-1 in vitro. Phosphatase treatment diminished the level of mortalin phosphorylation, as well as its interaction with FGF-1, though mortalin itself was unaffected (Fig. 4A). Analysis of the dose dependency and time course for the enzyme digestion revealed that treating cell extracts with 0.1 U of tyrosine phosphatase for 30 min (Fig. 4B, left panels) or with 0.2 U for 5 min (Fig. 4B, right panels) was sufficient to significantly diminish tyrosine phosphorylation and formation of mortalin-FGF-1 complexes.
1206
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
gether with our earlier observations (29), these findings indicate that tyrosine phosphorylation of mortalin promotes, but is not essential for, the interaction of mortalin with FGF-1. Nevertheless, as it enhances the interaction by more than fourfold, mortalin phosphorylation should be the determining factor regulating the degree of interaction. Mortalin represents two variants (mot-1 and mot-2) of a 75 kDa protein, both of which are expressed by BALB/c3T3 cells and interact with FGF-1. In the present study we used an antibody against mthsp75 that recognizes both mortalin variants to demonstrate the cell cycle-dependency of mortalin tyrosine phosphorylation for the first time. There is ample precedent for the phosphorylation of mortalin and related proteins. For example, a 75 kDa protein similar to mortalin/GRP75/mthsp70/PBP74 undergoes tyrosine phosphorylation when cells are treated with hydrogen peroxide and vanadate (38), and rat liver mthsp70
FIG. 3. Sodium orthovanadate enhances FGF-1-induced tyrosine-phosphorylation of mortalin and interaction of FGF-1 with mortalin. (A) Enhanced tyrosine-phosphorylation of mortalin by sodium orthovanadate (Va). Quiescent BALB/c3T3 cells were cultured for 1 h in the presence or absence of sodium orthovanadate (50 M), after which the cultures were treated with FGF-1 (20 ng/ml) and heparin (10 g/ml). Lysates were prepared at the indicated times and then incubated with anti-mthsp70 antibody. Immunocomplexes were analyzed by immunoblotting with anti-phosphotyrosine (4G10) (upper panels) or anti-mthsp70 (lower panels) antibodies. (B) Enhanced mortalin-FGF-1 binding evoked by sodium orthovanadate. Lysates prepared as described in (A) were incubated with anti-FGF-1 polyclonal antibody. The resultant precipitates were examined by immunoblotting with anti-mthsp70 antibody (upper panels). The amount of internalized FGF-1 in each sample was determined by immunoblotting with anti-FGF-1 antibody (mAb1) (lower panels). Relative levels of FGF-1-mortalin binding in the presence (F) or absence (E) of vanadate were determined by measuring the optical density of the bands and normalizing the results to the values obtained at 8 h, which were assigned a value 1.0.
DISCUSSION We have shown that FGF-1 present in culture medium is taken up by BALB/c3T3 cells, where it interacts with intracellular mortalin, just as endogenouslyexpressed, intracellularly-localized FGF-1 interacts with GRP75 (a rat homolog of mortalin) in rat L6 and Rat-1 cells (29). This interaction correlated with the level of mortalin tyrosine phosphorylation, which peaked in late G1 phase, suggesting that cell cycledependent tyrosine phosphorylation of mortalin regulates formation of the mortalin-FGF-1 complex. To-
FIG. 4. Tyrosine phosphatase decreases the interaction of FGF-1 with mortalin in vitro. (A) Mortalin in lysates of quiescent (0 h) or FGF-1-stimulated (12 h) BALB/c3T3 cells was immobilized on protein A ⫹ G Sepharose beads with anti-mthsp70 antibody. The immunocomplexes were then washed with high salt buffer (0.5 M NaCl, 50 mM Tris–HCl pH 7.4) and incubated for 30 min with tyrosine phosphatase (0.2 U), after which the beads were incubated with FGF-1 (200 ng/ml) in binding buffer (150 mM NaCl, 50 mM Tris–HCl pH 7.4) containing BSA (200 g/ml), and washed. Bound proteins were separated on SDS–PAGE and analyzed by immunoblotting with anti-FGF-1 (mAb1) (upper panel), anti-phosphotyrosine (4G10) (middle panel), or anti-mthsp70/mortalin (lower panel) antibody. (B) Immunocomplex prepared as described in (A) were incubated for 30 min with selected amounts of tyrosine phosphatase (left panels) or with fixed amount of enzyme (0.2 U) for various periods (right panels). FGF-1-binding assays were then carried out as described in (A).
1207
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
exhibits Ca 2⫹-dependent autophosphorylation (34), suggesting mortalin’s involvement in a variety of diverse signaling pathways. Indeed mortalin is known to associate with the IL-1 receptor and may affect the accessibility of IL-1 receptor associated kinase (IRAK) substrates (40). As a member of HSP 70 family, mortalin appears to play significant roles in regulating various signaling cascades, including those mediating cell senescence (30, 41). Because only critical regulatory molecules are tyrosine phosphorylated, our findings that mortalin is tyrosine phosphorylated in late G1-phase and that the phosphorylation affects its interaction with FGF-1 suggests formation of the mortalin-FGF-1 complex is crucial for the transduction of FGF-1-induced cell growth and differentiation. One possible function of the mortalin-FGF-1 complex is suggested by its distribution within cells. Confocal immunofluorescence microscopy showed that mortalin and FGF-1 colocalize in juxtanuclear regions, and that this colocalization is apparent 12 h after the onset of G1 phase. Interestingly, the juxtanuclear distribution of mortalin resembles the distributions of tyrosinephosphorylated FGFR-1 (42) and FGFR-4 (16); mortalin may thus regulate the availability of FGF-1 to FGFRs. It has been reported that not only immediate early events but also so-called late events, taking place in the middle to late G1 phase, are induced following growth factor binding to its cognate tyrosine kinase receptor (10 –13, 20). Apparently these late events are crucial for inducing cells to enter S phase and complete the cell cycle (20, 43). Consequently, short term exposure to FGF-1 is not sufficient to establish DNA synthesis, though it does induce cell migration (43). The late events mediating the mitogenic activity of FGF-1 may include nuclear localization of FGF-1 (17, 20) and tyrosine hyperphosphorylation of several cellular proteins, including cortactin and Src (6, 20, 42– 43). Furthermore, continuous uptake of FGF-1 into the cytosol and nucleus enables cells to progress through the cell cycle (17, 20); thus regulated association of mortalin with FGF-1 may be a part of late events and modulate this as yet uncharacterized activity of cytosolic and/or nuclear FGF-1. ACKNOWLEDGMENTS We thank Drs. Ai-Jun Li and Akira Ishisaki at NIBH for confocal image analysis and discussion. The text was edited by Dr. William Goldman at MST Editing Company. This study was supported by a AIST grant to T.I.
REFERENCES 1. Baird, A., and Klagsbrun, M. (1991) Cancer Cells 3, 239 –243. 2. Emoto, H., Tagashira, S., Mattei, M. G., Yamasaki, M., Hashimoto, G., Katsumata, T., Negoro, T., Nakatsuka, M., Birnbaum, D., Coulier, F., and Itoh, N. (1997) J. Biol. Chem. 272, 23191– 23194.
3. Smallwood, P. M., Munoz-Sanjuan, I., Tong, P., Macke, J. P., Hendry, S. H., Gilbert, D. J., Copeland, N. G., Jenkins, N. A., and Nathans, J. (1996) Proc. Natl. Acad. Sci. USA 93, 9850 –9857. 4. Mason, I. J. (1994) Cell 78, 547–552. 5. Fiesel, R., and Maiciag, T. (1995) FASEB J. 9, 919 –925. 6. Liu, J., Huang, C., and Zhan, X. (1999) Oncogene 18, 6700 – 6706. 7. Mckeehan, W. L., Wang, F., and Kan, M. (1998) Prog. Nucleic Acid Res. Mol. Biol. 59, 135–176. 8. Imamura, T., Tokita, Y., and Mitsui, Y. (1988) Biochem. Biophys. Res. Commun. 155, 583–590. 9. DiGabriele, A. D., Lax, I., Chen, D. I., Svahn, C. M, Jaye, M., Schlessinger, J., and Hendrickson, W. A. (1998) Nature 393, 812– 817. 10. Wang, J. K., Gao, G., and Goldfarb, M. (1994) Mol. Cell. Biol. 14, 181–188. 11. Wang, J. K., Xu, H., Li, H. C., and Goldfarb, M. (1996) Oncogene 13, 721- 729. 12. 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. 11, 5068 –5078. 13. Kouhara, H., Hadari, Y. R., Spivak-Kroizman, T., Schilling, J., Bar-Sagi, D., Lax, I., and Schlessinger, J. (1997) Cell 89, 693– 702. 14. Wiedlocha, A., Falnes, P. O., Madshus, I. H., Sandvig, K., and Olsnes, S. (1994) Cell 76, 1039 –1051. 15. Imamura, T., Engleka, K., Zhan, X., Tokita, Y., Forough, R., Roeder, D., Jackson, A., Maier, J. A., Hla, T., and Maciag, T. (1990) Science 249, 1567–1570. 16. Citores, L., Wesche, J., Kolpakova, E., and Olsnes, S. (1999) Mol. Biol. Cell 10, 3835–3848. 17. Imamura, T., Oka, S., Tanahashi, T., and Okita, Y. (1994) Exp. Cell Res. 215, 363–372. 18. Zhan, X., Hu, X., Friedman, S., and Maciag, T. (1992) Biochem. Biophys. Res. Commun. 188, 982–991. 19. Klingenberg, O., Wiedolocha, A., Citores, L., and Olsnes, S. (2000) J. Biol. Chem. 275, 11972–11980. 20. Zhan, X., Hu, X., Friesel, R., and Maciag, T. (1993) J. Biol. Chem. 268, 9611–9620. 21. Komi, A., Suzuki, M., and Imamura, T. (1998) Exp. Cell Res. 243, 408 – 414. 22. Lin, Y. Z., Yao, S. Y., and Hawiger, J. (1996) J. Biol. Chem. 271, 5305–5308. 23. Kolpakova, E., Wiedolocha, A., Stenmark, H., Klingenberg, O., Falsnes, P. O., and Olsnes, S. (1998) Biochem. J. 336, 213–222. 24. Tarantini, F., LaVallee, T. M., Jackson, A., Gamble, S., Carreira, M. C., Garfinkel, S., Burgess, W. H., and Maciag, T. (1998) J. Biol. Chem. 273, 22209 –22216. 25. Carreira, M. C., LaValee, T. M., Tarantini, F., Jackson, A., Lathrop, J. T., Hampton, B., Burgess, W. H., and Maciag, T. (1998) J. Biol. Chem. 273, 22224 –22231. 26. Bonnet, H., Filhol, O., Truchet, I., Brethenou, P., Cochet, C., Amalric, F., and Bouche, G. (1996) J. Biol. Chem. 271, 24781– 24787. 27. Chevet, E., Lemaitre, G., Cailleret, K., Dahan, S., Bergeron, J. J., and Katinka, M. D. (1999) Biochem. J. 341, 713–723. 28. Antoine, M., Reimers, K., Dickson, C., and Kiefer, P. (1997) J. Biol. Chem. 272, 29475–29481. 29. Mizukoshi, E., Suzuki, M., Loupatov, A., Uruno, T., Hayashi, H., Misono, T., Kaul, S. C., Wadhwa, R., and Imamura, T (1999) Biochem. J. 343, 461– 466. 30. Wadhwa, R., Kaul, S. C., Ikawa, Y., and Sugimoto, Y. (1993) J. Biol. Chem. 268, 6615– 6621.
1208
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
31. Massa, S. M., Longo, F. M., Zuo, J., Wang, S., Chen, J., and Sharp, F. R. (1995) J. Neurosci. Res. 40, 807– 819. 32. Bhattacharyya, T., Karnezis, A. N., Murphy, S. P., Hoang, T., Freeman, B. C., Phillips, B., and Morimoto, R. I. (1995) J. Biol. Chem. 270, 1705–1710. 33. Domanico, S. Z., DeNagel, D. C., Dahlseid, J. N., Green, J. M., and Pierce, S. K. (1993) Mol. Cell. Biol. 13, 3598 –3610. 34. Webster, T. J., Naylor, D. J., Hartman, D. J., Hoj, P. B., and Hoogenraad, N. J (1994) DNA Cell Biol. 13, 1213–1220. 35. Martin, J., and Hartl, F. U. (1997) Curr. Opin. Struct. Biol. 7, 41–52. 36. Rutherford, S. L., and Zuker, C. S. (1994) Cell 79, 1129 –1132. 37. Rousseau, S., Houle, F., Landry, J., and Huot, J. (1997) Oncogene 15, 2169 –2177.
38. Hadari, Y. R., Haring, H. U., and Zick, Y. (1997) J. Biol. Chem. 272, 657– 662. 39. Egerton, M., Moritz, R. L., Druker, B., Kelso, A., and Simpson, R. J. (1996) Biochem. Biophys. Res. Commun. 224, 666 – 674. 40. Sacht, G., Brigelius-Flohe, R., Kiess, M., Sztajer, H., and Flohe, L. (1999) Biofactors 9, 49 – 60. 41. Wadhwa, R., Kaul, S. C., Sugimoto, Y., and Mitsui, Y. (1993) J. Biol. Chem. 268, 22239 –22242. 42. Prudovsky, I., Savion, N., Zhan, X., Friesel, R., Xu, J., Hou, J., McKeehan, W. L., and Maciag, T. (1994) J. Biol. Chem. 269, 31720 –31724. 43. LaValee, T. M., Prudovsky, I. A., McMahon, G. A., Hu, X., and Maciag, T. (1998) J. Cell Biol. 141, 1647–1658.
1209
Cell-Cycle Dependent Tyrosine Phosphorylation on Mortalin Regulates Its Interaction with Fibroblast Growth Factor-1 Eiichi Mizukoshi,* ,† Masashi Suzuki,* Tomoko Misono,* Alexei Loupatov,* Eisuke Munekata,† Sunil C. Kaul,* Renu Wadhwa,‡ and Toru Imamura* ,1 *National Institute of Bioscience and Human Technology, AIST, Tsukuba, Ibaraki 305-8566, Japan; †Institute of Applied Biochemistry, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan; and ‡Chugai Research Institute for Molecular Medicine, Inc., Niihari, Ibaraki 300-4101, Japan
Received November 24, 2000
We previously reported that endogenously expressed, intracellularly localized fibroblast growth factor (FGF)-1 interacts with mortalin. Here we report that FGF-1 added to the culture medium of quiescent BALB/c3T3 cells is taken up by the cells and interacts with mortalin in the cells in a regulated manner. Although both the internalized FGF-1 and mortalin were present at high levels throughout the FGF-1-initiated cell cycle, their interaction became apparent only in late G1 phase. Interestingly, mortalin was preferentially tyrosine phosphorylated at the same time, and when its normally weak phosphorylation in early G1 phase was augmented by treating the cells with vanadate, a strong interaction between mortalin and FGF-1 was established. Conversely, when phosphorylated mortalin was treated with tyrosine phosphatase, its interaction with FGF-1 was abrogated. These results indicate that FGF-1 taken up by cells preferentially interacts with mortalin in late G1 phase of the cell cycle, and that tyrosine phosphorylation of mortalin regulates this interaction. © 2001 Academic Press Key Words: fibroblast growth factor; FGF; mortalin; GRP75; internalization; interaction; tyrosine phosphorylation; cell-cycle.
The fibroblast growth factor (FGF) family of proteins is composed of over 20 members, including FGF homologous factors, that control such cellular processes as growth, differentiation, migration, and survival (1– 6). FGFs stimulate these biological responses by binding to surface receptor tyrosine kinases (FGFRs) and heparan sulfate sugar chains, thereby activating intracel1 To whom correspondence should be addressed at Biosignaling Department, National Institute of Bioscience and Human Technology, AIST, 1-1 Higashi, Ibaraki 305-8566, Japan. Fax: ⫹81-298-616149. E-mail: [email protected].
lular signaling cascades (7–9). The intracellular molecules involved in these signaling pathways include phospholipase gamma, FRS2/Snt, Grb-2, Shc, and MAPK (10 –13); however, it is now clear that the immediate-early events occurring downstream of the FGFR are not sufficient to exert the mitogenic activity of FGF-1 (14 –15). Instead, after binding to cell-surface FGFRs, receptor-bound FGF-1 is transported across the plasma membrane via an as yet unidentified mechanism, and is deposited in the cytosol where it is then able to enter the nucleus (14 –20); the mitogenic activity of FGF-1 correlates closely with its appearance in the cytosol and with the presence of a nuclear localization sequence in its primary structure (14 –15). We and others have reported that delivery of the FGF-1 nuclear localization signal into living cells stimulates DNA synthesis (21–22), while the ability of FGF-1 to induce DNA synthesis requires its continuous presence in the cytosol and nucleus during the entire G1 phase (17, 20). Thus, internalized FGF-1 likely acts in a cooperative and synergetic fashion with receptortriggered intracellular processes to exert its full mitogenic activity (20). Increasing evidence indicates that intracellular FGF-1 binds to several proteins. FGF-1 intracellular binding protein (FIBP) interacts with FGF-1 and may serve to shuttle FGF-1 between the cytosol and nucleus (23), while synaptotagmin-1 and S100A13 also aggregate with intracellular FGF-1 (24 –25). Other FGF family ligands interact with other intracellular proteins. For example, FGF-2 bound to Casein kinase II and might regulate nuclear protein and exert its activity in nucleus (26). Intracellular proteins, including p79 and Cystein-rich FGF receptor (ESL), form complex with FGF-2 (27). FGF-3 was coimmunoprecipitated with karyopherin alpha, a nuclear localization receptor (28).
1203
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
We previously identified mortalin/glucose-regulated protein (GRP)75/mitochondrial hsp (mthsp)70/PBP-74 as an intracellular FGF-1-binding protein (29). As its various names suggest, this molecule appears to be involved in diverse regulatory processes including cellular senescence, glucose regulation, mitochondrial transport, nephrotoxicity, and antigen processing (30 – 34). While heat shock proteins (HSPs) have been generally considered to function in the folding, assembly and degradation of newly synthesized proteins (35), increasing evidence indicates that they also associate with regulatory proteins such as receptor kinases and transcription factors (36). Hsp27 and Hsp90, moreover, are phosphorylated on Ser/Thr residues (36 –37), while Hsp75 and Heat shock cognate protein 70 undergo acute tyrosine phosphorylation in response to vanadate and hydrogen peroxide (38 –39). These results suggest that HSP proteins may be involved in signaling pathway in regulated manners. In the present study, we examined the interaction between mortalin and FGF-1 taken up from culture medium by BALB/ c3T3 cells with the aim of better understanding the regulation of mortalin-FGF-1 complex formation. MATERIALS AND METHODS Reagents. The expression and purification of human recombinant FGF-1 and the production of an anti-mortalin polyclonal antibody and anti-FGF-1 monoclonal antibodies (mAb1, mAb15) have all been described previously (15, 17, 30). Anti-mthsp70 monoclonal antibody was purchased from Affinity Bioreagents (Golden, CA). Anti-FGF-1 polyclonal antibody, anti-phosphotyrosine monoclonal antibody (PT66), and rabbit kidney-derived protein tyrosine phosphatase were from Sigma Chemical (St. Louis, MO). Anti-phosphotyrosine monoclonal antibody (4G10) was from Upstate Biotechnology (Lake Placid, NY). Anti-cyclin D1 monoclonal antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). Alexa Fluor 488 goat antimouse IgG (H⫹L) conjugate was from Molecular Probes (Eugene, OR). Texas Red-conjugated donkey anti-rabbit IgG antibody was from Amersham Pharmacia Biotechnology (Uppsala, Sweden). Cell culture. BALB/c3T3 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in Dulbecco’s modified essential medium (DMEM, Sigma, St. Louis, MO) supplemented with 10% calf serum (CS, Hyclone Laboratories, Logan, UT). DNA synthesis assay. To assay the kinetics of DNA synthesis, BALB/c3T3 cells were plated to a density of 7.9 ⫻ 10 4 cells/cm 2 in 48-well plates (Sumilon, Tokyo) and cultured for 24 h in DMEM with 10% CS. The medium was then replaced with DMEM supplemented with 100 nM insulin, and the cells incubated for an additional 48 h before stimulation with FGF-1. DNA synthesis was initiated by addition of recombinant human FGF-1 (100 ng/ml) and heparin (50 g/ml). The cultures were then pulse-labeled for 2 h with [ 3H]thymidine (2 Ci/ml; Moravek Biochemicals, CA), and at the end of labeling period DNA synthesis was determined as described previously (21). Immunoblot analysis. Samples resolved by SDS–polyacrylamide gel electrophoresis (SDS–PAGE) were electrotransferred onto nitrocellulose membranes. The membranes were then blocked overnight at 4°C in buffer containing 5% skim milk, 150 mM NaCl, 10 mM Tris–HCl pH 7.4 (TBS), incubated for 60 min at room temperature with the appropriate antibody, washed for 20 min with 5% skim milk
in TBS containing 0.1% Tween 20 (TBS-T), and incubated with peroxidase-labeled goat anti-mouse IgG (Chemicon, Temecula, CA) appropriately diluted with TBS-T buffer. The membranes were then washed with TBS-T four times, and the signals were visualized using enhanced chemiluminescence (ECL, Amersham life science, Buckinghamshire, England) and Kodak X-O mat AR film (Kodak, Rochester, NY). Immunoprecipitation. Quiescent monolayers of BALB/c3T3 cells were treated with recombinant FGF-1 (20 ng/ml) and heparin (10 g/ml). At the times indicated, the cells were washed three times with phosphate-buffered saline (PBS), harvested using a cell scraper and lysed for 20 min on ice in 1 ml of lysis buffer (0.5% Triton X-100, 0.1% sodium deoxycolate, 150 mM NaCl, 10 mM Tris–HCl, pH 7.4, 1 mM EDTA, 50 mM NaF, Protease Inhibitor Cocktail (Boehringer Manheim, Manheim, Germany), 1 mM sodium orthovanadate and phosphatase inhibitor cocktail (Sigma)). The lysates were clarified by centrifugation for 10 min at 12,000g and incubated for 1 h at 4°C with the appropriate antibodies. Immunocomplexes were precipitated with protein A⫹G Sepharose (CALBIOCHEM, La Jolla, CA) and then washed three times with lysis buffer. The immunoprecipitated proteins were resolved by SDS–PAGE and analyzed by immunoblotting. Immunofluorescence analysis. BALB/c3T3 cells plated on coverslips and maintained in DMEM supplemented with 100 nM insulin were stimulated with FGF-1 (100 ng/ml) and heparin (50 g/ml) for the indicated times, after which they were fixed for 20 min at room temperature with 4% paraformaldehyde in PBS and permeabilized for 10 min with 0.2% Triton X-100 in PBS. The coverslips were then blocked for 30 min at 37°C in PBS containing 6% BSA, incubated with polyclonal anti-mortalin antibody, and either monoclonal antiFGF-1 or anti-phosphotyrosine antibody, washed and double-labeled with Alexa Fluor-conjugated anti-mouse IgG and Texas Redconjugated anti-rabbit antibodies. The coverslips were then mounted in FA mounting fluid (DIFCO LABORATORIES, Detroit, MI), and confocal microscopy was carried out using Carl Zeiss (Jena, Germany) equipment as described previously (29). In vitro binding assay. Serum-starved BALB/c3T3 cells were stimulated with FGF-1 (20 ng/ml) and heparin (10 g/ml) for 12 h and then lysed. Mortalin in the cell lysate was then exposed to anti-mortalin antibody, and the resultant immunocomplexes were washed first with high salt buffer (50 mM Tris–HCl, pH 7.4, 0.5 M NaCl) and then with phosphatase buffer (50 mM Tris–HCl pH 7.4), after which they were incubated for 30 min at 37°C with tyrosine phosphatase (0.1 units) in phosphatase buffer. After washing extensively with binding buffer (50 mM Tris–HCl pH 7.4, 150 mM NaCl), the immunocomplexes were incubated for 1 h in binding buffer containing FGF-1 (200 ng/ml) and BSA (200 g/ml), and washed again with binding buffer. The bound proteins were resolved by SDS–PAGE and then subjected to immunoblotting using antiFGF-1, anti-phosphotyrosine and anti-mortalin antibodies as probes.
RESULTS FGF-1 Protein Taken Up by BALB/c3T3 Cells Interacts with Endogenous Mortalin in a Cell Cycle-Dependent Manner In an earlier study we identified mouse mortalin and GRP75 (a rat homolog of mouse mortalin) as intracellular proteins that bind to endogenously-expressed, intracellularly-localized FGF-1 (29). And as FGF-1 added to culture medium is known to be taken up by the cells (14 –20), in the present study we investigated whether exogenously-applied FGF-1 would also bind mortalin. For this purpose, we used BALB/c3T3 cells,
1204
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Cell-cycle dependent interaction of FGF-1 with mortalin in BALB/c3T3 cells. (A) Time course of the interaction of FGF-1 with mortalin in BALB/c3T3 cells. Quiescent BALB/c3T3 cells were stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml), then harvested and lysed at the indicated times. Cell lysates were immunoprecipitated with anti-FGF-1 polyclonal antibody. Immunocomplexes were separated on 7.5% SDS–PAGE and analyzed by immunoblotting with an anti-mthsp70 antibody that recognizes both mortalin variants (mot-1 and mot-2). (B) Time course of internalization of FGF-1. Immunocomplexes obtained in (A) were analyzed by immunoblotting with anti-FGF-1 monoclonal antibody (mAb1). (C) Time course of mortalin expression in BALB/c3T3 cells. The amount of mortalin in the cell lysates obtained in (A) were examined by immunoblotting with anti-mthsp70 antibody. (D) Colocalization of internalized FGF-1 and mortalin. Quiescent BALB/c3T3 cells were stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml). After 12 h of stimulation, cells were fixed and labeled with polyclonal anti-mortalin and anti-FGF-1 (mAb15) antibodies and corresponding Texas red- and Alexa Fluor-conjugated secondary antibodies. Shown are confocal images of immunofluorescence depicting the distributions of mortalin (left panel), FGF-1 (middle panel), and the superimposition of the two (right panel). (E) Time course of cyclin D1 expression. Serum-starved cells were stimulated at the indicated times with FGF-1 (100 ng/ml) and harvested. Cyclin D1 expression was determined by immunoblotting with anti-cyclin D1 antibody. (F) Kinetics of DNA synthesis in BALB/c3T3 cells. Cells were serum-starved for 48 h, stimulated with FGF-1 (100 ng/ml) in the presence of heparin (50 g/ml) for 2–32 h, and labeled with [ 3H]thymidine during the final 2 h of the stimulation period. The incorporated radioactivity was determined with the mean ⫹/⫺ SE of triplicate cultures.
which express moderate levels of mortalin but not FGF-1. Once the cells were made quiescent by serum starvation, FGF-1 (100 ng/ml) was added to the culture medium for 12 h, after which the cells were solubilized, and the cell-associated FGF-1 was immunoprecipitated using a specific antibody. As shown in Fig. 1A, FGF-1 was indeed immunoprecipitated as part of a complex
with mortalin. Interestingly, however, though levels of cell-associated FGF-1 were maximal within 30 min of its addition (Fig. 1B) in good agreement with the results in our earlier study (17), and levels of mortalin remained constant throughout the FGF-1-initiated cell cycle (Fig. 1C), formation of mortalin-FGF-1 complexes developed gradually over the 12-h stimulation period
1205
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Double-immunolabeling with anti-mortalin and anti-phosphotyrosine antibodies of the cells after 12 h of FGF-1 stimulation revealed that there was substantial overlap between the distributions of mortalin and phosphotyrosine-containing proteins in the juxtanuclear region (results not shown). Inhibition of Phosphatase Augments the Interaction between Mortalin and FGF-1 FIG. 2. Cell-cycle dependent tyrosine-phosphorylation of mortalin in BALB/c3T3 cells. (A) Time course of tyrosinephosphorylation of mortalin in BALB/c3T3 cells. Quiescent BALB/ c3T3 cells were treated with FGF-1 (20 ng/ml) and heparin (10 g/ml) for the indicated times. Cell lysates were then prepared and incubated with anti-phosphotyrosine antibody (PT66). The immunoprecipitates were analyzed by immunoblotting with anti-mthsp70 antibody. (B) Time course of tyrosine-phosphorylation of mortalin in BALB/c3T3 cells. Cell lysates prepared in (A) were precipitated with anti-mthsp70 antibody. The immunocomplexes were then analyzed by immunoblotting with anti-phosphotyrosine antibody (4G10). Mortalin is indicated by an arrow. (C) Mortalin levels in BALB/c3T3 cells treated with FGF-1 for various times. The immunoprecipitates in (B) were probed with anti-mthsp70 antibody to determine mortalin levels.
(Fig. 1A). Moreover, when FGF-1 and mortalin were localized immunohistochemically by labeling FGF-1stimulated (12 h) cells with anti-FGF-1 and antimortalin antibodies, considerable overlap between the FGF-1 and mortalin signals was observed in the juxtanuclear region (Fig. 1D). Because the interaction profile suggested a possible correlation between the interaction of FGF-1 and mortalin and cell cycle progression, we monitored the cell cycle by analyzing cyclin D1 expression and DNA synthesis. As shown in Fig. 1E, cyclin D1 expression peaked after about 16 h FGF-1 stimulation, and S phase, as determined by DNA synthesis, occurred after 20 –24 h of stimulation (Fig. 1F). The interaction of internalized FGF-1 and endogenous mortalin thus appears to correlate with the late G1 and early S phases of the cell cycle. Mortalin Is Tyrosine Phosphorylated during Late G1-S Phase of the FGF-1-Initiated Cell Cycle The cell cycle-dependent interaction of FGF-1 and mortalin suggested the occurrence of cell cycledependent, post-translational modification of these proteins. And since tyrosine-phosphorylation of cellular proteins is reported to increase in late G1 phase, we analyzed the proteins and found that murine mortalin was indeed tyrosine phosphorylated (Figs. 2A and 2B), though FGF-1 was not (data not shown). The level of mortalin phosphorylation increased with progression of the cell cycle, peaking 12 h after the onset of G1 (Figs. 2A and 2B), while expression of mortalin remained constant (Fig. 2C).
The results described so far suggest that the interaction between mortalin and internalized FGF-1 is regulated by tyrosine phosphorylation of the former. We therefore examined the extent to which modulation of mortalin tyrosine phosphorylation affected formation of the mortalin-FGF-1 complex. We first tested the effect of sodium orthovanadate, a phosphatase inhibitor, and found that inhibition of phosphatase activity significantly augmented tyrosine phosphorylation of mortalin (Fig. 3A). Moreover, treating cells with vanadate significantly increased the interaction of mortalin with FGF-1 in early to middle G1 phase—i.e., after 1, 2 and 4 h of FGF-1 stimulation (Fig. 3B). For example, mortalin-FGF-1 binding after 1 h of stimulation was increased approximately 4.5-fold in the presence of vanadate (Fig. 3B). By contrast, vanadate had no effect after 8 h of stimulation, when substantial tyrosine phosphorylation of mortalin was observed even in its absence (Figs. 3A and 3B). Uptake of FGF-1 by the cells was unaffected by vanadate, confirming that the increased interaction between mortalin and FGF-1 was not the result of a vanadate-induced increase in the availability of FGF-1 (Fig. 3B). Treatment of Phosphorylated Mortalin with Tyrosine Phosphatase Abrogates Complex Formation Our finding that an inhibitor of tyrosine phosphatase augmented the interaction between FGF-1 and mortalin in early G1 phase strongly suggests the involvement of phosphotyrosine in the formation of the mortalin-FGF-1 complex. This was further confirmed by examining the effect of tyrosine phosphatase. After stimulating cells with FGF-1 for 12 h, the tyrosinephosphorylated mortalin was extracted, digested with tyrosine phosphatase, and analyzed for its interaction with FGF-1 in vitro. Phosphatase treatment diminished the level of mortalin phosphorylation, as well as its interaction with FGF-1, though mortalin itself was unaffected (Fig. 4A). Analysis of the dose dependency and time course for the enzyme digestion revealed that treating cell extracts with 0.1 U of tyrosine phosphatase for 30 min (Fig. 4B, left panels) or with 0.2 U for 5 min (Fig. 4B, right panels) was sufficient to significantly diminish tyrosine phosphorylation and formation of mortalin-FGF-1 complexes.
1206
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
gether with our earlier observations (29), these findings indicate that tyrosine phosphorylation of mortalin promotes, but is not essential for, the interaction of mortalin with FGF-1. Nevertheless, as it enhances the interaction by more than fourfold, mortalin phosphorylation should be the determining factor regulating the degree of interaction. Mortalin represents two variants (mot-1 and mot-2) of a 75 kDa protein, both of which are expressed by BALB/c3T3 cells and interact with FGF-1. In the present study we used an antibody against mthsp75 that recognizes both mortalin variants to demonstrate the cell cycle-dependency of mortalin tyrosine phosphorylation for the first time. There is ample precedent for the phosphorylation of mortalin and related proteins. For example, a 75 kDa protein similar to mortalin/GRP75/mthsp70/PBP74 undergoes tyrosine phosphorylation when cells are treated with hydrogen peroxide and vanadate (38), and rat liver mthsp70
FIG. 3. Sodium orthovanadate enhances FGF-1-induced tyrosine-phosphorylation of mortalin and interaction of FGF-1 with mortalin. (A) Enhanced tyrosine-phosphorylation of mortalin by sodium orthovanadate (Va). Quiescent BALB/c3T3 cells were cultured for 1 h in the presence or absence of sodium orthovanadate (50 M), after which the cultures were treated with FGF-1 (20 ng/ml) and heparin (10 g/ml). Lysates were prepared at the indicated times and then incubated with anti-mthsp70 antibody. Immunocomplexes were analyzed by immunoblotting with anti-phosphotyrosine (4G10) (upper panels) or anti-mthsp70 (lower panels) antibodies. (B) Enhanced mortalin-FGF-1 binding evoked by sodium orthovanadate. Lysates prepared as described in (A) were incubated with anti-FGF-1 polyclonal antibody. The resultant precipitates were examined by immunoblotting with anti-mthsp70 antibody (upper panels). The amount of internalized FGF-1 in each sample was determined by immunoblotting with anti-FGF-1 antibody (mAb1) (lower panels). Relative levels of FGF-1-mortalin binding in the presence (F) or absence (E) of vanadate were determined by measuring the optical density of the bands and normalizing the results to the values obtained at 8 h, which were assigned a value 1.0.
DISCUSSION We have shown that FGF-1 present in culture medium is taken up by BALB/c3T3 cells, where it interacts with intracellular mortalin, just as endogenouslyexpressed, intracellularly-localized FGF-1 interacts with GRP75 (a rat homolog of mortalin) in rat L6 and Rat-1 cells (29). This interaction correlated with the level of mortalin tyrosine phosphorylation, which peaked in late G1 phase, suggesting that cell cycledependent tyrosine phosphorylation of mortalin regulates formation of the mortalin-FGF-1 complex. To-
FIG. 4. Tyrosine phosphatase decreases the interaction of FGF-1 with mortalin in vitro. (A) Mortalin in lysates of quiescent (0 h) or FGF-1-stimulated (12 h) BALB/c3T3 cells was immobilized on protein A ⫹ G Sepharose beads with anti-mthsp70 antibody. The immunocomplexes were then washed with high salt buffer (0.5 M NaCl, 50 mM Tris–HCl pH 7.4) and incubated for 30 min with tyrosine phosphatase (0.2 U), after which the beads were incubated with FGF-1 (200 ng/ml) in binding buffer (150 mM NaCl, 50 mM Tris–HCl pH 7.4) containing BSA (200 g/ml), and washed. Bound proteins were separated on SDS–PAGE and analyzed by immunoblotting with anti-FGF-1 (mAb1) (upper panel), anti-phosphotyrosine (4G10) (middle panel), or anti-mthsp70/mortalin (lower panel) antibody. (B) Immunocomplex prepared as described in (A) were incubated for 30 min with selected amounts of tyrosine phosphatase (left panels) or with fixed amount of enzyme (0.2 U) for various periods (right panels). FGF-1-binding assays were then carried out as described in (A).
1207
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
exhibits Ca 2⫹-dependent autophosphorylation (34), suggesting mortalin’s involvement in a variety of diverse signaling pathways. Indeed mortalin is known to associate with the IL-1 receptor and may affect the accessibility of IL-1 receptor associated kinase (IRAK) substrates (40). As a member of HSP 70 family, mortalin appears to play significant roles in regulating various signaling cascades, including those mediating cell senescence (30, 41). Because only critical regulatory molecules are tyrosine phosphorylated, our findings that mortalin is tyrosine phosphorylated in late G1-phase and that the phosphorylation affects its interaction with FGF-1 suggests formation of the mortalin-FGF-1 complex is crucial for the transduction of FGF-1-induced cell growth and differentiation. One possible function of the mortalin-FGF-1 complex is suggested by its distribution within cells. Confocal immunofluorescence microscopy showed that mortalin and FGF-1 colocalize in juxtanuclear regions, and that this colocalization is apparent 12 h after the onset of G1 phase. Interestingly, the juxtanuclear distribution of mortalin resembles the distributions of tyrosinephosphorylated FGFR-1 (42) and FGFR-4 (16); mortalin may thus regulate the availability of FGF-1 to FGFRs. It has been reported that not only immediate early events but also so-called late events, taking place in the middle to late G1 phase, are induced following growth factor binding to its cognate tyrosine kinase receptor (10 –13, 20). Apparently these late events are crucial for inducing cells to enter S phase and complete the cell cycle (20, 43). Consequently, short term exposure to FGF-1 is not sufficient to establish DNA synthesis, though it does induce cell migration (43). The late events mediating the mitogenic activity of FGF-1 may include nuclear localization of FGF-1 (17, 20) and tyrosine hyperphosphorylation of several cellular proteins, including cortactin and Src (6, 20, 42– 43). Furthermore, continuous uptake of FGF-1 into the cytosol and nucleus enables cells to progress through the cell cycle (17, 20); thus regulated association of mortalin with FGF-1 may be a part of late events and modulate this as yet uncharacterized activity of cytosolic and/or nuclear FGF-1. ACKNOWLEDGMENTS We thank Drs. Ai-Jun Li and Akira Ishisaki at NIBH for confocal image analysis and discussion. The text was edited by Dr. William Goldman at MST Editing Company. This study was supported by a AIST grant to T.I.
REFERENCES 1. Baird, A., and Klagsbrun, M. (1991) Cancer Cells 3, 239 –243. 2. Emoto, H., Tagashira, S., Mattei, M. G., Yamasaki, M., Hashimoto, G., Katsumata, T., Negoro, T., Nakatsuka, M., Birnbaum, D., Coulier, F., and Itoh, N. (1997) J. Biol. Chem. 272, 23191– 23194.
3. Smallwood, P. M., Munoz-Sanjuan, I., Tong, P., Macke, J. P., Hendry, S. H., Gilbert, D. J., Copeland, N. G., Jenkins, N. A., and Nathans, J. (1996) Proc. Natl. Acad. Sci. USA 93, 9850 –9857. 4. Mason, I. J. (1994) Cell 78, 547–552. 5. Fiesel, R., and Maiciag, T. (1995) FASEB J. 9, 919 –925. 6. Liu, J., Huang, C., and Zhan, X. (1999) Oncogene 18, 6700 – 6706. 7. Mckeehan, W. L., Wang, F., and Kan, M. (1998) Prog. Nucleic Acid Res. Mol. Biol. 59, 135–176. 8. Imamura, T., Tokita, Y., and Mitsui, Y. (1988) Biochem. Biophys. Res. Commun. 155, 583–590. 9. DiGabriele, A. D., Lax, I., Chen, D. I., Svahn, C. M, Jaye, M., Schlessinger, J., and Hendrickson, W. A. (1998) Nature 393, 812– 817. 10. Wang, J. K., Gao, G., and Goldfarb, M. (1994) Mol. Cell. Biol. 14, 181–188. 11. Wang, J. K., Xu, H., Li, H. C., and Goldfarb, M. (1996) Oncogene 13, 721- 729. 12. 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. 11, 5068 –5078. 13. Kouhara, H., Hadari, Y. R., Spivak-Kroizman, T., Schilling, J., Bar-Sagi, D., Lax, I., and Schlessinger, J. (1997) Cell 89, 693– 702. 14. Wiedlocha, A., Falnes, P. O., Madshus, I. H., Sandvig, K., and Olsnes, S. (1994) Cell 76, 1039 –1051. 15. Imamura, T., Engleka, K., Zhan, X., Tokita, Y., Forough, R., Roeder, D., Jackson, A., Maier, J. A., Hla, T., and Maciag, T. (1990) Science 249, 1567–1570. 16. Citores, L., Wesche, J., Kolpakova, E., and Olsnes, S. (1999) Mol. Biol. Cell 10, 3835–3848. 17. Imamura, T., Oka, S., Tanahashi, T., and Okita, Y. (1994) Exp. Cell Res. 215, 363–372. 18. Zhan, X., Hu, X., Friedman, S., and Maciag, T. (1992) Biochem. Biophys. Res. Commun. 188, 982–991. 19. Klingenberg, O., Wiedolocha, A., Citores, L., and Olsnes, S. (2000) J. Biol. Chem. 275, 11972–11980. 20. Zhan, X., Hu, X., Friesel, R., and Maciag, T. (1993) J. Biol. Chem. 268, 9611–9620. 21. Komi, A., Suzuki, M., and Imamura, T. (1998) Exp. Cell Res. 243, 408 – 414. 22. Lin, Y. Z., Yao, S. Y., and Hawiger, J. (1996) J. Biol. Chem. 271, 5305–5308. 23. Kolpakova, E., Wiedolocha, A., Stenmark, H., Klingenberg, O., Falsnes, P. O., and Olsnes, S. (1998) Biochem. J. 336, 213–222. 24. Tarantini, F., LaVallee, T. M., Jackson, A., Gamble, S., Carreira, M. C., Garfinkel, S., Burgess, W. H., and Maciag, T. (1998) J. Biol. Chem. 273, 22209 –22216. 25. Carreira, M. C., LaValee, T. M., Tarantini, F., Jackson, A., Lathrop, J. T., Hampton, B., Burgess, W. H., and Maciag, T. (1998) J. Biol. Chem. 273, 22224 –22231. 26. Bonnet, H., Filhol, O., Truchet, I., Brethenou, P., Cochet, C., Amalric, F., and Bouche, G. (1996) J. Biol. Chem. 271, 24781– 24787. 27. Chevet, E., Lemaitre, G., Cailleret, K., Dahan, S., Bergeron, J. J., and Katinka, M. D. (1999) Biochem. J. 341, 713–723. 28. Antoine, M., Reimers, K., Dickson, C., and Kiefer, P. (1997) J. Biol. Chem. 272, 29475–29481. 29. Mizukoshi, E., Suzuki, M., Loupatov, A., Uruno, T., Hayashi, H., Misono, T., Kaul, S. C., Wadhwa, R., and Imamura, T (1999) Biochem. J. 343, 461– 466. 30. Wadhwa, R., Kaul, S. C., Ikawa, Y., and Sugimoto, Y. (1993) J. Biol. Chem. 268, 6615– 6621.
1208
Vol. 280, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
31. Massa, S. M., Longo, F. M., Zuo, J., Wang, S., Chen, J., and Sharp, F. R. (1995) J. Neurosci. Res. 40, 807– 819. 32. Bhattacharyya, T., Karnezis, A. N., Murphy, S. P., Hoang, T., Freeman, B. C., Phillips, B., and Morimoto, R. I. (1995) J. Biol. Chem. 270, 1705–1710. 33. Domanico, S. Z., DeNagel, D. C., Dahlseid, J. N., Green, J. M., and Pierce, S. K. (1993) Mol. Cell. Biol. 13, 3598 –3610. 34. Webster, T. J., Naylor, D. J., Hartman, D. J., Hoj, P. B., and Hoogenraad, N. J (1994) DNA Cell Biol. 13, 1213–1220. 35. Martin, J., and Hartl, F. U. (1997) Curr. Opin. Struct. Biol. 7, 41–52. 36. Rutherford, S. L., and Zuker, C. S. (1994) Cell 79, 1129 –1132. 37. Rousseau, S., Houle, F., Landry, J., and Huot, J. (1997) Oncogene 15, 2169 –2177.
38. Hadari, Y. R., Haring, H. U., and Zick, Y. (1997) J. Biol. Chem. 272, 657– 662. 39. Egerton, M., Moritz, R. L., Druker, B., Kelso, A., and Simpson, R. J. (1996) Biochem. Biophys. Res. Commun. 224, 666 – 674. 40. Sacht, G., Brigelius-Flohe, R., Kiess, M., Sztajer, H., and Flohe, L. (1999) Biofactors 9, 49 – 60. 41. Wadhwa, R., Kaul, S. C., Sugimoto, Y., and Mitsui, Y. (1993) J. Biol. Chem. 268, 22239 –22242. 42. Prudovsky, I., Savion, N., Zhan, X., Friesel, R., Xu, J., Hou, J., McKeehan, W. L., and Maciag, T. (1994) J. Biol. Chem. 269, 31720 –31724. 43. LaValee, T. M., Prudovsky, I. A., McMahon, G. A., Hu, X., and Maciag, T. (1998) J. Cell Biol. 141, 1647–1658.
1209