|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 279, Issue 25, 26445-26452, June 18, 2004
Bi-directional Regulation of Ser-985 Phosphorylation of c-Met via Protein Kinase C and Protein Phosphatase 2A Involves c-Met Activation and Cellular Responsiveness to Hepatocyte Growth Factor*![]() ![]()
From the
Division of Molecular Regenerative Medicine, Course of Advanced Medicine, Osaka University Graduate School of Medicine, Osaka 565-0871 and
Received for publication, December 29, 2003 , and in revised form, March 29, 2004.
Previous studies indicated that treatment of cells with 12-O-tetradecanoylphorbol-13-acetate induced phosphorylation of Ser-985 at the juxtamembrane of c-Met, the receptor tyrosine kinase for hepatocyte growth factor (HGF), and this was associated with decreased tyrosine phosphorylation of c-Met. However, the regulatory mechanisms and the biological significance of the Ser-985 phosphorylation in c-Met remain unknown. When A549 human lung cancer cells were exposed to oxidative stress with H2O2, H2O2 treatment induced phosphorylation of Ser-985, but this was abrogated by an inhibitor for protein kinase C (PKC). Likewise, treatment of cells with NaF (an inhibitor of protein phosphatases) allowed for phosphorylation of Ser-985, and a protein phosphatase responsible for dephosphorylation of Ser-985 was identified to be protein phosphatase 2A (PP2A). The effects of PKC inhibitors revealed that PKC and - were responsible for the Ser-985 phosphorylation of c-Met, and pull-down analysis indicated that associations of PKC and - with c-Met may be involved in the regulation of Ser-985 phosphorylation of c-Met. Instead, PP2A was constitutively associated with c-Met, whereas its activity to dephosphorylate Ser-985 of c-Met was decreased when cells were exposed to H2O2. Addition of HGF to A549 cells in culture induced c-Met tyrosine phosphorylation, the result being mitogenic response and cell scattering. In contrast, in the presence of H2O2 stress, HGF-dependent tyrosine phosphorylation of c-Met was largely suppressed with a reciprocal relationship to Ser-985 phosphorylation, and this event was associated with abrogation of cellular responsiveness to HGF. These results indicate that Ser-985 phosphorylation of c-Met is bi-directionally regulated through PKC and PP2A, and the Ser-985 phosphorylation status may provide a unique mechanism that confers cellular responsiveness/unresponsivenss to HGF, depending on extracellular conditions.
Hepatocyte growth factor (HGF),1 originally identified and cloned as a mitogenic protein for hepatocytes (13), evokes multiple cellular responses, including mitogenesis, morphogenesis, migration, and anti-apoptosis (46). These biological activities of HGF are triggered by tyrosine phosphorylation of c-Met, a specific receptor tyrosine kinase for HGF (7). Biological activities of HGF support tissue organization during development and regeneration of organs, including the liver, kidney, placenta, and skeletal muscle (46), but unregulated and/or constitutive activation of the c-Met receptor endows tumor cells with invasive and metastatic characteristics (8, 9).
The c-Met receptor is a heterodimeric protein composed of extracellular
In the present study, we found that the phosphorylation status of juxtamembrane Ser-985 of the c-Met receptor is bidirectionally regulated through reverse activities of PKC
MaterialsA polyclonal anti-human c-Met antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal anti-phospho-c-Met (pY1234/1235) antibody, monoclonal anti-phosphotyrosine (4G10) and anti-PP2A antibodies, recombinant PKC isoforms, recombinant protein phosphatase 1 (PP1), and protein phosphatase 2A (PP2A) proteins were purchased from Upstate Biotechnology, Inc. A polyclonal anti-phospho-c-Met (pY1365) antibody was purchased from BIOSOURCE International (Camarillo, CA). A monoclonal anti-bromodeoxyuridine (BrdUrd) antibody and monoclonal anti-PKC antibodies were purchased from BD Biosciences. Bisindolylmaleimide (BIM), KN-62, Gö6976, Gö6983, calyculin A, and cyclosporin A were from Calbiochem. TO-PRO-3 was from Molecular Probes (Eugene, OR). A polyclonal anti-phospho-Ser-985 c-Met antibody against synthetic peptide corresponding to the 11 amino acid sequence at the juxtamembrane domain of c-Met (amino acids 978988) was raised in rabbit wherein Ser-985 was phosphorylated. The antibody was purified using an affinity column immobilized with the antigenic peptide. Cell CultureA549 cells obtained from the Japanese Cancer Research Resources Bank were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum at 37 °C in a humidified atmosphere containing 5% CO2.
Immunoprecipitation and Western BlottingCells were lysed in lysis buffer consisting of 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 25 mM 5-Bromo-2'-deoxyuridine (BrdUrd) LabelingCells were stimulated 10 ng/ml HGF for 12 h with or without 0.5 mM H2O2 and labeled with 10 µM BrdUrd during the last 2 h of the culture period. The cells were then fixed with 4% paraformaldehyde for 10 min and permeabilized with 0.2% Triton X-100 for 3 min. Cells were further treated with 2 M HCl for 15 min and neutralized with 0.1 M sodium tetraborate for 30 min. After washing with phosphate-buffered saline, these cells were incubated with anti-BrdUrd antibody for 2 h and subsequently with fluorescence labeled second antibody for 1 h. The cells were washed with phosphate-buffered saline, and the nuclei were then stained with TO-PRO-3. The number of cells stained positively for BrdUrd and nuclei was counted using fluorescence microscopy.
Protein Kinase C AssayImmunoprecipitated c-Met from nonstimulated A549 cells were incubated with recombinant PKC isoforms in the reaction buffer (20 mM Tris-HCl, 10 mM MgCl2, 20 µM ATP, 25 mM
Changes in PKC activity were assayed by measuring the transfer of 32Pi from [ Protein Phosphatase AssaySer-985 of the c-Met receptor was phosphorylated by treatment of the cells with 25 mM NaF for 10 min. Ser-985-phosphorylated c-Met was immunoprecipitated from the cells by using an anti-c-Met antibody, and incubated with recombinant PP1 and PP2A in the buffer (50 mM Tris-HCl, 0.1 mM Na2EDTA, 5 mM dithiothreitol, 0.01% Brij 35, and 2 mM MnCl2 (pH 7.5)), respectively, at 30 °C for 10 min. Samples were subjected to SDS-PAGE and subsequent Western blotting by using anti-phospho-Ser-985 antibody.
Inhibition of Cellular Response to HGF by H2O2Oxidative stress induces growth arrest in proliferating cells (14). We first asked whether the mitogenic stimulus of HGF was influenced under oxidative stress. DNA synthesis was measured by Br-dUrd incorporation into A549 cells cultured in the absence or presence of HGF and/or H2O2 (Fig. 1). In the control culture, the number of cells labeled with BrdUrd reached 32%. When the cells were cultured in the presence of 0.5 mM H2O2 for 12 h, numbers of BrdUrd-positive cells decreased to 4%, indicating that serum-dependent proliferation of cells was strongly inhibited by H2O2. On the other hand, addition of HGF (10 ng/ml) stimulated BrdUrd labeling to 60%, indicating that HGF exerts mitogenic activity for A549 cells even in the presence of fetal bovine serum. However, exposure of the cells to H2O2 in the presence of HGF inhibited BrdUrd uptake to a level similar to that seen in cells treated with H2O2. These results demonstrate that the mitogenic stimulus of HGF, as well as serum components, was down-regulated by H2O2.
Because cell scattering is a unique biological response through activation of the c-Met receptor and serum components do not induce cell scattering for A549 cells (see below), we also asked if the cell scattering induced by HGF is influenced under the oxidative stress. A549 cells were cultured under the absence or presence of HGF and/or H2O2 for 12 h (Fig. 1B). HGF (10 ng/ml) significantly induced scattering of cells, although it was inhibited in the presence of 0.5 mM H2O2. Addition of H2O2 alone did not induce cell scattering. Taken together, in addition to the mitogenic response, H2O2 treatment diminished biological responses to HGF in A549 cells.
PKC-mediated Ser-985 Phosphorylation Under H2O2 StressTo search for mechanisms by which H2O2 treatment suppressed biological responses to HGF in A549 cells, we focused on the Ser-985 phosphorylation of c-Met, as well as tyrosine phosphorylation of the c-Met receptor, because phosphorylation of Ser-985 induced by TPA treatment was associated with suppression of tyrosine phosphorylation of the c-Met receptor (13). To specifically detect phosphorylated Ser-985, we prepared a polyclonal antibody against a synthetic peptide (amino acids 978988 in the c-Met), which contains phosphorylated Ser-985. A549 cells were treated with or without TPA, and the cell lysate was subjected to immunoprecipitation for c-Met and subsequent Western blot with anti-phosphorylated Ser-985 antibody. The positive signal was detectable only in TPA-treated cells (Fig. 2A, left panel), and reprobing with anti-c-Met antibody indicated that the positive signal for phosphorylated Ser-985 coincided with the size of the c-Met
By using specific antibodies, phosphorylation of tyrosines and Ser-985 of c-Met was analyzed by immunoprecipitation and subsequent Western blots (Fig. 3A). Tyrosine phosphorylation of c-Met was induced within 5 min with HGF stimulus; however, the phosphorylation level strongly decreased when the cells were pretreated with H2O2 for 60 min prior to HGF stimulus (Fig. 3A, upper panel). In contrast to tyrosine phosphorylation of c-Met, Ser-985 was not phosphorylated by HGF but was strongly phosphorylated in the presence of H2O2 (Fig. 3A, middle panel). Therefore, H2O2 stimulus induces Ser-985 phosphorylation, and Ser-985 phosphorylation is reciprocally associated with the tyrosine phosphorylation status of the c-Met receptor.
When cells were treated with 0.5 mM H2O2 alone, phosphorylation of Ser-985 was within 10 min and was decreased but evidently sustained until 120 min after exposure to H2O2 (Fig. 3B). Ser-985 phosphorylation was induced by 0.1 mM H2O2 and increased in a dose-dependent manner (Fig. 3C). On the other hand, Ser-985 phosphorylation was almost completely inhibited by BIM (an inhibitor of PKC), indicating that PKC is involved in H2O2-induced Ser-985 phosphorylation (Fig. 3D). Reprobing of Western blots with anti-c-Met antibody revealed that the c-Met level was mostly equal at each immunoprecipitation procedure. These results indicate that changes in phosphorylation levels of Ser-985 and tyrosine residues in c-Met by H2O2 and HGF could be attributed to phosphorylation/dephosphorylation events and that PKC is responsible for H2O2-induced Ser-985 phosphorylation. Transient Phosphorylation of Ser-985 by HGF StimulusBecause PKCs are downstream signal transducers of c-Met following HGF stimulus, we speculated that the HGF stimulus would induce Ser-985 phosphorylation to some extent even without H2O2 treatment. When cells were treated with HGF for 10 min, Ser-985 was weakly phosphorylated in a concentration-dependent manner (Fig. 4A). In the time course experiment (Fig. 4B), Ser-985 phosphorylation was detected within 10 min by stimulus with 10 ng/ml HGF and sustained for up to 60 min; however, it disappeared 120 min after the stimulus. When the same immunoblots were reprobed using anti-phosphotyrosine antibody (Fig. 4B, middle panel), c-Met was tyrosine-phosphorylated within 1 min and then gradually increased until 60 min after the stimulus. Thereafter, the tyrosine phosphorylation level slightly declined in parallel to a slight decrease in the expression level of c-Met but remained at a high level until 120 min after the stimulus. Taken together with the finding that HGF stimulus allows for a mitogenic response in A549 cells in the absence of H2O2, these results indicate that HGF weakly induces Ser-985 phosphorylation at least to some extent, but it is transient and allows for HGF-dependent tyrosine phosphorylation of c-Met, which is capable of inducing a mitogenic stimulus.
When phosphorylation levels of Ser-985 induced by HGF and TPA were compared, Ser-985 phosphorylation induced by HGF was much weaker than that induced by TPA (Fig. 4C). Tyrosine phosphorylation of c-Met was induced by HGF, whereas when cells were simultaneously treated with HGF and TPA, HGF-dependent tyrosine phosphorylation was much weaker than that seen with HGF alone (Fig. 4C). Thus, the level of HGF-induced tyrosine phosphorylation of c-Met was again in a reciprocal relationship to the Ser-985 phosphorylation level when Ser-985 phosphorylation was induced by TPA. Because the amino acid sequence encompassing Ser-985 (Ser-Ala-Arg) coincides with the consensus sequences in substrates of PKC ((S/T)X(R/K)) and Ca2+/calmodulin kinase II (CaMK II) ((S/T)XR), involvement of PKC and CaMK II in Ser-985 phosphorylation was analyzed by using specific inhibitors (Fig. 4D). BIM almost completely inhibited Ser-985 phosphorylation induced by TPA and HGF. In contrast, KN-62 (CaMK II inhibitor) could not reduce Ser-985 phosphorylation induced by HGF and TPA. These results suggest that PKC is responsible for HGF-induced Ser-985 phosphorylation.
PKC Isoforms Responsible for Ser-985 PhosphorylationBecause TPA targets both conventional and novel types of PKCs, we addressed PKC isoforms responsible for Ser-985 phosphorylation of c-Met. Western blot analysis using cell lysate revealed that PKC isoforms,
We addressed mechanisms coordinating these activities for the regulation of Ser-985 phosphorylation. A potential mechanism would be association of PKC and c-Met. To determine whether PKCs are associated with c-Met, c-Met was immunoprecipitated from H2O2-treated cells, and the immunoprecipitates were subjected to Western blotting with anti-PKC antibodies. Association of PKC and - with c-Met was marginal in unstimulated cells, whereas it was detectable 10 min after H2O2 stimulus and was sustained for up to 60 min (Fig. 5D). These results suggested that the association of PKC and - with c-Met may be involved in the regulation of Ser-985 phosphorylation of c-Met.
To address further the PKC isoform-dependent Ser-985 phosphorylation, we measured changes in activity of PKC Involvement of Protein Phosphatase 2A in Ser-985 DephosphorylationBecause the HGF-induced Ser-985 phosphorylation was transient and the level was weak compared with that induced by H2O2 (Fig. 3) and TPA (Fig. 4), we speculated that Ser/Thr protein phosphatases might attenuate the level of phosphorylation of Ser-985 during HGF stimulation. To address this possibility, the effect of NaF was analyzed, which is a broad inhibitor for protein phosphatases on phosphorylation levels of Ser-985 and tyrosine of c-Met receptor (Fig. 6A). In this experiment, phosphorylation of tyrosine residues at 1234/1235 and 1365 of c-Met, which are critical autophosphorylation sites to evoke intracellular signals of HGF, was analyzed by using specific antibodies. In control culture without NaF, HGF induced transient Ser-985 phosphorylation and phosphorylation of Tyr-1234/1235 and Tyr-1365. On the other hand, pretreatment with NaF for 10 min allowed for Ser-985 phosphorylation, and the level was higher than that induced by HGF stimulus, suggesting that protein phosphatase(s) is involved in the regulation of Ser-985 phosphorylation. HGF stimulation in the presence of NaF further increased Ser-985 phosphorylation to higher levels than that seen with NaF alone. Most important, in the presence of NaF, phosphorylation levels of Tyr-1234/1235 and Tyr-1365 induced by HGF were much weaker than those seen in the absence of NaF, again demonstrating a reciprocal relationship between phosphorylation levels of Ser-985 and tyrosine residues in the c-Met receptor.
We then attempted to identify the phosphoserine phosphatase responsible for the attenuation of Ser-985 phosphorylation. Because NaF broadly inhibits activities of PP1, PP2A, and PP2B, the effects of selective inhibitors, calyculin A (specific to PP1/2A) and cyclosporin A (specific to PP2B), were evaluated (Fig. 6B). Treatment with calyculin A further enhanced the Ser-985 phosphorylation induced by HGF, but cyclosporin A did not influence the Ser-985 phosphorylation induced by HGF, thereby suggesting the involvement of PP1 and/or PP2A in Ser-985 dephosphorylation. To address this, the c-Met receptor possessing phosphorylated Ser-985 was immunoprecipitated from A549 cells pretreated with NaF, and the catalytic activity of purified recombinant PP1 and PP2A to dephosphorylate of Ser-985 was analyzed, respectively (Fig. 6C). Treatment with PP2A but not PP1 almost completely dephosphorylated Ser-985 of c-Met; hence, PP2A is presumed to be responsible for dephosphorylation of phospho-Ser-985 of c-Met. Association of PP2A with c-Met and PP2A Activity Modulated by HGF- and H2O2 StimulationThe above results indicate that the phosphorylation status of Ser-985 is bi-directionally regulated by phosphorylation and dephosphorylation mediated by PKC and PP2A, respectively. We therefore addressed mechanisms coordinating these activities for the regulation of Ser-985 phosphorylation. A potential mechanism would be association of PKC and/or PP2A to the c-Met receptor. To determine whether PP2A is associated with c-Met, c-Met was immunoprecipitated, and the immunoprecipitates were subjected to Western blotting with anti-PP2A antibody. PP2A was clearly associated with the c-Met receptor even in unstimulated cells, and HGF stimulus did not significantly alter the association of PP2A with c-Met (Fig. 6D). Hence, we suggest a constitutive association of PP2A with the c-Met receptor. When changes in the association of PP2A were analyzed in cells treated with H2O2, H2O2 stimulation did not change the association of PP2A with c-Met following H2O2 stimulus (Fig. 6D), again suggesting a constitutive association of PP2A with c-Met regardless of any extracellular stimuli with HGF and H2O2. We then considered possible changes in PP2A activity even though PP2A is constitutively associated with c-Met. To address this possibility, the catalytic activity of PP2A required to dephosphorylate phosphorylated Ser-985 of c-Met was analyzed. PP2A was immunoprecipitated from A549 cells untreated or treated with HGF or H2O2, and PP2A activity was analyzed by Western blots, using phosphorylated Ser-985 of c-Met from NaF-treated cells as a substrate (Fig. 6E). Ser-985 was equally dephosphorylated by PP2A prepared from cells untreated or treated with HGF for up to 60 min, indicating that PP2A associated with c-Met was active in unstimulated cells, and this activity was unchanged upon HGF stimulation. On the other hand, PP2A prepared from H2O2-treated cells decreased in activity to dephosphorylate Ser-985, thereby increasing the phosphorylation level of Ser-985 (Fig. 6E). These results indicate that PP2A associated with c-Met was active in cells under stress-free conditions, but this activity to dephosphorylate Ser-985 in cells decreased in the presence of oxidative stress.
A previous study demonstrated that Ser-985 phosphorylation of c-Met was induced by TPA stimuli, and this event resulted in a decrease of c-Met tyrosine phosphorylation (13). In our work, strong Ser-985 phosphorylation induced by H2O2 stimulus and inhibition of protein phosphatase, as well as TPA, was associated with a decreased c-Met tyrosine phosphorylation upon HGF stimulus. Thus, the phosphorylation status of Ser-985 seems to play a functional regulatory role in activation of the c-Met receptor, thereby involved in distinct cellular responsiveness to HGF, which depends on distinct extracellular conditions. However, the regulation of Ser-985 phosphorylation by extracellular stimuli, potential molecules responsible for Ser-985 phosphorylation/dephosphorylation, and involvement of Ser-985 phosphorylation in regulation of cellular responsiveness mediated by c-Met remain to be addressed.
In the present study, we found that Ser-985 was phosphorylated and dephosphorylated, respectively, by PKC (PKC
A variety of cellular damages and stresses such as oxidative stress, UV irradiation, and heat induce cell cycle arrest (14, 17, 18). Cell cycle arrest is a critical event, which allows cells to be restored from the damages (e.g. DNA-repair and refolding of denatured enzymes) or eliminated by apoptotic pathways; however, precise mechanisms of cell cycle arrest have yet to be elucidated. We found that the growth-promoting effects of HGF were abolished by suppression of the tyrosine phosphorylation event of c-Met when cells were exposed to oxidant stress, and the suppression was mediated by phosphorylation of Ser-985. To our knowledge, this is the first finding that H2O2-induced growth arrest is primarily regulated by receptor inactivation even when cells are exposed to mitogens. We propose that attenuation of the cellular responsiveness to growth factors may be an initial step to avoid disadvantages associated with cell cycle transition in damaged cells. Previous studies indicated that Ser/Thr phosphorylation in receptor tyrosine kinases is involved in inhibition of tyrosine phosphorylation events following ligand stimulus, suggesting that a specific Ser/Thr phosphorylation provides an autoinhibitory mechanism in distinct receptor tyrosine kinases. Tyrosine phosphorylation events in epidermal growth factor receptor (19), insulin receptor (20), c-Kit (21), and RET (22) are reciprocally inhibited by PKC-mediated Ser/Thr phosphorylation. Although related mechanisms remain to be addressed, the inhibition of receptor dimerization and enhancement of internalization in the epidermal growth factor receptor have been noted (23, 24). As for c-Met, pretreatment of cells with TPA and NaF induced extensive Ser-985 phosphorylation, and this was associated with suppression of tyrosine phosphorylation of c-Met upon HGF stimulus without any accompanying c-Met receptor degradation (Figs. 4C and 5A). Furthermore, TPA and NaF treatment after HGF stimulus also induced extensive Ser-985 phosphorylation and resulted in tyrosine dephosphorylation without c-Met degradation.2 Thus, we speculate that phosphorylation of Ser-985 may mediate the inhibition of receptor dimerization and/or association of protein-tyrosine phosphatases.
c-Met has a structural variant that lacks a cytoplasmic juxtamembrane region (
PKC is activated by diverse extracellular stimuli such as cellular stress, cell-cell adhesion, cell-extracellular matrix interaction, and cellular (tissue) injury (16, 2931). Likewise, c-Met-mediated cellular responsiveness is modified depending on extracellular conditions, including cellular stress, cell-cell adhesion, and cell extracellular matrix interaction (3234). Taken together, Ser-985 phosphorylation mediated by activation of PKC
* This work was supported by a research grant for studies on science and cancer from the Ministry of Education, Science, Technology, Sports and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 The abbreviations used are: HGF, hepatocyte growth factor; TPA, 12-O-tetradecanoylphorbol-13-acetate; PKC, protein kinase C; PP, protein phosphatase; BrdUrd, bromodeoxyuridine; BIM, bisindolylmaleimide; CaMK II, Ca2+/calmodulin kinase II.
2 A. Hashigasako, M. Machide, T. Nakamura, K. Matsumoto, and T. Nakamura, unpublished data.
We are grateful to M. Ohara (Fukuoka, Japan) for comments and for language assistance.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||