Title: p27 cytoplasmic localization is regulated by phosphorylation on Ser10 and is not a prerequisite for its proteolysis
Abstract: Article3 December 2001free access p27 cytoplasmic localization is regulated by phosphorylation on Ser10 and is not a prerequisite for its proteolysis Geneviève Rodier Geneviève Rodier Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Alessia Montagnoli Alessia Montagnoli Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Lucia Di Marcotullio Lucia Di Marcotullio Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Philippe Coulombe Philippe Coulombe Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Molecular Biology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Giulio F. Draetta Giulio F. Draetta European Institute of Oncology, 20141 Milan, Italy Search for more papers by this author Michele Pagano Michele Pagano Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Sylvain Meloche Corresponding Author Sylvain Meloche Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Molecular Biology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Pharmacology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Geneviève Rodier Geneviève Rodier Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Alessia Montagnoli Alessia Montagnoli Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Lucia Di Marcotullio Lucia Di Marcotullio Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Philippe Coulombe Philippe Coulombe Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Molecular Biology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Giulio F. Draetta Giulio F. Draetta European Institute of Oncology, 20141 Milan, Italy Search for more papers by this author Michele Pagano Michele Pagano Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA Search for more papers by this author Sylvain Meloche Corresponding Author Sylvain Meloche Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Molecular Biology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Department of Pharmacology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 Search for more papers by this author Author Information Geneviève Rodier1, Alessia Montagnoli2, Lucia Di Marcotullio2, Philippe Coulombe1,3, Giulio F. Draetta4, Michele Pagano2 and Sylvain Meloche 1,3,5 1Institut de recherches cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 2Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, NY, 10016 USA 3Department of Molecular Biology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 4European Institute of Oncology, 20141 Milan, Italy 5Department of Pharmacology, University of Montreal, Montreal, Quebec, Canada, H2W 1R7 *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:6672-6682https://doi.org/10.1093/emboj/20.23.6672 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The activity of the cyclin-dependent kinase inhibitor p27 is controlled by its concentration and subcellular localization. However, the mechanisms that regulate its intracellular transport are poorly understood. Here we show that p27 is phosphorylated on Ser10 in vivo and that mutation of Ser10 to Ala inhibits p27 cytoplasmic relocalization in response to mitogenic stimulation. In contrast, a fraction of wild-type p27 and a p27(S10D)-phospho-mimetic mutant translocates to the cytoplasm in the presence of mitogens. G1 nuclear export of p27 and its Ser10 phosphorylation precede cyclin-dependent kinase 2 (Cdk2) activation and degradation of the bulk of p27. Interestingly, leptomycin B-mediated nuclear accumulation accelerates the turnover of endogenous p27; the p27(S10A) mutant, which is trapped in the nucleus, has a shorter half-life than wild-type p27 and the p27(S10D) mutant. In summary, p27 is efficiently degraded in the nucleus and phosphorylation of Ser10 is necessary for the nuclear to cytoplasmic redistribution of a fraction of p27 in response to mitogenic stimulation. This cytoplasmic localization may serve to decrease the abundance of p27 in the nucleus below a certain threshold required for activation of cyclin–Cdk2 complexes. Introduction The cyclin-dependent kinase (Cdk) inhibitor p27Kip1 (p27) is an important regulator of the mammalian cell cycle (reviewed in Sherr and Roberts, 1995, 1999; Hengst and Reed, 1998). p27 negatively regulates G1 progression by binding to cyclin–Cdk2 complexes and preventing their activity. Accordingly, the levels of p27 are high in quiescent cells and decline upon mitogenic stimulation. In addition, both p27 and its homolog p21Cip1 positively regulate cell cycle progression by promoting the assembly and activity of cyclin D–Cdk4/6 complexes (Soos et al., 1996; LaBaer et al., 1997; Cheng et al., 1999). The activity of p27 is controlled by its concentration, its distribution among different cellular complexes and its subcellular localization (Ekholm and Reed, 2000; Slingerland and Pagano, 2000). One of the key mechanisms involved in the regulation of p27 abundance is proteolysis by the ubiquitin–proteasome pathway (Pagano et al., 1995). p27 is phosphorylated on Thr187 by cyclin E–Cdk2 (Muller et al., 1997; Sheaff et al., 1997; Vlach et al., 1997; Montagnoli et al., 1999), and is then recognized and targeted for ubiquitylation by the SCFSkp2 ubiquitin–protein ligase and its cofactor, the Cdk subunit 1 (Cks1; Carrano et al., 1999; Sutterluty et al., 1999; Tsvetkov et al., 1999; Nakayama et al., 2000; Ganoth et al., 2001; Spruck et al., 2001). Thus, the timing of p27 degradation is dependent on both the accumulation of cyclin E and concomitant activation of Cdk2, and the mitogen-stimulated induction of Skp2 and Cks1 expression. The biosynthesis of p27 is also subject to regulation by transcriptional (Servant et al., 2000) and translational mechanisms (Agrawal et al., 1996; Hengst and Reed, 1996; Millard et al., 1997; Miskimins et al., 2001). In addition to changes in expression levels, inactivation of p27 also occurs through sequestration by cyclin D–Cdk4/6 complexes (Sherr and Roberts, 1999). To exert its inhibitory action, p27 needs to be transported into the nucleus (Reynisdottir and Massague, 1997; Orend et al., 1998; Soucek et al., 1998; Tomoda et al., 1999). The nuclear import of p27 is dependent on the presence of a nuclear localization signal (NLS) localized near the C-terminus of the protein (Zeng et al., 2000). Moreover, association of p27 with nuclear pore-associated protein 60 (NPAP60) also contributes to nuclear import (Muller et al., 2000). p27 was shown to physically interact with Jab1, a component of the COP9–signalosome complex, whose overexpression induces p27 translocation to the cytoplasm and promotes p27 degradation (Tomoda et al., 1999). Thus, it is generally believed that degradation of p27 requires its cytoplasmic relocalization. Despite these observations, the regulation of p27 cellular localization remains poorly understood. Here, we show that phosphorylation of Ser10 is a necessary signal for nuclear export of p27 upon cell cycle re-entry. In addition, we demonstrate that p27 is efficiently degraded in the nucleus and that cytoplasmic localization is not a prerequisite for p27 degradation. Results p27 is phosphorylated on Ser10 in vivo We have analyzed the phosphorylation state of endogenous p27 in Rat1 fibroblasts. Results of in vivo labeling studies indicated that p27 is phosphorylated in quiescent cells (Figure 1A). The level of p27 phosphorylation transiently increases upon cell cycle re-entry. After 6–12 h of mitogenic stimulation, p27 phosphorylation declines with no apparent change in the amount of protein. At the G1/S transition, the amount of p27 protein and phosphorylation fall concomitantly (Figure 1A). Analysis of tryptic phosphopeptide maps revealed that p27 is phosphorylated on two major peptides (Figure 1B, spots 2 and 3) and one minor peptide of variable intensity (Figure 1B, spot 1). The two major labeled peptides were found to contain exclusively phosphoserine (data not shown). Figure 1.p27 is phosphorylated on Ser10 in vivo. (A) Quiescent Rat1 cells were labeled with 0.5 mCi/ml [32P]phosphoric acid for 4 h and then stimulated with 10% serum for the times indicated. The cells were lysed and the endogenous p27 was immunoprecipitated with anti-p27 antibody. The precipitated proteins were resolved by SDS–gel electrophoresis, transferred to PVDF membrane and analyzed by autoradiography (top). The abundance of p27 protein was monitored by immunoblotting (bottom). (B) The 32P-labeled p27 protein band (from cells stimulated for 12 h with serum) was excised from the membrane and digested with trypsin. The resulting phosphopeptides were separated by thin layer electrophoresis followed by ascending chromatography. The arrow denotes the position of sample application. (C) Rat1 cells were transfected with HA-tagged wild-type p27 or p27(S10A) mutant. After 48 h, the cells were labeled with [32P]phosphoric acid and ectopic p27 was immunoprecipitated from cellular lysates with an anti-HA antibody. The extent of phosphorylation and abundance of ectopic p27 were analyzed as in (A). (D) Phosphopeptide mapping analysis of ectopically expressed HA-tagged p27 and p27(S10A) mutant. (E) Specificity of p27 phospho-Ser10 antibody. Recombinant purified wild-type p27 and p27(S10A) mutant were phosphorylated in vitro by either cyclin E–Cdk2 (lanes 1 and 2) or ERK2 (lanes 3 and 4). Phosphorylation products were analyzed by immunoblotting with anti-phospho-Ser10-specific antibody (top), anti-phospho-Thr187 specific antibody (middle) or with a mouse anti-p27 antibody (bottom). (F) NIH 3T3 cells were transfected with human wild-type p27 (lane 1), p27(S10A; lane 2), p27(S10D; lane 3) or p27(S10E; lane 4) mutants. Protein extracts were analyzed by immunoblotting with either anti-phospho-Ser10 antibody or anti-p27 antibody. The human p27 can be distinguished from the endogenous mouse p27 because it migrates slower on SDS gels. (G) Lysates from NIH 3T3 cells transfected with wild-type p27 (wt) or p27(S10A) mutant were subjected to immunoprecipitation with either anti-p27 antibody (lanes 1 and 2) or anti-phospho-Ser10 antibody (lanes 3 and 4). The precipitated proteins were analyzed by immunoblotting with anti-p27 antibody. Download figure Download PowerPoint A site-directed mutagenesis approach was used to identify the phosphorylation sites in p27. Because of the important role of proline-directed kinases in cell cycle regulation, we mutated the three serine and threonine residues (Ser10, Ser178 and Thr187) that are immediately followed by a Pro residue. Rat1 cells were transfected with expression vectors encoding wild-type or alanine mutants of hemagglutinin (HA)-tagged p27, metabolically labeled with [32P]orthophosphate, and the ectopically expressed p27 was immunoprecipitated with anti-HA antibody. The extent of phosphorylation of p27-HA was analyzed by autoradiography and normalized to the amount of immunoprecipitated protein. As shown in Figure 1C, replacement of Ser10 by Ala resulted in a strong reduction in 32P incorporation, indicating that Ser10 is a major site of p27 phosphorylation. No significant change in the total level of p27-HA phosphorylation was observed for the p27(S178A) mutant, while the p27(T187A) mutant showed a decrease in phosphorylation only when cyclin E and Cdk2 were co-transfected to phosphorylate this site (data not shown). The phosphopeptide map of HA-tagged p27 was comparable with that of the endogenous protein, except for the presence of four additional minor spots labeled 4–7 (Figure 1D). The physiological significance of these four minor phosphopeptides remains to be established, as they were reproducibly detected only upon overexpression of the protein. The two major phosphopeptides (spots 2 and 3) observed in wild-type p27 were absent in the p27(S10A) mutant, confirming that p27 is phosphorylated on Ser10 in vivo (Figure 1D). Further analysis revealed that spots 2 and 3 are derived from the same peptide containing phosphorylated Ser10 exclusively (data not shown). During the course of this work, Ishida et al. (2000) also reported that Ser10 is a major phosphorylation site of p27. To explore the regulation of p27 phosphorylation on Ser10, we raised a phospho-specific antibody against a synthetic peptide that spans the phosphorylated Ser10 residue of p27. The specificity of the purified antibody is illustrated in Figure 1E. The purified antibody recognized p27 recombinant protein only after phosphorylation in vitro by the MAP kinase ERK2 (Figure 1E, lane 3, top), and substitution of Ser10 by Ala completely abolished the immunodetection of the protein (Figure 1E, lane 4). In addition, p27 was not recognized by the anti-phospho-Ser10 antibody when phosphorylated in vitro by cyclin E–Cdk2 (Figure 1E, lanes 1 and 2, top), which phosphorylated both wild-type and p27(S10A) protein on Thr187 efficiently instead, as detected with a phospho-Thr187-specific antibody (Carrano et al., 1999; Montagnoli et al., 1999; Figure 1E, lanes 1 and 2, middle). We further tested the specificity of the antibody on cellular p27. Asynchronous NIH 3T3 cells were transfected with expression plasmids encoding wild-type p27, p27(S10A), p27(S10D) or p27(S10E) mutants. Immunoblot analysis of total cell lysates revealed that the anti-phospho-Ser10 antibody recognizes endogenous murine p27 as well as ectopically expressed human wild-type p27 (which migrates slower than mouse p27), but fails to detect the p27(S10A) mutant, thereby confirming the specificity of the antibody for phosphorylated p27 (Figure 1F, lanes 1 and 2). Notably, this antibody recognized p27(S10D) and p27(S10E) mutants (Figure 1F, lanes 3 and 4), indicating that these substitutions effectively mimic the negative charge of the phosphate in position 10. The specificity of the antibody was also demonstrated in immunoprecipitation (Figure 1G) and in immunofluorescence experiments (not shown). Phosphorylation of p27 on Ser10 is predominant in G0/G1 cells To determine whether p27 is phosphorylated on Ser10 in a cell cycle-dependent manner, quiescent primary T lymphocytes were stimulated with mitogens (Figure 2A). Progression of T lymphocytes into the cell cycle was associated with an increase in the levels of cyclin E and cyclin A, and a gradual decrease in the abundance of p27. The decrease in p27 expression coincided with a marked reduction in phosphorylation of the protein on Ser10. Similar results were obtained in synchronized NIH 3T3 and HeLa cells (data not shown). To demonstrate that changes in p27 phospho-Ser10 immunoreactivity reflect changes in the stoichiometry of Ser10 phosphorylation, not just variations in the amount of protein, extracts of NIH 3T3 cells, synchronized in G0/G1 by serum starvation and released into the cell cycle for the times indicated, were assayed for their ability to phosphorylate a constant amount of recombinant p27 protein in vitro. Immunoblot analysis with anti-phospho-Ser10- and anti-phospho-Thr187-specific antibodies revealed that the kinase activity responsible for Ser10 phosphorylation was elevated in G0/G1 and declined as cells advanced to S phase (16 h after serum re-addition; Figure 2B). Figure 2.Phosphorylation of p27 on Ser10 is cell cycle regulated. (A) Mouse T lymphocytes were stimulated with concanavalin A and interleukin-2 for the times indicated. Cellular extracts were analyzed by immunoblotting with specific antibodies to the indicated proteins. (B) NIH 3T3 fibroblasts were synchronized in G0/G1 by serum starvation and restimulated with 10% serum for the times indicated. Cell extracts were prepared and incubated with recombinant purified p27 in kinase assay buffer as described in Materials and methods. The reaction products were analyzed by immunoblotting using anti-phospho-Ser10- (upper) and Thr187 (lower)-specific antibodies. (C) MCF-7 breast cancer cells were depleted of estradiol and treated with 1 μM tamoxifen for the times indicated (lanes 1–4). After 48 h, arrested cells were stimulated to re-enter the cell cycle by addition of 500 nM estradiol for the intervals of time indicated (lanes 5–9). Cell extracts were analyzed by immunoblotting with specific antibodies to the indicated proteins. (D) Cell extracts (50 μg protein) from proliferating (AS) or tamoxifen-treated (TMX) MCF-7 cells were sequentially immunoprecipitated four times with anti-phospho-Ser10 antibody. The resultant pellets (lanes 1–4 and 6–9) and the final supernatant (lanes 5 and 10), as well as the input lysate (lanes 11 and 12), were analyzed by immunoblotting with anti-p27 antibody. (E) MCF-7 cell extracts described in (C) (lanes 1–4) were subjected to immunoprecipitation with anti-p27 antibody and analyzed by immunoblotting with specific antibodies to the indicated proteins. (F) The same MCF-7 cell extracts described in (E) were subjected to immunoprecipitation with anti-phospho-Ser10 antibody and analyzed by immunoblotting as indicated. Download figure Download PowerPoint Additional evidence for the cell cycle regulation of p27 phosphorylation on Ser10 was obtained in MCF-7 breast cancer cells. Asynchronous MCF-7 cells were arrested in G1 by treatment with tamoxifen combined with estrogen deprivation, as monitored by cyclin A down-regulation (Figure 2C, lanes 1–4). In this cell line, the total abundance of p27 did not change significantly during cell cycle exit, but the level of its phosphorylation on Ser10 increased significantly (Figure 2C, lanes 1–4). When the cells were re-stimulated to enter the cell cycle by addition of estradiol, p27 protein abundance decreased in parallel with a decrease in Ser10 phosphorylation (Figure 2C, lanes 5–9). To quantify more accurately the extent of p27 Ser10 phosphorylation in exponentially proliferating versus G1-arrested MCF-7 cells, cellular lysates were subjected to successive rounds of immunoprecipitation with the anti-phospho-Ser10-specific antibody to ensure that all the phosphorylated p27 was precipitated effectively. The pellets and the final supernatant were analyzed by immunoblotting with anti-p27 antibody to estimate the percentage of p27 phosphorylated on Ser10 (Figure 2D). From four different experiments using both MCF-7 and NIH 3T3 cells and densitometry analysis of the bands, we calculated that ∼33% of total p27 was phosphorylated on Ser10 in arrested cells, while only 5% of the protein was phosphorylated at this site in proliferating cells (Figure 2D and data not shown). Importantly, we determined that the fraction of p27 phosphorylated on Ser10 is still able to bind to cyclin D1 and cyclin E, and that the anti-phospho-Ser10 antibody effectively precipitates cyclin–Cdk complexes together with phosphorylated p27 (Figure 2E and F). Phosphorylation on Ser10 stabilizes p27 in vivo but does not affect p27 ubiquitylation in vitro Phosphorylation of p27 on Thr187 by cyclin E–Cdk2 promotes p27 degradation by the ubiquitin–proteasome pathway (Carrano et al., 1999; Montagnoli et al., 1999; Sutterluty et al., 1999). While this work was in progress, Nakayama and co-workers suggested that phosphorylation on Ser10 stabilizes p27 (Ishida et al., 2000). We confirmed that the phospho-mimetic p27(S10D) mutant was more stable than the non-phosphorylatable p27(S10A) mutant; the estimated half-lives of p27(S10D) and p27(S10A) were 9.5 and 5.9 h, respectively, whereas wild-type p27 was degraded at a rate intermediate between that of p27(S10D) and p27(S10A) mutants (Figure 3). Figure 3.Phosphorylation of p27 on Ser10 increases its half-life. (A) Rat1 cells were transfected with HA-tagged p27 wild-type or p27(Ser10) mutants and serum deprived for 24 h. The cells were then pulse-labeled for 2 h with [35S]methionine or [35S]cysteine and chased for the times indicated in fresh medium containing 10% serum. Cell lysates were subjected to immunoprecipitation with anti-HA antibody and the labeled p27 protein was analyzed by SDS–gel electrophoresis and fluorography. (B) Densitometric analysis of p27 degradation rate. Data points correspond to the experiment shown in (A). The half-life values of wild-type and mutant p27 represent the mean ± SEM of four independent pulse–chase experiments. Download figure Download PowerPoint We next explored the molecular basis for the stabilization effect of Ser10 phosphorylation on p27 turnover in vivo. We first examined the impact of the S10D mutation on the phosphorylation of p27 on Thr187 and its association with the F-box protein Skp2. For these experiments, cellular extracts prepared from Rat1 cells transiently transfected with the different p27-HA mutants were incubated with in vitro translated Skp2. The p27-HA complexes were isolated by anti-HA immunoprecipitation and analyzed by autoradiography and immunoblotting. Mutation of Ser10 to Ala or Asp did not affect the ability of p27-HA to associate with Skp2, while under the same conditions, the T187A mutant failed to co-precipitate labeled Skp2 (Figure 4). These experiments also showed that p27 Ser10 mutants are phosphorylated on Thr187 to the same extent as the wild-type protein, consistent with the results of in vitro phosphorylation (Figure 1E). These data indicate that phosphorylation of Ser10 does not impact on the targeting of p27 by the SCFSkp2 ubiquitin–protein ligase complex. Figure 4.Phosphorylation of p27 on Ser10 does not affect its interaction with the F-box protein Skp2. Rat1 cells were transfected with HA-tagged p27 wild-type or p27 mutants and treated for 5 h with 5 μM MG132 (to increase the amount of p27 phosphorylated on Thr187). Lysates were prepared, incubated with in vitro translated [35S]Myc6Skp2 for 3 h and subjected to immunoprecipitation with anti-HA antibody. The precipitated proteins were resolved by SDS–gel electrophoresis, transferred to nitrocellulose membrane and visualized by autoradiography (upper). The membrane was further subjected to immunoblot analysis with antibodies to p27 (middle) and phospho-Thr187 (lower). Download figure Download PowerPoint To confirm that p27 Ser10 mutants are still ubiquitylated, we used an established purified recombinant system (Ganoth et al., 2001), which includes the four components of the SCF ubiquitin ligase (Skp1, Cul1, Skp2 and Roc1/Rbx1), Cks1, the ubiquitin-conjugating enzyme Ubc3, the ubiquitin-activating enzyme E1 and cyclin E–Cdk2 complex. We performed in vitro ubiquitylation assays using this purified system and in vitro translated 35S-labeled wild-type p27 and Ser10 mutants as substrates (Figure 5). Both p27(S10A) and p27(S10D) mutants were efficiently ubiquitylated in vitro to the same extent as the wild-type protein (∼75% of ubiquitin conjugation after 1 h reaction). Accordingly, both Ser10 mutants were degraded in vitro with the same kinetics as the wild-type protein (data not shown). Thus, phosphorylation on Ser10 does not affect p27 ubiquitylation or stability in vitro. Figure 5.Phosphorylation on Ser10 does not affect p27 in vitro ubiquitylation. 35S-labeled in vitro translated p27 wild-type (lanes 1–5), p27(S10A) mutant (lanes 6–10) or p27(S10D) mutant (lanes 11–15) were subjected to a ubiquitylation reaction for the times indicated using purified components as described in Materials and methods. The reaction products were analyzed by SDS–gel electrophoresis followed by autoradiography. The bracket on the left side marks a ladder of bands corresponding to polyubiquitylated p27. Download figure Download PowerPoint Phosphorylation on Ser10 is necessary for the nuclear export of p27 Since Ser10 mutants of p27 are ubiquitylated and degraded efficiently in vitro while expression of p27(S10D) is stabilized in vivo, we reasoned that in the cell-free system a regulatory step was missing. One such regulatory mechanism could be p27 transport, since the subcellular localization of p27 appears to play an important role in controlling its stability (Tomoda et al., 1999). Therefore, we evaluated the role of Ser10 phosphorylation on p27 cellular localization. We first examined the subcellular distribution of endogenous p27 during cell cycle re- entry of Rat1 cells. In quiescent cells, p27 was localized almost exclusively in the nucleus (Figure 6A, t = 0 h). Stimulation with serum promoted the redistribution of a fraction of p27 to the cytoplasm in all cells (Figure 6A, t = 4 and 8 h, top). After 12 h of stimulation, the staining of p27 decreased in parallel with the abundance of the protein (data not shown). Detailed kinetic studies revealed that the cytoplasmic relocalization of p27 precedes activation of Cdk2, induction of Skp2 and degradation of the bulk of p27 protein (Figure 6B). Figure 6.Mitogenic stimulation promotes the nuclear export of a fraction of p27. (A) Rat1 cells were made quiescent by serum starvation and restimulated to enter the cell cycle by addition of 10% serum for the times indicated. The subcellular localization of endogenous p27 was determined by fluorescence microscopy. When present, LMB (2 ng/ml) was added simultaneously with serum. (B) Rat1 cells were made quiescent and restimulated with serum for the times indicated. The expression of p27 and Skp2 was monitored by immunoblotting using specific antibodies. The activity of Cdk2 was assayed using histone H1 as substrate. Download figure Download PowerPoint To confirm these findings, p27-HA was ectopically expressed in Rat1 cells and its localization was visualized by immunofluorescence with an antibody against the HA epitope. Under the transfection conditions used for these studies, ∼50% of cells expressing p27 enter S phase after 20 h of serum stimulation (data not shown). In serum-starved cells, p27-HA was predominantly nuclear (Figure 7A and B). Addition of serum caused a time-dependent translocation of p27-HA to the cytoplasm, with ∼35% of cells displaying cytoplasmic staining after 24 h (Figure 7A and B). A higher proportion of cells exhibited cytoplasmic redistribution of endogenous p27 as compared with ectopic p27-HA, which possibly reflects the inhibitory effect of ectopic p27 expression on cell cycle progression. Treatment of Rat1 cells with leptomycin B (LMB), a specific inhibitor of CRM1-dependent nuclear export (Nishi et al., 1994), completely prevented the cytoplasmic relocalization of endogenous and ectopic p27 (Figures 6A, bottom and 7C). Figure 7.Phosphorylation of Ser10 is necessary for the nuclear to cytoplasmic translocation of p27 upon cell cycle re-entry. (A) Rat1 cells were transfected with HA-tagged p27 wild-type or Ser10 mutants and serum starved for 24 h. The cells were then stimulated with 10% serum for the times indicated. The subcellular localization of ectopic p27 was determined by fluorescence microscopy after staining with anti-HA antibody. (B) Quantitative evaluation of the cellular localization of p27 wild-type and Ser10 mutants. At least 300 cells were scored for each coverslip. The results are expressed as the percentage of cells showing both nuclear and cytoplasmic staining. The graph represents the mean ± SEM of three separate experiments. (C) Effect of LMB. Rat1 cells were transfected with HA-tagged p27 wild-type or Ser10 mutants. After 24 h, the cells were made quiescent and restimulated with 10% serum for 24 h in the absence or presence of 2 ng/ml LMB. The results are expressed as the percentage of cells showing both nuclear and cytoplasmic staining. The graph represents the mean ± SEM of five separate experiments. Download figure Download PowerPoint We next evaluated the effect of Ser10 mutations on the subcellular localization of p27. The phospho-mimetic p27(S10D) mutant behaved as the wild-type protein and was relocalized to the cytoplasm upon serum stimulation (Figure 7A and B). In contrast, the Ser10 to Ala mutation almost completely abolished the cytoplasmic redistribution of the protein, indicating that phosphorylation of Ser10 is necessary for nuclear export of p27. We also tested the effect of LMB on the localization of Ser10 mutants. Treatment with LMB completely inhibited the serum-dependent relocalization of the p27(S10D) mutant to the cytoplasm (Figure 7C). Finally, we examined the localization of endogenous p27 phosphorylated on Ser10 in MCF-7 cells. In quiescent tamoxifen-treated cells, ∼90% of total p27 was detected in the nuclear compartment (Figure 8). Treatment