p27Kip1 Regulates T Cell Proliferation
2001; Elsevier BV; Volume: 276; Issue: 24 Linguagem: Inglês
10.1074/jbc.m009788200
ISSN1083-351X
AutoresSubhra Mohapatra, Deepak Agrawal, W. J. Pledger,
Tópico(s)Cancer Research and Treatments
ResumoOur studies addressed the mechanism by which serum acts in conjunction with T cell receptor (TCR) agonists to promote the proliferation of primary splenic T cells. When added to resting splenocytes, TCR agonists initiated G0/G1 traverse and activated cyclin D3-cdk6 complexes in a serum-independent manner. On the other hand, both TCR agonists and 10% serum were required for the activation of cyclin E-cdk2 and cyclin A-cdk2 complexes and the entry of cells into S phase. Serum facilitated cdk2 activation by maximizing the extent and extending the duration of the TCR-initiated down-regulation of the cdk2 inhibitor, p27Kip1. Although p27Kip1 levels were reduced (albeit submaximally) in cells stimulated in serum-deficient medium, nearly all of the cdk2 complexes in these cells contained p27Kip1. In contrast, in cells receiving TCR agonist and 10% serum, little if any p27Kip1was present in cyclin-cdk2 complexes. Unlike wild-type splenocytes, p27Kip1-null splenocytes did not require serum for cdk2 activation or S phase entry whereas loss of the related cdk2 inhibitor, p21Cip1, did not override the serum dependence of these responses. We also found that cdk2 activation was both necessary and sufficient for maximal expression of cdk2 protein. These studies provide a mechanistic basis for the serum dependence of T cell mitogenesis. Our studies addressed the mechanism by which serum acts in conjunction with T cell receptor (TCR) agonists to promote the proliferation of primary splenic T cells. When added to resting splenocytes, TCR agonists initiated G0/G1 traverse and activated cyclin D3-cdk6 complexes in a serum-independent manner. On the other hand, both TCR agonists and 10% serum were required for the activation of cyclin E-cdk2 and cyclin A-cdk2 complexes and the entry of cells into S phase. Serum facilitated cdk2 activation by maximizing the extent and extending the duration of the TCR-initiated down-regulation of the cdk2 inhibitor, p27Kip1. Although p27Kip1 levels were reduced (albeit submaximally) in cells stimulated in serum-deficient medium, nearly all of the cdk2 complexes in these cells contained p27Kip1. In contrast, in cells receiving TCR agonist and 10% serum, little if any p27Kip1was present in cyclin-cdk2 complexes. Unlike wild-type splenocytes, p27Kip1-null splenocytes did not require serum for cdk2 activation or S phase entry whereas loss of the related cdk2 inhibitor, p21Cip1, did not override the serum dependence of these responses. We also found that cdk2 activation was both necessary and sufficient for maximal expression of cdk2 protein. These studies provide a mechanistic basis for the serum dependence of T cell mitogenesis. T cell receptor interleukin-2 interleukin-2 receptor cyclin-dependent kinase CDK inhibitor phosphate-buffered saline glutathione S-transferase concanavalin A T cell proliferation, and the consequent expansion of T cell populations, are fundamental events in the generation of an immune response. In vivo, T cell mitogenesis is initiated by the interaction of T cell receptors (TCRs)1 with peptides coupled to major histocompatibility antigen complexes located on the surface of antigen-presenting cells (1Davis M.M. Bjorkman P.J. Nature. 1988; 334: 395-402Crossref PubMed Scopus (2466) Google Scholar). In vitro, the actions of natural ligands are mimicked by a variety of agents including anti-CD3, calcium ionophore, and lectins such as concanavalin A (ConA) and phytohemagglutinin. TCR stimulation allows resting T cells to enter G1 and induces the expression of interleukin-2 (IL-2) and the α subunit of the IL-2 receptor (IL-2Rα) (2Meuer S.C. Hussey R.E. Cantrell D.A. Hodgdon J.C. Schlossman S.F. Smith K.A. Reinherz E.L. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 1509-1513Crossref PubMed Google Scholar, 3Leonard W.J. Kronke M. Peffer N.J. Depper J.M. Greene W.C. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 6281-6285Crossref PubMed Scopus (133) Google Scholar, 4Karnitz L.M. Abraham R.T. Adv. Immunol. 1996; 61: 147-199Crossref PubMed Google Scholar). In the presence of IL-2, IL-2Rα combines with IL-2Rβ and IL-2Rγ to form the high affinity IL-2R complex, which elicits the secondary signals required for continued G1 traverse and S phase entry. T cell proliferation also requires additional factors, many of which are routinely provided to cultured T cells in the form of serum (5Herzberg V.L. Smith K.A. J. Immunol. 1987; 139: 998-1004PubMed Google Scholar). Although previous studies have shown that serum enhances the production of IL-2 and IL-2Rα (5Herzberg V.L. Smith K.A. J. Immunol. 1987; 139: 998-1004PubMed Google Scholar), the potential contribution of serum to other cell cycle-regulatory processes has yet to be explored. In T cells, as in other cell types, cell cycle traverse is governed by the ordered activation of cyclin-dependent kinases (CDKs) (6Sherr C.J. Science. 1996; 274: 1672-1677Crossref PubMed Scopus (4976) Google Scholar). Activation of the CDKs requires their interaction with cyclins, whose levels fluctuate during the cell cycle, and their phosphorylation at specific threonine residues by constitutively expressed cyclin-activating kinases. CDKs also interact with a group of proteins collectively termed CDK inhibitors (CKIs). CKI levels, like cyclin levels, vary during the cell cycle and thus contribute to the periodicity of CDK activation (7Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5144) Google Scholar). Two classes of CKIs have been defined: the INK proteins, which target cdk4 and cdk6, and the Cip/Kip proteins, which inactivate cdk2-containing complexes (8Ruas M. Peters G. Biochim. Biophys. Acta. 1998; 1378: F115-F177Crossref PubMed Scopus (874) Google Scholar, 9Hengst L. Reed S.I. Curr. Top. Microbiol. Immunol. 1998; 227: 25-41PubMed Google Scholar). Traverse of G0/G1 and entry into S phase is controlled by the sequential activation of complexes containing the D cyclins and cdk4 or cdk6, cyclin E and cdk2, and cyclin A and cdk2. Addition of mitogens to G0-arrested cultures induces the expression of the D cyclins (D1, D2, and D3) and, in some cell types (e.g. T cells), of cdk4 and cdk6 (10Winston J.T. Pledger W.J. Mol. Biol. Cell. 1993; 4: 1133-1144Crossref PubMed Scopus (89) Google Scholar, 11Ajchenbaum F. Ando K. DeCaprio J.A. Griffin J.D. J. Biol. Chem. 1993; 268: 4113-4119Abstract Full Text PDF PubMed Google Scholar, 12Kwon T.K. Buchholz M.A. Ponsalle P. Chrest F.J. Nordin A.A. J. Immunol. 1997; 158: 5642-5648PubMed Google Scholar, 13Modiano J.F. Domenico J. Szepesi A. Terada N. Lucas J.J. Gelfand E.W. Ann. N. Y. Acad. Sci. 1995; 766: 134-148Crossref PubMed Scopus (14) Google Scholar). Mitogenic stimulation also down-regulates p27Kip1, a Cip/Kip protein that accumulates in quiescent cells (14Firpo E.J. Koff A. Solomon M.J. Roberts J.M. Mol. Cell. Biol. 1994; 14: 4889-4901Crossref PubMed Scopus (275) Google Scholar, 15Kato J.Y. Matsuoka M. Polyak K. Massague J. Sherr C.J. Cell. 1994; 79: 487-496Abstract Full Text PDF PubMed Scopus (709) Google Scholar, 16Agrawal D. Hauser P. McPherson F. Dong F. Garcia A. Pledger W.J. Mol. Cell. Biol. 1996; 16: 4327-4336Crossref PubMed Scopus (161) Google Scholar, 17Winston J. Dong F. Pledger W.J. J. Biol. Chem. 1996; 271: 11253-11260Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar). D cyclin complexes become active in mid G1 and phosphorylate the anti-oncogene Rb (18Lundberg A.S. Weinberg R.A. Mol. Cell. Biol. 1998; 18: 753-761Crossref PubMed Scopus (856) Google Scholar). Pre-existing cyclin E-cdk2 complexes become active after p27Kip1 levels decrease and further phosphorylate Rb in late G1. When phosphorylated by these kinases, Rb no longer represses the activity of the E2F transcription factors, and a variety of genes, including those encoding cyclins E and A, are expressed (19Weinberg R.A. Cell. 1995; 81: 323-330Abstract Full Text PDF PubMed Scopus (4311) Google Scholar, 20DeGregori J. Kowalik T. Nevins J.R. Mol. Cell. Biol. 1995; 15: 4215-4224Crossref PubMed Scopus (839) Google Scholar, 21Chellappan S.P. Hiebert S. Mudryj M. Horowitz J.M. Nevins J.R. Cell. 1991; 65: 1053-1061Abstract Full Text PDF PubMed Scopus (1095) Google Scholar). At this point, cells pass through the restriction point in late G1 and in a manner dependent on cyclin E-cdk2 and cyclin A-cdk2 activity enter and traverse S phase (22Planas-Silva M.D. Weinberg R.A. Curr. Opin. Cell Biol. 1997; 9: 768-772Crossref PubMed Scopus (208) Google Scholar). The drop in p27Kip1 levels and the consequent activation of pre-existing and newly formed cyclin E-cdk2 and cyclin A-cdk2 complexes are critical aspects of G0/G1 traverse. For example, prevention of p27Kip1 down-regulation by agents such as rapamycin and cyclic AMP blocks cdk2 activation and cell cycle traverse, as does ectopic expression of p27Kip1 (15Kato J.Y. Matsuoka M. Polyak K. Massague J. Sherr C.J. Cell. 1994; 79: 487-496Abstract Full Text PDF PubMed Scopus (709) Google Scholar,23Nourse J. Firpo E. Flanagan W.M. Coats S. Polyak K. Lee M.H. Massague J. Crabtree G.R. Roberts J.M. Nature. 1994; 372: 570-573Crossref PubMed Scopus (904) Google Scholar, 24LaBaer J. Garrett M.D. Stevenson L.F. Slingerland J.M. Sandhu C. Chou H.S. Fattaey A. Harlow E. Genes Dev. 1997; 11: 847-862Crossref PubMed Scopus (1221) Google Scholar, 25Zhang X. Wharton W. Donovan M. Coppola D. Croxton R. Cress W.D. Pledger W.J. Mol. Biol. Cell. 2000; 11: 2117-2130Crossref PubMed Scopus (30) Google Scholar). Conversely, ablation of p27Kip1 function in both whole animals and cultured cells has been shown to promote proliferation by impeding G0 arrest (26Coats S. Flanagan W.M. Nourse J. Roberts J.M. Science. 1996; 272: 877-880Crossref PubMed Scopus (651) Google Scholar, 27Rivard N. L'Allemain G. Bartek J. Pouyssegur J. J. Biol. Chem. 1996; 271: 18337-18341Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar). The mitogen-induced decrease in p27Kip1 levels is a complex and not fully understood process that involves translational inhibition, accelerated degradation, and perhaps also transcriptional repression (12Kwon T.K. Buchholz M.A. Ponsalle P. Chrest F.J. Nordin A.A. J. Immunol. 1997; 158: 5642-5648PubMed Google Scholar, 16Agrawal D. Hauser P. McPherson F. Dong F. Garcia A. Pledger W.J. Mol. Cell. Biol. 1996; 16: 4327-4336Crossref PubMed Scopus (161) Google Scholar, 28Hengst L. Reed S.I. Science. 1996; 271: 1861-1864Crossref PubMed Scopus (822) Google Scholar, 29Sheaff R.J. Groudine M. Gordon M. Roberts J.M. Clurman B.E. Genes Dev. 1997; 11: 1464-1478Crossref PubMed Scopus (794) Google Scholar, 30Vlach J. Hennecke S. Amati B. EMBO J. 1997; 16: 5334-5344Crossref PubMed Scopus (607) Google Scholar, 31Appleman L.J. Berezovskaya A. Grass I. Boussiotis V.A. J. Immunol. 2000; 164: 144-151Crossref PubMed Scopus (167) Google Scholar). Additional studies have shown that p27Kip1 down-regulation is not sufficient for cdk2 activation, which also requires sequestration of residual p27Kip1 molecules by complexes containing the D cyclins and their CDK partners (7Sherr C.J. Roberts J.M. Genes Dev. 1999; 13: 1501-1512Crossref PubMed Scopus (5144) Google Scholar). Whether cyclin D/CDK activity is also repressed by p27Kip1 is unclear, because past reports differ in this regard (24LaBaer J. Garrett M.D. Stevenson L.F. Slingerland J.M. Sandhu C. Chou H.S. Fattaey A. Harlow E. Genes Dev. 1997; 11: 847-862Crossref PubMed Scopus (1221) Google Scholar, 32Cheng M. Olivier P. Diehl J.A. Fero M. Roussel M.F. Roberts J.M. Sherr C.J. EMBO J. 1999; 18: 1571-1583Crossref PubMed Scopus (974) Google Scholar). Previous studies have shown that TCR agonists down-regulate p27Kip1 and activate cdk6 and cdk2 when added to resting T cells (11Ajchenbaum F. Ando K. DeCaprio J.A. Griffin J.D. J. Biol. Chem. 1993; 268: 4113-4119Abstract Full Text PDF PubMed Google Scholar, 12Kwon T.K. Buchholz M.A. Ponsalle P. Chrest F.J. Nordin A.A. J. Immunol. 1997; 158: 5642-5648PubMed Google Scholar, 13Modiano J.F. Domenico J. Szepesi A. Terada N. Lucas J.J. Gelfand E.W. Ann. N. Y. Acad. Sci. 1995; 766: 134-148Crossref PubMed Scopus (14) Google Scholar, 33Meyerson M. Harlow E. Mol. Cell. Biol. 1994; 14: 2077-2086Crossref PubMed Scopus (734) Google Scholar). However, these studies were done on cells stimulated in the presence of 10% serum. Thus, they leave open the possibility that at least some of these responses are induced by TCR agonists in a serum-dependent manner. This is an important issue, because it addresses a potential mechanism by which serum might promote T cell proliferation. In the studies described here, we examined the contribution of serum to cyclin/CDK activation in primary splenocytes and purified T cells exposed to TCR agonists such as ConA or anti-CD3. We show that serum acts in conjunction with mitogenic amounts of ConA or anti-CD3 to induce the sustained down-regulation of p27Kip1 and the activation of cdk2-containing complexes. Consistent with the premise that serum facilitates T cell proliferation by targeting p27Kip1, we also find that splenocytes lacking p27Kip1 no longer require serum for cdk2 activation or S phase entry. A single cell suspension of mouse spleen cells was prepared by passage through nylon mesh. Red cells were depleted using a whole blood erythrocyte lysing kit (R&D Systems). For isolation of purified T cells, spleen cell suspensions were loaded onto T cell enrichment columns, and T cells were purified via high affinity negative selection as specified by the manufacturer (R&D Systems). Splenocytes and purified T cells were plated at 107 cells/ml and 5 × 106cells/ml, respectively, in RPMI 1640 supplemented with 2 mml-glutamine, 50 units/ml penicillin, and 10% fetal bovine serum. For assessment of DNA synthesis, triplicate cultures in microtiter plates were pulsed with 1 μCi/ml [3H]thymidine (PerkinElmer Life Sciences) for the indicated times. Incorporation was determined by scintillation counting, and each experiment was repeated at least twice. To determine cell cycle position, cells were washed with cold phosphate-buffered saline (PBS) and fixed with 70% ethanol overnight at 4 °C. Fixed cells were resuspended in PBS containing 1% bovine serum albumin, 0.5% Tween 20, 1 μg/ml propidium iodide, and 1 μg/ml RNase A and incubated at room temperature for 2 h. Total DNA content was analyzed on an EPICS 753 flow cytometer (Coulter Electronics, Inc.). Cells were rinsed with cold PBS and lysed in buffer containing 50 mm Hepes (pH 7.5), 150 mm NaCl, 1 mm EDTA, 1 mm EGTA, 10% glycerol, 0.5% Nonidet P-40, 1 mm dithiothreitol, 0.1 mm phenylmethylsulfonyl fluoride, 2.5 μg/ml leupeptin, 0.5 mm NaF, and 0.1 mm sodium vanadate (lysis buffer). After a 30-min incubation, insoluble material was removed by centrifugation. Proteins (30 μg) were resolved by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane or polyvinylidene difluoride membrane for cdk2. Blots were blocked in PBST (PBS plus 0.05% Tween) containing 5% instant milk and incubated with primary antibody in PBST. Proteins recognized by the antibody were detected by enhanced chemiluminescence using a horseradish peroxidase-coupled secondary antibody as specified by the manufacturer (Pierce). Cell extracts (80 μg of protein in lysis buffer) were incubated with antibody to the indicated cyclin or CDK for 4–12 h at 4 °C and subsequently with protein A-agarose beads. Immune complexes were washed twice with lysis buffer and once with either histone kinase buffer (50 mmTris (pH 7.4), 10 mm MgCl2, 1 mmdithiothreitol) or Rb kinase buffer (50 mm Tris (pH 7.4), 10 mm MgCl2, 5 mmMnCl2, 1 mm dithiothreitol). Washed complexes were incubated in 15 μl of kinase buffer containing 20 μm ATP, 0.1 μCi/ml [γ-32P]ATP, and either 100 μg/ml histone H1 (Roche Molecular Biochemicals) for 10 min at 37 °C or 2.5 μg/ml GST-Rb for 30 min at 30 °C. Reactions were stopped by boiling in Laemmli buffer and proteins were separated by SDS-polyacrylamide gel electrophoresis. Radiolabeled proteins were visualized by autoradiography. For p27Kip1 immunodepletion, cell extracts (150 μg of protein) were incubated with antibody to p27Kip1 (or for mock depletion, with preimmune serum) for 4 h at 4 °C, and immune complexes were removed by centrifugation with protein A-agarose beads. Removal of p27Kip1 was confirmed by Western blotting. ConA and anti-CD3 were obtained from Sigma Chemical Co. and PharMingen, respectively. Antibodies to cyclin D3, cyclin E, and cdk6 were purchased from Santa Cruz Biotechnology. p27Kip1 and cdk2 antibodies were from Transduction Laboratories. Cyclin A antibody was provided by E. Leof and GST-Rb by D. Cress. p27kip-deficient mice were obtained from by A. Koff (34Kiyokawa H. Kineman R.D. Manova-Todorova K.O. Soares V.C. Hoffman E.S. Ono M. Khanam D. Hayday A.C. Frohman L.A. Koff A. Cell. 1996; 85: 721-732Abstract Full Text Full Text PDF PubMed Scopus (1149) Google Scholar), and p21Cip1-deficient mice were obtained from T. Jacks (35Brugarolas J. Chandrasekaran C. Gordon J.I. Beach D. Jacks T. Hannon G.J. Nature. 1995; 377: 552-557Crossref PubMed Scopus (1150) Google Scholar). Initial experiments confirmed that TCR agonists and serum act in a concerted manner to stimulate the proliferation of primary splenocytes and purified T cells. The cells used in these experiments were derived from Balb/c mice. As assessed by incorporation of [3H]thymidine into DNA, resting splenocyte populations did not initiate DNA synthesis when treated for 40 h with serum (0.1–10%) alone or with 0.1% serum and various concentrations of ConA (Fig. 1 A) or anti-CD3 (Fig. 1 B). However, co-stimulation of splenocytes with optimal amounts of serum (10%) and either ConA (2.5 μg/ml) or anti-CD3 (10 μg/ml) produced a significant increase in [3H]thymidine incorporation. Anti-CD3 and serum also synergistically stimulated the cell cycle traverse of T cell-enriched populations containing greater than 95% T cells (Fig. 1 C). Maximal [3H]thymidine incorporation occurred in T cell cultures receiving 5 μg/ml anti-CD3 and 10% serum. The lower amount of [3H]thymidine incorporation seen in cultures receiving 10 μg/ml as compared with 5 μg/ml anti-CD3 presumably reflects the fact that many of the cells in the former condition have already exited S phase at the time of the pulse (36Williams M.E. Lichtman A.H. Abbas A.K. J. Immunol. 1990; 144: 1208-1214PubMed Google Scholar). Serum did not simply substitute for IL-2, because ConA (2.5 μg/ml) and IL-2 (500 units/ml) did not induce DNA synthesis in medium containing 0.1% serum (data not shown). When exposed to 10% serum and either ConA or anti-CD3, quiescent splenocytes and T cells entered S phase after a lag of 20–24 h (Fig. 1, D, E, andF). As determined by fluorescence-activated cell sorting analysis, 30–40% of the cells were in S phase or G2/M at 40 h after stimulation. In contrast, cultures stimulated in medium containing 0.1% serum did not appreciably initiate DNA synthesis during the experimental period. These findings show that mixed splenocyte populations respond similarly to ConA and anti-CD3 and behave similarly to purified T cells. The next set of experiments addressed the possibility that serum promoted T cell proliferation by facilitating cyclin/CDK activation. In these experiments, quiescent splenocytes and T cells were stimulated with ConA or anti-CD3 in the presence of 10% or 0.1% serum. These serum concentrations were chosen because they allow maximal and minimal S phase entry, respectively (see Fig. 1). Serum at 0.1% (rather than no serum) was used as the negative control, because serum at this concentration prevents adsorption of IL-2 to culture dishes (5Herzberg V.L. Smith K.A. J. Immunol. 1987; 139: 998-1004PubMed Google Scholar). However, events induced by ConA or anti-CD3 in medium containing 0.1% serum were also induced by these agents in serum-free medium (data not shown) and thus are considered serum-independent. Cyclin, CDK, and p27Kip1 levels were determined by Western blotting, and cyclin/CDK activity was assessed byin vitro kinase assay. It is noted that treatment of resting cells with 10% serum in the absence of TCR agonist had no effect on the responses examined below (data not shown). p27Kip1 was present in quiescent splenocytes, and its levels decreased to a similar extent in cells stimulated with ConA and either 10% or 0.1% serum for 10 h (Fig.2 A; 68% versus51% decrease, respectively). After this time, p27Kip1levels fell much more precipitously in cultures receiving ConA and 10% as compared with 0.1% serum. For example, at 25 h after stimulation, p27Kip1 levels were 10-fold lower in the serum-supplemented culture than in the serum-deficient culture. Although the kinetics differed somewhat, the pattern of p27Kip1 down-regulation was similar in ConA-treated splenocytes (Fig. 2 B), anti-CD3-treated splenocytes (Fig.2 C), and anti-CD3-treated T cells (Fig. 2 D). Note that p27Kip1 levels re-accumulated at later times in cells stimulated in serum-deficient medium (Fig. 2, A–D). These findings show that p27Kip1 down-regulation in T cells is a two-stage process, consisting of serum-independent and serum-dependent segments. Cyclin D3, cdk6, and cyclin E were expressed at low levels in resting cells and were substantially up-regulated by ConA or anti-CD3 regardless of serum concentration (Fig. 2, B–D). These agents also increased cyclin D3-cdk6 activity in cells receiving either 10% or 0.1% serum. Cyclin A, on the other hand, was barely detectable in quiescent cells and was up-regulated by ConA or anti-CD3 to a greater extent in medium containing 10% as compared with 0.1% serum (Fig. 2, B–D). cdk2 expression was even more dependent on serum; elevations in the basal levels of cdk2 were seen in cells treated with ConA and 10% but not 0.1% serum (see Fig.3 B). In cells stimulated in medium containing 10% serum, increases in the expression of cyclin E, cyclin A, and cdk2 were paralleled by increases in the activities of both cyclin E-cdk2 and cyclin A-cdk2. In contrast, these complexes were not active in cells receiving ConA or anti-CD3 in serum-deficient medium (Fig. 2, B–D). Collectively, these findings demonstrate that TCR agonists induce an initial but transient loss of p27Kip1, the expression of cyclin D3, cyclin E, and cdk6, and the activation of cyclin D3-cdk6 complexes in a serum-independent manner. In contrast, serum is clearly required for maximal and sustained p27Kip1 down-regulation, maximal expression of cyclin A and cdk2, and activation of cyclin E-cdk2 and cyclin A-cdk2 complexes. Because cdk2 activity is obligatory for cell cycle traverse, it is apparent from these studies that serum promotes T cell proliferation by enabling this response. The data presented above demonstrate that serum is required for events that occur in late G0/G1 (e.g. cdk2 activation) but not for events that begin earlier in G0/G1(e.g. cdk6 activation). Consistent with these results, we found that quiescent splenocytes initiated but did not complete G0/G1 traverse in serum-deficient medium. In these experiments, resting splenocytes were treated continuously with ConA and either 10% or 0.1% serum, or were pretreated with ConA and 0.1% serum for 20 h prior to addition of 10% serum. DNA synthesis was assessed by [3H]thymidine incorporation. Similar to the data in Fig. 1 D, splenocytes co-treated with ConA and 10% serum entered S phase after an approximate 20 h lag (Fig. 3 A). In contrast, cultures receiving ConA and 0.1% serum remained (for the most part) in G0/G1. However, when exposed to 10% serum, serum-deprived cultures initiated DNA synthesis within 8–12 h. This observation indicates that ConA-treated splenocytes partially traverse G0/G1 in medium containing 0.1% serum and that the serum-dependent "checkpoint" is temporally located in mid to late G1. In accord with its capacity to induce S phase entry, serum replenishment also elevated the expression of cyclin A and cdk2, maximized the loss of p27Kip1, and restored both cyclin A-cdk2 and cyclin E-cdk2 activities (Fig. 3 B). The capacity of splenocytes pretreated with ConA and 0.1% serum to elicit these responses when exposed to 10% serum indicates that a substantial portion of the population remains viable in serum-deficient medium. Our data suggest that serum elicits cdk2 activation by persistently down-regulating p27Kip1 and by elevating the expression of cdk2 and cyclin A. Such conditions would favor the formation of cyclin-cdk2 complexes that do not contain p27Kip1 (designated "p27Kip1-free") and thus are catalytically active. On the other hand, due to higher p27Kip1 levels and lower cdk2 and cyclin A levels, inactive p27Kip1-associated complexes would predominate in cells receiving ConA and 0.1% serum. These predictions are confirmed by the data presented in Fig.4. In these experiments, the amounts of total and p27Kip1-free cyclin A-cdk2 complexes were determined in unfractionated and p27Kip1-immunodepleted cell lysates, respectively. As shown in Fig. 4, most if not all of the cyclin A-cdk2 complexes in cells stimulated with ConA and 10% serum for 24 or 40 h were present in p27Kip1-depleted extracts and thus are not bound to p27Kip1 (comparelanes 3 and 9, and 6 and12). Cyclin A-cdk2 complexes were not detected in cells receiving ConA and 0.1% serum for 24 h (lane 1) but were present at low levels in cells treated in this manner for 40 h (lane 4). However, the majority of these complexes were associated with p27Kip1 (i.e. were removed by p27Kip1 antibody; compare lanes 4 and10). Addition of serum (to 10%) to cells treated with ConA and 0.1% serum for 20 h increased both the total amount of cyclin A-cdk2 and the percentage of cyclin A-cdk2 complexes that do not contain p27Kip1 (lanes 5 and 11). Similar results were obtained in experiments examining the interaction of p27Kip1 with cyclin E-cdk2 complexes (data not shown). These findings demonstrate that the lack of cdk2 activity in splenocytes stimulated in serum-deficient medium results from the presence of p27Kip1 in cdk2-containing complexes. Previous studies have shown that T cells derived from p27Kip1-null mice exhibit constitutive cyclin E-cdk2 activity (37Fero M.L. Rivkin M. Tasch M. Porter P. Carow C.E. Firpo E. Polyak K. Tsai L.H. Broudy V. Perlmutter R.M. Kaushansky K. Roberts J.M. Cell. 1996; 85: 733-744Abstract Full Text Full Text PDF PubMed Scopus (1341) Google Scholar, 38Coats S. Whyte P. Fero M.L. Lacy S. Chung G. Randel E. Firpo E. Roberts J.M. Curr. Biol. 1999; 9: 163-173Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). Because our data indicate that serum promotes T cell proliferation by allowing cdk2 activation, it was of interest to determine if splenocytes lacking p27Kip1 initiated DNA synthesis in a serum-independent manner. For these experiments, we used splenic cells derived from C57b1/6 mice that express an N-terminally truncated form of p27Kip1 that does not interact with or inhibit the activity of cyclin-cdk2 complexes (34Kiyokawa H. Kineman R.D. Manova-Todorova K.O. Soares V.C. Hoffman E.S. Ono M. Khanam D. Hayday A.C. Frohman L.A. Koff A. Cell. 1996; 85: 721-732Abstract Full Text Full Text PDF PubMed Scopus (1149) Google Scholar). Initial experiments characterized the "cyclin/CDK profiles" of p27+/+, p27+/−, and p27−/− splenocytes. In all three populations, and similar to results obtained with Balb/c splenocytes (Fig.2 B), levels of cyclin D3 and cdk6 and of cyclin D3-cdk6 activity increased in response to ConA in a serum-independent manner (data not shown). In unstimulated p27−/− cells, levels of cdk2 and of cyclin E-cdk2 activity were comparable to those of p27+/+ cells treated with ConA and 10% serum (Fig.5 A; compare lanes 3and 8). Thus, in the absence of p27Kip1, cdk2 is constitutively expressed and, as described previously, cyclin E-cdk2 is constitutively active (37Fero M.L. Rivkin M. Tasch M. Porter P. Carow C.E. Firpo E. Polyak K. Tsai L.H. Broudy V. Perlmutter R.M. Kaushansky K. Roberts J.M. Cell. 1996; 85: 733-744Abstract Full Text Full Text PDF PubMed Scopus (1341) Google Scholar, 38Coats S. Whyte P. Fero M.L. Lacy S. Chung G. Randel E. Firpo E. Roberts J.M. Curr. Biol. 1999; 9: 163-173Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). In contrast, levels of cyclin A and of cyclin A-cdk2 activity were not substantially elevated in unstimulated p27−/− cells. However, addition of ConA and 0.1% serum to p27−/− cells resulted in increases in both cyclin A expression and associated activity that were comparable to those seen in p27+/+ cells receiving ConA and 10% (but not 0.1%) serum (compare lanes 3 and 11). Thus, although TCR stimulation is still required, abrogation of p27Kip1function renders cyclin A expression and cyclin A-cdk2 activation serum-independent. The need for TCR signaling for these events presumably reflects the dependence of cyclin A expression (and hence cyclin A-cdk2 activity) on cyclin D3-cdk6 activity and consequent Rb phosphorylation and E2F activation (18Lundberg A.S. Weinberg R.A. Mol. Cell. Biol. 1998; 18: 753-761Crossref PubMed Scopus (856) Google Scholar, 20DeGregori J. Kowalik T. Nevins J.R. Mol. Cell. Biol. 1995; 15: 4215-4224Crossref PubMed Scopus (839) Google Scholar). p27+/− cells behaved similarly to p27−/−cells in terms of cyclin A expression and cyclin A-cdk2 activity (Fig.5 A, lanes 4–7). On the other hand, p27+/− cells, unlike p27−/− cells, required ConA (but not 10% serum) for cdk2 up-regulation and cyclin E-cdk2 activation. p27Kip1, however, was present in quiescent p27+/− cells at levels comparable to those seen in quiescent p27+/+ cells (compare lanes 1 and4), and its down-regulation was mediated by ConA (lanes 5–7). Interestingly, ConA and 0.1% serum decreased p27Kip1 levels in p27+/− cells to an extent greater than that observed in p27+/+ cells stimulated with ConA and 10% serum (compare lanes 3 and 7). Together, the findings
Referência(s)