Estrogens induce cell proliferation in target tissues by stimulating progression through G1 phase of the cell cycle, but the underlying molecular targets remain undefined. To determine the role of the cyclin/cyclin-dependent kinase (CDK)/retinoblastoma protein (pRB) pathway in this response we treated MCF-7 breast cancer cells with the pure estrogen antagonist ICI 182780 to inhibit estrogen-induced gene expression and induce G1 phase arrest. Subsequent treatment with 17β-estradiol resulted in the synchronous entry of cells into S phase commencing at 12 h. The proportion of cells in S phase reached a maximum of 60% at 21–24 h. Cells subsequently completed mitosis and entered a second semisynchronous round of replication. Entry into S phase was preceded by increased activity of both Cdk4 and cyclin E-Cdk2 and hyperphosphorylation of pRB, all within the first 3–6 h of estradiol treatment. The increase in Cdk4 activity was accompanied by increases in cyclin D1 mRNA and protein, indicating that an initiating event in the activation of Cdk4 was increased cyclin D1 gene expression. In contrast, the levels of Cdk2 and the CDK inhibitors p21 (WAF1/CIP1/SDI1) and p27 (KIP1) in total cell lysates and in cyclin E immunoprecipitates were unaltered at these early time points. However, an inhibitory activity was present in antiestrogen-pretreated cell lysates toward recombinant cyclin E-Cdk2 and was relieved by estradiol treatment. This activity was attributable predominantly to p21. These apparently conflicting data were resolved by performing gel filtration chromatography, which revealed that only a minority of cyclin E-Cdk2 complexes were active following estradiol treatment. Active complexes eluted at a higher molecular weight than inactive complexes, were relatively deficient in both p21 and p27, and contained Cdk2 with increased threonine 160 phosphorylation, consistent with a mechanism of activation of cyclin E-Cdk2 involving both reduced CDK inhibitor association and CDK-activating kinase-mediated phosphorylation of Cdk2. These results provide an explanation for the early activation of both cyclin D1-Cdk4 and cyclin E-Cdk2 complexes that accompany G1-S phase progression in response to estradiol. Estrogens induce cell proliferation in target tissues by stimulating progression through G1 phase of the cell cycle, but the underlying molecular targets remain undefined. To determine the role of the cyclin/cyclin-dependent kinase (CDK)/retinoblastoma protein (pRB) pathway in this response we treated MCF-7 breast cancer cells with the pure estrogen antagonist ICI 182780 to inhibit estrogen-induced gene expression and induce G1 phase arrest. Subsequent treatment with 17β-estradiol resulted in the synchronous entry of cells into S phase commencing at 12 h. The proportion of cells in S phase reached a maximum of 60% at 21–24 h. Cells subsequently completed mitosis and entered a second semisynchronous round of replication. Entry into S phase was preceded by increased activity of both Cdk4 and cyclin E-Cdk2 and hyperphosphorylation of pRB, all within the first 3–6 h of estradiol treatment. The increase in Cdk4 activity was accompanied by increases in cyclin D1 mRNA and protein, indicating that an initiating event in the activation of Cdk4 was increased cyclin D1 gene expression. In contrast, the levels of Cdk2 and the CDK inhibitors p21 (WAF1/CIP1/SDI1) and p27 (KIP1) in total cell lysates and in cyclin E immunoprecipitates were unaltered at these early time points. However, an inhibitory activity was present in antiestrogen-pretreated cell lysates toward recombinant cyclin E-Cdk2 and was relieved by estradiol treatment. This activity was attributable predominantly to p21. These apparently conflicting data were resolved by performing gel filtration chromatography, which revealed that only a minority of cyclin E-Cdk2 complexes were active following estradiol treatment. Active complexes eluted at a higher molecular weight than inactive complexes, were relatively deficient in both p21 and p27, and contained Cdk2 with increased threonine 160 phosphorylation, consistent with a mechanism of activation of cyclin E-Cdk2 involving both reduced CDK inhibitor association and CDK-activating kinase-mediated phosphorylation of Cdk2. These results provide an explanation for the early activation of both cyclin D1-Cdk4 and cyclin E-Cdk2 complexes that accompany G1-S phase progression in response to estradiol. Estrogenic steroids have several major roles in mammalian physiology. These include control of development of the reproductive tract and secondary sex organs, e.g. the mammary gland, regulation of the estrus cycle, control of lactation, and effects on the bone, liver, and cardiovascular systems (1Sutherland R.L. Watts C.K.W. Clarke C.L. van der Molen H.J. King R.J.B. Cooke B.A. Hormones and Their Actions: Part 1. Elsevier Science Publishing B. V., Amsterdam1988: 197-215Google Scholar). Estrogens are also intimately linked with the development and progression of a number of human cancers, notably breast cancer. Observations made a century ago showed that ovariectomy in cases of premenopausal breast cancer could lead to tumor regression (2Beatson G. Lancet. 1896; ii: 104-107Crossref Google Scholar). Subsequent research in experimental models of carcinogen-induced mammary carcinoma revealed that estrogen was essential for both the initiation and progression of the disease (3Henderson B. Ross R. Bernstein L. Cancer Res. 1988; 48: 246-253PubMed Google Scholar). More recently, studies with human breast cancer cell lines xenografted to athymic nude mice have demonstrated an absolute requirement for estrogen in tumor formation and growth (4Soule H.D. McGrath C.M. Cancer Lett. 1980; 10: 177-189Crossref PubMed Scopus (165) Google Scholar). Such observations provided the rationale for the introduction of antiestrogen therapy, the current treatment of choice for hormone-responsive breast cancer. The subsequent demonstration of the efficacy of the antiestrogen tamoxifen in decreasing both disease progression and the development of contralateral breast cancer (5Early Breast Cancer Trialists' Collaborative Group, Lancet, 339, 1992, 1, 15, 71–85.Google Scholar) provide further compelling evidence for the pivotal role of estrogen-regulated cell proliferation in breast cancer. Despite this, little is known of the molecular basis of cell proliferation control by estrogen. Early studies on cell proliferation in the rodent uterus and mammary gland in vivo demonstrated that estrogen increases the proportion of cells synthesizing DNA by recruiting noncycling cells into the cell cycle and reducing the duration of G1 phase in already cycling cells (reviewed in Ref. 6Sutherland R.L. Reddel R.R. Green M.D. Eur. J. Cancer Clin. Oncol. 1983; 19: 307-318Abstract Full Text PDF PubMed Scopus (100) Google Scholar). Using breast cancer cells synchronized at the G1/S boundary or at G2/M to test the effect of estrogen added at different stages of the cell cycle, Leung et al. (7Leung B.S. Potter A.H. J. Cell. Biochem. 1987; 34: 213-225Crossref PubMed Scopus (37) Google Scholar) concluded that the sensitive cells were in early G1 phase, immediately following mitosis. These data supported observations that both nonsteroidal (8Sutherland R.L. Hall R.E. Taylor I.W. Cancer Res. 1983; 43: 3998-4006PubMed Google Scholar) and steroidal antiestrogens (9Wakeling A.E. Dukes M. Bowler J. Cancer Res. 1991; 51: 3867-3873PubMed Google Scholar, 10Watts C.K.W. Brady A. Sarcevic B. deFazio A. Musgrove E.A. Sutherland R.L. Mol. Endocrinol. 1995; 9: 1804-1813Crossref PubMed Scopus (153) Google Scholar) arrest ER-positive breast cancer cells in G1 phase. More precise mapping of the point of antiestrogen action within G1 phase using cells synchronized by mitotic selection identified a window of sensitivity in early to mid-G1 phase (11Taylor I.W. Hodson P.J. Green M.D. Sutherland R.L. Cancer Res. 1983; 43: 4007-4010PubMed Google Scholar). Together these data are compatible with a model whereby estrogens and antiestrogens, through their interactions with the ER, regulate the transcription of genes that control key points in G1 progression. Candidate genes that might fulfill this role include "immediate early" and "delayed early" genes with established roles in signal transduction and cell cycle control, particularly c-myc, which encodes a transcription factor (c-Myc) with an established role in control of proliferation and apoptosis, genes encoding components of the AP-1 transcription complex, c-fos and c-jun, and the more recently described cyclin-dependent kinase (CDK) 1The abbreviations used are: CDK, cyclin-dependent kinase; CDKI, cyclin-dependent kinase inhibitor; estradiol or E2, 17β-estradiol; pRB, retinoblastoma protein; CAK, Cdk-activating kinase; p21, WAF1/CIP1/SDI1; p27, KIP1; ER, estrogen receptor; DTT, dithiothreitol; GST, glutathione S-transferase; kb, kilobase pair(s); PAGE, polyacrylamide gel electrophoresis; ICI 182780, 7α-[9-(4,4,5,5,5-pentafluoropentylsulfinyl)nonyl]estra-1,3,5,(10)-triene-3,17β-diol. 1The abbreviations used are: CDK, cyclin-dependent kinase; CDKI, cyclin-dependent kinase inhibitor; estradiol or E2, 17β-estradiol; pRB, retinoblastoma protein; CAK, Cdk-activating kinase; p21, WAF1/CIP1/SDI1; p27, KIP1; ER, estrogen receptor; DTT, dithiothreitol; GST, glutathione S-transferase; kb, kilobase pair(s); PAGE, polyacrylamide gel electrophoresis; ICI 182780, 7α-[9-(4,4,5,5,5-pentafluoropentylsulfinyl)nonyl]estra-1,3,5,(10)-triene-3,17β-diol. complexes involved in G1progression. Previously published studies have focused predominantly on c-myc and c-fos. Regulation of these genes is among the earliest detectable responses to estrogens and antiestrogens, being apparent within 30 min. Expression of c-myc and c-fos is apparently directly transcriptionally regulated by estrogen in immature rat uteri (12Weisz A. Bresciani F. Mol. Endocrinol. 1988; 2: 816-824Crossref PubMed Scopus (213) Google Scholar, 13Murphy L.J. Murphy L.C. Friesen H.G. Endocrinology. 1987; 120: 1882-1888Crossref PubMed Scopus (128) Google Scholar) and in breast cancer cell lines (14Dubik D. Dembinski T.C. Shiu R.P.C. Cancer Res. 1987; 47: 6517-6521PubMed Google Scholar) with kinetics that mimic those following treatment of growth arrested cells with peptide mitogens in other cell types (15Eisenman R.N. Weinberg R.A. Oncogenes and the Molecular Origins of Cancer. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989: 175-221Google Scholar). Furthermore, treatment with antisense c-myc oligonucleotides specifically inhibits the increase in cell numbers observed after estrogen treatment of MCF-7 breast cancer cells, providing strong evidence that c-myc expression is necessary for estrogen-induced proliferation (16Watson P.H. Pon R.T. Shiu R.P.C. Cancer Res. 1991; 51: 3996-4000PubMed Google Scholar). The induction of c-fos and c-myc by estrogen in some respects parallels the response to growth factor mitogens. Nevertheless, there are differences in the signaling pathways utilized by estrogen and growth factors, illustrated by the different mechanisms for regulation of c-fos and c-myc,i.e. direct transcription regulation by the ERversus activation of a cascade of signaling molecules. Treatment of ZR-75-1 breast cancer cells with a series of growth factors elicited ribosomal S6 kinase activation and S6 phosphorylation, but activation of the kinase by estrogen could not be detected (17Novak-Hofer I. Küng W. Fabbro D. Matter A. Eppenberger U. Experientia. 1988; 44: 162-166Crossref PubMed Scopus (5) Google Scholar). Although transient activation of mitogen-activated protein kinases following estrogen treatment of MCF-7 cells has been demonstrated (18Migliaccio A. Di Domenico M. Castoria G. de Falco A. Bontempo P. Nola E. Auricchio F. EMBO J. 1996; 15: 1292-1300Crossref PubMed Scopus (853) Google Scholar), another recent study using MCF-7 cells rescued with estrogen from cell cycle arrest induced by hydroxymethylglutaryl-CoA reductase inhibitors showed that estrogen-induced reinitiation of cell cycle progression was independent of activation of mitogen-activated protein kinases (19Bonapace I.M. Addeo R. Altucci L. Cicatiello L. Bifulco M. Laezza C. Salzano S. Sica V. Bresciani F. Weisz A. Oncogene. 1996; 12: 753-763PubMed Google Scholar). Thus, cell cycle control by estrogen will not necessarily parallel that by growth factors. Progress through G1 phase requires inactivation of the pRB protein by phosphorylation and the consequent release of a number of factors including the E2F family of transcription factors (20Weinberg R.A. Cell. 1995; 81: 323-330Abstract Full Text PDF PubMed Scopus (4233) Google Scholar, 21Riley D.J. Lee E.Y.-H.P. Lee W.-H. Annu. Rev. Cell Biol. 1994; 10: 1-29Crossref PubMed Scopus (202) Google Scholar). These transcription factors then activate transcription of genes whose products are required for S phase progression (20Weinberg R.A. Cell. 1995; 81: 323-330Abstract Full Text PDF PubMed Scopus (4233) Google Scholar, 21Riley D.J. Lee E.Y.-H.P. Lee W.-H. Annu. Rev. Cell Biol. 1994; 10: 1-29Crossref PubMed Scopus (202) Google Scholar). Phosphorylation of pRB is mediated by the action of the G1 phase CDKs (Cdk4, Cdk6, and Cdk2), which are activated by cyclin binding (Cdk4 and Cdk6 by D-type cyclin binding, and Cdk2 by cyclin E binding). Control of G1 CDK activity is achieved by several mechanisms including transcriptional activation of D-type cyclins and cyclin E, the rate-limiting regulatory subunits of the G1 cyclin-CDK complexes; activation and inactivation of the enzyme complexes by phosphorylation/dephosphorylation events; and the abundance and action of two families of CDK inhibitors (CDKIs) (22Morgan D.O. Nature. 1994; 374: 131-134Crossref Scopus (2893) Google Scholar, 23Sherr C.J. Roberts J.M. Genes & Dev. 1995; 9: 1149-1163Crossref PubMed Scopus (3185) Google Scholar). Modulation at any of these levels of regulation could regulate pRB phosphorylation and hence G1 phase progression. D-type cyclins are induced as delayed early response genes by a variety of mitogens in many cell types, and removal of growth factors in G1 phase leads to their rapid down-regulation (24Matsushime H. Roussel M.F. Ashmun R.A. Sherr C.J. Cell. 1991; 65: 701-713Abstract Full Text PDF PubMed Scopus (985) Google Scholar), consistent with the notion that these cyclins act as mitogenic sensors linking extracellular signals with cell cycle progression (25Sherr C.J. Cell. 1993; 73: 1059-1065Abstract Full Text PDF PubMed Scopus (1977) Google Scholar). An essential role for cyclin D1 in mammary gland development is demonstrated by the absence of lobular-alveolar structures in transgenic mice with disruption of the cyclin D1 gene (26Sicinski P. Donaher J.L. Parker S.B. Li T. Fazeli A. Gardner H. Haslam S.Z. Bronson R.T. Elledge S.J. Weinberg R.A. Cell. 1995; 82: 621-630Abstract Full Text PDF PubMed Scopus (881) Google Scholar, 27Fantl V. Stamp G. Andrews A. Rosewell I. Dickson C. Genes & Dev. 1995; 9: 2364-2372Crossref PubMed Scopus (597) Google Scholar). There is accumulating evidence that D-type cyclins may also play a role in mediating the effects of growth factors and steroid hormones on breast cancer cell cycle progression by binding and activating Cdk4 and Cdk6. The abundance of cyclin D1 increases following growth factor and progestin stimulation of breast cancer cell proliferation (28Musgrove E.A. Hamilton J.A. Lee C.S.L. Sweeney K.J.E. Watts C.K.W. Sutherland R.L. Mol. Cell. Biol. 1993; 13: 3577-3587Crossref PubMed Scopus (277) Google Scholar) and declines rapidly following exposure to growth-inhibitory antiestrogens (10Watts C.K.W. Brady A. Sarcevic B. deFazio A. Musgrove E.A. Sutherland R.L. Mol. Endocrinol. 1995; 9: 1804-1813Crossref PubMed Scopus (153) Google Scholar, 28Musgrove E.A. Hamilton J.A. Lee C.S.L. Sweeney K.J.E. Watts C.K.W. Sutherland R.L. Mol. Cell. Biol. 1993; 13: 3577-3587Crossref PubMed Scopus (277) Google Scholar). Similarly, cyclin E expression also increases following growth factor stimulation of breast cancer cells (28Musgrove E.A. Hamilton J.A. Lee C.S.L. Sweeney K.J.E. Watts C.K.W. Sutherland R.L. Mol. Cell. Biol. 1993; 13: 3577-3587Crossref PubMed Scopus (277) Google Scholar) but at times later than for cyclin D1 and compatible with its established role in the control of the G1 to S phase transition. Finally, ectopic expression of cyclin D1 in T-47D breast cancer cells is sufficient for Cdk4 and Cdk2 activation, pRB phosphorylation and G1-S phase progression (29Musgrove E.A. Sarcevic B. Sutherland R.L. J. Cell. Biochem. 1996; 60: 363-378Crossref PubMed Scopus (30) Google Scholar). Thus, G1 cyclins and their associated kinases are potential downstream targets of estrogen-induced mitogenesis. Our recent demonstration that a decreased rate of G1 progression in antiestrogen-treated breast cancer cells is preceded by decreased cyclin D1 gene expression, cyclin D1-Cdk4 activity, and reduced pRB phosphorylation supports such a view (10Watts C.K.W. Brady A. Sarcevic B. deFazio A. Musgrove E.A. Sutherland R.L. Mol. Endocrinol. 1995; 9: 1804-1813Crossref PubMed Scopus (153) Google Scholar). To further test this hypothesis, we exploited the unique properties of steroidal antiestrogens that, in marked contrast to their nonsteroidal predecessors, are pure antagonists devoid of estrogen agonist activity (9Wakeling A.E. Dukes M. Bowler J. Cancer Res. 1991; 51: 3867-3873PubMed Google Scholar, 30Wakeling A.E. Bowler J. J. Steroid Biochem. 1988; 31: 645-653Crossref PubMed Scopus (203) Google Scholar). Thus, exposure of estrogen-dependent cells to these compounds will block estrogen-induced gene expression with consequent inhibition of cell proliferation. Using a model in which subsequent "rescue" of these growth-arrested cells with estradiol leads to synchronous progression of cells through the cell cycle, we identified activation of both cyclin D1-Cdk4 and cyclin E-Cdk2 complexes as early events in estradiol action. Subsequent experiments were directed at identifying potential mechanisms of estrogen-induced CDK activation. Stock solutions of 7α-[9-(4,4,5,5,5-pentafluoropentylsulfinyl)nonyl]estra-1,3,5,(10)-triene-3,17β-diol (ICI 182780) and 17β-estradiol (estradiol) were prepared as follows. ICI 182780 (a kind gift from Dr. Alan Wakeling, Zeneca Pharmaceuticals, Alderley Park, Cheshire, United Kingdom) was dissolved in ethanol to 10−2m, and a working dilution of 10−5m in RPMI 1640 medium was prepared from this stock just prior to each experiment. Estradiol (Sigma, Castle Hill, New South Wales, Australia) was dissolved in ethanol at 2 × 10−4m. Stock cultures of MCF-7 cells (Michigan Cancer Foundation, Detroit, MI) were cryopreserved and maintained as described previously (8Sutherland R.L. Hall R.E. Taylor I.W. Cancer Res. 1983; 43: 3998-4006PubMed Google Scholar). Cells were cultured in RPMI 1640 medium supplemented with 5% fetal calf serum, insulin (10 μg/ml), and gentamicin (10 μg/ml). For experiments investigating the effects of ICI 182780 and estradiol on cell cycle phase distribution, 25-cm2 flasks were seeded with 0.8 × 105 cells. Experiments investigating the effects of estradiol on mRNA and protein levels, cyclin-CDK complex formation, and kinase activity employed 150-cm2 tissue culture flasks seeded with 1.5 × 106 cells. The growth kinetics, including changes in cell cycle phase distribution, were identical under both experimental conditions. Cells were allowed to proliferate for 2 days. ICI 182780 or vehicle was then added to the medium. Following 48 h of ICI 182780 pretreatment and without a change of medium, estradiol (100 nm), or vehicle was added directly to the medium. The final concentration of ethanol in the tissue culture medium was less than 0.06% and had no effect on the rate of cell proliferation. At the completion of experiments, cells were harvested by brief incubation with 0.05% (w/v) trypsin, 0.02% (w/v) EDTA for flow cytometry as described previously (8Sutherland R.L. Hall R.E. Taylor I.W. Cancer Res. 1983; 43: 3998-4006PubMed Google Scholar) or as described below. Cell cycle phase distribution was determined by analytical DNA flow cytometry as described previously (31Musgrove E.A. Wakeling A.E. Sutherland R.L. Cancer Res. 1989; 49: 2398-2404PubMed Google Scholar). Total cellular RNA was extracted from duplicate 150-cm2 flasks by a guanidinium isothiocyanate-cesium chloride procedure, and Northern blot analysis employing 10–20 μg of total RNA/lane was performed as described previously (32Musgrove E.A. Lee C.S.L. Sutherland R.L. Mol. Cell. Biol. 1991; 11: 5032-5043Crossref PubMed Google Scholar). Filters were hybridized with cDNA labeled with [α-32P]dCTP to a specific activity of approximately 109 dpm/μg by random primer extension (Multiprime, Amersham Australia, North Ryde, Australia). Messenger RNA abundance was quantitated by analysis of autoradiographs using PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA). Plasmids containing the human cDNAs used in this study were provided by the following investigators: cyclins D1 (pHsCYCD1-H124) and D3 (pD3-H347), Drs. Yue Xiong and David Beach (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY); cyclin E (in pBSKS), Dr. Steven Reed (Scripps Research Institute, La Jolla, CA); c-myc (p9C-myc), Dr. Geoff Symonds (R. W. Johnson Pharmaceutical Research Institute, Sydney, Australia). Restriction enzyme digestion was used to obtain fragments for labeling as follows: for cyclin D1, a 1.3-kb EcoRI fragment comprising the entire open reading frame; for cyclin D3, a 0.5-kbPstI fragment corresponding to 60% of the open reading frame; for cyclin E, a 2.5-kb EcoRI fragment comprising the entire open reading frame; for c-myc, a 0.45-kbPstI fragment corresponding to the entire second exon. Equivalent loading of RNA samples was confirmed by hybridization with an oligonucleotide probe complementary to a region of 18 S ribosomal RNA. Cells were lysed as follows. Cell monolayers were washed twice in ice-cold phosphate-buffered saline and then scraped into ice-cold lysis buffer A (50 mm HEPES, pH 7.5, 150 mm NaCl, 10% (v/v) glycerol, 1% Triton X-100, 1.5 mm MgCl2, 1 mm EGTA, 10 μg/ml aprotinin, 10 μg/ml leupeptin, 1 mm phenylmethylsulfonyl fluoride, 200 μm sodium orthovanadate, 10 mmsodium pyrophosphate, 100 mm NaF, and 1 mmDTT). The lysates were incubated for 5 min on ice, and the cellular debris was cleared by centrifugation (15,000 × g, 5 min, 4 °C). Equal amounts of total protein (20–40 μg) were separated by SDS-PAGE and then transferred to nitrocellulose filters. Proteins were visualized using the ECL detection system (Amersham, Australia) after incubation (2 h at room temperature or overnight at 4 °C) with the following primary antibodies: cyclin D1 (PRAD1), cyclin D2 (C-17), cyclin D3 (C-16), cyclin E (HE12), cyclin B1 (GNS1), p15 (K-18), p27 (C-19), Cdk2 (M2), Cdk4 (C-22), and Cdk6 (C-21) from Santa Cruz Biotechnology Inc. (Santa Cruz, CA); cyclin D1 (DCS-6) from Novacastra Laboratories Ltd. (Newcastle-on-Tyne, UK); p21 (a kind gift from Dr. David Beach, Cold Spring Harbor Laboratory, NY) or p21 (catalog number C24420; Transduction Laboratories, Lexington, KY); c-Myc (9E10; ATCC, Rockville, MD); or pRB (G3–245) and cyclin A (BF683) from PharMingen (San Diego, CA). Protein abundance was quantitated by analysis of Western blots using densitometry (Molecular Dynamics, Sunnyvale, CA). Quantitation of protein levels by this method was linear over the protein concentrations and exposure times tested. A GST-pRB fusion protein was employed as the substrate for the Cdk4 activity assay.Escherichia coli transformed with a pGEX expression vector containing a GST-pRB construct encoding amino acids 773–923 (kindly supplied by Dr. Ed Harlow, Massachusetts General Hospital Cancer Center, Charlestown, MA) were induced by the addition of 0.4 mm isopropyl-β-d-thiogalactopyranoside and incubated for 3 h at room temperature. The bacterial pellets were then lysed by sonication, and the fusion protein was purified by affinity chromatography on glutathione-agarose beads and then eluted with 15 mm reduced glutathione. GST-p16 was prepared as described previously (33Lilischkis R. Sarcevic B. Kennedy C. Warlters A. Sutherland R.L. Int. J. Cancer. 1996; 66: 249-254Crossref PubMed Scopus (68) Google Scholar). cDNAs for Cdk2 (from Dr. Tony Hunter, Salk Institute, La Jolla, CA) and cyclin E were amplified by polymerase chain reaction and cloned separately into the baculoviral transfer vector pVL1392 (PharMingen). Cyclin E was cloned with an N-terminal GST cassette and Cdk2 with an N-terminal 6-histidine tag. The identity of both vectors was confirmed by sequencing. Viruses were generated according to the manufacturer's instructions, and Sf9 insect cells were co-infected with high titer pVL1392 GST-cyclin E and pVL1392 6His-Cdk2 viral stocks. Active cyclin E-Cdk2 complexes were purified using glutathione affinity chromatography. For Cdk4 activity assays, MCF-7 cell monolayers were washed twice with phosphate-buffered saline and then scraped into ice-cold phosphate-buffered saline and pelleted by centrifugation (15,000 × g, 5 min), and the pellets were frozen in liquid nitrogen and then resuspended in 1 ml of ice-cold lysis buffer B (50 mm HEPES, pH 7.5, 150 mm NaCl, 1 mmEDTA, 2.5 mm EGTA, 0.1% Tween 20, 10% glycerol, 10 mm β-glycerophosphate, 1 mm NaF, 0.1 mm orthovanadate, 10 μg/ml leupeptin, 10 μg/ml aprotinin, 1 mm DTT, and 0.1 mmphenylmethylsulfonyl fluoride). The lysate was placed on ice and vortexed vigorously at intervals for 60 min and then centrifuged at 15,000 × g for 5 min at 4 °C, and the supernatant was stored at −80 °C. Equivalent amounts of protein were precleared by incubation with preimmune rabbit serum conjugated to protein A-Sepharose (Zymed, San Francisco, CA) for 45 min at 4 °C. Cdk4 complexes were immunoprecipitated with rabbit polyclonal anti-human Cdk4 antiserum (catalog number 14936E, PharMingen) conjugated to protein A-Sepharose for 3 h at 4 °C. Immunoprecipitates were washed four times with ice-cold lysis buffer B and three times with ice-cold 50 mm HEPES pH 7.5, 1 mm DTT and then either incubated with 30 μl of 50 mm HEPES, pH 7.5, 1 mm DTT or 30 μl of the same buffer containing approximately 10−5m bacterially expressed GST-p16 or GST for 1 h at 30 °C. The supernatant was aspirated, and the immunoprecipitates were used for the kinase assay. For both the cyclin E-Cdk2 kinase activity and Cdk2 activity assays, lysates were prepared with lysis buffer A as described above. Equivalent amounts of lysate were precleared with protein A-Sepharose (1 h, 4 °C) and then immunoprecipitated with protein A-Sepharose conjugated to either an anti-cyclin E polyclonal antibody (C-19; Santa Cruz Biotechnology) or an anti-Cdk2 polyclonal antibody (M2; Santa Cruz Biotechnology) for 3 h at 4 °C. The immunoprecipitates were then washed once with ice-cold lysis buffer A, twice with ice-cold lysis buffer A containing 1 m NaCl, once again with ice-cold lysis buffer A, and then three times with ice-cold 50 mm HEPES, pH 7.5, 1 mm DTT. The kinase reactions were initiated by resuspending the beads in 30 μl of kinase buffer (50 mm HEPES, pH 7.5, 1 mm DTT, 2.5 mm EGTA, 10 mmMgCl2, 20 μm ATP, 10 μCi of [γ-32P]ATP, 0.1 mm orthovanadate, 1 mm NaF, 10 mm β-glycerophosphate) containing either 10 μg of GST-pRB773–923 (Cdk4 immunoprecipitates), 10 μg of histone H1 (from Sigma; cyclin E immunoprecipitates), or 3 μg of histone H1 (Cdk2 immunoprecipitates) as a substrate. After incubation for either 30 min (Cdk4), or 15 min (cyclin E and Cdk2) at 30 °C, the reactions were terminated by the addition of 15 μl of 3 × SDS sample buffer (187 mmTris-HCl, pH 6.8, 30% (v/v) glycerol, 6% SDS, 15% (v/v) β-mercaptoethanol). Unlabeled ATP (1 μl of a 10 mmstock solution) was added to the Cdk4 activity assays following termination to reduce background. The samples were then heated at 95 °C for 2 min and separated using 12% SDS-PAGE, and the dried gel was exposed to x-ray film. Relative band intensities were quantitated by densitometric analysis as described above. Immunoprecipitation of cyclin D1, cyclin E, and Cdk2 was performed using the method described above (for immunoprecipitating cyclin E and Cdk2 for kinase activity assays), except that the cyclin E antibodies were chemically cross-linked with dimethyl pimelimidate (Sigma) to protein A-Sepharose to reduce background (34Harlow E. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1988: 522-523Google Scholar). Antibodies used were rabbit polyclonal antisera to human cyclin D1 (29Musgrove E.A. Sarcevic B. Sutherland R.L. J. Cell. Biochem. 1996; 60: 363-378Crossref PubMed Scopus (30) Google Scholar), human cyclin E (C-19; Santa Cruz Biotechnology), and human Cdk2 (M2; Santa Cruz Biotechnology). Supernatants were Western blotted to determine the degree of immunodepletion achieved with each antibody. The immunoprecipitated proteins were resuspended in 1 × SDS sample buffer, separated by SDS-PAGE, and transferred to a nitrocellulose membrane, and then proteins were detected using the antibodies described for Western blotting above. A rabbit polyclonal anti-p27 antiserum (C-19; Santa Cruz Biotechnology) and a mouse monoclonal anti-p27 antibody (catalog number K25020; Transduction Laboratories) were both used on cyclin E immunoprecipitates. Protein A-horseradish peroxidase (Zymed Laboratories, San Francisco, CA) was employed rather than a goat anti-rabbit secondary antibody to reduce background following the use of primary rabbit polyclonal antibodies. Equivalent lane loading was demonstrated by the rabbit polyclonal IgG used for immunoprecipitation that cross-reacted with secondary antibodies used for Western blotting. Cell lysates were prepared with lysis buffer A as described above. He