Distinct Phosphorylation Events Regulate p130- and p107-mediated Repression of E2F-4
2002; Elsevier BV; Volume: 277; Issue: 30 Linguagem: Inglês
10.1074/jbc.m200381200
ISSN1083-351X
AutoresThomas Farkas, Klaus Hansen, Karin Holm, Jiří Lukáš, Jiří Bártek,
Tópico(s)Cancer, Hypoxia, and Metabolism
ResumoThe "pocket proteins" pRb (retinoblastoma tumor suppressor protein), p107, and p130 regulate cell proliferation via phosphorylation-sensitive interactions with E2F transcription factors and other proteins. We previously identified 22 in vivo phosphorylation sites in human p130, including three sites selectively targeted by cyclin D-Cdk4(6) kinases. Here we assessed the effects of alanine substitution at the individual or combined Cdk4(6)-specific sites in p130, compared with homologous sites in p107 (Thr369/Ser650/Ser964). In U-2-OS cells, the triple p107ΔCdk4* mutant strongly inhibited E2F-4 activity and imposed a G1 arrest resistant to cyclin D1 coexpression. In contrast, the p130ΔCdk4mutant still responded to cyclin D1, suggesting the existence of additional phosphorylation sites critical for E2F-4 regulation. Extensive mutagenesis, sensitive E2F reporter assays, and cell cycle analyses allowed the identification of six such residues (serines 413, 639, 662, 1044, 1080, and 1112) that, in addition to the Cdk4-specific sites, are necessary and sufficient for the regulation of E2F-4 and the cell cycle by p130. Surprisingly, 12 of the in vivophosphorylation sites seem dispensable for E2F regulation and probably modulate other functions of p130. These results further elucidate the complex regulation of p130 and provide a molecular mechanism to explain the differential control of p107 and p130 by cyclin-dependent kinases. The "pocket proteins" pRb (retinoblastoma tumor suppressor protein), p107, and p130 regulate cell proliferation via phosphorylation-sensitive interactions with E2F transcription factors and other proteins. We previously identified 22 in vivo phosphorylation sites in human p130, including three sites selectively targeted by cyclin D-Cdk4(6) kinases. Here we assessed the effects of alanine substitution at the individual or combined Cdk4(6)-specific sites in p130, compared with homologous sites in p107 (Thr369/Ser650/Ser964). In U-2-OS cells, the triple p107ΔCdk4* mutant strongly inhibited E2F-4 activity and imposed a G1 arrest resistant to cyclin D1 coexpression. In contrast, the p130ΔCdk4mutant still responded to cyclin D1, suggesting the existence of additional phosphorylation sites critical for E2F-4 regulation. Extensive mutagenesis, sensitive E2F reporter assays, and cell cycle analyses allowed the identification of six such residues (serines 413, 639, 662, 1044, 1080, and 1112) that, in addition to the Cdk4-specific sites, are necessary and sufficient for the regulation of E2F-4 and the cell cycle by p130. Surprisingly, 12 of the in vivophosphorylation sites seem dispensable for E2F regulation and probably modulate other functions of p130. These results further elucidate the complex regulation of p130 and provide a molecular mechanism to explain the differential control of p107 and p130 by cyclin-dependent kinases. retinoblastoma protein cyclin-dependent kinase hemagglutinin glutathione S-transferase The retinoblastoma protein (pRb)1 and the related proteins p107 and p130, collectively known as pocket proteins, are involved in regulation of the cell cycle, differentiation, cellular senescence, and cell death (1Wang J.Y. Curr. Opin. Genet. Dev. 1997; 7: 39-45Crossref PubMed Scopus (71) Google Scholar, 2Classon M. Dyson N. Exp. Cell Res. 2001; 264: 135-147Crossref PubMed Scopus (202) Google Scholar, 3Grana X. Garriga J. Mayol X. Oncogene. 1998; 17: 3365-3383Crossref PubMed Scopus (283) Google Scholar, 4Lipinski M.M. Jacks T. Oncogene. 1999; 18: 7873-7882Crossref PubMed Scopus (342) Google Scholar). Whereas pRb is a tumor suppressor, a potentially analogous role of p107 and p130 in preventing tumorigenesis is still a matter of debate (5Paggi M.G. Giordano A. Cancer Res. 2001; 61: 4651-4654PubMed Google Scholar). Nevertheless, evidence for essential functions of p107 and p130 in negative control of cell cycle progression has recently emerged (2Classon M. Dyson N. Exp. Cell Res. 2001; 264: 135-147Crossref PubMed Scopus (202) Google Scholar, 6Bruce J.L. Hurford R.K., Jr. Classon M. Koh J. Dyson N. Mol. Cell. 2000; 6: 737-742Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar, 7Sage J. Mulligan G.J. Attardi L.D. Miller A. Chen S. Williams B. Theodorou E. Jacks T. Genes Dev. 2000; 14: 3037-3050Crossref PubMed Scopus (500) Google Scholar, 8Dannenberg J.H. van Rossum A. Schuijff L. te Riele H. Genes Dev. 2000; 14: 3051-3064Crossref PubMed Scopus (349) Google Scholar). pRb, p107, and p130 exert their functions through interactions with a large number of cellular proteins via multiple independent binding sites (3Grana X. Garriga J. Mayol X. Oncogene. 1998; 17: 3365-3383Crossref PubMed Scopus (283) Google Scholar, 9Morris E.J. Dyson N.J. Adv. Cancer Res. 2001; 82: 1-54Crossref PubMed Scopus (294) Google Scholar). The so-called A/B pocket is such a binding site, originally defined in pRb. This pocket structure encompasses an A- and a B-domain, separated by an insert sequence (10Hu Q.J. Dyson N. Harlow E. EMBO J. 1990; 9: 1147-1155Crossref PubMed Scopus (247) Google Scholar, 11Huang S. Wang N.P. Tseng B.Y. Lee W.H. Lee E.H. EMBO J. 1990; 9: 1815-1822Crossref PubMed Scopus (153) Google Scholar, 12Kaelin W.G., Jr. Ewen M.E. Livingston D.M. Mol. Cell. Biol. 1990; 10: 3761-3769Crossref PubMed Scopus (163) Google Scholar, 13Lee J.O. Russo A.A. Pavletich N.P. Nature. 1998; 391: 859-865Crossref PubMed Scopus (368) Google Scholar). The A/B pocket is necessary for binding to viral oncoproteins and other proteins containing the motif LXCXE and is conserved in p107 and p130 (14Ewen M.E. Xing Y.G. Lawrence J.B. Livingston D.M. Cell. 1991; 66: 1155-1164Abstract Full Text PDF PubMed Scopus (346) Google Scholar, 15Li Y. Graham C. Lacy S. Duncan A.M. Whyte P. Genes Dev. 1993; 7: 2366-2377Crossref PubMed Scopus (302) Google Scholar, 16Hannon G.J. Demetrick D. Beach D. Genes Dev. 1993; 7: 2378-2391Crossref PubMed Scopus (405) Google Scholar, 17Mayol X. Grana X. Baldi A. Sang N., Hu, Q. Giordano A. Oncogene. 1993; 8: 2561-2566PubMed Google Scholar). Many of the known effects of pocket proteins depend on their ability to repress gene expression by binding members of the E2F family of transcription factors. The direct targets of E2Fs include genes whose products are critical for embryonic development and tissue homeostasis (18Müller H. Bracken A.P. Vernell R. Moroni M.C. Christians F. Grassilli E. Prosperini E. Vigo E. Oliner J.D. Helin K. Genes Dev. 2001; 15: 267-285Crossref PubMed Scopus (627) Google Scholar, 19Ishida S. Huang E. Zuzan H. Spang R. Leone G. West M. Nevins J.R. Mol. Cell. Biol. 2001; 21: 4684-4699Crossref PubMed Scopus (491) Google Scholar). The high affinity DNA binding form of E2F is a dimer consisting of one subunit from the DP protein family (DP-1 or DP-2) and one of six E2Fs (E2F-1 to E2F-6). The carboxyl terminus of the E2F subunit binds directly to the so-called "large A/B pocket" of pRb, p107, and p130 in a manner compatible with simultaneous binding of the pocket with LXCXE containing proteins (20Wang J.Y. Knudsen E.S. Welch P.J. Adv. Cancer Res. 1994; 64: 25-85Crossref PubMed Google Scholar). Whereas p130 and p107 preferentially bind E2F-4 and E2F-5, pRb can interact with all E2Fs except for E2F-6, which lacks the transactivating domain. p130 is highly expressed in quiescent cells, where it forms the predominant E2F complex. When cells enter the cell cycle and pass the so-called restriction point in middle to late G1 (21Bartek J. Bartkova J. Lukas J. Exp. Cell Res. 1997; 237: 1-6Crossref PubMed Scopus (228) Google Scholar), p107 is induced and becomes the major E2F binding partner at the G1/S transition. Complexes of pRb are mainly found in G1, but they also exist in quiescence (G0) and in S phase (22Nevins J.R. Cell Growth Differ. 1998; 9: 585-593PubMed Google Scholar, 23Helin K. Curr. Opin. Genet. Dev. 1998; 8: 28-35Crossref PubMed Scopus (427) Google Scholar). p107 and p130 redundantly repress a subset of E2F targets distinct from the subset of genes controlled by pRb (24Hurford R.K., Jr. Cobrinik D. Lee M.H. Dyson N. Genes Dev. 1997; 11: 1447-1463Crossref PubMed Scopus (380) Google Scholar). The pocket proteins repress transcription by direct inhibition of the transactivating domain of E2Fs (25Flemington E.K. Speck S.H. Kaelin W.G., Jr. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6914-6918Crossref PubMed Scopus (284) Google Scholar, 26Helin K. Harlow E. Fattaey A. Mol. Cell. Biol. 1993; 13: 6501-6508Crossref PubMed Scopus (400) Google Scholar, 27Hiebert S.W. Chellappan S.P. Horowitz J.M. Nevins J.R. Genes Dev. 1992; 6: 177-185Crossref PubMed Scopus (464) Google Scholar) and by recruitment of nucleosome-remodeling enzymes to the E2F-pocket protein complex, thereby changing chromatin into a transcriptionally less active state (28Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (949) Google Scholar). A characteristic difference distinguishing p107 and p130 from pRb is their binding to cyclin E-Cdk2 and cyclin A-Cdk2, a function that cooperates with repression of E2Fs to inhibit the cell cycle (29Smith E.J. Nevins J.R. Mol. Cell. Biol. 1995; 15: 338-344Crossref PubMed Scopus (48) Google Scholar, 30Zhu L. Enders G. Lees J.A. Beijersbergen R.L. Bernards R. Harlow E. EMBO J. 1995; 14: 1904-1913Crossref PubMed Scopus (133) Google Scholar, 31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar, 32Woo M.S. Sanchez I. Dynlacht B.D. Mol. Cell. Biol. 1997; 17: 3566-3579Crossref PubMed Scopus (102) Google Scholar). The regulation of pRb function by cell cycle-dependent phosphorylation has provided a framework for the understanding of the regulation of the other two pocket proteins. The inactivation of pRb is carried out in middle to late G1 by phosphorylation of serine and threonine residues, mediated by the sequentially activated cyclin D-Cdk4(6) and cyclin E-Cdk2 kinase complexes. When hyperphosphorylated, pRb loses its binding potential, and induction of E2F target gene transcription is one of the downstream effects (3Grana X. Garriga J. Mayol X. Oncogene. 1998; 17: 3365-3383Crossref PubMed Scopus (283) Google Scholar). p107 and p130 are also phosphorylated in a cell cycle-dependent manner (33Beijersbergen R.L. Carlee L. Kerkhoven R.M. Bernards R. Genes Dev. 1995; 9: 1340-1353Crossref PubMed Scopus (234) Google Scholar, 34Mayol X. Garriga J. Grana X. Oncogene. 1995; 11: 801-808PubMed Google Scholar), and G1 phase cyclin-Cdk-mediated phosphorylations are required to dissociate the E2F-4(5)-p130 complex (35Smith E.J. Leone G. DeGregori J. Jakoi L. Nevins J.R. Mol. Cell. Biol. 1996; 16: 6965-6976Crossref PubMed Scopus (207) Google Scholar). The importance of p107 and p130 phosphorylation for cell cycle progression is suggested by the growth arrest imposed by overexpression of phosphorylation-deficient mutants of p107 and p130 in tumor cell lines resistant to the corresponding wild-type proteins (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar, 36Ashizawa S. Nishizawa H. Yamada M. Higashi H. Kondo T. Ozawa H. Kakita A. Hatakeyama M. J. Biol. Chem. 2001; 276: 11362-11370Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar). Although the cell cycle regulation of p130 and p107 by phosphorylation is broadly reminiscent of the control over pRb, some important differences exist in the ways phosphorylation affects the individual pocket proteins. First, cyclins A and E are not dissociated by hyperphosphorylation of p107 and p130, thereby providing a phosphorylation-insensitive function (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). Second, p130 harbors a unique region in the B-domain, with several phosphorylation sites targeted by non-Cdk kinases (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar, 37Canhoto A.J. Chestukhin A. Litovchick L. DeCaprio J.A. Oncogene. 2000; 19: 5116-5122Crossref PubMed Scopus (44) Google Scholar). Third, p107 and p130 can be mutually distinguished through their differential responses to G1 cyclin-Cdks in vivo. Thus, phosphorylation of p130 and dissociation of E2F-4 can be achieved by either Cdk4(6) or Cdk2, and it occurs even in cells deficient in cyclin D-associated kinase activity. In contrast, Cdk2 activity is insufficient to release E2F-4 from p107, and this process requires the activity of Cdk4(6) (33Beijersbergen R.L. Carlee L. Kerkhoven R.M. Bernards R. Genes Dev. 1995; 9: 1340-1353Crossref PubMed Scopus (234) Google Scholar,38Xiao Z.X. Ginsberg D. Ewen M. Livingston D.M. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4633-4637Crossref PubMed Scopus (96) Google Scholar, 39Cheng L. Rossi F. Fang W. Mori T. Cobrinik D. J. Biol. Chem. 2000; 275: 30317-30325Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). Fourth, SV40 virus T-antigen binds to hyperphosphorylated forms of p107 and p130 but not of pRb, indicating a nonequivalent regulation of LXCXE-mediated protein interactions by phosphorylation (40Knudsen E.S. Wang J.Y. Oncogene. 1998; 16: 1655-1663Crossref PubMed Scopus (21) Google Scholar). Other LXCXE-mediated interactions, however, are controlled by phosphorylation, as exemplified by the reduced affinity of phosphorylated p107 for histone deacetylase (41Ferreira R. Magnaghi-Jaulin L. Robin P. Harel-Bellan A. Trouche D. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10493-10498Crossref PubMed Scopus (221) Google Scholar). Fifth, the fact that only three out of 22 in vivo phosphorylation sites in p130 are conserved in pRb (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar) suggests that p130 is subject to unique modes of regulation. There is a growing appreciation of the involvement of p107 and p130 in cell cycle regulation, and better understanding of how their activities are controlled is needed. In this study, we identify a distinct subset of in vivo phosphorylation sites in p130 necessary for regulation of E2F-4 and the cell cycle. In addition, a complementary subset of phosphorylation sites of p130 has only a minor influence on E2F-4-mediated transcription and cell cycle control. The latter phosphorylation sites may be involved in regulating processes distinct from E2F-4 binding. Comparative analysis of p130 with p107 suggests a functional conservation of the Cdk4-specific phosphorylation sites but also a striking difference in phosphorylation site usage between p107 and p130, providing an explanation for the differential kinase sensitivities of the two proteins. The human cell lines U-2-OS and T98G were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, penicillin, and streptomycin. Except for in vivo labeling experiments, transfections were performed by the calcium phosphate precipitation method (42Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Struhl K. Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York1996: 9.1.4-9.1.6Google Scholar) in dishes of cells grown to 30–50% confluence. The plasmid used for expression of HA-tagged wild-type p130 (HAp130wt) in eukaryotic cells, pcDNA.1-HAp130wt, was reported previously (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). The expression plasmid for HA-tagged p107 was created by digesting a synthetic double-stranded DNA encoding the HA-tag withNsiI and HpaI (sense strand, GCATAAAATGCATACCCCTACGACGTGCCCGACTACGCCTGTTAACTGCAGT) and inserting it downstream of the last sense codon in a previously described p107 expression plasmid (43Zhu L. van den Heuvel S. Helin K. Fattaey A. Ewen M. Livingston D. Dyson N. Harlow E. Genes Dev. 1993; 7: 1111-1125Crossref PubMed Scopus (470) Google Scholar). The NsiI andHpaI sites were introduced using a pair of complementary mutagenesis primers (sense strand, GAAAGAGCAAATCATGCATGTTGTTGTTAACTCTATGATAAAAGCAC) as described (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). Other point mutations in the p130 and p107 genes were created similarly. Mutants with more than one of the four clusters of phosphorylation sites completely substituted with alanines were created by joining regions using restriction enzymes as follows: p1303600 was made by joiningHindIII-Eco81I fragments from p1303000 and p1300600; p1303006 by joining HindIII-Eco81I fragments from p1303000 and p1300006; p1300670 by joining HindIII-PaeI fragments from p1300600 and p1300070; p1300606 by joining HindIII-PaeI fragments from p1300600 and p1300006; p1303606 by joining HindIII-Eco81I fragments from p1303000 and p1300606; p1303670 by joining HindIII-PaeI fragments from p1303606 and p1300070; p1303676 by joining HindIII-ScaI fragments from p1303670 (completely digested) and p1300006(partial digestion with ScaI). HAp130PM9A was constructed by joining the HindIII-PaeI fragments of a plasmid with mutations in codons 401, 413, 639, 662, 672 and a plasmid with mutations in codons 1035, 1044, 1080, 1112. HAp130PM12A was made similarly from a plasmid mutated in codons 642, 688, and 694 and a plasmid mutated in codons 948, 952, 962, 966, 973, 982, 986, 1068, and 1097. A Gal4-E2F-4 fusion protein expression plasmid was created by inserting a PCR product encoding full-length E2F-4 in the plasmid pM (CLONTECH) in the EcoRI and BamHI sites. The PCR primers contained these sites as 5′-extensions. The reporter construct 5×GAL-luc was created by PCR amplification of the region in pG5CAT (CLONTECH) containing the five GAL4 binding sites and the minimal promoter of the adenovirus E1b gene and insertion into the SacI and HindIII sites of pGL3-basic (Promega). The PCR primers contained these sites as 5′-extensions. A plasmid for expression of GSTp130wt fusion protein in bacteria (pGEX-2tp130wt) was created by joining aBamHI-XbaI fragment (partial digestion withXbaI) of pcDNA.1-HAp130wt to the vector pGEX-2t opened with BamHI and XbaI. The plasmid for expression of GSTp130PM22A was made similarly from pcDNA.1-HAp130PM22A. Immunoblotting and immunoprecipitation were performed as in Ref. 31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar. Antibodies 12CA5 and SC805 or SC7392 against the HA tag (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were used in immunoprecipitation and immunoblotting, respectively. Antibodies to p130 (SC317), E2F-4 (SC866), and cyclin A (SC751) were purchased from Santa Cruz Biotechnology. GST-p130wt and GST-p130PM22A were expressed in the Escherichia coli strain Bl21pLys from pGEX-2t. The E. coli cells were grown to A600 = 1 and harvested. Isopropyl-β-d-thiogalactopyranoside induction was omitted to avoid the appearance of incomplete fusion proteins. Cell extract was prepared as described (44Farkas T. Kutskova Y.A. Zimarino V. Mol. Cell. Biol. 1998; 18: 906-918Crossref PubMed Scopus (61) Google Scholar) in E. coli lysis buffer (20 mm Hepes, pH 7.2, 1 mm dithiothreitol, aprotinin (2.5 μg/ml), and leupeptin (2.5 μg/ml)). One pull-down reaction contained an amount of fusion protein derived from 10 ml ofE. coli culture, precoupled to glutathione-Sepharose beads, and 250 μg of protein from U-2-OS. The reaction was incubated at 4 °C for 90 min with gentle end-over-end mixing. Beads were washed three times in immunoprecipitation lysis buffer (50 mmHepes, pH 7.5, 150 mm NaCl, 1 mm EDTA, 2.5 mm EGTA, 0.1% Tween 20) supplemented with aprotinin (2.5 μg/ml), leupeptin (2.5 μg/ml), and dithiothreitol (1 mm) and boiled in 40 μl of 2× Laemmli sample buffer; 20 μl was analyzed by immunoblotting. U-2-OS cells were seeded in 6-cm dishes and transfected with the indicated type and amount of plasmid DNA. Sonicated salmon sperm DNA was added to a total of 8 μg of DNA per transfection. Forty hours after transfection, the cells were harvested and processed for reporter assays as described (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). Briefly, luciferase activity was measured using the Luciferase Assay System (Promega) and a Berthold Lumat LB95d instrument. The β-galactosidase assay was performed in 300 μl as follows: 30 μl of cleared lysate in 270 μl of reaction buffer (67 mm phosphate buffer, pH 7.5, 1.5 mg/mlo-nitrophenyl-β-d-galactopyranoside (Sigma), 1 mm MgCl2, 0.25% β-mercaptoethanol). The reaction was stopped by adding 500 μl of Na2CO3 (1 m). Optical density at 420 nm was measured. Relative luciferase activity was obtained by normalizing to β-galactosidase activity. Each transfection was performed in duplicate and repeated independently at least three times. Within one experiment, the fluctuation of the measured values did not exceed 10%. All measurements were performed so that the light intensities and optical densities were recorded within their respective linear ranges. The presented data were calculated as ratios between the activities of the pocket protein containing samples to those obtained from control transfected cells (transfected with either empty vector or with wild type p130 plasmid (see legends in Figs. 1B, 3, and 4B)). The latter control had been included repeatedly in each separate experiment. Because its value differs in separate measurements due to varying transfection efficiencies, it was arbitrarily set to 100 to allow comparison of multiple independent experiments.Figure 3Functional mapping of individual phosphorylation sites in p130 important for E2F-4 regulation.Plasmids expressing the indicated p130 constructs (500 ng) were tested for repression of cotransfected 5×Gal-luc (1 μg) and Gal4-E2F-4 (25 ng) in U-2-OS cells. A CMV-LacZ construct (500 ng) was cotransfected and used to normalize for transfection efficiency. Thediagrams in the top, middle, andbottom show repression data for alanine substitutions made in the A-domain proximal cluster of phosphorylation sites, the spacer domain, and the C-domain, respectively. The phosphorylation sites substituted for alanine in the mutants are indicated beloweach lane. The results are mean and S.D. values of three independent experiments. Effects of substitutions on reporter activity were related directly to p130wt, thereby defining the activity of Gal4-E2F-4 in the presence of p130wt as 100. The immunoblotsbelow the diagrams were made as follows. U-2-OS cells were transfected with 1 μg of the indicated p130 expression plasmid and 0.1 μg of the luciferase expression plasmid pCMV-luc. The p130 derivatives were immunoprecipitated with one anti-HA antibody (12CA5) and immunoblotted using another anti-HA antibody (Y-11). The input volumes of cell lysate in immunoprecipitations were normalized for transfection efficiency by measuring the luciferase activity.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4p130PM9A and p130PM12A have strong and weak ability, respectively, to regulate and bind E2F-4 and show specific loss of phosphopeptides corresponding to mutated residues. A, the diagram indicates the alanine substituted residues in p130PM9A and p130PM12A. Phosphopeptides in the maps are numbered according to their appearance in the primary amino acid sequence; identical numbers are used for residues located on the same tryptic phosphopeptide. Anasterisk indicates sample application point. Thearrows indicate the location of phosphopeptides giving rise to only a weak signal at the selected exposure. Thearrowhead indicates the location of a phosphopeptide of unknown identity (The "B2" spot described by Hansen et al. (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar)). We cannot explain the specific loss of B2 in the p130PM12A map. The existence of two phosphopeptides numbered 3 is due to a single and a double phosphorylation on the peptide containing Ser639 and Thr642. The existence of two phosphopeptides numbered 8 is due to an incomplete tryptic digest of the most positively charged peptide (the spot 8 migrating outermost right). B, repression of E2F-4 transactivation. Left part, U-2-OS cells were transfected with the indicated p130 constructs (0.5 μg), 5×Gal-luc (1 μg), Gal4-E2F-4 (25 ng), and CMV-LacZ (0.5 μg). The results are mean and S.D. values of three independent experiments. The effects of mutants on reporter activity were related directly to the effect of p130wt, defined as 100. Right part, same as left but cotransfected with either 2 μg of pX-p16 expression vector or pX empty vector as indicated.C, expression level of p130 mutants and coimmunoprecipitation of endogenous E2F-4. Left part, expression plasmids (1 μg) for the indicated HA-tagged p130 constructs were transfected into U-2-OS cells together with pCMV-luc (0.1 μg). HA-tagged p130 was immunoprecipitated with the anti-HA antibody, 12CA5, and immunoblotted for the HA-tag or E2F-4. To obtain equal signals in the p130wt, p130PM12A, p130PM9A, and p130PM22A lanes, volumes of extract with the following predetermined relative amounts of luciferase activity were immunoprecipitated: 1; 1.2; 3.5; 3.5, respectively. The total protein input in immunoprecipitations was kept constant by supplementing with U-2-OS lysate from untransfected cells.Right part, same as left part but cotransfected with 2 μg of pX-p16 expression vector; the input ratios of transfection units in immunoprecipitations were all equal to 1.View Large Image Figure ViewerDownload Hi-res image Download (PPT) U-2-OS cells were seeded in 10-cm dishes and transfected with 5 μg of pCMV-CD20 and the indicated amounts and types of pocket protein expression plasmids. Empty vector was added to a total of 15 μg of DNA per transfection. Occasionally 0.1 μg of pCMV-Luc was cotransfected as transfection control; in that case, one-third of the cells were analyzed for luciferase activity, as described above, and two-thirds were processed for flow cytometry, essentially as described (45Sorensen C.S. Lukas C. Kramer E.R. Peters J.M. Bartek J. Lukas J. Mol. Cell. Biol. 2001; 21: 3692-3703Crossref PubMed Scopus (110) Google Scholar). T98G cells were electroporated in the presence of the indicated plasmid DNA and labeled 42 h later for 4 h with [32P]orthophosphate (2 mCi/ml) in phosphate-free medium containing 10% dialyzed fetal calf serum. Phosphopeptide maps were made as described previously (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). Previously, we identified 22 in vivophosphorylation sites in p130 (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar). Among those, three were specifically targeted by Cdk4(6) (referred to here as Cdk4-specific), whereas most of the other phosphorylation sites were more general Cdk sites targeted by both Cdk4(6) and Cdk2. In vivo phosphorylation sites in p107, on the other hand, have not yet been systematically studied. Sequence alignments revealed conservation of 10 of the in vivo phosphorylation sites between p130 and p107, including all three Cdk4-specific sites (Fig.1A, top). A strict dependence on cyclin D-Cdk4(6) kinase activity for dissociation of p107 from E2F-4 has been described (33Beijersbergen R.L. Carlee L. Kerkhoven R.M. Bernards R. Genes Dev. 1995; 9: 1340-1353Crossref PubMed Scopus (234) Google Scholar, 38Xiao Z.X. Ginsberg D. Ewen M. Livingston D.M. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4633-4637Crossref PubMed Scopus (96) Google Scholar), whereas Cdk2-associated kinase activity alone is sufficient to dissociate p130 from E2F-4 (39Cheng L. Rossi F. Fang W. Mori T. Cobrinik D. J. Biol. Chem. 2000; 275: 30317-30325Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). This difference between the two related pocket proteins prompted us to compare the influence of the Cdk4-specific sites in p107 and p130 on E2F-4 activity, using the mutant p130ΔCdk4 with the three Cdk4-specific sites substituted with alanine (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20: 422-432Crossref PubMed Scopus (86) Google Scholar) and the homologous triple-alanine mutant p107ΔCdk4* made in this study (the asterisk indicates that the notion of Cdk4 specificity here is based on homology to p130). Our assay measured repression of a luciferase reporter with five Gal4 binding sites in front of a minimal promoter. Coexpressing the fusion protein Gal4-E2F-4, which contains the DNA binding domain of Gal4 fused to full-length E2F-4, induces transcription 250-fold higher than the Gal4 domain alone (data not shown). We refer to this activity as E2F-4-induced. This reporter type was chosen for its high sensitivity and ability to assess the E2F-4 pocket protein interaction. Before activity measurements, we estimated the impact of the triple-alanine substitution (T369A,S650A,S964A) on the global phosphorylation status of p107 by tryptic phosphopeptide mapping of in vivo 32P-labeled p107 versus p130 as a reference. The four phosphopeptides representing the Cdk4-specific sites are, as previously published (31Hansen K. Farkas T. Lukas J. Holm K. Ronnstrand L. Bartek J. EMBO J. 2001; 20:
Referência(s)