Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

p73 Is Regulated by Phosphorylation at the G2/M Transition

2003; Elsevier BV; Volume: 278; Issue: 49 Linguagem: Inglês

10.1074/jbc.m304921200

ISSN

1083-351X

Autores

Marcella Fulco, Antonio Costanzo, Paola Merlo, Rosamaria Mangiacasale, Sabrina Strano, Giovanni Blandino, Clara Balsano, Patrizia Lavia, Massimo Levrero,

Tópico(s)

Cell death mechanisms and regulation

Resumo

p73 is a p53 paralog that encodes proapoptotic (transactivation-competent (TA)) and antiapoptotic (dominant negative) isoforms. TAp73 transcription factors mediate cell cycle arrest and/or apoptosis in response to DNA damage and are involved in developmental processes in the central nervous system and the immune system. p73 proteins may also play a role in the regulation of cell growth. Indeed, p73 expression is itself modulated during the cell cycle and TAp73 proteins accumulate in S phase cells. In addition, the function of p73 proteins is also regulated by post-translational modifications and protein-protein interactions in different cellular and pathophysiological contexts. Here we show that p73 is a physiological target of the p34cdc2-cyclin B mitotic kinase complex in vivo. Both p73β and p73α isoforms are hyperphosphorylated in normal mitotic cells and during mitotic arrest induced by microtubule-targeting drugs. p34cdc2-cyclin B phosphorylates and associates with p73 in vivo, which results in a decreased ability of p73 to both bind DNA and activate transcription in mitotic cells. Indeed, p73 is excluded from condensed chromosomes in meta- and anaphase, redistributes throughout the mitotic cytoplasm, and unlike p53, shows no association with centrosomes. Together these results indicate that M phase-specific phosphorylation of p73 by p34cdc2-cyclin B is associated with negative regulation of its transcriptional activating function. p73 is a p53 paralog that encodes proapoptotic (transactivation-competent (TA)) and antiapoptotic (dominant negative) isoforms. TAp73 transcription factors mediate cell cycle arrest and/or apoptosis in response to DNA damage and are involved in developmental processes in the central nervous system and the immune system. p73 proteins may also play a role in the regulation of cell growth. Indeed, p73 expression is itself modulated during the cell cycle and TAp73 proteins accumulate in S phase cells. In addition, the function of p73 proteins is also regulated by post-translational modifications and protein-protein interactions in different cellular and pathophysiological contexts. Here we show that p73 is a physiological target of the p34cdc2-cyclin B mitotic kinase complex in vivo. Both p73β and p73α isoforms are hyperphosphorylated in normal mitotic cells and during mitotic arrest induced by microtubule-targeting drugs. p34cdc2-cyclin B phosphorylates and associates with p73 in vivo, which results in a decreased ability of p73 to both bind DNA and activate transcription in mitotic cells. Indeed, p73 is excluded from condensed chromosomes in meta- and anaphase, redistributes throughout the mitotic cytoplasm, and unlike p53, shows no association with centrosomes. Together these results indicate that M phase-specific phosphorylation of p73 by p34cdc2-cyclin B is associated with negative regulation of its transcriptional activating function. p73 is a nuclear protein that shares substantial sequence homology and functional similarities with p53 (1.Irwin M.S. Kaelin W.G. Cell Growth Differ. 2001; 12: 337-349PubMed Google Scholar, 2.Levrero M. De Laurenzi V. Costanzo A. Gong J. Wang J.Y. Melino G. J. Cell Sci. 2000; 113: 1661-1670Crossref PubMed Google Scholar, 3.Strano S. Rossi M. Fontemaggi G. Munarriz E. Soddu S. Sacchi A. Blandino G. FEBS Lett. 2001; 490: 163-170Crossref PubMed Scopus (88) Google Scholar). Multiple TA 1The abbreviations used are: TAtransactivation-competentIKKIκB kinaseDAPI4-6-diamidino-2-phenylindoleCDKcyclin-dependent kinaseCMVcytomegalovirusGSTglutathione S-transferase.1The abbreviations used are: TAtransactivation-competentIKKIκB kinaseDAPI4-6-diamidino-2-phenylindoleCDKcyclin-dependent kinaseCMVcytomegalovirusGSTglutathione S-transferase. (transactivation-competent, pro-apoptotic, and anti-proliferative) p73 COOH-terminal splicing isoforms (α, β, γ, δ, ϵ, ζ) exist (4.De Laurenzi V. Costanzo A. Barcaroli D. Terrinoni A. Falco M. Annicchiarico-Petruzzelli M. Levrero M. Melino G. J. Exp. Med. 1998; 188: 1763-1768Crossref PubMed Scopus (361) Google Scholar, 5.De Laurenzi V.D. Catani M.V. Terrinoni A. Corazzari M. Melino G. Costanzo A. Levrero M. Knight R.A. Cell Death Differ. 1999; 6: 389-390Crossref PubMed Scopus (131) Google Scholar). In addition, dominant negative variants, which are expressed from a second promoter, lack the amino-terminal transactivation domain, act as trans-repressors of p53- and p73-dependent transcription, and possess anti-apoptotic and pro-proliferative potential (6.Pozniak C.D. Radinovic S. Yang A. McKeon F. Kaplan D.R. Miller F.D. Science. 2000; 289: 304-306Crossref PubMed Scopus (406) Google Scholar, 7.Yang A. Kaghad M. Wang Y. Gillett E. Fleming M.D. Dotsch V. Andrews N.C. Caput D. McKeon F. Mol. Cell. 1998; 2: 305-316Abstract Full Text Full Text PDF PubMed Scopus (1834) Google Scholar, 8.Vossio S. Palescandolo E. Pediconi N. Moretti F. Balsano C. Levrero M. Costanzo A. Oncogene. 2002; 21: 3796-3803Crossref PubMed Scopus (71) Google Scholar). This complex expression strategy of the p53 family of tumor suppressors may explain why the attempt to correlate p53 status with prognosis and response to anticancer treatments has proved of limited clinical impact in the past. transactivation-competent IκB kinase 4-6-diamidino-2-phenylindole cyclin-dependent kinase cytomegalovirus glutathione S-transferase. transactivation-competent IκB kinase 4-6-diamidino-2-phenylindole cyclin-dependent kinase cytomegalovirus glutathione S-transferase. Although exogenously expressed TAp73 proteins transactivate endogenous targets of p53, several lines of evidence indicate that in vivo p53 and p73 may be differentially regulated and carry out specialized, non-overlapping functions. Indeed, p73- and p63-deficient mice exhibit specific phenotypes related to abnormal differentiation of brain and skin, respectively, but do not show defects related to spontaneous or x-ray-induced apoptosis during embryogenesis, nor do they develop spontaneous cancers (9.Mills A.A. Zheng B. Wang X.J. Vogel H. Roop D.R. Bradley A. Nature. 1999; 398: 708-713Crossref PubMed Scopus (1694) Google Scholar, 10.Yang A. Schweitzer R. Sun D. Kaghad M. Walker N. Bronson R.T. Tabin C. Sharpe A. Caput D. Crum C. McKeon F. Nature. 1999; 398: 714-718Crossref PubMed Scopus (1900) Google Scholar, 11.Yang A. Walker N. Bronson R. Kaghad M. Oosterwegel M. Bonnin J. Vagner C. Bonnet H. Dikkes P. Sharpe A. McKeon F. Caput D. Nature. 2000; 404: 99-103Crossref PubMed Scopus (880) Google Scholar). Second, p53 is induced in response to a variety of DNA-damaging agents and cellular stressors, whereas p73 is induced only by a subset of agents, including cisplatin, doxorubicin, and irradiation, in a p53-independent/c-abl-dependent manner (12.Agami R. Blandino G. Oren M. Shaul Y. Nature. 1999; 399: 809-813Crossref PubMed Scopus (504) Google Scholar, 13.Gong J.G. Costanzo A. Yang H.Q. Melino G. Kaelin Jr., W.G. Levrero M. Wang J.Y. Nature. 1999; 399: 806-809Crossref PubMed Scopus (831) Google Scholar, 14.Irwin M.S. Kondo K. Marin M.C. Cheng L.S. Hahn W.C. Kaelin W.G. Cancer Cell. 2003; 3: 403-410Abstract Full Text Full Text PDF PubMed Scopus (362) Google Scholar). Finally, the specificity of p73 proteins role in the apoptotic response to DNA damage is further underlined by the recent evidence that p73 is required for p53-dependent apoptosis induced by DNA damage (15.Flores E.R. Tsai K.Y. Crowley D. Sengupta S. Yang A. McKeon F. Jacks T. Nature. 2002; 416: 560-564Crossref PubMed Scopus (722) Google Scholar) and that blocking p73 function with a dominant negative or by specific small interfering RNA leads to chemoresistance of human tumor cells, irrespective of their p53 status (14.Irwin M.S. Kondo K. Marin M.C. Cheng L.S. Hahn W.C. Kaelin W.G. Cancer Cell. 2003; 3: 403-410Abstract Full Text Full Text PDF PubMed Scopus (362) Google Scholar). In addition, p73 mediates T-cell receptor activation-induced cell death (TCR-AICD), a p53-independent apoptotic process that occurs in mature peripheral T cells (16.Lissy N.A. Davis P.K. Irwin M. Kaelin W.G. Dowdy S.F. Nature. 2000; 407: 642-645Crossref PubMed Scopus (288) Google Scholar). Thus, p73 and p53 act through specific and distinctive pathways, although some of their functions clearly overlap. p73 proteins are regulated both at the transcriptional level and by post-translational modifications in different cellular and pathophysiological contexts. The p73 gene has been reported to be transcriptionally regulated by the transcription factor E2F1 in apoptosis induced by E2F1 overexpression (17.Stiewe T. Putzer B.M. Nat. Genet. 2000; 26: 464-469Crossref PubMed Scopus (307) Google Scholar, 18.Irwin M. Marin M.C. Phillips A.C. Seelan R.S. Smith D.I. Liu W. Flores E.R. Tsai K.Y. Jacks T. Vousden K.H. Kaelin Jr., W.G. Nature. 2000; 407: 645-648Crossref PubMed Scopus (534) Google Scholar), in response to DNA damage (19.Pediconi N. Ianari A. Costanzo A. Belloni L. Gallo R. Cimino L. Porcellini A. Screpanti I. Balsano C. Alesse E. Gulino A. Levrero M. Nat. Cell Biol. 2003; (in press)PubMed Google Scholar), and in physiological conditions during the S phase of the cell cycle (18.Irwin M. Marin M.C. Phillips A.C. Seelan R.S. Smith D.I. Liu W. Flores E.R. Tsai K.Y. Jacks T. Vousden K.H. Kaelin Jr., W.G. Nature. 2000; 407: 645-648Crossref PubMed Scopus (534) Google Scholar). Specific phosphorylation and acetylation events contribute to the activation of the p73 gene product in response to DNA damage (12.Agami R. Blandino G. Oren M. Shaul Y. Nature. 1999; 399: 809-813Crossref PubMed Scopus (504) Google Scholar, 13.Gong J.G. Costanzo A. Yang H.Q. Melino G. Kaelin Jr., W.G. Levrero M. Wang J.Y. Nature. 1999; 399: 806-809Crossref PubMed Scopus (831) Google Scholar, 20.Costanzo A. Merlo P. Pediconi N. Fulco M. Sartorelli V. Cole P.A. Fontemaggi G. Fanciulli M. Schiltz L. Blandino G. Balsano C. Levrero M. Mol. Cell. 2002; 9: 175-186Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar, 21.Yuan Z.M. Shioya H. Ishiko T. Sun X. Gu J. Huang Y.Y. Lu H. Kharbanda S. Weichselbaum R. Kufe D. Nature. 1999; 399: 814-817Crossref PubMed Scopus (539) Google Scholar) and bolster p73 apoptotic functions by potentiating the selective recruitment of p73 onto the promoters of apoptotic target genes versus genes involved in cell cycle arrest and re-entry (20.Costanzo A. Merlo P. Pediconi N. Fulco M. Sartorelli V. Cole P.A. Fontemaggi G. Fanciulli M. Schiltz L. Blandino G. Balsano C. Levrero M. Mol. Cell. 2002; 9: 175-186Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar). A number of positive and negative regulators have been shown to interact with or modulate the activity of p73. Hmdm2 inhibits p73 transcriptional activity but does not affect its stability (22.Balint E. Bates S. Vousden K.H. Oncogene. 1999; 18: 3923-3929Crossref PubMed Scopus (194) Google Scholar, 23.Dobbelstein M. Wienzek S. Konig C. Roth J. Oncogene. 1999; 18: 2101-2106Crossref PubMed Scopus (144) Google Scholar, 24.Ongkeko W.M. Wang X.Q. Siu W.Y. Lau A.W. Yamashita K. Harris A.L. Cox L.S. Poon R.Y. Curr. Biol. 1999; 9: 829-832Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 25.Zeng X. Chen L. Jost C.A. Maya R. Keller D. Wang X. Kaelin Jr., W.G. Oren M. Chen J. Lu H. Mol. Cell. Biol. 1999; 19: 3257-3266Crossref PubMed Scopus (303) Google Scholar), and a subset of tumor-derived p53 mutants block p73 functions (26.Di Como C.J. Gaiddon C. Prives C. Mol. Cell. Biol. 1999; 19: 1438-1449Crossref PubMed Scopus (380) Google Scholar, 27.Gaiddon C. Lokshin M. Ahn J. Zhang T. Prives C. Mol. Cell. Biol. 2001; 21: 1874-1887Crossref PubMed Scopus (381) Google Scholar, 28.Strano S. Munarriz E. Rossi M. Cristofanelli B. Shaul Y. Castagnoli L. Levine A.J. Sacchi A. Cesareni G. Oren M. Blandino G. J. Biol. Chem. 2000; 275: 29503-29512Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar, 29.Marin M.C. Jost C.A. Brooks L.A. Irwin M.S. O'Nions J. Tidy J.A. James N. McGregor J.M. Harwood C.A. Yulug I.G. Vousden K.H. Allday M.J. Gusterson B. Ikawa S. Hinds P.W. Crook T. Kaelin Jr., W.G. Nat. Genet. 2000; 25: 47-54Crossref PubMed Scopus (468) Google Scholar, 30.Bergamaschi D. Gasco M. Hiller L. Sullivan A. Syed N. Trigiante G. Yulug I. Merlano M. Numico G. Comino A. Attard M. Reelfs O. Gusterson B. Bell A.K. Heath V. Tavassoli M. Farrell P.J. Smith P. Lu X. Crook T. Cancer Cell. 2003; 3: 387-402Abstract Full Text Full Text PDF PubMed Scopus (383) Google Scholar). On the other hand, the WW domain containing YES-associated protein (YAP) binds p73, but not p53, and potentiates its transcriptional activity toward apoptotic target genes (31.Strano S. Munarriz E. Rossi M. Castagnoli L. Shaul Y. Sacchi A. Oren M. Sudol M. Cesareni G. Blandino G. J. Biol. Chem. 2001; 276: 15164-15173Abstract Full Text Full Text PDF PubMed Scopus (352) Google Scholar). More recently, it has been also reported that p73 interacts physically with various cyclins in vitro, including cyclins A, B, D, and E, and that multiple p73 isoforms are phosphorylated by CDK1- and CDK2-containing complexes at threonine 86; however, the physiological relevance of these interactions in vivo has not yet been clearly established (32.Gaiddon C. Lokshin M. Gross I. Levasseur D. Taya Y. Loeffler J.P. Prives C. J. Biol. Chem. 2003; 278: 27421-27431Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). Here we show that p73 is a target of the p34cdc2-cyclin B mitotic kinase complex in vivo. Both p73β and p73α isoforms are hyperphosphorylated during physiological mitosis and during mitotic arrest induced by microtubule-targeting drugs. p34cdc2-cyclin B phosphorylates and associates with p73 in vivo, which results in a decreased ability of p73 to both bind DNA and activate transcription in mitotic cells. This is associated with p73 detachment from condensed chromosomes and redistribution throughout the mitotic cytoplasm in meta- and anaphase. Cell Culture and Transfections—Human HCT116(3)(6) (colon carcinoma), U2OS (osteosarcoma), MCF7 (breast cancer), and HeLa (cervix carcinoma) cells were cultured in Dulbecco's modified Eagle's medium plus antibiotics and 10% heated-inactivated fetal bovine serum. HCT-116(3) cells are a subclone of the HCT-116 colon carcinoma cell line in which chromosome 3 was reintroduced to restore the ability to activate c-abl in response to cisplatin (33.Boyer J.C. Umar A. Risinger J.I. Lipford J.R. Kane M. Yin S. Barrett J.C. Kolodner R.D. Kunkel T.A. Cancer Res. 1995; 55: 6063-6070PubMed Google Scholar). The human lung carcinoma cell line H1299 was maintained in RPMI medium supplemented with 10% fetal calf serum. H1299 ponasterone-inducible p73α and wild-type p53 cell lines were described elsewhere (28.Strano S. Munarriz E. Rossi M. Cristofanelli B. Shaul Y. Castagnoli L. Levine A.J. Sacchi A. Cesareni G. Oren M. Blandino G. J. Biol. Chem. 2000; 275: 29503-29512Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar, 34.Fontemaggi G. Kela I. Amariglio N. Rechavi G. Krishnamurthy J. Strano S. Sacchi A. Givol D. Blandino G. J. Biol. Chem. 2002; 277: 43359-43368Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). Transient transfections were performed by either calcium phosphate precipitation or using the LipofectAMINE Plus reagent (Invitrogen). Samples for flow cytometry analysis of cell cycle distribution were prepared by fixing the cells in 70% ethanol (30 min, 4 °C), washing twice in phosphate-buffered saline plus 0.5% Tween 20, and stained with 50 μg/ml propidium iodide. Samples were analyzed in a FACStar Plus (BD Biosciences) flow cytometer (10,000 events/sample) using the WinMDI software. Antibodies, Plasmids, Recombinant Proteins, and Reagents—Antibodies directed against human p34cdc2 (17.Stiewe T. Putzer B.M. Nat. Genet. 2000; 26: 464-469Crossref PubMed Scopus (307) Google Scholar), cyclin B1 (clone GNS1), p21 (clone H164), IKKα (clone C20), E2F1 (clone C20), actin (clone I19), and the HA tag (clone Y11) were purchased from Santa Cruz Biotechnology, Inc. The mouse monoclonal anti-p73 antibody (clone 1288, Imgenex Inc.) recognizes all known splicing variants of the human and mouse p73. 2Y. J. Y. Wang and A. Costanzo, unpublished observations. The polyclonal anti-p73 H79 antibody (Santa Cruz) was used in immunoprecipitation experiments. Murine anti-γ-tubulin antibodies were from Dako, Inc. Secondary anti-mouse fluorescein isothiocyanate- and rhodamine-conjugated antibodies were from Sigma and Jackson Immunoresearch, respectively. The pCDNAHA, pCDNAHA-p73α, pCDNAHA-p73β, and pCDNAHA-p53 expression vectors and the p21-luc reporter plasmid were described previously (4.De Laurenzi V. Costanzo A. Barcaroli D. Terrinoni A. Falco M. Annicchiarico-Petruzzelli M. Levrero M. Melino G. J. Exp. Med. 1998; 188: 1763-1768Crossref PubMed Scopus (361) Google Scholar, 13.Gong J.G. Costanzo A. Yang H.Q. Melino G. Kaelin Jr., W.G. Levrero M. Wang J.Y. Nature. 1999; 399: 806-809Crossref PubMed Scopus (831) Google Scholar). The pCDNAHA-p73DD plasmid expresses a p73 miniprotein that acts as a selective dominant negative of p73-dependent, but not p53-dependent, transcription (Ref. 18.Irwin M. Marin M.C. Phillips A.C. Seelan R.S. Smith D.I. Liu W. Flores E.R. Tsai K.Y. Jacks T. Vousden K.H. Kaelin Jr., W.G. Nature. 2000; 407: 645-648Crossref PubMed Scopus (534) Google Scholar and data not shown). pCDNAHA-p73α(T86A) was generated by PCR mutagenesis. The CMVneoHAp34cdc2 and CMVneoHAp34cdc2dominant negative plasmids were kindly provided by Dr. van den Heuvel (Boston, MA) (35.van den Heuvel S. Harlow E. Science. 1993; 262: 2050-2054Crossref PubMed Scopus (970) Google Scholar). CMV-neo-myc-cyclin B1 was kindly provided by G. Piaggio. Procaryotic expression vectors for GST-p73α, GST-p73β, GST-p73-(1–120), GST-p73-(1–321), GST-p73-(120–321), GST-p73-(321–636), GST-p73-(1–249), and GST-p73-(1–249)T86A, were constructed by standard PCR techniques (primers available upon request) and verified by sequencing. GST fusion proteins were purified as described previously (28.Strano S. Munarriz E. Rossi M. Cristofanelli B. Shaul Y. Castagnoli L. Levine A.J. Sacchi A. Cesareni G. Oren M. Blandino G. J. Biol. Chem. 2000; 275: 29503-29512Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar). Histone H1, nocodazole, paclitaxel (taxol), propidium iodide, hydroxyurea, and olomoucine were purchased from Sigma. Ponasterone A was from Alexis Biochemical, Inc. Immunoblotting, Immunoprecipitations, and Kinase Assays—Immunoblotting and immunoprecipitation assays were performed as described (13.Gong J.G. Costanzo A. Yang H.Q. Melino G. Kaelin Jr., W.G. Levrero M. Wang J.Y. Nature. 1999; 399: 806-809Crossref PubMed Scopus (831) Google Scholar). For p34cdc2-cyclin B1 kinase assay, immunoprecipitates and purified GST fusion proteins were incubated in kinase buffer (20 μl) containing 50 mm Tris-HCl (pH 7.5), 10 mm MgCl2, 1 mm EGTA, 1 mm dithiothreitol, 100 μm ATP, and 15 μCi of [γ-32 P]ATP in the presence of 20 units of highly purified recombinant p34cdc2-cyclin B complex (New England Biolabs). The reaction was incubated for 30 min at 30 °C and terminated by adding 10 μl of 3× SDS sample buffer. Samples were resolved through 10% SDS-PAGE, blotted on polyvinylidene difluoride membranes and analyzed by autoradiography. For pull-down assays, 2 μg of either immobilized purified GST fusion proteins or wild-type GST were incubated for 2 h at 4 °C in the presence of 0.5 μg of a 55-kDa full-length p34cdc2 fusion protein produced in Escherichia coli (Santa Cruz). After 6 washes in radioimmune precipitation buffer, proteins were resolved by SDS-PAGE and probed with anti-p34cdc2 antibody. Binding to Biotinylated Oligonucleotides—200 ng of biotinylated oligonucleotides encompassing the p53-binding sites of the p21 promoter were immobilized to Dynabeads M-280 streptavidin (Dynal, AS, Oslo, Norway) and then incubated for 30 min at room temperature in the presence of in vitro translated (TnT, Promega) p73α and p73α-T86A proteins. After three washes, oligonucleotide-bound p73 proteins were incubated in 20 μl of kinase buffer with (20 units) or without highly purified recombinant Cdc2-cyclin B complex (New England Biolabs). Where indicated the p34cdc2-cyclin B complex was preincubated with 0.5 mm olomoucine for 10 min at room temperature. After a 30-min incubation at 30 °C, oligonucleotide-bound p73 proteins were washed twice and resolved through 10% SDS-PAGE, blotted to polyvinylidene difluoride membranes, and analyzed using the 1288 anti-p73 antibody. Cytological Preparations and Microscopy—Cells were fixed in 3.7% paraformaldehyde and processed for indirect immunofluorescence using antibodies against p73 (1:100, Imgenex) and p53 (1:200, Dako) γ-tubulin (3.5 μg/ml in phosphate-buffered saline, 0.1% Tween 20, 5% fetal calf serum) followed by Texas Red- and fluorescein isothiocyanate-conjugated secondary anti-mouse IgG antibody (800-fold dilution in phosphate-buffered saline, 0.1% Tween 20, 5% fetal calf serum) before counterstaining in 0.1 μg/ml DAPI. Images were analyzed using an Olympus AX70 microscope configured for epifluorescence and equipped with a cooled camera device (Photometrics). p73α and -β Are Phosphorylated during Mitosis—As a first step toward extending our current knowledge of p73 regulation during the cell cycle (see the Introduction and references therein), we examined p73 steady state levels in cells that were drug-synchronized in different cell cycle phases. As shown in Fig. 1A, the amount of both p73α and -β isoforms increases in S phase-enriched cultures as compared with exponentially growing asynchronous cells (Fig. 1A, compare nt and S in rows a and b). The increase in p73 parallels the increase in E2F1 protein levels, consistent with the notion that p73 is regulated by E2F1 during G1/S progression (18.Irwin M. Marin M.C. Phillips A.C. Seelan R.S. Smith D.I. Liu W. Flores E.R. Tsai K.Y. Jacks T. Vousden K.H. Kaelin Jr., W.G. Nature. 2000; 407: 645-648Crossref PubMed Scopus (534) Google Scholar). p73 protein levels decrease in cells progressing from S into the G2 phase (Fig. 1A, compare lanes S and G2 in rows a and b). In nocodazole-arrested M phase cells, p73β levels increase, and both p73α and p73β isoforms display a slower electrophoretic mobility (Fig. 1A, lane M in rows a and b). λ-Phosphatase treatment shifted the migration of mitotic p73α back to that of its interphase counterpart (Fig. 1B, compare lanes 3 and 4 with lanes 1 and 2), indicating that p73 proteins are modified by phosphorylation in nocodazole-treated cells. To establish whether p73 phosphorylation observed in nocodazole-arrested cells reflected a mitotic specific modification, we separated mitotic from interphase cells by vigorous shaking and found that p73 hyperphosphorylation was confined to mitotic cells and was absent in culture samples that remained adherent and were enriched in G2 phase cells (Fig. 1C, left panel). Both p73α and -β displayed a similar mobility shift in taxol-treated cells (Fig. 1C, middle panel); thus, p73 phosphorylation was associated with mitotic arrest, irrespective of the drug specificity in the mechanism of action. Most importantly, p73 shift was also observed in mitotic cells collected by "shake-off" from an asynchronous population of untreated HCT116(3) cells (Fig. 1C, right panel). Similar results were obtained in human HeLa and breast cancer MCF7 cells (data not shown). Overall these results indicate that p73α and -β are subjected to mitosis-specific phosphorylation, both during the physiological mitotic division and during the mitotic arrest induced by microtubule-targeting drugs. p73 Is Phosphorylated by the p34cdc2-Cyclin B Complex in Vivo and in Vitro—p34cdc2-cyclin B is the key regulator of mitosis and is the kinase complex responsible for the G2/M transition. A number of proteins involved in transcriptional regulation, including RNA PolII, the basal factor TFIID (transcription factor IID), and several sequence-specific transcription factors, are subjected to mitosis-specific phosphorylation, which in most cases is performed by the cdc2-cyclin B kinase (see Ref. 36.Gottesfeld J.M. Forbes D.J. Trends Biochem. Sci. 1997; 22: 197-202Abstract Full Text PDF PubMed Scopus (310) Google Scholar for review). To test whether p34cdc2-cyclin B is involved in p73 mitotic phosphorylation, we first carried out an in vitro kinase assays using a highly purified p34cdc2-cyclin B complex. This complex efficiently phosphorylates histone H1, indicating that its activity is maintained in vitro (data not shown). p34cdc2-cyclin B complex phosphorylates a bacterially expressed GST-p73α fusion protein (Fig. 2A, left panel). In addition, we found that endogenous p73α protein immunoprecipitated from HCT116(3) cells acts as an efficient phosphorylation substrate of p34cdc2-cyclin B, contrary to IKKα protein immunoprecipitated from the same cultures (Fig. 2A, right panel). These results therefore identify p34cdc2-cyclinB as the candidate kinase that phosphorylates p73 in vivo during mitosis. By assaying several p73α deletion mutants expressed as GST fusion proteins (depicted in Fig. 2C) in an in vitro phosphorylation assay, we found that the amino-terminal region of p73 (residues 1–120) is targeted by p34cdc2-cyclin B phosphorylation (Fig. 2B, left panel). This region contains a single TPEH sequence matching the consensus Ser/Thr-Pro-X-basic amino acid phosphorylation site for p34cdc2 (37.Holmes J.K. Solomon M.J. J. Biol. Chem. 1996; 271: 25240-25246Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar) with the potentially phosphorylated threonine in position 86. We found that a T86A substitution in the GST-p73α-(1–120) fusion protein abolished phosphorylation by p34cdc2-cyclin B (Fig. 2D, left panel), consistent with results recently reported by Gaiddon et al. (32.Gaiddon C. Lokshin M. Gross I. Levasseur D. Taya Y. Loeffler J.P. Prives C. J. Biol. Chem. 2003; 278: 27421-27431Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). To assess the physiological relevance of these findings exogenous HA-tagged T86A-p73α, which cannot be phosphorylated by p34cdc2 in vitro, was expressed in cell cultures subjected to nocodazole synchronization. No electrophoretic variation was detected (Fig. 2D, right panel), indicating that no other phosphorylation event contributes to the M phase-specific pattern of p73 migration observed in vivo (see Fig. 1A). These results indicate that p73 is an in vivo target of phosphorylation by p34cdc2-cyclin B on threonine 86. p34cdc2-Cyclin B1 Complex Binds p73 in Mitosis—Several p73 isoforms have been shown to interact with purified cyclin-Cdk complexes in vitro (32.Gaiddon C. Lokshin M. Gross I. Levasseur D. Taya Y. Loeffler J.P. Prives C. J. Biol. Chem. 2003; 278: 27421-27431Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). Because in vivo p73 displays an obvious susceptibility to modifications during mitosis (see Fig. 1), we next asked whether p73 and the p34cdc2-cyclin B complex do associate physically in vivo. As shown in Fig. 3A, p34cdc2 immunoprecipitates from asynchronously growing HCT116(3) cells contained p73α. It is noteworthy that the amount of coassociated p73α was significantly higher in p34cdc2 immunoprecipitates from nocodazole-arrested cells compared with asynchronously growing cultures (Fig. 3A), further indicating that p73 does associate with p34cdc2-cyclin B complex during mitosis. The association of p73 with p34cdc2 was followed up during mitotic exit and re-entry into G1 after release from nocodazole arrest (Fig. 3B). Levels of p34cdc2-cyclin B were down-regulated, as expected. In parallel, we observed a progressive reduction in the amount of p34cdc2-associated p73α at 6 h after nocodazole release; when all cells were beyond mitosis and into G1 phase, p34cdc2-p73α association was barely detectable (Fig. 3B). To assess whether the association between p73 and p34cdc2-cyclin B observed in vivo was direct or mediated by other cellular proteins, we performed pull-down experiments using a panel of GST-p73 fusion proteins and a commercial 55-kDa full-length p34cdc2 fusion protein produced in E. coli. p34cdc2 efficiently binds to GST-p73α and GST-p73β (Fig. 3C), and the association is maintained only with the GST p73-(1–120) deletion mutant (Fig. 3C). Together these results demonstrate that p73 associates directly with p34cdc2-cyclin B complex through the amino-terminal region of p73 and the association is physiologically established in vivo. p34cdc2-Cyclin B-mediated Phosphorylation Interferes with p73 Functions—Although mitosis is accompanied by a global repression of nuclear RNA synthesis in higher eukaryotes, increasing evidence indicate that some genes are stably transcribed even in the context of the highly condensed mitotic chromatin. Indeed, the human cyclin B1 gene itself is actively transcribed during mitosis (38.Sciortino S. Gurtner A. Manni I. Fontemaggi G. Dey A. Sacchi A. Ozato K. Piaggio G. EMBO Rep. 2001; 2: 1018-1023Crossref PubMed Scopus (50) Google Scholar). To assess the impact of p34cdc2-cyclin B-mediated phosphorylation on p73 transcriptional activity in vivo, we co-transfected the HA-p73α, HA-p34cdc2, and Myc-cyclin B expression vectors together with a luciferase reporter gene driven by the p53–73-responsive p21 promoter. p73-mediated transcriptional activation of the p21 promoter was reduced by 50% when p73 was co-transfected with either HA-p34cdc2 alone or with HA-p34cdc2 and Myc-cyclin B (Fig. 4A). The nonphosphorylatable p73-T86A mutant retained a somewhat lower ability to activate p21 promoter-driven transcription, which was not affected by co-transfection of HAp34cdc2 alone or with Myc-cyclin B (Fig. 4A). Transcriptional repression is thought to reflect both the increased condensation of mitotic chromatin (39.Martinez-Balbas M.A. Dey A. Rabindran S.K. Ozato K. Wu C. Cell. 1995; 83: 29-38Abstract Full Text PDF PubMed Scopus (361) Google Scholar) and phosphorylation events that inhibit the binding of transcriptional activators to their cognate DNA target (36.Gottesfeld J.M. Forbes D.J. Trends Biochem. Sci. 1997; 22: 197-202Abstract Full Text PDF PubMed Scopus (310) Google Scholar). However, the mitotic dispersal of RNA polymerase II factors and sequence-specific transcription factors from the condensed mitotic chromosomes is not general. For example, AP2 (39.Martinez-Balbas M.A. Dey A. Rabindran S.K. Ozato K. Wu C. Cell. 1995; 83: 29-38Abstract Full Text PDF PubMed Scopus (361) Google Scholar) and NF-Y (38.Sciortino S. Gurtner A. Manni I. Fontemaggi G. Dey A. Sacchi A. Ozato K. Piaggio G. EMBO Rep. 2001; 2: 1018-1023Crossref PubMed Scopus (50) Google Scholar), serum response factor p67SRF (40.Gauthier-Rouviere C. Cavadore J.C. Blanchard J.M. Lamb N.J. Fernandez A. Cell Regul. 1991; 2: 575-588Crossref PubMed Scopus (53) Google Scholar), and a significant proportion of the TATA-binding protein (TBP) and TBP-associated factor 250 (TAFII250) (41.Chen D. Hinkley C.S. Henry R.W. Huang S. Mol. Biol. Cell. 2002; 13: 276-284Crossref PubMed Scopus (124) Google Scholar) remain associated with the condensed and transcriptionally inactive mitotic chromatin. To assess whether mitotic p34cdc2-cyclin B-dependent phosphorylation of p73 interferes with its DNA binding ability, we tested the effect of purified active p34cdc2-cyclin B on preassembled DNA-protein complexes containing in vitro translated p73 and a biotinylated p73-binding oligonucleotide from the p21 promoter. As shown in Fig. 4B, the amount of p73α bound to its cognate site was drastically reduced in the presence of purified active p34cdc2-cyclin B (compare lane 2 with lane 3). The amount of oligonucleotide-bound p73α remained instead unchanged when the pre-assembled DNA-protein complexes were incubated with p34cdc2-cyclin B preincubated with 0.5 μm olomoucine, a specific inhibitor of the p34cdc2 mitotic kinase (Fig. 4B, compare lanes 4 and 3). Thus, the reduced affinity of p73 for its DNA target sites is indeed mediated by p34cdc2 kinase activity. To further demonstrate that p34cdc2-cyclin B-mediated phosphorylation of p73 was responsible for its reduced DNA binding ability, the same experiment was performed using the in vitro translated p73-T86A mutant, which cannot be phosphorylated by p34cdc2-cyclin B. In these experiments, the amount of p73-T86A bound to p21-specific oligonucleotides was unchanged even in the presence of active p34cdc2-cyclin B (Fig. 4B, compare lanes 7, 8, and 9), consistent with the finding that Thr86 is the key p73 phosphorylation site during mitosis. To assess whether p34cdc2-cyclin B-dependent reduction in the DNA binding ability of p73 is reflected by displacement from mitotic chromatin in vivo, we studied p73 localization both in MCF7 (Fig. 5A) and H1299-derived p73α ponasterone A-inducible cells (Fig. 5B). Interphasic cells display a homogenous p73 nuclear staining that colocalizes with DAPI-stained chromatin and is excluded from nucleoli (Fig. 5, A, row a, and B, row a). In the metaphase and anaphase stages, both endogenous and inducible p73 are excluded from condensed chromosomes (Fig. 5, A, row b, and B, rows c and d). The mitotic localization of p73 differs from that of p53, which is associated with centrosomes in mitosis (Ref. 42.Ciciarello M. Mangiacasale R. Casenghi M. Zaira Limongi M. D'Angelo M. Soddu S. Lavia P. Cundari E. J. Biol. Chem. 2001; 276: 19205-19213Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar and Fig. 5C, row a). Differential Behavior of p73 and p53 in Response to Spindle Damage—The association of p53 with centrosomes during normal mitosis is disrupted by nocodazole treatment (Ref. 42.Ciciarello M. Mangiacasale R. Casenghi M. Zaira Limongi M. D'Angelo M. Soddu S. Lavia P. Cundari E. J. Biol. Chem. 2001; 276: 19205-19213Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar and Fig. 5C, row b) and is not re-established in cells that achieve mitotic exit after transient exposure to nocodazole. p53 delocalization and prolonged activation are responsible for the G1 block that follows the removal of the spindle-damaging agent (42.Ciciarello M. Mangiacasale R. Casenghi M. Zaira Limongi M. D'Angelo M. Soddu S. Lavia P. Cundari E. J. Biol. Chem. 2001; 276: 19205-19213Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). We asked whether p73 localization would also be affected by nocodazole. As shown in Fig. 5B (row e), p73 remained excluded from chromatin in nocodazole-arrested prometaphase cells. The exclusion of p73 from chromatin during nocodazole induced mitotic block, together with its reduced DNA binding activity, is expected to result in a reduction of p73 transcriptional activity. Indeed, ponasterone A-induced p73 cells accumulated in mitosis by nocodazole, despite a level of p73 induction similar to that of untreated cells, showed a lower ability to up-regulate p21 (Fig. 6A). When the same experiment was performed in the p53 H1299-derived cell line, the induced p53 retained its intact transcriptional activity, as reflected by p21 protein levels in nocodazole-treated cells (Fig. 6B). The latter result suggests that some of the p53 delocalized from centrosomes is transcriptionally active in nocodazole-arrested mitotic cells. Together these findings suggest that p73 phosphorylation by cdc2-cyclin B and dispersal from mitotic chromatin contribute to down-regulate its transcriptional potential. Conclusions—It is becoming increasingly clear that p53 and p73, despite their similarities in both primary amino acid sequence and DNA binding properties, execute partially overlapping, yet distinct functions (for review see Levrero et al. (2.Levrero M. De Laurenzi V. Costanzo A. Gong J. Wang J.Y. Melino G. J. Cell Sci. 2000; 113: 1661-1670Crossref PubMed Google Scholar) and Strano et al. (3.Strano S. Rossi M. Fontemaggi G. Munarriz E. Soddu S. Sacchi A. Blandino G. FEBS Lett. 2001; 490: 163-170Crossref PubMed Scopus (88) Google Scholar)). Differently from p53, p73 has been shown to be regulated along the cell cycle. p73 is a transcriptional target of transcription factor E2F1 during the S phase of the cell cycle, and TAp73 proteins accumulate in S phase cells (18.Irwin M. Marin M.C. Phillips A.C. Seelan R.S. Smith D.I. Liu W. Flores E.R. Tsai K.Y. Jacks T. Vousden K.H. Kaelin Jr., W.G. Nature. 2000; 407: 645-648Crossref PubMed Scopus (534) Google Scholar). These results suggest that as yet unknown S phase-specific target genes may be regulated by TAp73. In vitro p73 has also been shown to interact physically with and to be regulated negatively by various CDK2- and CDK1-containing complexes, although the physiological role of these interactions in the cell cycle has not yet been clearly established (32.Gaiddon C. Lokshin M. Gross I. Levasseur D. Taya Y. Loeffler J.P. Prives C. J. Biol. Chem. 2003; 278: 27421-27431Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). Here we show that cyclinB-p34cdc2-dependent phosphorylation of both p73α and p73β isoforms is physiologically relevant in G2/M and governs its subcellular localization and DNA binding features. Most importantly, p34cdc2-cyclinB-mediated phosphorylation of p73 on threonine 86 inhibits its transcriptional activity. At nuclear envelope breakdown, the endogenous phosphorylated p73 is excluded from condensed chromosomes and, unlike p53, disperses throughout the mitotic cytoplasm with no specific association with centrosomes. Taken together our results suggest that p73 phosphorylation by cdc2-cyclin B, and the ensuing dispersal from mitotic chromatin contributes to the down-regulation of its transcriptional potential. Instead, transcriptional inactivity of p53 in a normal mitosis is most likely ensured by its transient association with centrosomes. When p53-centrosome association is disrupted by nocodazole, p53 again becomes transcriptionally active, despite the persistence of chromatin condensation. p73 localization and transcriptional activity is not unaffected by nocodazole, suggesting that p73-dependent transcription has to be abrogated during physiological, as well as drug-induced, mitosis. We have also observed that after nocodazole removal, p73 can activate p21 in p53-null cells 3M. Fulco and P. Merlo, unpublished observations. and that up-regulation of p21 peaks several hours after release from arrest, i.e. at a time when most cells have achieved mitotic exit and have reached the following G1 phase. These results suggest that p73 can contribute to sustaining the postmitotic checkpoint activated by transient spindle damage, as already established for p53. Besides this shared function common to both p53 and p73 during G1 re-entry, several features reported here, i.e. p73 inhibitory phosphorylation at the G2-to-M transition by the p34cdc2-cyclin B complex, its lack of association with mitotic microtubules and centrosomes, and the differential behavior in response to nocodazole treatment, strongly suggest that p53 and p73 undergo specific and distinct cellular fates during mitotic division. We thank P. L. Puri, V. Sartorelli, and E. Cundari for critical reading of the manuscript and helpful suggestions.

Referência(s)