Portal de Periódicos da CAPES

Oligomerization Is Required for p53 to be Efficiently Ubiquitinated by MDM2

1999; Elsevier BV; Volume: 274; Issue: 23 Linguagem: Inglês

10.1074/jbc.274.23.16531

1083-351X

Carl G. Maki,

Epigenetics and DNA Methylation

Wild-type p53 is degraded in part through the ubiquitin proteolysis pathway. Recent studies indicate that MDM2 can bind p53 and promote its rapid degradation although the molecular basis for this degradation has not been clarified. This report demonstrates that MDM2 can promote the ubiquitination of wild-type p53 and cancer-derived p53 mutants in transiently transfected cells. Deletion mutants that disrupted the oligomerization domain of p53 displayed low binding affinity for MDM2 and were poor substrates for ubiquitination. However, efficient MDM2 binding and ubiquitination were restored when an oligomerization-deficient p53 mutant was fused to the dimerization domain from another protein. These results indicate that oligomerization is required for p53 to efficiently bind and be ubiquitinated by MDM2. p53 ubiquitination was inhibited in cells exposed to UV radiation, and this inhibition coincided with a decrease in MDM2 protein levels and p53·MDM2 complex formation. In contrast, p53 dimerization was unaffected following UV treatment. These results suggest that UV radiation may stabilize p53 by blocking the ubiquitination and degradation of p53 mediated by MDM2. Wild-type p53 is degraded in part through the ubiquitin proteolysis pathway. Recent studies indicate that MDM2 can bind p53 and promote its rapid degradation although the molecular basis for this degradation has not been clarified. This report demonstrates that MDM2 can promote the ubiquitination of wild-type p53 and cancer-derived p53 mutants in transiently transfected cells. Deletion mutants that disrupted the oligomerization domain of p53 displayed low binding affinity for MDM2 and were poor substrates for ubiquitination. However, efficient MDM2 binding and ubiquitination were restored when an oligomerization-deficient p53 mutant was fused to the dimerization domain from another protein. These results indicate that oligomerization is required for p53 to efficiently bind and be ubiquitinated by MDM2. p53 ubiquitination was inhibited in cells exposed to UV radiation, and this inhibition coincided with a decrease in MDM2 protein levels and p53·MDM2 complex formation. In contrast, p53 dimerization was unaffected following UV treatment. These results suggest that UV radiation may stabilize p53 by blocking the ubiquitination and degradation of p53 mediated by MDM2. Inactivation of p53 is considered an important step in the development of many human cancers. It is therefore important to determine how p53 levels are normally regulated and how this regulation is altered in cancer. The activity most associated with the tumor suppressor function of p53 is its ability to bind DNA and activate gene transcription (1Funk W. Pak D.T. Karas R.H. Wright W.E. Shay J.W. Mol. Cell. Biol. 1992; 12: 2866-2871Crossref PubMed Scopus (677) Google Scholar, 2Kern S.E. Pietenpol J.A. Thiagalingam S. Seymour A. Kinzler K.W. Vogelstein B. Science. 1992; 256: 827-830Crossref PubMed Scopus (889) Google Scholar, 3Pietenpol J.A. Tokino T. Thiagalingam S. El-Deiry W.S. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1988-2002Crossref Scopus (382) Google Scholar). This activity increases in response to DNA damage because of stabilization of the p53 protein (4Maltzman W. Czyzyk L. Mol. Cell. Biol. 1984; 4: 1689-1694Crossref PubMed Scopus (815) Google Scholar, 5Maki C.G. Howley P.M. Mol. Cell. Biol. 1997; 17: 355-363Crossref PubMed Scopus (299) Google Scholar). The effect of increasing p53 is to stop cell proliferation, either through a G1-phase cell cycle arrest or apoptotic cell death (6Lane D.P. Nature. 1992; 358: 15-16Crossref PubMed Scopus (4487) Google Scholar, 7Levine A.J. Cell. 1997; 88: 323-331Abstract Full Text Full Text PDF PubMed Scopus (6727) Google Scholar). Current models suggest that p53 is part of a DNA damage checkpoint pathway that monitors the genome integrity and protects cells from accumulating genetic damage. p53 carries out this function by temporarily halting cell proliferation to allow DNA repair, or, in the case of irreparable damage, by inducing cell death through apoptosis. Two proteolytic pathways have been implicated in p53 degradation: the calpain proteolytic pathway (8Kubbutat M.H. Vousden K.H. Mol. Cell. Biol. 1997; 17: 460-468Crossref PubMed Scopus (275) Google Scholar, 9Pariat M. Carillo S. Molinari M. Salvat C. Debussche L. Bracco L. Milner J. Piechaczyk M. Mol. Cell. Biol. 1997; 17: 2806-2815Crossref PubMed Scopus (147) Google Scholar) and the ubiquitin proteolytic pathway (10Chowdary D.R. Bermody J.J. Jha K.K. Ozer H.L. Mol. Cell. Biol. 1994; 14: 1997-2003Crossref PubMed Scopus (266) Google Scholar, 11Maki C.G. Huibregtse J.M. Howley P.M. Cancer Res. 1996; 56: 2649-2654PubMed Google Scholar). The relative contribution of these two pathways to p53 turnover has not been clarified. The hallmark of the ubiquitin pathway is the covalent attachment of ubiquitin (an 8-kDa protein) to a substrate, which "marks" that substrate for degradation (reviewed in Ref. 12Hochstrasser M. Annu. Rev. Genet. 1996; 30: 405-439Crossref PubMed Scopus (1458) Google Scholar). Ubiquitin is first sequentially transferred through a series of ubiquitin system enzymes, designated E1, E2, and E3. The E3 enzyme then transfers the ubiquitin molecule to one or more lysine residues in the substrate (13Scheffner M. Nuber U. Huibregtse J.M. Nature. 1995; 373: 81-83Crossref PubMed Scopus (743) Google Scholar). Multiple ubiquitins are attached to one another to form ubiquitin chains, and the multi-ubiquitinated substrate is degraded by the 26 S proteasome. Proof that p53 is a ubiquitin-system target came with the demonstration of wild-type p53:ubiquitin conjugates in vivo (11Maki C.G. Huibregtse J.M. Howley P.M. Cancer Res. 1996; 56: 2649-2654PubMed Google Scholar). Factors that regulate and/or participate in p53 ubiquitination have not been fully characterized. Increasing evidence suggests that MDM2 can regulate p53 stability. MDM2 forms an autoregulatory feedback loop in which p53 activates MDM2 transcription, and increased levels of MDM2 protein then bind p53 and inhibit its transcriptional activity (14Momand J. Zambetti G.P. Olson D. George D. Levine A.J. Cell. 1992; 69: 1237-1245Abstract Full Text PDF PubMed Scopus (2787) Google Scholar, 15Oliner J.D. Kinzler K.W. Meltzer P.S. George D. Vogelstein B. Nature. 1992; 358: 80-83Crossref PubMed Scopus (1796) Google Scholar). DNA-damaging agents that stabilize p53 can promote p53 phosphorylation at serines 15 and 37 (16Shieh S.Y. Ikeda M. Taya Y. Prives C. Cell. 1997; 91: 325-334Abstract Full Text Full Text PDF PubMed Scopus (1752) Google Scholar). Interestingly, MDM2 was unable to inhibit the activity of p53 when p53 was phosphorylated at these sites. This could result if MDM2 is unable to bind p53 phosphorylated at Ser-15 and Ser-37. Other studies demonstrated that MDM2 can promote p53 degradation in transient transfection assays (17Haupt Y. Maya R. Kazaz A. Oren M. Nature. 1997; 387: 296-299Crossref PubMed Scopus (3705) Google Scholar, 18Kubbutat M.H.G. Jones S.N. Vousden K.H. Nature. 1997; 387: 299-303Crossref PubMed Scopus (2839) Google Scholar). Though the basis of this degradation was not clarified, Honda et al. (19Honda R. Tanaka H. Yasuda H. FEBS Lett. 1997; 420: 25-27Crossref PubMed Scopus (1597) Google Scholar) reported that MDM2 could promote p53 ubiquitination in vitro in the presence of E1 and E2 ubiquitin system enzymes. Further, MDM2 formed a thioester bond with ubiquitin, a characteristic of E1, E2, and E3 enzymes. Mutations in MDM2 that abrogated thioester bond formation also abrogated p53 ubiquitination, suggesting that MDM2 functions as an E3 enzyme in p53 ubiquitination. Taken together, these studies provide a provocative model for the regulation of p53 stability. According to this model, MDM2 binds p53 under normal conditions and promotes its ubiquitination and subsequent degradation by the proteasome. In response to DNA damage, p53 is phosphorylated at Ser-15 and Ser-37, preventing MDM2 binding and thus stabilizing p53. The stabilized p53 can then promote transcription of its downstream target genes, resulting in cell cycle arrest or apoptosis. This report demonstrates the presence of p53:ubiquitin conjugates in cells transiently transfected with p53 expression DNAs. High levels of p53 ubiquitination were observed when cells were cotransfected with MDM2, providing evidence that MDM2 can promote p53 ubiquitinationin vivo. Deletion mutants that disrupted the oligomerization domain of p53 had low MDM2 binding affinity and were poor substrates for ubiquitination. However, efficient MDM2 binding and ubiquitination were restored when oligomerization-deficient p53 mutants were fused to a heterologous dimerization domain. Finally, p53 ubiquitination was inhibited in UV-irradiated cells, and this coincided with decreased MDM2 protein levels and p53·MDM2 complex formation. UV radiation may stabilize p53 by inhibiting the ubiquitination of p53 mediated by MDM2. Expression DNAs encoding wild-type p53, mutant p53s, and Myc-tagged ubiquitin were from Peter Howley (Harvard Medical School). HA-tagged p53 expression DNA was from Christine Jost (Dana Farber Cancer Institute). DNAs encoding p53-(1–363) and -(1–333) (20Pellegata N.S. Cajot J.-F. Stanbridge E.J. Oncogene. 1995; 11: 337-349PubMed Google Scholar) were from Eric Stanbridge (University of California-Irvine). DNAs encoding p53-(1–353) and p53–333CC (3Pietenpol J.A. Tokino T. Thiagalingam S. El-Deiry W.S. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1988-2002Crossref Scopus (382) Google Scholar) were from Jennifer Pietenpol (Vanderbilt University). Human MDM2 expression DNA was from Steve Grossman (Dana Farber Cancer Institute). Human osteosarcoma cell lines Saos-2 (p53-null) and U2OS (wild-type p53) were grown in Dulbecco's modified Eagle's media (DMEM) 1The abbreviations used are: DMEM, Dulbecco's modified Eagle's medium; RIPA, radioimmune precipitation buffer. supplemented with 10% fetal bovine serum and 100 μg/ml each of penicillin and streptomycin. Transfections were done using the calcium-phosphate method (5Maki C.G. Howley P.M. Mol. Cell. Biol. 1997; 17: 355-363Crossref PubMed Scopus (299) Google Scholar) in 60- or 100-mm dishes with cells approximately 80% confluent. For 60-mm dishes, 4–5 μg of each p53 plasmid was transfected with an equal quantity of MDM2 plasmid, and the final DNA amount was adjusted to 12–15 micrograms with herring sperm DNA. For 100-mm dishes, 10 μg of p53 plasmid was transfected with 10 μg of MDM2, and the final DNA amount was adjusted to 25 micrograms. Saos-2 cells were washed twice with DMEM minus serum 6–8 h after addition of the DNA precipitate and then refed with DMEM plus 10% fetal bovine serum. Cell extracts were prepared 30–36 h later. U2OS cells were washed and refed 16 h after addition of the DNA precipitate. Where indicated, U2OS cells were UV irradiated (5Maki C.G. Howley P.M. Mol. Cell. Biol. 1997; 17: 355-363Crossref PubMed Scopus (299) Google Scholar) and extracts prepared 3–7 h later. For detection of p53:ubiquitin conjugates, cells were rinsed with phosphate-buffered saline and scraped into 800-μl radioimmunoprecipitation buffer (RIPA) (2 mm Tris, pH 7.5, 5 mm EDTA, 150 mm NaCl, 1.0% Nonidet P-40, 1.0% deoxycholate, 0.025% SDS, 1 mm phenylmethylsulfonyl fluoride). Cells were then sonicated for 10 pulses at setting 5, 50% output, with a Branson 450 sonifier, and spun at 15,000 ×g for 15 min. One-twentieth (40 μl) of the resulting supernatant was examined directly by immunoblotting. p53 was immunoprecipitated from the remaining supernatant using the p53 antibody Ab-6 (Oncogene Science). For co-immunoprecipitation experiments, cells were rinsed with phosphate-buffered saline, scraped into 750 μl of lysis buffer (50 mm Tris, pH 7.5, 5 mm EDTA, 150 mm NaCl, 0.5% Nonidet P-40, 1 mm phenylmethylsulfonyl fluoride), and placed on ice for 30 min. Cells were then spun at 15,000 × g for 15 min, and the supernatant was immunoprecipitated with either the p53 antibody 1801 (Oncogene Science) or the anti-HA polyclonal antibody HA.11 (Babco). Immunoprecipitates from RIPA or lysis buffer were resolved by SDS-polyacrylamide gel electrophoresis and transferred to an Immobilon-P membrane. For detection of p53:ubiquitin conjugates, the membrane was autoclaved in water for 15 min prior to blocking with milk and probing with the p53 antibody Ab-6. MDM2 was detected in p53 immunoprecipitates using the MDM2 polyclonal antibody N-20 (Santa Cruz Biotechnology). MDM2 can promote p53 degradation in transfected cells (17Haupt Y. Maya R. Kazaz A. Oren M. Nature. 1997; 387: 296-299Crossref PubMed Scopus (3705) Google Scholar, 18Kubbutat M.H.G. Jones S.N. Vousden K.H. Nature. 1997; 387: 299-303Crossref PubMed Scopus (2839) Google Scholar). The purpose of this study was to determine the basis for this degradation and the amino acid determinants in p53 required for this effect. Saos-2 cells (p53 null) were transfected with expression DNAs encoding p53, MDM2, and myc-tagged ubiquitin. p53 was then immunoprecipitated and examined by immunoblotting with the p53 antibody Ab-6 (Fig. 1 A). A ladder of protein bands with molecular weights larger than p53 were recognized when p53 was coexpressed with MDM2. These bands ranged in size from ∼63 to 90 kDa, consistent with the addition of one to five ubiquitin moieties to p53. When Myc-tagged ubiquitin was coexpressed with p53 and MDM2, these bands shifted to a slightly larger molecular weight, consistent with the additional Myc-tag. To confirm these bands as p53:ubiquitin conjugates, cells were transfected with epitope-tagged p53 (HA-p53) and Myc-tagged ubiquitin. p53 was then immunoprecipitated with an anti-HA antibody and examined by immunoblotting for the myc epitope. The myc antibody recognized a ladder of bands in the immunoprecipitates, and these same bands were recognized by the p53 antibody Ab-6 (Fig. 1 B). These results prove that the ladder of bands is ubiquitinated p53 and thus prove that MDM2 can promote p53 ubiquitination. Cancer-derived p53 point mutants were also tested for MDM2-mediated ubiquitination (Fig. 2). The p53 mutants V143A and R248W were ubiquitinated to approximately equal levels when coexpressed with MDM2, while the R273H mutant was ubiquitinated to a much lesser extent. R273H ubiquitination could be seen on long gel exposures, but the extent of R273H ubiquitination was consistently lower than that of either V143A or R248W. This suggests that specific p53 mutations can affect the susceptibility of p53 to MDM2-mediated ubiquitination. The relative ability of each p53 mutant to bind MDM2 was assessed to determine whether the lower ubiquitination of R273H was because of a decreased ability to bind MDM2. This involved immunoprecipitation of mutant p53 from transfected cells, followed by immunoblotting for MDM2 (Fig. 3). Approximately equal levels of MDM2 were detected in V143A and R273H immunoprecipitates, while slightly lower MDM2 levels were detected in R248W immunoprecipitates. These results suggest that the decreased ubiquitination of R273H is not because of a decreased ability to bind MDM2.Figure 3MDM2 binding by mutant p53s. Saos-2 cells were transfected with the indicated p53 mutant alone, or cotransfected with MDM2, and cell lysates were prepared. A, p53 was immunoprecipitated with the p53 antibody 1801, and the immunoprecipitates were examined by immunoblotting with the MDM2 antibody N20. B, 50 micrograms of protein extract were examined without prior immunoprecipitation by immunoblotting for p53 and MDM2.View Large Image Figure ViewerDownload (PPT) Deletion analysis was used to identify p53 regions required for MDM2-mediated ubiquitination (Fig. 4). p53 contains four main functional domains (21Ko L.J. Prives C. Genes Dev. 1996; 10: 1054-1072Crossref PubMed Scopus (2288) Google Scholar). Amino acids 1–42 comprise the transcriptional activation domain, and it is this region that binds MDM2. Residues102–292 encode the sequence-specific DNA binding domain. p53 binds DNA as a tetramer, and oligomerization is mediated by residues 324–355. The carboxyl terminus of p53 (residues 367–393) binds DNA nonspecifically and can allosterically regulate DNA binding by the central region. p53 proteins lacking 30 (p53-(1–363)), 40 (p53-(1–353)), or 60 (p53-(1–333)) C-terminal amino acids were tested for MDM2-mediated ubiquitination. p53-(1–363) and p53-(1–353) maintain the oligomerization function of p53, while oligomerization is lost in p53-(1–333) (22Marston N.J. Jenkins J.R. Vousden K.H. Oncogene. 1995; 10: 1709-1715PubMed Google Scholar). All three deletion mutants maintain the major nuclear localization signal (NLSI, residues 316–325) of p53 (23Shaulsky G. Goldfinger N. Ben-Ze'ev A. Rotter V. Mol. Cell. Biol. 1990; 10: 6565-6577Crossref PubMed Scopus (292) Google Scholar). Wild-type p53 and the deletion mutants p53-(1–363) and p53-(1–353) were efficiently ubiquitinated by MDM2 (Fig. 4), indicating that sequences between 353 and 393 are not critical for ubiquitination. In contrast, ubiquitination was drastically reduced in p53-(1–333), indicating that sequences between 333 and 353 are required for efficient ubiquitination. There are three possible explanations for these results. First, deletion of residues 333 to 353 may have removed critical lysine residues that are sites of p53 ubiquitination. There is one lysine (position 351) between residues 333 and 353, and a p53 protein with this site mutated to isoleucine was efficiently ubiquitinated by MDM2 (not shown). Second, deletion of residues 333 to 353 may have abrogated MDM2 binding. Third, p53 oligomerization may be required for ubiquitination by MDM2, and deleting this function in p53-(1–333) may have thus prevented ubiquitination. Coimmunoprecipitation experiments were done to determine the relative ability of the p53 deletion mutants to bind MDM2 (Fig.5). High levels of MDM2 were detected in immunoprecipitates of wild-type p53, p53-(1–363), and p53-(1–353). In contrast, very low levels of MDM2 coimmunoprecipitated with p53-(1–333). These results establish a correlation between MDM2 binding and p53 ubiquitination and suggest that deletion of residues 333 to 353 abrogated p53 ubiquitination by diminishing the ability of p53 to bind MDM2. To determine whether oligomerization affected either p53 ubiquitination or p53·MDM2 binding, the effect of replacing the oligomerization domain of p53 with the dimerization domain from another protein was determined (Fig. 6). p53–333CC contains amino acids 1–333 of p53 fused to the dimerization domain from the yeast transcription factor GCN4 (3Pietenpol J.A. Tokino T. Thiagalingam S. El-Deiry W.S. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1988-2002Crossref Scopus (382) Google Scholar). As shown in Fig. 6 A, wild-type p53 and p53–333CC were efficiently ubiquitinated by MDM2, while p53-(1–333) was not. Further, MDM2 co-immunoprecipitated with wild-type p53 and p53–333CC to comparable levels but did not coimmunoprecipitate with p53-(1–333) (Fig. 6 B). These results indicate that a heterologous dimerization domain can restore MDM2 binding and MDM2-mediated ubiquitination to an oligomerization-deficient p53. This is consistent with the findings of Karen Vousden and co-workers (24Kubbutat M.H.G. Ludwig R.L. Ashcroft M. Vousden K.H. Mol. Cell. Biol. 1998; 18: 5690-5698Crossref PubMed Scopus (171) Google Scholar) who recently reported that p53 proteins with disrupted oligomerization function displayed decreased MDM2 binding affinity and lower susceptibility to MDM2-mediated degradation. We conclude that the ability to oligomerize is required for p53 to efficiently bind MDM2 and be targeted for ubiquitination. DNA-damaging agents stabilize p53 through an unknown mechanism. Given that p53 is targeted for ubiquitination by MDM2, one possibility is that MDM2-mediated ubiquitination of p53 is inhibited in DNA-damaged cells. To test this possibility, cells were transfected with p53 alone, or in combination with MDM2, and subsequently DNA-damaged by exposure to UV light. The effect of UV exposure on p53 ubiquitination, p53·MDM2 complex formation, and p53 dimerization was then assessed (Fig. 7). U2OS cells were used for these experiments because p53 could be ubiquitinated in these cells in the absence of MDM2 coexpression. p53 ubiquitination was inhibited in the transfected cells between 1 and 7 h after UV treatment (Fig.7 A). This decrease in ubiquitination coincided with decreased MDM2 protein levels, and a corresponding decrease in p53·MDM2 complexes (Fig. 7 B). The effect of UV radiation on p53 dimerization was also assessed. Cells were cotransfected with DNAs encoding epitope-tagged p53-(HA-p53) and nontagged p53 and subsequently were exposed to UV radiation. Lysates were immunoprecipitated with an anti-HA antibody, followed by immunoblot analysis for p53. The ability of HA-tagged p53 to immunoprecipitate nontagged p53 was used as a measure of p53 dimerization in UV-irradiated and nonirradiated cells. As shown in Fig. 7 C, HA-tagged p53 could dimerize nontagged p53 in transfected cells, and this dimerization was not diminished after UV exposure. These results indicate that the decrease in p53·MDM2 complex formation following UV treatment did not involve a decrease in the ability of p53 to homodimerize. The activity of wild-type p53 is regulated in large part through changes in protein stability. It is of interest therefore to identify cellular factors that regulate p53 turnover. The most compelling evidence that MDM2 normally regulates p53 levels comes from David Lane and colleagues (25Bottger V. Bottger A. Howard S.F. Picksley S.M. Chene P. Garcia-Echeverria C. Hochkeppel H.K. Lane D.P. Oncogene. 1996; 13: 2141-2147PubMed Google Scholar, 26Bottger A. Bottger V. Sparks A. Liu W.L. Howard S.F. Lane D.P. Curr. Biol. 1997; 7: 860-869Abstract Full Text Full Text PDF PubMed Scopus (353) Google Scholar). Phage-display was used in these studies to identify a 12-amino acid peptide that mimics the MDM2 binding site in p53 and blocks p53·MDM2 binding. When expressed in cells containing low p53 levels, this peptide caused an accumulation of p53, activation of p53-responsive genes, and a cell cycle arrest. In contrast, a mutant peptide unable to block p53·MDM2 binding had no effect. These results provided strong evidence that p53 levels and activity are normally regulated by MDM2 interaction. Other studies demonstrated that MDM2 can promote p53 degradation in transient transfection assays (17Haupt Y. Maya R. Kazaz A. Oren M. Nature. 1997; 387: 296-299Crossref PubMed Scopus (3705) Google Scholar, 18Kubbutat M.H.G. Jones S.N. Vousden K.H. Nature. 1997; 387: 299-303Crossref PubMed Scopus (2839) Google Scholar) although the basis of this degradation was not fully clarified. The current report demonstrates that MDM2 can promote the formation of p53:ubiquitin conjugates in vivo. These results are consistent with those of Honda et al. (19Honda R. Tanaka H. Yasuda H. FEBS Lett. 1997; 420: 25-27Crossref PubMed Scopus (1597) Google Scholar), which suggested that MDM2 functions as an E3 ubiquitin protein ligase in p53 ubiquitination. Cancer-derived p53 mutant proteins varied in their susceptibility to MDM2-mediated ubiquitination, with the V143A and R248W mutants being more prone to ubiquitination than the R273H mutant. This did not appear to result from a lower ability of R273H to bind MDM2, suggesting that factors other than MDM2 binding can affect p53 ubiquitination. One possibility is that these p53 mutants vary in their subcellular localization, and this accounts for their differences in ubiquitination. It has been suggested that p53 ubiquitination by MDM2 occurs in the nucleus (27Freedman D.A. Levine A.J. Mol. Cell. Biol. 1998; 18: 7288-7293Crossref PubMed Scopus (458) Google Scholar), and certain p53 mutants may be retained in the cytoplasm (28Martinez J. Georgoff I. Martinez J. Levine A.J. Genes Dev. 1991; 5: 151-159Crossref PubMed Scopus (493) Google Scholar) where they may not be ubiquitinated. The conformation of p53 might also affect its susceptibility to ubiquitination. It is worth noting that the V143A and R248W p53 mutants were more highly ubiquitinated than either wild-type p53 or R273H. Perhaps R273H adopts a more wild-type conformation than V143A or R248W and so is less susceptible to ubiquitination. Deletion analyses indicated that the C-terminal oligomerization domain of p53 was required for efficient MDM2 binding and ubiquitination. This is perhaps surprising given that the MDM2 binding site in p53 is located in the N terminus of p53 (29Chen J. Marechal V. Levine A.J. Mol. Cell. Biol. 1993; 13: 4107-4114Crossref PubMed Scopus (622) Google Scholar). Further, MDM2 can interact with peptides from the N terminus of p53, indicating that C-terminal sequences are not essential for the interaction to occur. Perhaps the MDM2 binding domain in p53 assumes a more complex tertiary structure when p53 is oligomerized, which enhances its affinity for MDM2. These results are consistent with those of Karen Vousden and her coworkers (24Kubbutat M.H.G. Ludwig R.L. Ashcroft M. Vousden K.H. Mol. Cell. Biol. 1998; 18: 5690-5698Crossref PubMed Scopus (171) Google Scholar) who recently demonstrated that p53 proteins with disrupted oligomerization function displayed decreased binding affinity for MDM2 and were less susceptible to MDM2-mediated degradation. p53 is stabilized in response to various DNA damaging agents, including ionizing radiation (5Maki C.G. Howley P.M. Mol. Cell. Biol. 1997; 17: 355-363Crossref PubMed Scopus (299) Google Scholar), UV radiation (5Maki C.G. Howley P.M. Mol. Cell. Biol. 1997; 17: 355-363Crossref PubMed Scopus (299) Google Scholar), actinomycin D (30Kessis T.D. Slebos R.J. Nelson W.G. Kastan M.B. Plunkett B.S. Han S.M. Lorincz A.T. Hedrick L. Cho K.R. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3988-3992Crossref PubMed Scopus (505) Google Scholar), and cisplatin (31Perego P. Giarola M. Righetti S.C. Supino R. Caserini C. Delia D. Pierotti M.A. Miyashita T. Reed J.C. Zunino F. Cancer Res. 1996; 56: 556-562PubMed Google Scholar). The molecular basis of this stabilization is unknown. Two proteolytic pathways have been implicated in p53 degradation; the calpain proteolysis system (8Kubbutat M.H. Vousden K.H. Mol. Cell. Biol. 1997; 17: 460-468Crossref PubMed Scopus (275) Google Scholar, 9Pariat M. Carillo S. Molinari M. Salvat C. Debussche L. Bracco L. Milner J. Piechaczyk M. Mol. Cell. Biol. 1997; 17: 2806-2815Crossref PubMed Scopus (147) Google Scholar) and the ubiquitin proteolysis system (10Chowdary D.R. Bermody J.J. Jha K.K. Ozer H.L. Mol. Cell. Biol. 1994; 14: 1997-2003Crossref PubMed Scopus (266) Google Scholar, 11Maki C.G. Huibregtse J.M. Howley P.M. Cancer Res. 1996; 56: 2649-2654PubMed Google Scholar). Given that p53 is degraded through these two pathways, one possibility is that DNA damaging agents signal a repression of p53 degradation through the calpain system, the ubiquitin system, or both. This report demonstrates that p53 ubiquitination is inhibited in response to UV radiation. This inhibition coincided with decreased expression of MDM2, and a decrease in p53·MDM2 complexes. This suggests that UV radiation may stabilize p53, at least in part, by inhibiting MDM2 expression and thus blocking the ubiquitination and degradation of p53. DNA-damaging agents can promote p53 phosphorylation, and it has been suggested that this phosphorylation could stabilize p53 by blocking p53·MDM2 complex formation (16Shieh S.Y. Ikeda M. Taya Y. Prives C. Cell. 1997; 91: 325-334Abstract Full Text Full Text PDF PubMed Scopus (1752) Google Scholar). It remains possible that such post-translational modifications might also contribute to the UV-induced loss of p53 ubiquitination observed in the current report. Current models suggest that p53 functions in a pathway that monitors the DNA and protects cells from accumulating genetic damage. p53 carries out this function in response to DNA damage by temporarily halting cell proliferation to allow DNA repair or by eliminating the damaged cell through apoptosis (6Lane D.P. Nature. 1992; 358: 15-16Crossref PubMed Scopus (4487) Google Scholar, 7Levine A.J. Cell. 1997; 88: 323-331Abstract Full Text Full Text PDF PubMed Scopus (6727) Google Scholar). This model predicts that p53 inactivation would be necessary for cells to resume cycling following repair of the damaged DNA. In this regard, it is worth noting the studies of Perry et al. (32Perry M.E. Piette J. Zawadzki J.A. Harvey D. Levine A.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11623-11627Crossref PubMed Scopus (241) Google Scholar) in which the kinetics of MDM2 and p53 induction were examined in UV-irradiated cells. Three important findings emerged: 1) the induction of MDM2 was delayed in UV-treated cells even though p53 levels rose almost immediately; 2) MDM2 induction was delayed for longer periods of time with increasingly larger UV doses; and 3) the time of MDM2 induction was correlated with the recovery of normal rates of DNA synthesis, presumably after DNA repair. Based on these results, it is reasonable to speculate that MDM2 expression may be inhibited in UV-irradiated cells to allow an accumulation of p53 and a subsequent cell cycle arrest and DNA repair. Once DNA repair is completed, the induction of MDM2 may then allow for p53 ubiquitination and degradation and for continued cell cycle progression. It will be of interest to determine whether MDM2 induction following UV radiation is coincident with repair of the UV-induced DNA lesions. This would suggest that MDM2-mediated ubiquitination and degradation of p53 is important for cells to resume cycling after DNA damage.