A ribonucleotide reductase gene is a transcriptional target of ... - Nature

Oncogene (2000) 19, 4283 ± 4289 ã 2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00 www.nature.com/onc

A ribonucleotide reductase gene is a transcriptional target of p53 and p73 Katsunori Nakano1, EÂva BaÂlint1, Margaret Ashcroft1 and Karen H Vousden*,1 1

Regulation of Cell Growth Laboratory, NCI-FCRDC, Frederick, Maryland, MD 21702-1201, USA

Many p53-inducible genes have been identi®ed that might play a role in mediating the various downstream activities of p53. We have identi®ed a close relative of ribonucleotide reductase, recently named p53R2, as a p53-inducible gene, and show that this gene is activated by several stress signals that activate a p53 response, including DNA damaging agents and p14ARF. p53R2 expression was induced by p53 mutants that are defective for the activation of apoptosis, but retain cell cycle arrest function, although no induction of p53R2 was seen in response to p21WAF1/CIP1-mediated cell cycle arrest. Several isoforms of the p53 family member p73 were also shown to induce p53R2 expression. Transient ectopic expression of either wild type p53R2 or p53R2 targeted to the nucleus, did not signi®cantly alter cell cycle progression in unstressed cells. The identi®cation of this gene as a p53 target supports a direct role for p53 in DNA repair, in addition to inhibition of growth of damaged cells. Oncogene (2000) 19, 4283 ± 4289. Keywords: p53; p73; DNA damage; ribonucleotide reductase; p53R2 Introduction The p53 tumor suppressor gene plays an important role in preventing cancer development, and most tumor cells show either mutation in p53 or defects in the pathways responsible for the activation of p53 (Hollstein et al., 1999). p53 has been shown to participate in several cellular responses that could contribute to the suppression of tumor development, including DNA repair, senescence, cell cycle arrest and apoptosis (Ashcroft and Vousden, 1999). Activation of p53 in response to stress signals such as DNA damage or oncogene activation is thought to prevent the replication of abnormal cells and either allow their repair or target their elimination. In this way p53 can prevent the growth of cells that show uncontrolled proliferation or harbor potentially oncogenic mutations. The importance of p53 is emphasized by the dramatically enhanced cancer incidence in mice deleted of p53 (Attardi and Jacks, 1999), or humans with a germline mutation in one p53 allele (Evans and Lozano, 1997). Understanding how p53 mediates its downstream eects has been of considerable interest. It is clear that one activity of p53 that plays an important role is the ability to function as a sequence speci®c DNA binding

*Correspondence: KH Vousden, Regulation of Cell Growth Laboratory, NCI-FCRDC, Building 560, Room 22-96, West 7th Street, Frederick, MD 21702-1201, USA Received 3 April 2000; revised 15 May 2000; accepted 4 July 2000

protein and transcription factor (Bates and Vousden, 1996; El-Deiry, 1998). Although several studies have suggested the existence of transcriptionally independent activities of p53, particularly in mediating the apoptotic response, in many systems activation of gene expression by p53 seems to be essential to mediate the p53 response (Bates and Vousden, 1999). Many p53 target genes have now been described, and in some cases it is possible to attribute p53-mediated responses to the activation of a known target gene. For example, activation of the cyclin-dependent kinase inhibitor p21WAF1/CIP1 (El-Deiry et al., 1993) leads to the G1 and G2 cell cycle arrest characteristic of p53 activation, but does not clearly participate in activation of the apoptotic response (Bates et al., 1998; Niculescu III et al., 1998). Deletion of p21WAF1/CIP1 substantially reduced p53-mediated cell cycle arrest (Waldman et al., 1995) (Bunz et al., 1998), but does not completely abolish this response, indicating that activation of other genes, such as 14-3-3 sigma (Hermeking et al., 1997), also contribute to the inhibition of cell cycle progression. Several excellent candidate genes for mediators of p53dependent cell death have also been described, including Bax, Fas and DR5 (El-Deiry, 1998), although no single gene has been identi®ed so far that could be considered the principal mediator of p53-induced apoptosis (Bates and Vousden, 1999). It therefore seems likely that each p53-induced response re¯ects the combined eects of activation of several target genes. We examined gene expression in response to p53 expression, and identi®ed a novel p53 inducible gene encoding a protein closely related to ribonucleotide reductase. Ribonucleotide reductase is necessary for the production of deoxyribonucleotides and is important for both DNA replication and repair (Jordan and Reichard, 1998), and the identi®cation of this new p53 target provides a direct mechanism through which p53 could contribute to DNA repair. Results and discussion In order to identify novel p53 target genes that might mediate one or more p53 responses we generated a p53 null Saos-2 cell line inducible for expression of wild type p53, similar to the line described previously (Stott et al., 1998) (Figure 1a,b). Dierences in gene expression in this line in the presence or absence of p53 were determined by analysing microarrays (Incyte) hybridized with cDNA derived from uninduced and induced cells. One of the tags identi®ed was used as a probe in Northern blotting and detected an approximately 5.0 kb transcript whose expression was clearly enhanced following activation of wild type p53 in the inducible system (data not shown). 5' RACE was then carried out to isolate the full-length c-DNA. A novel

Activation of a ribonucleotide reductase by p53 K Nakano et al

4284

Figure 2 Expression of a newly identi®ed p53-inducible gene in response to p53. (a) Amino acid sequence of the protein expressed by the p53-inducible gene. This gene has recently been named p53R2 by Tanaka et al. (Tanaka et al., 2000). (b) Northern blot showing expression of the identi®ed p53-inducible gene (p53R2) in Saos-2 cells (Saos-2p53) without treatment (7) and after induction with doxycycline (Dox) at 2.5 mg/ml, and RKO cells, RKO/E6 cells, MCF-7 cells and U2OS cells treated without or with actinomycin-D (5 nM) for 24 h. GAPDH expression was examined as a loading control

Figure 1 Saos-2 cells with doxycycline inducible expression of wild type p53. (a) Flow cytometric analysis of Saos-2 cells (Saos2p53) without treatment and after induction with doxycycline treatment (2.5 mg/ml) for 24 h, showing an increase of apoptotic cells with a sub-G1 DNA content, represented as a percentage (%) of the total population. (b) Western blot analysis of p53 expression in the inducible cells without treatment (7) and after induction with doxycycline (Dox) at 2.5 mg/ml for 24 h. p53 levels were assessed using monoclonal antibody DO-1

4.9 kb cDNA was obtained that contained a single open reading frame encoding a putative protein of 351 amino acids (Figure 2a). This protein showed a high degree of similarity to the small subunit of human ribonucleotide reductase, and very recently the same gene was independently isolated as p53 inducible by dierential display, and termed p53R2 (Tanaka et al., 2000). To determine if activation of endogenous p53 would also regulate expression of this gene, we examined three wild type p53 expressing tumor cell lines (RKO, U2OS and MCF-7). Treatment of each of these lines with 5 nM actinomycin-D, which stabilizes and activates the endogenous p53 (Ashcroft et al., 2000), clearly induced expression of p53R2 (Figure 2b). The p53 dependence of this actinomycin-D response was con®rmed by showing that expression of E6, which inactivates p53, resulted in loss of induction of this transcript in RKO cells (Figure 2b). Interestingly, the cell lines with endogenous p53 (RKO, MCF-7 and U2OS) all showed higher basal levels of p53R2 Oncogene

expression than the p53 null Saos-2 cells, and correspondingly higher levels of p53R2 expression following p53 induction was also seen in these cells (Figure 2b). p53 has been shown to activate expression of a large number of genes and it has been suggested that dierent p53 targets might contribute to dierent activities of p53. Analysis of mutant p53 proteins that retain cell cycle arrest activity but fail to eciently induce cell death suggests that p53 target genes that contribute to cell cycle arrest can be activated dierentially from those that induce apoptosis (Friedlander et al., 1996; Ludwig et al., 1996). Analysis of several such p53 mutants indicated that while they retained wild type activity in the activation of transcription from the p21WAF1/CIP1 promoter, they showed a severe defect in the activation of other p53 responsive promoters such as Bax (Ryan and Vousden, 1998). To determine whether p53R2 was also dierentially regulated in this way, we examined expression of p53R2 in cells inducibly expressing wild type p53, p53 175P, a mutant that retains cell cycle arrest activity but has severely diminished apoptotic function, and p53 175H, a mutant that fails to mediate either response (Figure 3). Each of the p53 proteins was expressed to similar levels following induction with doxycycline (Figure 3a). A Saos-2 line stably expressing p53 143A was also examined. The p53 143A mutant is temperature sensitive, showing loss of all activity at the non-permissive temperature (378C), but acquiring the ability to activate cell cycle arrest, but not apoptosis, at the permissive temperature (308C)


Figure 3 Activation of p53R2 expression by p53 mutants. (a) Western blot analysis showing expression of p53 in Saos-2 cells following induction of wild type p53 (Saos-2p53), p53 175P, p53 175H without treatment (7) and after treatment with doxycycline (Dox) at 2.5 mg/ml for 24 h. Actin expression was examined as a loading control. (a) Northern blot analysis showing expression of p53R2 in Saos-2 cells following induction of wild type p53 (Saos2p53), p53 175P, p53 175H without treatment (7) and after treatment with doxycycline (Dox) at 2.5 mg/ml for 24 h. GAPDH expression was examined as a loading control. (b) Northern blot analysis showing expression of p53R2 in Saos-2 cells following induction of wild type p53 (Saos-2p53), or p21WAF1/CIP1 without treatment (7) and after treatment with doxycycline (Dox) at 2.5 mg/ml, and in Saos-2 cells expressing the temperature sensitive p53 143A grown at the non-permissive (378C) and permissive (308C) temperatures. GAPDH expression was examined as a loading control

(Friedlander et al., 1996). p53R2 expression was found to be activated in Saos-2 cells following induction of wild type p53 and p53 175P, and in cells expressing p53 143A that were incubated at the permissive temperature (Figure 3b,c). No activation of p53R2 expression was seen in response to p53 175H expression, or in the p53 143A cells grown at the non-permissive temperature. These results raise the possibility that activation of p53R2 is related to cell cycle arrest, since wild type p53, p53 175P and p53 143A at the permissive temperature activate p21WAF1/CIP1 and block cells in the G1 and G2 phases of the cell cycle. We therefore examined the expression of p53R2 in Saos-2 cells

inducibly expressing p21WAF1/CIP1, that show G1 and G2 cell cycle arrest indistinguishable from that induced by p53 (Bates et al., 1998). Activation of p21WAF1/CIP1 in these cells did not result in the elevation of p53R2 expression (Figure 3c), indicating that arrest of cell cycle progression per se does not lead to activation of p53R2. Dierential activation of p53 target gene expression can also be detected depending on the stress signal activating p53. Treatment of cells with actinomycin-D, for example, leads to elevated expression of two well established p53 targets, p21WAF1/CIP1 and MDM2. In contrast, treatment of the same cells with the topoisomerase inhibitor camptothecin (CPT), results in stabilization of p53 and activation of p21WAF1/CIP1, but no induction of MDM2 expression (Ashcroft et al., 2000). Given the potential importance of p53R2 in mediating the DNA repair function of p53, we examined the activation of p53R2 expression in response to various dierent stress signals (Figure 4). Although each treatment induced some increase in the expression of p53R2, actinomycin-D led to a higher elevation of p53R2 expression than CPT or UV treatment at this time point. This dierence was not so clearly seen in the induction of p21WAF1/CIP1 expression, which was equivalent following actinomycin-D and UV treatment and only slightly lower in CPT treated cells (Figure 4a). Another way to stabilize p53 is through expression of p14ARF, a protein that binds MDM2 and prevents degradation (Sherr, 1998). Induction of p14ARF in a cell line expressing wild type p53 (Stott et al., 1998) leads to stabilization of p53 and transcriptional activation of both p21WAF1/CIP1 and p53R2 (Figure 4b,c). Taken together, these results show that diverse stress signals can activate p53R2 expression. Our observation that the level of activation of p53R2 expression depends on the signal inducing p53 might re¯ect a dierence in the requirement for p53R2 in mediating the response to dierent forms of stress. p53 has recently been shown to belong to a family of proteins including p63 and p73, although unlike p53 neither p63 nor p73 show clear activity as tumor suppressor proteins (Lohrum and Vousden, 1999). However, deletion of p63 or p73 in mice has shown that both proteins are important during normal development (Mills et al., 1999; Yang et al., 1999, 2000). To determine whether p53R2 might be induced by activation of p73 we analysed a series of Saos-2 cell lines inducible for expression of several isoforms of p73 (Figure 5a). Induction of p73a, b and g expression led to the activation of p21WAF1/CIP1 expression (Figure 5a) as previously reported (Zhu et al., 1999). Northern blot analysis of these lines showed that p53R2 expression was also activated by each of the p73 isoforms (Figure 5b), although the levels of activation were somewhat lower than that seen following activation of p53. By comparison, p73a and b activated p21WAF1/CIP1 mRNA as eciently as p53 in this system, suggesting that the p73 proteins are slightly less ecient than p53 in the activation of p53R2. p73a has been shown to be activated by some forms of DNA damage through the activity of c-Abl (Agami et al., 1999; Gong et al., 1999; Yuan et al., 1999) and these results open the possibility that p53R2 might also function to mediate the cellular response to p73.

4285

Oncogene


4286

Figure 5 Activation of p53R2 expression by p73 isoforms. (a) Western blot analysis showing inducible expression of p73a, b and g in Saos-2 cells without treatment (7) and after treatment with doxycycline (Dox) at 2 mg/ml. Induction of endogenous p21WAF1/CIP1 in response to p73a, b and g is also shown. Actin expression was examined as a loading control. (b) Northern blot analysis showing expression of p53R2 and p21WAF1/CIP1 in response to p73a, b and g induction without treatment (7) and after treatment with doxycycline (Dox) at 2.5 mg/ml. GAPDH expression was examined as a loading control

Figure 4 Activation of p53R2 expression in wild type p53 expressing cells exposed to p53-activating signals. (a) Northern blot analysis showing expression of p53R2 and p21WAF1/CIP1 in MCF-7 cells treated with CPT (2 mM), UV (50 J/m72) or actinomycin-D (5 nM) for 12 h. GAPDH expression was examined as a loading control. (b) Expression of p53 and p14ARF in NARF2 cells (U2OS cells with inducible p14ARF) following activation of p14ARF by treatment with 1 mM IPTG for the indicated times. (c) Northern blot analysis of cells treated as in (b) showing p21WAF1/CIP1 and p53R2 expression in response to p14ARF. GAPDH expression was examined as a loading control

To examine the eects of p53R2 expression on cell growth, we cloned the full-length p53R2 cDNA with an N-terminal Flag tag to facilitate detection. Transient expression of this construct in p53 null Saos-2 cells, followed by ¯ow cytometric analysis of Oncogene

transfected cells (selected by sorting for co-transfected CD20) showed only a very modest eect of p53R2 expression on the cell cycle pro®le (Figure 6a), with a small enhancement of cells in the S and G2 phases of the cell cycle compared to vector-only transfected cells. These results also indicated that p53R2 does not induce the apoptotic response, in agreement with a recent study (Tanaka et al., 2000). Our immunocytochemical analysis of the transfected cells showed that the p53R2 protein was expressed in both the cytoplasm and nucleus (Figure 6b). Tanaka et al. showed that p53R2 contributes to the activation of a G2 cell cycle arrest in response to DNA damage (Tanaka et al., 2000). This function of p53R2 was correlated with the translocation of p53R2 from the cytoplasm to the nucleus. We therefore generated a p53R2 protein containing a nuclear localization signal derived from SV40 LT, named p53R2nls. Although this protein was almost exclusively localized to the nucleus (Figure 6b), cells transfected with p53R2nls showed only the slight


4287

Figure 6 Expression of p53R2 in Saos-2 cells. (a) Flow cytometric analysis showing cell cycle pro®le of Saos-2 cells transfected with (10 mg) vector control, ¯ag-p53R2 or ¯ag-p53R2nls expression constructs. (b) Cellular localization of Saos-2 cells transfected with (10 mg) vector control, ¯ag-p53R2 or ¯ag-p53R2nls expression constructs. Localization of protein was detected using an anti¯ag antibody and visualized using a FITC-conjugated secondary antibody. Cell nuclei were stained with DAPI

elevation of cells in S and G2 phases as observed following expression of wild type p53R2 (Figure 6a). The activation of a clear G2 arrest in response to DNA damage would therefore appear to require signals in addition to nuclear expression of p53R2.

We have identi®ed a new p53 inducible gene encoding a ribonucleotide reductase. This gene was also identi®ed recently by Tanaka et al. (Tanaka et al., 2000), and shown to play a role in the repair of DNA damage. We show that p53R2 is activated in response to stress signals, Oncogene


4288

such as actinomycin-D treatment and p14ARF expression, that may activate p53 without causing DNA damage, indicating that p53R2 may have functions additional to those that contribute to DNA repair. The activation of p53R2 by p73a, b and g also suggests a role for this protein in mediating the p53-independent DNA damage response pathway activated through c-Abl (Agami et al., 1999; Gong et al., 1999; Yuan et al., 1999). It will be of interest to determine whether activation of p53R2 is necessary for the developmental functions of p73 that were recently revealed by the generation of p73 de®cient mice (Yang et al., 2000).

Materials and methods Plasmids N-terminal hemagglutinin (HA)-tagged human p73a, p73b or p73g full-length cDNAs were released from the plasmids pcDNA-HAp73a, b and g, (obtained from G Melino, University of Rome, Rome, Italy) by digestion with HindIII, followed by blunt-end ®lling using Klenow, and digestion with XbaI. The fragments were ligated into the tet-responsive expression plasmid, pTRE (Clontech Laboratories, Inc.) at the EcoRI, (which was blunt-end ®lled using Klenow), and XbaI sites. Full-length human cDNA of p53 was subcloned into the pTRE plasmid digested at EcoRI and BamHI. The expression construct for 14ARF was described previously (Stott et al., 1998). The full-length Flag-tagged p53R2 was obtained by RT ± PCR using sense primer, 5'TATGGATCCCGCACCATGGACTACAAGGA CGACGATGACAAGGGCGACCCGGAAAGGCC'3 and anti-sense primer, 5'-TATGGATCCTTGGGAATATGCAGGGCGAG'3, then subcloned into pCMVneoBam to create pCMVneoBam-¯ag53R2. pCMVneo Bam-¯ag53R2nls containing nuclear localization signal derived from SV40 LT at its C-terminus was also constructed by PCR using sense primer (as above) and anti-sense primer, TATGGATCCTCAGACCTTTCGCTTCTTCTTTGGAAAATCTGCATCCAAGGTGAAGACG and the pCMVneoBam-¯ag53R2 construct as a template. All PCR-derived constructs were con®rmed by sequencing. Cell lines, transfections and flow cytometry MCF-7, RKO, RKO/E6 (stably expressing E6), U2OS, Saos2, Saos2-143A (expressing the temperature-sensitive p53 mutant p53 143A), NARF2 (U2OS cells with inducible p14ARF) and Saos-2 cells inducible for p53 175P, p53 175H and p21WAF1/CIP1 were described previously (Stott et al., 1998; Bates et al., 1998; Ashcroft et al., 2000; Ryan et al., 2000). Saos-2 cells inducible for wild type p53 and various p73 isoforms were generated by transfection of the Saos-tet primary cell line (provided by K Ryan (Bates et al., 1998)) with 10 mg of pTREp53, pTREp73a, pTREp73b, or pTREp73g, and 2 mg pSV2Hyg for stable selection of transfected cells. Cells were selected in hygromycin at 100 mg/ml for approximately 2 weeks. Independent clonal lines inducible for p53, p73a, p73b, and p73g were analysed for inducible protein expression by Western blotting. All cell lines were maintained in DMEM supplemented with 10% FCS and grown at 378C in an atmosphere of 10% CO2 in air. The temperature sensitive p53 was activated by moving cells to the permissive temperature (308C). To induce protein expression the tetracycline inducible cell lines were treated with the tetracycline analog doxycycline at 2.0 ± 2.5 mg/ml as indicated. p14ARF expression was induced in NARF2 cells by treatment with 1 mM IPTG. For induction of endogenous p53 protein in MCF-7, RKO, RKO/E6 (stably expressing E6), and U2OS cells, cells were treated with either actinomycin-D (5 nM), camptothecin (CPT, 2 mM), or UV Oncogene

(50 J/m72) and harvested at the times indicated. Cells were transfected using calcium phosphate coprecipitation and harvested for either ¯ow cytometry, or immuno-staining at the indicated times. For cell cycle analysis Saos-2 cells were transfected with 10 mg of total test plasmid and 1 mg pCMV CD20. Cells were harvested at 36 h after transfection, stained for CD20 expression and analysed by ¯ow cytometry (FACScalibur, Becton Dickinson), as previously described (Rowan et al., 1996; Zhu et al., 1993). Array screening and cloning of p53R2 Total RNA was collected from the p53 inducible Saos-2 cell line 24 h after treatment with or without 2.5 mg/ml doxycycline. The total RNA was sent to Incyte who isolated mRNA, generated cDNA and screened the human UniGEM V Microarray, and returned results showing fold changes in gene expression following p53 induction, using GEMTools software. One of the EST tags (GenBank accession number; AA768847) was identi®ed and subjected for 5'RACE (SMART RACE, Clontech) according to the manufacturer's instructions to obtain the p53R2 cDNA. RNA and protein analysis Total RNA was isolated from cells grown to 70 ± 80% con¯uency on a 15-cm diameter dish using TRIzol reagent (GIBCO ± BRL), and 20 mg of total RNA per lane was examined by Northern blot analysis. 32P-dCTP-labeled fulllength human p53R2 or p21WAF1/CIP1 cDNA probes were prepared using the Megaprime DNA labeling system according to the manufacturer's instructions (Amersham), and used for hybridization. The GAPDH cDNA probe (G3PDH) was purchased from Clontech. Western blot analysis was performed using total cell extracts as described previously (Ashcroft et al., 2000). Human p53, p21WAF1/CIP1 and beta-actin proteins were detected using the monoclonal antibodies DO-1, Ab-1 (Calbiochem), and C4 (Chemicon International Inc.), respectively. Human HA-tagged p73a, b and g isoforms were detected using the F-7 anti-HA antibody (Santa Cruz Biotechnology, Inc.) or anity puri®ed rabbit polyclonal antibody that was raised against the C-terminal 19 amino acids of human p73b (CTPPPPYHADPSLVRTWGP). The p14ARF polyclonal antibody has been described previously (Stott et al., 1998). Immunostaining Saos-2 cells were plated onto 10-cm diameter dishes (56105 cells) containing 1-cm diameter sterile glass coverslips, and transfected as described above. Cells were washed three times with PBS, then ®xed in 100% methanol for 10 min at room temperature. After ®xation, cells were washed three times with PBS and blocked in PBS containing 0.5% BSA for 30 min at room temperature. Cells were incubated for 30 min at room temperature with M2 anti-¯ag antibody (at 1 : 500, Sigma) in blocking solution, washed three times with PBS, then incubated for 30 min at room temperature with a rabbit anti-mouse conjugated FITC antibody (at 1 : 100, DAKO) in blocking solution containing 1 mg/ml DAPI (Sigma). Cells were washed three times with PBS and slides were mounted with PBS/glycerol mount.

Acknowledgments We would like to thank Gerry Melino (University of Rome, Italy), for generously providing the human pcDNAHAp73a, b and g constructs, and Gordon Peters for the gift of NARF2 cells. We are extremely grateful to Andy Phillips, Kevin Ryan and all other members of the Vousden Lab for their invaluable help, advice and encouragement during the course of these studies.


4289

References Agami R, Blandino G, Oren M and Shaul Y. (1999). Nature, 399, 809 ± 813. Ashcroft M and Vousden KH. (1999). Oncogene, 18, 7637 ± 7643. Ashcroft M, Taya Y and Vousden KH. (2000). Mol. Cell. Biol., 20, 3224 ± 3233. Attardi LD and Jacks T. (1999). Cell. Mol. Life Sci., 55, 48 ± 63. Bates S, Ryan KM, Phillips ADC and Vousden KH. (1998). Oncogene, 17, 1691 ± 1703. Bates S and Vousden KH. (1996). Curr. Opin. Genet. Dev., 6, 1 ± 7. Bates S and Vousden KH. (1999). Cell. Mol. Life Sci., 55, 28 ± 37. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown JP, Sedivy JM, Kinzler KW and Vogelstein B. (1998). Science, 282, 1497 ± 1501. El-Deiry W, Tokino T, Velculescu VE, Levy DB, Parson VE, Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein B. (1993). Cell, 75, 817 ± 825. El-Deiry WS. (1998). Semin. Cancer Biol., 8, 345 ± 357. Evans SC and Lozano G. (1997). Mol. Med. Today, 3, 390 ± 395. Friedlander P, Haupt Y, Prives C and Oren M. (1996). Mol. Cell. Biol., 16, 4961 ± 4971. Gong JG, Costanzo A, Yang H-Q, Melino G, Kaelin WG, Levrero M and Wang JYJ. (1999). Nature, 399, 806 ± 809. Hermeking H, Lengauer C, Polyak K, He T-C, Zhang L, Thiagalingam S, Kinzler KW and Vogelstein B. (1997). Mol. Cell, 1, 3 ± 11. Hollstein M, Hergenhahn M, Yang Q, Bartsch H, Wang ZQ and Hainaut P. (1999). Mutat. Res., 431, 199 ± 209. Jordan A and Reichard P. (1998). Annu. Rev. Biochem., 67, 71 ± 98. Lohrum MAE and Vousden KH. (1999). Cell Growth Dier., 6, 1162 ± 1168.

Ludwig RL, Bates S and Vousden KH. (1996). Mol. Cell. Biol., 16, 4952 ± 4960. Mills AA, Zheng B, Wang X-J, Vogel H, Roop DR and Bradley A. (1999). Nature, 398, 708 ± 713. Niculescu III AB, Chen X, Smeets M, Hengst L, Prives C and Reed SI. (1998). Mol. Cell. Biol., 18, 629 ± 643. Rowan S, Ludwig RL, Haupt Y, Bates S, Lu X, Oren M and Vousden KH. (1996). EMBO J., 15, 827 ± 838. Ryan KM, Ernst MK, Rice NR and Vousden KH. (2000). Nature, 404, 892 ± 897. Ryan KM and Vousden KH. (1998). Mol. Cell. Biol., 18, 3692 ± 3698. Sherr CJ. (1998). Genes Dev., 12, 2984 ± 2991. Stott F, Bates SA, James M, McConnell BB, Starborg M, Brookes S, Palmero I, Hara E, Ryan KM, Vousden KH and Peters G. (1998). EMBO J., 17, 5001 ± 5014. Tanaka H, Arakawa H, Yamaguchi T, Shiraishi K, Fukuda S, Matsui K, Takei Y and Nakamura Y. (2000). Nature, 404, 42 ± 49. Waldman T, Kinzler KW and Vogelstein B. (1995). Cancer Res., 55, 5187 ± 5190. Yang A, Schweitzer R, Sun D, Kaghad M, Walker N, Bronson RT, Tabin C, Sharpe A, Caput D, Crum C and McKeon F. (1999). Nature, 398, 714 ± 717. Yang A, Walker N, Bronson R, Kaghad M, Oosterwegel M, Bonnin J, Vagner C, Bonnet H, Dikkes P, Sharpe A, McKeon F and Caput D. (2000). Nature, 404, 99 ± 103. Yuan Z-M, Shioya H, Ishiko T, Sun X, Gu J, Huang YY, Lu H, Kharbanda S, Weichselbaum R and Kufe D. (1999). Nature, 399, 814 ± 817. Zhu J, Jiang J, Zhou W, Zhu K and Chen X. (1999). Oncogene, 18, 2149 ± 2155. Zhu L, Vandenheuvel S, Helin K, Fattaey A, Ewen M, Livingston D, Dyson N and Harlow E. (1993). Genes Dev., 7, 1111 ± 1125.

Oncogene