Novel Gain of Function Activity of P53 Mutants: Activation of the Dutpase Gene Expression Leading to Resistance to 5-ﬂuorouracil

Home , DUT (gene)

Oncogene (2002) 21, 4595 – 4600 ª 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc ORIGINAL PAPERS Novel gain of function activity of p53 mutants: activation of the dUTPase gene expression leading to resistance to 5-ﬂuorouracil

Elena N Pugacheva1,4,5, Alexey V Ivanov2,6, Julia E Kravchenko1, Boris P Kopnin3, Arnold J Levine4,7 and Peter M Chumakov*,1,2,6

1Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, 32 Vavilov Street, 117984 Moscow, Russia; 2Department of Molecular Genetics, University of Illinois at Chicago, Chicago, Illinois, IL 60607, USA; 3Institute of Cancerogenesis, Russian Cancer Research Center, Moscow, Kashirskoye shosse 24, 115478 Moscow, Russia; 4Department of Molecular Biology, Princeton University, Princeton, New Jersey, USA

Mutated forms of p53 are often expressed in a variety of Introduction human tumors. In addition to loss of function of the p53 tumor suppressor, mutant p53s contribute to malignant The p53 tumor suppressor gene plays a key role in process by acquisition of novel functions that enhance maintaining genetic stability and protection against transformed properties of cells and resistance to antic- cancer. Accumulation and activation of functional p53 ancer therapy in vitro, and increase tumorigenecity, protein occurs in response to a variety of stresses and invasiveness and metastatic ability in vivo. Searching for insults potentially leading to genetic alterations genes that change expression in response to p53 gain of (Levine, 1997; Pugacheva et al., 2000). Acting mainly function mutants may give a clue to the mechanisms (but not exclusively) as a sequence-specific transcrip- underlying their oncogenic effects. Recently by subtraction factor, p53 is capable of activating a set of genes tion hybridization cloning we found that the dUTPase required for cell cycle arrest or apoptosis (for review gene is transcriptionally upregulated in p53-null mouse see el-Deiry (1998); Sionov et al. (1999)). In this fibroblasts expressing the exogenous human tumor- manner the p53 gene controls elimination of potential derived His175 p53 mutant. Here we show that progenitors of malignant tumor cells. p53 deficient cells conditional expression of His175 and Trp248 hot-spot show greatly increased genetic instability, and most of p53 mutants in p53-negative mouse 10(1) fibroblasts and the p53 knockout mice develop malignant diseases by 6 human SK-OV3 and H1299 tumor cells results in months of age (Donehower et al., 1992). Different increase in dUTPase gene transcription, an important types of abnormalities in the p53 signaling pathways marker predicting the efficacy of cancer therapy with are observed in a majority of human malignancies. The fluoropyrimidine drugs. Using tetracycline-regulated most frequent type of defect represents missense retroviral vectors for conditional expression of p53 mutations within the p53 gene itself. Unexpectedly a mutants, we found that transcription of the dUTPase small fraction of mutations in the tumors corresponds gene is increased within 24 h after tetracycline with- to nonsense or frame shift mutations and that drawal, and the cells acquire higher resistance to 5-FU. discriminates the p53 gene from other tumor-suppres- Additional inactivation of the N-terminal transcription sors, such as RB or APC (Levine et al., 1995). This fact activation domain of mutant p53 (substitutions in amino- suggests that the presence of a mutated protein in acid residues 22 and 23) results in abrogation of both addition to the loss of wild type p53 protein might induction of dUTPase transcripts and 5-FU resistance. confer selective growth advantage and promote clonal Oncogene (2002) 21, 4595 – 4600. doi:10.1038/sj.onc. expansion (Greenblatt et al., 1994). In contrast to a 1205704 classical tumor suppressor gene, a missense mutation in the p53 gene can lead not only to loss of function (both Keywords: mutant p53; drug resistance; gain of in cis by mutation itself, or in trans by a dominant- function negative mechanism), but also to a gain of function. In particular, novel activities of mutant p53 are capable of promoting tumorigenicity when introduced into p53- *Correspondence: PM Chumakov, Department of Molecular Biol- null cell types (Dittmer et al., 1993; Michalovitz et al., ogy, Lerner Research Institute, Cleveland Clinic Foundation, 9500 1991). Also, over-expression of mutant p53 proteins in Euclid Avenue, Cleveland, Ohio, OH 44195, USA; p53-null cells of several lineages enhances their E-mail: [email protected] Current addresses: 5Fox Chase Cancer Center, 7701 Burtholme transformed properties and plating efficiency in culture, Avenue, Philadelphia, Pennsylvania, PA 19111, USA; 6Department and increases tumorigenicity and tissue invasiveness in of Molecular Biology, Lerner Research Institute, Cleveland Clinic mice (Dittmer et al., 1993; Gloushankova et al., 1997; Foundation, 9500 Euclid Avenue, Cleveland, Ohio, OH 44195, USA; Hsiao et al., 1994; Kremenetskaya et al., 1997; Sun et 7Rockefeller University, 1230 York Avenue, New York, NY 10021, USA al., 1993; Wolf et al., 1984). Introduction into p53-null Received 2 January 2002; revised 15 May 2002; accepted 22 May mice of a mutant p53 allele changes the tumor 2002 spectrum and results in higher frequencies of metas- Mutant p53s increase resistance to 5-FU EN Pugacheva et al 4596 tases (Liu et al., 2000). Some of the p53 mutants of dUTPase expression in colon cancer showed its provide increased resistance to p53-independent apop- strong correlation with poor response to 5-FU therapy tosis induced by a variety of treatments, including and a worse prognosis (Ladner et al., 2000). certain chemotherapeutic drugs (Kremenetskaya et al., In this study we show that certain types of mutant 1997; Li et al., 1998; Lotem and Sachs, 1995; Peled et p53s are capable of inducing expression of the al., 1996). For example, in p53 negative lung carcinoma dUTPase gene and, as a result, increasing resistance cell line H1299 introduction of His175 and His179 p53 of cells to fluoropyrimidine drugs. This observation mutants significantly reduces the rate of etoposide- provides a possible mechanism for increased resistance induced apoptosis, although His273 and Trp248 of certain malignant tumors to 5-FU, and may have mutants provide a much milder effect, indicating an potential value for prediction of sensitivity to allele specific impact, which provides further evidence chemotherapy based on analysis of mutant p53 that mutant p53 protein confers this activity (Blandino expression. et al., 1999). The mechanisms underlying gain of function activities of p53 mutants remain poorly understood. Some Results and discussion studies emphasize the importance of the N-terminal transcription-activation domain for expression of We tested whether expression of different forms of mutant p53 gain of function phenotype (Lanyi et al., tumor-derived p53 mutants introduced into p53 1998; Lin et al., 1995). Mutant p53s seem to be capable negative cells can affect dUTPase expression and of activating promoters of genes that are usually not sensitivity to 5-FU. Constitutive ectopic expression of activated by the wild type protein. Among these are His175 and Ala281 p53 mutants in p53-negative mouse promoters of MDR1 (Chin et al., 1992; Dittmer et al., cell line 10(3) results in significant increase in dUTPase 1993; Kopnin et al., 1995), PCNA (Deb et al., 1992), transcripts as seen by Northern blot hybridization HSP70 (Tsutsumi-Ishii et al., 1995), 15-lipoxygenase (Figure 1a). This result is consistent with the fact that (Kelavkar and Badr, 1999), BAG-1 (Yang et al., 1999) the dUTPase cDNA clone was present in a high and c-myc (Frazier et al., 1998) genes. It is not clear proportion of a subtraction library from 10(3)-His175 whether mutant p53 acts directly as a transcription vs 10(3) cell lines. factor with altered sequence specificity, or modifies Conditional expression vectors were then used to expression of the genes by interaction with factors study immediate effects of mutant p53s. Various forms responsible for positive or negative transcriptional of mutant p53 cDNAs were inserted into self- regulation. Taking into account how common missense inactivating retroviral vectors designed for tetracy- p53 mutants are in human cancer, it is clear that cline-regulated gene expression. Three different p53 understanding the mechanisms underlying this gain of deficient cell lines bearing a tetracycline dependent function as well as identification of genes that are transactivator protein, tTA, were infected with recom- regulated by p53 mutants may help in development of binant retroviral stocks obtained after transfection of rational strategies of cancer treatment. the DNA constructs into the packaging cell line In a recent study using subtraction hybridization Phoenix. Mass cultures obtained after selection with cloning we have isolated several cDNA clones G418 were induced to express p53 by withdrawal of corresponding to genes that are activated in p53- doxacycline from the medium. Twenty-four hours after negative mouse fibroblast cell line 10(3) following induction, strong signals corresponding to p53 were expression of exogenous tumor-derived human mutant seen on Western blots (Figure 1b,d). In the mouse p53 p53-His175 (Pugacheva et al., 2000). One of the most deficient fibroblast cell line 10(1) induction of His175 abundant cDNAs differentially expressed in the 10(3)- p53 mutant expression correlated with a significant p53His175 cell line corresponded to the mouse increase in the dUTPase transcript revealed by semi- homolog of the dUTPase gene. dUTPase mediates quantitative RT – PCR (Figure 1b). Disruption of the conversion of dUTP to dUMP, which can then be used transcriptional activation domain by a double muta- by thymidylate synthase (TS) for synthesis of dTMP. tion in amino acids 22 and 23 eliminated this effect Thus, dUTPase lowers dUTP/dTTP ratio preventing indicating that this domain is required for induction of mis-incorporation of uracil into DNA. Uracil mis- dUTPase. The effect was limited to mutant p53s, as incorporation plays a leading role in cytotoxicity wild type p53 expression did not increase dUTPase mediated by TS inhibitors, such as FUdR and a expression while increasing transcription of p21-Waf1. broadly used anticancer drug 5-FU (Aherne and A similar response to expression of His175 mutant was Browne, 1999). Inhibition of TS leads to increase in observed in two p53-negative human cell lines derived dUTP levels and incorporation of uracil into DNA. from lung (H1299) and ovarian (SK-OV3) carcinoma This initiates DNA excision repair process mediated by cell lines (Figure 1c,d). In SK-OV3 cells the importance uracil-DNA glycosylase (Lindhal, 1982). Massive DNA of the transcriptional activation domain of p53 for repair generates multiple DNA strand breaks that induction of dUTpase was also demonstrated. ultimately lead to cell death. Ectopic expression of To test whether there could be some specific effects dUTPase results in significant protection from FUdR- on dUTPase expression depending on the type of p53 induced DNA strand breaks and increased viability missense mutation, two additional tumor derived hot- over control cells (Canman et al., 1994). Comparison spot mutant p53 cDNAs, His273 and Trp248, were

Oncogene Mutant p53s increase resistance to 5-FU EN Pugacheva et al 4597

Figure 1 Changes in the dUTPase gene transcripts in p53 negative cells after ectopic expression of p53 mutants. (a) Northern blot hybridization of the total RNA (20 mg) from mouse cells 10(3) expressing Ala281 and His175 p53 mutants with labeled mouse dUT- Pase cDNA probe. (b) RT – PCR analysis of the dUTPase transcripts in mouse cells 10(1) harboring tetracycline-regulated constructs expressing wt-p53, 22/23-p53, and 22/23/His175 p53 in the presence (+) and following 24 h after removal (7)of1mg/ ml of tetracycline. RT – PCR with p21-Waf1 primers was used as control for transcriptional activity of the wt-p53. Beta-actin primers were used for normalization of cDNA samples. (c) RT – PCR analysis of the dUTPase transcripts in human lung carcinoma cells H1299 harboring tetracycline-regulated wt-p53, His175 p53 and Trp248 p53. RT – PCR with primers specific for human p53 were used for control of induction of p53 expression. (d) RT – PCR analysis of the dUTPase transcripts in human ovary carcinoma cells SK-OV3 harboring tetracycline-regulated wt-p53, 22/23/His175 p53, His175, His273 p53 and Trp 248 p53. RT – PCR with primers specific for the human c-myc were used to monitor expression of one of the genes known to be upregulated by mutant p53s expressed in p53 negative cells. While the Trp248 p53 formation was observed when a similar assay was mutant showed similar induction of dUTPase tran- performed with doxorubicin (Figure 2a,b,c). The scripts, there was no visible effect from His273 both in His175 p53 mutant induced increase in colony SK-OV3 (Figure 1d) and H1299 cell lines (not shown). formation in the presence of 5-FU was comparable This differential effect of p53 mutants was specific for with that observed with 10(1) cells expressing exogen- the dUTPase gene, as c-myc, another gene known to ous human dUTPase gene (Figure 2c). Expression of respond to p53 mutants (Frazier et al., 1998) was His175 p53 mutant with a disrupted transcriptional- induced by all three mutants equally well (Figure 1d). activation domain (mutation in amino-acids 22 and 23) The result was similar with SK-OV3 and H1299 cell did not show any increase in 5-FU resistance. This lines (not shown). result is in agreement with a lack of dUTPase Next, we examined how the increase in dUTPase induction in 10(1) cells expressing His175 mutant expression induced by p53 mutants affects sensitivity of containing mutations in amino acids 22 and 23 (Figure cells to a chemotherapeutic drug 5-FU. Testing 1b). Therefore, it is likely that an intact transcription sensitivity to another chemotherapeutic drug, doxor- activation domain is important for either direct or ubicin, was used as a control. In the 10(1) cells indirect induction of dUTPase expression leading to induction of the His175 p53 mutant resulted in a increased resistance to 5-FU. This result is also in good significant increase in the number of colonies formed in agreement with recent study showing that the integrity the presence of 5-FU, but no increase in colony of N-terminal transcription domain of p53 is required

Oncogene Mutant p53s increase resistance to 5-FU EN Pugacheva et al 4598

Figure 2 Colony formation of 10(1) cell sublines in the presence of increasing concentrations of 5-FU (a) and doxorubicin (b). Colony formation in the presence of 5-FU and doxorubicin (DOX) of 10(1) cells (c), H1299 cells (d) and SK-OV3 cells (e) with tetracycline-regulated expression of p53 mutants with (tet+) and without (tet-) tetracycline in the medium

for mutant p53 interference with cisplatin-induced FU (Figure 3) or 15 nM of doxorobucin (not shown) in apoptosis (Matas et al., 2001). a sub-confluent culture for 3 days, and then re-plated In two p53 negative human cell lines H1299 and SK- to allow further 4 – 5 divisions to reach confluence. The OV3, a good correlation was found between the ability proportion of EGFP positive cells was determined by of p53 mutants to increase expression of the dUTPase FACScan. It was found that while treatment of 10(1)- gene, and to increase number of colonies formed in the EGFP cells mixed in equal proportion with unlabeled presence of 150 nM of 5-FU (Figure 2d,e). The 10(1), or 10(1)-22/23-His175 cells yielded nearly 50% strongest increase in colony formation was observed of EGFP-positive cells, only 25 – 30% of cells showed with the His175 p53 mutant. Another p53 mutant, EGFP fluorescence in mixtures of 10(1)-EGFP cells Trp248, was also capable of increasing the number of with either 10(1)-His175, or 10(1)-dUTPase cells 5-FU resistant colonies, although the effect was (Figure 3a). When 10(1)-His175-EGFP cells were moderate. In contrast, His273 p53 mutant over- mixed with 10(1) cells, the proportion of labeled cells expressed in SK-OV3 cells produced neither an increase after 5-FU treatment increased to 65 – 70% showing in dUTPase transcripts nor increased survival of cells greater resistance of mutant-p53 expressing cells. In in the presence of 150 nM 5-FU (Figure 1d, and Figure mixtures of 10(1)-His175-EGFP and 10(1)-dUTPase 2e). It should be noted that expression of either p53 cells the proportion of EGFP positive cells was 55 – mutants in the cell lines tested produced no changes in 60% indicating that mutant p53 might confer even number of colonies formed in the presence of higher resistance to 5-FU than expression of the doxorubicin. Therefore, the effect was not due to some dUTPase gene alone (Figure 3b). Therefore, although nonspecific increase in cell survival, but was limited to expression of dUTPase seems to be the major factor resistance to fluoropyrimidine drugs. responsible for greater viability of cells in the presence To test the possibility that treatment with 5-FU may of 5-FU, some additional activities of mutant p53 may result in preferential selection of mutant p53 expressing contribute to the resistance. Other genes whose cells, an experiment was set up that monitors changes expression was altered by mutant p53s might mediate in proportion of different types of cells in cell mixtures these activities. Treatment of similar cell mixtures with after treatment with different drugs. 10(1) and 10(1)- 15 nM of doxorubicin did not reveal any significant His175 cells were labeled by introduction of retroviral changes in proportion of EGFP-expressing cells (result construct expressing EGFP (pPS-EGFP-neo). Both not shown). types of fluorescent-labeled cells were mixed with an The results obtained in this study reveal a novel gain equal number of non-labeled cells expressing either p53 of function activity conferred by certain p53 mutants mutants (His175 or 22/23-His175), or exogenous that directly links expression of mutant p53 activity human dUTPase in a sense or an anti-sense orienta- with resistance to chemotherapeutic drugs of the tion. The cell mixtures were exposed to 150 nM of 5- fluoro-pyrimidine group. Clinical studies have revealed

Oncogene Mutant p53s increase resistance to 5-FU EN Pugacheva et al 4599 (Gossen and Bujard, 1992). Selected functional tTA expressing clones were infected with self-inactivating retroviral vectors (pSIT-neo) carrying diﬀerent mutant p53s (His175, Trp248, His273, Leu22/Trp23, His175/Leu22/Trp23) under control of a tetracycline-dependent promoter. G418-resistant cultures were selected and maintained in the presence of 250 ng/ml of doxycycline to suppress p53 expression. For induction of the p53 expression, the cells were incubated in doxycycline-free medium for 24 – 48 h.

Figure 3 Changes in proportion of EGFP-labeled cells 10(1) and Western blot analysis 10(1)-His175 in mixtures with unlabeled 10(1) cell sublines expres- Cells were washed twice with PBS, scraped from Petri dishes sing p53 mutants, or recombinant mouse dUTPase cDNA in sense and antisense. Labeled and unlabeled cells were mixed in cold PBS followed by brief centrifugation. The 10(1) cell 50 : 50 before treatment with 5-FU (see Materials and Methods). pellets were lysed in buffer (50 mM Tris-HCl pH7.5, 150 mM Proportion of EGFP-labeled cells was determined by FACScan NaCl, 0.5% NP-40 and 1 mM EDTA with addition of 1 mM dithiothreitol, 100 mM PMSF, 1 mM pepstatin A and 1 mM E64). The SK-OV3 and H1299 cell pellets were lysed in the RIPA buffer. Equal amounts of total protein (approximately higher resistance to fluoropyrimidine therapy of tumors 600 mg) were used for overnight immunoprecipitation with expressing p53 mutants (Benhattar et al., 1996; the p53 specific monoclonal antibodies PAb421 cross-linked Cabelguenne et al., 2000; Zheng et al., 1999). In to Protein A-Sepharose CL4B at 48C. The precipitated several cases such effects of p53 mutants were proteins were separated on 10% SDS – PAAG electrophor- attributed to dominant-negative suppression of the esis, transferred onto Immobilon-P (Millipore) membrane, wild type p53 activity that resulted in abrogation of probed with the p53 specific monoclonal antibodies PAb421 p53-induced apoptosis (Bunz et al., 1999), and an and developed using peroxidase conjugated anti-mouse IgG increase in activity of the MDR1 gene (Bottini et al., and ECL-Plus Western blot reagents (Amersham). 2000). Clearly, induction of dUTPase by certain p53 mutants can additionally contribute to resistance to 5- Northern hybridization FU therapy. It has yet to be established whether there Total RNA was isolated with Trizol reagent (GIBCO – BRL). is any correlation between the type of p53 mutation in Separation on 1% agarose gel, capillary transfer to Hybond- human tumors and the outcome of 5-FU therapy. N membrane, and hybridization with a corresponding 32P- Differential effects of hot-spot p53 mutants on labeled cDNA probes were performed according to published expression of dUTPase and resistance to 5-FU protocols (Sambrook et al., 1989). Each sample applied on a (His174 and Trp248 vs His273 mutants) indicate the gel contained 20 mg of total RNA. The dUTPase cDNA used potential value of information about a particular type as a probe was obtained by RACE PCR. of p53 mutation for the development of rational therapeutic schemes. Testing other types of p53 Semi-quantitative RT – PCR mutations are necessary to extend these observations, cDNA was synthesized on 5 mg of total RNA, with oligo- and to reveal common structural features required for (dT)18 – 23 primer and AMV reverse transcriptase according to this particular gain of function phenotype. Identifica- manufacturer’s recommendations (Amersham PLC). PCR tion of dUTPase as a specific mutant p53 up-regulated amplifications were performed with primers that yield 126 bp gene may provide a basis for design of rational drugs fragment of the mouse dUTPase cDNA: 5’-AGCATTTGGT- and development of novel therapeutic strategies. In GTTCTAGGATGCAGGA; 5’-ACCAAGTCTGCATGTCA- addition, these results validate the approach of study- ATGCCATGCTC; 520 bp fragment of the human dUTPase ing p53 gain of function by identification of new cDNA: 5’-CGGGATCCCGCCTTCTGGCTCTGCCATGC- transcriptional targets of mutant p53 and further direct CCTGC-3’;5’-GGAATTCGCATAAATTTTAATTCTTTC- CAGTG-3’;5’-GGAATTCGCATAAATTTTAATTCTTTC- testing of their contribution to the malignant pheno- CAGTG-3’, 278 bp fragment of mouse p21-WAF1 cDNA: 5’- type. CCGAGAACGGTGGAACTTTGA; 5’-GTTTTCGGCCCT- GAGATGTTG; 589 bp fragment of human c-myc cDNA: 5’- AAGATGAGGAAGAAAACGATGTTG; 5’-ACATTTCT- Materials and methods GTTAGAAGGAATCGT; and 604 bp fragment corresponding to the beta-actin cDNA: 5’-AAGATGACCCAGATCAT- Cell lines GTTTGAGACC; 5’-GCAGTAATCTCCTTCTGCATCCT- All cell lines were grown in DMEM supplemented with 10% GTCA. Amplification was carried out with HotTub DNA FBS at 378C and 5% CO2. Mouse p53-null fibroblast cell polymerase (Amersham) for 20, 25 and 30 cycles. The PCR lines 10(3) (Harvey and Levine, 1991) with introduced His175 products were separated by agarose gel electrophoresis. and Ala281 p53 mutants under control of MuMLV LTR Colonogenic survival assays of the mouse 10(1), the human were described earlier (Pugacheva et al, 2000). For tetra- SK-OV3 and H1299 cells with tetracycline-regulated expres- cycline-regulated expression of p53 mutants, mouse 10(1) and sion of different p53 mutants were performed in 6-well plates. 10(3) (Harvey and Levine, 1991) and human SK-OV3 (ATTC One6103 for the 10(1) and the H1299 cells, 46103 for the HTB-77) and NCI-H1299 (ATCC CRL-5803) p53-null cell SK-OV3 cells were seeded onto each well with and without lines were first infected with retroviral vector pPS-tTA-hygro 250 ng/ml of doxycycline. Forty-eight hours later the medium expressing tetracycline-dependent transactivator protein tTA containing drugs at LD90 concentration (for 10(1) 250 nM,

Oncogene Mutant p53s increase resistance to 5-FU EN Pugacheva et al 4600 for the SK-OV3 and the H1299 150 nM of 5-FU; and for the after three weeks of further cultivation. The cells of SK-OV3 1 nM, for 10(1) and the H1299 10 nM of corresponding EGFP negative and EGFP-positive lines were doxorubicin) was added and changed every 4 days. Eighteen mixed in equal proportion (50 : 50) and were allowed to grow days later the cells were fixed with methanol, stained with for 2 days. For the experiments 56104 cells of mixed cell Giemsa, and the number of colonies was counted. Quantita- populations were seeded onto 6-well plates with and without tion of EGPF-positive cells by FACS. EGFP labeled cells doxycycline. Twenty-four hours later half of the wells were were obtained by retroviral-mediated infection with super- treated with 250 nM of 5-FU. After an additional 24 h the natants of the packaging cell line Phoenix-ampho after cells were washed twice with PBS and fresh medium with lipofection of the pPS – EGFP retroviral vector. Five days doxycycline was added. After 5 – 7 days of recovery the after infection the cells were subjected to fluorescent activated cultures approaching monolayer were spitted, and the cell sorting (FACS Vantage, BD). More than 90% of the cell percentage of the EGFP-positive cells was determined by populations obtained had maintained green fluorescence even FACS.

References

Aherne GW and Browne S. (1999). Anticancer Drug Kremenetskaya OS, Logacheva NP, Baryshnikov AY, Development Guide: Antifolate Drugs in Cancer Therapy. Chumakov PM and Kopnin BP. (1997). Oncol. Res., 9, Jackman AL (ed). Totowa, NJ: Humana Press, Inc, pp. 155 – 166. 409 – 421. Ladner RD, Lynch FJ, Groshen S, Xiong YP, Sherrod A, Benhattar J, Cerottini JP, Saraga E, Metthez G and Givel JC. Caradonna SJ, Stoehlmacher J and Lenz HJ. (2000). (1996). Int. J. Cancer, 69, 190 – 192. Cancer Res., 60, 3493 – 3503. Blandino G, Levine AJ and Oren M. (1999). Oncogene, 18, Lanyi A, Deb D, Seymour RC, Ludes-Meyers JH, Subler 477 – 485. MA and Deb S. (1998). Oncogene, 16, 3169 – 3176. Bottini A, Berruti A, Bersiga A, Brizzi MP, Brunelli A, Levine AJ. (1997). Cell, 88, 323 – 331. GorzegnoG,DiMarcoB,AgugginiS,BolsiG,CirilloF, Levine AJ, Wu MC, Chang A, Silver A, Attiyeh EF, Lin J Filippini L, Betri E, Bertoli G, Alquati P and Dogliotti L. and Epstein CB. (1995). Ann. NY Acad. Sci., 768, 111 – (2000). Clin. Cancer Res., 6, 2751 – 2758. 128. Bunz L, Hwang PM, Torrance C, Waldman T, Zhang Y, Li R, Sutphin PD, Schwartz D, Matas D, Almog N, Dillehay L, Williams J, Lengauer C, Kinzler KW and Wolkowicz R, Goldﬁnger N, Pei H, Prokocimer M and Vogelstein B. (1999). J. Clin. Invest., 104, 263 – 269. Rotter V. (1998). Oncogene, 16, 3269 – 3277. Cabelguenne A, Blons H, de Waziers I, Carnot F, Houllier Lin J, Teresky AK and Levine AJ. (1995). Oncogene, 10, AM, Soussi T, Brasnu D, Beaune P, Laccourreye O and 2387 – 2390. Laurent-Puig P. (2000). J. Clin. Oncol., 18, 1465 – 1473. Lindhal T. (1982). Annu. Rev. Biochem., 51, 61 – 87. Canman CE, Radany EH, Parsels LA, Davis MA, Lawrence Liu G, McDonnell TJ, Montes de Oca Luna R, Kapoor M, TS and Maybaum J. (1994). Cancer Res., 54, 2296 – 2298. Mims B, El-Naggar AK and Lozano G. (2000). Proc. Natl. Chin KV, Ueda K, Pastan I and Gottesman MM. (1992). Acad. Sci. USA, 97, 4174 – 4179. Science, 255, 459 – 462. Lotem J and Sachs L. (1995). Proc.Natl.Acad.Sci.USA,92, DebS,JacksonCT,SublerMAandMartinDW.(1992).J. 9672 – 9676. Virol., 66, 6164 – 6170. Matas D, Sigal A, Stambolsky P, Milyavsky M, Weisz L, DittmerD,PatiS,ZambettiG,ChuS,TereskyAK,Moore Schwartz D, Goldﬁnger N and Rotter V. (2001). EMBO J., M, Finlay C and Levine AJ. (1993). Nat. Genet., 4, 42 – 46. 20, 4163 – 4172. Donehower LA, Harvey M, Slagle BL, McArthur MJ, Michalovitz D, Halevy O and Oren M. (1991). J. Cell Montgomery Jr CA, Butel JS and Bradley A. (1992). Biochem., 45, 22 – 29. Nature, 356, 215 – 221. Peled A, Zipori D and Rotter V. (1996). Cancer Res., 56, el-Deiry WS. (1998). Semin. Cancer Biol., 8, 345 – 357. 2148 – 2156. Frazier MW, He X, Wang J, Gu Z, Cleveland JL and Pugacheva EN, Snegur IN, Kopnin BP and Chumakov PM. Zambetti GP. (1998). Mol. Cell. Biol., 18, 3735 – 3743. (2000). Mol. Biol., 34, 143 – 151. Gloushankova N, Ossovskaya V, Vasiliev J, Chumakov P Sambrook J, Fritsch E and Maniatis T. (1989). Molecular and Kopnin B. (1997). Oncogene, 15, 2985 – 2989. Cloning. New York: Cold Spring Harbor Laboratory Gossen M and Bujard H. (1992). Proc. Natl. Acad. Sci. USA, Press. 89, 5547 – 5551. Sionov RV, Moallem E, Berger M, Kazaz A, Gerlitz O, Ben- Greenblatt MS, Bennett WP, Hollstein M and Harris CC. Neriah Y, Oren M and Haupt Y. (1999). J. Biol. Chem., (1994). Cancer Res., 54, 4855 – 4878. 274, 8371 – 8374. Harvey DM and Levine AJ. (1991). Genes Dev., 5, 2375 – Sun Y, Nakamura K, Wendel E and Colburn N. (1993). 2385. Proc. Natl. Acad. Sci. USA, 90, 2827 – 2831. Hsiao M, Low J, Dorn E, Ku D, Pattengale P, Yeargin J and Tsutsumi-Ishii Y, Tadokoro K, Hanaoka F and Tsuchida N. Haas M. (1994). Am.J.Pathol.,145, 702 – 714. (1995). Cell Growth Differ., 6, 1–8. Kelavkar UP and Badr KF. (1999). Proc. Natl. Acad. Sci. Wolf D, Harris N and Rotter V. (1984). Cell, 38, 119 – 126. USA, 96, 4378 – 4383. Yang X, Pater A and Tang SC. (1999). Oncogene, 18, 4546 – Kopnin BP, Stromskaya TP, Kondratov RV, Ossovskaya 4553. VS, Pugacheva EN, Rybalkina EY, Khokhlova OA and Zheng M, Wang H, Zhang H, Ou Q, Shen B, Li N and Yu B. Chumakov PM. (1995). Oncol. Res., 7, 299 – 306. (1999). J. Cancer Res. Clin. Oncol., 125, 357 – 360.

Oncogene