High frequency in esophageal cancers of p53 alterations inactivating ...

Carcinogenesis vol.21 no.4 pp.563–565, 2000

High frequency in esophageal cancers of p53 alterations inactivating the regulation of genes involved in cell cycle and apoptosis

Vale´rie Robert, Pierre Michel1, Jean Michel Flaman, Anne Chiron1, Cosette Martin, Francoise Charbonnier, Bernard Paillot1 and Thierry Frebourg2 INSERM EPI 9906, Faculte´ de Me´decine et de Pharmacie, 22 Boulevard de Gambetta, 76183 Rouen and IFRMP, 76821 Mont-Saint-Aignan Cedex and 1Centre de De ´ pistage et de Traitement des Tumeurs Digestives, CHU de Rouen, 76031 Rouen, France 2To

whom correspondence should be addressed at INSERM EPI 9906, Faculte´ de Me´decine et de Pharmacie, 22 Boulevard de Gambetta, 76183 Rouen, France Email: [email protected]

Somatic mutations of the tumor suppressor gene p53 have been frequently detected in esophagal cancers, but their biological significance remains to be established. The tumor suppressor activity of p53 results in part from its ability to transactivate genes involved in the cell cycle and apoptosis, such as p21, bax and PIG3, and some p53 mutations may have a differential effect on the transactivation of these target genes. We developed yeast strains in which the activation by wild-type p53 of reporter plasmids containing p53 binding sites present within these target genes induces a change in the color of the colonies (red/ white). Using these strains, we analyzed 56 esophageal cancers from patients residing in Normandy, France, a high incidence geographic area. Forty-seven tumors (84%), scored as mutant with the p21, bax and PIG3 reporter strains and in most of the cases (76%), the percentage of red colonies suggested that both p53 alleles were inactivated. Sequencing analysis allowed the identification of a p53 mutation in each positive sample, and the spectrum of mutations was in agreement with the etiological role of tobacco and alcohol. These results confirm the high frequency of biallelic p53 mutations in esophageal carcinoma and strongly suggest that their biological consequence is the complete alteration of the transactivation of genes involved in the cell cycle and apoptosis, which indicates that p53 alteration is a key event in esophagus carcinogenesis.

Introduction Somatic mutations of the p53 tumor suppressor gene are the most frequently observed molecular alterations in human tumors, and p53 appears to be an essential molecular target of carcinogens (for review see refs 1–3). The tumor suppressor activity of p53 results, at least in part, from its ability to upregulate transcription of genes involved in the cell cycle and apoptosis, such as p21, bax and PIG3. p21 encodes a cdk/ cyclin inhibitor which controls the G1/S and G2/M transitions (4–6). The bax gene, which encodes a mitochondrial protein regulating caspase activation via cytochrome c release (7,8), was the first identified target gene of p53 involved in apoptosis, but extensive analysis of transcripts induced by p53 before apoptosis led more recently to the identification of genes such as PIG3, potentially related to the redox status of the cell (9). © Oxford University Press

Although more than 700 different mutations of p53 have been described, and listed in databases (10,11), their biological consequences have not been characterized. This point is especially critical to analyze the prognostic value of p53 mutations and to demonstrate the real involvement of p53 alterations in malignant transformation. It has been shown that some p53 mutations selectively impaired the transactivation of bax and therefore the apoptotic function of p53 (12–14), indicating that some p53 mutations may have a differential effect on the transactivation of the target genes. In northwestern France, esophagus cancer has a high incidence and is attributable to the consumption of alcoholic beverages and to tobacco smoking. The incidence of p53 mutations had previously been shown to be particularly high in esophageal cancers (1,15–20), which suggests that p53 inactivation is a key event in esophagus carcinogenesis. Using functional assays developed in yeast, we characterized in this study the biological effects of p53 mutations in these tumors. Materials and methods Clinical samples Biopsies of esophagus tumors were collected from 56 untreated patients undergoing diagnostic endoscopy. Histopathological examination subsequently revealed that these samples corresponded to 50 squamous cell carcinomas and six adenocarcinomas. These 56 tumors were staged according to the TNM classification into four stage I, nine stage II, 11 stage IIA, four stage IIB, 17 stage III and 11 stage IV tumors. All patients, except six, were habitual tobacco and/or alcohol consumers. Samples were directly placed in 400 µl of RNA extraction buffer (Pharmacia) in order to prevent RNA degradation and stored at –20°C until analyzed. Plasmids, strains and media The RGC, p21, bax reporter plasmids and the corresponding yIG397RGC, YPH-p21 and YPH-bax yeast strains have previously been described (14,21,22). The PIG3 reporter plasmid contains, upstream of the CYC1 minimal promoter and the ADE2 open reading frame, one copy of a 20 bp sequence (5-CAG CTT GCC CAC CCA TGC TC-3⬘) containing the p53 binding site found 308 bp upstream of the transcription start site of the PIG3 gene (9). The PIG3 reporter plasmid was cut at the ApaI restriction site and was integrated in vivo by homologous recombination at the URA3 locus of the ade– yeast strain YPH500 and selected for uracil prototrophy, giving YPHPIG3. The YPH-PIG3 strain, like the yIG397-RGC, YPH-p21 and YPH-bax strains, is spontaneously red when growing on media containing limiting adenine. These strains should routinely be cultured on complete medium supplemented with 200 µg/ml adenine to avoid selection of spontaneous suppressors of the endogenous mutant ADE2 locus. p53 functional assay To validate the YPH-PIG3 strain, we PCR amplified from plasmids wild-type or defined mutant p53 cDNA without any detectable transcriptional activity in YPH-p21 and YPH-bax yeast strains (14) (131 Asn, 135 Tyr, 143 Ala, 177 Leu, 179 Arg, 220 Cys, 245 Asp, 245 Cys, 248 Gln, 248 Trp, 248 Leu, 258 Lys, 273 His, 273 Gly and 282 Trp). The YPH-PIG3 strain was then cotransformed with unpurified PCR product, pSS16 expression vector linearized by HindIII and StuI and carrier DNA using the lithium acetate procedure, as described previously (21,22). Activation of the PIG3 reporter plasmid by wild-type p53 resulted in white colonies, whereas all mutants tested failed to activate the reporter system, as indicated by red colonies, indicating that their phenotype was (p21–/PIG3–/bax–). We then analyzed in this strain two p53 mutants with a differential effect on p21 and bax reporter strains (p21⫹/bax–), 175 Leu and 283 His (14). Expression of the 175 Leu mutant in the YPHPIG3 yielded white colonies whereas red colonies were observed with the

563

V.Robert et al. 283 His mutant, indicating that the phenotypes of these mutants were (p21⫹/ PIG3⫹/bax–) and (p21⫹/PIG3–/bax–), respectively. mRNA extraction from esophagus tumors, cDNA synthesis, PCR-amplification of p53 cDNA and functional assay in the different reporter yeast strains were performed as described previously (21,22). For each transformation, the percentage of red colonies was estimated from at least 200 yeast colonies. Sequencing analysis Sequencing analysis was performed on p53 cDNA derived either from RT– PCR or from p53 plasmids rescued from three independent red yeast colonies, as described previously (21,22). Codons 53–142, 133–291 and 257–364 of p53 were respectively amplified with primers P3 (21) and 2 Rb (5⬘-TGT AAA ACG ACG GCC AGT CAA CCC ACA GCT GCA CAG-3⬘), 5F (5⬘-CAG GAA ACA GCT ATG ACC TCC CCT GCC CTC AAC AAG ATG-3⬘) and 8R (5⬘-TGT AAA ACG ACG GCC AGT TCG TGG TGA GGC TCC CCT TTC-3⬘), 5Fb (5⬘-CAG GAA ACA GCT ATG ACC CCA TCC TCA CCA TCA TCA CA-3⬘) and P4 (21). Underlined nucleotides correspond to additional M13R or M13-21 sequences. The PCR consisted of 35 cycles of 30 s at 94°C, 30 s at 58°C and 50 s at 72°C, preceded by 3 min at 95°C and followed by 5 min at 72°C. PCR products were purified after electrophoresis using QIAquick Gel Extraction Kit (Qiagen) and directly sequenced on both strands using the PRISM AmpliTaqFS Ready Reaction Dye Primer or Big Dye Terminator Cycle Sequencing kits (PE Applied Biosystems, Perkin Elmer) and an Applied Biosystems model 373A automated sequencer.

Results In this study, we analyzed 56 esophageal tumors, including 50 squamous cell carcinomas and six adenocarcinomas, using functional assays developed in yeast. In these types of assays (21,22), reporter yeast strains are co-transformed with p53 cDNA derived from RT and amplified between codons 53 and 364 and a gapped expression vector linearized between codons 67 and 346, cDNA are cloned in vivo by homologous recombination, and the activation by wild-type p53 of the reporter system containing the ADE2 open reading frame and a p53 binding site changes the colour of the yeast colonies (red→white). Samples containing wild-type p53 commonly give a background of 5–10% red colonies due to PCR-induced errors and the presence of an alternatively spliced p53 mRNA, and a value of ⬎10% red colonies therefore indicates the presence of p53 mutation (21,22). To detect p53 mutations, we first used in this study the yIG397-RGC reporter strain containing the RGC sequence (21,22), which corresponds to a low affinity sequence for p53. Forty-seven tumors (84%), corresponding to 41/50 squamous cell carcinomas (82%) and the six adenocarcinomas, generated a percentage of red colonies ⬎10%. In most of the cases (35/47), the percentage of red colonies was ⬎50%. We then analyzed the 47 positive tumors with the YPH-p21, YPH-bax and YPH-PIG3 reporter strains to determine the specific effects of the p53 alterations present in the tumors on the transactivation of the corresponding target genes. We used, as controls, three p53 mutants with a different phenotype according to their ability to transactivate the reporter systems, 248 Trp (p21–/PIG3–/bax–), 283 His (p21⫹/PIG3–/ bax–) and 175 Leu (p21⫹/PIG3⫹/bax–). All the tumors yielded in these strains a percentage of red colonies ⬎10%, indicating that they contain only (p21–/PIG3–/bax–) mutants. To determine the mutational spectrum in our series of esophageal tumors, we sequenced p53 between codons 53 (exon 4) and 364 (exon 10) in the 47 positive samples. This analysis, performed either directly on the RT–PCR product (when the percentage of red colonies was ⬎50%) or on plasmids rescued from red colonies, identified a p53 mutation in all the samples (Table I). Discussion Using functional assays in yeast, we confirmed the high frequency of somatic p53 alterations in esophagal cancers and 564

Table I. p53 mutations identified in 47 oesophageal cancers Tumor

Codon

Nucleotide changea

Amino-acid change

O122b O137b O49b O62d O68b O109b O59b O55b O95b O25b O180b O83b O22b O32b O171b O121b O134d O41b O87b O99b O47d O103b O114b O19b O44b O13b O191b O90b O35b O80b O131d O130d O1b O71d O106b O65b O101b O119b O143b O161b O7b O4b O155b O29b O57b O158b O77b

110 148 151 155 155 155 158 163 163 173 175 176 179 179 179 187 192 195 205 212 213 214 220 230 232 236 236 239 244 245 245 246 248 248 248 255 255 265 265 265 272 273 273 282 282 296 343

CGT→CTT GAT→GTT CCC→CAC ACC→CCC ACC→CCC ACC→CCC CGC→CTC del1Be TAC→TGC GTG→TTG CGC→CAC TGC→TGG CAT→CGT CAT→CGT CAT→CGT del exon 6 CAG→TAG ATC→TTC TAT→TGT del2BCe CGA→TGA del1Ce TAT→TGT del1Ce ATC→TTC TAC→TGC TAC→TCC AAC→AGC GGC→AGC GGC→GTC GGC→GAC ATG→GTG CGG→TGG CGG→CAG CGG→CAG ATC→TTC ATC→TTC CTG→CCG CTG→CCG CTG→CCG GTG→ATG CGT→CTT CGT→TGT CGG→TGG CGG→TGG CAC→CTC GAG→TAG

Arg→Leu Asp→Val Pro→Hisc Thr→Proc Thr→Proc Thr→Proc Arg→Leuc Stop at 169 Tyr →Cysc Val→Leuc Arg→Hisc Cys→Trp His→Argc His→Argc His→Argc Stop at 190 Gln→Stopc Ile→Phec Tyr→Cysc Stop at 214 Arg→Stopc Stop at 246 Tyr→Cysc Stop at 246 Ile→Phe Tyr→Cysc Tyr→Ser Asn→Serc Gly→Ser Gly→Valc Gly→Aspc Met→Valc Arg→Trpc Arg→Glnc Arg→Glnc Ile→Phec Ile→Phec Leu→Pro Leu→Pro Leu→Pro Val→Metc Arg→Leuc Arg→Cysc Arg→Trpc Arg→Trpc His→Leuc Glu→Stop

aDetected

in the cDNA sequence (see Materials and methods). For each tumor, except O55, O22, O121, O114, O44, O143 and O161, mutations were confirmed by a second independent RT–PCR. bSquamous cell carcinoma. cPreviously reported in esophageal cancers (10). dAdenocarcinoma. eThe number of bases deleted and the position of this deletion in the codon (A, B or C) are indicated according to Beroud et al. (11).

we show that these mutations abolish the ability of p53 to activate p21, bax and PIG3 reporter plasmids. The percentage of p53 mutations in esophagal cancers that we report in this study (84%) is the highest reported with that reported by Audrezet et al. (16), who had analyzed exons 2–8 of p53 using DGGE. It should be noted that patients analyzed in both studies originated from the West part of France (Normandy and Brittany, respectively) where esophagus cancer has a high incidence. In p53 functional assays, the percentage of red colonies derived from a tumor reflects the inactivation of one or both alleles and depends on the presence of normal cells in

Inactivating p53 mutations in oesophageal cancers

the sample. Among 47 esophagus biopsies which scored positive in the functional assay, 35 samples generated a percentage of red colonies ⬎50% which may suggest that, in most of the tumors, both p53 alleles were inactivated. A percentage of red colonies between 10 and 50% probably reflects the presence of a functional allele in the tumor and the presence of normal cells in the biopsy. Under this circumstance, the assay facilitates the identification of the mutation, since p53 cDNA is cloned by homogous recombination and mutant cDNA can specifically be rescued from red colonies. The spectrum of p53 mutations, that we have identified in squamous cell carcinomas (Table I), is in agreement with the etiological role of tobacco and alcohol and is similar to that previously reported in most of the studies (1), with (i) the predominance of transitions or transversions occurring at A:T base pairs (41%), which might be related to ethanol metabolites such as acetaldehyde (19); (ii) the presence of G→T transversions (15%) observed in tobacco-related cancers (1,2) and attributable to benzo[a]pyrene present in cigarette smoke (23); and (iii) the low frequency of C→T transitions at CpG sites (10%), which represent the predominant p53 mutations observed in other gastrointestinal tumors such as colorectal cancers (1,2). The most significant result emerging from this study is probably the fact that the mutations detected in esophagal cancers abolish the transactivation of the p21, bax and PIG3 reporter systems. The p53 protein has a higher affinity for the p21 site than for the bax site (14,24,25) or the RGC site (14). Previous reports have shown that some p53 mutants have retained the ability to transactivate high affinity targets, such as the p21 gene (12–14). The yeast strains used in this study allowed us to easily show that such mutations with differential effects, if they exist, are not frequent in esophagal cancers. It could be argued that the analysis of the transcriptional activity of p53 has been performed in yeast and not in mammalian cells, which may have introduced some artefacts in the results. Nevertheless, our previous studies performed with yeast strains containing RGC (T. Frebourg and T. Soussi, unpublished data), p21 or bax reporters plasmids, have showed an excellent correlation between the analyses of the transcriptional activity of p53 performed in yeast and mammalian cells. Therefore, our results suggest that p53 mutations detected in esophagal cancers alter the control of the cell cycle and apoptosis. Considering the frequency of p53 mutations in esophagal cancers, this study indicates that p53 alteration, attributable to tobacco and alcohol, is a key event in esophagus carcinogenesis. Acknowledgements This work was supported by l’Association pour la Recherche sur le Cancer, La Ligue Nationale Contre le Cancer, Le Groupement des Entreprises Franc¸ aises dans la Lutte Contre le Cancer.

References 1. Greenblatt,M.S., Bennett,W.P., Hollstein,M. and Harris,C.C. (1994) Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res., 54, 4855–4878. 2. Hollstein,M., Sidransky,D., Vogelstein,B. and Harris,C.C. (1991) p53 mutations in human cancers. Science, 253, 49–53. 3. Levine,A.J. (1997) p53, the cellular gatekeeper for growth and division. Cell, 88, 323–331. 4. El-Deiry,W.S., Tokino,T., Velculescu,V.E., Levy,D.B., Parsons,R.,

Trent,J.M., Lin,D., Mercer,W.E., Kinzler,K.W. and Vogelstein,B. (1993) WAF1, a potential mediator of p53 tumor suppression. Cell, 75, 817–825. 5. Harper,J.W., Adami,G.R., Wei,N., Keyomarsi,K. and Elledge,S.J. (1993) The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclindependent kinases. Cell, 75, 805–816. 6. Bunz,F., Dutriaux,A., Lengauer,C., Waldman,T., Zhou,S., Brown,J.P., Sedivy,J.M., Kinzler,K.W. and Volgestein,B. (1998) Requirement for p53 and p21 to sustain G2 arrest after DNA damage. Science, 282, 1497–1501. 7. Miyashita,T. and Reed,J.C. (1995) Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell, 80, 293–299. 8. Jurgensmeier,J.M., Xie,Z., Deveraux,Q., Ellerby,L., Brendesen,D. and Reed,J.C. (1998) Bax directly induces release of cytochrome C from isolated mitochondria. Proc. Natl Acad. Sci. USA, 95, 4997–5002. 9. Polyak,K., Xia,Y., Zweier,J.L., Kinzler,K.W. and Vogelstein,B.A. (1997) Model for p53-induced apoptosis. Nature, 389, 300–305. 10. Hainaut,P., Hernandez,T., Robinson,A., Rodriguez-Tome,P., Flores,T., Hollstein,M., Harris,C.C. and Montesano,R. (1998) IARC Database of p53 gene mutations in human tumors and cell lines: updated compilation, revised formats and new visualisation tools. Nucleic Acids Res., 26, 205–213. 11. Beroud,C., Verdier,F. and Soussi,T. (1998) p53 gene mutation: software and database. Nucleic Acids Res., 26, 200–204. 12. Ludwig,R.L., Bates,S. and Vousden,K.H. (1996) Differential activation of target cellular promoters by p53 mutants with impaired apoptotic function. Mol. Cell. Biol., 16, 4952–4960. 13. Rowan,S., Ludwig,R.L., Haupt,Y., Bates,S., Lu,X., Oren,M. and Vousden,K.H. (1996) Specific loss of apoptotic but not cell-cycle arrest function in a human tumor derived p53 mutant. EMBO J., 15, 827–838. 14. Flaman,J.M., Robert,V., Lenglet,S., Moreau,V., Iggo,R. and Frebourg,T. (1998) Identification of human p53 mutations with differential effects on the bax and p21 promoters using functional assays in yeast. Oncogene, 16, 1369–1372. 15. Hollstein,M.C., Metcalf,R.A., Welsh,J.A., Montesano,R. and Harris,C.C. (1990) Frequent mutation of the p53 gene in human esophageal cancer. Proc. Natl Acad. Sci. USA, 87, 9958–9961. 16. Audrezet,M.P., Robaszkiewicz,M., Mercier,J.P., Nousbaum,J.B., Bail,J.P., Hardy,E., Volant,A., Lozac’h,P., Charles,J.F., Gouerou,H. and Ferec,C. (1993) TP53 gene mutation profile in esophageal squamous cell carcinomas. Cancer Res., 53, 5745–5749. 17. Sarbia,M., Porschen,R., Borchard,F., Horstmann,O., Willers,R. and Gabbert,H.E. (1994) p53 protein expression and prognosis in squamous cell carcinoma of the esophagus. Cancer, 74, 2218–2223. 18. Ribeiro,U., Finkelstein,S.D., Safatle-Ribeiro,A.V., Landreneau,R.J., Clark,M.R., Bakker,A., Swalsky,P.A., Gooding,W.E. and Posner,M.C. (1998) p53 sequence analysis predicts treatment response and outcome of patients with esophageal carcinoma. Cancer, 83, 7–18. 19. Montesano,R., Hollstein,M. and Hainaut,P. (1996) Genetic alterations in esophageal cancers and their relevance to etiology and pathogenesis: a review. Int. J. Cancer, 69, 2225–2235. 20. Shi,S.T., Yang,G.Y., Wang,L.D., Xue,Z., Feng,B., Ding,W., Xing,E.P. and Yang,C.S. (1999) Role of p53 gene mutations in human esophageal carcinogenesis: results from immunohistochemical and mutation analyses of carcinomas and nearby non-cancerous lesions. Carcinogenesis, 20, 591–597. 21. Flaman,J.M., Frebourg,T., Moreau,V., Charbonnier,F., Martin,C., Chappuis,P., Sappino,A.P., Limacher,J.M., Bron,L., Benhattar,J., Tada,M., Van Meir,E.G., Estreicher,A. and Iggo,R.D. (1995) A simple p53 functional assay for screening cell lines, blood and tumors. Proc. Natl Acad. Sci. USA, 92, 3963–3967. 22. Frebourg,T., Flaman,J.M., Estreicher,A. and Iggo,R.D. (1998) Functional assay of the p53 tumor suppressor gene. In Cotton,R.G.H., Edkins,E. and Forrest,S.M. (eds) Mutation Detection: A Practical Approach. Oxford University Press, Oxford, pp. 211–222. 23. Denissenko,M.F., Pao,A., Tang,M. and Pfeifer,G.P. (1996) Preferential formation of benzo[a]pyrene adducts at lung cancer mutational hotspots in p53. Science, 274, 430–432. 24. Friedlander,P., Haupt,Y., Prives,C. and Oren,M. (1996) A mutant p53 that discriminates between p53-responsive genes cannot induce apoptosis. Mol. Cell. Biol., 16, 4961–4971. 25. Thukral,S.K., Lu,Y., Blain,G.C., Harvey,T.S. and Jacobsen,V.L. (1995) Discrimination of DNA binding sites by mutant p53 proteins. Mol. Cell. Biol., 15, 5196–5202. Received May 11, 1999; revised November 15, 1999; accepted December 20, 1999

565