ANTICANCER RESEARCH 27: 1559-1564 (2007)

Association of and p21(CDKN1A/WAF1/CIP1) Polymorphisms with Oral in Taiwan Patients

DA-TIAN BAU1,4*, MING-HSUI TSAI2*, YEN-LI LO6, CHIN-MOO HSU1, YUHSIN TSAI4, CHENG-CHUN LEE1 and FUU-JEN TSAI3,5

Departments of 1Medical Research, 2Otolaryngology and 3Pediatrics, China Medical University Hospital, Taichung; 4Graduate Institute of Chinese Medical Science and 5College of Chinese Medicine, China Medical University, Taichung; 6Division of Biostatistics and Bioinformatics, National Health Research Institutes, Taiwan, R.O.C.

Abstract. Background: The tumor suppressor p53 and Oral cancer is one of the most commonly diagnosed its downstream effector p21(CDKN1A/WAF1/CIP1) are in the world, the fourth most common in Taiwan. Its rising thought to play major roles in the development of human incidence and mortality poses a formidable challenge to malignancy. Polymorphic variants of p53, at codon 72, and oncologists. Early premalignant oral lesions, such as CDKN1A, at codon 31, have been associated with cancer leukoplakia, appear as a white patch in the oral cavity of susceptibility, but few studies have investigated their effect on betel and tobacco consumers; five to ten percent progress oral cancer risk. Materials and Methods: In this hospital-based toward malignancy (1). Therefore, the identification of a case-control study, the association of p53 codon 72 and biomarker for screening the high-risk population for CDKN1A codon 31 polymorphisms with oral cancer risk in a increased predisposition to cancer is of utmost importance Taiwanese population were investigated. In total, 137 patients for primary prevention and early anticancer intervention. with oral cancer and 105 age-matched controls recruited from The p21(Cdkn1a/Waf1/Cip1), encoded by the the Chinese Medical Hospital in Central Taiwan were CDKN1A locus, is a universal inhibitor of cyclin-dependent genotyped. Results: We found a significant difference in the kinases (Cdks), which suggests its widespread role in frequency of the p53 genotype, but not the CDKN1A genotype, regulating the cycle. The human CDKN1A gene consists between the oral cancer and control groups. Those who had of three exons of 68, 450 and 1600 bp (2). In normal cells, Arg/Arg at p53 codon 72 showed a 2.68-fold (95% confidence p21 exists predominantly in quaternary complexes with interval=1.19-6.01) increased risk of oral cancer compared to cyclins, Cdks, and PCNA to inhibit the activity of Cdks and those with Pro/Pro. The distribution of the combination of p53 control the G1- to S-phase transition (3). The CDKN1A codon 72 and CDKN1A codon 31 was different in the oral gene has a p53 transcriptional regulatory motif and cells cancer and control groups. The percentages of three subgroups lacking functional p53 tumor suppressor protein express with the p53 GG homozygote were all higher in the oral cancer very low levels of p21, suggesting that p53 regulates group, and the risky double homozygote, p53/CDKN1A CDKN1A expression directly (4). The expression of p21 GG/CC form, was almost 9-fold higher than the control group. induces differentiation of normal and transformed cells, and Conclusion: Our findings suggest that the homozygous Arg the involvement of p21 in terminal differentiation has been allele of the p53 codon 72 may be associated with the observed in several cell systems (5, 6). Differential development of oral cancer and be a useful marker for primary regulation of p21 by p53 and retinoblastoma has been prevention and anticancer intervention. reported in cellular response to oxidative stress (7). In addition, several recent studies suggest a role for p21 in (8). Quercetin-induced apoptosis in hepatocytes was also associated with the regulation of p21 protein *Both authors contributed equally to this work. expression in a p53-independent pathway (9). In view of the central role of p21 in inducing growth Correspondence to: Fuu-Jen Tsai, MD, Ph.D., Department of arrest, terminal differentiation, or apoptosis, alterations in Pediatrics, Medical Genetics and Medical Research, China Medical the CDKN1A gene and its expression may play a vital role in University Hospital, 2 Yuh-Der Road, Taichung, 404 Taiwan, R.O.C. Tel: +88 6422052121 Ext 7080, Fax: +88 6422033295, the pathogenesis of cancer. Alterations in CDKN1A e-mail: [email protected] expression have been observed in a wide variety of human cancers, including ovarian, uterine, cervical, colorectal, Key Words: p53, p21, polymorphism, oral cancer, carcinogenesis. hepatocellular, and head and neck carcinomas (10-12).

0250-7005/2007 $2.00+.40 1559 ANTICANCER RESEARCH 27: 1559-1564 (2007)

Currently, knowledge of the biological and clinical relevance categorical variables. Information on these factors was identified of CDKN1A alterations in oral lesions is meager (13, 14). as established habits when occurring at least twice a week for years. Differential expression of p21 protein in tobacco-related Our study was approved by the Institutional Review Board of the China Medical University Hospital and written-informed consent oral tumorigenesis in the Indian population was reported was obtained from all participants. (15). As well known, in response to DNA damage, p21 is a key mediator of the G0-G1 cell cycle arrest induced by p53. Genotyping assays. Genomic DNA was prepared from peripheral Further study revealed that p21 also interacted with PCNA blood leukocytes using a QIAamp Blood Mini Kit (Blossom, Taipei, to cause both G1 and G2 cell cycle arrest in p53-deficient Taiwan) and further processed according to a previous paper (29). cells (5, 16, 17). A novel polymorphism of the CDKN1A Briefly, for p53 codon 72, the primers 5'-TCCCCCTT gene in codon 149 was found in the Indian population and GCCGTCCCAA-3' and 5'-CGTGCAAGTCACAGACTT-3' were used (30), and for CDKN1A codon 31, the primers 5'-GTCAGAA was considered as a marker of genetic susceptibility to CCGGCTGGGGATG-3' and 5'-CTCCTCCCAA CTCATCCCGG- esophageal and oral cancers (18, 19). Bioinformatics 3' were used (31). The following cycling conditions were performed: analysis revealed that the so-called polymorphism of codon one cycle at 94ÆC for 5 min; 35 cycles of 94ÆC for 20 sec, 58ÆC for 149 is not a susceptible site (20). 20 sec and 72ÆC for 20 sec; and a final extension at 72ÆC for 10 min. The p53 gene is a , whose function The PCR products were studied after digestion with BstU1 restriction is partially mediated by transactivating the CDKN1A enzyme for p53 codon 72, or with BlpI for CDKN1A codon 31. promoter which controls the cell cycle and prevents tumor Statistical analyses. To ensure that the controls used were formation (21). It was found that premalignant lesions and representative of the general population and to exclude the squamous cell carcinomas (SCCs) with mutant p53 protein possibility of genotyping error, deviation of the genotype overexpressed p21 (19). p21 immuno-positive well- frequencies of p53 codon 72 or CDKN1A codon 31 single differentiated tumors with p53 missense mutations probably polymorphism in the control subjects from those harbor a p21-dependent differentiation pathway activated expected under the Hardy-Weinberg equilibrium was assessed through a p53-independent mechanism (22). Taken using the goodness-of-fit test. Pearson's ¯2 test or Fisher's exact test together, these effects reflect the complexity of the p53/p21 (when the expected number in any cell was less than five) was used to compare the distribution of the p53 and CDKN1A genotypes pathways of cell cycle regulation and differentiation in the between cases and controls. Cancer risk associated with the pathogenesis of oral cancer. Mutations in either p53 or genotypes was estimated as odds ratio (ORs) and 95% confidence CDKN1A were detected in some tumor cells (21, 23) and intervals (CIs) using unconditional logistic regression with polymorphisms of p53 codon 72 or CDKN1A codon 31 were adjustment for age, smoking, alcohol consumption and betel found to be associated with many tumors (24-27). chewing habits. Data was recognized as significant when the Since p53 gene mutations are the most common cancer- statistical p-value was less than 0.05. related genetic alterations, found in ~50% of human cancer (28), and p53 regulates p21 expression (4), we were Results interested in looking for a susceptible site in p53 and CDKN1A together in Taiwan population, especially in The mean ages of the oral cancer patients and the controls oral cancer patients. Thus, the main goal of the study was were 53.27 (SD=12.27) and 53.02 (SD=10.08) years, to examine the joint effect of polymorphisms in codon 72 of respectively. The proportion of men and women were the p53 gene and codon 31 of the CDKN1A gene with oral almost the same in the patient and control groups (i.e., 117 cancer in a Central Taiwan population. men and 20 women). Characteristically, the most common sites of primary tumor were the tongue and buccal mucosa Materials and Methods in this Central Taiwan population. Approximately 38% (52 of 137) of the study cohort consisted of patients diagnosed Study population and sample collection. One hundred and thirty- with TNM late stage (III and IV) oral cancer (data not seven patients diagnosed with oral cancer were recruited at the shown). outpatient clinics of general surgery between 1997-2005 at the China Medical University Hospital, Taichung, Taiwan, Republic of The frequency of the alleles for the p53 codon 72 and China. All patients voluntarily participated, completed a self- CDKN1A codon 31 between oral cancer and control groups administered questionnaire and provided peripheral blood samples. is shown in Table I. The Arg allele at p53 codon 72 was One hundred and five non-oral cancer healthy people as controls significantly associated with oral cancer risk (p=0.02). In were selected by matching for age and gender after initial random contrast, Arg or Ser at CDKN1A codon 31 was not sampling from the Health Examination Cohort, which consisted of differently distributed in the oral cancer patient and control people who voluntarily visited the health-screening clinic at the groups (p>0.05). same hospital. The questionnaire administered included questions on alcohol consumption, betel chewing habits, as well as smoking The frequency of the genotype of p53 codon 72 and history and frequency. Self-reported alcohol consumption, betel CDKN1A codon 31 polymorphisms in the oral cancer and chewing and smoking habits were evaluated and classified as control groups is shown in Table II. Using 72Pro as the

1560 Bau et al: p53 and p21 Polymorphisms in Oral Cancer

Table I. Allele frequencies for p53 codon 72 and CDKN1A codon 31 Table II. Association of p53 codon 72 and CDKN1A codon 31 polymorphisms in the oral cancer and control groups. polymorphisms and oral cancer risk.

Allele Cases (%) Controls (%) p-valuea Genotype Cases Controls Crude OR Adjusted OR N=274 N=210 (%) (%) (95% CI)a (95% CI)b p53 codon 72 p53 Allele C (Pro) 112 (40.9) 109 (51.9) 0.020 Pro/Pro 21 (15.3) 22 (21.0) 1.00 (ref) 1.00 (ref) Allele G (Arg) 162 (59.1) 101 (48.1) Pro/Arg 70 (51.1) 65 (61.9) 1.13 (0.57-2.24) 1.09 (0.78-2.04) CDKN1A codon 31 Arg/Arg 46 (33.6) 18 (17.1) 2.68 (1.19-6.01)c 2.76 (1.44-5.63)c Allele A (Arg) 125 (45.6) 107 (51.0) 0.284 with Arg 116 (84.7) 83 (79.0) 1.46 (0.76-2.84) 1.53 (0.89-2.63) Allele C (Ser) 149 (54.4) 103 (49.0) CDKN1A Arg/Arg 26 (19.0) 21 (20.0) 1.00 (ref) 1.00 (ref) ap-value based on ¯2 test. Aer/Ser 73 (53.3) 65 (61.9) 0.91 (0.47-1.76) 1.03 (0.74-1.81) Ser/Ser 38 (27.7) 19 (18.1) 1.62 (0.73-3.58) 1.64 (0.94-4.27) with Ser 111 (81.0) 84 (80.0) 1.07 (0.56-2.03) 1.21 (0.79-2.36)

aOR, odds ratio; CI, confidence interval. reference group, there was an obvious association between bAdjusted for age and environmental habits (betel quid chewing, the homozygotes and heterozygotes of 72Arg of p53 and smoking and alcohol consumption) c oral cancer risk. After controlling age, smoking, betel p<0.05. chewing and alcohol consumption, the adjusted OR was still significant (Table II). Neither hetero- nor homozygotes of Table III. Distribution of combinations of p53 codon 72 and CDKN1A 31Ser of CDKN1A seemed to be risky genotypes for oral codon 31 polymorphisms in the oral cancer and control groups. cancer (p>0.05). Since p53 and CDKN1A may be closely related to each Genotype Cases (%) Controls (%) p-valuea other in the same pathway, the gene – gene interaction was N=137 N=105 also investigated. The results of analysis of the genotype P53/CDKN1A 0.012 combinations of p53 codon 72 and CDKN1A codon 31 are GG/AA 8 (5.8) 4 (3.8) shown in Table III; there was a significant difference GG/AC 27 (19.7) 13 (12.4) between oral cancer and control groups (p=0.012). The GG/CC 11 (8.0) 1 (0.9) percentages of three subgroups with p53 GG were all higher GC/AA 12 (8.8) 13 (12.4) in the oral cancer group and the risky double homozygote GC/AC 38 (27.7) 39 (37.1) GC/CC 20 (14.6) 13 (12.4) (p53/CDKN1A=GG/CC) was almost 9-fold higher than the CC/AA 6 (4.4) 4 (3.8) control group (Table III). The results in Table III again CC/AC 8 (5.8) 13 (12.4) show a major role of p53 codon Arg72 and a minor role of CC/CC 7 (5.1) 5 (4.8) CDKN1A codon Ser31 in oral carcinogenesis (Table I). ap-value based on ¯2 test. Discussion

Cell proliferation and death are essential aspects in the that the majority (76%) of female papillomavirus-associated understanding of carcinogenesis. Considerable evidence cancer patients were Arg homozygotes while only 37% of now links the activities of the p53 gene to regulation of the the controls were Arg homozygotes (30). In contrast, a few cell cycle and mutations in this gene are the most common studies have reported the Pro allele to be a risky genotype. genetic changes known to occur in human cancers (28). In The Pro allele is associated with a 1.37- to 11.29-fold higher this study, it was found participants homozygous for p53 risk for (34-36), a 3.7-fold higher risk for codon 72 Arg had a 2.68-fold higher risk of oral cancer nasopharyngeal carcinoma (37), and an 11-year earlier age (Table II). As for the Arg/Pro heterozygotes, there was only of onset for oral cancers in a non-Hispanic white population a 1.16-fold increased risk. The data suggested that 72Arg (38). There is controversy about the role of the p53 codon homozygotes were much more susceptible to oral cancer 72 genotype in various cancers. risk than heterozygotes or 72Pro homozygotes. Our findings The p53Arg72 and p53Pro72 do not differ in are consistent with those of other studies which reported the their ability to bind to DNA in a sequence-specific manner Arg allele to be a risky genotype. The Arg allele has been but do differ in other ways. The p53Arg72 protein induces reported to be associated with a 4.69-fold increased risk for apoptosis faster and suppresses transformation more (32) and 3.1-fold higher risk for gastric efficiently than the p53Pro72 protein (39). Conversely, the cardia adenocarcinoma (33). In addition, it was reported p53Arg72 protein is more susceptible than p53Pro72 protein

1561 ANTICANCER RESEARCH 27: 1559-1564 (2007) to HPV E6 protein-mediated degradation and this groups, and SY Lin for her technical assistance. This study was degradation is correlated with increased risk of HPV- supported by research grants from China Medical University associated cancers (30). Mutations of p53 were detected in Hospital (DMR-95-097) and National Science Council (NSC 95- 2320-B-039-014-MY3). about one-third of oral cancer patients (40, 41) and preferentially occurred at the 72Arg allele (40). References There is accumulating evidence that p53 genetic alteration is closely associated with oral carcinogenesis. 1 Daftary DK, Murti PR, Bhonsle RR, Gupta PC, Mehra FS and Mutations of the p53 gene were found in 35% of oral Pindborg JJ: Risk factors and risk markers for oral cancer in cancer cases, suggesting that mutation of p53 is involved in high risk areas of the world. In: Oral Cancer: The Detection of the carcinogenesis of the oral cavity (41). It was also found Patients and Lesions at Risk. Johnson NW (ed.). Cambridge, that p53 protein was overexpressed in most p53-mutated United Kingdom, Cambridge University Press, pp 29-63, 1991. oral cancer cases (41). Even the cases of oral tumor cells 2 Chen J, Jackson PK, Kirschner MW and Dutta A: Separate domains of p21 involved in the inhibition of Cdk kinase and with no p53 gene mutation also overexpressed p53 protein, PCNA. Nature 374: 386-388, 1995. but the expression of p21 was unrelated to p53 status in 3 Xiong Y, Zhang H and Beach D: Subunit rearrangement of the oral cancer (41). Genetic alteration in another region of the cyclin-dependent kinases is associated with cellular exon, such as the promoter and intron of the p53 gene, transformation. Genes Dev 7: 1572-1583, 1993. could result in a high expression of wild-type p53 (42, 43). 4 El-Deiry WS, Harper JW, O'Connor PM, Velculescu VE, Another possibility is that binding of the wild-type p53 Canman CE, Jackman J, Pietenpol JA, Burrell M, Hill DE and protein to other proteins, such as MDM2 or E6 viral Wang Y: WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis. Cancer Res 54: 1169-1174, 1994. oncoprotein, can stabilize and inactivate it resulting in 5 Jiang H, Lin J, Su ZZ, Collart FR, Huberman E and Fisher PB: overexpression (44, 45-47). Induction of differentiation in human promyelocytic HL-60 The p21 protein inhibits two different targets, the cyclin- cells activates p21, WAF1/CIP1, expression in the CDK complexes and PCNA, which control cell cycle absence of p53. 9: 3397-3406, 1994. transitions and DNA replication. Two particularly conserved 6 Michieli P, Chedid M, Lin D, Pierce JH, Mercer WE and Givol regions in the human CDKN1A gene were found on 60 D: Induction of WAF1/CIP1 by a p53-independent pathway. amino acids near codon 21 and 164 amino acids near codon Cancer Res 54: 3391-3395, 1994. 130 (48). Our experimental result has shown that there was 7 Yin Y, Solomon G, Deng C and Barrett JC: Differential regulation of p21 by p53 and Rb in cellular response to no significance of codon 31 of CDKN1A in the Taiwanese oxidative stress. Mol Carcinog 24: 15-24, 1999. population we surveyed and it is not a marker of 8 Zhang Y, Fujita N and Tsuruo T: -mediated cleavage susceptibility for oral carcinogenesis (Table II). These data of p21Waf1/Cip1 converts cancer cells from growth arrest to may not exclude the possibility that CDKN1A may play a undergoing apoptosis. Oncogene 18: 1131-1138, 1999. role in oral cancer, neither can all the other mediators of 9 Iwao K and Tsukamoto I: Quercetin inhibited DNA synthesis p53 function, such as p27 and alpha-catenin, be ignored in and induced apoptosis associated with increase in c-fos mRNA WAF1CIP1 further study. The genetic polymorphisms in the PCNA level and the upregulation of p21 mRNA and protein expression during liver regeneration after partial hepatectomy. binding motif, located between codons 144±151 (2, 49), Biochim Biophys Acta 1427: 112-120, 1999. need further investigation. PCNA is essential for DNA 10 El-Deiry WS, Tokino T, Waldman T, Oliner JD, Velculescu replication and repair, acting as an auxiliary factor for DNA VE, Burrell M, Hill DE, Healy E, Rees JL and Hamilton SR: polymerase ‰ and Â, facilitating the loading of these Topological control of p21WAF1/CIP1 expression in normal and polymerases onto DNA templates and increasing their neoplastic tissues. Cancer Res 55: 2910-2919, 1995. activity in both DNA replication and repair (50). If there 11 Elbendary AA, Cirisano FD, Evans AC, Jr-Davis PL, Iglehart were any significant SNPs in the PCNA-binding motif of JD, Marks JR and Berchuck A: Relationship between p21 expression and mutation of the p53 tumor suppressor gene in p21, the structural changes in the protein product might normal and malignant ovarian epithelial cells. Clin Cancer Res affect the binding between p21 and PCNA. As a 2: 1571-1575, 1996. consequence, these significant SNPs (such as p53 codon 72), 12 Erber R, Klein W, Andl T, Enders C, Born AI, Conradt C, combined with other biomarkers, will be useful for Bartek J and Bosch FX: Aberrant p21(CIP1/WAF1) protein screening, prediction and anticancer gene therapy of oral accumulation in head-and-neck cancer. Int J Cancer 74: 383- carcinogenesis. 389, 1997. 13 Heinzel PA, Balaram P and Bernard HU: Mutations and polymorphisms in the p53, p21 and p16 genes in oral carcinomas Acknowledgements of Indian betel quid chewers. Int J Cancer 68: 420-423, 1996. 14 Facher EA, Becich MJ, Deka A and Law JC: Association We thank Dr. WK Yang for his critical review of the manuscript between human cancer and two polymorphisms occurring and his precious scientific advice and English editing, Dr. WC together in the p21Waf1/Cip1 cyclin-dependent kinase inhibitor Chen for his assistance in recruiting and recording the control gene. Cancer 79: 2424-2429, 1997.

1562 Bau et al: p53 and p21 Polymorphisms in Oral Cancer

15 Agarwal S, Mathur M, Shukla NK and Ralhan R: Expression of 30 Storey A, Thomas M, Kalita A, Harwood C, Gardiol D, cyclin dependent kinase inhibitor p21waf1/cip1 in premalignant Mantovani F, Breuer J, Leigh IM, Matlashewski G and Banks and malignant oral lesions: relationship with p53 status. Oral L: Role of a p53 polymorphism in the development of human Oncol 34: 353-360, 1998. papillomavirus-associated cancer. Nature 393: 229-234, 1998. 16 Sheikh MS, Li XS, Chen JC, Shao ZM, Ordonez JV and Fontana 31 Li YJ, Laurent-Puig P, Salmon RJ, Thomas G and Hamelin JA: Mechanisms of regulation of WAF1/Cip1 in R: Polymorphisms and probable lack of mutation in the human breast carcinoma: role of p53-dependent and independent WAF1-CIP1 gene in colorectal cancer. Oncogene 10: 599-601, pathways. Oncogene 9: 3407-3415, 1994. 1995. 17 Zeng YX and El-Deiry WS: Regulation of p21WAF1/CIP1 32 Soulitzis N, Sourvinos G, Dokianakis DN and Spandidos DA: expression by p53-independent pathways. Oncogene 12: 1557- p53 codon 72 polymorphism and its association with bladder 1564, 1996. cancer. Cancer Lett 179: 175-183, 2002. 18 Bahl R, Arora S, Nath N, Mathur M, Shukla NK and Ralhan 33 Zhang ZW, Newcomb P, Hollowood A, Feakins R, Moorghen R: Novel polymorphism in p21(waf1/cip1) cyclin dependent M, Storey A, Farthing MJ, Alderson D and Holly J: Age- kinase inhibitor gene: association with human esophageal associated increase of codon 72 arginine p53 frequency in cancer. Oncogene 19: 323-328, 2000. gastric cardia and non-cardia adenocarcinoma. Clin Cancer Res 19 Ralhan R, Agarwal S, Mathur M, Wasylyk B and Srivastava A: 9: 2151-2156, 2003. Association between polymorphism in p21(Waf1/Cip1) cyclin- 34 Wu X, Zhao H, Amos CI, Shete S, Makan N, Hong WK, dependent kinase inhibitor gene and human oral cancer. Clin Kadlubar FF and Spitz MR: p53 Genotypes and haplotypes Cancer Res 6: 2440-2447, 2000. associated with lung cancer susceptibility and ethnicity. J Natl 20 Xi YG, Ding KY, Ren YH, Shen Y and Ke Y: Esophageal Cancer Inst 94: 681-690, 2002. cancer in a Chinese population: no polymorphism in codon 149 35 Fan R, Wu MT, Miller D, Wain JC, Kelsey KT, Wiencke JK of P21(Waf1/Cip1) cyclin dependent kinase gene. Oncogene 21: and Christiani DC: The p53 codon 72 polymorphism and lung 7745-7748, 2002. cancer risk. Cancer Epidemiol Biomarkers Prev 9: 1037-1042, 21 Greenblatt MS, Bennett WP, Hollstein M and Harris CC: 2000. Mutations in the p53 tumor suppressor gene: clues to cancer 36 Jin X, Wu X, Roth JA, Amos CI, King TM, Branch C, Honn etiology and molecular pathogenesis. Cancer Res 54: 4855-4878, SE and Spitz MR: Higher lung cancer risk for younger African- 1994. Americans with the Pro/Pro p53 genotype. Carcinogenesis 16: 22 Sato M, Kawamata H, Harada K, Nakashiro K, Ikeda Y, Gohda 2205-2208, 1995. H, Yoshida H, Nishida T, Ono K, Kinoshita M and Adachi M: 37 Tsai MH, Lin CD, Hsieh YY, Chang FC, Tsai FJ, Chen WC Induction of cyclin-dependent kinase inhibitor, p21WAF1, by and Tsai CH: Prognostic significance of the proline form of p53 treatment with 3,4-dihydro-6-[4-(3,4)-dimeth-oxybenzoyl)-1- codon 72 polymorphism in nasopharyngeal carcinoma. piperazinyl]-2(1H)-quinoline (vesnarinone) in a human salivary Laryngoscope 112: 116-119, 2002. cancer cell line with mutant p53 gene. Cancer Lett 112: 181-189, 38 Shen H, Zheng Y, Sturgis EM, Spitz MR and Wei Q: p53 1997. codon 72 polymorphism and risk of squamous cell carcinoma of 23 Hollstein M, Rice K, Greenblatt MS, Soussi T, Fuchs R, Sorlie the head and neck: a case-control study. Cancer Lett 183: 123- T, Hovig E, Smith-Sorensen B, Montesano R and Harris CC: 130, 2002. Database of p53 gene somatic mutations in human tumors and 39 Thomas M, Kalita A, Labrecque S, Pim D, Banks L and cell lines. Nucleic Acids Res 22: 3551-3555, 1994. Matlashewski G: Two polymorphic variants of wild-type p53 24 Lai KC, Chen WC, Jeng LB, Li SY, Chou MC and Tsai FJ: differ biochemically and biologically. Mol Cell Biol 19: 1092- Association of genetic polymorphisms of MK, IL-4, p16, p21, 1100, 1999. p53 genes and human gastric cancer in Taiwan. Eur J Surg 40 Schneider-Stock R, Mawrin C, Motsch C, Boltze C, Peters B, Oncol 31: 1135-1140, 2005. Hartig R, Buhtz P, Giers A, Rohrbeck A, Freigang B and 25 Jones JS, Chi X, Gu X, Lynch PM, Amos CI and Frazier ML: Roessner A: Retention of the arginine allele in codon 72 of the p53 polymorphism and age of onset of hereditary nonpolyposis p53 gene correlates with poor apoptosis in head and neck colorectal cancer in a Caucasian population. Clin Cancer Res cancer. Am J Pathol 164: 1233-1241, 2004. 10: 5845-5849, 2004. 41 Yook JI and Kim J: Expression of p21WAF1/CIP1 is unrelated to 26 Huang SP, Wu WJ, Chang WS, Wu MT, Chen YY, Chen YJ, p53 tumour suppressor gene status in oral squamous cell Yu CC, Wu TT, Lee YH, Huang JK and Huang CH: p53 codon carcinomas. Oral Oncol 34: 198-203, 1998. 72 and p21 codon 31 polymorphisms in prostate cancer. Cancer 42 Beenken SW, Karsenty G, Raycroft L and Lozano G: An intron Epidemiol Biomarkers Prev 13: 2217-2224, 2004. binding protein is required for transformation ability of p53. 27 Roh JW, Kim JW, Park NH, Song YS, Park IA, Park SY, Kang Nucleic Acids Res 19: 4747-4752, 1991. SB and Lee HP: p53 and p21 genetic polymorphisms and 43 Barnes DM, Hanby AM, Gillett CE, Mohammed S, Hodgson susceptibility to endometrial cancer. Gynecol Oncol 93: 499- S, Bobrow LG, Leigh IM, Purkis T, MacGeoch C and Spurr 505, 2004. NK: Abnormal expression of wild-type p53 protein in normal 28 Hollstein M, Sidransky D, Vogelstein B and Harris CC: p53 cells of a cancer family patient. Lancet 340: 259-263, 1992. mutations in human cancers. Science 253: 49-53, 1991. 44 Zambetti GP and Levine AJ: A comparison of the biological 29 Tzai TS, Tsai YS and Chow NH: The prevalence and activities of wild-type and mutant p53. Faseb J 7: 855-865, 1993. clinicopathologic correlate of p16INK4a, retinoblastoma and p53 45 Chen J, Wu X, Lin J and Levine AJ: Mdm-2 inhibits the G1 immunoreactivity in locally advanced urinary bladder cancer. arrest and apoptosis functions of the p53 tumor suppressor Urol Oncol 22: 112-118, 2004. protein. Mol Cell Biol 16: 2445-2452, 1996.

1563 ANTICANCER RESEARCH 27: 1559-1564 (2007)

46 Momand J, Zambetti GP, Olson DC, George D and Levine AJ: 49 Goubin F and Ducommun B: Identification of binding domains The mdm-2 oncogene product forms a complex with the p53 on the p21Cip1 cyclin-dependent kinase inhibitor. Oncogene 10: protein and inhibits p53-mediated transactivation. Cell 69: 2281-2287, 1995. 1237-1245, 1992. 50 Funk JO, Waga S, Harry JB, Espling E, Stillman B and 47 Scheffner M, Werness BA, Huibregtse JM, Levine AJ and Galloway DA: Inhibition of CDK activity and PCNA-dependent Howley PM: The E6 oncoprotein encoded by human DNA replication by p21 is blocked by interaction with the papillomavirus types 16 and 18 promotes the degradation of HPV-16 E7 oncoprotein. Genes Dev 11: 2090-2100, 1997. p53. Cell 63: 1129-1136, 1990. 48 Huppi K, Siwarski D, Dosik J, Michieli P, Chedid M, Reed S, Mock B, Givol D and Mushinski JF: Molecular cloning, Received October 19, 2006 sequencing, chromosomal localization and expression of mouse Revised January 17, 2007 p21 (Waf1). Oncogene 9: 3017-3020, 1994. Accepted January 22, 2007

1564