Expression of p21WAF1/CIP1 is unrelated to p53 tumoursuppressor gene status in oral squamous cell carcinomas

J.I. Yook, J. Kim*Department of Oral Pathology, Institute of Oral Cancer Research, College of Dentistry, Yonsei University, Seodaemoon-ku, Shinchon-dong 134,

120-752, Seoul, Korea

Received 23 July 1997; accepted 13 October 1997

Abstract

The p53 tumour suppressor gene is frequently mutated in oral squamous cell carcinomas. However, the downstream mechanismof p53 during oral carcinogenesis is not fully understood. The cyclin-dependent kinase inhibitor p21WAF1/CIP1 (p21), which can

be induced by wild-type p53, functions as a downstream mediator of the antiproliferative and apoptosis-inducing actions of wild-type p53. To learn more about the roles of the p53 gene and its downstream mechanism, we evaluated p53 gene mutation andimmunohistochemical expression of p53 and p21 in 20 cases of oral squamous cell carcinoma. p53 gene mutations were observed in7 cases (35%). Overexpression of p53 was found in 4 of 13 cases with wild-type p53, and in 6 of 7 cases with p53 mutations. p21

expression was detected in 15 of 20 cases (75%). The expression of p21 correlated neither with mutated p53 mutation nor with p53protein overexpression. p21 was expressed even in carcinomas in which molecular analysis revealed a nonsense mutation. In normaloral mucosa, p21 expression was limited in the di�erentiating spinous cell layer. However, dysplastic or hyperplastic epithelium

adjacent to the tumour demonstrated the increased expression of p21 even in the proliferating basal cell layer. These molecular andimmunohistochemical data did not show any correlation with various clinico-pathologic parameters. These results suggest that p53gene mutations and altered expression of p21 are commonly involved in oral carcinogenesis, but do not correlate with each other or

with the clinico-pathologic parameters. They also suggest that p21 expression in oral squamous cell carcinomas may be induced bya p53-independent pathway. # 1998 Elsevier Science Ltd. All rights reserved.

Keywords: Oral squamous cell carcinomas; p53 tumour suppressor gene; p21WAF1/CIP1; Mutation; Overexpression; PCR-SSCP; Immunohis-

tochemistry

1. Introduction

The inactivation of the tumour suppressor gene isimportant during multistage carcinogenesis. Althoughmutation in a highly conserved region of the p53 gene isa frequent event in a wide variety of human malig-nancies [1], the precise manner in which p53 actioncontributes to suppression of tumorigenesis is not fullyunderstood. According to extensive previous studies,the p53 gene has been implicated in a variety of cellularprocesses, including G1 arrest, apoptosis, senescence,and di�erentiation [2].

Since wild-type p53 acts as a transcription factor andcontrols expression of other genes, downstream medi-ators of the p53 gene have been extensively studied [2±4].p21WAF1/CIP1, which has been described as the crit-ical downstream mediator of wild-type p53, is known to

suppress DNA replication and arrest the G1 cell cycleby the quaternary complex with cyclin D, cyclin-depen-dent kinase (cdk) and proliferating cell nuclear antigen(PCNA) [3,5±8]. At least two wild-type p53-responsiveelements exist in the promoter region of p21, suggestingthat the antiproliferative function of the p53 gene actsthrough the expression of p21WAF1/CIP1 [9].

The mutation or overexpression of the p53 tumoursuppressor gene are frequently found in squamous cellcarcinoma of the head and neck [10±12]. However, thedownstreammechanismof the p53 gene in head-and-neckcarcinoma is not well known. Therefore, the study ofp21WAF1/CIP1 expression in association with themutation of p53 may contribute to the understanding ofthe head-and-neck carcinogenesis. In this study we exam-ined the mutation of the p53 gene in oral squamous cellcarcinomas by PCR-SSCP and direct sequencing ana-lyses, and its association with immunohistochemicalexpression of p53 and p21WAF1/CIP1. Our purpose

ORAL

ONCOLOGY

Oral Oncology 34 (1998) 198±203

1368-8375/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved

PII: S1368-8375(97)00091-2

* Corresponding author.

was to explore the roles of the p53 gene and p21WAF1/CIP1 expression as a downstream mediator of p53 inoral carcinogenesis.

2. Materials and methods

Tumour tissues of oral squamous cell carcinoma wereobtained from the ®les of the Department of OralPathology at the College of Dentistry, Yonsei Uni-versity between 1991 and 1995. The patients included 14men and 6 women, aged between 36 and 74 years. Noinitial chemotherapy or radiation therapy was per-formed prior to tumour excision. Clinical data (age, sex,smoking history) were obtained from patient medicalrecords. All patients were retrospectively staged usingthe 1988 AJCC criteria [13]. All specimens were ®xed in10% neutral formalin, processed routinely, and embed-ded in para�n.

2.1. DNA extraction and polymerase chain reaction

DNA was extracted from para�n sections aftermicrodissection using proteinase K and phenol-chloro-form. p53 exons 5±9 were each ampli®ed in 20 ml volumewith 100 ng genomic DNA and 2 pM primers, in 10mMTris±HCl (pH 8.3), 50mM KCl, 1.8mM MgCl2,0.001% gelatin, 2mM deoxyribonucleoside triphosphatecontaining 1.0U Taq polymerase (Perkin-Elmer CetusCorp., Norwalk, CT, U.S.A.). PCR conditions were 35cycles of 94�C denaturation for 1min, 58±62�C anneal-ing for 1min, and 72�C extension for 45 s in an auto-mated thermal cycler. Primers for PCR and sequencingwere determined based on reported nucleotide sequenceinformation [14]. To control against DNA contamina-tion of PCR reactions, all experiments included onereaction tube in which no DNA was added. The primersequences and PCR fragment sizes were as follows:exon 5 (245 bps): 50 TTCCTCTTCCTGCAGTACTC 30

(sense) 50 ACCTGGGCAACCAGCCCTGT 30 (anti-sense); exon 6 (175 bps): 50 ACAGGGCTGGTTGCC-CAGGGT 30 (sense), 50 AGTTGCAAACCAGACCT-CAG 30 (antisense); exon 7 (190 bps): 50 GTGTTGTCT-CCTAGGTTGGC 30 (sense) 50 GTCAGAGGCAA-GCAGAGGCT 30 (antisense); exon 8 (212 bps): 50

TATCCTGAGTAGTGGTAATC 3 (sense) 50 AAGT-GAATCTGAGGCATAAC 30 (antisense); exon 9(138 bps): 50 GCAGTTATGCCTCAGATTCAC 30

(sense) 50 AAGACTTAGTACCTGAAGGGT 30 (anti-sense).

2.2. Single strand conformation polymorphism

SSCP analysis was performed using a modi®ed non-radioisotopic method [15,16]. Brie¯y, PCR product(1 ml) was diluted with 9ml of denaturing solution (95%formamide, 20mM EDTA, 0.25% bromophenol blue,

0.25% xylene cyanol) and 1 ml of 0.5N NaOH. Sampleswere then heated in boiling water for 5min, chilled in anice±ethanol bath, and immediately loaded onto a 15%acrylamide±TBE gel with or without 10% glycerol. Thesize of the gel was 80�80�1mm. Electrophoresis wasthen performed at a constant voltage of 200V underthermal controlling. The gels were stained with a silver-staining kit (BioRad, CA, U.S.A.) according to themanufacturer's recommendations. All mutations werecon®rmed by a complete repeat of the experimentalprocedure. HT3 (codon 245 of exon 7: GGC!GTC)and Caski (wild-type p53 gene) cervical cell lines wereused as a positive and negative control for SSCP analy-sis, respectively [17].

2.3. Direct sequencing of ampli®ed DNA

80 ml of PCR-ampli®ed DNA fragments whichshowed bands with altered mobilities by SSCP analysiswere puri®ed on 1.5% agarous gels using a Gene Cleankit (Bio 101, CA, U.S.A.) and sequenced directly with-out any intermediate cloning steps. Template DNA wasannealed with 25 nmol of primer and sequenced in bothdirections by the dideoxy chain-termination methodwith a T7 DNA sequencing kit (Pharmacia, CA,U.S.A.). Radioactive label incorporation was achievedby a 5-min preincubation at room temperature of thesequencing reaction containing the a-35S-labeled deox-ynucleotide (Amersham, U.K.). The reaction sampleswere denatured in boiling water for 3min and electro-phoresed in 8M urea±8% polyacrylamide gel for 1±3 h.Dried gels were exposed to Kodak X-AR 5 ®lm at roomtemperature for 1±3 days.

2.4. Immunohistochemical staining

Immunohistochemical staining for p53 and p21 wasperformed on formalin-®xed, para�n-embedded sec-tions and a heat-based antigen enhancement system thatutilized citrate bu�er (pH=6.0). Endogenous perox-idase was inhibited with 3% hydrogen peroxide andnon-speci®c binding blocked with normal horse serum.The monoclonal antibody DO-7 (Novocastra, New-castle, U.K.) was used in studies of p53 protein expres-sion. Antibody to p21WAF1/CIP1 was obtained fromSanta Cruz Biotechnology (sc-187, CA, U.S.A.). Thedilution and incubation time of primary antibody weredetermined on the basis of preliminary studies(1:150±200 for 2 h at room temperature) so as to yield amaximal signal. Staining was achieved using a biotin-conjugated goat anti-mouse secondary antibody (Vec-tor) and the ABC horseradish peroxidase method(Vector, CA, U.S.A.). p53 staining was nuclear and p21staining was reproducibly strong in vascular smoothmuscle cells, which served as positive internal controls[18]. Immunostaining for p53 and p21 were scored asabsent, focal or strong.

J.I. Yook, J. Kim/Oral Oncology 34 (1998) 198±203 199

3. Results

3.1. Mutations of p53 gene in oral squamous cellcarcinoma

Altered mobility of PCR-SSCP analysis and muta-tions in the p53 ampli®ed sample were identi®ed in 7 ofthe 20 tumours (35%). The results of SSCP-directsequencing with clinico-pathologic status are summar-ized in Table 1. 4 of the 7 mutations occurred in exon 6and one each in exon 5, 7 and 8. Each mutation resultedeither in an amino acid change in the p53 protein or in astop codon. In sequencing analysis, the wild-typesequence was also present, indicating either that thenormal allele was retained in the tumour or that the tum-our tissue sample contained a signi®cant number ofstromal cells (Fig. 1). There was no correlation betweenp53 mutations and smoking history. Positive control(HT3) and negative control (Caski) of the p53 gene inSSCP showed the expected pattern under di�erent con-ditions (data not shown).

3.2. Immunohistochemical detection of p53 andp21WAF1/CIP1 protein in oral squamous cellcarcinomas

p53 and p21WAF1/CIP1 expression with the corre-sponding p53 mutational status for each of the 20 casesof oral squamous cell carcinoma are summarized inTable 2. p53 overexpression was seen in 10 of 20 oralsquamous cell carcinomas, and it was limited to tumourcell nuclei (Fig. 2). Among 7 cases showing p53 genemutation, all of them demonstrated immunohistochem-ical expression, but one case of a nonsense mutation didnot demonstrate p53 accumulation. Moreover, 4 casesshowing a wild-type p53 gene showed di�use or focalexpression of the p53 protein. Among the 10 casesshowing p53 expression, only 6 cases contained adjacentmucosa, and 3 of those 6 cases showed p53 expression inthe adjacent mucosa in the basal layer. Moreover, thesurrounding mucosa in 2 cases which did not showp53 overexpression in the tumour revealed weak posi-tive reactivity in the basal cell layer. The HT3 cell line

was consistently positive to p53 antibody and Caski wasnegative.

p21WAF1/CIP1 staining was reproducibly strong invascular smooth muscle and ductal simple epithelium ofthe minor salivary gland, which served as internal posi-tive controls. 15 of 20 carcinomas demonstrated theexpression of p21WAF1/CIP1. Most of the p21WAF1/CIP1 positive expressions were localized in the nucleiof tumour cells (Fig. 3). The correlation betweenp21WAF1/CIP1 expression and p53 mutation or expres-sion is summarized in Table 3. Of 13 tumours whichwere wild-type for p53, 9 showed immunoreactivity for

Table 1

p53 gene mutations in oral squamous cell carcinomas

Case no. Age/sex Primary state Histologic

differentiation

TNM stage Smoking history Mutated codon

(exon)

Base change Amino acid

change

3 50/F Tongue W I No 206 (VI) TIG!ATG Leu!Met

4 36/M Givgiva W IV 15PY 213 (VI) CGA!CTA Arg!Leu

7 62/F Gingiva M IV No 143 (V) GTG!ATG Val!Met

9 59/M Tongue M III No 196 (VI) CGA!TGA Arg!Stop

12 51/M Mouth floor W IV No 260 (VII) TCC!ACC Ser!Thr

16 53/M Soft plate W III 30PY 278 (VIII) CCT!ACT Pro!Thr

19 57/M Mouth floor W III 10PY 213 (VI) CGA!CAA Arg!Gln

W, well di�erentiated; M, moderately di�erentiated; PY pack/day� years.

Fig. 1. Demonstration of point mutations in the p53 gene by direct

sequencing. A: TCC!ACC transversion in codon 260 in exon 7

resulting in Ser!Thr substitution. B: GTG!ATG transition in codon

143 in exon 5 in case 7 resulting in a Val!Met substitution.

Fig. 2. Immunohistochemical staining of oral squamous cell carcino-

mas with anti-p53 antibody DO-7. Neoplastic tissue shows strong

nuclear staining, whereas adjacent normal stromal tissue shows no

reaction product. �100.

200 J.I. Yook, J. Kim/Oral Oncology 34 (1998) 198±203

p21WAF1/CIP1, and 6 of 7 cases with demonstratedp53 mutations stained similarly for 21WAF1/CIP1.Even the single case with stop codon showedp21WAF1/CIP1 expression, while 4 cases with wild-type p53 gene displayed negative for p21WAF1/CIP1.

In normal mucosal epithelium, p21WAF1/CIP1staining was con®ned to the spinous cell layer, not tothe basal cell layer (Fig. 4). Interestingly, strong expres-sion of p21WAF1/CIP1 was seen in the entire layer,even in the basal cell layer, of adjacent hyperplastic ordysplastic mucosa to the squamous cell carcinomaswhile these cells were not expressed in p53 (Fig. 4).

4. Discussion

Cell proliferation and death are essential aspects inthe understanding of carcinogenesis. Considerableevidence now links the activities of the p53 gene to reg-ulation of the cell cycle and mutations in this gene arethe most common genetic changes known to occur inhuman cancers [1]. In this experiment, we found thatp53 gene mutations were found in 35% of our cases,and on the basis of this analysis we suggest that muta-tion of p53 tumour suppressor gene is involved in the

carcinogenesis of the oral cavity [11,12]. We found thatoverexpression of p53 protein was found in all p53-mutated cancers, except one case which had a nonsensemutation. But 4 cases of tumour cells which did notshow p53 gene mutation in PCR-SSCP were also posi-tively expressed for p53 protein. There are several pos-sible mechanisms other than a point mutation which

Table 2

Summary of molecular and immunostaining data

Case p53 mutation p53 expression p21 expression

5 WT ÿ +

6 WT ÿ ++

8 WT ÿ ++

11 WT ÿ ++

13 WT ÿ ++

17 WT ÿ +

1 WT ÿ ÿ15 WT ÿ ÿ18 WT ÿ ÿ2 WT + +

10 WT ++ ++

14 WT + +

20 WT ++ ÿ9 MT (stop codon) ÿ +

3 MT ++ +

4 MT ++ ++

7 MT ++ ++

16 MT ++ ++

19 MT + ++

12 MT ++ ÿ

MT, mutant type; WT, wild type; ÿ, absent; +, focal or weak; ++,

strong.

Table 3

Correlation between p21WAF1/CIP1 expression and p53 mutation

p53 mutation Wild type Mutant type

p21 expression

Positive 9 cases 6 cases

Negative 4 cases 1 case

Fig. 4. p21WAF1/CIP1 expression at normal mucosa and at the dys-

plastic mucosa adjacent to squamous cell carcinoma. A: Moderate

p21WAF1/CIP1 immunoreactivity is observed in the suprabasal and

spinous cells. �400. B: Adjacent dysplastic epithelium shows strong

expression of p21WAF1/CIP1 in mostly basal and suprabasal cells.

�400.

Fig. 3. Squamous cell carcinoma of the oral cavity immunohisto-

chemically positive for p21WAF1/CIP1 expression. �120.

J.I. Yook, J. Kim/Oral Oncology 34 (1998) 198±203 201

can result in overexpression of the p53 protein [19].Genetic alteration in another region of the exon, such asthe promoter and intron of the p53 gene could result ina high expression of wild-type p53 [20,21]. But 98% ofmutations are clustered between exons 5 and 8 in fourconserved regions of the p53 gene [1,2]. As to the pos-sibility of a false negative due to the sensitivity of SSCPor by excess normal tissue, our experiment was per-formed from microdissected specimens which containedmore than 40% tumour cells. Since we consistentlyfound a shifted band in exon 7 of HT3 cells under dif-ferent glycerol concentrations and temperature condi-tions [16], we excluded the possibility of a false negative.Another possibility is that binding of the wild-type p53protein to other proteins, such as MDM2 or E6 viraloncoprotein, can stabilize and inactivate a result inoverexpression [2,22±24].

p21WAF1/CIP1 is transcriptionally induced by DNAdamage in a p53-dependent manner and may mediatecell cycle arrest [3,6,25]. Recent evidence has shown thatp21WAF1/CIP1 can also be induced by p53-indepen-dent pathways [26,27]. In the present study, p21WAF1/CIP1 was expressed in 15 cases of oral squamous cellcarcinomas. But the p21WAF1/CIP1 expression did notcorrelate either with p53 mutation or with p53 over-expression by immunohistochemistry. Of the 13 casesthat were wild type for p53, 4 showed immunoreactivityfor p21WAF1/CIP1, and 6 of 7 cancers with p53 muta-tions displayed similarly for p21WAF1/CIP1. Even acase with stop codon also showed p21WAF1/CIP1expression. The data in this study do not exclude thepossibility of a p53-dependent pathway of p21WAF1/CIP1 expression. The lack of correlation among p53mutational status and p21WAF1/CIP1 expressionsuggests that p21WAF1/CIP1 might be regulated byp53-independent pathways. Recent studies also showedthat there is another pathway on p21WAF1/CIP1expression in colorectal carcinomas, pancreatic carcino-mas, and squamous cell carcinomas of the skin[18,28,29]. Although we did not evaluate the mutationalstatus of p21WAF1/CIP1, according to other studies,mutational inactivation of p21WAF1/CIP1 is notinvolved in carcinogenesis [30±32]. p21WAF1/CIP1overexpression in 15 of 20 cases in our study appears tobe a di�erent mechanism from the overexpression dueto mutational inactivation. p21WAF1/CIP1 mRNAand protein expression in normal and neoplastic lungtissues were strictly associated [26], while in humanbreast epithelial cells, wild-type p53 a�ects p21WAF1/CIP1 protein accumulation at post-transcriptional levels[33]. Therefore, modulation of p21WAF1/CIP1 mRNAand protein expression needs further study in normal andmalignant tumours. Recent studies have shown TGF-band p53 act through distinct elements in the p21WAF1/CIP1 promoters, and bcl-2 suppresses the expression ofp21WAF1/CIP1 [34,35]. These results suggest multiple

regulational pathways of p2WAF1/CIP1 and supportour studies which demonstrated p53-independentexpression of p21WAF1/CIP1 [26,27].

An important step toward understanding the mol-ecular mechanisms of cell proliferation was taken withthe recent discovery of cyclin-dependent kinases(CDKs) inhibitory elements [5±8]. Although, thep21WAF1/CIP1 gene has been suggested as a possibletumour suppressor gene, there is no mutational event orgene deletion in various malignant tumours [30±32]. Incontrast to the growth arrest in a normal cell, the pro-liferation index is not related to the p21WAF1/CIP1expression in ovarian cancer cells [36]. This ®nding, aswell as the results of our study, are explained by stoi-chiometry of p21WAF1/CIP1 and CDK complexeswhich may regulate its activity [37]. When the ratio ofp21 to the complex is more than one, p21 inhibits thekinase activities. If the ratio is less than one, p21 servesonly as an assembling factor of cdk complex and doesnot inhibit cdk activity [7,38]. Therefore, an under-standing of this relation with other regulatory productsis essential to investigating the roles of p21WAF1/CIP1during carcinogenesis.

We found basal cells of tumour adjacent mucosa, incases in which p53 was positively expressed, alsoshowed its overexpression. These ®ndings suggest thatp53 gene alteration may be an early genetic alterationduring carcinogenesis [39,40]. In this study, in contrastto the p53, p21WAF1/CIP1 is expressed only in thenon-proliferative compartment in normal oral mucosa[9]. But most of the cells, including basal cells in thedysplastic epithelium adjacent to a tumour, expressedp21WAF1/CIP1. These results suggest that theincreased expression of p21WAF1/CIP1 is also an earlygenetic event during oral carcinogenesis. Moreover,disrupted regulation between the proliferation and dif-ferentiation may be a critical feature of malignanttransformation.

In summary, alterations of the p53 tumour sup-pressor gene were associated with the development oforal squamous cell carcinoma. We did not ®nd a con-sistent relationship between mutation of p53 gene andp21WAF1/CIP1 expression, implying p21WAF1/CIP1expression may be regulated by an additional path-way(s) in oral squamous cell carcinomas. Further stud-ies, including premalignant lesions, will be necessary toelucidate the pathogenetic and clinical implications ofp53 gene status and p21WAF1/CIP1 expression in oralcarcinogenesis.

Acknowledgements

This work was supported by a grant from the YonseiAcademic Fund 1996 (96-215). We thank Mr Jun ChulHong for his excellent technical assistance and BobRoss for advice on the manuscript.

202 J.I. Yook, J. Kim/Oral Oncology 34 (1998) 198±203

References

[1] Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 muta-

tions in human cancers. Science 1991;253:49±53.

[2] Zambetti GP, Levine AJ. A comparision of the biological activ-

ities of wild-type and mutant p53. FASEB 1993;7:855±865.

[3] El-Deiry WS, Tokino T, Velculescu VE, et al. WAF1, a potential

mediator of p53 tumor suppression. Cell 1993;75:817±825.

[4] Farmer G, Bargonetti J, Zhu H, Friedman P, Prywes R, Prives C.

Wild-type p53 activates transcription in vitro. Nature 1992;358:

83±86.

[5] Dulic V, Kaufmann WK, Wilson SJ, et al. p53-dependent inhi-

bition of cyclin-dependent kinase activities in human ®broblasts

during radiation-induced G1 arrest. Cell 1994;76:1013±1023.

[6] Li Y, Jenkins CW, Nichols MA, Xiong Y. Cell cycle expression

and p53 regulation of the cyclin-dependent kinase inhibitor p21.

Oncogene 1994;9:2261±2268.

[7] Waga S, Hannon GJ, Beach D, Stillman B. The p21 inhibitor of

cyclin-dependent kinases controls DNA replication by interaction

with PCNA. Nature 1994;369:574±578.

[8] Skapek SX, Rhee J, Spicer DB, Lassar AB. Inhibition of myo-

genic di�erentiation in proliferating myoblasts by cyclin D1-

dependent kinase. Science 1995;267:1022±1024.

[9] El-Deiry WS, Tokino T, Waldman T, et al. Topological control

of p21WAF1/CIP1 expression in normal and neoplastic tissues.

Cancer Research 1995;55:2910±2919.

[10] Field JK, Pavelic ZP, Spandidos DA, Stambrook PJ, Jones AS,

Gluckman JL. The role of the p53 tumor suppressor gene in

squamous cell carcinoma of the head and neck. Archives of

Otolaryngology Head and Neck Surgery 1993;119:1118±1122.

[11] Sakai E, Tsuchida N. Most human squamous cell carcinomas in

the oral cavity contain mutated p53 tumor-suppressor genes.

Oncogene 1992;7:927±933.

[12] Somers KD, Merrick MA, Lopez ME, Incognito LA, Schechter

GL, Casey G. Frequent p53 mutations in head and neck cancer.

Cancer Research 1992;52:5997±6000.

[13] Beahrs OH, editor. Manual for Staging of Cancer, 3rd Edition.

Philadelphia: JB Lippincott, 1988.

[14] Buchman VL, Chumakov PM, Ninkina NN, Samarina OP,

Georgiev GP. A variation in the structure of the protein-coding

region of the human p53 gene. Gene 1988;70:245±252.

[15] Orita M, Iwahana H, Kanazawa H, Kayashi K, Sekiya T.

Detection of polymorphisms of human DNA by gel electrophor-

esis as single-strand conformation polymorphisms. Proceedings

of the National Academy of Science, USA 1989;86:2766±2770.

[16] Hongyo T, Buzard GS, Calvert RJ, Weghorst CM. ``Cold

SSCP'': a simple, rapid and non-radioactive method for opti-

mized single-strand conformation polymorphism analyses.

Nucleic Acids Research 1993;21:3637±3642.

[17] Yaginuma Y, Westphal H. Analysis of the p53 gene in human

uterine carcinoma cell lines. Cancer Research 1991;51:6506±6509.

[18] DiGiuseppe JA, Redston MS, Yeo CJ, Kern SE, Hruban RH.

p53-independent expression of the cyclin-dependent kinase inhi-

bitor p21 in pancreatic carcinoma. American Journal of Pathol-

ogy 1995;147:884±888.

[19] Bass IO, Mulder JWR, O�erhous JA, Vogelstein B, Hamilton

SR. An evaluation of six antibodies for immunohistochemistry of

mutant p53 gene product in archival colorectal neoplasms.

Journal of Pathology 1994;172:5±12.

[20] Beenken SW, Karsenty G, Raycroft L, Lozano G. An intron

binding protein is required for transformation ability of p53.

Nucleic Acid Research 1991;19:4747±4752.

[21] Barnes DM, Hanby AM, Gillett CE, et al. Abnormal expression

of wild type p53 protein in normal cells of a cancer family

patient. Lancet 1992;340:259±263.

[22] Momand J, Zambetti GP, Olson DC, George D, Levine AJ.

The mdm-2 oncogene product forms a complex with the p53

protein and inhibits p53-mediated transactivation. Cell 1992;69:

1237±1245.

[23] Chen J, Wu X, Lin J, Levine AJ. mdm-2 inhibits the G1 arrest

and apoptosis functions of the p53 tumor suppressor protein.

Molecular and Cellular Biology 1996;16:2445±2452.

[24] Sche�nerM,Werness BA,Huibregtse JM,LevineAJ,Howley PM.

The E6 oncoprotein encoded by human papillomavirus types 16

and 18 promotes the degradation of p53. Cell 1990;63:1129±1136.

[25] Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ. The

p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1

cyclin-dependent kinases. Cell 1993;75:805±816.

[26] Marchetti A, Doglioni C, Barbareshi M, et al. p21 RNA and

protein expression in non-small cell lung carcinomas: evidence of

p53-independent expression and association with tumoral di�er-

entiation. Oncogene 1996;12:1319±1324.

[27] Michieli P, Chedid M, Lin D, Pierce JH, Mercer WE, Givol D.

Induction of WAF1/CIP1 by a p53-indepedent pathway. Cancer

Research 1994;54:3391±3395.

[28] Matsuta M, Kon S, Sasaki K, Matsuta M. Immunohistochemical

detection of p21WAF1/CIP1 and p53 proteins in formalin-®xed

para�n-embedded tissue sections of squamous cell carcinoma of

the skin. Journal of Dermatological Science 1997;14:233±239.

[29] Sasaki K, Sato T, Kursoe A, Ikeda E. Immunohistochemical

detection of p21waf1/cip1/sdi1 and p53 proteins in formalin-

®xed, para�n-embedded tissue sections of colorectal carcinoma.

Human Pathology 1996;27:912±916.

[30] Shiohara M, El-Deiry WS, Wada M, et al. Absence of WAF1

mutations in a variety of human malignancies. Blood

1994;11:3781±3784.

[31] Sun Y, Hidesheim A, Li H, et al. No point mutation but a codon

31ser!arg polymorphism of the WAF-1/CIP-1/p21 tumor sup-

pressor gene in nasopharyngeal carcinoma (NPC): the poly-

morphism distinguishes Caucasians from Chinese. Cancer

Epidemiology and Biomarkers Prevention 1995;4:261±267.

[32] Li YJ, Laurent-Puig P, Salmon RJ, Thomas G, Hamelin R.

Polymorphisms and probable lack of mutation in the WAF1-

CIP1 gene in colorectal cancer. Oncogene 1995;10:599±601.

[33] Gudas J, Nhuyen H, Li T, Hill D, Cowan KH. E�ects of cell cycle,

wild-type p53 and DNA damage on p21CIP1/WAF1 expression in

human breast epithelial cells. Oncogene 1995;11:253±261.

[34] Datto MB, Li Y, Panus JF, Howe DJ, Xiong Y, Wang XF.

Transforming growth factor b induces the cyclin-dependent

kinase inhibitor p21 through a p53-independent mechanism.

Proceedings of the National Academy of Science, USA 1995;92:

5545±5549.

[35] Upadhyay S, Li G, Liu H, Chen YQ, Sarkar FH, Kim HRC.

bcl-2 suppresses expression of p21WAF1/CIP1 expression of

p21WAF1/CIP1 in breast epithelial cells. Cancer Research 1995;

55:4520±4524.

[36] Barboule N, Mazars P, Baldin V, et al. Expression of p21WAF1/

CIP1 is heterogeneous and unrelated to proliferation index in

human ovarian carcinoma. International Journal of Cancer

1995;63:611±615.

[37] Xiong Y, Zhang H, Beach D. Subunit rearragement of the cyclin-

dependent kinases is associated with cellular transformation.

Genes and Development 1993;7:1572±1583.

[38] Zhang H, Hannon GJ, Beach D. p21-containing cyclin kinases

exist in both active and inactive states. Genes and Development

1994;8:1750±1758.

[39] Chang SE, Foster S, Betts D, Marnock WE. DOK a cell line

established from human dysplastic oral mucosa, shows a partially

transformed non-malignant phenotype. International Journal of

Cancer 1992;52:896±902.

[40] Warnakulasuriya KAAS, Johnson NW. Expression of p53

mutant nuclear phosphoprotein in oral carcinoma and potentially

malignant oral lesions. Journal of Oral Pathology and Medicine

1992;21:404±408.

J.I. Yook, J. Kim/Oral Oncology 34 (1998) 198±203 203