Int. J. Cancer: 52,867-872 (1992) 0 1992 Wiley-Liss, Inc.

t

Publicationof the International Union Against Cancer Publicationde I'Union Internationale Contre le Cancel

THE p53 TUMOR-SUPPRESSOR GENE AND ras ONCOGENE MUTATIONS IN ORAL SQUAMOUS-CELL CARCINOMA Eiki SAKAI'.*, Koichi RIKIMARU?, Masayoshi UEDA?,Yukie MATSUMOTO',Naoko ISHII'*~, Shoji ENOMOTO?, Hajime YAMAMOTO'and Nobuo T S U C H I D A ~ , ~ 'Department of Molecular Cellular Oncology and Microbiology; 2Second Department of Oral and Mawillofucial Surgery; ?Department of Oral Pathology, Faculty of Dentistry, Tokyo Medical and Dental University, Tokyo 113, Japan. The frequencies of mutations in the p53 tumor-suppressor gene and ras proto-oncogenes were investigated systematically in surgically resected oral squamous-cell carcinomas (SCCs) using single-strand conformation polymorphism (SSCP) and/or dot-blot hybridization analysis of DNA fragments which had been amplified by the polymerase chain reaction (PCR). pS3 gene mutations, within the region of exons 5 to 8, were detected in 17 out of 27 (63%) tumor specimens. The role of p53 mutations in cell-line establishment was investigated. p53 gene mutations were detected in 5 out of 6 tissue samples from which cell lines were established and in 4 out of 5 specimens from which cell lines could not be established, suggesting that the presence of p53 gene mutations is not by itself sufficient for cell-line establishment. Tumor samples were also analyzed for point mutational activation of the ras proto-oncogenes. One out of 30 (3%) tumors showed an activating point mutation in codon I 2 of H-ras, this being consistent with reports from Europe and USA but not with any from India. Compared to frequencies of the other genetic changes so far reported for oral SCC, the p53 mutations have been observed most often to undergo genetic change. p53 gene mutation is thus intimately involved in the genesis of oral SCC and consequently should be useful as a marker for the diagnosis of this neoplasm.

o 1992 Wilty-Liss, lnc. Oral squamous-cell carcinoma is regarded as the sixth most common malignancy worldwide and is the leading cause of death from cancer in India (Sanghavi, 1981). This cancer is characterized by relatively poor prognosis, and treatment is generally followed by severe dysfunction and disfigurement (Field and Spandidos, 1987). In the search for effective means of prevention, diagnosis and treatment of this cancer, genetic changes of both proto-oncogenes and tumor-suppressor genes should be studied to determine the origins of oral SCC. Proto-oncogenes such as c-erbB-1, int-2, bcl-1, myc and ras have been studied for amplification and/or over-expression in oral SCC (Yamamoto et al., 1986; Somers et al., 1990; Berenson et al., 1989; Field et al., 1989; Field, 1991; Saranath et al., 1989). Point-mutational analysis of the ras gene family has been conducted and its frequency of mutation in oral SCC found to be low in Europe and the USA, but not in India (Rumsby et al., 1990; Saranath et a/., 1991; Chang et al., 1991). Although we previously observed 2 SCC cell lines with activating point mutation of H-ras (Tadokoro et uL, 1989), no systematic study on the frequency of ras mutations in Japan has yet been reported. Frequent genetic alteration of the p53 tumor-suppressor gene has been studied in many cancers (reviewed in Levine et a[., 1991), but no research of this type has been conducted on oral SCC except for our recent study on cell lines (Sakai and Tsuchida, 1992). Additionally, an immunocytochemical study of p53 expression has been reported for SCCs of the head and neck (Field et al., 1991). In this study, attention was directed to the frequencies of mutations of the p53 tumor-suppressor gene and ras oncogenes in surgically resected SCC tumor tissues. To screen for mutations, a rapid method using single-strand conformation polymorphism (SSCP) (Orita et al., 1989) was employed. In this study, the frequency of p53 gene mutations in oral SCC was high, while that of ras gene activating mutations was low.

An attempt was made to determine the role of p53 gene mutation in cell-line establishment, since mutated p53 genes function as immortalizing oncogenes when co-transfected with activated ras in primary rat embryo cells (Hinds et al., 1990). MATERIAL AND METHODS

Tumor tissues Tumor tissues of oral SCC were obtained from 32 patients treated at the 2nd Department of Oral and Maxillofacial Surgery, Faculty of Dentistry, Tokyo Medical and Dental University (Table I). Tumor specimens were obtained from the excised material, freed of surrounding normal tissue, immediately frozen in liquid nitrogen and stored at -80°C until extraction of DNA. The SCC patients (62.5% males, 37.5% females) ranged in age from 35 to 79 years (median 57.7 years). Tumor staging (TNM) was conducted according to the 1987 classification of UICC (Hermanek and Sobin, 1987). The numbers of SCC patients in stages I, 11, I11 and IV were 5,2, 10 and 15, respectively. Oral SCC cell lines Cell lines established at the 2nd Department of Oral and Maxillofacial Surgery were HOC313 and ZA (Tadokoro et al., 1989). HSC2, HSC3 and HSC4 were obtained through the courtesy of Dr. F. Momose (Tokyo Medical and Dental University; Momose et a/., 1989). Control cell lines The cell lines used as positive controls with p53 point mutations or activating point mutations at codons 12, 13 or 61 of ras genes were HOS (codon 156 of the p53 gene; Romano et al., 1989), A431 (codon 273 of the p53 gene; Harlow et al., 1985), T24 (codon 12 of H-ras), Hs242 (codon 61 of H-ras), Calu-1 (codon 12 of K-uas), PR310 (codon 61 of K-ras), Molt-4 (codon 12 of N-ras; Eva et al., 1983), HT1080 (codon 61 of N-ras), HL-60 (codon 61 of N-ras) and an NIH3T3 transfectant transformed with the D N A of a myelodysplastic syndrome patient (codon 13 of N-ras; Hirai et al., 1987). Mutated bases of ras genes in the above cell lines, except for Molt-4, have been summarized in Nishimura and Sekiya (1987).

DNA preparation Cellular DNAs from frozen tumor tissues, cell lines and paraffin-embedded sections were prepared as previously described (Tadokoro et al., 1989; Fasano et al., 1984; Shibata et al., 1989). PCR-primers and ampli$c ation The PCR-primers used to amplify exons 5-9 of the p53 gene, and exons 1 and 2 of the 3 ras genes are shown in Table 11. Cellular D N A (100 ng) was amplified in a l O - ~ lreaction mixture (Innis et al., 1990) under thermal cycling as shown in the same Table. 'To whom correspondence and reprint requests should be sent. Fax: 3-5684-35 18.

Received: June 2, 1992 and in revised form August 7, 1992.

Xh8

SAKAI L T A L . TABLE I - ~ 5 TUMOK-SUPPRESSOK 3 GENE AND r u GENE MUTATIONS IN ORAL SQUAMOUS-CELL CAKClNOMAS

Tumor nurnher

__-

Sex age

___

1

Fl4h

2

F146

3

Mi62 Fi71 Mi54 MI53 Fl39

4

5 h

7 8 9 10 11 12 13 14 1s

16 17 18

19 20 21 22 23 24 25 26 27 28 29 30 31 32

F139

Mi70 Fi5S Mi41

MIS6 Mi62 Fi51 F/38 MI71 MI67 MI66 Fl35 MI63

MI62 Mi69 Fi74 MI59 Fi78 MI54 MI62 MI79 MIS6 Fl59 MI40 Mi70

Tongue Tongue Tongue Tongue Mouth floor Mouth floor Buccal mucosa Buccal mucosa Tongue Maxilla Mouth floor Tongue Tongue Mouth floor Mandible Maxilla Tongue Tongue Tongue Maxilla Maxilla Mandible Mandible Tongue Mandible Mandible Tongue Mandible Tongue Mandible Tongue Mandible

Tongue M.T.? M.T. Tongue Mouth Boor Mouth floor Buccal mucosa M.T. Tongue Maxilla Mouth floor Tongue Tongue M.T. M.T. Maxilla M.T. Tongue Tongue Maxilla Maxilla M.T. Mandible Tongue Mandible Mandible M.T. Mandible Tongue Mandible M.T. M.T.

3 3 1 3 4

2 1 1 4 4 4 I 2 4 4

4 3 2 I 3 4

4 3 4 4 4 4

3 1 2 3 4

1 1 1 1 3 1

0 0 0 (1

2 0 1 1 1 0 3 0 0 0 0 2 0 0 0 1 1 (I 0 0 0 0

0 0 0

n

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

111

I11 111

I11 IV Ill

I I

ND’

ND -

IV IV

ND exon 8 exon 8

1v I

ND

I11 IV IV IV IV I1

exon 8’ exon 5’ exon 5 exon 6’ exon 8)

I I11

exon 5

-

-

IV

exon 5 exon 5 exon 5 exon 5 ND exon 5 exon 8

-

ND ND -

ND ND -

ND ND -

-

KD ND + (HOC927) + (NU)4

exon 8

111

IV

-

+ (HOC815)4 + (HOC719)4

-

IV I11 IV IV IV IV 111 I II

ND ND ND ND ND ND ND ND ND ND ND ND ND + (HOC313)4 + (HOC605)4

exon 8

-3

exon 6j

‘ND; not done.-’M.T.: metastasized tumor.-’Data from Sakai and Tsuchida, 1992.-4The p53 gene mutations of these cell lines had been sequenced (Sakai and Tsuchida, 1992). TABLE 11- PRIMERS AND CONDITIONS OF THERMAL CYCLING FOK PCK AMPLIFICATION

Nanic

Amplified gene

Length (hp)

ESS ESA E6S E6A E7S E7A EXS EXA E9S E9A HIS HIA H2S H2A KIS KIA K2S K2A NlS NlA N2S N2A

exon 5 of p53 gene exon 6 of p 53 gene exon I of p53 gene exon 8 of pS3 gene exon 9 of p53 gene exon 1 of H-rus gene exon 2 of H-rus gene exon 1 of K-rus gene exon 2 of K-ms gene exon 1 of N-ms gene Exon 2 of N-rus gene

269 181

171 229 210 63

73 108

128 109 103

~~

Prinirr sequence

Thermal cycling

95°C (60 sec) 60°C (60 sec) 35 cycles 95°C (60 sec) 60°C (60 sec) 72°C (30 sec) 5’-GCCTCT GATT C C T C ACT GAT -3’ sense) 5’-TTAACCCCTCCTCCCAGAGA-3f~ant,-sense~ 35 cycles 95°C (60 5ec) 60°C (60 sec) S’-ACT GGCCT C A T C T T GGGCCT -3‘ sense) 5 ‘-TGT GCAGGGT GGC AAGT GGC-3’ianti-wise) 35 cycles 9S”C (60 sec) 60°C (60 sec) S’-TAAATGGGACAGGTAGGACC-3’(sense) ‘ 35 cycles ‘ S’-TCCACCGCTTCTTGTCCTGC-3’(anti-sense) S’-ACTAAGCGAGGTAAGCAhGC-3’(sense) 95°C (60 sec) 60°C (60 sec) 5’-CTGGAAACTTT C C A C T T GAT -3’(anti-sense) ‘ 35 cycles ‘ 94°C (85 sec) 60°C (9.5 sec) 72°C (72 sec) S’-GACGGAATATAAGCTGGTGG-3’ sense) 40 cycles 5’-TGGAT GGT CAGCGCACT C T T -3’[anti-aense) S’-AGACGTGCCTGTTGGACATC-3’(sense) 94°C (95 sec) 60°C (1 10 sec) 71°C (80 sec) S‘-CGCATGT ACT GGTCCCGCAT-3’(anti-sense\ 40 cvcles ’ S’-GACT GAAT AT AAACTT GT GG-3’(sense) 94°C (100 sec) 55°C(1 15 sec) 72°C (85 sec) S’-CTA T T G T T GGAT CATATTCG-3’(anti-sense) 40 cvcles ’ 5 ’-TT C C T AC AGGAAGC AAGT AG-3‘(sense) 94°C (100 sec) 54°C (I 15 sec) 72°C (85 sec) S’-CACAAAGAAAGCCCTCCCCA-3’(anti-sense) 40 cvcles S-GACTGAGTACAAACTGGTGG-3’(sense) ’ 94 oc(100 sec) svc (1 10 sec) 7 2 ” (80 ~ sec) 40 cycles 5‘-CT C T AT GGT GGGAT CAT A T T -3’ S‘-GGTGAAACCTGTTTGTTGGA-3‘ 94°C I90 sec\ 59°C (100 sec) 72°C (70 sec) 5’-ATAC AC AGAGGAAGCCT T C G-3‘(anti-sense) ’ 40‘cycles ’

5’-TGTTCACTTGTGCCCTGACT-3’(sense) Sf-CAGCCCTGTCGTCTCTCCAG-3’(anti-sense)

PCR-SSCP anul\isis PCR-SSCp was used to analyze the cellular D N A ~ of oral SCC tumor tissues as previously described (Orita et al., 1989). Briefly, labeling PCR was carried out with DNA (10 ng/yl) and 0.2 pM of each primer in the presence of [a-3’P]dCTP (40-50 kBq/pI, ICN, Costa Mesa, CA), 20 mM Tris-HCI (pH

8.3), 1.5 mM MgCL, 25 mM KCI, 50 p M each of dATP, dCTP, dGTP and dTTl’, and 0.2 units of Tuq DNA polymerase. The labeled PCR product was assessed for amplification by 3% agarose gel electrophoresis, diluted IOO-fold, and subjected to SSCP analysis using a 6% polyacrylaniide gel with 5% glycerol, as described by Orita ef ul. (1989). Electrophoresis was carried

p53 AND r u MUTATIONS IN ORAL SCC

869

out at 50 W for 3.5 to 5.5 hr at 25°C using a thermostatic electrophoresis apparatus (Pharmacia LKB, Uppsala, Sweden). The gel was dried on filter paper and exposed to Kodak XAR film at -80°C with intensifying screens. DN,4 sequencing of thep53 gene Direct sequencing of single-stranded DNAs produced by asymmetric PCR was performed as described by Gyllensten and Erlich (1988). In addition to PCR-primers, an oligonuclewas used otidc (S-7: S’-GTGACTGCTTGTAGATGGCC-3’) a\ d primer to sequence exon 5 . Dor-blot hybridization analysis on the codons I 2 trrid 61 of H-, K- tirid N-ras genes Dot-blot hybridization analysis was conducted essentially as described by Verlaan-de Vries et al. (1986). PCR-amplified DNAs ( 1-p1 aliquots) were spotted on nitrocellulose membrane filters (Schleicher and Schuell, Dassel, Germany). The filters were dried at 80°C for 2 hr in a vacuum and prehybridized in 5x SSC (Ix SSC: 0.15 M NaCl and 0.015 M Densodium citrate, pH 7.0), 5x Denhardt’s solution (Ix hardt’s solution: 0.02% BSA, 0.02% Ficoll, and 0.02% polyvinylpyrolidone), 0.1% SDS and sonicated salmon testis D N A (100 pgiml) at 37°C for 4 hr. 32P-labeled A S 0 probes were then added (10 pmol; specific activity, 2.2 x loh dpmipmol) and hybridized for 12 hr at 37°C. A S 0 probes were labeled with [y-3’P]ATP using T4 polynucleotide kinase. The filters wcre washed with 2x SSPE ( l x SSPE: 180 mM NaCI, 10 mM NazPOl and I mM EDTA, pH 7.4) containing 0.1% SDS for 45 min at room temperature. and then with a solution containing 3 M tetramethylammonium chloride, 50 mM Tris-HCI (pH 8.0). 2 mM EDTA, and 0.1% SDS for 30 min at selected discriminating temperatures (ranging from 56°C to 64°C). They wcre then autoradiographed by exposure to Kodak XAR film at -80°C with intensifying screens.

F~CURE 1 - PCR-SSCP analysis of ( a ) E5F including exon 5 , (b) E8F including exon 8 of the p53 gene, PCR-amplified from the DNAs of surgically resected tumor tissues. Human placenta DNA

(HP) served as a wild-type.

RESULTS

PCR-SSC‘P arid direct sequeticirig analysis of the p53 gem’

Cellular DNAs prepared from 21 oral SCC tumor tissues (listed in Table I) were analyzed for p53 gene mutations in cxons 5-9 by PCR-SSCP analysis. Exon-flanking primers for PCR wcre dcsigned to detect mutations that fell within exons as well as those in flanking consensus splicing donor and acceptor sequences: the primers used were E5S and E5A for exon 5 , E6S and E6A for exon 6, E7S and E7A for exon 7, E8S and E8A for exon 8, and EYS and E9A for exon 9 (Table 11). To exclude fortuitous mutations caused by Taq polymerase (Saiki et al., 1988), the SSCP analysis was conducted by 2 independent experiments for each tumor sample. The results of the SSCP analyses are shown in Figure 1. Twelve tumors showed extra bands, indicative of the presence of mutations, in one pair of exon fragments, although the same fragments as the wild-type control were also observed to appear with varying intensity in the tumor tissues. SSCP analysis showed 7 tumors to have mutations in exon 5 (tumor numbers; 16, 20.22,23,24,25 and 27) and 5 tumors in exon 8 (tumor numbers; 5 , 9, 10, 28 and 30) (Fig. 1), but no tumor in exons 6, 7, and 9 fragments (data not shown). The results are summarized in Table 1. The Table also includes the result (Sakai and Tsuchida, 1992) of SSCP analysis of 6 tumor tissues from which cell lines have been established (see below). Upon combining the results of both experiments, 17 out of 27 tumor tissues (63%) were seen to have mutations in the p53 gene. Out of the total of 6 tumors, 4 tumors, which were used to generate cell lines and in which mutated bands with the same mobility shifts as those of the corresponding cell line fragments had been detected (Sakai and Tsuchida. 1992), were subjected to direct sequencing using single-stranded DNAs synthesized by asymmetric PCR as templates (Fig. 2). Mis-sense mutations were found in 3

F~GURE 2 - Direct sequencing analysis of p53 gene of oral SCC tumor tissues. Mis-sense mutations were detected in 14, 17 and 18, and a non-sense mutation in 15.

cases: in codons 285 (GAG to AAG), 205 (TAT to TGT) and 281 (GAC to GAG), respectively. The remaining case showed a non-sense mutation in codon 126 (TAC to TAG). In these 4 cell lines we thus confirmed that the mutations in the p53 gene were the same in both the original tumor and the corresponding cell line.

870

SAKAI E T A L .

p53 niutation in relation to capacity for tumor-cell establishment as a cell line A study was made to determine whether tumor cells with p53 mutations are more efficiently established as a cell line than those without mutations. From the 27 original SCC tumor tissues, 11 samples were randomly chosen to establish cell lines as previously described (Rikimaru et al., 1992). Six of these 11 samples were successfully established as cell lines, while 5 could not be established. p53 mutations were detected in 5 out of 6 samples from which cell lines were established, and in 4 out of 5 samples from which cell lines could not be established (Table I). ASOprobe hybridization and SSCP analysis of H-, Kand N-ras genes Cellular DNAs from 30 oral SCC tumor tissues (Table 11) were examined for point mutational activation of H-, K- and N-ras genes in codons 12, 13 and 61. DNAs were amplified by PCR in 2 regions; one set of primers was used to amplify the region containing codons 12 and 13, while the other set was used to amplify the sequence surrounding codon 61 (Table 11). Point mutation in codons 12 and 61 was analyzed and identified by dot-blot hybridization with 45 A S 0 probes which detect any mutation accompanied by amino acid substitution. Only one of the 30 tumor specimens was found to exhibit a point mutation in any rus gene. The tumor tissue that had been used to establish the oral SCC cell line HOC313 as well as the cell line displayed a mutation in codon 12 of H-ras (GG to AGC, Gly to Ser) (Fig. 3). The presence of a mutation in c$don 12 of H-ras in this cell line had been shown by DNA (Southern) blot hybridization analysis (Tadokoro et al., 1989). Since codon 13 within exon 1 of the rus genes is also a target for activating ras genes (Barbacid, 1987), the same PCRamplified fragments containing codons 12 and 13 that were used for the A S 0 probe hybridization analysis were also used for SSCP analysis. No mutation in this region could be detected in any tumor sample (Fig. 4). Mutated bands of the H-ras codon 12 mutation of HOC313 in the original tumor could also not be detected by this method (data not shown), as this mutation induced only marginal band-shift (Fig. 4), and

the sample contained excess non-tumor tissue. From the results of the 2 analyses, point mutational activation of the ras gene was thus detected in only one out of 30 (3%) tumor samples. DISCUSSION

Sensitive strategy using SSCP makes it possible to screen more quickly a large number of tissue samples for the mutation of specific genes. By this method mutation of the p53 tumorsuppressor gene was found in 17 out of 27 oral tumor tissues (63%). Compared to the frequencies of other genetic changes so far reported for oral SCC (Yamamoto et al., 1986; Somerset al., 1990; Berenson et al., 1989; Field et al., 1989, 1991; Saranath et al., 1989, 1991), this value is the highest reported, suggesting that p53 mutation is intimately involved in the genesis of oral SCC, and may thus be useful as a marker for diagnosis. This mutation was less frequent than in our report on oral SCC cell lines (Sakai and Tsuchida, 1992). This discrepancy may be due in part to the decrease in detection sensitivity associated with tumor samples, since in our cell lines we were able to analyze RNAs in addition to DNAs, and our lines were also expected to be free from non-tumor cells that were present in the tumors. Examination of lymphoid malignancy has indicated that tumor cells carrying p53 mutations are more amenable to the establishment of cell lines (Gaidano et al., 1991) and, in addition, p53 alteration is a common event in the spontaneous immortalization of primary murine embryo fibroblasts (Harvey and Levine, 1991), which may explain the discrepancy. The present study shows that p53 gene mutations were detected in 5 of 6 tissues from which cell lines were established and in 4 of 5 tissues from which lines could not be established. For oral SCC, other factor(s) would thus appear to have significant effects on cell-line establishment. Further, p53 mutationsperse may not be as important as the nature and position of mutations in this gene: depending on their position in the p53 gene, different mutations had different transforming activities in the collaboration assay with activated ras gene (Hinds et al., 1990).

FIGURE 3 - Dot-blot hybridization analysis of tumor tissues, cell lines and a paraffin-embedded section of oral SCC for activating point mutations in codon 12 of H-ras gene. DNAs from 29 tumor tissues, 5 cell lines (HOC313, ZA, HSC2, HSC3 and HSC4), a positive control cell line (T24) and human placenta DNA (HP) were amplified with a PCR-primer set (H1S and HlA), spotted on nitrocellulose filters as described in the left panel, hybridized to A S 0 probes. The results for 3 A S 0 probes with GGC (Gly; wild-type), AGC (Ser; mutated) and GTC (Val: mutated) sequences as codon 12 are shown.

871

pS3 AND rus MUTATIONS IN ORAL SCC

FIGURE 4 - PCR-SSCP analysis of the tumor tissues and cell lines of oral SCC for mutations of (a) H-, (h) K- and (c) N-rus genes in portions of exon 1 including codons 12 and 13. HlF, K1F and N1F indicate 2 denatured single-stranded DNAs of human placenta DNA (HP) which served as a negative control. T24, ZA and HOC313 served as positive controls for H-rus, Calu-1 for K-rus, and N13 (an NlH3T3 transfectant with the DNA of a myelodysplastic syndrome patient) and Molt-4 for N-rus. Residual double-stranded fragments are indicated as d( -).

In contrast to the frequent mutation of the pS3 gene, rus gene mutational activation, which has been reported to be a target in many human cancers (reviewed in Barbacid, 1987; Bos, 1989), is found infrequently in oral SCC. An activating point mutation in the H-ras gene was found in only 1 oral SCC tumor sample out of 30 (3%). After inclusion of the activating mutation in the previously reported ZA cell line (Tadokoro et ul.. 1989), the number of cases of rus gene activation in this cancer rises to 2 out of 31 (6.5%). This is consistent with studies in Europe and the USA, which show rus gene activation to be infrequent and activation to be primarily restricted to H-rus (Rumsby et a/., 1990; Somers et ul., 1990). Etiological factors in Japan might be similar to those in the above countries, but different from those in India (Saranath et ul., 1991). Besides the oral cavity, many organs such as the esophagus, lung, cervix, anus and skin can give rise to malignancies with squamous-cell histology. The high incidence of the pS3 gene mutation and low incidence of rus gene activation havc been reported for SCCs in these organs. This pattern is typical of lung cancers: squamous-cell carcinomas have a much higher frequency of the pS3 mutation than adenocarcinomas (Iggo et ul., 1990; Chiba et ul., 1990), and vice i w s u for rus gene activation (Bos, 1989). It should be pointed out that in HPV

(human papilloma virus)-positive SCCs of the cervix and anus, the degradation of pS3 protein, although normal, is enhanced by the E6 protein of HPV (Scheffner et ul., 1990). It thus follows that pS3 geneiprotein inactivation, but not rus gene activation, is intimately associated with the genesis of cancers with squamous-cell histology. In cervical cancers, inactivation of the R B gene may also be involved in carcinogenesis (Scheffner et ul., 1991). Whether the R B gene is also inactivated in oral SCC is thus a point that should be investigated. ACKNOWLEDGEMENTS

The authors express their sincere appreciation to Drs. K. Tadokoro, K. Hayashi, T. Sekiya and K. Yamato for their invaluable comments, and to Drs. F. Momose. H. Nakano, K. Shimotohno and the Japanese Cancer Research Resources Bank (JCRB)-Cell for providing cell lines. Thanks are also due to Dr. Y. Yuasa for kindly providing the Hs242 cellular DNA, and Mr. E.D.S. (Committee on Virology, Harvard Medical School) for his excellent editorial assistance. This work was supported in part by grants-in-aid from Japan Foundation for Aging and Health, and the Ministry of Education, Science and Culture of Japan.

REFERENCES

BARBACID, M., ras genes.Ann. Rev. Biochem., 56,779-827 (1987). BERENSON. J.R., YANO,J. and MICKEL.R.A., Frequent amplification of the hcl-I locus in head and neck squamous cell carcinomas. Oncogene, 4, 1111-1116(1989). Bas, J.L., ms oncogenes in human cancer: a review. Cuncer Res., 49, 4hX24689 (1989).

EVA,A., TRONICK, S.R., GOL,R.A., PIERCE. J.H. and AAKONSON, S.A., Transforming genes of human hematopoietic tumors: frequent detection of ras-related oncogenes whose activation appears to be independent of tumor phenotype. Proc. nut. Accid. Sci. (Wash.), 80,49264930 (1983).

FASANO, 0..BIRNBAUM, D., EDLCIND, L., FOGH,J. and WIGLER,M., New human transforming genes detected by a tumorigenicity assay.

CHANG,S.E., BHATIA, P., JOHNSON, N.W., MORGAN, P.R., MCCOR- Mol. cell. Bid., 4,1695-1705 (1983). MICK, F., YOUNG, B. and HIORNS, L., Rus mutations in United J.K., Oral cancer and detection of patients and lesions at risk. Kingdom examples of oral malignancies are infrequent. bit. J. Cancer, FIELD, Ir7: N.W. Johnson (ed.), The biology of oncogenes, pp. 257-293, 48,409412 (1991).

Cambridge University Press, Cambridge (1991). CHIBA, I., TAKAHASHI. T., NALI.M.M., D’AMICO, D., CLIRIEL, D.T., J.K. and SPANDIDOS. D.A., Expression of oncogenes in human MITSLJDOMI, T., BLICHHAGEN, D.L., CARBONE, D., PIANTADOSI, S., FIELD, KOGA, €I., R t l S S M A N . P.T.. SALMON, D.J., HOLMES, E.C. and MINNA, tumors with special reference to the head and neck region. J. oral J.D.. Mutations in the pS3 gene are frequent in primary, resected Purhol., 16,97-107 (1987). non-small cell lung cancer. Oncogene, 5, 1603-1610 (1990). FIELD, J.K., SPANDIDOS, D.A., MALLIRI, A,, GOSNEY, J.R., YIAGNISIS,

872

SAKAI E T A L .

M. and STELL, P.M., Elevated P.53 expression correlates with a history growth factor receptor in squamous cell carcinoma of the head and of heavy smoking in squamous cell carcinoma of the head and neck. neck. Head Neck, 14,8-13 (1992). Brit. J. Cuncer, 64,573-577 (1991). ROMANO, J.W., EHRHART. J.C., DUTHU.A., KIM, C.M.,APPELA.E. and FIELD,J.K., SPANDIDOS, D.A., STELL,P.M.. VAUGHAN. E.D., EVAN, MAY, P., Identification and characterization of a p53 gene mutation in (3.1. and MOORE,J.P.. Elevated expression of the c-myc oncoprotein a human osteosarcoma cell line. Oncogene, 4, 1483-1488 (1989). correlates with poor prognosis in head and neck squamous cell RUMSBY. G., CARTER,R.L. and GUSTERSON, B.A., Low incidence of carcinoma. Oncogene, 4, 1463-1468 (1989). ras oncogene activation in human squamous cell carcinomas. Brit. J. GAIDANO, G., BALLELINI, P., GONG,J.Z., INGHIRAMI.G.. NERI, A,, Cancer, 61,365-368 (1990). NEWCOMB,E.W., MAGRATH,I.T., KNOWLES,D.M. and DALLA- SAIKI, R.K., GELFAND, D.H., STOFFEL.S.J., HIGUCHI, R., HORN, G.T., FAVEKA,R.. p.53 mutations in human lymphoid malignancies: associa- MULLIS,K.B. and ERLICH,H.A., Primer-directed enzymatic amplification with Burkitt lymphoma and chronic lymphocytic leukemia. Proc. tion of DNA with a thermostable DNA polymerase. Science (Wash.), nut. Arad. Sci. (Wash.), 88, 5413-5417 (1991). 239,4871191 (1988). GYLLENSTEN, U.B. and ERLICH, H.A., Generation of single-stranded E. and TSUCHIDA, N..Most human squamous cell carcinomas in DNA by the polymerase chain reaction and its application to direct SAKAI, the oral cavity contain mutated p53 tumor-suppressor genes. Oncosequencing of the HLA-DQA locus. Proc. nut. Acad. Sci. (Wash.), 85, gene, 7,927-933 (1992). 7652-7656 (1988). SANGHAVI, L.D., Epidemiologic and intervention studies. Screening: HARLOW, E., WILLIAMSON, N.M., RALSTON, R.. HELFMAN, D.M. and cancer epidemiology: the Indian scene. J. Cancer Res. clin. Otzcol., 9, ADAMS,T.E., Molecular cloning and in vifro expression of a cDNA 1-14 (1981). clone for human cellular tumor antigen p53. Mol. cell. Biol., 5, SARANATH, D., CHANG,S.E.. BHOITE. L.T., PANCHAL, R.G., KERR, 1601-1610 (1985). I.B., MEHTA,A.R., JOHNSON, N.W. and DEO, M.G., High frequency HARVEY, D.M. and LEVINE, A.J., p53 alteration is a common event in mutation in codons 12 and 61 of H-ras oncogene in chewing-tobaccothe spontaneous immortalization of primary BALBic murine embryo related human oral carcinoma in India. Brit. J. Ccrncer, 63, 573-578 fibroblasts. Genes Devel., 5,2375-2385 (1991). (1991). HERMANEK, P. and SOBIN.L.H. (eds.), UICC: TNM clnssificxztion of SARANATH, D.. PANCHAL, R.G., NAIR.R., MEHTA,A.R., SANGHAVI. nialignant trmiors, 4th ed., Springer. Berlin (1987). V., SUMEGI, J., KLEIN,G. and DEO, M.G., Oncogene amplification in HINDS, P.W.. FINLAY,C.A., QUARTIN, R.S.. BAKER,S.J.. FEAKON,squamous cell carcinoma of the oral cavity. Jap. J. Cuncer Res., 80, E.R., V(x;ELsTEiN, B. and LEVINE.A.J., Mutant p53 DNA clones from 430-437 (1989). human colon carcinomas cooperate with ras in transforming primary R , MUNOER,K., BYRNE,J.C. and HOWLEY,P.M., The rat cells: a comparison of the “hot spot” mutant phenotypes. Cell S C H ~ F F N EM., state of the p53 and retinoblastoma genes in human cervical carcinoma Growth Difl, 1,571-580 (1990). cell lines. Proc. ,iat. Acad. Scr. (Wash.), 88,5523-5527 (1991). HIRAI,H., KOBAYASHI. Y., MANO,H., H A G I W A R A , K., MARU,Y., M., WERNESS, B.A., HUIBREGTSE, J.M.. LEVINE,A.J. and OMiNE, M., MIZOGUCHI, H., NISHIDA, J. and TAKAKU, F., A point SCHEFFNER, mutation at codon 13 of the N-rus oncogene in myelodysplastic HOw.EY. P.M., The E6 oncoprotein encoded by human papillomavirus type I6 and 18 promotes the degradation of p53. Cell, 63, syndrome. Nature (Lotid.), 327,430-432 (1987). 1129-1136 (1990). IGGO, R., GATTER, K., BARTEK,J.. LANE,D. and HARRIS,A.L.. SHIBATA, D., BRYNES,R.K., NATHWANI. B., KWOK.s.,SNINSKY, J. and Increased expression of mutant forms of p53 oncogene in primary lung A K N H E I M N.,, Human immunodeficiency viral DNA is readily found in cancer. Lancet, 335,675-679 (1990). lymphnode biopsies from seropositive individuals. Arner. J . Pathol., INNIS,M.A. and GELFAND,D.H., Optimization of PCRs. In: M.A. 135,697-702 (1989). Innis. D.H. Gelfand. J.J. Sninskv and T.J. White (eds.). PCR mntocols: K.D., CARTWRIGHT, S.L. and SCHECHTER. G.L., Amplification a guide to methods and application. pp. 3-12, Academic Press, San SOMERS, of the int-2 gene in human head and neck squamous cell carcinomas. Diego (1990). Oncogene. 5, 915-920 (1990). LEV IN^, A.J., MOMAND. J . and FINLAY, C.A.. The p53 tumor suppresTADOKORO, K., UEDA,M., OHSHIMA, T.. FLJJITA, K., ENOMOTO. S. and sor gene. Nnrure (Lotid.),351,453-456 (1991). TSUCHIDA, N., Activation of oncogenes in human oral cancer cells: a MOMOSE,F.. ARAIDA, T., NEGISHI,A,, ICHIJO, H., SHIODA, s. and novel codon 13 mutation of c-H-ras gene and concurrent amplification SASAKI,S., Variant sublines with different metastatic potentials of c-erbB-l and c-niyc. Oncogene, 4,499-505 (1989). selected in nude mice from human oral squamous cell carcinomas. J. VERLAAN-DE VRIES,M.. BOGAARD,M.E., V A N DEN ELST. H., VAN oral Pathol. Med., 18,391-395 (1989). BOOM,J.H., V A N DEK Eo, A.J. and Bos, J.L., A dot-blot screening NISHIMURA, s. and SEKIYA, T., Human cancer and cellular oncogenes. procedure for mutated ras oncogene using synthetic oligodeoxynucleBiochem. J., 243,313-327 (1987). otides. Gene, 50,3 13-320 (1986). ORITA.M.. SULUKI, Y., SEKIYA, T. and HAYASHI, K., Rapid and YAMAMOTO. T., KAMATA. N., KAWANO, H., SHIMIZU, S., KUKOKI, T., sensitive detection of point mutations and DNA polymorphism using TOYOSHIMA, K., RIKIMARU. K., NOMUKA, N., ISHIZAKI, R., PASrAN, I., the polymerase chain reaction. Genornics, 5,874-879 (1989). GAMOU.S. and SHiMizu, N., High incidence of amplification of the epidermal growth factor receptor gene in human squamous cell RIKIMARU, K., TADOKORO, K., YAMAMOTO, T., ENOMOTO, S. and TSUCHIDA. N., Gene amplification and overexpression of epidermal carcinoma cell lines. Cancer Res., 46,4141116 (1986).