204

Alterations of p53, Cyclin D1, Rb, and H-ras in Human

Oral Carcinomas Related to Tobacco Use
Jie Xu, M.S.1
Irma B. Gimenez-Conti, D.D.S., Ph.D.1
Joan E. Cunningham, B.S.C., M.S.C.2
Ana M. Collet, D.D.S.3
Mario A. Luna, M.D.4
Hector E. Lanfranchi, D.D.S.3
Margaret R. Spitz, M.D.2
Claudio J. Conti, D.V.M., Ph.D.1
1

The University of Texas M. D. Anderson Cancer

Center, Science Park-Research Division, Smithville, Texas.

Department of Epidemiology, the University of

Texas M. D. Anderson Cancer Center, Houston,
Texas.

Department of Pathology, School of Dentistry,

University of Buenos Aires, Buenos Aires, Argentina.

Department of Pathology, the University of Texas

M. D. Anderson Cancer Center, Houston, Texas.

BACKGROUND. Epidemiologic studies have indicated that environmental and personal habits, particularly tobacco use and alcohol abuse, are major etiologic factors
in the induction and progression of head and neck squamous cell carcinomas
(HNSCC). Molecular studies have focused on HNSCC related to smoking but not
those associated with smokeless tobacco.
METHODS. The authors studied immunohistochemical evidence of alterations of
p53, cyclin D1, and Rb in 34 human oral carcinomas related to tobacco use. They
also examined p53 and H-ras using single strand conformation polymorphism
(SSCP) and sequencing analysis.
RESULTS. Overexpression of cyclin D1 was found in 41% of cases, and accumulation of p53 was found in 59%. Only 9% of the samples did not show Rb staining. In
SSCP and sequencing analysis, 17 cases showed mutations in the conserved region
of the p53 gene. No mutations were detected in codons 12, 13, or 61 of the H-ras
gene.
CONCLUSIONS. Overexpression of cyclin D1 and p53 mutations are common alterations in HNSCC. In contrast, the loss of Rb function seems to occur infrequently,
and mutations in the H-ras gene apparently do not play a role in this cancer.
HNSCC associated with smokeless tobacco contained the same alterations as those
related to smoking. Cancer 1998;83:204 12. 1998 American Cancer Society.
KEYWORDS: head and neck, squamous cell carcinoma, cyclin, p53, Rb; H-ras,
tobacco, smoking, smokeless tobacco.

Supported by grants from DHS (CA 42157, 53123).

The authors thank Yolanda Valderrama and Melissa Bracher for helping with the preparation of the
manuscript, Judy G. Ing and the Art Department for
the artwork, the Histology Laboratory for the tissue
specimen preparation, and Dr. Michael LaBate and
the Molecular Biology Laboratory for technical assistance.
Address for reprints: Claudio J. Conti, D.V.M.,
Ph.D., the University of Texas M. D. Anderson
Cancer Center, Science Park-Research Division,
P.O. Box 389, Smithville, TX 78957.
Received November 28, 1997; revision received
December 8, 1997.
1998 American Cancer Society

ead and neck squamous cell carcinoma (HNSCC) is a common

neoplasm in humans. Epidemiologic studies have shown that
environment and personal habits, particularly tobacco use and alcohol abuse, seem to be major etiologic factors in the induction and
progression of HNSCC. The molecular mechanisms implicated in the
malignant transformation and progression of HNSCC are still not well
understood.
Neoplastic growth is characterized by alterations of two known
groups of genes: oncogenes and tumor suppressor genes. The interaction between activated oncogenes and mutations that result in a
loss of function of tumor suppressor genes appears to be the driving
force directing normal cells to uncontrolled growth and invasion.13
The two most-studied tumor suppressor genes are the retinoblastoma susceptibility gene (Rb) and the p53 gene.2 Mutations involving
the p53 gene are the most commonly identified events in various
human cancers.4 The Rb gene was the first tumor suppressor gene to
be isolated. Loss of heterozygosity and loss of expression of the Rb
gene are seen during the development of retinoblastoma and several
other types of human cancer.5
Of the oncogenes, the ras gene family is one of the most widely

Tobacco and Genes in Oral Tumors/Xu et al.

studied. Point mutations in codons 12, 13, and 61 of

the ras oncogene have been identified in many types
of human malignant tumors.6 The frequency of such
mutations in squamous cell carcinoma, however, appears to be substantially lower than in some other
types of neoplasia (e.g., adenocarcinomas).79
In addition to suppressor genes and oncogenes,
other genes controlling the entry of cells into the cell
cycle have recently been implicated in cancer development. One is the cyclin D1 gene, which has been
mapped to 11q13, where amplification of several
linked genes has been observed in human cancers.10
The product of the cyclin D1 gene forms a complex
with cyclin-dependent protein kinases that governs a
key transition in the cell cycle.11 The oncogenic potential of cyclin D1 has also been demonstrated in
vitro.12 The cyclin D1 gene is amplified and overexpressed in 30 40% of head and neck carcinomas.13,14
Previous investigations have shown that amplification and overexpression of the cyclin D1 gene and
mutations of the p53 gene are frequent events in
HNSCC.15,16 However, inactivation of the Rb gene and
mutations in ras oncogenes appear to be very rare
events in this type of cancer.17 Information regarding
alteration of Rb and cyclin D1 in head and neck carcinoma related to tobacco use is very limited. In addition, the correlations between inactivation of p53,
Rb, and overexpression of the cyclin D1 gene have not
previously been examined in HNSCC.
In this study, we investigated alterations in p53
and the H-ras gene at the genomic level and measured
the abundance of p53, Rb, and cyclin D1 protein by
immunohistochemical analysis to determine whether
there is a correlation among these genetic events and
whether it is associated with various types of tobacco
use.

MATERIALS AND METHODS

205

Immunohistochemical Analysis
After deparaffinization and dehydration of the tissue
samples, endogenous peroxidases were blocked with
0.3% hydrogen peroxide in methanol for 30 minutes.
Then the slides were washed three times with 0.1%
bovine serum albumin in 13 phosphate-buffered saline. Antigen retrieval was performed as described by
Cattoretti et al.18 and Robles and Conti.19 The slides
were washed in distilled water and boiled in a plastic
jar filled with 0.01 M sodium citrate buffer (pH 6.0) in
a water bath for 20 minutes or in a microwave oven.
The slides were then transferred to distilled water and
rinsed with phosphate-buffered saline. After antigen
retrieval, the slides were incubated with one of three
antibodiespolyclonal antihuman p53 antibody
CM-1 (Signet Laboratories, Dedham, MA), polyclonal
antihuman cyclin D antibody (Upstate Biotechnology,
Inc., Lake Placid, NY), or monoclonal antihuman Rb
antibody Ab-4 (Santa Cruz Biotechnology, Inc., Santa
Cruz, CA)for 1 hour at room temperature (overnight
at 4C for Rb staining). The human cyclin D antibody
reacts predominantly with cyclin D1 with little crossreaction to cyclin D2.19 The slides were then incubated
with the secondary antibodies, biotinylated antirabbit
(for p53 and cyclin D1) or antimouse (for Rb) immunoglobulin, for 30 minutes before the use of avidinbiotin complex (Vectastain Elite ABC Kit, Vector Laboratories Inc., Burlingame, CA). The specimens were
then incubated with a solution of diaminobenzidine
tetrahydrochloride and hydrogen peroxide for 510
minutes. As a control, the slides were stained without
the primary antibody to monitor background staining.
The accumulation of p53 protein and overexpression
of cyclin D1 protein were classified as 1 (520%
positive tumor cells), 11 (20 50% positive tumor
cells), and 111 (.50% positive tumor cells). In Rb
immunostaining, 1 and 2 were used to represent
the existence or absence, respectively, of Rb protein in
nuclei.

Tissue Specimens
Thirty-four oral carcinoma and two dysplasia samples
from 29 patients with oral cancer were studied for
changes in molecular markers (Table 1). The patients
were selected from among those who registered at the
M. D. Anderson Cancer Center (MDACC) between
1985 and 1992 and completed the Patient Risk Evaluation Program questionnaire and from whom tissue
samples were available. The tissue specimens had
been fixed with formalin and embedded in paraffin.
Histopathologic classification was assessed before immunostaining. From Patients 1005, 1010, and 1015,
samples from both original and subsequent primary
tumors were examined. Samples from the primary and
a metastatic tumor from Patient 1002 were examined.

DNA Extraction
DNA was extracted from the paraffin-embedded tissues as previously described by Gimenez-Conti et al.20
Tumor tissue was microdissected from unstained tissue sections. A total of approximately 10 20 mm2 of
tissue sections 5 mm thick were used in each case. The
paraffin was then removed from the dissected tissue
with xylene, and the slides were rinsed with 100%
ethanol and dried in a vacuum. The samples were
then resuspended in 5% Chelex TM 100 (Bio-Rad,
Hercules, CA) and incubated with proteinase K for 60
minutes. After being boiled for 10 minutes to destroy
the remaining proteinase activity, extracted DNA from

206

CANCER July 15, 1998 / Volume 83 / Number 2

TABLE 1
Histopathologic and Clinical Features
Exposure
Patient

Sample

Gender

Race

Age (yrs)

Smoker

Chew/snuff

Alcohol

XRT

Site

Histology

Type1

Grade2

1001
1002

1
2
3
4
5
6
7
8
10
11
12
13
14

M
M
M
M
M
F
F
M
M
F
M
M
M

W
W
W
W
W
W
W
W
W
W
W
W
W

64
64
64
79
77
61
61
70
49
58
78
52
52

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y

Y
Y
Y
Y
Y

Y
Y
Y

Y
Y

SCC
SCC
SCC
SCC
SCC
SCC
SCC
SCC
Dys
SCC
Dys
SCC
SCC

P
P
M
P
P
P
P
M

M
W
P
W
M
W
M
W

P
P

M
P

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

M
F
M
M
M
M
M
M
F
M
M
M
M
M
M
M
M
M
M
M

W
W
W
W
W
W
B
W
W
W
W
W
W
W
B
W
W
W
W
W

56
87
59
79
52
57
44
48
82
47
66
80
82
40
46
77
35
63
73
67

Y
Y
Y

Y
Y
Y
Y

Y
Y
Y
Y

Y
Y
Y
Y
Y

Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y

Y
Y

Y
Y
Y

Y
Y
Y

Y
Y

Tongue
Floor of mouth
Floor of mouth
Buccal mucosa
Tongue
Floor of mouth
Floor of mouth
Lymph node (neck)
Buccal mucosa
Palate
Palate
Retromolar mucosa
Retromolar mucosa
Fibroadipose of
neck
Buccal mucosa
Gingiva
Buccal mucosa
Gingiva
Buccal mucosa
Tongue
Buccal mucosa
Oral cavity
Oral cavity
Palate
Tongue
Retromolar mucosa
Tongue
Floor of mouth
Palate
Tongue
Tongue
Tongue
Floor of mouth

M
P
P
P
P
P
P
P
P
P

P
M
W
W
M
W
M
M
M
W

P
P
P
P
P
P
P
P
P

W
W
M
W
W
M
M
M
W

1003
1004
1005
1006
1007
1008
1009
1010

1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029

SCC
SCC
SCC
SCC
SCC
SCC
SCC
SCC
SCC
SCC
UN
SCC
SCC
SCC
SCC
Ver
SCC
SCC
SCC
Ver

Y: exposed; : not exposed; XRT: prior radiation therapy; SCC: squamous carcinomas; Dys: dysplasia; UN: unclassified epithelial neoplasm; Ver: verrucous carcinoma.
1
P: primary tumor; M: metastasis.
2
Tumor grade: W 5 well differentiated; M 5 moderately differentiated; P 5 poorly differentiated.

each specimen was used directly for the polymerase

chain reaction (PCR).

Single Strand Conformation Polymorphism

Single strand conformation polymorphism (SSCP)
analysis was performed according to a modified version of a previously reported method.21 Exons 5 8 of
the p53 gene were amplified by PCR. A nested (twostep) protocol was followed for more efficient DNA
amplification. The first step of the PCR reactions
was performed in a volume of 25 mL containing 15
pmol of each primer, 400 uM dNTPs (deoxynucleoside triphosphate), 10 mM Tris-Cl (pH 8.3), 50 mM
KCl, 1.5 mM MgCl2, 0.01% gelatin, and 0.75 U of Taq

polymerase (Perkin-Elmer-Cetus, Foster City, CA).

Samples were overlaid with 25 mL of mineral oil. The
primers used for the amplification of p53 exons 5 8
in this step were P5500 and P530 for exon 5,
P6 500 and P6 3 for exon 6, P750 and P730 for
exon 7, and P8 500 and P8 3 for exon 8. Twentyeight cycles were used for amplification, consisting
of 1 minute at 94C, 1 minute and 40 seconds for
annealing, and 1 minute at 72C for extension. All
the reactions were performed using an automated
DNA thermal cycler (Perkin-Elmer-Cetus).
Second-step reaction PCRs were performed using 1 mL from the first-step reaction as the template.
PCR reactions were carried out in a final volume of

Tobacco and Genes in Oral Tumors/Xu et al.

10 mL; the composition was as described above,

except for the addition of 5 uCi of a-32P-dNTP (Dupont/NEN products), 6 pmol of each primer, 100
mM dNTPs, and 6 U of Taq polymerase. The primers
used in this step were P550 and P531 for exon 5,
P6 50 and P6 3 for exon 6, P750 and P731 for
exon 7, and P8 51 and P8 3 for exon 8. Fifteen
cycles were used for amplification; these cycles consisted of 40 seconds at 94C, 40 seconds for annealing, and 40 seconds at 72C for extension. The reaction mixture was diluted (1:10) in 0.1% sodium
dodecyl sulfate/10 mM ethylenediaminetetraacetic
acid (EDTA) and then mixed 1:1 with a solution
containing 20 mM EDTA, 95% formamide, 0.05%
bromophenol blue, and 0.05% xylene cyanol. Samples were heated at 95C for 5 minutes, chilled on
ice, and immediately loaded (6 mL) onto a 6% acrylamide/Tris-borate-EDTA gel containing glycerol.
Gels were run at 4 W for 1215 hours at room temperature. Autoradiography was performed for 2 6
hours at room temperature. Genomic DNA from
control samples containing known wild-type and
mutant p53 alleles were processed in parallel in
every assay.

DNA Sequencing of p53 and H-ras Genes

Samples that showed a band shift in the SSCP analysis
were selected for DNA sequencing. One mL aliquot of
the first-step PCR products of the PCR-SSCP assays
were used as templates for second-step PCR. For these
separate reactions, the primers and conditions were as
described above for second-step PCRs, except that the
amplification was carried out for 30 cycles and without the addition of radiolabeled nucleotide. Amplified
products were purified with a Magic PCR minicolumn
(Promega, Madison, WI) before sequencing reaction.
Direct sequencing of PCR products was performed
using the AmpliTaqR Sequencing Kit (Perkin-ElmerCetus) according to the manufacturers specifications.
The PCR primers were used as sequencing primers,
and wild-type p53 alleles were sequenced in parallel as
control samples. Unlike p53, H-ras gene sequencing
was directly performed after PCR amplification without SSCP. Two-step PCR reactions were performed.
The conditions for the reactions were similar to p53
gene amplification. In the first-step PCR reaction,
primers 5H-ras 611 and 3H-ras 612 were used for
codon 61, and 5H-ras 121 and 3H-ras 122 were used
for codon 12 and 13. Primers 5H-ras 613, 3H-ras
61 4, 5H-ras 121, and 3H-ras 123 were used for
codon 61 and codon 12 in second-step reactions. After
purification of PCR products, direct sequencing of the
H-ras gene was performed. Mutations in codon 12,13,

207

and 61 of the H-ras gene were checked by comparison

with wild-type sequences.

RESULTS
Patient and Specimen Characteristics
There were 25 male and 4 female patients, ages 34 86
years (mean, 60 years), at the time of resection (Table
1). All patients had a history of considerable current or
previous use of tobacco. This was categorized as positive for smoking if the patient had smoked for at least
20 pack-years prior to resection, and positive for use of
smokeless tobacco if the patient reported use of snuff
or chewing tobacco. Alcohol use was categorized as
positive if the patient reported consuming more than
7 drinks per week (21 patients). Of these 21 patients,
20 also smoked.
Seven patients had been treated with radiation to
the area of the specimen prior to resection. None had
had chemotherapy prior to resection.
Histologic types of disease included squamous
carcinoma (28 specimens), verrucous carcinoma (2),
unclassified epithelial neoplasm of uncertain malignant potential (1), and dysplasia (2). Thirteen of the
tumors were well differentiated, 14 moderately differentiated, 3 poorly differentiated, and 1 not
graded. Twenty-six were classified as primary tumors on the MDACC pathology report, and four as
metastases; for one, classification was not possible.
Of the primary tumors and dysplasia, 9 were resected at least 5 months after the initial diagnosis
and surgery (maximum, 11.9 years; 9 were resected
at least 5 months after the initial diagnosis and
surgery (maximum, 11.9 years; median, 10 months),
and the other 20 were from, or within 2 months of,
the initial surgery.

Immunohistochemical Studies of p53, Cyclin D1, and Rb

in Human Oral Carcinomas
p53, Cyclin D1, and Rb immunostaining were performed on 33 tumor and dysplasia specimens from
patients with oral carcinomas (Table 2).
Twenty tumors had p53 protein accumulation
(65%), 12 showed positive staining for cyclin D1 (39%),
and only 3 (10%) were negative for Rb. In the two cases
of severe dysplasia, no staining for p53 or cyclin D1
was observed, and both were positive for Rb. Representative examples of p53, cyclin D1, and Rb immunostaining in positive or negative carcinomas are
shown in Fig. 1.
In Table 2, the specimens are categorized according to the patients tobacco and alcohol use and prior
radiation treatment. We observed a 1.71.8-fold increase in the frequency of p53 protein accumulation,
which was associated with smokeless tobacco use only

208

CANCER July 15, 1998 / Volume 83 / Number 2

TABLE 2
Immunostaining and Gene Mutations
p53
Patient

Sample

Mutation

Type

Codon

Nucleotides

Amino acids

Staining

Cyclin
D1

Rb

1001
1002
1002
1003
1004
1005
1005
1006
1007
1008
1009
1010
1010
1011
1012
1013
1014
1015
1015
1016

1
2
3
4
5
6
7
8
10
11
12
13
14
15
16
17
18
19
20
21

1
2
1
2
2
1
1
2
2
1
1
2
2
nd
nd
1
1
nd
1
1

250

CCC3TCC

PRO3SER

175

CGC3TGC

ARG3CYS

V
V

185
191

AGC3ATC
CCT3ACT

SER3ILE
PRO3THR

V
V

155
181

ACC3AAC
CGC3AGC

THR3ASN
ARG3SER

S
S

156
220

CGC3CAC
TAT3TGT

ARG3HIS

V
V

181
191 and 155

CGC3AGC
CCT3ACT and
ACC3AAC

ARG3SER
PRO3THR and
THR3ASN

111
111
111
111
11
11
11
2
2
111
2
2
2
2
1
2
1
2
1
111

2
2
2
2
111
2
2
1
2
111
2
2
11
2
2
2
2
111
1
111

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1017
1018
1019
1020
1021
1022
1023

22
23
24
25
26
27
28

2
2
2
nd
1
1
1

2
1
2
1
11
11
11

2
2
11
2
11
2
11

1
1
1
2
1
1
1

1024
1025
1026
1027
1028
1029

29
30
31
32
33
34

2
2
1
1
2
2

2
2
11
111
2
1

2
11
2
2
11
2

1
1
1
2
1
2

S
S
V

V
S

169
161
191 and 192

185
150

TCA3TTA
GCC3ACC
CCT3ACT and
CAG3CAT

AGC3ATC
ACA3ATA

SER3LEU
ALA3THR
PRO3THR and
GLU3HIS

SER3ILE
THR3ILE

1 1: positive for mutation; 2: negative for mutation; nd: not done; S: transition; V: transversion.

(P 5 0.10), with moderate or no alcohol consumption

(P 5 0.07), and with not having had radiation treatment (P 5 0.05). None of the other parameters were
different in the various risk groups.
Accumulation of p53 protein and overexpression
of cyclin D1 in oral mucosa adjacent to the tumors
were evaluated in 10 cases. Cyclin D1 and p53 staining
were absent in the adjacent mucosa. In contrast, Rb
staining was always present in the basal layer of the
adjacent mucosa (data not shown).

Mutation Analysis of p53 and H-ras Genes in Oral

Carcinomas
All 33 specimens were examined by SSCP analysis and
DNA sequencing. Point mutations (resulting in an

amino acid substitution in the p53 protein) were detected in 14 tumors, and 1 case of dysplasia and each
mutation was found to be a missense. Eleven mutations were in exon 5 (Fig. 2), five were in exon 6, and
one was in exon 7. Two mutations were found in
Specimens 21 and 28 in exons 5 and 6. The p53 protein
accumulation and gene mutations are summarized in
Table 2. Both gene mutation and protein accumulation occurred in 13 specimens, whereas protein accumulation without detected mutation was observed in
5 specimens. Two specimens, one tumor and one
dysplasia, had mutations but no protein accumulation. Sequencing data was not available for three specimens (two with protein accumulation).
In this study we also examined the mutations of

Tobacco and Genes in Oral Tumors/Xu et al.

209

FIGURE 1. Immunohistochemical staining of human oral carcinomas using p53, cyclin D1, and Rb antibodies is shown: (A) p53 positive carcinoma; (B) p53
negative carcinoma; (C) cyclin D1 positive carcinoma; (D) cyclin D1 negative carcinoma; (E) Rb-positive carcinoma; (F) Rb negative carcinoma. The mucosa
surrounding the tumor showed positive Rb staining in cell nuclei, but the carcinoma under the mucosa is negative. (All immunohistochemical reactions were
performed using dimethylaminobenzene in the chromogen and counterstained with hematoxylin; original magnification 3150).

the H-ras gene in codons 12, 13, and 61 in 30 oral

carcinomas. No mutations were found, which was
particularly relevant to patients who had used smokeless tobacco for several years (data not shown).

No association between p53 mutation and any

specific type of tobacco use was found. We found no
association between p53 mutation and either tumor
grade and stage or other risk factors.

210

CANCER July 15, 1998 / Volume 83 / Number 2

FIGURE 2. DNA sequencing analysis of

p53 mutations in human oral carcinomas
is shown. (A) Single strand conformation
polymorphism analysis of exon 5 is
shown. Numbers represent sample numbers from Table 1. wt: wild-type; p: positive control. The BT-20 cell line is a positive control for exon 5. The band shift can
be seen in Samples 25 and 32. (B) DNA
sequence of p53 mutation in exon 5 is
shown. Sequence is shown for Sample 25,
codon 183, and Sample 32, codon 150.

DISCUSSION
In this study, we have investigated alterations of p53,
cyclin D1, Rb, and H-ras genes in HNSCC from patients with documented exposure to smoke and
smokeless tobacco. The purpose of the study was to
determine alterations in these common genetic alteration in the same set of tumors and explore the possibility that exposure to different risk factors (i.e.,
smoke and smokeless tobacco) may change the profile
of the genetic alterations in these tumors.
High p53 mutation frequencies and protein accumulation have been reported previously in HNSCC
cell lines and tumor cells.16,23,24 This longer half-life of
the mutant p53 protein results in the accumulation of
this phosphoprotein in the nuclei, facilitating its detection by immunohistochemical analysis with specific antibodies. In this study, using a polyclonal rabbit
antibody to human p53 (CM-1), we found that 59% of
oral carcinomas showed nuclear immunoreactivity.
To define further the nature of p53 mutations in
this group of oral carcinomas, SSCP screening and
DNA sequencing were performed. Seventeen missense
mutations in the p53 gene were detected. All but 1
occurred at guanine residues; 64% (11 of 17) were
G3 T transversions, and 35% (5 of 17) were G3 A
transitions. This pattern of mutations agrees with the
results of previous molecular epidemiologic studies of
human tobacco-related cancers.2527 Furthermore, in
vitro experiments suggest that specific mutagens in
tobacco (benzo[a]pyrene and N-nitrosamines) may
form adducts predominantly with guanine residues in
the p53 gene and induce a high frequency of G3 T
mutations.26
Our study shows inconsistencies between p53
gene mutation and protein accumulation. Similar inconsistencies have been documented previously.28,29
Our data showed that p53 gene mutation was absent
in five tumor samples with accumulation of p53 protein. On the other hand, missense mutations in the
p53 gene were noted in two tumors that did not ex-

press detectable p53 protein. There are several possible explanations for this inconsistency. The tumor
cells without p53 mutations and with protein accumulation may have had p53 mutations outside exons 5 8.
Alternatively, the protein accumulation may be due to
the stabilization of wild-type p53 protein by binding to
another cellular protein in these tumors.
As previously reported, our study also showed a
high frequency of cyclin D1 overexpression in HNSCC.
Our immunochemical results also indicated that accumulation of p53 and overexpression of cyclin D1 did
not occur together in 62% (21 of 34) of the tumors
(Table 2). Positive staining of both proteins occurred
in 6 tumors (18%), and 5 tumors (15%) showed no
detectable p53 or cyclin D1 (Table 2). Although our
findings were based on a limited number of cases,
they suggest that during oral carcinogenesis there is
no selective advantage for cells to sustain both genetic
alterations, overexpression of cyclin D1, and mutation
of the p53 gene. It has been shown that p53 negatively
regulates cell cycle progression through WAF-1/CIP-1,
an inhibitor of cyclin-dependent kinases (CDKs).30
WAF-1 associates with and inhibits multiple CDKs,
including cyclin D dependent kinase, and so prevents
the cell from exiting G1. In tumors, loss of wild-type
p53 function deactivates this growth-control pathway
and is thought to play a role in the uncontrolled
growth of tumor cells. In contrast, cyclin D1 acts as
positive regulator of the transition from G1-phase to
S-phase by association with and activation of cyclin
D1 dependent kinase.11 Therefore, loss of wild-type
p53 function and overexpression of cyclin D1 may
have a similar effect on the cell cycle, namely, abrogation of G1 growth control.
In this investigation, a monoclonal Rb antibody
(Mb-4) was used to detect the absence of nuclear Rb
protein in tumor cells. Only 3 of 34 oral carcinomas
lacked detectable Rb protein. Moreover, most of the
tumors that overexpressed cyclin D1 still expressed Rb
protein (Table 2). Similar results were reported for

Tobacco and Genes in Oral Tumors/Xu et al.

211

TABLE 3
Molecular Changes, Use of Tobacco and Alcohol, and Prior Radiation Treatment

Use
Tobacco
Smoke only
Smokeless only
Both
Alcohol
Yes
No
Radiation
Yes
No
Total
1
2

Samples
n (%)

p53
Mutation
n (%)

p53
Staining
n (%)

Cyclin D1
Staining
n (%)

Rb
Absence
n (%)

H-ras1
Mutation
n (%)

16
4
13

6/14 (43)
2/4 (50)
7/11 (64)

9 (56)
4 (100)
6 (54)

6 (38)
1 (25)
5 (38)

1 (6)
1 (25)
1 (8)

0
0
0

25
8

12/22 (55)
3/7 (43)

13 (52)
7 (88)

9 (36)
3 (38)

2 (8)
1 (12)

0
0

7
26
33

4 (57)
11 (50)
15/29 (52)

2 (29)2
18 (69)
20 (61)

3 (43)
9 (26)
12 (36)

1 (14)
2 (8)
3 (9)

0
0
0

H-ras mutation was examined in 30 samples.

Significant at P50.05 (chi-square test).

human esophageal carcinoma.31 The low frequency of

Rb negative tumors was consistent with previous results showing a low frequency of Rb alterations in
HNSCC.17
Overexpression of the cyclin D1 gene in tumors
perturbs G1-related events, at least in part by interaction with the Rb tumor suppressor protein, which has
been implicated in negative control of cell cycle progression.32 Previous studies have demonstrated that
cotransfection of cyclin D1 with Rb can partially override the cell growth inhibition obtained by transfection of Rb only into Rb negative cell lines.33 A recent
study of human esophageal tumors suggested that the
Rb regulatory pathway may be disrupted by either Rb
inactivation or overexpression of cyclin D1.31 It is
thought that in some neoplasms the Rb pathway is
altered indirectly by cyclin D1 overexpression instead
of directly by Rb gene inactivation.
We examined H-ras mutations in 30 oral carcinomas related to tobacco use. No mutations were found
in codons 12, 13, or 61. A low prevalence of H-ras
mutations in oral malignancies has been reported in
several studies.8,33 However, a study conducted in India found very frequent H-ras mutations in primary
oral carcinomas associated with betel chewing.22 This
may be directly related to the process of betel chewing, in which a high concentration of carcinogens is
placed in direct contact with the oral mucosa, as they
are in tobacco chewing and snuff use. The use of
smokeless tobacco also exposes the oral mucosa to
concentrated levels of tobacco carcinogens for longer
periods of time than cigarette smoking. There are also
quantitative and qualitative differences between the
carcinogens present in tobacco smoke and unburned

tobacco extract.34 For these reasons, we have examined in this study human oral carcinomas from patients who used smokeless tobacco alone or in combination with cigarettes. It is noteworthy that the
profile of p53, cyclin D1, Rb, and H-ras in users of
smokeless tobacco was similar to that of patients who
smoked (Table 3). Furthermore, none of those cases
had H-ras mutations. This suggested that the H-ras
mutation in the squamous cell carcinomas from India
are probably due to specific carcinogens in betel. In
contrast, smoking tobacco seems to be associated with
ras mutations in unstratified epithelium rather than in
stratified epithelium.8
In summary, we investigated alterations of the
p53, cyclin D1, Rb, and H-ras genes in oral carcinomas
related to tobacco use. Our results suggest that p53
mutations and cyclin D1 overexpression are common
genetic events in this type of tumor. Although the Rb
and H-ras genes are involved in tumorigenesis in several different human cancers, Rb inactivation and Hras gene mutations are infrequent in oral carcinomas.

REFERENCES
1. Bishop M. Molecular themes in oncogenesis. Cell 1991;64:
235 48.
2. Marshall C, Tumor suppressor genes. Cell 1991;64:31326.
3. Caamano J, Zhang SY, Rosvold EA, Bauer B, Klein-Szanto
AJP. p53 alterations in human squamous cell carcinomas
and carcinoma cell lines. Am J Pathol 1993;142:11319.
4. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science 1991;253:49 53.
5. Horowitz JM, Park SH, Bogenmann E, Cheng JC, Yandell
DW, Kaye FJ, et al. Frequent inactivation of the retinoblastoma anti-oncogene is restricted to a subset of human tumor cells. Proc Natl Acad Sci U S A 1990;87:27759.

212

CANCER July 15, 1998 / Volume 83 / Number 2

6. Bos JL. The ras gene family and human carcinogenesis.

Mutat Res 1988;195:25571.
7. Rodenhuis S, Slebos RJC, Boot AJM, Evers SG, Moozz WJ,
Wagenaar SS, et al. Incidence and possible clinical significance of K-ras oncogene activation in adenocarcinoma of
human lung. Cancer Res 1988;48:5738 41.
8. Chang SE, Bhatia P, Johnson NW, Morgan PR, McCormick F,
Young B, et al. Ras mutations in United Kingdom: examples
of oral malignancies are infrequent. Int J Cancer 1991;48:
409 12.
9. Sakai E, Rikimaru K, Ueda M, Matsumoto Y, Lshii N, Enomoto S, et al. The p53 tumor-suppressor gene and ras oncongene mutations in oral squamous-cell carcinoma. Int J
Cancer 1992;52:86772.
10. Motokura T, Bloom T, Kim HG, Ruderman JV, Kronenkery
HM, Arnold A. A novel cyclin encoded by bc101-linked
candidate oncogene. Nature 1991;350:5125.
11. Sherr CJ. Mammalian G1 cyclins. Cell 1993;73:1059 65.
12. Jiang W, Kahn SM, Zhou P, Zhang YJ, Caccace AM, Infante
AS, et al. Overexpression of cyclin D1 in rat fibroblasts
causes abnormalities in growth control, cell cycle progression and gene expression. Oncogene 1993a;8:344757.
13. Jares P, Fernandes PL, Campo E, Nadal A, Bosch F, Cardesa
A. PRAD-1/Cyclin D1 gene amplification correlates with
messenger RNA overexpression and tumor progression in
human larygeal carcinomas. Cancer Res 1994;54:48137.
14. Callender T, El-Naggar AK, Lee MS, Frankenthaler R, Luna
MA, Batsakis J G. PRAD-1(CCND1)/cyclin D1 oncogene amplification in primary head and neck squamous cell carcinoma. Cancer 1994;74:152 8.
15. Maestro R, Dolcetti R, Gasparotto D, Doglioni C, Pelucchi S,
Barzan L, et al. High frequency of p53 gene alterations
associated with protein overexpression in human squamous
cell carcinoma of larynx. Oncogene 1992;7:1159 66.
16. Ahomadegbe JC, Barrois M, Fogel S, Bihan MLL, Douc-Rasy
S, Duvillard P, et al. High incidence of p53 alterations (mutation, deletion, overexpression) in head and neck primary
tumors and metastases: absence of correlation with clinical
outcome. Frequent protein overexpression in normal epithelium and in early non-invasive lesions. Oncogene 1995;
10:121727.
17. Yoo GH, Xu HJ, Brennan JA, Westra W, Hruban RH, Koch W,
et al. Infrequent inactivation of retinoblastoma gene despite
frequent loss of chromosome 13q in head and neck squamous cell carcinoma. Cancer Res 1994;54:4603 6.
18. Cattoretti G, Pileri S, Parraviccini C, Becker MHG, Poggi S,
Bifulco C, et al. Antigen unmasking of formalin-fixed, paraffin-embedded tissue sections. J Pathol 1993;171:8398.
19. Robles AI, Conti CJ. Early overexpression of cyclin D1 protein in mouse skin carcinogenesis. Carcinogenesis 1995;16:
781 6.
20. Gimenez-Conti IB, Bianchi AB, Stockman SL, Conti CJ, Slaga
TJ. Activating mutation of the H-ras gene in chemically
induced tumors of the hamster cheek pouch. Mol Carcinog
1992;5:259 63.

21. Orita M, Susuki Y, Sekiya T. Rapid and sensitive detection of

point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 1989;5:874 9.
22. Saranath D, Chang SE, Bhoite LT, Panchal RG, Rerr IB,
Mehta AR, et al. High frequency mutation in codons 12 and
61 of H-ras oncongene in chewing tobaccorelated human
oral carcinoma in India. Br J Cancer 1991;63:573 8.
23. Xu L, Davidson BJ, Murty VVVS, Sacks PG, Garin-Chesa P,
Schamtz SP, et al. Tp53 gene mutations and ccnp1 gene
amplification in head and neck squamous cell carcinoma
cell lines. Int J Cancer 1994b;59:3837.
24. Field JK, Spandidos DA, Malliri A, Gosney JR, Yiagnisis M,
Stell PM. Elevated p53 expression correlated with history of
heavy smoking in squamous cell carcinoma of the head and
neck. Br J Cancer 1991;64:5737.
25. Suzuki H, Takahashi T, Kuroishi T, Suyama M, Ariyoshi Y,
Takahashi T, et al.p53 mutations in non-small cell lung
cancer in Japan: association between mutations and smoking. Cancer Res 1992;52:734 6.
26. Ruggeri B, Dy M, Zhang SY, Baner B, Goodrow T, KleinSzanto AJP. Benzo[a]pyrene-induced murine skin tumors
exhibit frequent and characteristic G to T mutations in the
p53 gene. Proc Natl Acad Sci U S A 1993;90:10137.
27. Lozano JC, Nakazawa H, Cros MP, Cabral R, Yamasaki H.
G3 A mutations in p53 and H-ras genes in esophageal papillomas induce by N-Nitrosomethylbenzylamine in two
strains of rats. Mol Carcinog 1994;90:339.
28. Thompson AM, Anderson TJ, Contie A, Prosser J, Chetty U,
Carter DC, et al. p53 allele losses, mutations and expression
in breast cancer and their relationship to clinico-pathological parameters. Int J Cancer 1992;50:528 32.
29. Xu L, Chen YT, Huvos AG, Zlotolow IM, Rettig WJ, Old LJ, et
al. Overexpression of p53 proteins in squamous cell carcinomas of head and neck without apparent gene mutations.
Diagn Mol Pathol 1994a;3:8392.
30. El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsous R,
Trent JM, et al. WAF-1, a potential mediator of p53 tumor
suppression. Cell 1993;75:81725.
31. Jiang W, Zhang YJ, Kahn SM, Hollstein MC, Santella RM, Lu
SS, et al. Altered expression of the cyclin D1 and retinoblastoma genes in human esophageal cancer. Proc Natl Acad Sci
U S A 1993b;90:9026 30.
32. Weinberg RA. Tumor suppressor genes. Science 1991;254:
1138 46.
33. Hinds PW, Mittnacht S, Dulic V, Arnold A, Reed SI, Weinberg
RA. Regulation of retinoblastoma protein functions by ectopic expression of human cyclins. Cell 1992;70:9931006.
34. Hoffmann D, Brunnemann KD, Adams JD, Hecht SS. Formation and analysis of N-nitrosamines in tobacco products
and their endogenous formation in consumers. In: ONeill
IK, von Borstel RC, Miller CT, Long J, Bartsch H, editors.
N-Nitroso compounds: occurrence, biological effects and
relevance to human cancer. Lyon: International Agency for
Research on Cancer, 1984:743.