Professional Documents
Culture Documents
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.
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
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.
207
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.
208
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.
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
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).
210
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
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
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