Oral Med Pathol 13 (2009) 151

The absence of significant mutational events of the p53 gene in the

only two salivary gland tumors possessing radiation-related
development risks, mucoepidermoid carcinoma and Warthin tumor

Mayumi Abé1, Satoshi Maruyama2, Manabu Yamazaki1, Takanori Kobayashi2, Kamal Al-Eryani1,
Ahsan M. Shahidul1, Masayuki Tsuneki1, Mei Syafriadi1, Takashi Saku1, 2, Jun Cheng1
1
Division of Oral Pathology, Department of Tissue Regeneration and Reconstruction, Niigata University Graduate School of Medical and
Dental Sciences, Niigata, Japan
2
Oral Pathology Section, Department of Surgical Pathology, Niigata University Hospital, Niigata, Japan

Abstract: Since the risks of development of salivary mucoepidermoid carcinoma and

Warthin tumor have been significantly greater among atomic bomb survivors in a
dose-dependent manner, the p53 gene, an important proto-oncogene whose mutation
has been related to radiation, was analyzed by DNA sequencing and immuno­
histochemistry in 37 cases of mucoepidermoid carcinoma collected from Niigata and
Nagasaki and in 33 cases of Warthin tumor collected from Niigata. Immunohis­
tochemically, p53 gene products were heavily demonstrated in most of the tumor cell
nuclei of mucoepidermoid carcinomas but not in Warthin tumors. Mucoepidermoid
carcinomas had some point mutations (codons 136-137, 144, 232, 234, and 241) but
their incidences among the samples were not significantly high (2.7%-10.8%). In
contrast, three point mutations (codons 143, 151, and 229) were commonly found in
the Warthin tumor cases (80%-87%), but the former two mutations did not alter their
amino acid composition. Thus, there were no p53 mutations which were shared by the
two tumors. However, the mutations at exon 5 of mucoepidermoid carcinomas were
significantly higher in the cases from Nagasaki than those from Niigata, although
their highest frequencies at most were around 10%. The results suggest that point
mutations of p53 gene, as far as exons 5-7 were concerned, do not play any obviously
important roles in the radiation-based tumorigenetic processes shared by mucoepi­
dermoid carcinoma and Warthin tumor.
 [Oral Med Pathol 2009; 13: 151-158 doi: 10.3353/omp.13.151]

Key words: atomic bomb, mucoepidermoid carcinoma, p53, salivary gland, Warthin
tumor

Correspondence: Jun Cheng, Division of Oral Pathology, Department of Tissue Regeneration

and Reconstruction, Niigata University Graduate School of Medical and Dental
Sciences. 2-5274 Gakkocho-dori, Chuo-ku, Niigata 951-8514, Japan
Phone: +81-25-227-2832, Fax: +81-25-227-0805, E-mail: [email protected]

carcinomas and Warthin tumors, incidences of which were

Introduction shown to be dependent on the radiation doses (14-15).
Little is known about the causes of salivary gland It has generally been accepted that mutational events in
tumors. One of the most frequently investigated etiologic proto-oncogenes play important roles in human tumori­
relationships concerns the association of Epstein-Barr viral genesis. Among the proto-oncogenes, p53 gene has been
(EBV) infections with lymphoepithelial carcinomas, extensively investigated for its functional changes due to
although the pathogenetic mechanism of EBV still remains alteration in various kinds of human experimental animal
poorly understood (1-3). Another line of evidence reported tumors (16-17). However, there has been a significant
for salivary gland tumorigenesis is the later effect of amount of information about p53 mutations in salivary gland
exposure to ionizing radiation by atomic bombs (4-7) and tumors, and their important role in their pathogeneses has
therapeutic radiation (8-13). Based on the survey of atomic been indicated (18-33). In addition, there have been a large
bomb survivors in Hiroshima and Nagasaki, Japan, we have number of reports on the relationship between frequent p53
already found a causal role for ionizing radiation in salivary mutations and ultraviolet radiation-induced tumorigeneses
gland tumorigenesis, particularly in mucoepidermoid of human skin squamous cell carcinoma (34) or experimental
152 Abé et al. p53 gene mutation in radiation-related salivary tumors

murine skin tumors (35-36). sections were incubated in 0.002% hydrogen peroxide in
However, it is unknown whether radiation-related methanol to block endogenous peroxidase activities. Then,
salivary gland tumors have p53 mutations. The aim of this the sections were further incubated with 5% skim milk/PBS
study was to investigate mutational conditions of the p53 for 1 hr at 37℃ to block non-specific protein binding.
gene in mucoepidermoid carcinomas and Warthin tumors,
which have been closely related to radiation, in order to Polymerase Chain Reaction (PCR)
elucidate whether or not the developments of these two Total DNA was isolated from paraffin sections of
salivary tumors are dependent on the mutation of the p53 mucoepidermoid carcinomas, Warthin tumors and normal
gene. submandibular gland tissues using the phenol-chloroform
system. PCR was carried out in an Astec thermal cycler
PC-800 (Astec Co., Ltd., Fukuoka, Japan) as follows:
Materials and methods
reaction products of the reverse transcription were diluted
Surgical materials with 1× PCR buffer [50 mM KCl, 10 mM Tris-HCl (pH 8.5),
Thirty-seven surgical specimens of mucoepidermoid 0.01% Triton X-100] to a final volume of 50 μl, which
carcinoma were collected from the surgical pathology files contained 100 ng each of forward oligonucleotide primers
of the Department of Pathology, Faculty of Dentistry, and reverse primers for exons 5-7 of p53, additional dNTPs
Niigata University (21 cases) and Nagasaki University (final concentration of 0.2 mM), and 2.5 units of Taq DNA
Hospital (16 cases) during a 34-year period from 1965 to polymerase (Takara Bio Inc., Otsu, Japan). The PCR primers
1998 without any selection procedures. The specimens from for human p53 exon 5 were designed as follows: the sense
Nagasaki included those of atomic bomb survivors, although primer was 5̓-GTTTC TTTGC TGCCG TGTTC-3̓
individual information about the bombing victims was not (12972-12991), and the antisense primer was 3̓-ACGGG
available due to personal information protection regulations. TCCCA GGGGT CCGGA-5̓ (13275-13294). Those for
Thirty-three specimens of Warthin tumor were collected exon 6 were 5̓-TGGTT GCCCA GGGTC CCCAG-3̓
from only the Niigata University files during the same (13271-13290, forward) and 3̓-ACCAA CAGTCA CCGGG
period. Surgical materials were fixed in 10% formalin and AGG-5̓ (13475-13493, reverse). Those for exon 7 were
routinely processed for embedding in paraffin. Serial 5̓-CTTGC CACAG GTCTC CCCAA-3̓ (13941-13960,
sections cut at 4 μm were stained with hematoxylin-eosin forward) and 3̓-CGGTG AACGG TGGGA CGTGT-5̓
(HE), and stained immunohistochemically with the antibody (14117-14136, reverse). The thermocycling protocol during
described below. The paraffin sections were also used for 35 amplification cycles after denaturation at 94℃ for 4 min
DNA extraction. For control studies, 5 surgical specimens of was as follows: denaturation at 94℃ for 1 min; annealing at
normal submandibular gland obtained in radical neck 60℃ (62℃, exon 5) for 1 min; extension at 72℃ for 1 min;
dissections were used in the same manner as described and termination with a final cycle which involved annealing
above. The experimental protocol for isolation and analyses (60℃ or 62℃, exon 5, for 1 min) and extension (72℃ for 7
of tumor cells was reviewed and approved by the Niigata min). An exponential amplification for p53 was confirmed in
University Graduate School of Medical and Dental Sciences the 35 cycles. The amplified DNA fragments were analyzed
Ethical Board. by electrophoresis on 3% agarose gels (26).

Antibody Direct Sequencing of PCR products for p53

Bp53-11, a mouse monoclonal antibody (IgG2a) against All PCR products were subjected to cycle sequencing by
human p53 gene product, which recognizes the NH2-terminal using Thermo Sequenase Core Sequencing Kits with
domain, was purchased from Progen Biotechnik GmbH 7-deaza-dGTP (GE Healthcare Ltd./Amersham, Bucking­
(Heidelberg, Germany). hamshire, UK). The sequence primers were synthesized
based on the published data and labeled with Texas red at the
Immunohistochemistry 5̓-end by using the 5̓ Oligonucleotide Texas Red Labeling
The avidin-biotin peroxidase complex (ABC) technique, Kit (Amersham). The labeled primers for p53 exon 5 were
using biotinylated rabbit anti-mouse IgG (1:500, Dako, 5̓-TTCAA CTCTG TCTCC TTCCT-3̓ (forward); those for
Glostrup, Denmark) and peroxidase-conjugated streptavidin exon 6 were 5̓-GCCTC TGATT CCTCA CTGAT-3̓
complex (1:500, DAKO), was employed for immuno­ (forward); and those for exon 7 were 5̓-CTTGC CACAG
histochemical staining for the p53 gene products (P53) (37). GTCTC CCCAA-3̓ (forward). One reaction mixture
Briefly, sections were pretreated in 0.01M sodium citrate contained 3 μl of premixes (appropriate nucleotides/reaction
(pH 6.0) for 10 min at 120℃ (wet autoclave) to disclose buffer/Thermo Sequenase DNA polymerase); 1 μl of tem­
antigenic sites. The primary antibody was diluted to a plate PCR products, which were purified with GFX PCR
concentration of 50 ng/ml. For visualization of reaction DNA and Gel Band Purification Kits (Amersham); and 2 μl
products, sections were treated with 3.3̓-diaminobenzidine (2 pM) of Texas red-labeled primers. After denaturation at
in the presence of 0.05% hydrogen peroxide, and the sec­ 95℃ for 5 min, the reaction mixture was placed on a thermal
tions were counterstained with hematoxylin. For the control cycler for 25 cycles of denaturation at 95℃ for 30 sec and
sections, the primary antibody was replaced with normal annealing/extension at 60℃ for 30 sec. The reaction
mouse IgG. Prior to the reaction with the primary antibody, products were dissolved in 3 μl loading dye by vortexing,
 Oral Med Pathol 13 (2009) 153

and then the mixture was concentrated with vacuum directions of differentiation to mucous cells and squamous
desiccators. Then, 3 μl of samples for each lane were loaded epithelial cells. Their carcinoma cell nests also varied from
on a gel (7% Long Ranger (Nuseive FMC Bioproducts, solid to glandular or large-cystic with mucous contents.
Rocland, ME, USA)/6.1 M urea/1.2 × TBE buffer (10 mM Bizarre and hyperchromatic nuclei tended to be more often
Tris, 10 mM boric acid, and 2 mM EDTA)). The observed among squamous epithelial cells or mucous cells
electrophoresis was performed in a fluorescent DNA around cystic structures but not so frequently in regularly
sequencer (SQ-5500-S, Hitachi Ltd., Tokyo, Japan), and the aligned glandular cells and so-called intermediate cells with
sequencing data were analyzed by using the SQ-5500 clear cytoplasm. The stroma was partially hyaline, while
analysis software, ver. 3.03 (Hitachi) (2, 38). most of it was fibrous with a recognizable amount of
fibroblastic stromal cells (Fig. 1a). Immunohistochemically,
Statistical analysis positive staining for P53 was obtained in most of the cases
Statistical analysis was performed using Fisher̓s exact examined. The reaction products were restricted to tumor
test for the incidental difference of p53 mutational events in cell nuclei. In particular, atypical nuclei showed stronger
mucoepidermoid carcinomas between Nagasaki and Niigata staining intensities. However, extremely bizarre nuclei did
patients. A P-value of less than 0.05 was considered not show positive reactions for P53 (Fig.1b). All of the
statistically significant. examined cases showed the same tendency.
The histology of Warthin tumors was characterized by
glandular structures composed of a two-cell layer of
Results
eosinophilic ductal cells with oncocytic appearances and a
Immunohistochemistry dense lymphocytic stroma scattering lymphoid follicles. The
In mucoepidermoid carcinomas, differentiation of glandular structures were often distended with serous
carcinoma cells varied among cases, showing two major contents with lines of tumor cells infolded in a complicated

a b

Fig. 1. Immunohistochemical local­

ization of P53 in mucoepidermoid
carcinoma and Warthin tumor. (a,
b) Mucoepidermoid carcinoma, (c,
d) Warthin tumor. (a, c) Hematoxylin
and eosin (HE) stain, (b, d) immu­
noperoxidase stain for P53, hema­
toxylin counterstain, × 180. In mu­
coepidermoid carcinoma, P53 was
localized within nuclei of carcinoma
cells. The stainings were more
intensive in bizarre nuclei, while
some of them were not positive. In
Warthin tumor, P53 was not dem­
c d onstrated in tumor cells nor in
stromal cells.
154 Abé et al. p53 gene mutation in radiation-related salivary tumors

exon 5 exon 6 exon 7

413
311 323 Fig. 2. PCR-products for p53 gene,
249 exons 5, 6 and 7 in mucoepidermoid
223 carcinoma. (left) Exon 5, 323-bp,
200 196
200 (middle) exon 6, 223-bp, (right)
151 exon 7, 196-bp. Left lanes in each
column, molecular weight stan­
dards. PCR products of three exons
were successfully obtained from
DNA samples extracted from
paraffin sections of formalin-fixed
surgical materials.

manner, resulting in their papillary appearance (Fig. 1c). were applied for sequenase reactions and sequencing
Immunohistochemically, tumor cells did not show any procedures. The results from mucoepidermoid carcinoma
discernible staining for P53. Stromal lymphoid cells were samples are summarized in Table 1. In exon 5, two point
not positive, either (Fig. 1d). mutations were found. They were CAA CTG → CAG GTG
The results indicated that p53 gene products were over- at codons 136-137 and CAG → TAG at codon 144. These
expressed in mucoepidermoid carcinoma cells but not in mutations led to changes in the amino acid sequence from
Warthin tumor cells. This demonstrated that there seemed to Gln-Leu to Gln-Val and Gln to a stop codon, respectively.
be a significant difference in P53 turnover between benign The replacement of C with T at codon 144 was not always
and malignant tumors. caused by complete deletion of C but by double small peaks
of C and T, indicating a heterogeneous mutation at this point.
PCR and sequencing The incidence of the former was 2.7% among 37 cases and
Based on the histological evidence that over-expression that of the latter was 10.8%. There were no mutational
of p53 gene products took place in mucoepidermoid events in exon 6. In exon 7, three point mutations were
carcinomas, we carried out PCR amplification of the p53 revealed from mucoepidermoid carcinoma samples. They
gene fragments from genomic DNA extracts of paraffin were ATC → AGC at codon 232, TAC → TGC or CAC at
sections which were serial to those used for immuno­ codon 234, and TCC TGC → CC TGC AGT at codon
histochemistry. Five to twenty sections, depending on the 241-242. When translated into amino acids, they should
size, were dewaxed and extracted in a plastic tube. Total have altered from Ile to Ser (3.2%), from Tyr to Cys or His
DNA yields averaged about 16 μg per tube. Under the PCR (6.5%), and from Ser-Cys to nonsense sequences starting
condition described in Materials and Methods, exons 5 to 7 with Pro-Ala (3.2%). The replacement of A with G at codon
were successfully amplified as shown in electrophoresis 234 was not always caused by complete deletion of A,
images of Fig. 2. Single clear bands with molecular masses indicating a heterogeneous mutation at this point. All of
of 323 bp, 223 bp, and 196 bp were obtained for exons 5, 6, these mutations were severe enough to affect their translation
and 7, respectively (Fig. 2). into amino acids. However, their incidence among mucoepi­
For direct sequencing of the PCR products, they were dermoid carcinomas was not conspicuous. In addition, these
further purified with GFX PCR DNA and Gel Band mutations were not always C → T and CC → TT transitions,
Purification Kits. The purified DNA samples of exons 5 to 7 which were considered to be UV radiation specific (34, 36).

Table 1. p53 mutations in 37 cases of mucoepidermoid carcinoma

Exon Codon Bases (Amino acid) Mutation Frequency (%)
5 136-137 CAA CTG CAG GTG 2.7
(Gln Leu) (Gln Val)
5 144 CAG TAG 10.8
(Gln) (stop)
7 232 ATC AGC 3.2
(Ile) (Ser)
7 234 TAC TGC / CAC 6.5
(Tyr) (Cys) / (His)
7 241 TCC TGC _CC TGC AGT 3.2
(Ser Cys) (Pro Ala ........)
 Oral Med Pathol 13 (2009) 155

Table 2. p53 mutations in 33 cases of Warthin tumor

Exon Codon Bases (Amino acid) Mutation Frequency (%)
5 143 GTG GTC / GTA 87
(Val) (Val) / (Val)
5 151 CCC CCT / CCA 87
(Pro) (Pro) / (Pro)
7 229 TGT AGT / GGT 80
(Cys) (Ser) / (Gly)

In Warthin tumors, two point mutations were found from between the two areas.
exon 5. They were located in codons 143 and 151. In the
latter, CCC was changed to CCT or CCA. In the former,
Discussion
GTG was altered to GTC or GTA, and their incidences were
equally 87% of 33 cases. The replacement of G with C at In the present study, the over-expression of p53 gene
codon 143 was not always caused by complete deletion of G products was only obvious in tumor cells of mucoepidermoid
but by the presence of a small G peak, indicating that the carcinoma but not in those of Warthin tumor. This result was
mutation was heterogeneous. Similar to mucoepidermoid not surprising, because the over-expression of p53 has been
carcinomas, there was no exon 6 mutation in Warthin reported to be specific to lesions with cellular proliferation
tumors. In exon 7, there was a highly coincidental (80%) or malignancy (18-25). The over-expression of P53 in
mutation at codon 229, in which TGT (Cys) was replaced mucoepidermoid carcinomas has been well documented
with AGT (Ser) or GGT (Gly). Since the replacement of T with its frequencies ranging from 53% to 67% (25, 28,
with G at codon 229 was associated with incomplete deletion 30-31), while the sequencing data of the p53 gene have been
of T, the mutations were regarded as heterogeneous. rather limited, showing some sporadic point mutations (27,
Although the coincidence of these point mutations was quite 29, 32-33). In physiological conditions, p53 gene products
high, two of the three did not lead to any change in the amino are degraded soon after they are targeted to nuclei (39),
acid sequence. The results from Warthin tumor samples are hence, it is usually hard to detect P53 within a cell by
summarized in Table 2. conventional immunohistochemical methods. The intensive
Mutational events of p53 gene in mucoepidermoid staining for P53 within nuclei of mucoepidermoid carcinoma
carcinoma cases revealed as above were compared between cells thus indicates that the cells produced too much protein
patients from Niigata and Nagasaki, because the cases from in comparison to those degraded by its lyases, which were
Nagasaki should have included those from atomic bomb within normal levels of expression. Another possible
survivors. Since personal information about atomic bomb explanation is that the over-expressed P53 are mutant forms
radiation doses was unfortunately not obtained, the which are resistant to proteolytic cleavages. Thus, it was
comparison itself was not always worth performing. expected that we would analyze the p53 gene for mutations
However, as shown in Table 3, all of the mutations in exon 5 in mucoepidermoid carcinoma specimens with enhanced
were found in 4 patients from Nagasaki only. The mutations expression of its gene products, although it was uncertain
in exon 7 were found in 2 patients from Nagasaki and 1 from whether the over-expressed p53 was functional in G1 arrest
Niigata. The incidence of these point mutations was only of the cell cycling as well as apoptotic pathways (40).
8-10% of the cases, so it was unlikely that p53 mutations There have been no documents in the literature describing
played key roles in the pathogenesis of mucoepidermoid mutational events of the p53 gene in Warthin tumor. In the
carcinoma. However, when patients from Nagasaki and present study, however, we found p53 mutational points in
Niigata were compared, the incidence of mutational events Warthin tumor cases which were highly shared among the
in exon 5 of the p53 gene was significantly higher (P<0.05) sample group, although those point mutations did not affect
in patients from Nagasaki compared to those from Niigata. the amino acid translation so much, indicating that they were
In regard to exon 7, no significant difference was demonstrated just genetic polymorphisms. As shown in the sequencing

Table 3. Comparison of incidence of p53 mutations in mucoepidermoid

carcinomas between patients from Nagasaki and Niigata (%)
Patients from exon 5 exon 6 exon 7 Total
Nagasaki 25 0 15.4 31.3
(n=16) (4) (0) (2) (5)
Niigata 0 0 5.6 4.8
(n=21) (0) (0) (1) (1)
Total 10.8 0 23.1 16.2
(n=37) (4) (0) (3) (6)
156 Abé et al. p53 gene mutation in radiation-related salivary tumors

profiles, the mutations were not always homogenous but paid attention to the translocation t(11;19)(q21;p13), which
were heterogeneous in many of the instances. The clinical had been found in mucoepidermoid carcinoma and rarely in
features of Warthin tumors are quite different from those of Warthin tumor, as well (50). However, recently, Fehr et al
other types of salivary gland tumor. They often show have demonstrated that this translocation was not significant
bilaterally synchronous as well as metachronous development in the pathogenesis of Warthin tumors (51). Further
or unilateral multiple developments (41). These develop­ investigations on the detailed molecular basis for genetic
mental characteristics indicate some background of reactive instability are necessary for a better understanding of the
histogenesis, including delayed hypersensitivity or inflam­ molecular basis of radiation-induced tumorigenesis shared
mation in Warthin tumors (14, 42). The present data indicate by mucoepidermoid carcinoma and Warthin tumor.
a possibility that immunological or inflammatory stimuli The frequency and the severity of the p53 mutations in
tend to be associated with some intrinsic background such as mucoepidermoid carcinomas were not so high, although
p53 gene polymorphisms in the pathogenesis of Warthin they were greater than those in Warthin tumor. However,
tumor. their incidence in mucoepidermoid carcinoma among the
Together with Warthin tumors, mucoepidermoid patients from Nagasaki, which contained atomic bomb
carcinomas have been shown to be related to radiation in survivors, was higher than that of the patients from Niigata.
their pathogenesis (14-15). Therefore, these two tumors Unfortunately, it was not possible to collate those mutations
were simply expected to share certain common gene with the radiation doses among the patients from Nagasaki
mutational backgrounds caused by radiation. However, in in the present study. However, the present data suggest that a
contrast to the Warthin tumor data of silent mutations, all of history of irradiation predisposes p53 gene mutations. A
the point mutations in mucoepidermoid carcinoma cases number of studies have dealt with the relationship between
were missense, leading to changes in the amino acid p53 mutations and ultraviolet radiation-induced tumorigenesis
sequence of P53, but each of their frequencies was much of experimental murine skin tumors (35-36, 52) and human
lower than those found in Warthin tumors. The present cancers (34, 53). Although the p53 gene mutations in exons
results clearly showed that there was neither a common 5-7 did not seem to be important in the pathogeneses of
feature of genetic mutation between the two tumors nor mucoepidermoid carcinoma and Warthin tumor in the
UV-radiation specific mutations such as C → T and CC → present study, the definite enhancement of P53 expression in
TT transitions (34, 35), as far as exons 5-7 were concerned. mucoepidermoid carcinoma indicates at least a metabolic
In addition, the result that obviously atypical mucoepidermoid disturbance of p53 gene products in mucoepi­dermoid
carcinoma cells were not always immunopositive for P53 carcinoma cells. It is therefore suggested that some delayed
was noted. This may have been due to the mutated amino gene mutations caused by radiation took place somewhere in
acid sequences of the P53 NH2-terminal region, which could the upstream of p53 cascades, which stimulates over-
not be recognized by the antibody, Bp53-11 (43). expression of P53, although it still remains unknown
Different from the myelogenous or thyroid neoplasms whether the over-expressed P53 participate in the carcino­
that atomic bomb survivors suffered from soon after the genesis of mucoepidermoid carcinomas.
Hiroshima-Nagasaki bombings (5) or the Chernobyl accident
(44), direct and severe gene mutations after high dose
Acknowledgments
radiation exposure have not been considered to be causative
of salivary mucoepidermoid carcinomas and Warthin tumors. The authors are grateful to Dr. Hideo Miyazaki, Niigata
Our previous studies showed that these tumors developed University, for his valuable suggestion to our statistical
after long intervals since irradiation by the atomic bombs, study. They also thank Dr. Nobuo Tsuda, Nagasaki University
although their occurrences were highly dependent on the Hospital, for supplying the paraffin blocks of mucoepider­
radiation doses of the patients (14-15). These two tumors are moid carcinoma. This work was supported in part by
hence typical but rare examples of human diseases caused Grants-in-Aid for Scientific Research from the Japan Society
by delayed gene alterations due to radiation-induced genetic for the Promotion of Science and from the Ministry of
instability (45-46). It is unknown whether the molecular Education, Culture, Sports, Science and Technology, Japan.
mechanisms for such gene alterations develop after a long
interval from irradiation events. However, recent in-vitro
References
studies have demonstrated several lines of evidence for the
delayed gene mutation. Delayed lethal gene mutations were 1. Jen KY, Cheng J, Li J, et al. Mutational events in LMP1
confirmed in Chinese hamster ovary cells or BALB/3T3 gene of Epstein-Barr virus in salivary gland lymphoepi­
cells, which had undergone 30 mean population doublings thelial carcinomas. Int J Cancer 2003; 105: 654-60.
after X-ray irradiation (47). Carls and Schiestl reported 2. Nagao T, Ishida Y, Sugano I, et al. Epstein-Barr virus-
deletion of one specific DNA fragment in mice whose associated undifferentiated carcinoma with lymphoid
parents were irradiated (48), indicating that a high level of stroma of salivary gland in Japanese patients. Comparison
genetic instability was caused by a mutator phenotype with benign lymphoepithelial lesion. Cancer 1996; 78:
transmitted through many cell divisions or over generations. 695-703.
3. Saku T, Cheng J, Jen KY, et al. Epstein-Barr virus infected
More recently, chromosomal instability has also been shown
lymphoepithelial carcinomas of the salivary gland in the
in atomic bomb survivors with leukemia (49). We have also
 Oral Med Pathol 13 (2009) 157

Russia-Asia area: a clinicopathologic study of 160 cases. 21. Gallo O, Franchi A, Bianchi S, Boddi V, Gianelli E, Alajmo
Arch Pathol 2003; 65: 35-9. E. p53 oncoprotein expression in parotid gland carcinoma
4. Belsky JL, Tachikawa K, Cihak RW, Yamamoto T. Salivary is associated with clinical outcome. Cancer 1995; 75:
gland tumors in atomic bomb survivors, Hiroshima- 2037-44.
Nagasaki, 1957 to 1970. JAMA 1972; 219: 864-8. 22. Genetzakis M, Gomatos IP, Georgiou AN, et al. BCL-2,
5. Belsky JL, Takeichi N, Yamamoto T, et al. Salivary gland p53 and HLA-DR antigen expression in surgically treated
neoplasms following atomic radiation: Additional cases and parotid cancer patients. Eur Arch Otorhinolaryngol 2009;
reanalysis of combined data in a fixed population, 266: 417-24.
1957-1970. Cancer 1975; 35: 555-9. 23. Ishii K, Nakajima T. Evaluation of malignant grade of
6. Ohkita T, Takahashi N, Takeichi N, Hirose F. Prevalence of salivary gland tumors: studies by cytofluorometric nuclear
leukemia and salivary gland tumours among Hiroshima DNA analysis, histochemistry for nucleolar organizer
Atomic Bomb Survivors. In: Late Biological Effects of regions and immunohistochemistry for p53. Pathol Int
Ionizing Radiation. Volume 1. International Atomic Energy 1994; 44: 287-96.
Agency, Vienna, 1978; 71-81. 24. Kärjä VJ, Syrjänen KJ, Kurvinen AK, Syrjänen SM.
7. Takeichi N, Hirose F, Yamamoto H, Ezaki H, Fujikura T. Expression and mutations of p53 in salivary gland tumours.
Salivary gland tumors in atomic bomb survivors, Hiroshima, J Oral Pathol Med 1997; 26: 217-23.
Japan. II. Pathologic study and supplementary epidemiologic 25. Lee EJ, Kim J, Lee SA, et al. Characterization of newly
observations. Cancer 1983; 52: 377-85. established oral cancer cell lines derived from six squamous
8. Ju DM. Salivary gland tumors occurring after radiation of cell carcinoma and two mucoepidermoid carcinoma cells.
the head and neck area. Am J Surg 1968; 116: 518-23. Exp Mol Med 2005; 37: 379-90.
9. Katz AD, Preston-Martin S. Salivary gland tumors and 26. Li X, Tsuji T, Wen S, Sobhan F, Wang Z, Shinozaki F.
previous radiotherapy to the head or neck. Report of a Cytoplasmic expression of p53 protein and its morphological
clinical series. Am J Surg 1984; 147: 345-8. features in salivary gland lesions. J Oral Pathol Med 1995;
10. Palmer JA, Mustard RA, Simpson WJ. Irradiation as an 24: 201-5.
etiologic factor in tumours of the thyroid, parathyroid and 27. Matizonkas-Antonio LF, de Mesquita RA, de Souza SC,
salivary glands. Can J Surg 1980; 23: 39-42. Nunes FD. TP53 mutation in salivary gland neoplasms.
11. Preston-Martin S, White SC. Brain and salivary gland Braz Dent J 2005; 16: 162-6.
tumors related to prior dental radiography: implications for 28. Nordkvist A, Röijer E, Bang G, et al. Expression and
current practice. J Am Dent Assoc 1990; 120: 151-8. mutation patterns of P53 in benign and malignant salivary
12. Shore-Freedman E, Abrahams C, Recant W, Schneider AB. gland tumors. Int J Oncol 2000; 16: 477-83.
Neurilemmomas and salivary gland tumors of the head 29. Ohki K, Kumamoto H, Ichinohazama R, et al. Genetic
neck following childhood irradiation. Cancer 1983; 51: analysis of DNA microsatellite loci in salivary gland
2159-63. tumours: comparison with immunohistochemical detection
13. Spitz MR, Batsakis JG. Major salivary gland carcinoma. of hMSH2 and p53 proteins. Int J Oral Maxillofac Surg
Descriptive epidemiology and survival of 498 patients. 2001; 30: 538-44.
Arch Otolaryngol 1984; 110: 45-9. 30. Papadaki H, Finkelstein SD, Kounelis S, Bakker A,
14. Land CE, Saku T, Hayashi Y, et al. Incidence of salivary Swalsky PA, Kapadia SB. The role of p53 mutation and
gland tumors among atomic bomb survivors, 1950-1987. protein expression in primary and recurrent adenoid cystic
Evaluation of radiation-related risk. Radiat Res 1996; 146: carcinoma. Hum Pathol 1996; 27: 567-72.
28-36. 31. Soini Y, Kamel D, Nuorva K, Lane DP, Vähäkangas K,
15. Saku T, Hayashi Y, Takahara O, et al. Salivary gland tumors Pääkkö P. Low p53 protein expression in salivary gland
among atomic bomb survivors, 1950-1987. Cancer 1997; tumours compared with lung carcinomas. Virchows Archiv
79: 1465-75. A Pathol Anat Histopathol 1992; 421: 415-20.
16. Brash DE. Roles of the transcription factor p53 in 32. Suzuki T. p53 abnormality in salivary gland carcinoma and
keratinocyte carcinomas. Br J Dermatol 2006; 154 Suppl 1: its relation to tumor DNA aneuploidy and AgNOR. Nippon
8-10. Jibiinkoka Gakkai Kaiho 1994; 97: 2279-86.
17. Lubet RA, Zhang Z, Wiseman RW, You M. Use of p53 33. Yamamoto Y, Kishimoto Y, Wistuba II, et al. DNA analysis
transgenic mice in the development of cancer models for at p53 locus in carcinomas arising from pleomorphic
multiple purposes. Exp Lung Res 2000; 26: 581-93. adenomas of salivary gland: Comparison of molecular
18. Alves FA, Pires FR, De Almeida OP, Lopes MA, Kowalski study and p53 immunostaining. Pathol Int 1998; 48:
LP. PCNA, Ki-67 and p53 expressions in submandibular 265-72.
salivary gland tumours. Int J Oral Maxillofac Surg 2004; 34. Brash DE, Rudolph JA, Simon JA, et al. A role for sunlight
33: 593-7. in skin cancer: UV-induced p53 mutations in squamous cell
19. Arida M, Barnes EL, Hunt JL. Molecular assessment of carcinoma. Proc Natl Acad Sci USA 1991; 88: 10124-8.
allelic loss in Warthin tumors. Mod Pathol 2005; 18: 35. Benjamin CL, Ullrich SE, Kripke ML, Ananthaswamy HN.
964-8. p53 tumor suppressor gene: a critical molecular target for
20. Deguchi H, Hamano H, Hayashi Y. c-myc, ras and p53 UV induction and prevention of skin cancer. Photochem
expression in pleomorphic adenoma and its malignant form Photobiol 2008; 84: 55-62.
of the human salivary glands. Acta Pathol Jpn 1993; 43: 36. Kanjilal S, Pierceall WE, Cummings KK, Kripke ML,
413-22. Ananthaswamy HN. High frequency of p53 mutations in
158 Abé et al. p53 gene mutation in radiation-related salivary tumors

ultraviolet radiation-induced murine skin tumors: evidence 46. Plumb M, Cleary H, Wright E. Genetic instability in
for strand bias and tumor heterogeneity. Cancer Res 1993; radiation-induced leukemias: mouse models. Int J Radiat
53: 2961-4. Biol 1998; 74: 711-20.
37. Fisher CJ, Gillett CE, Vojtêsek B, Barnes DM, Millis RR. 47. Little JB, Gorgojo L, Vetrovs H. Delayed appearance of
Problems with p53 immunohistochemical staining: the lethal and specific gene mutations in irradiated mammalian
effect of fixation and variation in the methods of evaluation. cells. Int J Radiat Oncol Biol Phys 1990; 19: 1425-9.
Br J Cancer 1994; 69: 26-31. 48. Carls N, Schiestl RH. Effect of ionizing radiation on
38. Levine AJ. The p53 tumor suppressor gene and gene transgenerational appearance of p(un) reversions in mice.
product. Princess Takamatsu Symp 1989; 20: 221-30. Carcinogenesis 1999; 20: 2351-4.
39. Steele RJ, Thompson AM, Hall PA, Lane DP. The p53 49. Nakanishi M, Tanaka K, Shintani T, Takahashi T, Kamada
tumour suppressor gene. Br J Surg 1998; 85: 1460-7. N. Chromosomal instability in acute myelocytic leukemia
40. Moll UM, Schramm LM. p53 - an acrobat in tumorigenesis. and myelodysplastic syndrome patients among atomic
Crit Rev Oral Biol Med 1998; 9: 23-37. bomb survivors. J Radiat Res (Tokyo) 1999; 40: 159-67.
41. Gnepp DR, Schroeder W, Heffner D. Synchronous tumors 50. Enlund F, Behboudi A, Andrén Y, et al. Altered Notch
arising in a single major salivary gland. Cancer 1989; 63: signaling resulting from expression of a WAMTP1-MAML2
1219-24. gene fusion in mucoepidermoid carcinomas and benign
42. Teymoortash A, Werner JA. Tissue that has lost its track: Warthin̓s tumors. Exp Cell Res 2004; 292: 21-8.
Warthin̓s tumour. Virchows Arch 2005; 446: 585-8. 51. Fehr A, Röser K, Belge G, Löning T, Bullerdiek J. A closer
43. Stephen CW, Helminen P, Lane DP. Characterisation of look at Warthin tumors and the t(11;19). Cancer Genet
epitopes on human p53 using phage-displayed peptide Cytogenet 2008; 180: 135-9.
libraries: insights into antibody-peptide interactions. J Mol 52. Jiang W, Ananthaswamy HN, Muller HK, Kripke ML. p53
Biol 1995; 248: 58-78. protects against skin cancer induction by UV-B radiation.
44. Rybakov SJ, Komissarenko IV, Tronko ND, et al. Thyroid Oncogene 1999; 18: 4247-53.
cancer in children of Ukraine after the Chernobyl accident. 53. Ohnishi K, Nagata Y, Takahashi A, Taniguchi S, Ohnishi T.
World J Surg 2000; 24: 1446-9. Effective enhancement of X-ray-induced apoptosis in
45. Niwa O. Delayed effects and induction of genetic instability human cancer cells with mutated p53 by siRNA targeting
by radiation. J Toxicol Sci 1996; 21: 61-3. XIAP. Oncol Rep 2008; 20: 57-61.

Received May 12, 2009 Accepted July 3, 2009