Med Mol Morphol
DOI 10.1007/s00795-013-0058-4
ORIGINAL PAPER
The expression of cytokeratin in keratocystic odontogenic tumor,
orthokeratinized odontogenic cyst, dentigerous cyst, radicular cyst
and dermoid cyst
Kaname Tsuji • Masahiro Wato • Teruyoshi Hayashi • Norihiro Yasuda
Takumi Matsushita • Tomohiko Ito • Shoko Gamoh • Hiroaki Yoshida •
Akio Tanaka • Shosuke Morita
Received: 6 June 2013 / Accepted: 29 August 2013
Ó The Japanese Society for Clinical Molecular Morphology 2013
Abstract The epithelial lining of odontogenic keratocysts
exhibits either parakeratosis or orthokeratosis. In 2005, the
WHO classified odontogenic keratocysts with parakeratosis
as keratocystic odontogenic tumors (KCOT). Odontogenic
keratocysts with orthokeratosis were not classified as odon-
togenic tumors, but instead referred to as orthokeratinized
odontogenic cysts (OOC). To clarify the difference between
these two lesions, we investigated their biological charac-
teristics using immunohistochemical studies for cytokeratins
(CK) in KCOT and OOC as well as in dentigerous cysts
(DC), radicular cysts (RC) and dermoid cysts (DMC). We
examined twenty-five cases of KCOT, fifteen cases each of
OOC, DC and RC, and ten cases of DMC. We studied the
immunohistochemical expression of CK10, 13, 17 and 19.
To evaluate the immunohistochemical staining pattern, we
divided the epithelial lining of the lesions into three layers
(surface layer: su, spinous layer: sp, basal layer: ba). For
CK10, most OOC and DMC specimens of su and sp were
positive. For CK13 and 19, most KCOT, DC and RC spec-
imens of su and sp were positive. For CK17, most KCOT
specimens of su and sp were positive. The percentages of
total CK expression of su and sp, and ba of CK19 differed
K. Tsuji (&)  T. Hayashi  N. Yasuda  T. Matsushita 
T. Ito  H. Yoshida  S. Morita
First Department of Oral and Maxillofacial Surgery, Osaka
Dental University School of Dentistry, 5-17 Otemae 1-chome,
Chuo-ku, Osaka, Osaka 540-0008, Japan
e-mail: tsuji-k@cc.osaka-dent.ac.jp
M. Wato  A. Tanaka
Department of Oral Pathology, Osaka Dental University School
of Dentistry, Osaka, Japan
S. Gamoh
Department of Oral Radiology, Osaka Dental University School
of Dentistry, Osaka, Japan
•
significantly between the lesions (P \ 0.001). These results
support the hypothesis that OOC originate from not the
odontogenic apparatus, but the oral epithelial component.
Keywords Immunohistochemistry  Keratocystic
odontogenic tumor  Cytokeratin  Orthokeratinized
odontogenic cyst  Dentigerous cyst  Radicular cyst
Introduction
Odontogenic keratocysts were first defined in 1956 by
Philipsen [1] and included in the same category as follicular
dental cysts in the WHO classification published in 1992 [2].
Odontogenic keratocysts are divided into two types, para-
keratosis and orthokeratosis, according to the characteristics
of the epithelial lining. At present, odontogenic keratocysts
with parakeratosis are defined as a keratocystic odontogenic
tumor (KCOT), which is a new term; specifically, they are
classified not as cysts, but as a tumor [3]. Odontogenic
keratocysts with orthokeratosis are also defined as ortho-
keratinized odontogenic cysts (OOC). Only a few reports
have described the clinical and histopathological features of
KCOT and OOC since the adoption of the new WHO
classification. KCOT was reported potentially to exhibit
aggressive behavior and local recurrence [4]. Furthermore,
multiple KCOT are known to arise in basal cell nevus
syndrome. KCOT are thought to exhibit the characteristics
of a tumor because of the possible variation of tumor-related
genes [5, 6]. Several reports have presented data on the
histopathological analysis and expression of cytokeratin in
odontogenic keratocysts diagnosed according to the former
classification [7–13]. However, only a few reports have
presented data from histopathological analyses on KCOT
and OOC and the expression of cytokeratin in the epithelial
123
 Med Mol Morphol

lining of these lesions [10]. In the present study, we inves- Table 1 Clones and dilutions of the primary antibodies
tigated the expression of CK using immunohistochemical Source Clonality Clone Dilution
analyses in KCOT, OOC, dentigerous cysts (DC), radicular
cysts (RC) and dermoid cysts (DMC) to examine the dif- CK10 DAKO, Monoclonal DE-K10 1:100
Denmark (mouse)
ferences in the epithelial lining of KCOT and OOC. We then
compared our results with those of previous reports. CK13 NOVO, UK Monoclonal KS-1A3 1:100
(mouse)
CK17 NOVO, UK Monoclonal E3 1:40
(mouse)
Materials and methods CK19 NOVO, UK Monoclonal b170 1:100
(mouse)
Materials

We used 25 samples of KCOT, 15 samples of each of OOC,

DC and RC, and 10 samples of DMC obtained from Osaka
Dental University Hospital. KCOT and OOC were reclassified
on the basis of the 2005 WHO histopathological classification.
The specimens were fixed in 10 % formalin solution, dehy-
drated in a graded ethanol series and embedded in paraffin.
This research was approved by the Ethics Committee of Osaka
Dental University (approval number 080339).
Statistical analysis
Statistical analysis was accomplished using JMP07 (SAS
Institute, USA). The v2 test was used to compare the
staining of each layer. A P value of \0.05 was considered
to indicate statistical significance.

Immunohistochemistry Results

Four-lm-thick sections were deparaffinized in Hemo-DeÒ CK10 expression

(Falma, Tokyo, Japan) and rehydrated through a graded
ethanol series. Antigen retrieval was carried out by auto-
claving at 121 °C for 15 min in 0.01 M citrate buffer at pH
6.0. Endogenous peroxidase was blocked with 3 %
hydrogen peroxidase. Anti-human cytokeratin 10, 13, 17
and 19 mouse monoclonal antibodies (CK10, 13, 17 and
19) were diluted with Antibody Diluent solution (Dako-
Cytomation, Glostrup, Denmark), and each of the sections
was reacted for 60 min at room temperature. The sections
were then incubated with peroxidase dextran polymer
(Envision?; DakoCytomation, Carpinteria, CA, USA) for
30 min at room temperature. The sections were visualized
using 3,30 -diaminobenzidine-tetrahydrochloride (DAB,
DakoCytomation) and counterstained with hematoxylin.
The used antigens in terms of original sources, clones,
dilutions and manufacturers are summarized in Table 1.
Evaluation of immunohistochemistry
To evaluate the immunohistochemical staining, we divided
the lining epithelium of these lesions into three layers
(surface layer: su, spinous layer: sp, basal layer: ba).
Immunohistochemical expression was evaluated according
to the classification of Hayakawa et al. [11] and Koizumi
[10]. At least 1000 cells were counted in 5 high-power
fields (9400). In addition, we evaluated the staining
according to three levels: (-): less than 10 % positive cells
at each layer, (?): 10–50 % positive cells at each layer and
(2?): more than 50 % positive cells at each layer.
In each lesion, the basal layer (ba) was always (-). For the
KCOT specimens, 22 of the surface layers (su) were (-)
and 3 were (?), and all of the spinous layers (sp) were (-).
For the OOC specimens, 5 of the su were (?) and 10 were
(2?), and 7 of the sp were (?) and 8 were (2?). For the DC
and RC specimens, all of the sp and su were (-). For the
DMC specimens, 2 of the su were (?) and 8 were (2?),
and 7 of the sp were (?) and 3 were (2?).
CK13 expression
In each lesion, the ba was always (-). For the KCOT
specimens, 10 of the su were (?) and 5 were (2?), and 2 of
the sp were (-), 17 were (?) and 6 were (2?). For the
OOC specimens, 12 of the su were (-) and 3 were (?), and
all of the sp were (-). For the DC specimens, 8 of the su
were (?) and 7 were (2?), and 1 of the sp was (-) and 14
were (?). For the RC specimens, 13 of the su were (?) and
2 were (2?), and 5 of the sp were (-) and 10 were (?). For
the DMC specimens, all of the su and sp were (-).
CK17 expression
For the KCOT specimens, 2 of the su were (-), 21 were (?)
and 2 were (2?), and 22 of the sp were (?) and 3 were (2?);
all of the ba were (-). For the OOC, DC, RC and DMC
specimens, all of the layers were (-).

123
Med Mol Morphol
CK19 expression
For the KCOT specimens, 18 of the su were (?) and 7 were
(2?), and 1 of the sp was (-) and 24 were (?); 13 of the ba
were (-) and 12 were (?). For the OOC specimens, all of
the su and sp were (-), and 9 of the ba were (-) and 6 were
(?). For the DC specimens, 3 of the su were (?) and 12
were (2?), 8 of the sp were (?) and 7 were (2?), and 1 of
the ba was (-) and 14 were (?). For the RC specimens, 3
of the su were (?) and 12 were (2?), 5 of the sp were (?)
and 10 were (2?), and 4 of the ba were (-) and 11 were
(?). For the DMC specimens, all of the layers were (-).
The percentages of total CK expression of su and sp, and
ba of CK19 differed significantly between lesions
(P \ 0.001) (Figs. 1, 2).
Discussion
In this study, we considered the expression of CK10, 13, 17
and 19, which are related to the keratinization of squamous
cell epithelium, odontogenic epithelium and tumors arising
in the oral cavity.
CK10 (molecular weight, 56.5 kDa) is an acidic type of
cytokeratin that is specifically expressed on the orthoker-
atinized surface of the squamous epithelium in the cervicis
uteri and tongue [14]. Some reports have described the
expression of cytokeratin in each layer of the epithelial
lining [10, 13]. Koizumi [13] reported that CK10 was
expressed only in the su and sp of OOC and only sporad-
ically in the su of KCOT. Our results are consistent with
this finding, and in DMC, CK10 was expressed the same as
in OOC. The epithelial linings of OOC and DMC have
been suggested to be orthokeratinized. Stoll et al. [11]
studied the entire epithelial lining and reported relatively
low expression of CK10 in DC and RC. Thus, CK10
appears to be significantly expressed in OOC and DMC.
CK13 (molecular weight, 51 kDa) is an acidic type of
cytokeratin that is specifically expressed on the parakera-
tinized surface of squamous epithelium [14]. Previous
reports have indicated that CK13 is expressed in the epi-
thelial lining of odontogenic cysts [8, 10–12]. The CK13
expression rate in RC, DC and odontogenic keratocysts is
more than 80 % and CK13 was expressed in all of the RC
that were examined [11]. In our results, a high positive rate
for the expression of CK13 in RC and DC was also noted.
Furthermore, another report published after the adoption of
the 2005 WHO classification stated that CK13 was strongly
expressed in the su of KCOT and was almost negative in
the su and sp of OOC [10, 13]. Maera et al. [8] and Wato
et al. [15] studied the expression of CK13 in odontogenic
tumors such as ameloblastoma and suggested that it was
expressed in odontogenic cysts and tumors. In the present
study, CK13 was expressed in almost all of the KCOT
specimens (odontogenic tumors) as well as RC and DC
(odontogenic cysts), but was not expressed in OOC and
DMC, suggesting that OOC is not an odontogenic cyst or
tumor but an epithelial cyst, given the finding of the same
expression as in DMC, which originated from the epithelial
component.
CK17 (molecular weight, 46 kDa) has a relatively low
molecular weight and is an acidic type of cytokeratin. In
normal tissues, CK17 is reportedly expressed in basal
cells of the trachea, larynx and bronchi and in myoepi-
thelial cells of the salivary glands and sweat glands [14].
CK17 is also expressed in cancer of the uterine cervix
[16, 17] and in odontogenic tumors [15]. Reportedly,
CK17 is not expressed in normal oral mucosa, but is
expressed in oral squamous dysplasia and squamous cell
carcinoma with high proliferative activities [18]. There-
fore, CK17 is a remarkable CK for investigating the
neoplastic epithelium. In this study, CK17 was only
expressed in the su and sp of KCOT, suggesting that the
epithelial lining of KCOT has neoplastic characteristics
and other components do not. CK17 is expressed in the
basal cell layer of the larynx and bronchi [14], but few
reports have mentioned CK17 expression in KCOT,
OOC, RC, DC and DMC. We did not observe a sig-
nificant difference in CK17 expression in the ba among
the lesion types.
CK19 (molecular weight, 40 kDa) is an acidic type of
CK and is reportedly expressed in most secretory epithelia
and the basal cells of squamous epithelium, without or-
thokeratinized squamous cells [14]. A few reports have
described the expression of CK19, but the results varied
from almost positive to negative in odontogenic kera-
tocysts [11, 13]. Hayakawa et al. [10] reported that CK19
was expressed almost entirely positively in the su and sp of
KCOT and almost entirely negatively in the su and sp of
OOC. It is said that CK19 is expressed in cervicis uteri
cancers and odontogenic epithelium [18, 19]. The expres-
sion of CK19 in the su and sp of KCOT, RC and DC
suggests that these lesions originate from odontogenic
epithelium as odontogenic cysts or odontogenic tumors.
Meanwhile, OOC is unlikely to originate from odontogenic
epithelium because of the absence of CK19 expression, the
same as in DMC. CK19 is positive in the basal layer of
squamous cell epithelium. However, CK19 is not always
positive, depending on the site and lesion [14]. In our data,
however, there was a significant difference in ba, although
there was no regular tendency regarding positivity between
each lesion.
This is the first report on a study of CK expression in
odontogenic tumor, odontogenic cyst, epithelial cyst and
OOC. There were significant differences in total CK
expression among the different lesion types, and OOC and
123
 Med Mol Morphol

CK10
25 2
CK10 P value
20
15 2 su 73.426 <.001
10
5 sp 84.242 <.001

0
su sp ba su sp ba su sp ba su sp ba su sp ba ba NA NA

KCOT OOC DC RC DMC

CK13
25
2
20 CK13 P value

15
2 su 73.983 <.001
10
5 sp 65.456 <.001
0
su sp ba su sp ba su sp ba su sp ba su sp ba ba NA NA
KCOT OOC DC RC DMC

CK17
25
20 CK17 2
P value
15
2
10 su 71.018 <.001

5
sp 80 <.001
0
su sp ba su sp ba su sp ba su sp ba su sp ba
ba NA NA
KCOT OOC DC RC DMC

CK19
25
2
CK19 P value
20
15 2 su 101.807 <.001
10
5 sp 106.436 <.001
0
su sp ba su sp ba su sp ba su sp ba su sp ba ba 24.863 <.001
KCOT OOC DC RC DMC

Fig. 1 Expression of cytokeratins

DMC exhibited almost the same CK expression profiles. histopathological findings of dermoid cyst resembled those
This suggests that OOC may be a lesion that originates of OOC. As such, this study should contribute to the
from the oral epithelial component. In addition, classification of OOC and decisions on its treatment.

123
Med Mol Morphol

Fig. 2 Immunohistochemical
staining for CK10, 13, 17 and
19 in all lesions (original mag.
9200). 1 CK10 positivity in su
and sp of OOC. 2 CK13
positivity in su and sp of KCOT.
3 CK13 positivity in su and sp
of DC. 4 CK13 positivity in su
and sp of RC. 5 CK17 positivity
in su and sp of KCOT. 6 CK19
positivity in all layers of KCOT.
7 CK19 positivity in all layers
of DC. 8 CK19 positivity in all
layers of RC

123

References
1. Philipsen HP (1956) Om Keratocyst (Kolesteatomer) i kæberne.
Tandlaegebladet 60:963–980
2. Kramer IRH, Pindborg JJ, Shear M (1992) Histological typing of
odontogenic tumors, 2nd edn. Springer, Berlin, pp 12–34
3. Barnes L, Everson JW, Reichart P, Sidransky D (2005) Pathology
and genetics of tumors of the head and neck. International
Agency for Research on Cancer, Ryon, pp 284–327
4. Shear M (2002) The aggressive nature of the odontogenic kera-
tocyst: is it a benign cystic neoplasm? Part 1: Clinical and early
experimental evidence of aggressive behavior. Oral Oncol
38:219–226
5. Barreto DC, Gomes RS, Bale AE, Boson WL, Marco LD (2000)
PTCH gene mutations in odontogenic keratocysts. J Dent Res
79:1418–1422
6. Ohki K, Kumamoto H, Ichinohasama R, Sato T, Takahashi N,
Ooya K (2004) PTC gene mutations and expression of SHH,
PTC, SMO, and GLI-1 in odontogenic keratocysts. Int J Oral
Maxillofac Surg 33:584–592
7. MacDonald AW, Fletcher A (1989) Expression of cytokeratin in
the epithelium of dentigerous cysts and odontogenic keratocysts:
an aid to diagnosis. J Clin Pathol 42:736
8. Meara JG, Pilch BZ, Shah SS, Cunningham MJ (2000) Cyto-
keratin expression in the odontogenic keratocyst. J Oral Max-
illofac Surg 58:862–865
9. Kolar Z, Geierova M, Bouchal J, Pazdera J, Zboril V, Tvrdy P
(2006) Immunohistochemical analysis of the biological potential
of odontogenic keratocysts. J Oral Pathol Med 35:75–80
10. Hayakawa M, Okada H (2006) Cytokeratin expression and pro-
liferative activity of keratocystic odontogenic tumor. Int J Oral
Med Sci 5:43–49
123
Med Mol Morphol
11. Stoll C, Stollenwerk C, Riediger D, Mittermayer C, Alfer J
(2005) Cytokeratin expression patterns for distinction of odon-
togenic keratocysts from dentigerous and radicular cysts. J Oral
Pathol Med 34:558–564
12. Silva MJA, Sousa SOM, Correa L, Carvalhosa AA (2002)
Immunohistochemical study of the orthokeratinized odontogenic
cyst: a comparison with the odontogenic keratocyst. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 94:732–737
13. Koizumi Y (2004) Odontogenic keratocyst, orthokeratinized
odontogenic cyst and epidermal cyst: an immunohistochemical
study including markers of proliferation, cytokeratin and apop-
tosis related factors. Int J Oral Med Sci 2:14–22
14. Chu PG, Weiss LM (2002) Keratin expression in human tissue
and neoplasms. Histopathology 40:403–439
15. Wato M, Chen Y, Fang Y, He Z, Wu L, Bamba Y, Hida T,
Hayashi H, Ueda M, Tanaka A (2006) Immunohistochemical
expression of various cytokeratins in ameloblastomas. Oral Med
Pathol 11:67–74
16. Smedts F, Ramaekers F, Troyanovsky S, Pruszczynski M, Link
M, Lane B, Leigh I, Schijf C, Vooijs P (1992) Keratin expression
in cervical cancer. Am J Pathol 141:497–511
17. Carrilho C, Alberto M, Buane L, David L (2004) Keratins 8, 10,
13, and 17 are useful markers in the diagnosis of human cervix
carcinomas. Human Pathol 35:546–551
18. Berkovitz BK, Barret AW (1998) Cytokeratin intermediate fila-
ments in oral and odontogenic epithelia. Bull Group Int Rech Sci
Stomatol Odontol 40:4–23
19. Mittal KR, Demopoulos RI, Goswami S (1992) Patterns of ker-
atin 19 expression in normal, metaplastic, condylomatous,
atrophic, dysplastic, and malignant cervical squamous epithe-
lium. Am J Clin Pathol 98:419–423