Академический Документы
Профессиональный Документы
Культура Документы
Marilena Vered DMD, Sylvie Polak-Charcon PhD, Tania Babushkin PhD & Dan
Dayan DMD, MSc
To cite this article: Marilena Vered DMD, Sylvie Polak-Charcon PhD, Tania Babushkin PhD &
Dan Dayan DMD, MSc (2008) 4NQO-Induced Rat Tongue Carcinoma: An Ultrastructural Study,
Ultrastructural Pathology, 32:5, 199-205, DOI: 10.1080/01913120802034645
Article views: 49
Download by: [Sistema Integrado de Bibliotecas USP] Date: 30 August 2016, At: 10:17
Ultrastructural Pathology, 32:199205, 2008
Copyright r Informa Healthcare USA, Inc.
ISSN: 0191-3123 print /1521-0758 online
DOI: 10.1080/01913120802034645
ABSTRACT
The 4-nitroquinoline 1-oxide (4NQO)-induced rat tongue carcinoma, in which the carcinogen is
administered systemically in drinking water, is the most comparable animal model to the develop-
ment of human oral carcinoma. This is the first study to report the ultrastructural changes in this
model. The most significant changes were observed in the carcinoma cells at the invasion front and
included unique modifications in the basal lamina, presence of micropinocytotic vesicles (plasma-
lemmal caveolae), and emergence of cytoplasmic microfilaments featuring a parallel arrange-
ment. The microfilaments, in both appearance and organization, were consistent with contractile
microfilaments. These observations may be the morphological reflection of the phenotypic modifi-
cations occurring within the carcinoma cells, approaching smooth muscle differentiation.
Oral cancer consistently ranks as one of the top ten [46]. However, the ultrastructural changes have been far
cancers worldwide [1]. Lately, an alarming increase less studied. In regard to tongue cancer, only two studies
in the incidence of oral cancer, particularly tongue in the English literature were published in which the
cancer, in patients under the age of 40, has been carcinogenic agent was administered by topical applica-
reported [2]. Despite the recently reported drop in the tion [7, 8]. Nevertheless, Otsubo and Kameyama used
overall death rate from cancer, the estimated survival 7,12-dimethylbenzo(a)anthracene (DMBA) and hamsters
rate (B50%) and number of deaths from oral cancer [7], while Jiang et al., employed 4-nitroquinoline 1-oxide
remain virtually unchanged [3]. (4NQO) and rats [8]. The main findings in these studies
Chemically induced oral cancer in animal models is included presence of cytoplasmic projections of the basal
aimed to assist us to further understand the events lead- and intercellular aspects of the cancer cells, considerable
ing to the development of oral cancer in human patients. decrease in the number of desmosomes, occurrence of
The changes at the light-microscope level and histo- atypical cytoplasmic tonofilaments, increased intercellular
chemistry, immunohistochemistry, and modifications in spaces, and changes in the basal lamina ranging from
the genetic profile during experimental oral cancer have absence to discontinuity to normal appearance and to
been vastly investigated and thoroughly documented increased thickness or multilayering.
The 4NQO tongue carcinogenesis model, in which the grades of ethanol, and washed in 0.1 M buffered
carcinogen is systemically administered in drinking cacodylate. The specimens were fixed in 2.5% glutar-
water, is recognized as the most reliable simulator of oral aldehyde in 0.1 M buffered cacodylate, postfixed in
cancer development in human patients [5, 9, 10]. In this 1% osmium tetroxide for 1 h, dehydrated in a series
model, small amounts of carcinogen are ingested over a of increasing ethanol concentrations, and finally
long period of time, during which premalignant lesions embedded in epoxy resinAgar mix (Agar Scientific,
of increasing severity develop and culminate with the Essex, UK) for 2 days at 601C. From the blocks,
appearance of a true malignant tumor of squamous cell semithin sections were prepared and relevant areas
carcinoma (SCC). The pattern of the clinical, histopatho- were selected for ultrathin sections and stained with
logical, and nuclear changes of the rat tongue carcinoma uranyl acetate and lead citrate. Examination was
have been found to be equivalent to those described in performed in a Jeol 1200 EX transmission electron
human patients with a fair degree of consistency. In our microscope (Jeol, Peabody, MA, USA).
previous study on this model, we observed that evolu-
tion of tongue SCC was accompanied by a sudden sig-
nificant increase in the stromal myofibroblasts. The cells
were identified by routine immunohistochemical proce-
RESULTS
dure with an antibody against alpha-smooth muscle Figure 1 shows representative regions of normal
actin (a-SMA), a cytoskeleton protein characteristic of tongue (A) and SCC (B) as seen on the hematoxylin
myofibroblasts [13]. In addition, at the tumorstroma and eosin-stained slides, corresponding to tissue
interface we have observed a-SMA-stained cells that blocks submitted for electron microscope examination.
displayed an epithelioid morphology and were remark-
ably similar to SCC cells. Determining whether these cells
were myofibroblasts with an epithelioid morphology or
SCC cells positive for a-SMA was problematical. The
present study was aimed at examining the ultrastructural
changes in the systemic 4NQO tongue SCC model to
look for support for the presence of a-SMA immuno-
reactivity in the tumor cells. In addition, we compared
findings in relation to the corresponding ultrastructural
studies in human oral carcinoma.
EM Preparation
Areas of squamous cell carcinoma (SCC) and of
normal tongue epithelium were marked on the tissue Figure 1. Representative fields of normal tongue epithelium
blocks, respectively. These were extracted, deparaffi- (A) and SCC at the tumor connective tissue interface (B)
nized in xylene, rehydrated in a series of decreasing (hematoxylin and eosin, 400 original magnification).
200
M. Vered et al.
DISCUSSION
This is the first study in the English literature of the
ultrastructural changes in the 4NQO rat tongue
Figure 6. SCC, front of invasion: irregular-sized, congested carcinogenesis model, where the carcinogen was
peripheral cells form a ragged epithelial-connective tissue administered systemically, in drinking water. The
interface. At the basal aspect of these cells, a band of basal importance of this study lies in the observation that
lamina-like material is present. CT, connective tissue.
this particular model reproduces in the closest way the
stages of evolution of oral cancer in human patients.
together to form an intercellular thicker band, ulti- Cellular features corresponding to the state of maturation
mately creating a network in which the neoplastic cells of the epithelial layers were observed in the normal
202
M. Vered et al.
seems that this external lamina served as a means of pattern or an aggregation into bundles that ran parallel
connecting between the tumor cells. The many to the cell membrane. This organization conformed to
micropinocytotic vesicles found in the cytoplasm in the morphological appearance of the contractile com-
the vicinity of the cell membrane may be evidence ponents observed in cells with a recognized contractile
for the existence of an active transport of products potential, such as myofibroblasts [18]. The microfila-
synthesized within the cells and required for the ments in SCC cells might be consistent with actin
establishment of the external lamina. filaments, thus explaining our previous observation of
epithelioid, actin-stained cells at the tumorconnective
Studies on experimental tongue cancer performed tissue interface [10]. This, together with the presence of
with topically applied carcinogen, either in rats [8] or the micropinocytotic vesicles, sited in a pattern remi-
hamsters [7], reported changes in the basal lamina at niscent of smooth muscle cells [19], further supports
the tumorconnective tissue interface that ranged from the phenotypic changes occurring in the SCC cells,
normal ultrastructural morphology, to discontinuity, which approach a smooth muscle differentiation. In
and to prominent focal thickening/multilayering. Cell- the study by Jiang et al., in which 4NQO was topically
to-cell adhesion in the topically applied carcinogen applied on the rat tongues, atypical tonofilament
model was usually provided by desmosomes located bundles that appeared as linear structures in the
at the lateral cell borders, although they were fewer perinuclear and peripheral cytoplasm of the neoplas-
than in the normal tongue epithelium [8]. It seems that tic cells were reported [8]. In that study, they were
our finding of an external lamina network encircling assumed to indicate either a disturbance of epithe-
the tumor cells mainly at the invasion front is exclusive lial maturation or an association with cytoplasmic
for the development of tongue carcinoma following streaming in locomotive tumor cells. Recognition of
systemic administration of 4NQO. a considerably increased number of contractile micro-
Evidence of focal basal lamina multilayering/reduplica- filaments (e.g., actin) in human SCC cells (including
tion/thickening has been previously reported in experi- those of oral epithelium) as compared to their normal
mental carcinogenesis [8, 11] as well as in human oral counterparts was done by Shenk [16] and Gabbiani
SCC [12, 13]. In general, this can be explained in terms of et al. [20, 21] about 3 decades ago. Furthermore, Harris
either an increased turnover with synthesis of additional investigated spindle cell carcinomas of the skin and
cellular/extracellular products by the cells exposed to breast and found that the morphological changes were
micro-environmental insults, or may reflect the cellular accompanied by ultrastructural evidence of partial or
response to the dysplastic changes [14]. An additional complete mesenchymal metaplasia of the carcinoma
theory for the excessive production of external lamina cells [22]. Chen and Harwick also reported presence
relates to the possibility of locomotion of the SCC cells of cytoplasmic microfilaments in oral SCC, but unfor-
in a centrifugal pattern [12]. Areas with thickening of tunately their findings were not properly illustrated [12].
external lamina may present the trailing edge of the Our identification of contractile microfilaments may be
cancer cells. As these cells move forward, the trailing the morphological reflection of the invasive potential
edge will leave behind the original external lamina and of the neoplastic cells. It can be assumed that at any
form successive layers of new lamina. On the other hand, given time SCC cells could open a new invasive front
thinning and absence of this lamina probably represent as evidenced by the uneven appearance of the tumor
the leading portion of cancer cells, which are in the connective tissue interface. This would depend, for
process of active locomotion. Further support to this is example, on the number of contractile microfilaments
provided by the observation that following local inva- present in the cytoplasm of the SCC cells. As the
sion cells may cease migration and produce a distinct number of microfilaments ranged among the SCC cells
basal lamina, and that its disappearance seems restricted from none to many, it seems that those with a higher
to foci of ongoing tumor invasion [15]. Accordingly, in concentration of contractile microfilaments will be
the present study, where the external lamina network burdened with the task of drawing their neighboring
was of various thickness it could be assumed that SCC cells that lack cytoplasmic microfilaments. The net-
cells were in different phases of migration. work of external lamina surrounding the SCC cells
might serve as an intercellular adherence system to
Active locomotion of migrating SCC cells is probably a
transmit the locomotion force from cell to cell in the
central contributing factor to their invasiveness. This
absence of desmosomes. Changes in the thickness of
can be facilitated mainly by the presence of a con-
the external lamina in certain regions could indicate
tractile microfilament system [16]. Contractile cyto-
the movement of the most potently invasive SCC cells
plasmic microfilaments have been identified in
at a given time point.
experimental carcinogenesis of the dermis [17]. Our
results are the first to show occurrence of this type of The present study was the first to describe the ultra-
microfilaments in the rat systemic 4NQO-induced structural changes in the SCC cells in the systemic
SCC. The microfilaments displayed either a parallel 4NQO rat tongue model. Extensive production of
204
M. Vered et al.
basal lamina-like material and presence of contractile 9. Dayan D, Hirshberg A, Kaplan I, Rotem N, Bodner L. Experimen-
cytoplasmic microfilaments parallel the changes tal tongue cancer in desalivated rats. Oral Oncol. 1997;33:105109.
described in oral SCC in human patients. This further 10. Vered M, Allon I, Buchner A, Dayan D. Stromal myofibroblasts
and malignant transformation in a 4NQO rat tongue carcinogen-
emphasizes the potential of this animal model to esis model. Oral Oncol. 2007;43:9991006.
contribute to our understanding the mechanisms of 11. Ferguson JW, McMillan MD, Smillie AC. Ultrastructural
carcinogenesis in humans. Classification of the cyto- examination of experimentally induced premalignant lesions. Int
plasmic microfilaments and their functions in this J Oral Maxillofac Surg. 1991;20:112118.
model should be performed to further explain the 12. Chen SY, Harwick RD. Ultrastructure of oral squamous-cell
carcinoma. Oral Surg. 1977;44:744753.
evolution of the malignant phenotype. 13. Frithiof L. Electron microscopic observations on structures
related to the epithelial basement membrane in squamous cell
carcinoma. Acta Laryngol. 1972;73:323334.
14. Pierce GB, Nakane PK. Basement membranes: synthesis and
REFERENCES deposition in response to cellular injury. Lab Invest. 1969;21:2741.
15. McArdle JP, Roff BT, Muller HK, Murphy HK. The basal
1. Rodrigues VC, Moss SM, Tuomainen H. Oral cancer in the lamina in basal cell carcinoma, Bowens disease, squamous cell
UKto screen or not to screen. Oral Oncol. 1998;34:454465. carcinoma and keratoacanthoma: an immunoperoxidase study
2. Shiboski CH, Shiboski SC, Silverman S Jr. Trends in oral cancer using an antibody to type IV collagen. Pathology. 1984;16:6772.
rates in the United States, 19731996. Community Dent Oral 16. Shenk P. Microfilaments in human epithelial cancer cells.
Epidemiol. 2000;28:249256. Z Krebsforsch. 1974;84:241256.
3. Lodi G, Sardella A, Bez C, Demarosi F, Carrassi A. Interven- 17. Malech HL, Lentz TL. Microfilaments in epidermal cancer cells.
tions for treating leukoplakia. Cochrane Database Syst Rev. J Cell Biol. 1974;60:473482.
2004;3:CD001829. 18. Dayan D, Bodner L. Nodular fasciitis of the oral mucosa: light
4. Vered M, Yarom N, Dayan D. 4NQO oral carcinogenesis: and electron microscopy study. Head Neck. 1991;13:434438.
animal models, molecular markers and future expectations. Oral 19. Ghadially FN. Ultrastructural Pathology of the Cell and Matrix:
Oncol. 2005;41:337339. A Text and Atlas of Physiological and Pathological Alterations in the
5. Mognetti B, Di Carlo F, Berta GN. Animal models in oral Fine Structure of Cellular and Extracellular Components. London:
cancer research. Oral Oncol. 2006;42:448460. Butterworths; 1988.
6. Kanojia D, Vaidya MM. 4-Nitroquinoline-1-oxide induced experi- 20. Gabbiani G, Trenchev P, Holborow EJ. Increase of contractile
mental oral carcinogenensis. Oral Oncol. 2006;42:655667. proteins in human cancer cells. Lancet. 1975;25:796797.
7. Otsubo Y, Kameyama Y. Ultrastructural changes of epithelium- 21. Gabbiani G, Csank-Brassert J, Schneeberger J-C, Kapanci Y,
connective tissue junction in experimental lingual tumors. J Oral Trenchev P, Holborow EJ. Contractile proteins in human cancer
Pathol. 1982;11:159173. cells. Immunofluorocent and electron microscopic study. Am J
8. Jiang DJ, Wilson DF, Wiebkin OW. Ultrastructural features of Pathol. 1976;83:457474.
normal epithelium and 4-nitroquinoline 1-oxide-induced carci- 22. Harris M. Spindle cell squamous carcinoma: ultrastructural
nomas of the rat tongue. J Comp Pathol. 1993;108:375381. observations. Histopathology. 1982;6:197210.
205