- Email: [email protected]
Structure of hamster palatal gingiva and intermolar mucosa after intraperitoneal dosing with N-methyl-N-nitrosourea: a scanning electron microscope study
J. Comp. Path. 1995 Vol. 112, 403 415
Structure o f H a m s t e r P a l a t a l G i n g i v a a n d I n t e r m o l a r M u c o s a after I n t r a p e r i t o n e a l D o s i n g with N-Methyl-N-Nitrosourea: a Scanning Electron Microscope Study G. C. Symes, M. D. McMillan, A. C. Smillie and N. M. Boyd* Department of Oral Biology and Oral Pathology, School of DentisOy, Universityof Otago, Dunedin, New Zealand, and *Faculty of Dentistry, Universityof WesternAustralia, Perth, Australia
Summary The palatal gingiva and intermolar mucosa from normal hamsters and from hamsters that had received N-methyl-N-nitrosourea (NMU) intraperitoneally were examined by scanning electron microscopy over a 22-week period. The normal gingiva and rostral two-thirds of the intermolar mucosa were covered by flat polygonal cells that had a regular honeycomb surface pattern of interconnecting micro-ridges, distinct cell boundaries and imprints of cells that had been desquamated. The caudal third of the intermolar mucosa in normal and NMU-treated hamsters was covered by soft-palate type mucosa whose smooth surfaced cells surrounded scattered fungiform-like papillae. In NMU-treated hamsters changes were more common in the rostral two-thirds of the intermolar mucosa than in the gingiva. At 10 weeks there were sessile and conical surface projections and saucer-shaped and conical epitheliallined depressions. At 16 weeks these projections and depressions were larger and more numerous, and groups of conical projections formed papillomatouslike lesions. At 22 weeks the projections and depressions were further increased in number and size and there were distinct papillomas. At 10 and 16 weeks the entire epithelium showed cells, cell boundaries and cell imprints resembling those in the controls, except that there were defects or dilated intercellular spaces at the base of the conical depressions and some of the cells were thicker. Much of the 22-week epithelia had a similar structure, but in some areas that did not show the projections or depressions the cells varied in size and shape and were covered by elongated micro-ridges. Here the surface was irregular as were the cell boundaries. Examination of sections by light microscopy, both from these irregular areas and from areas that appeared normal by scanning electron microscopy, revealed that the deeper epithelial strata could be either normal, dysplastic or at times malignant. However, scanning electron microscopy failed to reveal dysplastic or malignant epithelium.
Introduction D i m e t h y l b e n z - a n t h r a c e n e (DMBA) application to the h a m s t e r c h e e k p o u c h is the most c o m m o n l y used m o d e l for the study o f oral c a n c e r a n d p r e c a n c e r (Eveson, 1981; B o y d a n d R e a d e , 1988), but the topical application o f 4 - n i t r o q u i n o l i n e - l - o x i d e ( 4 N Q O ) to the h a r d palates o f mice (Steidler and 0021-9975/95/040403 + 13 $08.00/0
9 1995 Academic Press Limited
404
G. CI. Syrnes et al.
Reade, 1984) and rats (Svensson and Heyden, 1982) and the tongues of rats (Wilson et al., 1993) is also used. All of these models, however, are based on the topical application of the carcinogen and some of the changes, especially the earlier ones, may be due, in part at least, to the local irritant effects of either the carcinogen, or in the case of DMBA the vehicle, or both. The use of a systemic carcinogen would eliminate these possibilities. N-methyl-N-nitrosourea (NMU) has been used to induce experimental oral carcinoma in rats and hamsters by the intravenous (Herrold, 1968), subcutaneous (Herrold, 1966), intragastric (Herrold, 1968; Suzuki, 1976; Kohgo et al., 1990) or intraperitoneal (Edwards and Germain, 1980) route. However, the results of these and related studies are equivocal; the best route of administration is intraperitoneal, but only recently was a dose rate established that produced tumours in the palatal gingiva and intermolar mucosa of hamsters yet allowed most animals to survive (McMillan et al., 1994). All of the studies with N M U and m a n y of those with DMBA and 4 N Q O have been based on light microscopy, but transmission electron microscopy has been used in some with the latter two carcinogens. The use of scanning electron microscopy in such studies is limited to the work of Hassanin et al. /1987) and Hassanin and Ashrafi (1988) on the early changes in the hamster cheek pouch after DMBA application. A limited number of studies on premalignant and malignant epithelium from the h u m a n oral cavity have been based on scanning electron microscopy (Morgenroth and Morgenroth, 1970; Matravers and Tyldesley, 1977; Banoczy et al., 1980; Reichart et al., 1981; Nakao, 1983; Reichart and Althoff, 1983;Jungell et al., 1987). The aim of the present study was to examine the structure of the palatal gingival and intermolar mucosa in hamsters by scanning electron microscopy after intraperitoneal dosing with NMU. Materials and Methods
Thirty-five 4 6-week-old male BIO 87-20 hamsters were used. Twenty-three were given intraperitoneal injections (12"5mg/kg body weight) of a freshly prepared 0'06 mol/1 solution of NMU in 0-1 M citrate buffer (pH 6"2) three times a week for 4 weeks. The remaining 12 animals (controls) were given a similar amount of the citrate buffer. All animals were carefully observed and regularly weighed throughout the entire experimental period. Any that showed undue weight loss or appeared unwell were killed by an overdose of anaesthetic followed by cervical dislocation. At 10, 16 and 22 weeks after the first injection, five NMU-treated animals and three controls were anaesthetized and killed and their maxillae removed and trimmed. The mucosal surfaces were washed with a gentle stream of freshly prepared phosphatebuffered saline. The maxillae were then fixed in glutaraldehyde 2% in 0-1 M cacodylate buffer (pH 7"4) at 4~ for at least 24 h, washed in distilled water, dehydrated in graded concentrations of acetone, and critical-point dried in liquid carbon dioxide. They were then glued to large aluminium stubs, sputter-coated with gold and examined in a Cambridge 360 scanning electron microscope. After this the maxillae were removed from the stubs, rehydrated in graded concentrations of ethanol, and demineralized in 10% formic acid. The specimens were then processed routinely for light microscopy and cut in either the sagittal or coronal plane. Sections were stained with haematoxylin and eosin (HE). Specimens from control animals were selected so that representative samples from all areas of the palatal gingiva and intermolar mucosa could be examined. Specimens from NMU-treated animals were selected so that a
Induced
Fig. 1.
Changes
in the Hamster
Palate
405
Control. Intermolar mucosa with rugae (R) and sessile and conical projections (arrows). Palatal gingiva (G) which merges with the intermolar mucosa and rugae; soft palate-like mucosa (S) with fungiform-like papilla (P); molar teeth (M). SEM. x 25.
representative sample of all the features not seen in control animals by scanning electron microscopy could be viewed by light microscopy. Results
Control Animals The appearance of the palatal gingiva and intermolar mucosa was similar in all animals and in the rostral two-thirds there was no sharp boundary between them (Fig. 1). In the rostral two-thirds, rugae extended from the gingiva medially although they never met in the midline. The rostral rugae were more prominent than the caudal rugae. There were clusters of small sessile and conical projections on the more rostral-lateral aspects of the intermolar mucosa and adjacent gingiva. The morphology of the cells covering the palatal gingiva and rostral two-thirds of the intermolar mucosa was similar. Regular flattened polygonal cells had a distinct honeycomb surface pattern of interconnecting micro-ridges that surrounded depressions. There were distinct cell boundaries and imprints of overlying cells that had been desquamated. The former consisted of two parallel linear ridges that were often separated by a narrow cleft. The latter were seen as a single linear ridge (Fig. 2). The junction between the rostral two-thirds and caudal third of the intermolar mucosa was sharp. The mucosa covering the caudal third was similar to that covering the soft palate and had fungiform-like papillae scattered over it (Fig. 1). Although the epithelium covering these latter structures was similar to that on the palatal gingiva and rostral two-thirds of the intermolar mucosa,
406
Fig. 2.
G . C . S y m e s e t al.
Control. Flattened polygonal cells with honeycomb surface pattern. Cell boundaries formed by two linear ridges (A); imprints of desquamated cells formed by a single linear ridge (B). SEM. x 3000.
that between the fungiform-like papillae was not. Here the epithelium was more uneven. The surface of the cells was smooth and lacked a honeycomb surface pattern. Distinct cell boundaries and cell imprints were only found occasionally. The fungiform-like papillae had one or two taste pore openings in their centre. By light microscopy the structure of all areas was as previously described (McMillan et al., 1994).
N M U - Treated Animals The majority underwent an initial small weight loss then showed a steady weight gain. However, two animals had to be killed in the first 10 weeks because of undue weight loss and five had to be killed near the end of the 22week period because of lethargy, loss of coat sheen and weight loss. At 10 weeks the surface of the palatal gingiva and rostral two-thirds of the intermolar mucosa was more irregular than in the control animals (Fig. 3). This irregularity was due to the presence of sessile surface projections, groups of small conical projections and saucer-shaped and conical depressions. The groups of conical projections were most common laterally between the more rostral rugae. The saucer-shaped depressions were relatively shallow (Fig. 4), had smooth rounded edges and were, in many instances, lined by a continuous layer of flattened epithelial cells. Some had dilated intercellular spaces at their base. The conical depressions extended for a greater distance into the epithelium than the saucer-shaped ones and usually had one or more defects or dilated intercellular spaces at their base (Fig. 5). T h e y were lined by a
I n d u c e d Changes in the H a m s t e r Palate
407
Fig. 3.
Ten weeks after N M U treatment. Irregular intermolar mucosa with sessile projections (arrows), conical projections (arrowheads) and saucer-shaped (S) and conical (C) depressions. Rugae (R); Palatal gingiva (G). SEM. x 40.
Fig. 4.
Ten weeks after N M U treatment. Epithelial lined saucer-shaped depression (S) with desquamating cells (arrowheads). SEM. x 290.
whorled arrangement of epithelial cells, a number of which were desquamating. The diameter of the oral opening of these depressions varied from 250 gm to less than 20 gm. Variable numbers of bacteria were found within them.
408
G . C . S y m e s et al.
Fig. 5.
Ten weeks after N M U treatment. Conical depression with desquamating cells (arrowheads) and basal defects and dilated intercellular spaces (arrows). SEM. • 250.
Fig. 6.
Sixteen weeks after N M U treatment. Irregular intermolar mucosa with conical depression (arrowhead) and conical projections (arrow's). Gingival lesions (L); palatal gingiva (G); unaltered soft palate-like mucosa (S). SEM. x 25.
Epithelial cells in all areas had a honeycomb surface pattern and cell boundaries and cell imprints were similar to those in the control animals. The changes at 16 weeks were similar to those at 10 weeks but more pronounced (Fig. 6). The conical depressions occurred more frequently, were
I n d u c e d C h a n g e s in the H a m s t e r P a l a t e
]Fig. 7.
409
Sixteen weeks after NMU treatment. Conical depression with basal defects and dilated intercellular spaces (D). SEM. x 650.
often of greater diameter, frequently extended for a greater distance into the epithelium and had larger defects or dilated intercellular spaces in their base (Fig. 7). The lateral margins of the cells lining these depressions were often thickened and separated by dilated intercellular spaces. However, they still had a honeycomb surface pattern together with distinct cell boundaries and cell imprints (Fig. 8). The conical projections in the 16-week specimens (Fig. 6) were larger than those seen at 10 weeks. They were often grouped together and formed papillomatous-like lesions. Towards the apex of these projections the most superficial cells were often desquamating, revealing an irregular folded surface. The surfaces of the deeper cells in these areas were crossed by elongated ridges rather than having a honeycomb surface pattern. Cell boundaries were also indistinct. All of the changes seen at i0 and 16 weeks were present at 22 weeks. However, at 22 weeks they were more extensive, involved more of the mucosa and often masked the normal rugal pattern. Some of the conical depressions were up to 500 gm in diameter at their oral opening. Distinct papillomas were a common finding (Fig. 9). The papillomas were demarcated from the surrounding mucosa by a rim-like arrangement of peripheral cells. The epithelium of the papillomas had the characteristic honeycomb surface pattern, together with distinct cell boundaries and cell imprints similar to those in control animals. The surface structure of those parts of the mucosa that did not show any of the previously described changes varied. In some areas it was similar to that in control animals, although the number of desquamating cells was increased. Elsewhere the size and shape of the cells varied, the surface
410
a . C . Symes e t
al.
Fig. 8.
Sixteen weeks after N M U treatment. Cells lining the conical depression in Fig. 7 that have a honeycomb surface pattern. Cell imprints (arrows). SEM. x 5500.
Fig. 9.
Twenty-two weeks after N M U treatment. Papilloma with numerous projections (P). SEM. x 190.
was irregular, and there were irregular cell boundaries and elongated microridges on the surface of the cells (Fig. I 0). The changes in the N M U animals all affected the rostral two-thirds of the intermolar mucosa and to a lesser extent the palatal gingiva (Fig. 6). Changes
Induced
Fig. 10.
Changes
in the Hamster
Palate
411
Twenty-two weeks after NMU treatment. Area ofmucosa with cells that have elongated microridges. SEM. x 5000.
were not seen on the soft palate-type intermolar mucosa covering the caudal third of the hard palate. Light microscopy showed changes similar to those described previously (McMillan et al., 1994). The saucer-shaped and conical depressions were associated with either keratin plugging, focal areas of individual cell keratinization, or both. Occasionally the conical depressions were associated with epithelial lined duct-like structures that extended into the underlying connective tissue. In the 22-week specimens there were areas of dysplastic and at times malignant change in the deeper epithelial strata, irrespective of the surface structure of the epithelium as seen by scanning electron microscopy. Discussion
The mucosa covering the caudal third of the intermolar area was similar to that of the soft palate. A similar arrangement is found in some strains of rats (Wong and Wilson, 1983) but not in others (McMillan, 1974). The intermolar mucosa, palatal gingiva and fungiform-like papillae in control hamsters were characterized by flattened polygonal cells that had a honeycomb surface pattern, were separated by distinct cell boundaries and were crossed by the imprints of the more superficial cells that had been desquamated. This appearance resembled that found in similar areas of rat oral mucosa (McMillan, 1974, 1979) and hamster cheek pouch (McMillan et al., 1982). The structure of the rest of the soft palate-type mucosa, which was folded and lacked a honeycomb surface pattern and obvious cell boundaries and imprints, was similar to that in the rat (McMillan, 1979).
412
G.C. Syrnes
e t al.
Although a honeycomb surface pattern is characteristic of keratinized oral epithelium in a number of species (Cleaton-]ones and Fleisch, 1973; KullaaMikkonen, 1986), a similar pattern is present on human buccal mucosa (Matravers and Tyldesley, 1977) and on monkey cheek and lip mucosa (Nair and Schroeder, 1981), which are not keratinized. In the rat, oral epithelium with a honeycomb surface pattern is completely orthokeratinized, but oral epithelium without such a pattern is incompletely orthokeratinized (McMillan, 1979). All of the oral mucosa in the hamster is also keratinized, but whether there is a similar association between the surface pattern and degree of keratinization is not known. The proportion of NMU-treated animals that had to be killed (7/23) was similar to that reported in a previous study (McMillan el al., 1994), as were the changes seen by light microscopy in palatal gingiva and intermolar mucosa. However, in the present study there was a longer latent period and the number of exophytic lesions was reduced, as were the number and depth of invasive lesions. These differences probably resulted from the use of different strains of hamsters in the two studies. The principal surface changes that occurred after the administration of N M U were the appearance of saucer-shaped and conical depressions and sessile, conical and papillomatous projections. These gradually increased in size and number, distinct papillomas being found in the 22-week specimens. Some of the conical depressions were the oral openings of the keratinfilled duct-like structures, but others were associated with extensive areas of individual cell keratinization or keratin plugging, which are characteristic of this experimental model (McMillan et al., 1994). The saucer-shaped depressions were associated with less extensive areas of keratin plugging or individual keratinization, or both. The areas of individual cell keratinization often resembled the focal acantholytic dyskeratosis of Darrier's disease (Philipsen and Fisker, 1983) which is a benign, usually inherited, slowly progressive kerototic skin disorder that may also involve the oral and genital mucosa. These features are not characteristic of human oral dysplasia and squamous cell carcinoma. As the structural changes associated with the saucer-shaped and conical depressions are rarely if ever seen in DMBA-treated hamster cheek pouches, it is not surprising that neither type was seen in the scanning electron microscope studies of Hassanin et al. (1987) and Hassanin and Ashrafi (1988). That exophytic lesions were only infrequently found on the palatal gingiva confirms the findings ofMcMillan et al. (1994). Suzuki (1976) reportd exophytic lesions of the gingiva in hamsters after N M U treatment, but Kohgo et al. (1990) found only invasive ones. Exophytic lesions of the palatal intermolar mucosa in hamsters after N M U treatment was also reported by Edwards and Germain (1980). The presence of exophytic lesions, including papillomas, is typical of DMBA-initiated carcinogenesis in mouse skin (Di Giovanni, 1992) and hamster cheek pouch (Boyd and Reade, 1988). Such changes are uncommon in carcinomas of the human skin and oral cavity. The sessile and conical projections in the more rostral lateral aspects of the intermolar mucosa were probably derived from similar small structures, as seen in the controls.
Induced Changes in the H a m s t e r Palate
413
However, those either more medially or caudally situated must have arisen de novo. By 16 weeks many of these projections were much larger than the ones in the controls and in some instances a number of the conical ones were grouped together to form papillomatous-like lesions which presumably gave rise to the distinct papillomas seen at 22 weeks. The soft palate-like mucosa that covered the caudal third of the intermolar mucosa was unaffected in all of the NMU animals, indicating the varying susceptibility of different oral sites to chemical carcinogens. Such. variability is found in animals treated with either N M U (Edwards and Germain, 1980; Kohgo et al., 1990), DMBA (Eveson, 1981) or 4 N Q O (Steidler and Reade, 1984; Boyd and Reade, 1988). Examination of DMBA-treated hamster cheek pouches by scanning electron microscopy showed that by 8 weeks the mucosal surface was totally altered, most of the cells being spindle-shaped and covered by microvilli (Hassanin et al., 1987; Hassanin and Ashrafi, 1988). Variation in cell size, thickening of individual cells, disruption of the normal intercellular relationships, loss of distinct cell boundaries and replacement of the normal surface pattern of micro-ridges by pebbly surface projections or microvilli have been described in many of the scanning electron microscope studies ofdysplastic and malignant epithelial cells from the human oral cavity (Morgenroth and Morgenroth, 1970; Matravers and Tyldesley, 1977; Banoczy et al., 1980; Reichart et al., 1981; Nakao, 1983; Reichart and Althoff, 1983), nasal cavity (Boysen and Reith, 1982) and uterine cervix (Jordan and Williams, 1971; Rubio and Kranz, 1976). Similar changes were seen in experimental carcinogenesis in the rat bladder (Jacobs et al., 1976). In the 10- and 16-week N M U specimens, scanning electron microscopy showed that the surface structure of the epithelium, including that covering the three types of exophytic lesion and the two types of depression, generally resembled the surface structure seen in the controls; slight differences included an increase in the number ofdesquamating cells and, in the conical depressions, dilated intercellular spaces and thickened epithelial cells. At 22 weeks much of the epithelium also had a surface structure similar to that of the controls. However, some of the epithelium between the exophytic lesions and depressions showed variations in cell size and shape and irregular cell boundaries, the characteristic honeycomb surface pattern on the cells being replaced by more elongated micro-ridges. By light microscopy the deeper strata of this more irregular epithelium varied. Sometimes it was normal, sometimes it was dysplastic and sometimes it was malignant. There were also similar variations in the structure of the deeper epithelial strata in areas where the surface structure, as seen by scanning electron microscopy, was similar to that in the controls. No changes indicative of epithelial dysplasia or malignancy were found by scanning electron microscopy, although such changes were in fact present in the deeper strata of the epithelium. However, had the hamsters been allowed to survive longer, or had a strain of hamster in which the carcinomas arose earlier been used, more of the surface changes said to be characteristic of epithelial dysplasia and malignancy might have developed. The relative lack of overt change in the morphologT of most of the surface
414
G.C. Symes e t al.
cells in this study serves to emphasize that in a thick keratinized stratified squamous epithelium, dysplastic and malignant change in the deeper cell layers may, initially at least, be masked by the more normal superficial layers.
Acknowledgments This work was supported by a University of Otago Research Grant. The paper was written while one of the authors (M. D. McMillan) was on leave in the Department of Oral Pathology, University of Sheffield, and Professor C.J. Smith is thanked for making this" possible. Mrs P.K. Cosgriff and Mrs E.A. McMillan are thanked for typing the manuscript.
References Banoczy, J., Lapis, K. and Albrecht, M. (1980). Scanning electron microscopic study of oral leukoplakia. Journal of Oral Pathology, 9, 145-154. Boyd, N. M. and Reade, P. C. (1988). Mechanisms of carcinogenesis with particular reference to the oral mucosa. Journal of Oral Pathology, !7, 193 201. Boysen, M. and Reith, A. (1982). The surface structure of the human nasal mucosa II. Metaplasia, dysplasia and carcinoma in nickel workers. A correlated study by scanning/transmission electron and light microscopy. VzrchowsArchiv B Cell Pathology, 40, 295-309. Cleaton-Jones, P. and Fleisch, L. (1973). A comparative study of the surface of keratinized and non-keratinized oral epithelia. Journal of Periodontal Research, 8, 366-370. Di Giovanni, J. (1992). Multistage carcinogenesis in mouse skin. Pharmacology and Therapeutics, 54, 63-128. Edwards, M. B. and Germain, J. P. (1980). Oral carcinogenesis by intraperitoneal Nmethyl-N-nitrosourea in the Syrian hamster. Archives of Oral Biology, 25, 631 633. Eveson, J. W. (1981). Animal models of intra-oral chemical carcinogenesis: a review. Journal of Oral Pathology, 10, 129-146. Hassanin, M.B. and Ashrafi, S.H. (1988). Comparative light, scanning and transmission electron microscopic study of chemically induced premalignant lesions in the hamster's cheek pouch. Ultrastructural Pathology, 12, 341 350. Hassanin, M.B., Ashrafi, S.H. and Das, A. (1987). Surface ultrastructural changes in chemically induced premalignant lesions in the hamster cheek pouch. Journal of Oral Pathology, 16, 273-277. Herrold, K.M. (1966). Carcinogenic effect of N-methyl-N-nitrosourea administered subcutaneously to Syrian hamsters, journal of Pathology and Bacteriology, 92, 35-41. Herrold, K.M. (1968). Odontogenic tumors and epidermoid carcinomas of the oral cavity. Oral Surgery, Oral Medicine and Oral Pathology, 25, 262-272. Jacobs, J.B., Arai, M., Cohen, S.M. and Friedell, G.H. (1976). Early lesions in experimental bladder cancer: scanning electron microscopy of cell surface markers. CancerResearch, 36, 2512-2517. Jordan, J. A. and Williams, A. E. (1971). Scanning electron microscopy in the study of cervical neoplasia. Journal of Obstetrics and Gynaecology, 78, 940-946. Jungell, P., Malmstrom, M., Wartiovaara, J. and Kouttinen, Y. (1987). Ultrastructure of oral leukoplakia and lichen planus. II. A correlated scanning and transmission electron microscopic study of epithelial surface cells. Journal of Oral Pathology, 16, 287 293. Kohgo, T., Uy, G.H., Iizuka, T., Shindoh, M. and Amemiya, A. (1990). Histopathologic study of squamous cell carcinoma of the gingiva in edentulous and dentulous jaws of hamsters treated with N-methylnitrosourea. Journal of Oral Pathology, 19, 202 207. Kullaa-Mikkonen, A. (1986). Scanning electron microscopic study of surface of human oral mucosa. ScandinavianJournal of Dental Research, 94, 50 56.
Induced Changes in the H a m s t e r Palate
415
Matravers,J. and Tyldesley, W. (1977). Scanning electron microscopy of oral epithelial cells. Part I. Normal and malignant tissues. British Journal of Oral Surgery, 15, 193-202. McMillan, M.D. (1974). A scanning electron microscopic study of keratinized epithelium of the hard palate of the rat. Archives of Oral Biology, 19, 225-229. McMillan, M.D. (1979). The surface structure of the completely and incompletely orthokeratinized oral epithelium in the rat. A light, scanning and transmission electron microscope study. American Journal of Anatomy, 156, 337-351. McMillan, M. D., Smillie, A. C. and Gray, D. W. (1982). The surface structure of the epithelium of the hamster cheek pouch. Archives of Oral Biology, 27, 623 634. McMillan, M. D., Smith, C.J. and Ramirez, V. (1994). The effect of intraperitoneal N-methyl-N-nitrosourea on hamster palatal mucosa. Archives of Oral Biology, 39, 519-528. Morgenroth, K. von and Morgenroth, K. (1970). Rasterelektronenmikroskopische untersuchungen zur morphologie der leukoplakie der mundhohle. Deutsche Zahn aerztliche Zeitschrift, 25, 1054-1060. Nair, P. N. R. and Schroeder, H. E. (1981). Variation and density of microplications in superficial cells of the normal oral lining mucosa in the monkey Macacus fascicularis. Archives of Oral Biology, 26, 837-843. Nakao, I. (1983). Comparative studies on exfoliated cells from the oral mucosa by light and scanning electron microscopy. Tsurumi Shigaku, 9, 151-177. Philipsen, H.P. and Fisker, A.V. (1983). Focal acantholytic dyskeratosis in experimental oral carcinogenesis in rats. Journal of Oral Pathology, 12, 30 36. Reichart, P. A. and Althoff, J. (1983). Oral leukoplakia: a scanning electron microscopic study of epithelial surface patterns. International Journal of Oral Surgery, 12, 159-164. Reichart, P. von, Althoff, J. and Reilmann, B. (1981). Klinische und rasterelektronenmikroskopische untersuchungen oraler papillomatosen. Deutsche Zah naerztliche Zeitschrift, 36, 386-388. Rubio, C.A. and Kranz, I. (1976). The exfoliating cervical epithelial surface in dysplasia, carcinoma in situ and invasive squamous carcinoma I. Scanning electron microscopic study. Acta Cytologica, 20, 144-150. Steidler, N. E. and Reade, P. C. (1984). Experimental induction of oral squamous cell carcinomas in mice with 4-nitroquinoline-1-oxide. Oral Surgery, Oral Medicine and Oral Pathology, 57, 524-531. Suzuki, M. (1976). Histological studies of early changes of the hamster gingival epithelium by N-nitrosomethylurea. Bulletin of Tokyo Medical and Dental University, 23, 115 134. Svensson, S. and Heyden, G. (1982). Experimental induction of irreversible precancerous changes in the palatal epithelium of the rat. International Journal of Oral Surgery, 11, 52 58. Wilson, D. F., Jiang, D.J., Leong, A.S. and Wiebkin, O.W. (1993). Laminin and type IV collagen in experimental rat oral carcinomas. Journal of Comparative Pathology, 108, 375-381. Wong, P.N.C. and Wilson, D.F. (1983). 4-Nitroquinoline-l-oxide-induced carcinogenesis in the rat palate. Journal of Oral Pathology, 12, 375 384.
Received, September 6 th, 1994] Accepted, February 8th, 1995 J
Structure o f H a m s t e r P a l a t a l G i n g i v a a n d I n t e r m o l a r M u c o s a after I n t r a p e r i t o n e a l D o s i n g with N-Methyl-N-Nitrosourea: a Scanning Electron Microscope Study G. C. Symes, M. D. McMillan, A. C. Smillie and N. M. Boyd* Department of Oral Biology and Oral Pathology, School of DentisOy, Universityof Otago, Dunedin, New Zealand, and *Faculty of Dentistry, Universityof WesternAustralia, Perth, Australia
Summary The palatal gingiva and intermolar mucosa from normal hamsters and from hamsters that had received N-methyl-N-nitrosourea (NMU) intraperitoneally were examined by scanning electron microscopy over a 22-week period. The normal gingiva and rostral two-thirds of the intermolar mucosa were covered by flat polygonal cells that had a regular honeycomb surface pattern of interconnecting micro-ridges, distinct cell boundaries and imprints of cells that had been desquamated. The caudal third of the intermolar mucosa in normal and NMU-treated hamsters was covered by soft-palate type mucosa whose smooth surfaced cells surrounded scattered fungiform-like papillae. In NMU-treated hamsters changes were more common in the rostral two-thirds of the intermolar mucosa than in the gingiva. At 10 weeks there were sessile and conical surface projections and saucer-shaped and conical epitheliallined depressions. At 16 weeks these projections and depressions were larger and more numerous, and groups of conical projections formed papillomatouslike lesions. At 22 weeks the projections and depressions were further increased in number and size and there were distinct papillomas. At 10 and 16 weeks the entire epithelium showed cells, cell boundaries and cell imprints resembling those in the controls, except that there were defects or dilated intercellular spaces at the base of the conical depressions and some of the cells were thicker. Much of the 22-week epithelia had a similar structure, but in some areas that did not show the projections or depressions the cells varied in size and shape and were covered by elongated micro-ridges. Here the surface was irregular as were the cell boundaries. Examination of sections by light microscopy, both from these irregular areas and from areas that appeared normal by scanning electron microscopy, revealed that the deeper epithelial strata could be either normal, dysplastic or at times malignant. However, scanning electron microscopy failed to reveal dysplastic or malignant epithelium.
Introduction D i m e t h y l b e n z - a n t h r a c e n e (DMBA) application to the h a m s t e r c h e e k p o u c h is the most c o m m o n l y used m o d e l for the study o f oral c a n c e r a n d p r e c a n c e r (Eveson, 1981; B o y d a n d R e a d e , 1988), but the topical application o f 4 - n i t r o q u i n o l i n e - l - o x i d e ( 4 N Q O ) to the h a r d palates o f mice (Steidler and 0021-9975/95/040403 + 13 $08.00/0
9 1995 Academic Press Limited
404
G. CI. Syrnes et al.
Reade, 1984) and rats (Svensson and Heyden, 1982) and the tongues of rats (Wilson et al., 1993) is also used. All of these models, however, are based on the topical application of the carcinogen and some of the changes, especially the earlier ones, may be due, in part at least, to the local irritant effects of either the carcinogen, or in the case of DMBA the vehicle, or both. The use of a systemic carcinogen would eliminate these possibilities. N-methyl-N-nitrosourea (NMU) has been used to induce experimental oral carcinoma in rats and hamsters by the intravenous (Herrold, 1968), subcutaneous (Herrold, 1966), intragastric (Herrold, 1968; Suzuki, 1976; Kohgo et al., 1990) or intraperitoneal (Edwards and Germain, 1980) route. However, the results of these and related studies are equivocal; the best route of administration is intraperitoneal, but only recently was a dose rate established that produced tumours in the palatal gingiva and intermolar mucosa of hamsters yet allowed most animals to survive (McMillan et al., 1994). All of the studies with N M U and m a n y of those with DMBA and 4 N Q O have been based on light microscopy, but transmission electron microscopy has been used in some with the latter two carcinogens. The use of scanning electron microscopy in such studies is limited to the work of Hassanin et al. /1987) and Hassanin and Ashrafi (1988) on the early changes in the hamster cheek pouch after DMBA application. A limited number of studies on premalignant and malignant epithelium from the h u m a n oral cavity have been based on scanning electron microscopy (Morgenroth and Morgenroth, 1970; Matravers and Tyldesley, 1977; Banoczy et al., 1980; Reichart et al., 1981; Nakao, 1983; Reichart and Althoff, 1983;Jungell et al., 1987). The aim of the present study was to examine the structure of the palatal gingival and intermolar mucosa in hamsters by scanning electron microscopy after intraperitoneal dosing with NMU. Materials and Methods
Thirty-five 4 6-week-old male BIO 87-20 hamsters were used. Twenty-three were given intraperitoneal injections (12"5mg/kg body weight) of a freshly prepared 0'06 mol/1 solution of NMU in 0-1 M citrate buffer (pH 6"2) three times a week for 4 weeks. The remaining 12 animals (controls) were given a similar amount of the citrate buffer. All animals were carefully observed and regularly weighed throughout the entire experimental period. Any that showed undue weight loss or appeared unwell were killed by an overdose of anaesthetic followed by cervical dislocation. At 10, 16 and 22 weeks after the first injection, five NMU-treated animals and three controls were anaesthetized and killed and their maxillae removed and trimmed. The mucosal surfaces were washed with a gentle stream of freshly prepared phosphatebuffered saline. The maxillae were then fixed in glutaraldehyde 2% in 0-1 M cacodylate buffer (pH 7"4) at 4~ for at least 24 h, washed in distilled water, dehydrated in graded concentrations of acetone, and critical-point dried in liquid carbon dioxide. They were then glued to large aluminium stubs, sputter-coated with gold and examined in a Cambridge 360 scanning electron microscope. After this the maxillae were removed from the stubs, rehydrated in graded concentrations of ethanol, and demineralized in 10% formic acid. The specimens were then processed routinely for light microscopy and cut in either the sagittal or coronal plane. Sections were stained with haematoxylin and eosin (HE). Specimens from control animals were selected so that representative samples from all areas of the palatal gingiva and intermolar mucosa could be examined. Specimens from NMU-treated animals were selected so that a
Induced
Fig. 1.
Changes
in the Hamster
Palate
405
Control. Intermolar mucosa with rugae (R) and sessile and conical projections (arrows). Palatal gingiva (G) which merges with the intermolar mucosa and rugae; soft palate-like mucosa (S) with fungiform-like papilla (P); molar teeth (M). SEM. x 25.
representative sample of all the features not seen in control animals by scanning electron microscopy could be viewed by light microscopy. Results
Control Animals The appearance of the palatal gingiva and intermolar mucosa was similar in all animals and in the rostral two-thirds there was no sharp boundary between them (Fig. 1). In the rostral two-thirds, rugae extended from the gingiva medially although they never met in the midline. The rostral rugae were more prominent than the caudal rugae. There were clusters of small sessile and conical projections on the more rostral-lateral aspects of the intermolar mucosa and adjacent gingiva. The morphology of the cells covering the palatal gingiva and rostral two-thirds of the intermolar mucosa was similar. Regular flattened polygonal cells had a distinct honeycomb surface pattern of interconnecting micro-ridges that surrounded depressions. There were distinct cell boundaries and imprints of overlying cells that had been desquamated. The former consisted of two parallel linear ridges that were often separated by a narrow cleft. The latter were seen as a single linear ridge (Fig. 2). The junction between the rostral two-thirds and caudal third of the intermolar mucosa was sharp. The mucosa covering the caudal third was similar to that covering the soft palate and had fungiform-like papillae scattered over it (Fig. 1). Although the epithelium covering these latter structures was similar to that on the palatal gingiva and rostral two-thirds of the intermolar mucosa,
406
Fig. 2.
G . C . S y m e s e t al.
Control. Flattened polygonal cells with honeycomb surface pattern. Cell boundaries formed by two linear ridges (A); imprints of desquamated cells formed by a single linear ridge (B). SEM. x 3000.
that between the fungiform-like papillae was not. Here the epithelium was more uneven. The surface of the cells was smooth and lacked a honeycomb surface pattern. Distinct cell boundaries and cell imprints were only found occasionally. The fungiform-like papillae had one or two taste pore openings in their centre. By light microscopy the structure of all areas was as previously described (McMillan et al., 1994).
N M U - Treated Animals The majority underwent an initial small weight loss then showed a steady weight gain. However, two animals had to be killed in the first 10 weeks because of undue weight loss and five had to be killed near the end of the 22week period because of lethargy, loss of coat sheen and weight loss. At 10 weeks the surface of the palatal gingiva and rostral two-thirds of the intermolar mucosa was more irregular than in the control animals (Fig. 3). This irregularity was due to the presence of sessile surface projections, groups of small conical projections and saucer-shaped and conical depressions. The groups of conical projections were most common laterally between the more rostral rugae. The saucer-shaped depressions were relatively shallow (Fig. 4), had smooth rounded edges and were, in many instances, lined by a continuous layer of flattened epithelial cells. Some had dilated intercellular spaces at their base. The conical depressions extended for a greater distance into the epithelium than the saucer-shaped ones and usually had one or more defects or dilated intercellular spaces at their base (Fig. 5). T h e y were lined by a
I n d u c e d Changes in the H a m s t e r Palate
407
Fig. 3.
Ten weeks after N M U treatment. Irregular intermolar mucosa with sessile projections (arrows), conical projections (arrowheads) and saucer-shaped (S) and conical (C) depressions. Rugae (R); Palatal gingiva (G). SEM. x 40.
Fig. 4.
Ten weeks after N M U treatment. Epithelial lined saucer-shaped depression (S) with desquamating cells (arrowheads). SEM. x 290.
whorled arrangement of epithelial cells, a number of which were desquamating. The diameter of the oral opening of these depressions varied from 250 gm to less than 20 gm. Variable numbers of bacteria were found within them.
408
G . C . S y m e s et al.
Fig. 5.
Ten weeks after N M U treatment. Conical depression with desquamating cells (arrowheads) and basal defects and dilated intercellular spaces (arrows). SEM. • 250.
Fig. 6.
Sixteen weeks after N M U treatment. Irregular intermolar mucosa with conical depression (arrowhead) and conical projections (arrow's). Gingival lesions (L); palatal gingiva (G); unaltered soft palate-like mucosa (S). SEM. x 25.
Epithelial cells in all areas had a honeycomb surface pattern and cell boundaries and cell imprints were similar to those in the control animals. The changes at 16 weeks were similar to those at 10 weeks but more pronounced (Fig. 6). The conical depressions occurred more frequently, were
I n d u c e d C h a n g e s in the H a m s t e r P a l a t e
]Fig. 7.
409
Sixteen weeks after NMU treatment. Conical depression with basal defects and dilated intercellular spaces (D). SEM. x 650.
often of greater diameter, frequently extended for a greater distance into the epithelium and had larger defects or dilated intercellular spaces in their base (Fig. 7). The lateral margins of the cells lining these depressions were often thickened and separated by dilated intercellular spaces. However, they still had a honeycomb surface pattern together with distinct cell boundaries and cell imprints (Fig. 8). The conical projections in the 16-week specimens (Fig. 6) were larger than those seen at 10 weeks. They were often grouped together and formed papillomatous-like lesions. Towards the apex of these projections the most superficial cells were often desquamating, revealing an irregular folded surface. The surfaces of the deeper cells in these areas were crossed by elongated ridges rather than having a honeycomb surface pattern. Cell boundaries were also indistinct. All of the changes seen at i0 and 16 weeks were present at 22 weeks. However, at 22 weeks they were more extensive, involved more of the mucosa and often masked the normal rugal pattern. Some of the conical depressions were up to 500 gm in diameter at their oral opening. Distinct papillomas were a common finding (Fig. 9). The papillomas were demarcated from the surrounding mucosa by a rim-like arrangement of peripheral cells. The epithelium of the papillomas had the characteristic honeycomb surface pattern, together with distinct cell boundaries and cell imprints similar to those in control animals. The surface structure of those parts of the mucosa that did not show any of the previously described changes varied. In some areas it was similar to that in control animals, although the number of desquamating cells was increased. Elsewhere the size and shape of the cells varied, the surface
410
a . C . Symes e t
al.
Fig. 8.
Sixteen weeks after N M U treatment. Cells lining the conical depression in Fig. 7 that have a honeycomb surface pattern. Cell imprints (arrows). SEM. x 5500.
Fig. 9.
Twenty-two weeks after N M U treatment. Papilloma with numerous projections (P). SEM. x 190.
was irregular, and there were irregular cell boundaries and elongated microridges on the surface of the cells (Fig. I 0). The changes in the N M U animals all affected the rostral two-thirds of the intermolar mucosa and to a lesser extent the palatal gingiva (Fig. 6). Changes
Induced
Fig. 10.
Changes
in the Hamster
Palate
411
Twenty-two weeks after NMU treatment. Area ofmucosa with cells that have elongated microridges. SEM. x 5000.
were not seen on the soft palate-type intermolar mucosa covering the caudal third of the hard palate. Light microscopy showed changes similar to those described previously (McMillan et al., 1994). The saucer-shaped and conical depressions were associated with either keratin plugging, focal areas of individual cell keratinization, or both. Occasionally the conical depressions were associated with epithelial lined duct-like structures that extended into the underlying connective tissue. In the 22-week specimens there were areas of dysplastic and at times malignant change in the deeper epithelial strata, irrespective of the surface structure of the epithelium as seen by scanning electron microscopy. Discussion
The mucosa covering the caudal third of the intermolar area was similar to that of the soft palate. A similar arrangement is found in some strains of rats (Wong and Wilson, 1983) but not in others (McMillan, 1974). The intermolar mucosa, palatal gingiva and fungiform-like papillae in control hamsters were characterized by flattened polygonal cells that had a honeycomb surface pattern, were separated by distinct cell boundaries and were crossed by the imprints of the more superficial cells that had been desquamated. This appearance resembled that found in similar areas of rat oral mucosa (McMillan, 1974, 1979) and hamster cheek pouch (McMillan et al., 1982). The structure of the rest of the soft palate-type mucosa, which was folded and lacked a honeycomb surface pattern and obvious cell boundaries and imprints, was similar to that in the rat (McMillan, 1979).
412
G.C. Syrnes
e t al.
Although a honeycomb surface pattern is characteristic of keratinized oral epithelium in a number of species (Cleaton-]ones and Fleisch, 1973; KullaaMikkonen, 1986), a similar pattern is present on human buccal mucosa (Matravers and Tyldesley, 1977) and on monkey cheek and lip mucosa (Nair and Schroeder, 1981), which are not keratinized. In the rat, oral epithelium with a honeycomb surface pattern is completely orthokeratinized, but oral epithelium without such a pattern is incompletely orthokeratinized (McMillan, 1979). All of the oral mucosa in the hamster is also keratinized, but whether there is a similar association between the surface pattern and degree of keratinization is not known. The proportion of NMU-treated animals that had to be killed (7/23) was similar to that reported in a previous study (McMillan el al., 1994), as were the changes seen by light microscopy in palatal gingiva and intermolar mucosa. However, in the present study there was a longer latent period and the number of exophytic lesions was reduced, as were the number and depth of invasive lesions. These differences probably resulted from the use of different strains of hamsters in the two studies. The principal surface changes that occurred after the administration of N M U were the appearance of saucer-shaped and conical depressions and sessile, conical and papillomatous projections. These gradually increased in size and number, distinct papillomas being found in the 22-week specimens. Some of the conical depressions were the oral openings of the keratinfilled duct-like structures, but others were associated with extensive areas of individual cell keratinization or keratin plugging, which are characteristic of this experimental model (McMillan et al., 1994). The saucer-shaped depressions were associated with less extensive areas of keratin plugging or individual keratinization, or both. The areas of individual cell keratinization often resembled the focal acantholytic dyskeratosis of Darrier's disease (Philipsen and Fisker, 1983) which is a benign, usually inherited, slowly progressive kerototic skin disorder that may also involve the oral and genital mucosa. These features are not characteristic of human oral dysplasia and squamous cell carcinoma. As the structural changes associated with the saucer-shaped and conical depressions are rarely if ever seen in DMBA-treated hamster cheek pouches, it is not surprising that neither type was seen in the scanning electron microscope studies of Hassanin et al. (1987) and Hassanin and Ashrafi (1988). That exophytic lesions were only infrequently found on the palatal gingiva confirms the findings ofMcMillan et al. (1994). Suzuki (1976) reportd exophytic lesions of the gingiva in hamsters after N M U treatment, but Kohgo et al. (1990) found only invasive ones. Exophytic lesions of the palatal intermolar mucosa in hamsters after N M U treatment was also reported by Edwards and Germain (1980). The presence of exophytic lesions, including papillomas, is typical of DMBA-initiated carcinogenesis in mouse skin (Di Giovanni, 1992) and hamster cheek pouch (Boyd and Reade, 1988). Such changes are uncommon in carcinomas of the human skin and oral cavity. The sessile and conical projections in the more rostral lateral aspects of the intermolar mucosa were probably derived from similar small structures, as seen in the controls.
Induced Changes in the H a m s t e r Palate
413
However, those either more medially or caudally situated must have arisen de novo. By 16 weeks many of these projections were much larger than the ones in the controls and in some instances a number of the conical ones were grouped together to form papillomatous-like lesions which presumably gave rise to the distinct papillomas seen at 22 weeks. The soft palate-like mucosa that covered the caudal third of the intermolar mucosa was unaffected in all of the NMU animals, indicating the varying susceptibility of different oral sites to chemical carcinogens. Such. variability is found in animals treated with either N M U (Edwards and Germain, 1980; Kohgo et al., 1990), DMBA (Eveson, 1981) or 4 N Q O (Steidler and Reade, 1984; Boyd and Reade, 1988). Examination of DMBA-treated hamster cheek pouches by scanning electron microscopy showed that by 8 weeks the mucosal surface was totally altered, most of the cells being spindle-shaped and covered by microvilli (Hassanin et al., 1987; Hassanin and Ashrafi, 1988). Variation in cell size, thickening of individual cells, disruption of the normal intercellular relationships, loss of distinct cell boundaries and replacement of the normal surface pattern of micro-ridges by pebbly surface projections or microvilli have been described in many of the scanning electron microscope studies ofdysplastic and malignant epithelial cells from the human oral cavity (Morgenroth and Morgenroth, 1970; Matravers and Tyldesley, 1977; Banoczy et al., 1980; Reichart et al., 1981; Nakao, 1983; Reichart and Althoff, 1983), nasal cavity (Boysen and Reith, 1982) and uterine cervix (Jordan and Williams, 1971; Rubio and Kranz, 1976). Similar changes were seen in experimental carcinogenesis in the rat bladder (Jacobs et al., 1976). In the 10- and 16-week N M U specimens, scanning electron microscopy showed that the surface structure of the epithelium, including that covering the three types of exophytic lesion and the two types of depression, generally resembled the surface structure seen in the controls; slight differences included an increase in the number ofdesquamating cells and, in the conical depressions, dilated intercellular spaces and thickened epithelial cells. At 22 weeks much of the epithelium also had a surface structure similar to that of the controls. However, some of the epithelium between the exophytic lesions and depressions showed variations in cell size and shape and irregular cell boundaries, the characteristic honeycomb surface pattern on the cells being replaced by more elongated micro-ridges. By light microscopy the deeper strata of this more irregular epithelium varied. Sometimes it was normal, sometimes it was dysplastic and sometimes it was malignant. There were also similar variations in the structure of the deeper epithelial strata in areas where the surface structure, as seen by scanning electron microscopy, was similar to that in the controls. No changes indicative of epithelial dysplasia or malignancy were found by scanning electron microscopy, although such changes were in fact present in the deeper strata of the epithelium. However, had the hamsters been allowed to survive longer, or had a strain of hamster in which the carcinomas arose earlier been used, more of the surface changes said to be characteristic of epithelial dysplasia and malignancy might have developed. The relative lack of overt change in the morphologT of most of the surface
414
G.C. Symes e t al.
cells in this study serves to emphasize that in a thick keratinized stratified squamous epithelium, dysplastic and malignant change in the deeper cell layers may, initially at least, be masked by the more normal superficial layers.
Acknowledgments This work was supported by a University of Otago Research Grant. The paper was written while one of the authors (M. D. McMillan) was on leave in the Department of Oral Pathology, University of Sheffield, and Professor C.J. Smith is thanked for making this" possible. Mrs P.K. Cosgriff and Mrs E.A. McMillan are thanked for typing the manuscript.
References Banoczy, J., Lapis, K. and Albrecht, M. (1980). Scanning electron microscopic study of oral leukoplakia. Journal of Oral Pathology, 9, 145-154. Boyd, N. M. and Reade, P. C. (1988). Mechanisms of carcinogenesis with particular reference to the oral mucosa. Journal of Oral Pathology, !7, 193 201. Boysen, M. and Reith, A. (1982). The surface structure of the human nasal mucosa II. Metaplasia, dysplasia and carcinoma in nickel workers. A correlated study by scanning/transmission electron and light microscopy. VzrchowsArchiv B Cell Pathology, 40, 295-309. Cleaton-Jones, P. and Fleisch, L. (1973). A comparative study of the surface of keratinized and non-keratinized oral epithelia. Journal of Periodontal Research, 8, 366-370. Di Giovanni, J. (1992). Multistage carcinogenesis in mouse skin. Pharmacology and Therapeutics, 54, 63-128. Edwards, M. B. and Germain, J. P. (1980). Oral carcinogenesis by intraperitoneal Nmethyl-N-nitrosourea in the Syrian hamster. Archives of Oral Biology, 25, 631 633. Eveson, J. W. (1981). Animal models of intra-oral chemical carcinogenesis: a review. Journal of Oral Pathology, 10, 129-146. Hassanin, M.B. and Ashrafi, S.H. (1988). Comparative light, scanning and transmission electron microscopic study of chemically induced premalignant lesions in the hamster's cheek pouch. Ultrastructural Pathology, 12, 341 350. Hassanin, M.B., Ashrafi, S.H. and Das, A. (1987). Surface ultrastructural changes in chemically induced premalignant lesions in the hamster cheek pouch. Journal of Oral Pathology, 16, 273-277. Herrold, K.M. (1966). Carcinogenic effect of N-methyl-N-nitrosourea administered subcutaneously to Syrian hamsters, journal of Pathology and Bacteriology, 92, 35-41. Herrold, K.M. (1968). Odontogenic tumors and epidermoid carcinomas of the oral cavity. Oral Surgery, Oral Medicine and Oral Pathology, 25, 262-272. Jacobs, J.B., Arai, M., Cohen, S.M. and Friedell, G.H. (1976). Early lesions in experimental bladder cancer: scanning electron microscopy of cell surface markers. CancerResearch, 36, 2512-2517. Jordan, J. A. and Williams, A. E. (1971). Scanning electron microscopy in the study of cervical neoplasia. Journal of Obstetrics and Gynaecology, 78, 940-946. Jungell, P., Malmstrom, M., Wartiovaara, J. and Kouttinen, Y. (1987). Ultrastructure of oral leukoplakia and lichen planus. II. A correlated scanning and transmission electron microscopic study of epithelial surface cells. Journal of Oral Pathology, 16, 287 293. Kohgo, T., Uy, G.H., Iizuka, T., Shindoh, M. and Amemiya, A. (1990). Histopathologic study of squamous cell carcinoma of the gingiva in edentulous and dentulous jaws of hamsters treated with N-methylnitrosourea. Journal of Oral Pathology, 19, 202 207. Kullaa-Mikkonen, A. (1986). Scanning electron microscopic study of surface of human oral mucosa. ScandinavianJournal of Dental Research, 94, 50 56.
Induced Changes in the H a m s t e r Palate
415
Matravers,J. and Tyldesley, W. (1977). Scanning electron microscopy of oral epithelial cells. Part I. Normal and malignant tissues. British Journal of Oral Surgery, 15, 193-202. McMillan, M.D. (1974). A scanning electron microscopic study of keratinized epithelium of the hard palate of the rat. Archives of Oral Biology, 19, 225-229. McMillan, M.D. (1979). The surface structure of the completely and incompletely orthokeratinized oral epithelium in the rat. A light, scanning and transmission electron microscope study. American Journal of Anatomy, 156, 337-351. McMillan, M. D., Smillie, A. C. and Gray, D. W. (1982). The surface structure of the epithelium of the hamster cheek pouch. Archives of Oral Biology, 27, 623 634. McMillan, M. D., Smith, C.J. and Ramirez, V. (1994). The effect of intraperitoneal N-methyl-N-nitrosourea on hamster palatal mucosa. Archives of Oral Biology, 39, 519-528. Morgenroth, K. von and Morgenroth, K. (1970). Rasterelektronenmikroskopische untersuchungen zur morphologie der leukoplakie der mundhohle. Deutsche Zahn aerztliche Zeitschrift, 25, 1054-1060. Nair, P. N. R. and Schroeder, H. E. (1981). Variation and density of microplications in superficial cells of the normal oral lining mucosa in the monkey Macacus fascicularis. Archives of Oral Biology, 26, 837-843. Nakao, I. (1983). Comparative studies on exfoliated cells from the oral mucosa by light and scanning electron microscopy. Tsurumi Shigaku, 9, 151-177. Philipsen, H.P. and Fisker, A.V. (1983). Focal acantholytic dyskeratosis in experimental oral carcinogenesis in rats. Journal of Oral Pathology, 12, 30 36. Reichart, P. A. and Althoff, J. (1983). Oral leukoplakia: a scanning electron microscopic study of epithelial surface patterns. International Journal of Oral Surgery, 12, 159-164. Reichart, P. von, Althoff, J. and Reilmann, B. (1981). Klinische und rasterelektronenmikroskopische untersuchungen oraler papillomatosen. Deutsche Zah naerztliche Zeitschrift, 36, 386-388. Rubio, C.A. and Kranz, I. (1976). The exfoliating cervical epithelial surface in dysplasia, carcinoma in situ and invasive squamous carcinoma I. Scanning electron microscopic study. Acta Cytologica, 20, 144-150. Steidler, N. E. and Reade, P. C. (1984). Experimental induction of oral squamous cell carcinomas in mice with 4-nitroquinoline-1-oxide. Oral Surgery, Oral Medicine and Oral Pathology, 57, 524-531. Suzuki, M. (1976). Histological studies of early changes of the hamster gingival epithelium by N-nitrosomethylurea. Bulletin of Tokyo Medical and Dental University, 23, 115 134. Svensson, S. and Heyden, G. (1982). Experimental induction of irreversible precancerous changes in the palatal epithelium of the rat. International Journal of Oral Surgery, 11, 52 58. Wilson, D. F., Jiang, D.J., Leong, A.S. and Wiebkin, O.W. (1993). Laminin and type IV collagen in experimental rat oral carcinomas. Journal of Comparative Pathology, 108, 375-381. Wong, P.N.C. and Wilson, D.F. (1983). 4-Nitroquinoline-l-oxide-induced carcinogenesis in the rat palate. Journal of Oral Pathology, 12, 375 384.
Received, September 6 th, 1994] Accepted, February 8th, 1995 J