As it is impossible to perform a longitudinal experiment of carcinogenesis of human samples, experimentally induced carcinoma has been used for this purpose. DMBA induced carcinoma on hamster buccal pouch mucosa has been frequently cited for its stability and high induction rate. We have been attending to this aspect of research. This article presents the findings from PAS stained sections on the process of DMBA induced hamster buccal pouch carcinogenesis.
The PAS-positive granules in the cytoplasm of epithelial cells, regarded as glycogen, were consumed by digestion of amylase. There was no difference between the experimental groups painted DMBA for no more than 6 weeks and the control groups which contained very little glycogen. But on the groups painted for more than 8 weeks, papillomas accumulated a large amount of glycogen in the cytoplasm of the spinous cell layer, and carcinomas did a little in the keratinized part. Epithelium with atypia revealed less glycogen than that of the other part of the epithelium. Carcinoma in situ contained extremely less.
Concerning on the PAS-basement membrane, that is PAS positive membraneous structure at the epithelium-connective tissue junction, there was no distinct difference between the controls and the up to 6-week-painted groups. In the groups painted more than 8 weeks, papillomas occasionally had thin membranes, and carcinomas were absent or faint.
Glycocalyx in the intercellular spaces was clearly observed in the controls and the up to 6-week-painted groups. Papillomas reduced the intercellular cohesion, and carcinomas showed wide intercellular spaces without any glycocalyx. The epithelium with atypia and carcinoma in situ also reduced intercellular cohesion.
Although PAS staining on the DMBA induced carcinogenesis could not predict the malignant potentiality of the epithelium, it is clarified that the induced tumors and epithelia with atypia closely resembled human oral carcinomas and precancerous lesions.
