Development of hepatocellular carcinomas in rats caused by a choline-deficient, L-amino aciddefined (CDAA) diet, usually associated with fatty liver, fibrosis, cirrhosis and oxidative DNA damage, has been recognized as a useful model of hepatocarcinogenesis caused by endogenous factors. In the present study, in order to further explore involved factors and genes, we established an equivalent model in spontaneous liver tumor-resistant C57BL/6J mice. Six-week-old males and females were continuously fed the CDAA diet and histological liver lesions and oxidative DNA damage due to 8-hydroxydeoxyguanosine (8-OHdG) were examined after 22, 65 and 84 weeks. In male mice, fatty change and fibrosis were evident at 22 weeks, and preneoplastic foci of altered hepatocytes were seen at an incidence of 8/8 (100%) and a multiplicity of 6.6 ± ± ± ±4.0 per mouse at 65 weeks. Hepatocellular adenomas and carcinomas developed at incidences of 16/24 (66.7%) and 5/ 24 (20.8%), and multiplicities of 1.42 ± ± ± ±1.32 and 0.29 ± ± ± ±0.62, respectively, at 84 weeks. The female mice exhibited resistance to development of these lesions. The CDAA diet also increased 8-OHdG levels in male but not female mice. These results indicate that a CDAA diet causes hepatocellular preneoplastic foci, adenomas and carcinomas associated with fibrosis and oxidative DNA damage in mice, as in rats, providing a hepatocarcinogenesis model caused by endogenous factors in mice.

The comparative carcinogenic activities of a choline‐deficient, l‐amino acid‐defined diet (CDAADD) and a purified choline‐deficient diet (CDD) for rat liver were studied in terms of both 8‐hydroxy‐deoxyguanosine induction, a marker of DNA damage induced by oxidative stress, and development of γ‐glutamyltransferase(GGT)‐positive putative preneoplastic lesions, including foci and hyperplastic nodules. Twelve weeks after the beginning of treatment, DNA damage could be detected in the liver DNA of rats receiving either CDAADD or CDD, the degree being significantly greater in the former case. Similarly, while GGT‐positive liver lesions were induced by both CDAADD and CDD, the numbers were higher and the areas of lesions were larger in rats receiving CDAADD than in those given CDD. Histologically, hyperplastic nodules were induced in the livers of animals administered CDAADD whereas only foci were seen in the CDD case. The results thus indicate that oxidative stress might be directly involved in rat liver carcinogenesis by CDD and, to a greater degree, with CDAADD.

No abstract

A rapid-production model incorporating the principle of selection by resistance to cytotoxicity demonstrated earlier for liver carcinogenesis in rats was established for pancreatic carcinoma development in Syrian hamsters. Adenocarcinomas were induced in 84% of treated animals by 10 weeks after initiation with 70 mg of N-nitroso-bis(2-oxopropyl)amine (BOP) per kg of body weight when augmentation pressure (choline-deficient diet combined with DL-ethionine and L-methionine and administration of 20 mg/kg BOP upon return to basal diet) was applied three times. A 52% yield of cholangiocellular tumors also resulted from this experimental protocol.

Expression of cyclooxygenase (COX)-2 protein during rat hepatocarcinogenesis associated with fatty change, fibrosis, cirrhosis and oxidative DNA damage, caused by a choline-deficient, L-amino acid-defined (CDAA) diet were investigated in F344 male rats, along with the chemopreventive efficacy of the specific COX-2 inhibitor, nimesulide (NIM). Nimesulide, which was administered in the diet at concentrations of 200, 400, 600 and 800 p.p.m. for 12 weeks, decreased the number and size of preneoplastic enzyme-altered liver foci, levels of oxidative DNA damage, and the grade and incidence of fibrosis in a dose-dependent manner. A preliminary long-term study of 65 weeks also revealed that 800 p.p.m. NIM decreased the multiplicity of neoplastic nodules and hepatocellular carcinomas and prevented the development of cirrhosis. Western blot analysis revealed that COX-2 protein was barely expressed in control livers and increased approximately 2.9-fold in the livers of rats fed on a CDAA diet for 12 weeks and approximately 4.5-5.4-fold in tumors, with a diameter larger than 5 mm, at 80 weeks. Immunohistochemically, COX-2 protein was positive in sinusoidal and stromal cells in fibrotic septa, which were identified by immunoelectron microscopy as Kupffer cells, macrophages, either activated Ito cells or fibroblasts, after exposure to the CDAA diet for 12 weeks, whereas it was only occasionally weakly positive in sinusoidal, probably Kupffer, cells in control livers. In neoplastic nodules in rats fed on a CDAA diet for 30 and 80 weeks, sinusoidal cells and cells with relatively large round nuclei and scanty cytoplasm were strongly positive for COX-2 protein, with the neoplastic hepatocytes in the minority of the nodules, but not the cancer cells, being moderately positive. These results clearly indicate that rat hepatocarcinogenesis, along with fatty change, fibrosis and cirrhosis, is associated with increased expression of COX-2 protein, and point to the chemopreventive efficacy of a selective COX-2 inhibitor against, at least, the early stages of hepatocarcinogenesis.