Search

Mutations in the region corresponding to the N-terminal phosphorylation sites (codons 1-51) of the rat beta-catenin gene (Ctnnb1) were investigated in rat colon tumors induced by 1-hydroxyanthraquinone (1-HA) plus methylazoxymethanol (MAM) acetate, by using polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) analysis. The beta-catenin gene was also screened for mutations in rat brain and oral tumors induced by ethyl nitrosourea (ENU) and 4-nitroquinoline 1-oxide (4-NQO), respectively. In colon tumors, beta-catenin gene mutations were found in two of three adenomas (67%) and 26 of 28 adenocarcinomas (93%), with a total incidence of 90% (28 of 31 adenomas plus adenocarcinomas). Eight (29%) were (34)G-->T (second position), eight (29%) were (32)G-->A (first position), five (18%) were (34)G-->A (first position), five (18%) were (41)C-->T (second position), one (4%) was (34)G-->A (second position), and one (4%) was (32)A-->G (second position), mutations, resulting in the substitutions of Gly(34)-->Val, Asp(32)-->Asn, Gly(34)-->Arg, Thr(41)-->Ile, Gly(34)-->Glu, and Asp(32)-->Gly, respectively. The (34)G-->T (second position) mutations found in this study were unique compared to those found in other carcinogen-induced rat colon carcinogenesis models. In contrast, beta-catenin gene mutations were not found in either the brain or oral tumors. These results suggest that mutations in the beta-catenin gene in rat tumors occur in specific tissues or organ sites and in a carcinogen-specific manner. Thus, the mutation spectrum in the beta-catenin gene is organ- and chemical carcinogen-specific., (Copyright 2001 Wiley-Liss, Inc.)

Recent evidence suggests that the beta-catenin gene (CTNNB1) acts as an oncogene, and some human colon tumors with an intact APC gene have activating mutations in CTNNB1. In this study, mutations in the region corresponding to N-terminal phosphorylation sites (codons 1-51) of the rat Ctnnb1 gene were investigated in 20 colon tumors associated with ulcerative colitis and induced with methylazoxymethanol acetate and 1-hydroxyanthraquinone. Ninety percent (18 of 20) of the tumors induced in male F344 rats harbored mutations, which were detected in three of four adenomas (75%) and 15 of 16 adenocarcinomas (94%). Of 18 total missense mutations, 13 (72%) were G-->A transitions at position 101, three were G-->A transitions at position 94, and two were C-->T transitions at position 122, resulting in the amino acid substitutions Gly34-->Glu, Asp32-->Asn, and Thr41-->Ile, respectively. Although there were no mutations in the Apc gene, as we previously reported in the same tumor samples, the results obtained in this study strongly implicate the Apc-beta-catenin-T-cell factor (Tcf) signaling pathway in methylazoxymethanol acetate, 1-hydroxyanthraquinone-induced colon carcinogenesis.

A rat model for human ulcerative colitis (UC) has been developed by using 1-hydroxyanthraquinone (1-HA) to cause severe inflammation of colonic mucosa. 1-HA also has synergistic effects on the carcinogenicity of methylazoxymethanol (MAM) acetate in the rat colon. In this study, four adenomas and 16 adenocarcinomas induced in male F344 rats by 1-HA and MAM acetate were examined for mutations in the entire coding regions and introns flanking coding exons of the APC gene by polymerase chain reaction (PCR)-single strand conformation polymorphism (SSCP) and PCR-restriction-SSCP analyses. No mutations were found. These results, together with our previous observations of a relative lack of Ki-ras gene mutations in the same tumors, are similar to those found in human UC-associated colon cancer, suggest a common pathway in these two systems, although they are different in their implication of p53 mutations. Therefore, this model may have some relevance and application to the study of colon cancer in human inflammatory bowel disease, which is not associated with APC mutations or with Ki-ras or p53 mutations.

Using western blotting and immunochemical analysis, we investigated alterations in the expression of the apoptosis-related proteins bcl-2, bax, and bcl-X in colonic adenocarcinomas induced by subcutaneous injection of azoxymethane (AOM) (15 mg/kg body weight weekly for 2 wk) into male Sprague-Dawley rats. Expression of the apoptosis-repressor bcl-2 in the colonic tumors was significantly weaker (0.6-fold) than that in adjacent non-neoplastic mucosa. The expression of bax protein, an apoptosis accelerator, was significantly stronger (7.33-fold) in all the tumors than in the non-tumoral mucosa. bcl-XL protein, which functions as a repressor of apoptosis, was significantly upregulated (3.23-fold) in all the tumors when compared with the non-neoplastic mucosa. There was no significant difference between the expression of these proteins in the non-neoplastic mucosa of the AOM-treated rats and in the normal mucosa of saline-treated control rats. As determined by immunohistochemical analysis, the tumor cells had more bax and bcl-X protein. These findings indicate that the regulation of the apoptosis-related proteins bcl-2, bax, and bcl-XL was altered in the AOM-induced colonic neoplastic tissue. In terms of resistance to apoptosis, elevated levels bcl-XL protein may have considerable meaning in this experimental model as well as in human colorectal cancer.

Telomerase activity in tissues may be related to tumor development, especially malignant conversion, in humans. However, there are few reports about telomeres and telomerase activity in animals. In this study, we examined telomerase activity in rat colon carcinogenesis and in normal rat liver tissue and compared it with that of human colon cancer tissues. This is the first report concerning telomerase activity in rats. F344 rats were used, and colon neoplasms were induced with methylazoxymethanol acetate. There was telomerase activity in not only the induced colon neoplasms but also the colon mucosa and livers of untreated rats, in contrast with the results from normal human somatic tissues in previous reports. Indeed, we also observed negative results in normal human mucosa, despite the positive results in colon-cancer tissues. These findings suggest that there is a difference in the telomerase activities in humans and rats. Because rat telomeres are very long (20-100 kp, average 50 kp) compared with human telomeres (5-15 kp, average 12 kp), the difference in the telomere lengths of rats and humans might be related to their enzyme activities, although this is still unclear. Furthermore, because the inhibition of telomerase has been proposed as a novel cancer therapy for humans, the rat model presented here, in which telomerase is expressed in somatic tissues, may be useful for studies of telomerase inhibition, including inhibition by chemopreventive agents.

The alkylating agent 4-nitroquinoline 1-oxide (4-NQO) is a powerful carcinogen and induces squamous cell hyperplasia, squamous cell dysplasia, papilloma, and squamous cell carcinoma (SCC) in rat oral epithelia. Oral cancers induced by a single application of 4-NQO develop through a multistage process in a way similar to the development of this cancer in humans. In this study, mutations in exons 1 and 2 of Ki-ras, N-ras, and Ha-ras and exons 4-7 of p53 were examined by polymerase chain reaction (PCR)-single strand conformation polymorphism (SSCP) analysis, followed by PCR-direct sequencing for the confirmation of mutations. Samples for the mutation analysis were obtained from dysplasias, papillomas, and SCCs on the tongue epithelia induced in F344 rats by adding 4-NQO (20 ppm) to their drinking water for 8 wk. The Ha-ras mutations (61A-->T transversions in the second position) were found in five of 29 (17%) samples (one dysplasia and four SCCs). However, no mutations were detected in either Ki-ras, N-ras, or p53 under two different conditions of PCR-SSCP analysis. We suggest that some neoplasms in oral carcinogenesis induced by 4-NQO may involve Ha-ras mutations but not mutations in Ki-ras, N-ras, or p53. The 4-NQO-induced rat oral carcinogenesis model may provide a system for evaluation of the mechanisms of multistage oral carcinogenesis associated with Ha-ras mutation without Ki-ras, N-ras, or p53 mutation.

1-Hydroxyanthraquinone (1-HA), which is present in some herbs, and methylazoxymethanol (MAM) acetate, a metabolite of azoxymethane, show synergistic carcinogenicity in rat colon, and 1-HA induces ulcerative changes with simultaneous severe inflammation of the entire colon. In this study, mutations in Ki-ras (exons 1 and 2) and p53 (exons 4-7) were studied by polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) analysis. Of 18 adenomas and 38 adenocarcinomas induced in male F344 rats (52 tumors induced by 1-HA plus MAM acetate, three by 1-HA alone, and one by MAM acetate alone), no mutations in Ki-ras or p53 were detected under two conditions of PCR-SSCP analysis. Because human colon carcinomas from patients with ulcerative colitis have a very low incidence of Ki-ras mutation, this experimental system would be a good animal model of human colon carcinomas with ulcerative colitis and of human colon carcinomas without Ki-ras or p53 mutations.

Phospholipase C (PLC), which hydrolyzes phosphoinositides, has been implicated as a key enzyme in signal transduction. We examined the expression of an isozyme of PLC, PLC-delta, in rat colon neoplasms induced by methylazoxymethanol (MAM) acetate. Large-bowel neoplasms were observed in five of 10 rats given MAM acetate (25 mg/kg body weight, by interperitoneal injection at 6 and 7 wk of age) 40 wk after treatment. Expression of PLC-delta in the neoplasms was not detected by northern blot analysis, and a low level of expression was detected by immunoblot analysis, although PLC-delta expression was apparent in the non-neoplastic colon mucosae of MAM acetate-treated rats as well as in the colon mucosae of control rats. Furthermore, analysis by reverse transcriptase-polymerase chain reaction revealed that the ratio of the expression of PLC-delta to that of beta-actin in the neoplasms was significantly lower than the ratios in the non-neoplastic colon mucosae of carcinogen-treated and control rats (P < 0.01). However, the ornithine decarboxylase (ODC) activity in the neoplasms was significantly greater than that of the non-neoplastic and control mucosae (P < 0.001). The differences in the levels of PLC-delta expression in neoplastic and non-neoplastic tissues and the inverse correlation of PLC-delta expression with ODC activity may suggest that PLC-delta has little effect on the PLC-mediated mitogenic signaling system, at least in MAM acetate-induced colon neoplasms in rats.