Use of p53 transgenic mice in the development of cancer models for multiple purposes.

The tumor suppressor gene p53 is perhaps the most commonly mutated gene in human cancer, being mutated in a high percentage of colon, breast, skin, bladder, and many cancers of the aerodigestive tract. Individuals with Li-Fraumeni syndrome, who routinely have a germline mutation in the p53 tumor suppressor gene, are at high risk for lung cancer, confirming its intimate role in lung tumorigenesis in humans. In contrast, the majority of chemically induced or spontaneous cancers in rodents do not contain mutations in p53. Therefore, we examined a transgenic mouse that contains a dominant negative mutation (Arg135Val) in the p53 gene placed under the control of its own endogenous promoter. The resulting mice have 3 copies of the mutated transgene as well as 2 normal p53 alleles. In the chemical carcinogenesis studies, we employed mice containing the mutated p53 gene to examine for carcinogen susceptibility. We found that mice with the p53 mutation, on an A/J F1 background, were more susceptible to a number of potential lung carcinogens, including N-methyl-N-nitrosourea (MNU) and the known tobacco carcinogens 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and benzo(a)pyrene (BP). Mice with a mutant p53 developed larger tumors and roughly 3 times as many tumors, emphasizing the potential effects of a p53 mutation both on tumor initiation and progression. In addition, we examined 2 nonlung carcinogens, 1,2-dimethylhydrazine (DMH), a colon carcinogen, and N-butyl-N-(4-hydroxybutyl)-nitrosamine (OHBBN), a bladder carcinogen. Interestingly a germline p53 mutation increased the incidence of DMH-induced colon, lung, hepatic, and uterine tumors, while having limited effects on OHBBN-induced bladder tumors. Because of its heightened susceptibility we are examining the use of this model in smoke-induced tumorigenesis in A/J mice as well. Employing the lung adenomas induced by NNK, we found that mice with or without a p53 mutation were equally susceptible to the chemopreventive effects of dexamethasone plus myo-inisitol and green tea. These tumors, which arise in a highly reproducible manner in p53 transgenic mice following carcinogen treatment, have mutations in both p53 and the K-ras oncogene. Thus, this model appears useful for examining for potential chemotherapeutic agents. p53-mutated or wild-type mice were equally susceptible to the therapeutic effects of Taxol or Adriamycin. Interestingly, piroxicam was similarly effective in inhibiting colon tumor formation by DMH in mice with or without a mutation in the p53 tumor suppressor gene. In contrast, lung and uterine tumors developing in these mice were not susceptible to the chemopreventive effects of piroxicam. In summary, mice with mutations in the p53 tumor suppressor gene appear to be particularly applicable for basic mechanistic studies, for screening for potential carcinogens, and for screening for chemopreventive or chemotherapeutic agents.