Summary
Transgenic mice overexpressing a rWAP-p53 minigene containing an Arg-His mutation at amino acid 172 in the mammary gland develop normally, and exhibit few spontaneous tumors. Nevertheless, when these mice are given pituitary isografts and are treated with the chemical carcinogen DMBA, they develop mammary tumors significantly earlier than do wildtype controls, and have a greater number of tumors per animal than that seen in controls. Analysis of the two groups of tumors demonstrated that there were no significant differences in either apoptosis or proliferation. However, flow cytometric analysis revealed that aberrant ploidy was more often found in the tumors from the 172 R-H p53 transgenic mice. Furthermore, this aberrant ploidy was also seen when these transgenic mice were independently crossed with mice carrying either WAP-TGFa, MMTV-neu, or WAP-des-IGF-1, which strongly suggests that the 172 R-H p53 mutant predisposes these mice to genomic instability.

Citations
Li B, Murphy KL, Laucirica R, Kittrell F, Medina D, Rosen JM .  A transgenic mouse model for mammary carcinogenesis.  Oncogene 1998 Feb 26;16(8):997-1007

Background
p53 is the most frequently-mutated gene in human cancers, and is altered in approximately 40% of breast cancers. The p53 knockout mouse has provided a very valuable model in which to study the role of p53 in cancer initiation and progression, but is not ideally suited for the study of the role of this protein in breast cancer, as mice often die of lymphomas and sarcomas prior to developing mammary tumors. There are three 'hotspot' p53 mutations frequently found in breast cancer. In order to study the role of one of these mutations, 175 R-H (the murine 172 is equivalent), in the initiation and progression of breast cancer, rat whey acidic protein sequences were used to target a p53 minigene bearing the 172 R-H mutation to the mammary gland. Lines of mice expressing the transgene were generated, and proved to be developmentally normal, although more susceptible to mammary tumorigenesis as a result of either carcinogen treatment or coexpression of any of three oncogenes in the mammary gland.

Transgene
WAP-p53 172 R-H

mouse strain
FVB

Mammary phenotype

Mammary development
The 172 R-H transgene appears to have little or no effect on normal mammary gland development. The morphology of transgenic and nontransgenic mammary glands were compared by whole mount and H&E staining at 17dP and 2dL, as well as in the virgin gland and following involution, and no significant differences were observed. Cell proliferation in glands taken at 17dP and 2dL was assessed by BrdU incorporation, and no significant difference in proliferation was observed between the two groups. Finally, apoptosis in forced involution was also compared between the two groups, again revealing no significant difference. These mice develop few spontaneous mammary tumors prior to one year of age even following multiple pregnancies. However, when these mice and controls were given pituitary isografts and treated with the carcinogen DMBA, the transgenic mice developed tumors with significantly shorter latency, and also developed more tumors per animal on average. When the two groups of tumors were analyzed, there were no significant differences in apoptosis or proliferation (measured by PCNA staining). It was noted that many of the tumors from transgenic mice contained large, irregular nuclei with marked nuclear substructures, suggestive of genomic instability. When cellular DNA contents were measured by FACS, a greater number of the tumors from carcinogen-treated transgenic mice were aneuploid. Genomic instability was also observed in experiments in which the 172 R-H mice were independently crossed with mice carrying WAP-driven des-IGF-1 or TGF-a transgenes or MMTV-neu. In each case, a substantial number of tumors from bigenic mice displayed severe aneuploidy, which was not observed in tumors from mice carrying only the IGF-1, TGF-a, or neu transgene. In addition, the presence of the p53 mutant decreased tumor latency in the des-IGF-1 and neu models, but not in the case of TGF-a. In the latter model, tumor latency is short (approximately 100 days mean for both groups), which may explain why a latency shift cannot be seen.

Gene expression
Some p53 mutants, such as the p53 172 R-L mutant, can still transactivate p21 and inhibit PCNA, gene expression similar to that seen with the wild-type protein. Decreased levels of WAP and á-casein expression are also seen in the mammary glands of mice bearing the 172 R-L transgene. The 172 R-H mutant has lost these properties, in keeping with the common presence of this mutant in breast tumors. The 172 R-H mutant does not bind consensus p53 binding sites found in genes such as p21 and mdm-2, because the mutation results in a conformationally disrupted protein.

Mechanistic implications
Crystallographic solution of the structure of the p53 DNA-binding domain has shown that amino acid 175 (equivalent to murine 172) is located between loops 2 and 3 of the protein. This suggests that the arginine side chain is critical for the maintenance of protein conformation, as it forms hydrogen bonds with elements of both loops. Furthermore, the binding site for the zinc ion critical for protein function is also in the near vicinity, and is disrupted in the mutant protein. The 175 R-H p53 is thought to be a gain-of-function p53 mutant, and not just a functional null, as it appears to affect centrosome duplication and cell growth, among other things, to a greater extent than does the absence of the wild-type protein. As the 175 R-H p53 mutant does not bind consensus binding sites, even when attempts have been made to activate it by relieving C-terminal allosteric inhibition, it appears that its effects may be mediated at the level of protein-protein interactions. Current efforts are aimed at identifying potential interactions and their significance in tumorigenesis. Finally, the fact that this p53 mutation accelerates tumorigenesis caused by carcinogen administration or oncogene coexpression but produces few spontaneous tumors makes this an excellent model in which to examine early events in mammary tumorigenesis.

Submitted by
Kristen Murphy

Department of Cell Biology, M637
Baylor College of Medicine
One Baylor Plaza
Houston, TX 77030
USA

km691540@bcm.tmc.edu

last update: July 1998