Metastasis is a complex biological process for which there is minimal understanding at the molecular and cellular level. Although poorly understood, metastasis is often described as the most significant step in malignant progression as cancer mortality in general results directly from the growth of disseminated metastatic cells. This profound impact of metastatic disease is clearly the case in prostate cancer which will be diagnosed in 200,000 US males in 1994 and cause mortality in 38,000 US men. Recently, we have modified the mouse prostate reconstitution (MPR) model system to use urogenital sinus tissue derived from p53 """"""""knock-out"""""""" mice as target for the introduction of the ras and myc oncogenes. Using this in vivo model system, we have studied the tumorigenic effects of ras+myc initiation in a background of wild-type p53, heterozygous p53 mutants or homozygous p53 mutants. We demonstrated that ras+myc-initiated wild-type 129/Sv mice produce predominantly hyperplasia with low frequency (approximately 10%) of focal nonmetastatic cancer. In contrast, both heterozygous p53 as well as homozygous p53 MPRs produce metastatic carcinoma in nearly 100% of the cases. Progress of ras+myc + heterozygote p53 """"""""knock-out"""""""" MPRs correlated with either complete loss, partial loss or loss of expression of the wild-type allele. Overall, our data clearly indicate that complete loss of p53 function leads to metastasis in this experimental mouse prostate cancer model. Interestingly, the pattern of metastasis closely mimics that seen in human prostate including widespread metastasis to the lung, mesenteric lymph nodes, bone and occasionally liver. We have extended this model to facilitate mechanistic studies at the molecular and cellular level, by generating a series of cell lines derived from both primary site carcinomas as well as metastatic disease from different organ sites. We have thus far demonstrated that TGFbeta1 inhibited growth in vitro in all (n=6) primary site tumor-derived cell lines (mean inhibition of 45+/- 9% after 8 days) as well as control urogenital sinus epithelial (CUGE) cells but only inhibited 3 out of 6 cell lines derived from lung metastases (mean inhibition in sensitive lines 20 +/- 1%). Similar differential responses to TGF-beta1 were also observed for colony formation and motility assays. The results of crosslinking studies for TGF-beta1 which demonstrated specific TGF-beta1 binding for type l, 2 and 3 TGF-beta receptors in all primary site tumor- and metastasis-derived cell lines suggest that loss of these TGF-beta1 induced responses in metastasis-derived cell lines is downstream of receptor binding. We propose to use this novel prostate cancer metastasis model system to investigate the genetic mechanisms which underlie the loss of TGFbeta1 response and its relationship to metastatic progression. In addition, specific regions of homozygous deletion associated with metastatic progression will be identified and cloned using representational difference analysis and regions of loss of heterozygosity that occur during metastasis in vivo will be determined using microsatellite probes. From past studies using the MPR model system we can predict that the results of these studies will also be relevant to the human disease.
Showing the most recent 10 out of 40 publications
