Differential requirements for ß-catenin in murine prostate cancer originating from basal versus luminal cells.

A driver mutation occurring in different cells of origin may impact cancer progression differently. Previously, we demonstrated higher invasiveness in Pten-deficient prostate cancer (CaP) arising from basal cells compared to that arising from luminal cells in mice. Here, we show higher expression of epithelial-mesenchymal transition (EMT)-inducing transcription factors and stem/progenitor properties in basal-derived CaP compared to luminal-derived CaP. We further explore the requirement for ß-catenin in basal and luminal prostate cells during CaP progression. Genetic ablation and pharmacological inhibition of ß-catenin specifically suppress basal-derived CaP progression through reduction of stemness and cell proliferation and increased γH2Ax-associated apoptosis. Lineage tracing revealed that loss of ß-catenin in basal cells impairs basal-to-luminal differentiation; conversely, ß-catenin loss is dispensable for luminal-derived CaP progression. Our findings suggest that ß-catenin is required for basal-derived normal luminal cells and cancer cells, but not for luminal derivatives. Although the cellular origin of CaP in patients cannot be easily determined at present, the results imply that ß-catenin inhibition is a potential therapeutic option for a subset of patients with basal-derived CaP.

Conditionally ablated Pten in prostate basal cells promotes basal-to-luminal differentiation and causes invasive prostate cancer in mice.

Prostate glands comprise two major epithelial cell types: luminal and basal. Luminal cells have long been considered the cellular origin of prostate cancer (CaP). However, recent evidence from a prostate regeneration assay suggests that prostate basal cells can also give rise to CaP. Here, we characterize Pten-deficient prostate lesions arising from keratin 5-expressing basal cells in a temporally controlled system in mice. Pten-deficient prostate lesions arising from basal cells exhibited luminal phenotypes with higher invasiveness, and the cell fate of Pten-deficient basal cells was traced to neoplastic luminal cells. After temporally ablating Pten in keratin 8-expressing luminal cells, luminal-derived Pten-deficient prostate tumors exhibited slower disease progression, compared with basal-derived tumors, within 13 weeks after Pten ablation. Cellular proliferation was significantly increased in basal-derived versus luminal-derived Pten-deficient prostate lesions. Increased tumor invasion into the smooth muscle layer and aberrantly regulated aggressive signatures (Smad4 and Spp1) were identified exclusively in basal-derived Pten-deficient lesions. Interestingly, p63-expressing cells, which represent basal stem and progenitor cells of basal-derived Pten-deficient prostate lesions, were significantly increased, relative to cells of the luminal-derived prostate lesion. Furthermore, castration did not suppress cellular proliferation of either basal-derived or luminal-derived Pten-deficient prostate tumors. Taken together, our data suggest that, although prostate malignancy can originate from both basal and luminal populations, these two populations differ in aggressive potential.

ß-catenin activation drives thymoma initiation and progression in mice.

Thymoma is the most commonly identified cancer in the anterior mediastinum. To date, the causal mechanism that drives thymoma progression is not clear. Here, we generated K5-∆N64Ctnnb1/ERT2 transgenic mice, which express an N-terminal deletion mutant of ß-catenin fused to a mutated ligand-binding domain of estrogen receptor (ERT2) under the control of the bovine cytokeratin 5 (K5) promoter. The transgenic mouse lines named Tg1 and Tg4 were characterized. Forced expression of ∆N64Ctnnb1/ERT2 in the Tg1 and Tg4 mice developed small thymoma lesions in response to tamoxifen treatment. In the absence of tamoxifen, the Tg1 mice exhibited leaky activation of ß-catenin, which activated the TOP-Gal transgene and Wnt/ß-catenin-targeted genes. As the Tg1 mice aged in the absence of tamoxifen, manifest thymomas were found at 10-12 months. Interestingly, we detected loss of AIRE and increase of p63 in the thymomas of Tg1 mice, similar to that observed in human thymomas. Moreover, the ß5t protease subunit, which was reported as a differential marker for human type B3 thymoma, was expressed in the Tg1 thymomas. Thus, the Tg1 mice generated in this study accurately mimic the characteristics of human thymomas and may serve as a model for understanding thymoma pathogenesis.

Tumor spectrum, tumor latency and tumor incidence of the Pten-deficient mice.

BACKGROUND: Pten functionally acts as a tumor suppressor gene. Lately, tissue-specific ablation of Pten gene in mice has elucidated the role of Pten in different tumor progression models. However, a temporally controlled Pten loss in all adult tissues to examine susceptibility of various tissues to Pten-deficient tumorigenesis has not been addressed yet. Our goal was to explore the genesis of Pten-deficient malignancies in multiple tissue lineages of the adult mouse. METHODS AND FINDINGS: We utilized an inducible Cre/loxP system to delete Pten exon 5 in the systemic organs of ROSA26 (R26)-CreER(T);Pten(fx/fx) mice. On reaching 45 weeks 4OHT-induced Pten loss, we found that the R26-CreER(T);Pten(fx/fx) mice developed a variety of malignancies. Overall tumor mean latency was 17 weeks in the Pten-deficient mice. Interestingly, mutant females developed malignancies more quickly at 10 approximately 11 weeks compared with a tumor latency of 21 weeks for mutant males. Lymphoma incidence (76.9% in females; 40.0% in males) was higher than the other malignancies found in the mutant mice. Mutant males developed prostate (20.0%), intestinal cancer (35.0%) and squamous cell carcinoma (10.0%), whereas the mutant females developed squamous cell carcinoma (15.4%) and endometrial cancer (46.1%) in addition to lymphomas. Furthermore, we tested the pharmacological inhibition of the PTEN downstream effectors using LY294002 on Pten-deficient prostate hyperplasia. Our data revealed that, indeed, the prostate hyperplasia resulting from the induced Pten loss was significantly suppressed by LY294002 (p = 0.007). CONCLUSIONS: Through monitoring a variety of Pten-deficient tumor formation, our results revealed that the lymphoid lineages and the epithelium of the prostate, endometrium, intestine and epidermis are highly susceptible to tumorigenesis after the Pten gene is excised. Therefore, this R26-CreER(T); Pten(fx/fx) mouse model may provide an entry point for understanding the role of Pten in the tumorigenesis of different organs and extend the search for potential therapeutic approaches to prevent Pten-deficient malignancies.