The role of the oncogene ETV4 in the regulation of genes involved in prostate cancer pathogenesis

Prostate cancer is the second leading cause of death in the male population. In most prostate cancers, ETS proteins (ERG, ETV1, ETV4) are over-expressed because of recurrent chromosomal translocations that bring an ETS gene under the control of a promoter of a gene highly expressed in the prostate. The role of ERG and ETV1 in determining PC has been widely studied, but only little information is available about the tumorigenic role of ETV4. It has been previously demonstrated that ETV4 over-expression promotes neoplastic features in human prostate cell lines. In particular, ETV4 increases migration, invasion, proliferation, cell cycle progression, anchorage-independent growth, epithelial-mesenchymal transition and tumor growth in xenograft model. Moreover, ETV4 over-expression in prostate cell lines was associated with the deregulation of cell cycle progression and reduction of p21 and p27. In keeping, in two independent transgenic mouse lines that we have engineered to express human ETV4 in a prostate-specific manner, we have observed that half of 10 months old transgenic mice develop low-grade prostatic intraepithelial neoplasia (PIN) lesions. The first aim of this PhD work was to evaluate also in vivo the expression levels of p21 and p27 and to investigate the mechanisms through which ETV4 regulates p21 and p27. The proportion of Ki67+ prostate cells was significantly increased in 5 months-old ETV4-transgenic mice compared to wild type mice, confirming that also in vivo ETV4 is associated with increased proliferation. Moreover, in these mice cdkn1a (p21) was reduced at both mRNA and protein levels whereas p27 was reduced only at protein level. To determine how ETV4 regulates the expression of CDKN1A in human prostate cells, we performed ChIP and luciferase assays. In the 2 kb upstream the transcriptional start site of CDKN1A gene we identified 2 putative ETV4 binding sites (BS): distal (-1409/-1403) and proximal (-704/-696) BS. ChIP assay suggested that ETV4 binds the proximal but not the distal BS of CDKN1A promoter. The Luciferase assay performed in several cell lines using a vector, in which the firefly luciferase was expressed under the control of the WT proximal ETV4 BS, showed that ETV4 over-expression induces a reduction of the relative luciferase expression, while the normal level is restored when this site is either mutated or deleted. In addition, we found that ETV4 exerts also an indirect regulation of p21 expression through the downregulation of the protein level of p53. In order to further understand the molecular mechanisms by which ETV4 promotes prostate tumorigenesis, we compared, by micro-array analysis, the expression profile of prostate tissue of ETV4 mice with that of WT mice at 5 months of age. The second aim of this PhD work was to investigate, after validation, some of the genes found differentially expressed in ETV4 mice. In particular, we focused on the secretory leukocyte protease inhibitor (SLPI). SLPI is a serine protease that protects host tissues from the excessive damage by proteolytic enzymes released during inflammation and it has been found over-expressed in a variety of cancers. However, it is paradoxically reduced in localized PC, but it may increase during progression in patients resistant to androgen deprivation therapy. It is intriguing, that this is consistent with the reduced SLPI expression we found in our ETV4 mouse model of early PC. In order to demonstrate that ETV4 is responsible for SLPI downregulation we have studied two human prostate cell lines: (i) the PC3 prostate cancer cell line that express high levels of ETV4 and (ii) RWPE normal prostate cell line that does not express ETV4. In these mirror cellular models we found that ETV4 silencing in PC3 cells results in increased levels of SLPI mRNA and protein whereas their levels were reduced upon the overexpression of ETV4 in RWPE cells. Moreover, ETV4 reduces luciferase expression driven by the SLPI promoter in different cell lines but ChIP assay have not been able to demonstrate the direct binding of ETV4 to the SLPI promoter. In addition, we investigated ETV1, another ETS protein of Pea3 subfamily, and we found that also ETV1 reduces SLPI expression. In summary, this PhD work has clarified some of the oncogenic mechanisms of ETV4 in the prostate cancer. 1) ETV4 over-expression increases the proliferation rate of prostate cells leading to the development of Prostate Intraepithelial Neoplasia. This increased proliferation rate is associated with the down regulation of p21, that results from both the binding of ETV4 to the p21 promoter and from the ETV4-mediated reduction of p53. 2) ETV4 downregulates the expression of SLPI in mice and in human prostate cell lines. This role is likely shared with other ETS transcriptional factors, such as ETV1, as we demonstrated in human prostate cell lines; however, this regulation is not exerted by the direct regulation of the SLPI promoter. This negative regulation of SLPI exerted by ETS proteins could explain the clinical finding of the reduced levels of SLPI during the early phase of prostate cancer.