Abstract A16: Different epigenetic mechanisms involved in the regulation of SFRP1 gene in prostate cancer

Abstract

Abstract Among men, prostate cancer (PCa)[1] is the second most common cancer and the fifth cause of death worldwide [2]. Androgens play an important role in the development of the disease. For advanced PCa the standard therapy is androgen depletion. However, after 2 or 3 years a high percentage of patients become resistant to this therapy and castration resistant prostate cancer (CRPC) is developed. There is no successful treatment for CRPC, leading into death [3, 4]. Wingless pathway (WNT) is aberrantly activated in several cancer types and in PCa it is involved in AR activity modulation, promoting cancer development [5]. Secreted Frizzled Related Protein 1 (SFRP1) is commonly known as a WNT inhibitor and generally proposed as a tumor suppressor gene. SFRP1 can exert its tumor suppressor activity by inhibiting the AR transcriptional activity [6]. Usually in cancer, SFRP1 mRNA is down regulated and this is associated to increased DNA methylation at its promoter region [7]. In contrast, in PCa there are different reports suggesting SFRP1 repression in absence of DNA methylation [8]. In this work we aimed to investigate two different epigenetic mechanisms for SFRP1 silencing in PCa. For the in vitro model, we used normal prostate cells RWPE-1 and PCa cell lines: LNCaP, 22RV1 and PC3. We confirmed the correlation between mRNA expression loss and the increased DNA methylation of SFRP1 only in PC3 cells, but not in LNCaP and 22RV1, were instead of DNA methylation we found enrichment of H3K27me3, the Polycomb Repressive Complex 2 (PRC2) mark. To correlate this findings, we analyzed the expression of SFRP1 in fresh frozen tissue samples, 11 primary PCa, 10 Benign Prostatic Hyperplasia (BPH) and 4 normal prostates (NP). We found SFRP1 downregulation in all PCa and BPH samples compared to Normal Prostate, but DNA methylation was increased only in PCa but not in BPH, this suggests a novel repressive mechanism for SFRP1 independent of DNA methylation. To see if AR has a role in regulating mRNA SFRP1 expression, we silenced AR with a siRNA, but no SFRP1 expression levels were modified, so we concluded AR was not directly involved in SFRP1 regulation. All these data together suggest different repression mechanisms for SFRP1, one depending on DNA methylation and another one depending on H3K27me3 PRC2 repressive mechanism, Androgen receptor is not directly involved in SFRP1 downregulation, so other factors might be involved in SFRP1 expression loss in PCa.