|dc.description.abstract||Prostate cancer (PCa) progression is dependent on transcriptional activation of the androgen receptor (AR) in the majority of cases. Therefore, therapies include anti-androgens, which decrease levels of circulating androgens and inhibit AR activity. However, resistance inevitably develops, resulting in more aggressive, incurable late-stage disease with intact or constitutively active AR signalling. There is currently no cure for castration-resistant metastatic PCa and novel therapies are needed. In this work I aimed to develop a non-replicating adenovirus for delivery of a potent prodrug-converting enzyme expressed at high levels to specifically target and kill PCa cells.
TMPRSS2 is a major AR-regulated gene, commonly expressed at higher levels in cancer than normal prostate. In 40-70% of PCa cases the AR-regulated TMPRSS2 promoter is fused to oncogenic ERG. Based on this, we generated TMPRSS2 promoter and enhancer constructs, utilising promoter regions upstream of Exon1 and Exon2 of TMPRSS2 to drive the prodrug-converting chimeric enzyme cytosine deaminase uracil phosphoribosyl transferase (CD/UPRT) to investigate the therapeutic efficacy in various PCa models.
Prior work in the lab evaluated the selectivity and transcriptional efficiency of the various promoter and enhancer regions upstream of Exon1 and Exon2 in TMPRSS2 were investigated by Luciferase (Luc) expression. To further improve on promoter activity and transcription, the TMPRSS2 constructs were inserted into an expression cassette derived from the versatile expression vector VISA (VP16-GAL4-WPRE integrated systemic amplifier). When the Luc-gene was replaced with CD/UPRT in the VISA vector, specific and dose-dependent cell killing was observed in 22RV1 (AR+) cells when the non-toxic prodrug 5-flourocytosine (5-FC) was administered.
To increase transfection efficiency, I inserted the TMPRSS2-VISA-CD/UPRT expression cassette into a non-replicating adenovirus (Ad5-TV-CU) replacing the E1-genes. Ad5-TV-CU was characterized and protein expression of the CD/UPRT gene was detected at high levels post infection in AR-expressing 22RV1, LNCaP sublines, and VCaP cells, whilst remaining inactive in DU145 and HEK293 cells (AR-negative). Dose-dependent decreases in EC50-values were observed upon infection with Ad5-TV-CU in combination with low doses of 5-FC in a number of cell lines, including 22RV1, LNCaP-104-S, LNCaP-CDXR3 and LNCaP-104R1 cells, whilst demonstrating no cell killing at any concentration in PC3, PNT1a, PNT2 and PrEC cells (AR-).
To determine efficacy of the viral vector in vivo, I explored various AR-expressing PCa cell lines grown as xenografts in immunodeficient murine models, including 22RV1 and the LNCaP sublines LNCaP-104-S and LNCaP-CDXR3. However, establishment of the in vivo models was unsuccessful due to variable tumour take and growth (LNCaP) or tumours grew fast and studies had to be terminated prior to efficacy determination (22RV1).
Further comparison of the Exon1 TMPRSS2 promoter with the traditional PSA promoter and the chimeric PSA promoter/enhancer demonstrated the Exon1 TMPRSS2 construct to be superior by inducing 3.5-fold and 2.3-fold higher levels of luciferase expression than the PSA constructs, respectively. Investigation of both Exon1 and Exon2 TMPRSS2 promoters revealed that fusion of the two regions to form the chimeric Exon1/Exon2 promoter, drove higher levels of prostate specific expression, that have the potential to drive higher level expression than the SV40 promoter, this higher expression could improve efficacy of the current non-replicating Ad5-TV-CU.
In summary, Ad5-TV-CU has demonstrated efficacy in a number of AR expressing cell lines, with limited results in vivo. Further studies that incorporate the optimal LW chimeric promoter into a replicating adenoviral vector could dramatically improve the efficacy of this virus for future studies. Additionally the discovery of a more suitable in vivo model will help to establish the true therapeutic potential of the new optimal Ad5-TV-CU virus.||en_US