Background: Using in vivo mouse models, we have demonstrated that the insulin promoter-driven suicidal gene therapy (RIPTK) could be used in the treatment of mouse insulinoma and human pancreatic cancer cells. However, limitations of this therapy include tumor cells lack of sufficient PDX-1 protein and low levels of transgene expression mediated by liposome delivery system. The purpose of this study was to determine 1) whether transient transfection of PDX-1 into selected pancreatic cancer cells would lead to increased RIPTK cytotoxicity, and 2) whether an adenoviral delivery system would increase the overall RIPTK gene expression in vitro.
Material and methods: RIPlacZ and RSVlacZ plasmid DNA as well as AdCMVlacZ and AdRIPlacZ were used in transfection assays in human pancreatic cancer cell lines PANC-1 and MIA PaCa2 (n = 8). An expression plasmid DNA containing the mouse PDX-1 cDNA was also used. LacZ reporter assays were performed. RIPTK genes constructed either in plasmid or in adenoviral vectors were used in cytotoxic assays. RT-PCR assays were used to determine PDX-1 expression levels.
Results: PDX-1 protein was detected in the human pancreatic ductal carcinoma cell line PANC-1, a little in MIA PaCa2 cells. Liposome mediated (L) RSVlacZ and RIPlacZ transfection in PANC-1 cells resulted in 10.1% and 9.3% transgene expression, respectively. Co-delivery of PDX-1 had no significant effect on RSVlacZ expression (9.3%, P = NS) but significantly increased RIPlacZ gene expression (14.9% P < 0.05). Adenoviral mediated (Ad) RIPlacZ transgene was highly expressed in PANC-1 cells (66.1%) and the reporter activity was further enhanced when PDX-1 was co-delivered (70.2%, P < 0.05). Liposomal transfection of MIA PaCa2 cells using RSVlacZ and RIPlacZ reporter genes resulted in 9.3% and 1.0% gene expression, respectively. Co-transfection of PDX-1 in these cells resulted in a significant activation of RIPlacZ gene expression (14.5%, P < 0.05) with no effects on RSVlacZ treated cells (9.8%). AdCMVlacZ and AdRIPlacZ significantly increased reporter activities in MIA PaCa2 cells (63.0% and 9.8%, respectively). Transfection of PDX-1 also significantly enhanced the AdRIPlacZ activities (46.0%, P < 0.05), with no significant effect in AdCMVlacZ treated cells (68.2%). The cytotoxic effect of liposome-RIPTK/ganciclovir (GCV) in PANC-1 cells was 18.6% and increased to 22.8% when PDX-1 was co-transfected into the cells (P = NS). MIA PaCa2 cells treated with RIPTK alone resulted in 4.9% cell death and increased to 18.2% when exogenous PDX-1 was co-delivered (P < 0.05). The AdRIPTK gene delivery with GCV treatment caused significant cytotoxic effect in PANC-1 (29.3%) and MIA PaCa2 (12.4%) compared with untreated cells. The cytotoxic effects were further increased to 43.4% and 29.4% in PANC-1 and MIA PaCa2 cells, respectively, when PDX-1 was co-transfected (P < 0.05 for both).
Conclusions: These data demonstrated that adenoviral mediated gene delivery resulted in a significant increase of transgene expression compared with liposomal delivery systems. RIPTK mediated cytotoxicity was also significantly enhanced via co-delivery of exogenous PDX-1 in these cells. Thus, these results also indicated that PDX-1 plays critical roles in insulin promoter activation and demonstrated that PDX-1 production is essential for insulin promoter-directed gene therapy.