Targeted Suicide Gene Therapy with Retroviral Replicating Vectors for Experimental Canine Cancers

Abstract

Cancer in dogs has increased in recent years and is a leading cause of death. We have developed a retroviral replicating vector (RRV) that specifically targets cancer cells for infection and replication. RRV carrying a suicide gene induced synchronized killing of cancer cells when administered with a prodrug after infection. In this study, we evaluated two distinct RRVs derived from amphotropic murine leukemia virus (AMLV) and gibbon ape leukemia virus (GALV) in canine tumor models both in vitro and in vivo. Despite low infection rates in normal canine cells, both RRVs efficiently infected and replicated within all the canine tumor cells tested. The efficient intratumoral spread of the RRVs after their intratumoral injection was also demonstrated in nude mouse models of subcutaneous canine tumor xenografts. When both RRVs encoded a yeast cytosine deaminase suicide gene, which converts the prodrug 5-fluorocytosine (5-FC) to the active drug 5-fluorouracil, they caused tumor-cell-specific 5-FC-induced killing of the canine tumor cells in vitro. Furthermore, in the AZACF- and AZACH-cell subcutaneous tumor xenograft models, both RRVs exerted significant antitumor effects. These results suggest that RRV-mediated suicide gene therapy is a novel therapeutic approach to canine cancers.

Keywords: canine cancer; retroviral replicating vectors; suicide gene therapy.

Figure 1

Schematic structure of RRVs. Each RRV contains the full-length replication-competent AMLV or GALV provirus. An internal ribosomal entry site (IRES) and gene of interest (GFP, CD, or luciferase) were inserted between the env gene and the 3′-untranslated region. CMV: cytomegalovirus-derived promoter; LTR: long terminal repeat sequence, ψ: packaging signal; gag/pol/env: retroviral structural genes; GFP: green fluorescent protein; CD: yeast cytosine deaminase, luc: firefly luciferase.

Figure 2

Replication kinetics of RRVs in canine cells. Normal canine cells (fibroblasts and primary liver epithelial cells) and canine tumor cell lines (AZACF, AZACH, AZACL1, AZACL2, AZACB, AZACU, oSCC-1, oSCC-4, TSCCLN#1, and TSCCLN#4) were infected with GFP-expressing RRV (AMLV or GALV) at a multiplicity of infection of 0.01. Cell growth was maintained and the percentage of cells expressing GFP was analyzed with a flow cytometer at every passage. Representative data from three independent experiments are shown.

Figure 3

In vivo replicative spread of RRV in canine tumor xenograft models. Nude mice were inoculated under the back skin with AZACF, AZACH, and AZACL2 cells to create subcutaneous tumor models. Luc-expressing RRVs (AMLV or GALV) were administered intratumorally and observed with the IVIS imaging system from day 3 to day 28. A replication-defective lentiviral vector was used to construct the controls. (a) In vivo fluorescence imaging. Representative images are shown for each group. The scale indicates the numerical value of the relative luminescence units on a CCD camera. (b) Luminescent signal intensities in the tumors were quantified with the ROI tool in the Living Image software.

Figure 4

Suicide-gene-mediated cell-killing effects in canine cells. Fibroblasts, liver epithelial cells, and AZACF, AZACH, AZACL2, oSCC-4, and TSCCLN#4 cells were infected with CD-expressing RRVs (AMLV or GALV) at an MOI of 0.01. On day 15, the cells were incubated with various concentrations of 5-FC for 6 days, and then cell viability was evaluated with an alamarBlue assay. Data shown are means ± SEs of triplicate experiments.

Figure 5

Suicide-gene-mediated antitumor effects in a canine tumor xenograft model. The backs of nude mice were subcutaneously inoculated with AZACF or AZACH cells, and CD-expressing RRVs (or PBS for the control group) were administered intratumorally when the tumor diameter reached 5 mm. 5-FC (500 µg/g bodyweight/day) was then administered intraperitoneally three times a week, commencing when the tumor volume exceeded 100 mm³ (AZACF: from day 3; AZACH: from day 14). Tumor volume was measured three times a week. Data shown are the means ± SEs of triplicate experiments. * p < 0.05, ** p < 0.01.