Mesenchymal Stem Cells as Suicide Gene Therapy Vehicles for Organ-Confined and Metastatic Prostate Cancer (PCa).

The efficacy of mesenchymal stem cells (MSC) is currently being examined as a clinical regenerative medicine for multiple sclerosis, cirrhosis, liver disease, tibial fractures, heart failure and graft versus host disease. MSC display an inherent tumor-tropic property that has been exploited for the targeted delivery of therapeutic genes to metastatic melanoma, glioma, breast and colon carcinoma in animal models. Advantages of using MSC include their ability for: self-renewal, ease of propagation and gene modification ex vivo; secretion of high levels of therapeutic protein; evasion of immune rejection; differentiation into connective tissue and tumor stroma; and long-term engraftment in vivo. This study explores the utility of MSC to deliver reporter or suicide genes to advanced Pca which is currently incurable using the syngeneic RM1 mouse model. Sca-1^bright CD45⁻ cells sorted from adherent bone marrow cells were shown to be true MSC by their ability to undergo tri-lineage differentiation in adipogeneic, osteogenic and chondrogenic media and their characteristic CD44⁺, CD90⁺, CD73⁺ and CD106⁺ phenotype. Lentiviral vectors showed sustained green fluorescent protein (GFP) reporter gene expression in MSC (MSC-GFP) for 50 passages by flow cytometry. When MSC-GFP (10e6 cells) were implanted into the mouse prostate with or without RM1 tumor cells on day 0, examination by full body fluorescence imaging (IVIS 100; Xenogen/Caliper) showed MSC persisted only within the tumor-bearing prostate (p<0.05; day 18). No MSC were detected in any other organ examined. To test their systemic homing ability, MSC-GFP (10e6 cells) were infused every second day (2–14) via the tail vein of mice in the presence or absence of RM1 lung pseudometastases. MSC persisted within the lungs of RM1 tumor-bearing mice alone (p<0.01) with no detectable MSC in any other organ examined (day 18). These results suggest MSC engraft organ-confined Pca and home to metastatic Pca. Gene directed enzyme prodrug therapy (GDEPT) describes the transfer of a suicide gene, not expressed in mammalian cells, into tumors using viral vectors. This renders tumors sensitive to prodrugs that are non-toxic to normal tissues. In our pre-clinical study, prostate tumors established from RM1 tumor cells stably transfected with fusion suicide gene cytosine deaminase/uracil phosphoribosyltransferase (CDUPRT) followed by systemic treatment with prodrug 5-fluorocytosine, showed significant reductions in prostate tumor growth and pseudometastases in the lungs (1). More recently, we stably transfected MSC with CDUPRT prior to implantation into established RM1 prostate tumors. In the presence of prodrug MSC-CDUPRT showed similar levels of Pca killing observed in the published experiment. In both experiments CDUPRT in the presence of prodrug significantly reduced (∼75%; p<0.05) the weight of RM1 prostate tumors compared to the control gene or no prodrug control mice. These results demonstrate that MSC can deliver suicide genes to developing Pca and efficiently convert prodrug to a toxic diffusible metabolite to suppress tumor growth. MSC implantation was well tolerated without observed toxicity and therefore show promise as a novel form of cell-directed suicide gene therapy.