Herpes virus type 1 thymidine kinase (HSV1-TK) with ganciclovir (GCV) prodrug treatment is the most widely used approach for suicide gene therapy. This ‘suicide’ strategy allows direct reduction of tumors and clearance of donor cells should graft-versus-host disease (GvHD) arise after bone marrow transplantation. Given recent clinical outcomes, this suicide approach may also provide a key safety component for therapeutic gene transfer vectors that integrate. Although suicide gene therapy using HSV1-TK-encoding oncoretroviral vectors has been evaluated in the clinic, the success of this approach has been relatively modest. Reasons for this include: low gene transfer efficacy, reduced expression of the suicide gene, and insufficient conversion of substrate. Our goal is to overcome these limitations by using a novel lentiviral vector (LV) encoding an alternative kinase/prodrug combination. The rational for our innovative suicide gene therapy strategy is two-fold: 1) Lentiviral vectors can efficiently transduce not only dividing cells but also non-dividing cells. 2) Applying a faster viral enzyme like equine herpes virus type 4 thymidine kinase (EHV4-TK) could be advantageous as it has been shown to be kinetically superior to HSV1-TK at GCV phosphorylation. The aim of this study is to evaluate whether LV-mediated gene modification of target cells with EHV4-TK can lead to efficient killing following GCV treatment. We first constructed a LV expression system carrying the wild-type EHV4-TK cDNA with an IRES element followed by a truncated form of human CD19 (hCD19Δ). Human CD19 was chosen as a cell surface marker to allow functional titering of virus and for immuno-enrichment of transduced cells prior to infusion since it is not expressed in the T cell lineage. The truncated form lacks the intracellular domain and therefore does not signal. Use of an IRES element can abrogate some variegated expression seen with vectors having dual promoters. The LV was pseudotyped with VSV-g and concentrated by ultracentrifugation. After one infection, Jurkat cells (human T cell leukemia) showed a more than 80% functional and stable transduction efficiency (MOI = 10). Using hCD19Δ as a selective marker, transduced Jurkat cells were enriched to over 95% positive by immuno-affinity sorting. EHV4-TK-transduced Jurkat cells exhibited increased cell killing in response to GCV treatment (the apoptotic cell indexes with or without GCV were 69.4 ± 1.5 % and 18.8 ± 1.7 %, respectively; n=3). Highly efficient transduction (more than 60%) of primary human T cells was accomplished by a three time exposure to virus over 36 hours at MOI of 20. Next, we found that GCV efficiently killed transduced primary human T cells in a dose dependent manner. We are now comparing the efficiency of GCV conversion by HSV1-TK and EHV4-TK using LV-transduced cells that express the similar protein levels. We are also evaluating intracellular levels of GCV metabolites by HPLC. These results demonstrate that our novel suicide gene therapy strategy has significant potential for many clinical applications.

Author notes

Corresponding author