A Novel Suicide Gene Therapy Approach for Reduction of GvHD Using Lentiviral Delivery of a Modified Human Thymidylate Monophosphate Kinase.

The infusion of donor lymphocytes in allogenic BMT recipients provides potent antitumor activity to treat recurrent malignancies. One complication, however, is severe GvHD, which is mediated by T cells in the graft. One approach to control GvHD is to employ ‘suicide’ gene therapy. In the customary adaptation of this approach, the herpes simplex virus type 1 thymidine kinase (HSV1-tk) gene, combined with the antiviral prodrug ganciclovir (GCV), is used to control GvHD after introduction of this suicide gene into donor T lymphocytes. However, the efficiency of HSV1-tk is suboptimal and the issue of host immunogenicity against this heterologous effector gene product can hamper outcomes. In addition, prophylactic GCV is often used to control cytomegalovirus infection after BMT. This confounds the broad clinical implementation of this approach. Our novel suicide gene therapy strategy combines the use of human thymidylate monophosphate kinase (tmpk) and the prodrug Zidovudine (AZT) in a lentiviral vector (LV) format. Since tmpk is endogenously expressed in human cells, immunogenic responses will be limited. We constructed a LV expression system carrying wild-type or one of two modified forms of tmpk. These engineered tmpk mutants (F105Y and BL) show substantially increased catalytic conversion of AZT compared to wild-type tmpk. Our vector also includes a truncated form of human CD19 (hCD19Δ), not normally expressed on the T cell lineage, that can be used to enrich and track transduced cells. Highly efficient (95%) transduction of Jurkat cells (human T cell leukemia line) was attained by a single infection with our LVs (MOI of 10). Transduced cells were efficiently and selectively killed in a dose-dependent manner by AZT (IC50 of 2 μM), while wild-type tmpk transduced cells were unaffected by AZT up to 100 μM. In response to AZT treatment, the apoptotic cell indices of cells transduced with wild-type tmpk, F105Y, or BL were 6.2 ± 0.3 %, 40.7 ± 1.7 %, and 46.1 ± 4.6 %, respectively (n=3). We next established by HPLC that cells transduced with a LV encoding a mutant form of tmpk effectively convert AZT into its active anti-metabolite form, AZT-triphosphate (AZT-TP). Intracellular ratio of AZT-TP to AZT-monophosphate (MP) is 11.3 in cells transduced with a LV encoding the BL mutant of tmpk, compared to 0.02 in non-transduced cells and 0.10 in wild-type tmpk transduced cells. Our findings also revealed that following incubation with indirubin-3′-oxime, which inhibits cellular proliferation, and AZT treatment, transduced cells were successfully killed. Thus the cytotoxic mechanism differs from HSV1-tk mediated cell killing and is independent of cell proliferation. We also succeeded in the infection of primary mouse and human T cells to over 40% and 70% transduction efficiency, respectively. Lastly, we have shown that in vivo growth of tumor cells transduced with these mutant tmpk LVs was totally inhibited by treatment with AZT. These results demonstrate that our novel suicide gene therapy system has significant potential for many clinical applications.