1. ¹Molecular Medicine Program, Mayo Clinic College of Medicine, Rochester, MN
2. ²Department of Immunology, Mayo Clinic College of Medicine, Rochester, MN
3. ³Department of Neurology, Massachusetts General Hospital, Charlestown, MA

Abstract

Truly systemic gene transfer in vivo has been limited by the ability of vectors to localize to the tumor as well as immune inactivation, non-specific adhesion and loss of particles in the circulation. Toxicity can also result from an immune response to the systemic virus and transduction of other cells. We have previously shown that retroviral vectors can be carried to human xenografts by tumor targeted T cells. In immunodeficient mice we observed regression of disseminated tumors without evidence of toxicity. To address a more clinically relevant model, we have now used tumor specific T cells to deliver retroviral vectors to tumors in immunocompetent mice. We used T cells from OT-1 TCR transgenic mice, whose T cell receptor recognizes the SIINFEKL peptide epitope from chicken ovalbumin, ova, presented in the context of H2Db by B16 cells that stably express ova (B16ova). We have previously shown that the adoptive transfer of OT-1 T cells in vivo results in the partial regression of B16ova tumors. We have developed two methods to engineer OT-1 T cells to deliver retroviral vectors. In the first method, OT-1 T cells were generated to produce retrovirus by transducing them with an HSV-derived amplicon vector encoding the gag, pol and env genes, and a retroviral vector encoding the beta-galactosidase gene. In the second method, OT-1 T cells were incubated with retroviruses that can adsorb to the cell surface whilst entering the cell at very low efficiency. We have demonstrated that OT-1 T cells that have retrovirus adsorbed to their surface will hand this virus off to target cells in culture. For example, 5 |[times]| 105 OT-1 T cells loaded with retroviral stocks at an MOI of about 100, washed and then co-cultured with B16 cells resulted in transduction of target B16 cells at a level of 2.7 |[times]| 103 c.f.u. T cell hand off-mediated target cell infection is strictly dependent upon the presence of a functional envelope and is not the result of carry over of virus in the media. Handoff is more efficient when the viral envelope utilizes a receptor that is poorly, or not, expressed on the T cell itself. OT-1 T cells coated with retrovirus can both kill and transfer virus into B16ova cells. This transfer does not entirely depend upon the presence of an intact envelope on the viral particle surface.

To confirm that hand off of virus could occur in vivo, we coated OT-1 T cells with a retrovirus encoding the HSVtk gene under the control of a melanoma specific promoter. These cells were injected intravenously into mice bearing B16ova lung metastases and ganciclovir was given. OT-1 T cells coated with retrovirus increased the survival of mice bearing B16ova lung metastases compared to OT-1 T cells alone or intravenous administration of cell free retrovirus stocks.

In summary, we have shown that it is possible to combine the natural homing and effector functions of T cells with the delivery and transfer of gene therapy vectors to tumors.