Transduction of malignant cells with toxin genes provides a novel means to promote tumor cell destruction. The efficacy of a toxin gene is dependent on the cell type targeted, the quantity of exogenous protein synthesized, and the mechanisms of growth inhibition and bystander killing. To develop gene therapy for targeting metastatic lung adenocarcinoma, the toxic activity of herpes simplex virus type 1-thymidine kinase, Escherichia coli cytosine deaminase, and human deoxycytidine kinase were investigated in metastatic human lung adenocarcinoma cell lines H1437 and H2122. Cells were transduced stably with retroviral vectors containing the toxin gene cDNA under the control of either a strong [cytomegalovirus (CMV) immediate early promoter and enhancer] or an intermediate strength (Moloney murine leukemia virus long terminal repeat) promoter. A comparison of toxin gene efficacy was based on the level of specific enzyme activity, the concentration of prodrug required to inhibit cell growth by 50%, and the magnitude of the bystander effect. In lung adenocarcinoma cell lines, cytosine deaminase, driven by the CMV promoter, was superior to thymidine kinase and deoxycytidine kinase in its ability to achieve high levels of specific enzyme activity, to induce growth inhibition, and to affect neighboring cell growth. Therefore, cytosine deaminase expressed from the CMV promoter seems to be the most promising toxin gene for human lung adenocarcinoma gene therapy.
This work was supported by Glaxo-Wellcome Co. D. K. H. was supported by NIH Training Grant HL51818-02.