Israel Medical Association Journal

Background: The humanized SCID mouse model is an attractive tool for testing gene therapy to combat human immunodeficiency virus infection in vivo.

Objectives: To devise a more specific gene therapy directed against HIV, replacing the formerly used interferon with either soluble CD4 molecule immunoadhesin (sCD4-IgG) and/or anti-gp41 monoclonal antibody (2F5), or negative transdominants (Tat, Rev).

Methods: Human monocytoid cell line (U937) was transfected with IFNa[1], b or g genes. 3T3 murine fibroblastic cell line was transfected with sCD4-IgG or 2F5, or both genes, and a human T4 cell line (CEM) was grafted to SCID mice. Negative transdominant genes (Tat, Rev or both) were also transduced in CEM T cell line. Animals were then challenged with HIV-1[2]. Viral load was followed.

Results: IFNa or b were potent anti-HIV, reducing viral load in vivo and inhibiting reverse transcriptase activity in human-removed cells from animals. sCD4-IgG immunoadhesin and gp41 monoclonal antibody resulted in a dramatic reduction of HIV-1 cellular and plasmatic viral load in humanized SCID mice. The simultaneous introduction of negative Tat and Rev genes resulted in a synergistic inhibition of HIV-1 replication in vivo.

Conclusions: Despite the marked reduction of HIV-1 propagation by IFN genes or by negative Tat and Rev transdominants, the gene therapy using soluble CD4 immunoadhesin or anti-gp41 was a more efficient preventive treatment against HIV infection.

Background: The transfer of therapeutic genes into malignant brain tumors has been the subject of intense pre­clinical and clinical research in recent years. Most approaches have used direct intratumoral placement of a variety of vectors and genes, such as retroviruses or adenoviruses carrying drug-susceptibility genes, modified replication-competent herpes virus, and several vectors carrying tumor suppressor genes such as the p53 gene. However, clinical results have so far been disappointing, mainly due to the limited ability to effectively distribute the genetic material into the target cell population. Accordingly, alternative delivery approaches into the central nervous system, e.g., intravascular, are under investigation. Genetic vectors administered intravascularly are unlikely to penetrate the blood-brain barrier and transfer a gene into brain or tumor parenchyma. However, intravascular delivery of vectors may target endothelial cells lining the blood vessels of the brain. Since endothetial cells participate in a variety of physiological and pathological processes in the brain, their modulation by gene transfer may be used for a variety of therapeutic purposes. Angiogenically stimulated endothelial cells within tumors replicate rapidly and hence may become targets for retroviral-mediated gene transfer.

Objective: To assess the anti-tumor effect of transferring a drug-susceptibility gene into endothelial cells of the tumor vasculature.

Methods: As a model for this approach we delivered concentrated retroviral vectors carrying a drug-susceptibility gene via the internal carotid artery of rats with malignant brain tumors. The safety and efficacy of this approach, without and with subsequent treatment with a pro-drug (ganciclovir). was evaluated.

Results: No acute or long-term toxicity was observed after intraarterial infusion of the vector. Treatment with ganciclovir resulted in variable hemorrhagic necrosis of tumors, indicating preferential transduction of the angiogenically stimulated tumor vasculature. This was accompanied by severe toxicity caused by subarachnoid hemorrhage and intracerebral hemorrhage in vascular territories shared by the tumor and adjacent brain.

Conclusion: The data indicate that endothelial cells can be targeted by intraarterial delivery of retroviral vectors and can be used for devising new gene therapy strategies for the treatment of brain tumors.