Persistent Control of SIV Infection in Rhesus Macaques By Expressing a Highly Potent SIV Decoy Receptor after In Vivo HSC Transduction

Despite the success achieved by antiretroviral HIV-1 therapies, long-term and repeated drug administration is associated with toxicity, virus evasion and high cost. We aim to develop a gene therapy approach for persistent control against HIV-1 infection by delivering the transgene expressing a decoy receptor (eCD4-Ig) to hematopoietic stem cells (HSCs) by in vivo transduction. In this approach, HSCs are mobilized from the bone marrow into the peripheral blood stream and transduced with intravenously injected virus vectors. We use an integrating, helper-dependent adenovirus (HDAd5/35++) vector system that targets human CD46, a receptor that is abundantly expressed on primitive HSCs. Transgene integration is achieved by a hyperactive Sleeping Beauty transposase and transgene marking in peripheral blood cells can be increased by in vivo selection. The efficacy and safety of our in vivo HSC transduction/selection strategy has been previously demonstrated for the treatment of b-hemoglobinopathies, hemophilia A, and cancer in murine disease models. In non-human primates, we showed efficient transgene (g-globin) expression in peripheral red blood cells using this strategy. eCD4-Ig functions like a neutralizing antibody and shows broad neutralization spectrum against HIV-1, HIV-2 and SIV isolates. We have designed and produced an HDAd5/35++ vector expressing rhesus eCD4-Ig (rh-eCD4-Ig) from a constitutive and highly active EF1a promoter. In CD46-transgenic mice, over 50µg/mL eCD4-Ig in serum was measured after in vivo transduction and selection, with no obvious adverse events observed. Neutralization assay with serum samples showed that the produced eCD4-Ig effectively inhibited HIV-1 and SIV infection. We then performed studies in a rhesus macaque. After in vivo HSC transduction/selection of a rhesus macaque, eCD4-Ig serum levels were stable at 20-30 mg/ml. In vitro SIVmac239 neutralization assays using week 13 serum and recombinant eCD4-Ig protein determined that the IC50 of eCD4-Ig in rhesus serum is 1.0 mg/ml. This implies that the serum eCD4-Ig concentration at the time of SIVmac239 challenge was ~25-fold higher than the IC50. High-level eCD4-Ig expression had no clinical or hematological side effects. The first SIVmac239 challenge (20pg) was given on June 22 ^nd. Increasing rechallenge doses are currently injected monthly. So far, the viral load measured by quantitative RT-PCR is below detection limit. The animal is without symptoms and has normal lymphocyte/subset counts. Our study demonstrates an in vivo HSC transduction approach for potential long-term control of HIV-1 infection.