368. Lentiviral Vector Strain of Origin and Target Cell Species-Specific Restriction Factors Influence Transduction Efficiency

Abstract

The efficiency of transduction of both human and non-human primate cells by lentiviral vectors is influenced by interactions with members of the TRIM5|[alpha]| family of cytoplasmic proteins. This has been a particular problem when using HIV-1-derived vectors in rhesus macaque hematopoietic cells, where the vectors are strongly restricted by rhesus TRIM5|[alpha]|, although SIV-derived vectors are not. Current models to explain this variation in restriction postulate that the recruitment of cyclophilin A (CypA) by HIV-1 capsid protects the virus from TRIM5|[alpha]| restriction in human cells, but not in macaque cells. Conversely, although the SIVmac capsid does not bind CypA, it is not significantly restricted by human TRIM5|[alpha]|. We wanted to further investigate the role of the CypA-capsid interaction in the restriction of lentiviral vectors by transducing various cell types with VSV-G-pseudotyped lentiviral vectors expressing the enhanced green fluorescent protein (eGFP) gene and then analyzing gene expression by FACS. As expected, we found that treatment of human embryonic kidney cells (HEK293A) cells with cyclosporine A (CsA), which inhibits the CypA-capsid interaction, resulted in a |[sim]|2-fold drop in transduction with the HIV-1-based vector. Inclusion of either the HIV-2 or SIV nef proteins in the vector did not relieve its sensitivity to CsA. In contrast, transduction with HIV-1 vectors containing certain capsid substitutions within the CypA-binding site (Chatterji et al., JBC;280(48):40293|[ndash]|40300) was not negatively affected by CsA treatment, suggesting that the mutated vectors no longer require CypA for efficient transduction of human cells. These vectors also transduced owl monkey and rhesus macaque kidney cells more efficiently than the wild-type vector. Vectors derived from different SIVmac-strains were found to transduce rhesus macaque cell lines at equal efficiency, but transduction of primary rhesus CD34+ cells with an SIVmac1A11-derived vector (Hanawa et al., Blood;103:4062|[ndash]|4069) was significantly more efficient (30|[ndash]|60 %) than with an SIVmac251-derived vector (4|[ndash]|12%) (N|[egrave]|gre et al., Gene Therapy;7:1613|[ndash]|1623). Experiments in which the SIVmac transfer vectors were crosspackaged with capsids from the other SIVmac strains revealed that this property maps to the transfer vector and not to the packaging construct. CsA treatment increased transduction with SIVmac-derived vectors in both rhesus and owl monkey cell lines and in primary human CD34+ cells, suggesting that the SIVmac capsid is susceptible to restriction in these tissues. In contrast, CsA treatment had no effect on transduction with either HIV-1 or SIVmac-derived vectors in primary rhesus macaque CD34+ cells. Together, these results show differential restriction of lentiviral vectors that varies with the species and cell type transduced. These findings may be relevant for studies using lentiviral gene transfer into non-human primate models.