6. Retargeted Foamy Virus Vectors Integrate Less Frequently Near Proto-Oncogenes

Abstract

Hematopoietic stem cell gene therapy offers immense potential to treat many genetic diseases and has already shown efficacy in clinical trials. However, retroviral vector mediated gene dysregulation, known as genotoxicity, remains a major challenge and clinically relevant approaches to reduce integration near genes and proto-oncogenes are needed. The frequency at which a retroviral vector integrates near proto-oncogenes is a major factor in determining the safety of the vector. Here we describe a clinically relevant method to retarget foamy retroviral vector (FV) integration away from proto-oncogenes using modified Gag and Pol helper constructs. The chromatin binding site (CBS) of FV Gag was altered using a previously described triple alanine substitution mutation of RTY (Gag-RTY) shown to drastically reduce chromatin binding of the FV pre-integration complex. The modified FV Pol construct expresses chromobox protein homolog 1 fused to the C-terminus of integrase (CBX1-IN). CBX1 interacts with tri-methylated lysine 9 of histone H3 (H3K9me3) and is associated with gene sparse regions in the human genome. We hypothesized a modified FV expressing Gag-RTY and CBX1-IN would have an altered integration site profile and be retargeted to H3K9me3 regions and away from genes and proto-oncogenes. FV integration sites from modified and control FVs were compared in normal human fibroblasts and in CD34+ human cord blood cells. We observed retargeting of FV integration into H3K9me3 regions with the modified FV (Figure 1Figure 1). Importantly, retargeting FV integration significantly reduces the number of integration sites near proto-oncogene transcription start sites (Table 1Table 1). Retargeted FVs can be produced at clinically relevant titers (> 107 transducing units / ml), resulting in efficient transgene expression (75 % of control in CD34+ cells) with no evidence of vector silencing. Another published method to retarget lentiviral vector integration required using a modified cell line, which is not practical for clinical use. Our approach is cell line independent and the modified Gag and Pol helper constructs will allow an investigator to simply use these modified helper plasmids during vector production to retarget any therapeutic FV in any target cell. Retargeted FVs integrate less frequently near proto-oncogenes than gammaretroviral vectors and unmodified FVs, and may be the safest current option for hematopoietic stem cell gene therapy.Table 1Retargeted FV integrates further away from genes and proto-oncogenes than control FV.Percent of FV integration sites in human CD34*** cellsVectorTotal sitesIn RefSeq genes< 10 kb from TSS< 50 kb from proto-oncogene TSSControl FV421740.924.54.7Retargeted FV153430.6***12.5***2.2****statistically significant at p < 0.001 compared to control vector. FV, foamy retroviral vector; TSS, transcription start site.View Large Image | Download PowerPoint Slide