566. Targeted Lentiviral Vector Integration Mediated by the AAV Rep78 Protein

Our goal is to improve the safety of lentiviral vectors by shifting their integration preference to genomic “safe harbor” sites where therapeutic transgenes can integrate and function in a predictable manner. To do this, we have generated integrase-defective LV vectors (IDLVs) capable of inserting transgene sequences at predetermined genomic sites through homologous recombination (HR). To increase the frequency of HR, we employ genome editing approaches involving site-specific nucleases. A major drawback of this approach is safety concerns caused by inadvertent double strand breaks (DSBs) and non-homologous end-joining (NHEJ) events at unintended genomic (off target) sites. To bypass this risk, we are pursuing site-specific nucleases that do not involve DSBs as a potentially safer class of DNA-modifying agents for site-specific integration of transgenes in human cells. The sequence and strand-specific endonucleases Rep78 and Rep68 encoded by adeno-associated virus (AAV) have previously been reported to stimulate HR of plasmids bearing homology arms corresponding to the AAVS1 safe harbor locus on chromosome 19. We tested the integration efficiency of IDLVs containing short homology arms corresponding to the AAVS1 locus mediated by an engineered version of the AAV2-derived Rep78 protein in HEK293 cells. We compared the integration efficiency of IDLVs mediated by the Rep78 protein to that of an AAVS1-specific zinc finger nuclease (ZFN). Furthermore, we explored the potential of IDLV donor vectors for targeted integration mediated by Rep78 in human induced pluripotent stem cells (hiPSCs). The results obtained indicate that in the presence of the Rep78 protein, the integration efficiency of IDLVs was up to 8.3 fold higher compared to cells treated with the donor vector alone. Moreover, the efficiency of site-specific targeting mediated by Rep78 was higher than that of the AAVS1-specific ZFN. We also showed that Rep78 can successfully direct integration of IDLVs at the AAVS1 locus in hiPSCs. We anticipate that the approach used may ultimately contribute to the reduction of risks associated with LV vectors.