123. Helper-Dependent Ad5/35 Vectors for ZFN Mediated Gene Editing in Hematopoietic Stem Cells

Abstract

A homozygous [Delta]32 deletion in the CCR5 gene, found in about 1% of Caucasians, confers a natural resistance to HIV-1. In a recent pivotal study it was shown that transplantation of hematopoietic stem/progenitor cells (HSCs) from a donor who was homozygous for CCR5 [Delta]32 in a patient with acute myeloid leukemia and HIV-1 infection resulted in long-term control of HIV. This finding supports the development of gene therapy approaches to eliminate CCR5 in HIV target cells, including approaches based on CCR5-knockout by zinc finger nucleases (ZFNs). Recent trials involved the ex vivo transduction of patient CD4+ T-cells with a CCR5-ZFN expressing adenovirus Ad5/35 vectors. More recent attempts have focused on CCR5 gene knock-out in hematopoietic stem cells (HSCs). Because HSCs are the source for all blood cell lineages, CCR5 knock-out would protect not only CD4 cells but also all remaining lymphoid and myeloid cell types that are potential targets for HIV infection. We generated a helper-dependent, capsid-modified HD-Ad5/35 vector for CCR5-ZFN expression in HSCs from mobilized adult donors. The production of these vectors required that ZFN expression in HD-Ad5/35 producer 293-Cre cells was suppressed. To do this, we developed a miRNA-based system for regulation of gene expression based on miRNA expression profiling of 293-Cre and CD34+ cells.We demonstrated that, after in vitro CD34+ cell transduction, the HD-Ad5/35.ZFNmiR vector conferred ccr5 knock-out in primitive HSC (i.e. long-term culture initiating cells and NOD/SCID repopulating cells). Upon transplantation of in vitro transduced CD34+ cells into irradiated NOG mice, the ccr5 gene disruption frequency achieved in engrafted HSCs found in the bone marrow was about 12%. However, we also found in transplantation studies that the HD-Ad5/35.ZFNmiR vector decreased CD34+ engraftment rates 7-10-fold compared to HD-Ad5/35-GFP (control) transduced cells. A possible interpretation of these data is that sustained high-level ZFN expression afforded by the HD-Ad5/35 vector platform might be detrimental to the HSC, particularly if the cells do not actively divide allowing the extended persistence of HD-Ad5/35 vector genome. This deficit has not been observed with previous first generation recombinant adenoviral vector delivery of ZFNs to CD34+ HSCs (Li et al. Mol Ther 2013), thus the source of the engraftment inhibition is unclear. However, we plan to investigate and develop methods for maximal ccr5 knock-out efficiency in HCSs for future HIV challenge experiments.