300. Targeted Gene Correction in Mouse Embryonic Stem Cells Using Helper-Dependent Adenoviral Vectors

Abstract

Integration of a therapeutic gene into random sites on host chromosomes could potentially lead to serious problems such as cellular transformation and gene silencing. Gene targeting, the process of homologous recombination between exogenous and chromosomal DNA, would achieve safe, transcriptionally active and stable expression of an integrated gene. However, the low frequency of homologous recombination in mammalian cells prevents therapeutic application of this strategy. Recombinant adenoviral vectors (AdVs) might be an ideal vehicle for homologous recombination because of their highly efficient delivery of double-stranded DNA genomes into the nucleus of almost any cell type. We examined the ability of AdVs to correct an insertional mutation in exon 3 of the hypoxanthine phosphoribosyl transferase (Hprt) locus in male mouse embryonic stem (ES) cells. The cells were infected with four types of Hprt targeting AdVs (E1-deleted vectors with 1.7-kb and 6.7-kb wild-type Hprt sequences and helper-dependent (HD) AdVs with 6.7-kb and 18.6-kb wild-type Hprt sequences) at multiplicities of infection (MOI) of 10, 100 and 1000 genomes per cell. The ES cells that might have undergone homologous recombination and thus restored their HPRT activity were detected by HAT selection. The HD AdVs also encode the beta-geo marker gene, which flanks the Hprt sequence. Therefore, randomly integrated vectors could also be detected by G418 selection. The frequency of homologous recombination in cells infected with HD AdVs with an 18.6-kb homology was nearly 0.02%, an 800-fold increase compared to those cells infected by an E1-deleted vector with the 6.7-kb homology. No homologous recombination was detected with the E1-deleted vector with the 1.7-kb Hprt sequence. In contrast to the homologous recombination frequency, which showed an MOI-dependent increase, the random integration frequency reached maximum levels at an MOI of 100 and was similar between the HD AdVs. The ratio of random to targeted integration was 61% with the 18.6-kb HD AdV at an MOI of 1000. Treatment of the infected cells with trichostatin A, a histone deacetylase inhibitor, enhanced only the random integration frequency but had no effect on the frequency of targeted integration. The majority of HAT-resistant cell clones showed a pattern of faithful gene targeting. No randomly integrated vectors were observed in the cells with targeted vector integration, as analyzed by PCR. These findings suggest that adenovirus-mediated homologous recombination might be applicable for ex vivo gene therapy.