371. Optimizing Custom Zinc-Finger Nucleases for Use in Human Cells

Genome engineering is a powerful instrument to study basic biological questions or to create treatment options for genetic diseases. A safe way to correct genetic mutations is through homologous recombination (HR), during which a mutant sequence is replaced with the corresponding wild-type sequence. While HR is predominant in yeast, it is a rare event in mammalian cells where non-homologous recombination is much more frequent. The creation of a targeted DNA double-strand break (DSB) has been shown to locally stimulate HR several hundred-fold by activating the cellular machinery for double-strand break repair, but it presupposes the availability of a site-specific endonuclease to create the DSB. Some initial experiments with artificial nucleases in human cells revealed that cytotoxicity of such enzymes is a major issue, most likely through cleavage at non-specific sites. Here, we present a rapid two-step method to evaluate and optimize novel, rationally designed endonucleases. In a first step, the DNA-binding domain was synthesized by assembling predefined C2H2-zinc-finger modules and fused to a transcriptional activation domain. Each zinc-finger module is designed to recognize 3 bp of DNA and the modules can be assembled in any order necessary to recognize any given sequence in a target locus. DNA-binding specificity of these artificial transcription factors was assessed in a reporter assay. In a second step, the DNA-binding domain was attached to the catalytic domain of the FokI restriction endonuclease to generate the custom nucleases. Upon co-transfection with a repair matrix, they were assessed for their ability to stimulate HR in an episomal gene repair assay. We have engineered several kinds of custom nucleases and demonstrate that high specificity, fine-tuned expression kinetics and protein design are absolutely critical to promote HR. Importantly, the same parameters are also crucial for a low level of cytotoxicity. After optimization, expression of the custom nucleases stimulated both episomal and chromosomal gene repair and HR was achieved in 35% or 2.5% of transfected cells, respectively. Thus, custom nucleases will be a valuable tool to stimulate gene repair for therapeutic purposes. Moreover, they can be easily combined with viral vector systems for efficient delivery.