Zinc Finger Nuclease-Mediated Gene Targeting of Human Rhodopsin Mutations in Mice

Abstract

Purpose: : To Test and optimize the use of Zinc Finger Nucleases (ZFN) as potential agents for gene therapy of dominant mutations in photoreceptor cells.

Methods: : Mouse lines were engineered to have the mouse rhodopsin gene substituted by a mutant of human rhodopsin fused to EGFP at the C-terminus. Three of these encode point mutations (P23H, Q64X and Q344X)that cause Autosomal Dominant Retinitis Pigmentosa (ADRP) in humans. ZFNs were designed that bind and cause double strand breaks (DSBs) at target sequences close to mutation sites in cell culture. We used recombinant adeno-associated virus (rAAV) to deliver the ZFNs alone or in combination with a DNA fragment of the gene with the mutation corrected as a template for repair by homologous recombination (HR) into retinas of these mice. Retinas were examined for fluorescent rods by confocal microscopy at different times post-treatment.

Results: : Morphological, biochemical and functional characterization shows that the heterozygotes of our knock-in mice are very similar to wild type and have little cell death over time; as expected from the low expression levels of the mutant protein, due to engineered hairpin structures in the mRNA. hRhoP23H-EGFP mice express the full length mutant protein and revealed mislocalization to the inner segment and nucleus of photoreceptors. Treatment with rAAV-ZFNs led to ablation of EGFP signal presumably due to DNA repair by the error-prone non homologous end joining pathway (NHEJ), which usually causes frame shifts and premature termination. In contrast, retinas of hRhoQ64X-EGFP and hRhoQ344X-EGFP express a truncated protein with no fluorescence. Treatment with rAAV encoding ZFNs plus rAAV containing the HR DNA template, resulted in appearance of a large number of EGFP-positive rods. Gene repair was increased up to 10,000-fold above background.

Conclusions: : Our knock-in mouse lines are an extremely sensitive system for the detection of gene targeting and repair in vivo. We have demonstrated that ZFNs can be designed and used to cause targeted DSBs in the genome of photoreceptor cells, which can be repaired by either NHEJ or HR. Optimization of the efficiency of these genome correction procedures to therapeutic levels, could provide a novel approach to the treatment of dominant retinal and other neurological diseases.

Keywords: gene transfer/gene therapy • photoreceptors