CRISPR-Cas system is a robust platform for genome editing application. The introduction of a DNA double-strand break (DSB) in precise gene locations can be exploited to achieve targeted gene knockout by harnessing the error-prone Non-Homologous End-Joining (NHEJ) DNA repair mechanism. However, using this technology for precise genome editing remains challenging. This relies on the homology-directed repair (HDR) pathway that uses a properly designed DNA donor as a blueprint to install the desired modification. NHEJ predominance during DSB repair together with restriction of HDR to certain phases of the cell cycle often results in HDR frequencies far below clinically relevant frequencies. Common strategies to increase HDR-mediated DSB repair include the use of chemicals to either arrest the cells in those cell cycle phases when HDR is most active or to inhibit NHEJ. However, the global effects of these drugs may pose safety concerns for clinical applications. To address this issue, we devised a strategy to recruit HDR-promoting factors or NHEJ-inhibiting proteins at the DSB site. This is achieved via their direct fusion to the Cas9 nuclease to promote the engagement of HDR during DSB repair and increase the frequency of precise genome editing. We generated 16 different Cas9-fusion proteins (referred to as HDR-CRISPR) and extensively investigated their impact on DNA repair choice by using two reporter systems, the traffic light reporter (TLR) and the BFP-to-GFP (B2G) assay. These two assays allowed us to investigate the outcome of DNA repair mediated by a DNA donor supplied either as plasmid or oligodeoxynucleotides (ODN) respectively. Our results indicated that HDR-CRISPRs can affect DSB repair leading to a 3-fold increase in HDR frequency over baseline levels. The simultaneous reduction of NHEJ-mediated repair led to a 5-fold increase in the HDR: NHEJ ratio using our best performing HDR-CRISPR. Next, we evaluated the capability of HDR-CRISPR to precisely integrate a large GFP expression cassette into the endogenous AAVS1 safe harbor locus of K562 and Jurkat cell lines. Independently by the cell type used, HDR-CRISPR increased 2.5-fold the precise integration events. The most efficient HDR-CRISPR fusion was then delivered as mRNA in T lymphocytes and Hematopoietic Stem Cells (HSCs). Using an appropriate ODN as a repair template, we aimed at introducing a stop codon within the exon 3 of the CCR5 gene to generate immune cells resistant to HIV infection. Our results show that HDR-CRISPR is capable to alter the normal resolution of a DSB, leading to a 2-fold increase in precise genome editing events as compared to the standard Cas9. Our data support the hypothesis that DSB repair choice can be altered through the local recruitment of key factors capable of either promoting HDR or inhibit NHEJ. We envision this technology may contribute to increasing precise editing yield in clinically relevant settings.
