Tissue Specific DNA Repair Outcomes Shape the Landscape of Genome Editing

Abstract

The use of Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas9 has moved from bench to bedside in less than 10years, realising the vision of correcting disease through genome editing. The accuracy and safety of this approach relies on the precise control of DNA damage and repair processes to achieve the desired editing outcomes. Strategies for modulating pathway choice for repairing CRISPR-mediated DNA double-strand breaks (DSBs) have advanced the genome editing field. However, the promise of correcting genetic diseases with CRISPR-Cas9 based therapies is restrained by a lack of insight into controlling desired editing outcomes in cells of different tissue origin. Here, we review recent developments and urge for a greater understanding of tissue specific DNA repair processes of CRISPR-induced DNA breaks. We propose that integrated mapping of tissue specific DNA repair processes will fundamentally empower the implementation of precise and safe genome editing therapies for a larger variety of diseases.

Keywords: CRISPR-Cas9; DNA double-strand break; genome editing; homology directed repair; microhomology mediated end-joining; non-homologous end-joining; tissue specific DNA repair; tissue stem cells.

Figure 1

Repair outcomes after a Cas9-induced DNA double-strand break (DSB) and strategies for enhancing precise repair. (A) Cas9, targeted by a sgRNA, induces a DSB in a precise region of the genome. Non-homologous end-joining (NHEJ), promoted by 53BP1, is the default repair pathway. Through the coordinated action of factors such as DNA-PK and LIG4, NHEJ repairs the DSB by re-joining the DNA-ends in an error-prone manner. This results in small insertions and deletions (indels) that can generate a loss-of-function allele if a frameshift is generated. If end-resection occurs [mediated by CtIP and MRE11-Rad50-NBS1 (MRN)], microhomology-mediated end-joining (MMEJ), or homology-directed repair (HDR) function. The repair outcome following MMEJ-mediated repair can vary, although this can be predicted since it depends on regions of microhomology and leads to larger indels. HDR, mediated by factors such as BRCA1 and RAD51, relies on a repair template and hence is error-free, leading to precise genomic alterations. (B) The use of ssDNA oligonucleotides (ssODN) as donor templates has also been developed to harness HDR. This process is called single-stranded templated repair (SSTR). SSTR is generally more efficient due to the asymmetry of the Cas9-DNA complex, which leads to the release of the PAM-distal non-target strand. Therefore, a rational design of the ssODN donor template complementary to the strand that is first released improves precise editing. (C) The inhibition of NHEJ has been used to improve precise repair following Cas9-breaks. 53BP1 inhibition through ubiquitin variants, dominant negative forms, or expression of factors that displace 53BP1, has proven useful. Small molecule inhibitors against DNA-PKcs and LIG4 have also been used. (D) Cell cycle manipulation has also proved useful for enhancing HDR. HDR (depicted in blue) is only active in S/G2/M phases, contrary to NHEJ (depicted in orange), which is active throughout the cell cycle. Strategies to improve HDR have included the use of compounds (such as XL413, aphidicolin, and nocodazole) to block cells in HDR-permissive phases. A Cas9-CtIP fusion allows end-resection (and subsequently HDR) to occur throughout the entire cell-cycle. PAM, protospacer adjacent-motif.

Figure 2

Successful clustered regularly interspaced short palindromic repeats (CRISPR) applications require consideration of tissue-specific DNA repair and repair pathway accessibility. (A) Hematopoietic stem cells (HSCs) are extracted from the bone marrow and edited ex vivo for the treatment of sickle cell anaemia and β-thalassemia. Without stimulating cells to enter the cell cycle once before CRISPR-Cas9 editing, quiescent HSCs rely on error prone NHEJ to repair induced DSB. In a clinical application, the preference for NHEJ is leveraged to disrupt the transcription factor BCL11A, which represses the expression of foetal hemoglobin. The re-expression of foetal hemoglobin allows for the formation of normally shaped erythrocytes. (B) The toxic accumulation of fumarylacetoacetate in fatal hereditary tyrosinemia type I (HTI) leads to liver cirrhosis and liver failure due to a mutation in the fumarylacetoacetate hydrolase gene (FAH). Highly differentiated quiescent cells can be stimulated to re-enter the cell cycle upon DNA, or tissue, damage. Provided with a single-stranded repair template, few cycling hepatocytes have access to repair DSB via homologous directed repair. Precisely edited hepatocytes have a growth advantage over non-edited cells and reconstitute tissue homeostasis. (C) Leber congenital amauroris (LCA) is the first disease treated with an in vivo CRISPR approach. The post mitotic light sensitive cells in the retina degenerate with age, leading to impaired vision early on in life. Appropriating the propensity of post mitotic cells to repair DSBs via NHEJ, the therapy aims to disrupt an aberrant splicing site in exon 26 of CEP290, maintaining a functional retina.