Abstract
BACKGROUND: The use of programmable nucleases has transformed genome editing and functional genomics. Clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated 9 (Cas9) was developed such that targeted genomic lesions [usually DNA double-stranded breaks (DSBs)] could be introduced in vivo with ease and precision. In the presence of homology donors, these lesions facilitate high-efficiency precise genome editing (PGE) via homology-directed repair (HDR) pathways. Because DSBs can lead to genomic instability, however, a large amount of effort has been invested in methodologies (e.g., base editors) that only require nicking the chromosomal DNA on one strand. Indeed, we have demonstrated in human cells that oligodeoxynucleotide (ODN)-mediated PGE using nickase variants of Cas9 can proceed by at least two HDR subpathways termed synthesis-dependent strand annealing (SDSA) and single-stranded DNA incorporation (ssDI). Which pathway is utilized is determined by which chromosomal strand (sense or anti-sense/Watson or Crick) is nicked and by the strandedness (sense or anti-sense/Watson or Crick) of the donor ODN. RESULTS: While the mechanism of mammalian SDSA is moderately well understood, that of ssDI is not. To gain genetic insight into ssDI, we carried out a genome-wide CRISPR knockout screen to identify those genes which, when absent, enable increased ssDI. This screen identified the protein lysine methyl transferase (PKMT) Su(var)3-9, enhancer-of-zeste and trithorax (SET) domain bifurcated histone lysine methyltransferase 1 (SETDB1):activating transcription factor 7-interacting protein (ATF7IP) heterodimer and the downstream human silencing hub (HUSH) complex as strong negative regulators of ssDI. Consistent with their well-known biological effects, the negative regulation of ssDI by SETDB1/ATF7IP and HUSH was specific for transgenic reporters and for a HUSH-regulated single-copy gene, but was not observed at other (non-HUSH regulated) single-copy endogenous loci. CONCLUSIONS: In toto, these experiments underscore the profound impact that chromatin modifiers - and by extension, chromatin structure - have on PGE outcomes. Specifically, we have identified SETDB1/ATF7IP and the HUSH complex as major negative regulators of the HDR subpathway, ssDI, when the target is a transgene. These experiments are a proof-of-principle that chromatin can act as a potent barrier to genetic recombination and they strongly support the feasibility of extending similar chromatin modulating strategies to enhance PGE efficiency at endogenous single-copy loci.