Defining and improving the genome-wide specificities of CRISPR–Cas9 nucleases

Abstract

Wild-type CRISPR–Cas9 nucleases can induce high-frequency mutations at unintended off-target sites. Methods for defining genome-wide specificities of CRISPR–Cas9 nucleases include cell-based methods such as integrase-defective lentiviral vector (IDLV) capture, genome-wide unbiased identification of DSBs enabled by sequencing (GUIDE-seq), high-throughput genome-wide translocation sequencing (HTGTS), and breaks labelling, enrichment on streptavidin and next-generation sequencing (BLESS), and in vitro methods such as digested genome sequencing (Digenome-seq). Methods for improving genome-wide specificity include truncated guide RNAs (tru-gRNAs), extended gRNAs, paired Cas9 nickases (Cas9n), dimeric RNA-guided FokI–dCas9 ('dead' Cas9) nucleases (RFNs) and engineered variants with reduced non-specific protein–DNA interactions. For many research applications, simple controls (such as the use of multiple gRNAs) and/or genetic reversion or complementation experiments, may help to exclude the possibility of confounding off-target effects. For therapeutic applications, it will be important to carefully define even low-frequency off-target effects using sensitive, unbiased genome-wide methods.