Cas9 gRNA engineering for genome editing, activation and repression

nature methods | ADVANCE ONLINE PUBLICATION | We targeted Cas9 and a set of truncated gRNAs to the promoter of a transiently transfected fluorescent reporter. In agreement with previous results, Cas9 showed robust levels of nuclease activity with both 20-nt and 18-nt gRNAs and a sharp loss of function with ≤16-nt gRNAs (Supplementary Fig. 1a). To determine whether the lack of DNA modification observed with ≤16-nt guides was due to attenuated Cas9 nuclease activity, we fused a potent transcriptional activator (VPR) to Cas9 (ref. 7). We then targeted the Cas9-VPR fusion product to the same fluorescent reporter and quantified the effect of gRNA length on activation. As expected, Cas9-VPR showed minimal activation when a 20-nt gRNA was used, but when the gRNA length was decreased, a corresponding increase in activation was observed, with maximal activation achieved with 16-nt or 14-nt gRNAs (Supplementary Fig. 1b). Cas9-VPR showed nuclease activity similar to that of wild-type Cas9 with 20-nt or 18-nt gRNAs, and it demonstrated reporter activation equivalent to that of a fusion between nuclease-null Cas9 and VPR (dCas9VPR) when 16-nt or 14-nt gRNAs were used (Supplementary Fig. 1). To assess the generality of this approach, we tested the effects of shortened gRNAs using two other Cas9 orthologues8 and observed a similar capacity of shortened gRNAs to inhibit nuclease activity while still allowing interaction with DNA (Supplementary Fig. 2). We next sought to determine whether our gRNA engineering paradigm would enable us to modulate Cas9 activity at endogenous target genes. Using 20-, 16and 14-nt gRNAs, we targeted Cas9, Cas9-VPR and dCas9-VPR to the promoter regions of genes encoding structural proteins (ACTC1 and TTN), a long noncoding RNA (MIAT) and a protein critical to tissue oxygen delivery (HBG1). Cas9-VPR was able to induce target chromosomal gene expression with 16-nt and 14-nt gRNA, but not with 20-nt gRNA (Fig. 1a–c and Supplementary Fig. 3). In addition, Cas9-VPR in conjunction with a 14-nt gRNA was able to generate expression equivalent to at least 40% of the expression level for all targets tested when compared with dCas9-VPR with a 20-nt gRNA. In addition to measuring gene induction, we also examined the amount of Cas9-induced insertions and deletions (indels) in the targeted regions. For ACTC1 and MIAT, mutagenesis was observed only with 20-nt gRNAs (Fig. 1a,b), whereas for TTN and HBG1, indels were observed with both 20-nt and 16-nt gRNAs (Fig. 1c and Supplementary Fig. 3). To further characterize 14and 20-nt gRNAs, we generated a series of spacer-mismatched fluorescent reporter plasmids and cas9 grna engineering for genome editing, activation and repression