CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering

Abstract

Prokaryotic type II CRISPR-Cas systems can be adapted to enable targeted genome modifications across a range of eukaryotes. Here we engineer this system to enable RNA-guided genome regulation in human cells by tethering transcriptional activation domains either directly to a nuclease-null Cas9 protein or to an aptamer-modified single guide RNA (sgRNA). Using this functionality we developed a transcriptional activation-based assay to determine the landscape of off-target binding of sgRNA:Cas9 complexes and compared it with the off-target activity of transcription activator-like (TALs) effectors. Our results reveal that specificity profiles are sgRNA dependent, and that sgRNA:Cas9 complexes and 18-mer TAL effectors can potentially tolerate 1-3 and 1-2 target mismatches, respectively. By engineering a requirement for cooperativity through offset nicking for genome editing or through multiple synergistic sgRNAs for robust transcriptional activation, we suggest methods to mitigate off-target phenomena. Our results expand the versatility of the sgRNA:Cas9 tool and highlight the critical need to engineer improved specificity.

Fig. 1. RNA-guided transcriptional activation

(a) To generate a Cas9_N- fusion protein capable of transcriptional activation, we directly tethered the VP64 activation domain to the C terminus of Cas9_N-. (b) To generate sgRNA tethers capable of recruiting activation domains, we appended two copies of the MS2 bacteriophage coat-protein binding RNA stem-loop to the 3′ end of the sgRNA and expressed these chimeric sgRNAs together with Cas9_N- and MS2-VP64 fusion protein. (c) Design of reporter constructs used to assay transcriptional activation is shown. Note that the two reporters bear distinct sgRNA target sites, and share a control TALE-TF target site. (d) Cas9_N-VP64 fusions display RNA-guided transcriptional activation as assayed by both fluorescence-activated cell sorting (FACS) and immunofluorescence assays (IF). Specifically, whereas the control TALE-TF activated both reporters, the Cas9_N-VP64 fusion activates reporters in a sgRNA sequence specific manner. (e) As assayed by both FACS and IF we observed robust sgRNA sequence-specific transcriptional activation from reporter constructs only in the presence of all 3 components: Cas9_N-, MS2-VP64 and sgRNA bearing the appropriate MS2 aptamer binding sites. The bar in the micrographs is 100μm. (f) For the REX1 gene we designed 10 sgRNAs (positions indicated in the figure) targeting a ~5kb stretch of DNA upstream of the transcription start site (DNase hypersensitive sites are highlighted in green), and assayed transcriptional activation using both the above approaches via qPCR of the endogenous genes. Although introduction of individual sgRNAs modestly stimulated transcription, multiple sgRNAs acted synergistically to stimulate robust multi-fold transcriptional induction. Note that in the absence of the 2X-MS2 aptamers on the sgRNA we do not observe transcriptional activation via the sgRNA-MS2-VP64 tethering approach. Data are means +/− SEM (N=3).

Fig. 2. Evaluating the landscape of targeting by sgRNA:Cas9 complexes and TALEs

(a) The methodology of our approach is outlined (refer also Supplementary Fig. 6). (b) The targeting landscape of a sgRNA:Cas9 complex reveals that it is potentially tolerant to 1–3 mutations in its target sequences. (c) The sgRNA:Cas9 complex is also largely insensitive to point mutations, except those localized to the PAM sequence. Notably this data reveals that the predicted PAM for the S. pyogenes Cas9 is not just NGG but also NAG. (d) Introduction of 2 base mismatches significantly impairs the sgRNA:Cas9 complex activity, primarily when these are localized to the 8–10 bases nearer the 3′ end of the sgRNA target sequence (in the heat plot the target sequence positions are labeled from 1–23 starting from the 5′ end). (e) Similarly examining the TALE off-targeting data for an 18-mer TALE reveals that it can potentially tolerate 1–2 mutations in its target sequence, and fails to activate a large majority of 3 base mismatch variants in its targets. (f) The 18-mer TALE is, similar to the sgRNA:Cas9 complexes, largely insensitive to single base mismatched in its target. (g) Introduction of 2 base mismatches significantly impairs the 18-mer TALE activity. Notably we observe that TALE activity is more sensitive to mismatches nearer the 5′ end of its target sequence (in the heat plot the target sequence positions are labeled from 1–18 starting from the 5′ end). Statistical significance symbols are: *** for P<.0005/n, ** for P<.005/n, * for P<.05/n, and N.S. (Non-Significant) for P>= .05/n, where n is the number of comparisons (refer Supplementary Table 3).

Fig. 3. Off-set nicking

(a) We employed the traffic light reporter to simultaneously assay for HR and NHEJ events upon introduction of targeted nicks or breaks: DNA cleavage events resolved through the HDR pathway restore the GFP sequence (via a donor template), whereas mutagenic NHEJ causes frame-shifts rendering the GFP out of frame and the downstream mCherry sequence in frame. For the assay, we designed 14 sgRNAs covering a 200bp stretch of DNA: 7 targeting the sense strand (U1–7) and 7 the antisense strand (D1–7). Using the Cas9D10A mutant, which nicks the complementary strand, we used different two-way combinations of the sgRNAs to induce a range of programmed 5′ or 3′ overhangs (the nicking sites for the 14 sgRNAs are indicated). (b) Inducing off-set nicks to generate DSBs is highly effective at inducing gene disruption. Notably off-set nicks leading to 5′ overhangs result in more NHEJ events as opposed to 3′ overhangs. (c) Again, off-set nicks leading to 5′ overhangs also result in more HR and NHEJ events as opposed to 3′ overhangs. In (b,c) the predicted overhang lengths are indicated below the corresponding x-axis legends. Data are means +/− SEM (N=3).