Fig 1.
Library construction and screening of SaCas9-HF variants.
(A) Schematic of the protocol for CRISPR-SaCas9-mediated gene editing of EGFP. (B) Heatmap represents the relative SaCas9 cleavage efficiency by 63 single-mutated and one perfect-matched sgRNA each for 4 EGFP target sites (n = 3). Nonmutated guide sequence is shown above and highlighted in gray squares in the heatmap. Relative EGFP disruption efficiencies (increasing from white to blue) are normalized to the nonmutated guide sequence. The data underlying this figure can be found in S1 Data. (C) The schematic of constructing two mutational SaCas9 libraries by error-prone PCR with in-fusion methods. “*” represents the mutation. (D) The schematic strategy of screening SaCas9-HF variants. (E) Eight variants were investigated with perfect-matched sgRNA site 3 sgRNA and single-nt mismatched sgRNAs site 3 sgRNA. Relative disruption efficiencies are normalized to the perfect-matched guide sequence with WT SaCas9; error bars, SEM; n = 3. (F–H) Validation of enhanced fidelity of Mut268 variant. For each site, WT SaCas9 and variants were assessed by EGFP disruption assay with a perfect-matched sgRNA or single-nt mismatched sgRNAs. The data underlying this figure can be found in S2 Data; error bars, SEM; n = 3. 268, Mut268; EGFP, enhanced green fluorescence protein; FCM, flow cytometry; M, mismatched; M3-1, single-nt mismatched sgRNA site 3; PM, perfect-matched; PM3, perfect-matched sgRNA site 3; SaCas9, Staphylococcus aureus Cas9; sgRNA, single-guide RNA; single-nt, single nucleotide; W/O, without; WT, wild-type.
Fig 2.
Fidelity validation of improved specificity for Mut268 at ten endogenous loci.
(A) Illustration of CRISPR-SaCas9-mediated genome editing at endogenous genes. Ten endogenous-targeted sites were selected for fidelity test. (B) WT and Mut268 mediated cleavage at on-target and predicted off-target sites. Cleavage efficiency was examined by targeted deep sequencing. Edited products containing nucleotides substitution, in/del were analyzed. Mismatched nucleotides in predicted off-target sites are highlighted in red. The data underlying this figure can be found in S3 Data. Error bars, SEM; n = 2. (C) Comparison of WT and Mut268 for the enhanced specificity in HEK-293 cell lines. All 10 on-target sites were plotted to compare the on-target activity between WT and Mut268. Off-target sites edited by WT SaCas9 with obvious off-target effects and Mut268 were plotted. Sample size of 10 in on-target sites and 7 in off-target sites. (D) WT and Mut268 mediated cleavage at on-target and predicted off-target sites in HeLa cells; error bars, SEM; n = 2. (E) Comparison of WT and Mut268 for the enhanced specificity in HeLa cells. in/del, insertion/deletion; NC, negative control (without transfection of Cas9-expression plasmid); “On,” on-target; OT, predicted off-target; SaCas9, Staphylococcus aureus Cas9; WT, wild-type.
Fig 3.
Mut268 possesses lower off-target effects at genome-wide levels.
(A) Schematic of PEM-seq. (B) Edited efficiency for WT SaCas9 and Mut268 detected by PEM-seq. chr2 represents chr2:156,968,467–156,968,493 locus. (C) Off-target frequency for WT SaCas9 and Mut268 detected with PEM-seq. (D) Scatter plot of indicated off-target hotspots for WT and Mut268. y-Axis showed frequency of each hotspot per 100,000 editing events (in/dels plus translocation). in/del, insertion or deletion; PEM-seq, primer-extension-mediated sequencing; SaCas9, Staphylococcus aureus Cas9; WT, wild-type.
Fig 4.
Structural insights of enhanced fidelity.
(A) Strategy for mapping the responsible mutation(s). (B) Relative EGFP disruption of site 3 with variants harboring different mutation(s). Relative disruption efficiencies are normalized to the perfect-matched sgRNA target site 3 recognized by WT SaCas9; error bars, SEM; n = 3. (C) Effects of single amino-acid substitutions of N260 on SaCas9 editing activity and fidelity. (D) Structural context of efSaCas9. Left panel: Overall structure of a ternary complex formed by SaCas9 (grey), sgRNA (cyan), and target DNA (yellow). The REC1 and REC3 domains of the REC lobe are indicated. The PAM sequence is in magenta. Right panel: Zoomed-in view of the boxed region in (A), showing the interaction between the REC3 domain with the heteroduplex. Hydrogen bonds are indicated by yellow dotted lines. The cation interaction between N260 and Y256 are indicated by the partial charges and black dotted lines (PDB:5CZZ). (E) Fidelity comparisons of structure-guided additional SaCas9 variants with perfect-matched sgRNA target site 3 and single-nt mismatched sgRNAs; error bars, SEM; n = 3. efSaCas9, enhanced-fidelity SaCas9; M, single-nt mismatched sgRNA; PAM, protospacer adjacent motif; PM3, perfect-matched sgRNA target site 3; REC,; SaCas9, Staphylococcus aureus Cas9; sgRNA, single-guide RNA; single-nt, single nucleotide; WT, wild type.