Improving the specificity and efficacy of CRISPR/CAS9 and gRNA through target specific DNA reporter

Abstract

Genomic engineering by the guide RNA (gRNA)-directed CRISPR/CAS9 is rapidly becoming a method of choice for various biological systems. However, pressing concerns remain regarding its off-target activities and wide variations in efficacies. While next generation sequencing (NGS) has been primarily used to evaluate the efficacies and off-target activities of gRNAs, it only detects the imperfectly repaired double strand DNA breaks (DSB) by the error-prone non-homologous end joining (NHEJ) mechanism and may not faithfully represent the DSB activities because the efficiency of NHEJ-mediated repair varies depending on the local chromatin environment. Here we describe a reporter system for unbiased detection and comparison of DSB activities that promises to improve the chance of success in genomic engineering and to facilitate large-scale screening of CAS9 activities and gRNA libraries. Additionally, we demonstrated that the tolerances to mismatches between a gRNA and the corresponding target DNA can occur at any position of the gRNA, and depend on both specific gRNA sequences and CAS9 constructs used.

Keywords: CAS9; CRISPR; EGFP reporter; Specificity; gRNA.

Figure 1

Characterization of the psDNA EGFP reporter system. (A) Schematic of the psDNA EGFP reporter design and function mechanism. The template pEGFP-C1 plasmid was linearized at the dual BsaI sites created between the CMV promoter (open black bar) and the EGFP reporter gene (solid black bar when unexpressible and solid green bar when expressible). A target protospacer DNA (psDNA, open red bar) with a protospacer adjacent motif (PAM, solid vertical blue bar at the right side of the psDNA) is inserted at the BsaI sites by ligation to create the psDNA EGFP reporter (solid green bar). Co-transfection of the psDNA EGFP reporter with plasmid(s) expressing a CAS9 (blue crescent) and the corresponding guide RNA (gRNA, green line with a wiggled tail) into a cell (dotted square) results in formation of the gRNA and CAS9 protein complex, which binds to the target psDNA site homologous to the gRNA sequence and generates double strand DNA break (DSB). Consequently the EGFP gene expression is diminished due to the separation of the CMV promoter from the EGFP gene (solid black bar). (B) Optimization of pEGFP-C1 dosage (ng/rx) in transfection. Right panel: percentage of fluorescent cells transfected with various amount of pEGFP plasmid as assessed by flow cytometry. The same amounts of gRNA_GFP-T2 and hCAS9 were used in all transfections. The unrelated gRNA_Pfb was used as negative gRNA control. Horizontal axis: ng pEGFP plasmid used per transfection (rx). Left panel: schematic showing the target psGFP-T2 site of the EGFP gene. (C) Creation of the psAAVS1-T2 EGFP reporter. At the dual BasI sites created between the CMV promoter and the EGFP reporter gene (shown in solid arrows) the double strand target psDNA of the gRNA_AAVS1-T2, psAAVS1-T2 (20nt in italics) and the PAM sequence (TGG in bold) flanked by BsaI compatible ends were inserted. The dotted double head arrow indicates the predicated double strand DNA break (DSB) site by gRNA_AAVS1-T2 guided CAS9 complex.

Figure 2

Validation of the psDNA EGFP reporter system. (A) Schematic sequence alignment of px330-CAS9, hCAS9 and myc-CAS9 plasmid constructs. The black horizontal bars represent the common SpCAS9 sequences (not in scale). The 3xFlag for px330-CAS9 and the myc tag for myc-CAS9 constructs are in red and the nuclear localization signals are underlined. The identical sequences at either N or C termini of px330-CAS9 and myc-cas9 are in bold black. (B) Comparison of EGFP expressing cells transfected with equal amount of psAAVS1-T2 reporter and different combinations of gRNAs and CAS9 constructs as shown. The gRNA_AAVS1-T1 with no homologies to psAAVS1-T2 reporter served as negative control.

Figure 3

Dosage response profiling of DSB efficacies of hCAS9 and gRNA_AAVS1-T2 against the target psAAVS1-T2 reporter. A and C used the non-homologous gRNA, gRNA_AAVS1-T1, as negative controls. Horizontal axis: μg plasmid DNA or gRNA used per transfection reaction (rx). The total amount of plasmid of DNAs was kept constant by using the comparable amount of the non-relevant plasmid, pW25, to minimize variations in transfection efficiency.

Figure 4

Profiling of DSB efficacies for different combinations of gRNA and CAS9 constructs. (A) Heatmap of DSB activities displayed by different combinations of gRNA and CAS9 constructs (as shown) against their corresponding psDNA EGFP reporters as labeled. The psAAVS1-T2 reporter was used for gRNA_AAVS1-T2, gRNA_GFP-T1 and T2 constructs. Non-relevant plasmid (pW25) was used as negative control to show the basal level of GFP cells without CAS9 activities. Green, yellow, and red colors represent low, medium and high DSB activities respectively. The numbers in each colored cells represent the average percentages of GFP positive cells counted by flow cytometry analysis from three or more independent biological experiments. (B–D) The psAAVS1-T2 EGFP reporter was used. (E–I) Other psDNA EGFP reporters used: psPfb (E), psR7bp1 (F), psR7bp2 (G), psR7bp3 (H), and psR7bp4 (I). Comparisons with statistically significant differences at p< 0.05 level were labeled with (*) and at p< 0.01 level were labeled with (***).

Figure 5

Profiling of DSB specificities for different combinations of gRNA and CAS9 constructs. (A) Heatmap of the potential off-target DSB activities displayed by combinations of either gRNA_Pfb (top five rows) or gRNA_R7bp3 (bottow five rows) with different CAS9 constructs and their corresponding target psDNAs as shown. The mutated nucleotides of each psDNA are in red small letters at the positions as labeled. Non-relevant plasmid, pW25 was used in control reactions. Green, yellow, and red colors represent low, medium and high DSB activities respectively. The numbers in each colored cells represent the average percentages of GFP positive cells counted by flow cytometry analysis from three or more independent biological experiments. (B) Effects of psDNA mutations on DSB efficacies of gRNA_Pfb and different CAS9 constructs. (C) Effects of psDNA mutations on DSB efficacies of gRNA_R7bp3 and different CAS9 constructs. The suffixes M1-2, M10-11, and M19-20 refer to the mutated nt positions of a psDNA respectively. Comparisons with statistically significant differences at p< 0.05 level were designed as (*) and at p< 0.01 level were designed as (***).