Photocleavable Guide crRNAs for a Light-Controllable CRISPR/Cas9 System

Abstract

The design of controllable and precise RNA-targeted CRISPR/Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats) systems is an important problem of modern molecular biology and genetic technology. Herein, we have designed a series of photocleavable guide CRISPR RNAs (crRNA) and their 2'-modified (2'-fluoro and locked nucleic acid) analogs containing one or two 1-(2-nitrophenyl)-1,2-ethanediol photolabile linkers (PL). We have demonstrated that these crRNAs can be destroyed by relatively mild UVA irradiation with the rate constants 0.24-0.77 min^-1 and that the photocleavage markedly slows down the action of Cas9 nuclease in the model in vitro system. Two PLs provide more rapid crRNA destruction than a single linker. PLs in the crRNA structure improve the specificity of DNA cleavage by Cas9 nuclease for the fully complementary target. The application of photocleavable crRNA in CRISPR/Cas9 genome editing permits the system to be switched off in a spatiotemporally controlled manner, thus alleviating its off-target effects.

Keywords: 2′-modified crRNA; off-target effects; photocleavable crRNA; photoregulated CRISPR/Cas9 system.

Figure 1

(A) Sequences of native (R) and 2′-modified crRNA (F, L) and their analogs with one (R_PL1, F_PL1 and L_PL1) or two (R_PL2, F_PL2 and L_PL2) photolabile linkers, and native tracrRNA (trR). (B) Structures of ribonucleotide, 2′-fluoro-modified nucleotide, LNA-nucleotide and photolabile linker (PL) are presented. (C) Reaction of photocleavage of PL in the crRNA structure.

Figure 2

(A) Image of a representative agarose gel showing products of cleavage of a protospacer-containing plasmid by Cas9 bound to tracrRNA and L_PL1 or L_PL2 crRNA with or without UV irradiation. M, linear DNA size markers; C«−», no enzyme; C«+», Cas9 bound to non-irradiated native crRNA/tracrRNA. Arrows mark the mobility of the supercoiled, relaxed and linear forms of the plasmid. (B–D) extent of the plasmid DNA cleavage by Cas9 targeted by photocleavable crRNAs ((B) native; (C) 2′-fluoro-modified; (D) LNA-modified) and non-modified tracrRNA. The crRNAs were irradiated either prior to the complex assembly (UV1) or as part of Cas9 ribonucleoprotein (UV2). NC, negative control (plasmid without any treatment). Mean ± SD is shown (n = 3); *, p < 0.01; **, p < 0.001; ***, p < 0.0001 (one-way ANOVA) in comparison with non-photocleavable crRNAs of the same series (R, F or L). Schemes of the substrates, crRNA and tracrRNA, are presented in Supplementary Figure S3.

Figure 3

Schemes of model DNA duplexes (DNA1, DNA2, DNA3) bound to Cas9 with PL-containing crRNAs. Y₁/X₁ and Y₂/X₂ indicate the positions of the replaced base pairs in DNA2 and DNA3.

Figure 4

The extent of DNA duplex cleavage by nuclease Cas9 targeted by photocleavable native (A,B), 2′-fluoro-modified (C,D) and LNA-modified (E,F) crRNA with one (A,C,E) or two (B,D,F) photolabile linkers. The specificity of DNA cleavage was investigated using three model 50-nt DNA duplexes labeled by Cy5 fluorescent dye attached to the 3′-end of the non-target strand (Figure 3, Table S2), containing a 20-nt protospacer sequence (Psp2 without mismatches or with a single mismatch) and a PAM. DNA1, Psp2 fully complementary to crRNA; DNA2, a target with a single A:C mismatch in position 5; DNA3, a target with a single A:C mismatch in position 7. The nature of the DNA duplex is indicated in the figures. Mean ± SD is shown (n = 3); *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001 (two-way ANOVA) in comparison with the cleavage of DNA1.