Modified guide RNA for inhibiting off-target binding in crispr / cas9-based genome editing, and use thereof

Abstract

The present application provides a guide RNA for inhibiting off-target binding in CRISPR / Cas9-based genome editing, wherein the guide RNA comprises at least one abasic spacer in a crRNA sequence. The effects of the present invention include inhibiting unintended off-target binding while maintaining the target genome editing effect by CRISPR / Cas9 by weakening the binding force with the target DNA while maintaining the structure of the guide RNA.

Description

Modified guide RNA that inhibits off-target binding in CRISPR / CAS9-based genome editing and its uses

[0001] This application claims priority based on Korean Patent Application No. 10-2024-0125407 filed on September 13, 2024, and all contents disclosed in the specification and drawings of said application are incorporated into this application.

[0002] The present invention relates to a modified guide RNA that inhibits off-target binding in CRISPR / Cas9-based genome editing and its uses.

[0003] Genome editing, a technology that alters genomic sequences for desired purposes, involves artificially expressing a protein complex known as "gene scissors"—which recognizes and cuts specific DNA sequences—into a cell. This causes the DNA to cut at a specific desired location, inducing the cell's DNA repair mechanism and thereby altering the DNA base sequence as intended. For this process, it must be possible to easily modify the DNA sequence to recognize the desired genomic target location. Currently, genome editing technologies based on the CRISPR / Cas9 system are widely used because they allow for easy target modification through the use of guide RNA and the base sequences of the target DNA.

[0004] CRISPR / Cas9 consists of Cas9, a protein that cleaves DNA dimeric molecules, and guide RNA, which enables Cas9 to specifically recognize target DNA. When recognizing target DNA, Cas9 recognizes it only if it possesses a three-nucleotide DNA sequence called a Protospacer Adjacent Motif (PAM) (primarily NGG). Additionally, to specifically recognize a particular sequence, it recognizes it by aligning complementarily with the target DNA through approximately 20 bases of a small guide RNA called crRNA, which consists of about 35 bases. For crRNA to form a complex with Cas9 and act as a guide, it must bind to and co-occur with tracrRNA, an RNA molecule of about 65 bases. Therefore, for convenience, the two RNA molecules, crRNA and tracrRNA, are sometimes combined into a single guide RNA called sgRNA for use in genome editing. As such, the DNA targets recognized by Cas9 can be easily changed depending on the guide RNA sequence, and accordingly, the CRISPR / Cas9 system is widely used to correct DNA sequences in various cells and organisms. In other words, if 20 base pairs from the 5' end of the guide RNA are provided to bind complementarily to the DNA sequence immediately preceding the PAM sequence, gene editing by CRISPR / Cas9 can be induced in any desired region of the genome. Depending on the application, CRISPR / Cas9-based genome editing technology can restore disease mutations in specific genes to their original state and also modify the expression of specific genes; therefore, it can be widely utilized as a therapeutic agent to correct disease-causing genomic mutations back to normal.

[0005] To enable CRISPR / Cas9-based genome editing technology to specifically recognize a target sequence of specific DNA, guide RNAs are designed and used to have approximately 20 bases complementary to that sequence. The problem here is that the binding affinity can vary depending on the base sequence of the guide RNA binding to the target DNA. More seriously, guide RNAs can bind sufficiently to DNA sequences in other regions through partial base sequences even if they are not perfectly complementary. In other words, the off-target effect is well known, where guide RNA can recognize a target and cleave it via Cas9 even with up to four base sequence mismatches. Consequently, because the CRISPR / Cas9 system can cause critical problems and side effects by having guide RNA bind to off-targets and trigger DNA cleavage in unintended regions during the genome editing process, the issue of off-target binding is a challenge that must be addressed in genome editing technology.

[0006] The first aspect of the present invention provides a guide RNA for inhibiting non-target binding, wherein the guide RNA comprises one or more abasic spacers within a crRNA sequence that recognizes a target.

[0007] A second aspect of the present invention provides a composition for inhibiting non-target binding comprising a guide RNA according to the present invention.

[0008] The third aspect of the present invention provides a gene editing method using guide RNA according to the present invention.

[0009] The fourth aspect of the present invention provides a method for producing a guide RNA for inhibiting non-target binding, wherein the method comprises the step of including one or more abasic spacers in the crRNA sequence of the guide RNA.

[0010] The effect according to the present invention includes maintaining the structure of the guide RNA while weakening the binding affinity with the target DNA, thereby maintaining the target genome editing effect by CRISPR / Cas9 while suppressing unintended non-target binding.

[0011] FIG. 1a is the result of S. pyogenes genome sequence analysis according to one embodiment or one example of the present invention.

[0012] FIG. 1b is the result of identifying potential non-target sequences for a spacer sequence derived from crRNA according to one embodiment or one example of the present invention.

[0013] FIG. 1c is the result of identifying potential non-target sequences for a spacer sequence derived from crRNA according to one embodiment or one example of the present invention.

[0014] FIG. 1d is the result of an in vitro cleavage analysis using recombinant SpCas9 protein and synthetic guide RNA according to one embodiment or one example of the present invention.

[0015] FIG. 1e is the result of further analyzing the SF370 strain from which the CRISPR-Cas9 technology according to one embodiment or one example of the present invention originated.

[0016] FIG. 1f is the result of supplementing additional tracrRNA to the reaction according to one embodiment or one example of the present invention to block the activity of other potential RNAs that may be present in the fraction and enable corresponding target cleavage.

[0017] Figure 1g is the result of quantitatively measuring the extratarget activity of endogenous crRNA according to one embodiment or one example of the present invention.

[0018] Figure 1h shows the results of endogenous crRNA according to one embodiment or one example of the present invention showing a significant reduction in some off-target cleavage.

[0019] FIG. 1i is a result showing that endogenous crRNA according to one embodiment or one example of the present invention exhibits a significant reduction in some off-target cleavage.

[0020] Figure 1j is a result showing that endogenous crRNA according to one embodiment or one example of the present invention exhibits a significant reduction in some off-target cleavage.

[0021] Figure 1k shows the results of endogenous crRNA according to one embodiment or one example of the present invention showing a significant reduction in some off-target cleavage.

[0022] FIG. 11 is the result of comparing the cDNA sequences of the endogenous crRNA of SF370 and Bruno strains according to one embodiment or one example of the present invention with synthetic crRNA.

[0023] FIG. 1m is the result of dot blot and ELISA analysis according to one embodiment or one example of the present invention.

[0024] FIG. 1n is a Fenton reaction result according to one embodiment or one example of the present invention.

[0025] FIG. 10 is an experimental result using synthetic RNA containing a base-free modification according to one embodiment or one example of the present invention.

[0026] FIG. 2a is the result of analyzing the cDNA sequence end of an endogenous crRNA according to one embodiment or one example of the present invention.

[0027] FIG. 2b is the result of an in vitro experiment in which crRNA according to one embodiment or one example of the present invention was oxidized through a Fenton reaction.

[0028] FIG. 2c is a sequencing method ("ARP-crRNA Seq") for directly mapping base-free modifications of crRNA according to one embodiment or one example of the present invention.

[0029] FIG. 2d is an ARP-crRNA Seq result according to one embodiment or one example of the present invention.

[0030] FIG. 2e is an ARP-crRNA Seq result according to one embodiment or one example of the present invention.

[0031] Figure 2f shows the results of confirming reactive oxygen species (ROS) when Phage A1 infects a host SF370 strain according to one embodiment or one example of the present invention.

[0032] Figure 2g is the result of confirming base-free modification of RNA and crRNA according to one embodiment or one example of the present invention.

[0033] FIG. 2h is the result of treatment with the antioxidant N-acetylcysteine ​​(NAC) according to one embodiment or one example of the present invention.

[0034] FIG. 2i is the result of in vitro SpCas9 analysis of endogenous crRNA according to one embodiment or one example of the present invention.

[0035] FIG. 2j is the result of in vitro SpCas9 analysis of endogenous crRNA according to one embodiment or one embodiment of the present invention.

[0036] FIG. 2k is a Phage A1 infection result according to one embodiment or one example of the present invention.

[0037] FIG. 21 is a Phage A1 infection result according to one embodiment or one example of the present invention.

[0038] FIG. 2m is the result of oxidation of a base-free RNA fragment according to one embodiment or one example of the present invention.

[0039] FIG. 2n is the result of confirming an insertion-deletion mutation (indel) according to one embodiment or one example of the present invention.

[0040] FIG. 20 is the result of confirming the oxidation of crRNA in S. pyogenes by ROS induced during bacteriophage infection according to one embodiment or one example of the present invention.

[0041] FIG. 3a is a schematic diagram of forming hydrogen bonds with Q926 and R661 of a SpCas9 protein according to one embodiment or one example of the present invention.

[0042] FIG. 3b is a schematic diagram of the loss of protein interaction in a tru-gRNA (tru-gRNA; 17-nt spacer, X17) with shortened 5' end complementarity of gRNA according to one embodiment or one example of the present invention.

[0043] FIG. 3c is the result of confirming the potential for base pair formation when a base-free substitution is introduced at positions 18-20 according to one embodiment or one example of the present invention.

[0044] FIG. 3d shows Ø3X17 gRNA actually synthesized in an in vitro SpCas9 assay according to one embodiment or one example of the present invention.

[0045] FIG. 3e is the result of constructing a reporter system for target sites T2, T3, and EMX1 according to one embodiment or one example of the present invention.

[0046] FIG. 3f is the result of observing off-target removal at T3 according to one embodiment or one example of the present invention.

[0047] FIG. 3g is the result of analyzing the SpCas9-gRNA-target DNA structure (PDB ID: 5F9R) according to one embodiment or one example of the present invention.

[0048] FIG. 3h is the result of structural analysis (PDB ID: 5Y36) according to one embodiment or one example of the present invention.

[0049] FIG. 3i is the result of confirming cleavage activity in a target DNA strand according to one embodiment or one example of the present invention.

[0050] FIG. 3j is a schematic diagram of GGX20 according to one embodiment or one example of the present invention, in which the added positions 21 and 22 are substituted with base-free groups.

[0051] FIG. 3k is the result of confirming activity within the target in a SpCas9 test according to one embodiment or one example of the present invention.

[0052] FIG. 31 is the result of confirming activity within the target in a SpCas9 test according to one embodiment or one example of the present invention.

[0053] FIG. 3m is the result of confirming activity within the target in a SpCas9 test according to one embodiment or one example of the present invention.

[0054] FIG. 3n is the result of confirming extratarget activity in a SpCas9 test according to one embodiment or one example of the present invention.

[0055] FIG. 4a is a schematic diagram of introducing baseless modifications at positions 18 and 19 in S. pyogenes according to one embodiment or one example of the present invention, and extending them to position 22.

[0056] FIG. 4b is a schematic diagram of introducing baseless modifications at positions 18 and 19 in S. pyogenes according to one embodiment or one example of the present invention, and extending them to position 22.

[0057] FIG. 4c is the result of confirming the activity within the target in an in vitro analysis for ØXØ gRNA according to one embodiment or one example of the present invention.

[0058] FIG. 4d is the result of confirming the activity within the target in an in vitro analysis for ØXØ gRNA according to one embodiment or one example of the present invention.

[0059] FIG. 4e is a reporter analysis result according to one embodiment or one example of the present invention.

[0060] FIG. 4f is the result of analyzing insertion / deletion (indel) mutations at target and out-of-target locations using targeted sequencing according to one embodiment or one example of the present invention.

[0061] FIG. 4g is the result of analyzing insertion / deletion (indel) mutations at target and out-of-target locations using targeted sequencing according to one embodiment or one example of the present invention.

[0062] FIG. 4h is the result of analyzing insertion / deletion (indel) mutations at target and out-of-target locations using targeted sequencing according to one embodiment or one example of the present invention.

[0063] FIG. 4i is the result of analyzing insertion / deletion (indel) mutations at target and out-of-target locations using targeted sequencing according to one embodiment or one example of the present invention.

[0064] FIG. 5a is a CIRCLE-seq according to one embodiment or one example of the present invention.

[0065] FIG. 5b is the CIRCLE-seq analysis result for T2, T3 and EMX1 according to one embodiment or one example of the present invention.

[0066] FIG. 5c is the CIRCLE-seq analysis result for T2, T3 and EMX1 according to one embodiment or one example of the present invention.

[0067] FIG. 5d is the result of confirming the level of off-target cleavage within the genome according to one embodiment or one example of the present invention.

[0068] FIG. 5e is the result of confirming the level of off-target cleavage within the genome according to one embodiment or one example of the present invention.

[0069] FIG. 5f is the result of confirming the target cleavage result for EMX1 gRNA according to one embodiment or one example of the present invention.

[0070] FIG. 5g is a synthetic DNA target library containing all 1MM of a T2 target according to one embodiment or one example of the present invention.

[0071] FIG. 5h is a library of any mismatched sequences for T2, T3 and EMX1 according to one embodiment or one example of the present invention.

[0072] FIG. 5i is the result of in vitro SpCas9 analysis according to one embodiment or one example of the present invention.

[0073] FIG. 5j is the result of in vitro SpCas9 analysis according to one embodiment or one example of the present invention.

[0074] FIG. 5k is a heatmap analysis result according to one embodiment or one example of the present invention.

[0075] FIG. 5L is the result of evaluating the performance of ØXØ gRNA compared with SpCas9 variants according to one embodiment or one example of the present invention.

[0076] FIG. 5m is the result of evaluating the performance of ØXØ gRNA compared with SpCas9 variants according to one embodiment or one example of the present invention.

[0077] FIG. 5n is a random mismatch library according to one embodiment or one example of the present invention.

[0078] FIG. 50 is the result of in vitro SpCas9 analysis according to one embodiment or one example of the present invention.

[0079] FIG. 5p is a random mismatch library according to one embodiment or one example of the present invention.

[0080] FIG. 5q is a random mismatch library according to one embodiment or one example of the present invention.

[0081] FIG. 6a shows the conversion of positions 18 and 19 of gRNA according to one embodiment or one example of the present invention to dSpacer or rSpacer (Fig. 6a).

[0082] FIG. 6b is a result of confirming the T2 target site cutting ability according to one embodiment or one example of the present invention.

[0083] FIG. 6c is the result of confirming the reduction of the T2 outside the target area according to one embodiment or one example of the present invention.

[0084] FIG. 6d is a result of confirming the T2 target site cutting ability according to one embodiment or one example of the present invention.

[0085] FIG. 6e shows cases in which dSpacer substitution according to one embodiment or one example of the present invention is applied to positions 1 through 8 of gRNA, respectively.

[0086] FIG. 6f is a result of confirming the T2 target site cutting ability according to one embodiment or one example of the present invention.

[0087] FIG. 6g is the result of confirming the reduction of the T2 out-of-target area according to one embodiment or one example of the present invention.

[0088] FIG. 6h shows the case in which Ø3X17 gRNA (18-20 base-free substitutions) and Ø2X18 gRNA (19-20 base-free substitutions) having two or more dSpacer substitutions according to one embodiment or one example of the present invention applied consecutively to the gRNA terminals were introduced into a cell.

[0089] FIG. 6i is a reporter analysis result according to one embodiment or one example of the present invention.

[0090] FIG. 6j is a reporter analysis result according to one embodiment or one example of the present invention.

[0091] FIG. 7a shows the introduction of baseless conversion and baseless extension according to one embodiment or one example of the present invention applied in various ways.

[0092] FIG. 7b is the result of in-test analysis according to one embodiment or one example of the present invention.

[0093] FIG. 7c is the result of an in-test analysis according to one embodiment or one example of the present invention.

[0094] FIG. 7d is the result of in-test analysis according to one embodiment or one example of the present invention.

[0095] FIG. 7e shows Ø4X19 gRNA according to one embodiment or one example of the present invention.

[0096] Figure 7f shows the results of analysis using a reporter after introducing Ø3XØgRNA, Ø3X20 gRNA, and Ø4X19 gRNA into cells according to one embodiment or one example of the present invention.

[0097] FIG. 7g is the result of analysis using a reporter after introducing Ø3XØgRNA, Ø3X20 gRNA, and Ø4X19 gRNA into cells according to one embodiment or one example of the present invention.

[0098] The first aspect of the present invention provides a guide RNA for inhibiting non-target binding, wherein the guide RNA comprises one or more abasic spacers within a crRNA sequence that recognizes a target.

[0099] According to one embodiment, the guide RNA is,

[0100] 1) One or more of the sequences among the above crRNA sequences that are complementary to the target DNA base are substituted with non-base spacers, or

[0101] 2) One or more non-basic spacers extending to the 5' end of the above crRNA sequence may be added.

[0102] According to one embodiment, the guide RNA may be such that three or more sequences of the crRNA sequence that are arranged complementarily to the target DNA base are substituted with non-base spacers.

[0103] According to one embodiment, the guide RNA may have bases 18, 19, and 20 of the crRNA sequence substituted with non-base spacers.

[0104] According to one embodiment, the guide RNA may have two or more non-basic spacers added to the 5' end of the crRNA sequence.

[0105] A second aspect of the present invention provides a composition for inhibiting non-target binding comprising a guide RNA according to the present invention.

[0106] The third aspect of the present invention provides a gene editing method using guide RNA according to the present invention.

[0107] The fourth aspect of the present invention provides a method for producing a guide RNA for inhibiting non-target binding, wherein the method comprises the step of including one or more abasic spacers within a crRNA sequence that recognizes a target in the guide RNA.

[0108] Throughout this specification, "guide RNA" refers to a short RNA molecule that functions to guide Cas9 or a similar nuclease to a specific double-stranded cleavage site in a CRISPR-Cas system. The guide RNA contains a base sequence complementary to the target DNA, binds to the Cas9 protein to form a complex, and the complex recognizes a specific region within the genome to induce DNA cleavage at the precise location.

[0109] Generally, guide RNA consists of about 20 nucleotides, among which the 'spacer' region is a nucleotide sequence that has basic complementarity with the target DNA and determines target specificity. In addition, it includes structural elements (scaffolds or charged structures) that maintain binding with the Cas protein. Thanks to this structure, guide RNA plays a role in precisely designating target sites in CRISPR gene editing technology, thereby enabling accurate gene editing at desired genomic locations.

[0110] Guide RNA is sometimes synthesized in the form of a single guide RNA (sgRNA), which is an artificial combination of crRNA and tracrRNA and is modified and optimized in various ways for experimental and medical purposes.

[0111] In another embodiment of the present invention, the abasic spacer is a compound capable of maintaining a space for a nucleotide site, preferably an organic compound, most preferably a hydrocarbon chain containing a phosphate group or a sulfate group, and the hydrocarbon chain may be an alkyl group having at least three carbon atoms (C3 spacer).

[0112] In another embodiment of the present invention, the abasic spacer may be one that cannot bind to biological bases and form a base sequence, and may be a dSpacer molecule having a deoxyribonucleotide backbone or an rSpacer molecule having a ribonucleotide backbone as a nonbasic form.

[0113] [Experimental Method]

[0114] 1. Confirmation of off-target binding effects via crRNAs in S. pyogenes (Off-target prediction for crRNAs in S. pyogenes)

[0115] Annotations of CRISPR and Cas genes in S. pyogenes (258 strains) were obtained from the CRISPRCasdb database via the CRISPR-Cas++ website. Batch tables for CRISPR arrays, Cas clusters, and GenBank registry numbers were downloaded via SQL dump. Among strains with type II Cas clusters (n = 173), 159 strains with CRISPR-Cas9 systems possessing both high-reliability CRISPR arrays (evidence level = 4) and Cas9 genes were selected. To define spacer sequences in crRNA, among strains containing CRISPR loci adjacent to Cas9 loci (distance between starting points < 6.1 kb, observed in SF370, n = 111), only strains with CRISPR arrays of known sequence orientation (n = 107) were analyzed. Through this analysis, a total of 481 spacer sequences (171 independent sequences) were produced by extracting a 20 nt segment adjacent to the 5' end of the direct repeat (DR) sequence of crRNA.

[0116] To predict potential off-target sites for these spacer sequences, we performed a fuzzy search using PyPI's regex module to identify partially matching sites (≤6MM, n 4,827). This targeted spacer sequences or their inverse complements, followed by an investigation into the common PAM motifs (NGGs) present within the corresponding S. pyogenes genomes. The mismatch locations of potential off-target sites were analyzed relative to the unique spacer sequences. The location frequency distributions of sites including 2MM (n = 7), 3MM (n = 6), 4MM (n = 35), 5MM (n = 187), and 6MM (n = 1,372) were visualized as heatmaps generated using Morpheus. The heatmaps represent the distribution of mismatch locations in potential off-targets containing PAM sequences (NGGs) relative to the crRNA 5' spacer; the proportions are expressed as percentages relative to the total number of 1MMs and normalized according to the degree of mismatch. To provide a comprehensive understanding of the data, the distribution for ≤4MM was additionally summarized as a line graph. After performing hierarchical clustering based on position weights (starting at position 10 and gradually increasing to 11) in the PAM proximal region, the locations of potential off-target mismatches were visualized as a heatmap. To prevent redundant representation, only unique pairs of target and spacer sequences (n 1,607) were included in the analysis. The expected frequency of off-targets was calculated as the product of the probability of NGG PAM occurrence, the probability of mismatch within the spacer, the probability of a perfect match at the remaining locations, and the genome size.

[0117] 2. ROS measurement in S. pyogenes

[0118] CM-H2DCFDA was used to quantify ROS levels in S. pyogenes. 1 mL of S. pyogenes culture was sampled, harvested by centrifugation (3,500 × g, 10 min, 4°C), and resuspended in PBS (2.5 mM DFOM, 300 μL). SF370 was infected on Phage A1 (MOI = 0.1) for 2 hours, and samples were collected at 30-minute intervals. An equal amount of cells (OD600 nm = 0.5) were incubated with CM-H2DCFDA in the dark at 37°C for 30 minutes. Fluorescence signals were measured using a Qubit 2.0 Fluorometer.

[0119] 3. crRNA synthesis and modification

[0120] crRNA synthesis and modification were custom-made by TriLink Biotechnologies (USA) and Bioneer (Korea), and quality was monitored, reported, and verified via mass spectrometry. Due to high intracellular stability, baseless nucleotides (Ø) were synthesized as dSpacer (baseless deoxynucleotides lacking a hydroxyl group at the 1' position) and introduced at designated sites. To analyze baseless-induced mutations in cDNA, various Ø RNAs (5Ø, 5Ø BamHI, 7Ø, 7G RNA) and 7o8G were additionally synthesized to compare mutations induced by 8-oxoguanine mutations.

[0121] Synthetic gRNA in the form of a complex of synthetic crRNA and tracrRNA at a level of 25 μM was produced by mixing 2 μL of synthetic crRNA (100 μM), 2.2 μL of tracrRNA (100 μM, Sigma-Aldrich, #TRACRRNA05N), and 5 μL of ultrapure water in a volume ratio of 1:1.1:1.9, followed by annealing at 90°C for 4 minutes, 70°C for 10 minutes, and 37°C for 15 minutes, or cooling to 25°C at a rate of -0.1°C / s (10 minutes 50 seconds) and storing at 4°C (Table 1).

[0122] crRNA and Synthetic / Modified crRNA Information VEGFA(T2)crRNA5'p-GACCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'tru-crRNA5'p-CCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'Ø3X 17 crRNA5'p-ØØØCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'GGX 20 crRNA5'p-GGGACCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'ØØX 20 crRNA5'p-ØØGACCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'rØrØX 20 crRNA5'p-rØrØGACCCCCUCCACCCCGCCUCGUUUUUAGAGCUAUGCUGUUUUG-3'Ø3XØ crRNA5'p-ØØØGØCCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'Ø3X 20 crRNA5'p-ØØØGACCCCCUCCACCCCGCCUCGUUUUAGAGCUAUGCUGUUUUG-3'Ø4X 19crRNA5'p-ØØØACCCCCUCCACCGCCUCGUUUAGAGCUAUGCUGUUUUG-3'ØXØ crRNA5'p-ØØGØCCCCACCCCCGCCUCGUUUAGAGCUAUGCUGUUUUG-3'VEGFA(T3)crRNA5'p-GGUGAGUGAGUGUG UGCGUGGUUUUAGAGCUAUGCUGUUUUG-3'tru-crRNA5'p-GAGUGAGUGUGUGCGUGGUUUUAGAGCUAUGCUGUUUUG-3'Ø3X 17 crRNA5'p-ØØØGAGUGAGUGUGUGCGUGGUUUUAGAGCUAUGCUGUUUUG-3'GGX 20 crRNA5'p-GGGGUGAGUGAGUGUGUGCGUGGUUUUAGAGCUAUGCUGUUUUG-3'ØX 20 crRNA5'p-ØGGUGAGUGAGUGUGUGCGUGGUUUAAGAGCUAUGCUGUUUUG-3'ØXØ crRNA5'p-ØØØGAGUGAGUGUGUGCGUGGUUUAGAGCUAUGCUGUUUUG-3'EMX1crRNA5'p-GAGUCCGAGCAGAAGAA GAAGUUUAAGCUAUGCUGUUUUG-3'tru-crRNA5'p-UCCGAGCAGAAGAAGUUUUAGAGCUAUGCUGUUUUG-3'Ø3X 17 crRNA5'p-ØØØUCCGAGCAGAAGAAGAAGUUUUAGAGCUAUGCUGUUUUG-3'GGX 20 crRNA5'p-GGGAGUCCGAGCAGAAGAAAGUUUUAGAGCUAUGCUGUUUUG-3'ØØX 20 crRNA5'p-ØØGAGUCCGAGCAAGAAGAAGUUUUAGAGCUAUGCUGUUUUG-3'ØXØ crRNA5'p-ØØGØØUCCGAGCAGAAGAAGAAGUUUUAGAGCUAUGCUGUUUUG-3'

[0123] SEQ NO. 1 to 12 SEQ NO. Sequence Remarks 1nnnCCCCUCCACCCCGCCUCVEGFA(T2)2nnGACCCCCUCCACCCCGCCUC3nnnGnCCCCCUCCACCCCGCCUC4nnnGACCCCCUCCACCCCGCCUC5nnnnACCCCCUCCACCCCGCCUC6nnGnnCCCCUCCACCCCGCCUC7nnnGAGUGAGUGUGUGCGUGVEGFA(T3)8nnGGUGAGUGAGUGUGUGUGCGUG9nnGnnGAGUGAGUGUGUGCGUG10nnnUCCGAGCAGAAGAAGAAEMX111nnGAGUCCGAGCAGAAGAAGAA12nnGnnUCCGAGCAGAAGAAGAA

[0124] n: baseless

[0125] As control RNA (cont), tracrRNA or the self-hybridized product of cel-miR-67 (C. elegans-specific miRNA, provided by Dharmacon) was used instead of crRNA in mouse experiments. cel-miR-67 consists of 5'p-ucACAACCUCCUAGAAAGAGUAdTdT-3' and 5'′' (dT is a thymidine deoxynucleotide, lowercase 2'-O-methyl modification).

[0126] To generate a small RNA-Seq library, a 5' RNA adapter (5'-GUUCAGAGUUCUACAGUCCGACGAUc-3'), a 5' UMI adapter (5'-GUUCAGAGUUCUACAGUCCGACGAUCNNNNc-3'), and a 3' adapter (5'-(rApp)T(PMe)GGAATTCTCGGGTGCCAAGG(ddC)-3'; rApp is adenylate, PMe is methyl phosphate, and ddC is dideoxycytosine) were synthesized.

[0127] 4. In vitro RNA oxidation (using Fenton reaction)

[0128] In vitro RNA oxidation was induced using the Fenton reaction. 2.5–15 μg RNA was reacted with H2O2 (0.1–2.0 mM; Sigma-Aldrich, #216763) at 37°C for 1 hour, ensuring that 0.5 mM ferric citrate (III) (Sigma-Aldrich, #F6129), 5 mM ascorbic acid (Sigma-Aldrich, #A92902), and 10 mM sodium phosphate (pH 7.4) (Biosesang, #SR2154-100-74) were included in a total solution of 75 μL. To reflect physiologically relevant ROS concentrations in S. pyogenes, 0.5 mM and 2 mM H2O2 were used based on reported intracellular concentrations (≤ 0.5 mM) and plasma concentrations (1–5 mM). ROS levels induced upon bacteriophage infection were also within the intracellular concentration range (≤ 0.5 mM). For the in vitro oxidation of synthetic crRNA, RNA (2.5–5 μg) was purified using a MicroSpin G-25 column, followed by a Fenton reaction. The reaction was terminated by adding 25 μL of 40 mM DFOM and incubating on ice for 5 minutes. The oxidized RNA was further purified using the RNA purification protocols of the Qiagen RNeasy Mini kit or the RNeasy MinElute Cleanup Kit, and 1 μg of RNA was used for sequencing library construction. An untreated control group was also treated under the same conditions without H2O2.

[0129] 5. In vitro SpCas9 assays for crRNAs

[0130] Recombinant SpCas9 protein (EnGen Spy Cas9 NLS; NEB, #M0646) was used for in vitro cleavage studies with slight modifications to the manufacturer's protocol. For DNA substrate preparation, target sequences within the S. pyogenes genome (4 nt forward + 20 nt target region + PAM) were requested for synthesis, which included adapters (underlined portions) commonly used in small RNA sequencing library construction. The target region was designed with the 4 nt forward sequence as NNNN and the PAM as NGG. PCR was performed on the synthesized DNA oligonucleotides to generate extended double-stranded DNA substrates (~72 bp). The PCR conditions used were Q5 High-fidelity 2x master mix (NEB), 100 nM universal forward primers, and 100 nM barcode primers. The PCR products were purified using the QIAquick PCR Purification Kit (QIAGEN).

[0131] For the analysis using synthetic crRNA, 2 μL of 100 μM synthetic crRNA and 2.2 μL of 100 μM tracrRNA were mixed with 5 μL of ultrapure water, heated at 90°C for 4 minutes, cooled to 25°C at -0.1°C / s, and stored at 4°C to prepare 20 μM gRNA. Subsequently, 50 nM synthetic gRNA and 50 nM SpCas9 protein were hybridized in NEB r3.1 buffer at 25°C for 10 minutes (prepared as a master mix). 5 μL of 30 nM DNA substrate was added to the prepared SpCas9-gRNA complex at 37°C for 1 hour. The reaction was terminated by adding RNase A, followed by treatment at 37°C for 10 minutes and 60°C for 15 minutes. The reaction product was further purified using Oligo Clean & Concentrator Kits (Zymo Research). In vitro SpCas9 testing using synthetic crRNA oxidized by the Fenton reaction was also performed under the same conditions.

[0132] In the experiment using endogenous crRNA, SpCas9 protein (30 nM), endogenous small RNA (2.5 μg), and synthetic tracrRNA (30 nM) were mixed in NEB r3.1 buffer (25 μL) for 1 hour at 25°C (prepared as a master mix). Subsequently, 5 μL of 18 nM DNA substrate was added to the SpCas9-gRNA complex at 37°C for 4 hours. The reaction was terminated by sequential treatment at 37°C for 1 hour, 60°C for 15 minutes, and 37°C for 30 minutes after the addition of RNase Cocktail (Invitrogen).

[0133] Although the crRNA-tracrRNA complex was preserved intact even with the cold RNA extraction method, the complex's dynamic respiration (RNA breathing) significantly increased during subsequent incubation at 25°C (SpCas9 loading) and 37°C (cleavage reaction), leading to partial dissociation and reduced activity in in vitro SpCas9 tests. To prevent this, additional tracrRNA was added at the same concentration as the synthetic crRNA to block cross-hybridization with other RNAs within the small RNA fraction, thereby maintaining endogenous crRNA activity and enabling efficient targeted cleavage. Considering this, 0.25 nM of synthetic crRNA was used to achieve the same targeted cleavage activity. The DNA was purified after three replications.

[0134] Cleavage efficiency was quantified and analyzed using Capillary Electrophoresis with a Fragment Analyzer (Advanced Analytical Technologies), the HS Small Fragment Kit (Agilent), and PROsize 3.0 software. When using endogenous crRNA, precision was ensured by using approximately nine times more samples for DNA cleavage measurement compared to non-cleavage DNA measurement. The cleavage rate was calculated as follows: Cleavage Rate (%) = 100 × ([Cleaved DNA] / ([Cleaved DNA] + [Non-cleavage DNA])).

[0135] 6. In vitro SpCas9 assays for gRNAs

[0136] Basically, the procedure was performed under conditions similar to those used in "5. In Vitro SpCas9 Analysis," but DNA substrates were prepared from genomic PCR. Human genomic DNA extracted from HEK293 using the Blood & Cell Culture DNA Mini Kit (QIAGEN, #13323) was used to PCR amplify regions of interest using Phusion Hot Start II DNA polymerase (Thermo Scientific), AccuPrime Pfx Supermix (Invitrogen), or Q5 High-Fidelity 2X Master Mix (NEB), and purified using the QIAquick PCR Purification Kit (QIAGEN). Primer sequences were referenced from existing literature (excluding OT4-HCN1, OT4-10, and OT4-MFAP1), designed using the Primer3 program (https: / / bioinfo.ut.ee / primer3 / ), and the amplification product size was 300-500 bp.

[0137] For the in vitro SpCas9 cleavage assay for T2, T3, and EMX1 gRNA, 100 nM SpCas9 protein (0.15 μL of 20 μM stock) and 50 nM gRNA (0.075 μL of 20 μM stock) were pretreated (mixed master mix) in 20 μL of 1.5X Cas9 Nuclease Reaction Buffer (NEB, #B0386) at 25°C for 10 minutes. The 20 μM gRNA was prepared by mixing 2 μL of 100 μM synthetic crRNA and 2.2 μL of 100 μM tracrRNA (Sigma-Aldrich, #TRACRRNA05N) in 10 μL of ultrapure water at 90°C for 4 minutes, cooling to 25°C at a rate of -0.1°C / s, and storing at 4°C. Subsequently, 10 nM DNA substrate was added to 30 μL of 1X Cas9 Nuclease Reaction Buffer and reacted at 37°C for 1 hour. Changes in gRNA concentration were indicated in the results. The reaction was terminated by adding 1 μL of Proteinase K (20 mg / mL; Roche) and treating at room temperature for 10 minutes. The reaction product DNA was purified using Oligo Clean & Concentrator kits (Zymo Research), separated by agarose gel electrophoresis, and quantified using ImageJ software.

[0138] When preparing DNA substrates for GGX20(T2), double-stranded DNA was generated by the Klenow fill-in reaction using the following synthetic DNA oligonucleotides: 200 ng each of GG-substrate (turbid base sequence), GC-substrate, CG-substrate, and CC-substrate, 200 ng of T7 primer (5'-amine-TAATACGACTCACTATAGGG-3'), 1 μL of 650 μM dNTPs, 1 μL of Klenow enzyme (NEB, #M0210), and 10 μL of 1X NEB Buffer 2 at 25°C for 20 minutes, followed by heating at 75°C for 20 minutes, and then purified using the Oligo Clean & Concentrator Kit (Zymo).

[0139] For the preparation of the radiolabeled DNA substrate, 1 μL of T4 polynucleotide kinase (NEB, #M0201) and 1.5 μL of γ-32P ATP (PerkinElmer, #NEG-502A) were reacted with 10 μL of 1X T4 Polynucleotide Kinase Reaction Buffer at 37°C for 30 minutes, after which 0.2 μL of 2 mM ATP was added and the reaction was continued at 37°C for 5 minutes. The radiolabeled DNA was purified using an Illustra MicroSpin G-25 column (GE Healthcare) according to the manufacturer's instructions.

[0140] In the in vitro Cas9 cleavage assay using a radiolabeled DNA substrate, a complex composed of 1 μL of 1 μM SpCas9 protein and 2 μL of 12.5 μM gRNA (2.5 μL of 100 μM crRNA + 3 μL of 100 μM tracrRNA, annealed at 90°C for 4 min, 70°C for 10 min, 37°C for 15 min, and 4°C for 10 min) was reacted at 25°C for 10 min in 10 μL of 2X Cas9 Nuclease Reaction Buffer (NEB), followed by reaction at 37°C for 1 hour in 20 μL of 1X Cas9 Nuclease Reaction Buffer containing 5 ng of a labeled DNA substrate. After the reaction was complete, 1 μL of RNase A (10 mg / mL; Biosesang) was added and treated at 37°C for 15 min and 60°C for 15 min. The cutting products were separated by 20% native polyacrylamide gel electrophoresis, dried, and exposed to X-ray film at -80°C for 1 day. Quantification was performed using ImageJ.

[0141] 7. Construction of crRNA-targeted sequencing library

[0142] To construct a cDNA library specific to crRNA sequences, a target sequencing library was constructed using conserved repeat sequences within the crRNA as reverse transcription primers. 1 μg of purified small RNA (<200 nt; derived from SF370 or Bruno) or 1 μg of synthetic crRNA was ligated with a 5' RNA adapter (0.385 μM) after denaturation at 70°C for 2 minutes. The ligation reaction was performed using a total volume of 26 μL containing T4 RNA Ligase, Ligase Buffer, RNase Inhibitor, and BSA, treated at 28°C for 1 hour and 65°C for 20 minutes. The linked RNA was reverse transcribed using DR-specific RT primers for Bruno (primerB, primerBL) or for SF370 (primerS, primerSL), and the conditions were inactivation at 50°C for 1 hour and 70°C for 15 minutes in a 44 μL reaction containing SuperScript III reverse transcribe, dNTPs, DTT, RNase inhibitor, and First Strand Buffer.

[0143] To insert the Illumina TruSeq index adapter sequence, the generated cDNA was first pre-amplified for 5 cycles using Q5 High-Fidelity 2X Master Mix, universal forward primers, and crRNA RT primers (98°C 30 sec, 98°C 10 sec, 60°C 30 sec, 72°C 15 sec, 72°C 10 min). The PCR products were purified using the NucleoSpin Gel and PCR Clean-Up Kit. Subsequently, to determine the optimal number of cycles, qPCR was performed to confirm the number of amplification cycles within the linear range (including SYBR Green I). Afterward, multiplex barcode generation PCR was performed using 333 nM universal forward primers and 333 nM barcode primers (including 6-mer barcodes). If necessary, PCR products of an appropriate insertion size of ~42 nt were selected using a 3% agarose gel cache via the Pippin Prep system. The final PCR products were purified using the NucleoSpin Gel and PCR Clean-Up Kit, quantified using the Qubit dsDNA HS Assay Kit, and verified using a Fragment Analyzer. The libraries were self-sequencingd on an Illumina MiniSeq system with 50–75 cycle single reads. All FASTQ files were published in the SRA database (SRP532228, SRP532225).

[0144] 8. ARP conjugation reaction with abasic RNAs

[0145] For the quantification and isolation of base-free RNA, aldehyde reaction probes (ARPs) that specifically attach biotin to base-free sites were used. Typically, 20 μg RNA was incubated in 100 μL of solution at 37°C for 1 hour with 2 mM ARP (Invitrogen, #A10550; Cayman, #10009350), 50 mM sodium acetate (pH 5.5), 2.5 mM DFOM, and 40 U recombinant RNase inhibitor (Takara, #2313A). After the reaction was complete, 350 μL of RLT buffer was added, and the RNA was purified according to the Qiagen RNeasy Mini kit RNA purification protocol. The purified RNA was used for dot blot analysis and base-free RNA affinity purification.

[0146] 9. Construction of ARP-crRNA sequencing libraries

[0147] To analyze in depth the base-free, directly induced site-specific sequencing characteristics, a crRNA-targeted sequencing library ("ARP-crRNA Seq") was constructed from affinity-purified ARP-binding small RNAs. Generally, ARP binding was performed after ligating a 5' adapter to the small RNA ("linker-AP"). A 5' adapter was ligated to 6 μg of small RNA according to "Construction of crRNA-targeted sequencing library," and after adding 700 μL of RLT buffer, the RNA was purified using the RNeasy Mini kit RNA purification system for use in the ARP binding reaction (1 μg was set aside for input sequencing library construction).

[0148] ARP-bound small RNA was reacted with 2 mM ARP and purified using the RNeasy Mini Kit, followed by affinity purification using Dynabead MyOne Streptavidin T1 and multi-step washing. Bead-based ARP-bound base-free RNA was eluted with 700 μL of Qiazol Lysis Reagent (DFOM 2.5 mM, containing 20% ​​chloroform) and finally purified using the miRNeasy Mini Kit, or purified by adding RLT buffer after competitive elution twice for 1 hour in 1X PXL buffer containing 5 mM ARP.

[0149] For some samples (Experimental Batches 2 and 5), ARP binding and affinity purification were performed ("AP-linker") prior to 5' adapter ligation, and the process was carried out in parallel with the linker-AP method. First, 10 μg of SF370 small RNA was bound to ARP and purified to affinity, after which a 5' adapter was ligated to 30 μL of ARP-bound base-free RNA purified by competitive elution (see "Construction of crRNA-targeted sequencing library").

[0150] Subsequently, according to "Construction of crRNA-targeted sequencing library," each elution RNA (linker-AP), input RNA (1 μg), and 5' adapter linker RNA (AP-linker) were reverse transcribed using Bruno / SF370-specific DR RT primers. After PCR amplification with TruSeq index adapters, the results were quantified and verified using the Qubit dsDNA HS assay kit and Fragment Analyzer. Finally, 50- or 75-cycle single-end sequencing was performed using the Illumina MiniSeq system. If PCR product resizing was required, Pippin Prep (3% agarose gel cassette) was used. All FASTQ files were registered in the SRA database (SRP532072, SRP532054).

[0151] Example 1. Confirmation of base-free crRNA modifications with reduced extratarget activity

[0152] It is presumed that evading lethal autoimmunity by reducing out-of-target sequences within the genome is involved in the biological maintenance of S. pyogenes in the presence of the CRISPR-Cas9 system. However, analysis of the S. pyogenes genome sequence (CRISPR-Cas++ database; n = 107; Fig. 1a) revealed 1,607 potential out-of-target sequences (OT, ≤6MM; expected value 568) for 171 spacer sequences derived from high-reliability crRNAs located near the Cas9 locus (< 6 kb; n = 481) (Fig. 1b, Fig. 1c). Regardless of the degree of mismatch (≤6MM), potential out-of-target sites were observed approximately 3 to 13 times more frequently than expected. Mismatches were significantly accumulated primarily in the proximal PAM region (positions 11-16), which plays a crucial role in the initiation of SpCas9 activation (Fig. 1b), and significant accumulation was also observed in the distal PAM region (position 4) (Figs. 1b, 1c). In these cases, in in vitro cleavage analysis using recombinant SpCas9 protein and synthetic guide RNA, cleavage activity ranging from 1.9% to 95.8% was observed under conditions of up to 4 mismatches (≤4MM) (Fig. 1d). In addition, further analysis of the SF370 strain from which CRISPR-Cas9 technology originated revealed that potential extratarget sites for the expressed crRNAs (crRNA1, Spyo1h_001; crRNA4, Spyo1h_004; crRNA6, Spyo1h_006; Fig. 1e) exhibited in vitro cleavage activity (1.8-6.2%) despite containing 5 mm or 6 mm of the PAM proximal region, and 4 mm or 6 mm extratarget sites were functionally verified in vitro in S. pyogenes of the Bruno strain (crRNAb) containing the same spacer sequence. Notably, no cleavage activity was observed in some potential extratarget sites containing inconsistencies in the PAM proximal region (0%).

[0153] To evaluate the activity of endogenous bacterial crRNA, the inventors optimized the biochemical conditions of an in vitro SpCas9 test containing a complete crRNA-tracrRNA complex using a small RNA fraction (<200 nt) isolated from S. pyogenes via cold RNA extraction. To minimize cross-hybridization that could occur due to RNA breathing within the crRNA-tracrRNA complex during the reaction, additional tracrRNA was added to the reaction to block the activity of other potential RNAs present in the fraction and enable corresponding target cleavage (SF370; Fig. 1f, left panel). Through this approach, the extratarget activity of the endogenous crRNA could be quantitatively measured when bound to the SpCas9 protein at concentrations exhibiting the same intratarget activity levels as synthetic crRNAs (crRNA1 and 4; SF370, crRNAb; Bruno) (Fig. 1f, Fig. 1g). Interestingly, under these conditions, endogenous crRNA showed a significant decrease in some off-target cleavage (*P < 0.05, n=3; Fig. 1h–Fig. 1k). This suggests a potentially distinct biochemical context compared to synthetic crRNA.

[0154] Indeed, the cDNA sequences of the endogenous crRNAs of the SF370 and Bruno strains differed significantly in the frequency of sequence variations compared to synthetic crRNAs (Fig. 1l, left panel). Among the variations, G>T and C>T variations were observed most frequently and were particularly prominent under aerobic culture conditions (solid line dots; Fig. 1l, left panel). This suggests oxidative RNA modification. Specifically, baseless modifications (Ø) that could be induced by stepwise oxidation to 8-oxoguanine (o8G), 5-hydroxycytosine (h5C), 8-oxoadenine (o8A), and 5-hydroxyuridine (h5U) were suspected, and it was confirmed that G and C were most frequently oxidized in RNA (o8G 3.4, h5C = 2.0, o8A = 0.81, h5U = 0.45 (mg); yeast RNA (4 g))31. In addition, baseless modification (Ø) can explain the origin of Ø>T mutations in cDNA in that it induces Ø>T mutations according to the "A-rule" (Ø-A; Fig. 11, right panel) during reverse transcription.

[0155] To confirm the presence of base-free modifications, an aldehyde reaction probe (ARP) that specifically binds biotin to base-free sites was used. Dot blot and ELISA analyses revealed base-free modifications in small RNAs of the Bruno and SF370 strains, with levels increasing under aerobic culture conditions (Fig. 1m). Additionally, a similar increase in sequence variation was observed in synthetic crRNA oxidized via the Fenton reaction; simultaneously, in vitro off-target cleavage activity decreased, which was more pronounced than the decrease in on-target activity (Fig. 1n). Furthermore, using synthetic RNA containing base-free modifications, it was confirmed that base-free modifications induce Ø>T mutations (45-76%), deletions (16-55%), and incomplete cDNA (~1%) stopped at the synthesis site during reverse transcription, while in the case of o8G, only o8G>T mutations were induced (Fig. 1o). Taken together, these results show that oxidative base-free modification of crRNA can play a role in attenuating off-target cleavage mediated by CRISPR-Cas9 in S. pyogenes.

[0156] Example 2. Confirmation of reduction in off-target cleavage within bacteriophages due to base-free oxidation at the 5' end of crRNA

[0157] As incomplete 3' ends of cDNA occurred at synthetic baseless positions (Fig. 1o), analysis of the cDNA sequence ends of endogenous crRNA (compared to synthetic crRNA without modification; SF370 and Bruno) revealed a tendency for baseless modifications to generally be present near the 5' end (positions 16-18) in endogenous crRNA (Fig. 2a). This pattern was consistently observed in a redox-dependent manner in in vitro experiments in which synthetic crRNA was oxidized via the Fenton reaction (Fig. 2b). In particular, maximum 2-nucleotide elongation of the 5' end (positions 21 and 22, PAM reference) was detected only in endogenous crRNA, suggesting the presence of a 22-nucleotide-length spacer in S. pyogenes (Fig. 2a). To clearly distinguish mutations induced by base-free modifications, the inventors developed a sequencing method ("ARP-crRNA Seq") to directly map base-free modifications in crRNA (Fig. 2c). This method involves streptavidin affinity purification (ARP-AP) of base-free RNA using ARP, followed by targeting and analyzing base-free locations within crRNA based on incomplete 3' ends and base-free-induced mutations (Ø>T and deletions) during reverse transcription (Fig. 2c). When applied to the SF370 strain, accumulation of base-free modifications was observed at the 5' end of the crRNA, reaching a peak specifically at position 19 relative to PAM. This was supported by base-free mutations such as Ø>T, deletions, and Ø>C at the 3' end of the cDNA read (Fig. 2d) (Fig. 2e).

[0158] Interestingly, infection of the host SF370 strain by Phage A1 resulted in an immediate increase in reactive oxygen species (ROS) (Fig. 2f), which was consistent with ROS production reported during lethal viral attacks. Infection cumulatively induced base-free modifications of small RNAs and crRNAs within one hour (Fig. 2g), which were inhibited by treatment with the antioxidant N-acetylcysteine ​​(NAC) (Fig. 2h). Importantly, endogenous crRNAs isolated from SF370 strains infected with Phage A1 significantly reduced the out-of-target activity of crRNA1 and crRNA4 in in vitro SpCas9 analysis (Figs. 2i, 2j; compared to mock). In-target activity was also significantly reduced by infection (Inf / mock = 0.6–0.8), but the reduction in out-of-target activity was much more pronounced than the reduction in in-target activity (Inf / mock = 0.0–0.3). Considering the basal level of endogenous base-free crRNA activity, the induced base-free crRNA dramatically reduced off-target cleavage compared to synthetic crRNA (12-24%, crRNA1; 0%, crRNA4) while significantly preserving on-target cleavage (64%, crRNA1; 84%, crRNA4) (Figs. 1h-1k; Figs. 2i, 2j).

[0159] After Phage A1 infection, the level of baseless modification at the 5' end of crRNA increased, and significant accumulation was observed particularly at position 19, which was confirmed by baseless modifications detected at the 3' end of cDNA reads (Fig. 2k) and in ARP-crRNA Seqs (Fig. 2l). In particular, Phage A1 infection also significantly increased baseless modification at position 22 of the extended crRNA (22-nt spacer; Fig. 2k). This is consistent with the 5' end-shifted o8G oxidation phenomenon reported in microRNA, and this positional bias is presumed to be due to the inherent property of radical cations within nucleic acids migrating toward the 5' end. In fact, even when randomly generated baseless RNA fragments were oxidized in an artificial transcript in which the GG sequence was removed to exclude positional bias effects, sequencing features showed baseless modifications skewed to the 5' end, suggesting that oxidation was attenuated at the 3' end (Fig. 2m).

[0160] Endogenously, an increase in insertion-deletion mutations (indels) near the off-target cleavage site (OT5b) within the SF370 genome was observed within 1 hour of Phage A1 infection (Fig. 2n, left panel), which is a well-known feature of the repair process of activated CRISPR-Cas9-mediated cleavages. Interestingly, these indels were observed more frequently under NAC treatment conditions that inhibit baseless crRNA modification (Fig. 2n, left panel). Conversely, when oxidative stress was increased by adding 100 μM H2O2 during infection, indel mutations near the off-target cleavage sites (OT5b and OT6d) were significantly reduced (Fig. 2n, right and bottom panels). Notably, substitution mutations, known to occur rarely in CRISPR-Cas9-mediated cleavages, did not change near the off-target site regardless of redox status (Fig. 2n, right panel).

[0161] In summary, ROS induced during bacteriophage infection increase the oxidation of crRNA in S. pyogenes and induce base-free modification at the 5' end of crRNA, thereby inhibiting off-target cleavage of the bacterial genome while enabling CRISPR-Cas9 to effectively cleave the invading bacteriophage (Fig. 2o). Based on this natural control mechanism observed herein, designed gRNAs were applied to VEGFA (T2 and T3) and EMX1, which have been reported to exhibit prominent off-target activity among target sites of the human genome, by in vitro oxidation (Fenton reaction). As a result, the off-target activity of the gRNA was significantly reduced, but at the cost of a significant loss of in vitro activity (Fig. 2g). Accordingly, to resolve this trade-off issue, the inventors rationally analyzed the positional role of base-free modification at the 5' end of the gRNA to achieve optimal specificity and efficiency and to overcome the positional ambiguity of in vitro oxidation.

[0162] Example 3. Confirmation of target specificity relative to cleaved gRNA by base-free substitution

[0163] In the SpCas9-gRNA (20-nt spacer, X20)-target DNA ternary structure, it was confirmed that the nucleotide template at the 5' end forms hydrogen bonds with Q926 and R661 of the SpCas9 protein at positions 18–20 relative to PAM (Fig. 3a). These interactions critically contribute to the substitution and single-strand cleavage of the target DNA via R-loop formation. In the case of cleavage gRNA (tru-gRNA; 17-nt spacer, X17), in which the 5' end complement of the gRNA is shortened to reduce off-target effects, these protein interactions are lost (Fig. 3b). However, when a base-free substitution is introduced at positions 18–20 (Ø3X17), the interaction is maintained while the potential for base pairing can be limited (Fig. 3c). In fact, in an in vitro SpCas9 assay, synthesized Ø3X17 gRNA (T2, T3, EMX1; Fig. 3d) exhibited high in-target activity (83-92%) compared to tru-gRNA (66-85%) when synthesized using highly stable dSpacer (a base-free deoxynucleotide without a hydroxyl group at the 1' position) (Fig. 2i).

[0164] To quantitatively verify in-target efficiency, a reporter system was constructed for the T2, T3, and EMX1 target sites (Fig. 3e, top panel). This reporter vector was configured to enable GFP expression upon the occurrence of an indel caused by SpCas9 cleavage, and was measured by flow cytometry in contrast to the normal expression of RFP. The analysis results showed that Ø3X17 restored reduced in-target activity compared to tru-gRNA (Emax = 76-104% vs. 59-89% for X17) and demonstrated consistently superior performance across all guide RNA concentrations.

[0165] As a result of evaluating the off-target sites of T2, T3, and EMX1 (OT2-LAMA3, OT2-24, OT2-FMN1; OT3-COMDA, OT3-20, OT3-MAX, OT3-4; OT4-HCN1, OT4-10, OT4-MFAP1), Ø3X17 reduced off-target cleavage to a degree similar to that of tru-gRNA (average cleavage rates: Ø3X17 = 31.8%, X17 = 35.8%, X20 = ​​64.6%). In reporter validation, Ø3X17 also showed consistent reduction in off-target cleavage (Emax = 0.5-6.8%), and almost complete off-target cleavage was observed, particularly in T3 (Fig. 3f). Therefore, base-free substitution is proven to be a strategy that effectively lowers off-target cleavage while preserving intra-target activity.

[0166] Example 4. Confirmation of target specificity relative to GGX20 by base-free extension

[0167] In the analysis of the SpCas9-gRNA-target DNA structure (PDB ID: 5F9R), the 5' end of gRNA 20-nt (X20) is covered by a RuvC nuclease domain, which possesses structural characteristics that initiate single-stranded target DNA cleavage (Fig. 3g). It is known that elongation of the 5' end of gRNA increases specificity through stereochemical interactions with RuvC, and GGX20 (GG addition to the 5' end of gRNA; positions 21, 22) corresponds to this. However, structural analysis (PDB ID: 5Y36) indicates that GGX20 creates a protruding and distorted R-loop, and while this distortion may further hinder the unwinding of incomplete extratarget DNA, it actually leads to a decrease in activity within the target due to additional base pair formation (Fig. 3h). It is presumed that this distortion hinders the unwinding of incomplete extratarget DNA sequences. In particular, these extended bases at positions 21 and 22 can form base pairs with the target DNA or non-target strand within the ternary structure, which causes additional steric hindrance to the RuvC domain by accommodating additional base pairs (Fig. 3h, right panel). This additional steric effect may further inhibit intra-target activity. Accordingly, in in vitro SpCas9 analysis, all intra-target sequence variants (GG, CC, GC, and CG) capable of forming base pairs with positions 21-22 of GGX20 showed reduced cleavage activity (41-61%) compared to uncorrected gRNA (X20, 65-74%), with the lowest cleavage activity (41%) observed when perfectly matched with GG on the target DNA strand (Fig. 3i).

[0168] Inspired by base-free modifications of crRNA observed in bacteriophage infection, GGX20 was designed to maintain a 22-nt length while preventing unnecessary base pair formation by substituting base-free positions 21 and 22 (ØØX20) (Fig. 3j). ØØX20 restored in-target activity in the SpCas9 test and exhibited high in-target specificity (90-104%) comparable to gRNA (X20) in T2, T3, and EMX1 (Figs. 3k-3m). On the other hand, off-target cleavage was significantly reduced (X20 average 60.9% -> ØØX20 average 39.0%), which was similar to that of GGX20 (Fig. 3n). In particular, off-target cleavage corresponding to the 5' end mismatch site, such as in OT2-PAX6, was effectively inhibited by both ØØX20 and GGX20.

[0169] Example 5. Confirmation of Improvement in SpCas9 Fidelity by Base-Free Substitution and Extension (ØXØ)

[0170] Baseless substitution (Ø3X17) and baseless extension (ØØX20) appear to be inherently conflicting strategies. This is because the former shortens base pair formation (X17), while the latter extends length (GGX20). However, baseless modification can simultaneously fuse two mechanisms: "base pair restriction" and "nucleotide length extension." The inventors devised a new ØXØ design by introducing baseless modifications at positions 18 and 19, which are frequently observed in actual S. pyogenes (Fig. 21), and extending them to position 22 (Figs. 4a, 4b).

[0171] In in vitro analysis, ØXØ gRNA preserved in-target activity at levels of 78–92% (Figs. 4c, 4d) and reduced out-of-target cleavage to 16.5% (X20: 60.3%). In particular, out-of-target cleavage was completely eliminated at the EMX1 target (0.0% vs. 31.3%). In reporter analysis as well, ØXØ exhibited the lowest out-of-target activity (Emax = 0.0–3.3%) among all variants and was superior to tru-gRNA or GGX20 (Fig. 4e). Although in-target activity was slightly reduced, it outperformed both Ø3X17 and ØØX20 in terms of overall specificity.

[0172] Next, for T2, T3, and EMX1 gRNAs within the human genome (HEK293), indel / insertion mutations at in-target and out-of-target locations were analyzed using targeted sequencing in GFP-positive cells containing surrogate reporters. ØXØ induced effective indel / insertion frequencies at in-target locations, but at levels somewhat lower than those of uncorrected gRNA (T2, T3, and EMX1: ØXØ = 2.9%, 7.6%, and 28.8%; gRNA = 3.6%, 10.6%, and 36.9%; Fig. 4f). More importantly, ØXØ induced very low indel / insertion frequencies at out-of-target locations, achieving a 50-100% reduction compared to uncorrected gRNA (Figs. 4g–4i). In particular, for OT2-24 and OT4-10, a high proportion of unreported sequence variations were present at their genomic location (HEK293), making amplification by targeted sequencing impossible, resulting in extremely low detection frequencies. Overall, base-free substitutions and extensions (ØXØ) within the gRNA synergistically enhanced the target specificity of CRISPR-Cas9 and provided superior performance compared to other gRNA variants.

[0173] Example 6. Verification of enhanced target specificity of ØXØ at the genome-wide level

[0174] To extend validation to the genome-wide level, off-target sites for T2, T3, and EMX1 were analyzed using CIRCLE-seq as previously described (Fig. 5a). High-reliability off-target sites (discord score ≤ 6) were further selected based on the number of supporting reads from CRISPR-Cas9-mediated cleavages (P < 0.1 or P < 0.05), which were found to be significantly abundant compared to the background distribution estimated from sites with a discord score of 6 (see Methods). The unmodified gRNA exhibited indiscriminate off-target activity due to the high repeatability and low complexity sequences of the human genome (Figs. 5b, 5c). In contrast, ØXØ significantly reduced the number of off-target sites, achieving a 49–78% reduction for T2 (Fig. 5b) and a 6–29% reduction for T3 (Fig. 5c). Among these extratarget sites, when analyzed by targeted genome sequencing (Figs. 4g–4i), ØXØ showed a significant decrease in cleavage at 0–4.06% in T2 and 0–4.06% in T3, which was lower than that of unmodified gRNA at 0.09–11.78%. In particular, some extratarget sites showed higher cleavage frequencies than intratarget sites in T2 and T3, but after the introduction of the ØXØ modification, no extratarget site showed higher activity than intratarget cleavage, except for a specific case in T2 (including a mismatch at baseless nucleic acid positions 18 and 19). Genomic off-target cleavage levels were further evaluated using normalized read counts, and for T2, the mean log2 value at ØXØ was significantly reduced (12.0 vs. 5.6; 50 nM gRNA, 12.2 vs. 5.5; 10 nM gRNA; Fig. 5d). A similar reduction was observed for T3 (5.8 vs. 0.2; 50 nM gRNA; Fig. 5e). This reduction was consistent even when considering off-target sites identified in cells using GUIDE-seq.

[0175] In the case of EMX1 gRNA, only 5 out-of-target sites were identified via CIRCLE-seq, all of which were completely eliminated using ØXØ (read-count (RPM) 0), and in-target cleavage was almost perfectly preserved (925 vs. 949 RPM; Fig. 5f). In contrast, tru-gRNA showed only limited in-target activity (e.g., 56% for EMX1), and both tru-gRNA and GGX20 showed only partial effects in preventing out-of-target cleavage, reducing activity in only 3 of the 5 out-of-target sites (Fig. 5f). In particular, tru-gRNA and GGX20 generated 6 and 7 new out-of-target sites, respectively, most of which had discrepancies near positions 18-20, which is presumed to be due to base pair loss (tru-gRNA) or stereochemical distortion (GGX20) in that region.

[0176] Example 7. Comprehensive sequencing analysis of a random mismatch target library for ØXØ

[0177] Considering the limited sequence complexity of the human genome, a synthetic DNA target library containing all 1MMs in the T2 target was constructed. In vitro SpCas9 analysis of the library confirmed the expected depletion of intra-target and 1MM regions, which was proportional to the increase in uncleaved spike-in controls. The proportional depletion of 1MM extra-target regions was used to estimate Cas9-induced cleavage levels based on substrate abundance values ​​(log2 ratio, gRNA vs. control; Fig. 5g). The ØXØ modification significantly inhibited cleavage of 1MM extra-target sequences, which was reflected in a substantial increase in substrate abundance and a dramatic decrease in cleavage activity (gRNA vs. ØXØ: number of sequences with relative abundance < 6.5%, 31 vs. 1). In particular, compared to unmodified gRNA, ØXØ exhibited reduced cleavage activity in the PAM-distal region, specifically in sequences with a mismatch at the 5' end position modified to a base-free state.

[0178] Additionally, to investigate off-target specificity in depth, a library of random mismatch sequences up to 3 mm for T2, T3, and EMX1 was constructed by pooling 'doped' synthetic DNA (Fig. 5h). Each correct base could be substituted with another base with a 6% probability (2% for each), and designated contamination sites were substituted with selected incorrect bases with a 99% probability. The complexity and quantity of 1 mm, 2 mm, and 3 mm sequences were verified by sequencing, and the degree of depletion was measured following in vitro SpCas9 analysis with gRNA introduction. ØXØ significantly inhibited the cleavage of off-target sequences with up to 3 mm mismatch, which was evidenced by an increase in substrate abundance of the corresponding sequences compared to unmodified gRNA (Figs. 5i, 5j). Heatmap analysis showed that ØXØ reduced mismatch cleavage in both the proximal and distal regions of PAM and exhibited stronger tolerance to transversions than to transitions (Fig. 5k).

[0179] Example 8. Comparison of ØXØ and SpCas9 variants in off-target cleavage reduction

[0180] Next, the performance of ØXØ gRNA used with SpCas9 was evaluated in comparison to uncorrected gRNA and well-established SpCas9 variants (eSpCas9, SpCas9-HF1, HypaCas9, and Sniper-Cas9) (Figs. 5l, 5m). All of these variants were developed to reduce out-of-target activity. In a surrogate reporter assay (Emax > 25%) targeting potent out-of-target regions of T2 and T3 (OT2-LAMA3 and OT3-MAX), all SpCas9 variants achieved only limited inhibition of out-of-target editing (11–99% reduction) while significantly sacrificing intra-target activity (Emax > 57%). In contrast, ØXØ almost completely inhibited out-of-target editing (96–99% reduction, Figs. 5l, 5m, top panel) while maintaining excellent intra-target activity (Emax > 87%). Furthermore, a comprehensive comparison was made with commercial SpCas9 variant proteins (eSpCas9 (Sigma-Aldrich), Truecut HiFi Cas9 (Invitrogen), HiFi Cas9 (IDT), SpCas9-HF1 (NEB)) using a random mismatch library (≤3MM, Figs. 5n–5p). While some variants showed similar performance in increasing substrate abundance at the off-target site in in vitro SpCas9 assays (T2, Fig. 5n; T3, Fig. 5o), ØXØ was significantly superior to all variants in terms of inhibiting off-target cleavage, as confirmed by cumulative distribution function (CDF) analysis and Kolmogorov-Smirnov (KS) test (Fig. 5p). In addition, in the evaluation using a random mismatch library (CDF analysis), ØXØ was superior to the base-free substitution strategy (Ø3X17) and base-free extension strategy (ØØX20) in terms of inhibiting off-target cleavage (≤2MM; Fig. 5q). Overall, ØXØ was proven to be superior to several Cas9 variants in reducing off-target activity.

[0181] Example 9. Confirmation of target specificity relative to gRNA according to positional base-free substitutions

[0182] When positions 18 and 19 of gRNA were converted to dSpacer (a baseless deoxynucleotide without a hydroxyl group at the Ø, 1' position) or rSpacer (a nucleotide without a hydroxyl group at the rØ, 1' position) (Fig. 6a), thereby limiting the base pairing potential while maintaining interaction with the SpCas9 protein, the cleavage ability of the T2 target site increased (Fig. 6b), but a decrease in cleavage of non-target sites of T2 (OT2-FMN1, OT2-PAX6, OT2-24) was confirmed (Fig. 6c).

[0183] When positions 1, 2, 1,2, and 2,3 of gRNA were substituted with dSpacer or rSpacer, complete inhibition of cleavage and a significant reduction were observed in the non-target regions of T2 (OT2-FMN1, OT2-PAX6, OT2-LAMA3, OT2-24) (Fig. 6d).

[0184] When dSpacer substitutions were applied to gRNA positions 1 through 8 respectively (Fig. 6e), substitutions at gRNA positions 4, 7, and 8 of the T2 target enhanced target cleavage ability (Fig. 6f), while completely eliminating or significantly reducing cleavage of non-target regions of T2 (OT2-FMN1, OT2-LAMA3, OT2-CALY, OT2-24) (Fig. 6g).

[0185] Next, when Ø3X17 gRNA (18-20 base-free substitutions) and Ø2X18 gRNA (19-20 base-free substitutions) with two or more consecutive dSpacer substitutions applied to the gRNA terminus (Fig. 6h) were introduced into cells, reporter analysis showed high target specificity for the KU target (111-130%) (Fig. 6i). On the other hand, off-target cleavage was significantly reduced (3.0-4.9%) (Fig. 6j).

[0186] SEQ ID NO: 13 to 25 SEQ NO. Sequence Remarks 13GnnCCCCUCCACCCCCGCCUC, 6a14GACCCCCUCCACCCCCGCCnn15GACCCCCUCCACCCCCGCnnC16nnnCAACCGCAUCAGUAUUU, 6h17nnGCAACCGCAUCAGUAUUU18GACCCCCUCCACCCCCGCCUn 6e19GACCCCCUCCACCCCCGCCnC20GACCCCCUCCACCCCCGCnUC21GACCCCCUCCACCCCCGnCUC22GACCCCCUCCACCCCCnCCUC23GACCCCCUCCACCCCnGCCUC24GACCCCCUCCACCCnCGCCUC25GACCCCCUCCACCnCCGCCUC

[0187] n: baseless

[0188] Example 10. Confirmation of SpCas9 specificity by introduction of base-free substitution and extension

[0189] In order to observe the effects of introducing base-free transformation and base-free extension in various ways, in addition to ØXØ gRNA, 3X1 gRNA with base-free modifications of 18 and 20 and extension to 22, Ø4X18 gRNA with base-free modifications of 19 and 20 and extension to 22, Ø5X17 gRNA with base-free modifications of 18-20 and extension to 22, Ø3XØ gRNA with base-free modification of 19 and extension to 23, Ø3X20 gRNA with base-free extension to 23, and Ø4X19 gRNA with base-free modification of 20 and extension to 23 were designed (Figs. 7a, 7e).

[0190] When 3X1 gRNA, Ø4X18 gRNA, and Ø5X17 gRNA were used, in vitro analysis preserved intra-target activity (Fig. 7b) and, notably, completely eliminated off-target cleavage (Fig. 7c). Next, Ø3XØ gRNA, Ø3X20 gRNA, and Ø4X19 gRNA were introduced into cells, and reporter analysis confirmed that intra-target activity was preserved (Emax = 99–103%) (Fig. 7f) and off-target cleavage was inhibited (Emax = 21.4–34.8%) (Fig. 7g). Notably, the ØXØ variant exhibited the lowest off-target activity (Emax = 2.0) among all the aforementioned variant gRNAs (Fig. 7g).

[0191] CXC gRNA was designed by introducing another base-free modification, the C3 base-free (C3 spacer), at bases 18 and 19 and extending it to 22 (Fig. 7a). In vitro analysis of CXC gRNA confirmed that in-target activity was preserved (Fig. 7b) and that off-target cleavage was completely eliminated (Fig. 7c).

[0192] SEQ NO. 26 to 28 SEQ NO. Sequence Remarks 26 nnnAnCCCCUCCACCCCCGCCUC 7a 27 nnnnCCCCCUCCACCCCCGCCUC 28 nnnnnCCCCUCCACCCCCGCCUC

[0193] n: baseless

Claims

1. Guide RNA for inhibiting non-target binding of CRISPR / Cas9-based genome editing, The guide RNA is characterized by comprising one or more abasic spacers within a crRNA sequence that recognizes a target.

2. In Paragraph 1, The above guide RNA is, 1) One or more of the sequences among the above crRNA sequences that are complementary to the target DNA base are substituted with non-base spacers, or 2) A guide RNA having one or more non-basic spacers extended at the 5' end of the above crRNA sequence.

3. In Paragraph 2, The above guide RNA is, A guide RNA in which three or more sequences among the above crRNA sequences that are complementary to the target DNA bases are substituted with non-base spacers.

4. In Paragraph 3, The above guide RNA is, A guide RNA in which bases 18, 19, and 20 of the above crRNA sequence are substituted with non-base spacers.

5. In Paragraph 2, The above guide RNA is, A guide RNA having two or more non-basic spacers extended at the 5' end of the above crRNA sequence.

6. A composition for inhibiting non-target binding comprising guide RNA according to claim 1.

7. A gene editing method using guide RNA according to paragraph 1.

8. A method for constructing guide RNA for inhibiting non-target binding, The above method is, A method comprising the step of including one or more abasic spacers within a crRNA sequence that recognizes a target in the guide RNA.

9. In Paragraph 8, The above method is, 1) a step of substituting one or more of the sequences among the above crRNA sequences that are arranged complementarily to the target DNA base with non-base spacers; or 2) A method comprising the step of adding one or more non-basic spacers extending to the 5' end of the crRNA sequence.

10. In Paragraph 9, Step 1) above is, 1) A method comprising the step of substituting three or more sequences among the above crRNA sequences that are arranged complementarily to the target DNA bases with non-base spacers.

11. In Paragraph 10, Step 1) above is, A method comprising the step of substituting bases 18, 19, and 20 of the above crRNA sequence with non-base spacers.

12. In Paragraph 8, Step 2) above is, 2) A method comprising the step of adding two or more non-basic spacers extending to the 5' end of the crRNA sequence.

Images