Engineered protein

Abstract

The present invention provides an engineered protein that can be utilized as a genome engineering tool. The protein comprises the amino acid sequence represented by SEQ ID NO: 1 in which amino acid at position 188 is substituted with histidine, the amino acid sequence further including one selected from: substitution of amino acid at position 2 with tyrosine; substitution of amino acid at position 70 with tyrosine; substitution of amino acid at position 80 with arginine; substitution of amino acid at position 105 with threonine; substitution of amino acid at position 123 with histidine; substitution of amino acid at position 195 with lysine; substitution of amino acid at position 208 with arginine; substitution of amino acid at position 232 with alanine; substitution of amino acid at position 246 with methionine; substitution of amino acid at position 316 with methionine; and substitution of amino acid at position 337 with isoleucine.

Description

Engineered Proteins

[0001] The present invention relates to engineered substances, such as engineered proteins.

[0002] The CRISPR-Cas systems of bacteria and archaea are known to provide adaptive immunity against foreign nucleic acids. Patent Document 1 discloses that CRISPR-Cas systems are classified into two classes (classes 1 and 2) and six types (types I to VI). Class 2 systems include types II, V, and VI, and contain single multi-domain effector Cas proteins such as Cas9 (type II) and Cas12 (type V). As disclosed in Patent Document 1, Cas9 and Cas12a exhibit strong nuclease activity in eukaryotic cells and are therefore widely used as versatile genome engineering tools.

[0003] International Publication No. 2022 / 092317

[0004] Proteins used as genome engineering tools, such as Cas9 and Cas12, are required to have various properties. For example, SpCas9 derived from Streptococcus pyogenes is suitable as a genome engineering tool due to its strong activity. However, its molecular size is large, at 1,367 amino acid residues (the gene is 4.1 kb). On the other hand, the space available for a foreign gene in an AAV vector is less than 4.5 kb, making it difficult to even load SpCas9 onto an AAV vector and introduce it into animal cells. This size limits its use as a genome engineering tool. Therefore, there is a need to provide engineered proteins that can be used as genome engineering tools.

[0005] The present invention has been made under the above circumstances, and aims to provide an engineered substance that can be used as a genome engineering tool.

[0006] A protein according to one embodiment of the present invention has, in the amino acid sequence represented by SEQ ID NO: 1, a substitution with histidine at amino acid position 188, and further has one substitution selected from a substitution with tyrosine at amino acid position 2, a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 80, a substitution with threonine at amino acid position 105, a substitution with histidine at amino acid position 123, a substitution with lysine at amino acid position 195, a substitution with arginine at amino acid position 208, a substitution with alanine at amino acid position 232, a substitution with methionine at amino acid position 246, a substitution with methionine at amino acid position 316, or a substitution with isoleucine at amino acid position 337; the protein has a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO: 1 and has activity of cleaving a target site in DNA.

[0007] According to the present invention, it is possible to provide an engineered substance that can be used as a genome engineering tool.

[0008] A: Domain structure of AsCas12f. B: Complex structural analysis using cryo-electron microscopy. C: Structural model of the AsCas12f-sgRNA-target DNA ternary complex. D: Schematic of sgRNA and target DNA. A: Size-exclusion chromatography profile. B: Illustration of a representative cryo-electron microscopy image recorded on TitanKrios. C: Illustration of the single particle cryo-electron microscopy image processing workflow. D: Graph showing the Fourier shell correlation (FSC) curve for 3D reconstruction. E: Size-exclusion chromatography profile. F: Illustration of a representative cryo-electron microscopy image recorded on TitanKrios. G: Illustration of the single particle cryo-electron microscopy image processing workflow. H: Graph showing the Fourier shell correlation (FSC) curve for 3D reconstruction. I: Size-exclusion chromatography profile. J: Illustration of a representative cryo-electron microscopy image recorded on TitanKrios. K is an illustration of the single-particle cryo-electron microscopy image processing workflow, and L is a graph showing the Fourier shell correlation (FSC) curve for 3D reconstruction. A is an illustration of the structural comparison between AsCas12f and UnCas12f. B is an illustration of the dimer interface between AsCas12f.1 and AsCas12f.2 and the recognition site of the sgRNA scaffold. C is an illustration of the superposition of AsCas12f.1 and AsCas12f.2 based on the REC lobe. D is the dimer interface between REC.1 and REC.2. E is the dimer interface between RuvC.1 and RuvC.2. F is the recognition of Stem2 by AsCas12f.2. G is an illustration of the recognition of Stem2 by AsCas12f.1. A is a diagram of the guide RNA and target DNA recognition sites, and B is an illustration of the electrostatic surface potential of AsCas12f. C-E are diagrams of stem 1 and PK 1, stem 3, and guide RNA-target DNA heteroduplex recognition, respectively. F is a diagram comparing the structures of the TS, NTS, and RuvC active site of AsCas12f with those of UnCas12f derived from an uncultured archaeon. A is a diagram of the DMS library design for AsCas12f, and B is an overview of the DMS approach for genomic evaluation of gene editing efficiency in the context of GFP gene deletion.C is an illustration of the effect of single mutations on genome editing activity. D is an illustration of a split-GFP-reconstitution assay to evaluate genome editing activity on the VEGF gene in HEK293T cells. E is an illustration of the genome editing activity of AsCas12f mutations. F is a graph showing the time course of GFP deletion induced by AsCas12f-HKRA and AsCas12f in HEK293T cells expressing d2EGFP. A is a graph showing the correlation of the effect of single mutants on GFP deletion. B-1 is a list of single amino acid substitutions that increase editing efficiency by 20% or more compared to WT. B-2 is a list of single amino acid substitutions that increase editing efficiency by 20% or more compared to WT. C, D, and E are graphs showing the genome editing activity of double, triple, and quadruple AsCas12f mutants, respectively. A is an illustration of the sequence of the sgRNA mutant, B is an illustration showing the genome editing activity of the sgRNA mutant determined by flow cytometry, C is a graph showing the expression of sgRNA quantified by qPCR for common sequence sites (n = 3 means ± SD), D is a graph showing the time course of GFP deletion. E is a schematic illustration of wild-type sgRNA, F is a schematic illustration of sgRNA_ΔS3-5_v7. G is the structure of the wild-type sgRNA scaffold, H is an illustration of the structure of the sgRNA_ΔS3-5_v7 scaffold. I is the PK1 region of the wild-type sgRNA, J is the PK1 region of sgRNA_ΔS3-5_v7, K is the PK2 region of the wild-type sgRNA, L is an enlarged view of the stem 3 region of sgRNA_ΔS3-5_v7. A is an illustration of the target DNA and target DNA complex results using cryo-electron microscopy. B is the dimer interface of AsCas12f-YHAM, C is target DNA recognition of AsCas12f-YHAM, D is hydrophobic interaction of AsCas12f-YHAM, E is guide RNA target DNA heteroduplex recognition of AsCas12f-HKRA, F is target DNA recognition, G is an illustration of sgRNA scaffold recognition. Illustration of the results of data collection, processing, model improvement and validation. Illustration of the results of data collection, processing, model improvement and validation. A is an illustration of a self-targeting library, B is a graph showing editing frequency. C to E are graphs showing genome editing activity with various Cas.Graph A shows the results of comparing target editing efficiencies, and B is an explanatory diagram of the PAM duplex recognition results. C and D are graphs showing indel efficiencies at other target sites, etc. E is a graph showing indel efficiencies at other target sites, etc. A is a graph showing the effects of mismatches, and B is an explanatory diagram comparing the number of off-target edits. C-1 is an explanatory diagram of off-target sites identified by guide sequences. C-2 is an explanatory diagram of off-target sites identified by guide sequences. C-3 is an explanatory diagram of off-target sites identified by guide sequences. Diagrams illustrating the generation of muscular dystrophy model induced pluripotent stem cells (iPSCs) and their differentiation into cardiomyocytes (iPSC-CMs): A is an explanatory diagram of confirmation of complete deletion of the DMD exon by PCR, and B is an explanatory diagram of the genomic sequence of a DMD exon 44-deleted iPS cell line. Figure 1 is an explanatory diagram of the generation of muscular dystrophy model induced pluripotent stem cells (iPSCs) and their differentiation into cardiomyocytes (iPSC-CMs); C is an explanatory diagram of flow cytometry analysis of α-actinin-mCherry on day 14 of cardiac differentiation; D is a diagram showing Western blot analysis. E is a graph showing a comparison of luciferase activity; F is a schematic diagram of the plasmid construct; G is a graph showing the increase in luciferase activity in cells introduced with the plasmid vector. A to F are explanatory diagrams of treatment and knock-in using mutants. G to J are explanatory diagrams of treatment and knock-in using mutants. A to F are explanatory diagrams of FIX knock-in (mouse) at the Alb locus using the AsCas12f mutant; G to J are explanatory diagrams of gene transcription activation using the AsCas12f mutant. A is an explanatory diagram showing luciferase expression due to gene transcription activation in mice; B is a graph showing the ROI value.

[0033] Figure 1 illustrates the nucleic acid sequences used for structural analysis in connection with the STAR Methods.

[0034] Figure 2 illustrates the nucleic acid sequences used for genome editing in connection with the STAR Methods.

[0035] Figure 3 illustrates the primers and constructs used for genome editing in connection with the STAR Methods.

[0036] Figure 4 illustrates the primers and constructs used for genome editing in connection with the STAR Methods.

[0037] Figure 5 illustrates the primers and constructs used for genome editing in connection with the STAR Methods.1. An explanatory diagram of primers and constructs used for genome editing in relation to the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of key resources in the STAR Methods. An explanatory diagram of a schematic of genome editing in Nicotiana benthamiana using a viral vector expressing an AsCas12f mutant. (a) and (b) are explanatory diagrams showing the results of visualizing genome-edited cells. (a) is an explanatory diagram of the NGS analysis results when using the PVX-AsCas12f mutant and PVX-SpCas9, respectively, and (b) is an explanatory diagram of the CAPS analysis results. (a) is an explanatory diagram of the outline of genome editing in tomato using a viral vector expressing the AsCas12f mutant, and (b) is an explanatory diagram of the CAPS analysis results using the PVX-AsCas12f mutant.

[0009] Herein, substitution mutations in amino acid sequences may be expressed by the single-letter abbreviation of the original amino acid, followed by a one- to three-digit number for the position, followed by the single-letter abbreviation of the substituted amino acid. For example, a mutation in which aspartic acid (D) is replaced with asparagine (N) at amino acid position 1022 is represented as "D1022N," which is synonymous with "substitution of Asp at amino acid position 1022 with Asn." The -CO-NH- bond connecting amino acids is called a peptide bond. An amino acid bonded in this manner is referred to as an amino acid residue. Furthermore, "mutation" refers to a change in the starting amino acid or nucleic acid sequence. This term is intended to encompass substitutions, insertions, and deletions. In this embodiment, the engineered substance is a protein, such as engineered AsCas12f. The engineered substance may be used as a composition together with other substances, such as guide RNAs or vectors.

[0010] Bacterial and archaeal CRISPR-Cas systems provide adaptive immunity against foreign nucleic acids and are classified into two classes (classes 1 and 2) and six types (types I to VI). Class 2 systems include types II, V, and VI, and include single multi-domain effector Cas proteins such as Cas9 (type II) and Cas12 (type V). The uses of CRISPR-Cas are not limited to genome editing. Cas with lost cleavage activity functions as an RNA-dependent DNA-binding protein, and is therefore used in technologies such as regulating transcription activity and single-base editing by fusing it with various effectors. Genome editing enzymes use viral vectors such as AAV for introduction into target tissues, but there is a limit to the size of the gene that can be loaded, making it difficult to load Cas9 and Cas12a due to their size.

[0011] Cas9 binds to dual RNA guides (CRISPR RNA [crRNA] and transactivating crRNA [tracrRNA]) or single guide RNAs (sgRNAs) and cleaves double-stranded DNA (dsDNA) targets at sequences complementary to the 20-nt guide segment of the RNA guide and adjacent to an NGG (where N is any nucleotide) proto-spacer adjacent motif (PAM).

[0012] Among the various type V Cas12s, type V-A Cas12a (also known as Cpf1) binds to crRNA and cleaves dsDNA targets with a TTTV (where V is A, G, or C) PAM. Cas9 contains two nuclease domains, HNH and RuvC, that cleave the target strand (TS) and non-target strand (NTS) of dsDNA targets, respectively.

[0013] In contrast, Cas12a cleaves both the TS and NTS with a single RuvC nuclease domain. Cas9 and Cas12a exhibit potent nuclease activity in eukaryotic cells and are therefore widely used as versatile genome engineering tools.

[0014] In addition, Cas9 (type II) (SpCas9) derived from Streptococcus pyogenes and, among the various type V Cas12s, Cas12a (AsCas12a) derived from Acidaminococcus sp. are widely used as genome editing tools in human cells.

[0015] However, these are relatively large, limiting delivery by adeno-associated virus vectors, which have cargo size limitations. On the other hand, Type V-F Cas12f (AsCas12f) derived from Acidibacillus sulfuroxidans is very compact (422 amino acids) and functions in human cells. Table 1 shows a comparison of the size of AsCas12f with previously reported Cass. AsCas12f has the amino acid sequence shown in SEQ ID NO: 1, and is the smallest Cas currently reported, at 422 residues, less than half the size of Cas9, Cas12a, Cas12b, Cas12e, etc. However, the nuclease activity of wild-type AsCas12f is relatively low. Therefore, in this embodiment, two types of AsCas12f activity-enhanced (enAsCas12f) variants were developed by structural analysis using cryo-electron microscopy and mutant screening.

[0016] In particular, the enAsCas12f mutant exhibited genome editing activity in human cells comparable to or superior to that of SpCas9 and AsCas12a. Structural analysis revealed that the mutant stabilized dimer formation and enhanced interaction with nucleic acids, resulting in improved DNA cleavage activity. Furthermore, an all-in-one AAV vector co-packaged with a partner gene demonstrated efficient knock-in / knock-out activity and transcriptional activation in mice. Taken together, enAsCas12f may provide a minimal genome editing platform for in vivo gene therapy.

[0017] Cas (clustered regularly interspaced short palindromic repeats and CRISPR-associated proteins) systems, which provide adaptive immunity against mobile genetic elements in bacteria and archaea, are classified into two classes (class 1 and class 2) and six types (types I-VI). SpCas9, accompanied by dual RNA guides (CRISPR RNA [crRNA] and trans-activating crRNA [tracrRNA], or an artificially linked single-guide RNA [sgRNA]), cleaves double-stranded DNA (dsDNA) targets flanked by NGG (where N is any nucleotide) protospacer adjacent motifs (PAMs) using its HNH and RuvC nuclease domains. In contrast, AsCasl2a binds to crRNA and uses a single RuvC nuclease domain to cleave dsDNA targets with a TTTV (where V is A, G, or C) PAM.

[0018] SpCas9 and AsCasl2a exhibit robust nuclease activity in eukaryotic cells and are widely used as versatile genome engineering tools. However, the large gene size of both SpCas9 (1,368 amino acids) and AsCasl2a (1,307 amino acids) makes it difficult to efficiently package them into a single adeno-associated virus (AAV) vector, hindering their clinical application in in vivo gene therapy.

[0019] V-F type Cas12 effectors have been shown to be highly compact (400-700 amino acids) RNA-guided DNA endonucleases. Cas12f associates with dual crRNA:tracrRNA guides and cleaves target DNA containing a T-rich PAM. Previous structural studies of the 529-amino acid Cas12f (UnCas12f) from an uncultured archaea revealed that UnCas12f functions as a dimer to compensate for its small size. However, UnCas12f only cleaves DNA targets under low-salt conditions in vitro and lacks activity in human cells, limiting its application as a genome editing tool. On the other hand, as described below, the smallest Cas12Cf (AsCas12f) derived from Acidibacillus sulfuroxidans consists of only 422 amino acids and can cleave DNA targets with a TTR (R is A or G) PAM under physiological conditions in vitro. Furthermore, AsCas12f exhibits small but detectable genome editing activity in human cells. Therefore, AsCas12f is useful as a compact genome editing tool that can be packaged into a single AAV vector.

[0020] In this embodiment, we have improved the AsCas12f system for genome editing. By combining structural analysis and deep mutational scanning (DMS), we uncovered the molecular basis and identified a detailed landscape of amino acid substitutions that significantly enhance the nuclease activity of AsCas12f. The synergistic effect of these mutations and guide RNA engineering significantly improved the genome editing efficiency of AsCas12f in human cells, reaching levels comparable to or even exceeding those of SpCas9 and engineered Cas12a proteins. The compact size of AsCas12f makes it highly useful for AAV-deliverable gRNAs and partner genes such as base editors and epigenetic modifiers, making the AsCas12f system a promising genome editing platform.

[0021] <Cryo-electron microscopy structure of the AsCas12f-sgRNA-target DNA ternary complex> Figure 1-1 shows the domain structure of AsCas12f (A), and the cryo-electron microscopy map (B) of the AsCas12f-sgRNA-target DNA ternary complex. Figure 1-2 shows the structural model (C) of the AsCas12f-sgRNA-target DNA ternary complex. Figure 1-3 shows a schematic diagram of the sgRNA and target DNA (D), with the disordered region enclosed in a dashed frame, TS represents the target strand, NTS represents the non-target strand, and PK represents pseudo-binding. To distinguish between NTS and TS, NTS is indicated with an asterisk.

[0022] Figures 2-1 and 2-2 show cryo-electron microscopy analysis of AsCas12f-sgRNA-target DNA (A to D), Figures 2-3 and 2-4 show cryo-electron microscopy analysis of AsCas12f-YHAM-sgRNA_ΔS3-5_v7-target DNA (E to H), and Figures 2-5 and 2-6 show cryo-electron microscopy analysis of AsCas12f-HKRA-sgRNA_ΔS3-5_v7-target DNA complex (I to L). In detail, in Figures 2-1 to 2-6, A represents the size-exclusion chromatography profile of AsCas12f-sgRNA-target DNA, E represents the size-exclusion chromatography profile of EAsCas12f-YHAM-sgRNA_ΔS3-5_v7-target DNA, and I represents the size-exclusion chromatography profile of the AsCas12f-HKRA-sgRNA_ΔS3-5_v7-target DNA complex. The peak fractions were used for the following cryo-electron microscopy analysis. Also, in Figures 2-1 to 2-6, B, F, and J represent representative cryo-electron microscopy images recorded on a TitanKrios at 300 kV with a K3 camera. C, G, and K are illustrations of the single particle cryo-electron microscopy image processing workflow. D, H, and L are graphs showing Fourier shell correlation (FSC) curves for 3D reconstruction with a gold-standard cut-off (FSC=0.143) indicated by the black dotted line.

[0023] To understand the molecular mechanism of AsCas12f, we used cryo-electron microscopy (cryo-EM) to analyze the complex structure of AsCas12f with a 222-nucleotide (nt) sgRNA and a 38-nt dsDNA in which the DNA backbone surrounding the cleavage site was phosphorothioate-modified with TTA PAM, as shown in Figures 1-1 to 1-2A-C, 2-1 to 2-2A-D, and Figures 9 and 10, which represent the results of data collection, processing, model refinement, and validation. The complex structure was reconstructed at a total resolution of 3.1 Å (Angstroms). This structure revealed that two AsCas12f molecules (AsCas12f.1 and AsCas12f.2) and one sgRNA molecule assemble to form an asymmetric homodimer.

[0024] The Cas12f dimer has a bilobed structure with a recognition (REC) lobe and a nuclease (NUC) lobe, and the guide RNA-target DNA heteroduplex binds to a central channel between the two lobes. The REC lobe consists of the wedge (WED) and REC domains of AsCas12f.1 and AsCas12f.2 (WED.1 / WED.2 / REC.1 / REC.2), while the NUC lobe contains the RuvC and target nucleic acid binding (TNB) domains of AsCas12f.1 and AsCas12f.2 (RuvC.1 / RuvC.2 / TNB.1 / TNB.2). Furthermore, the AsCas12f-sgRNA-DNA complex structure was found to function as a dimer, similar to UnCas12f.

[0025] As shown in Figure 1-3D, the sgRNA consists of a 20-nt guide segment (G1 to C20) and a 202-nt sgRNA scaffold (A(-202) to C(-1)). The sgRNA scaffold has an unexpected architecture with two pseudoknots (PK1 and PK2) and five stems (Stem1 to Stem5), which were not predicted from its primary sequence. Stem1 contains a 6-base pair (bp) duplex (U(-192):A(-175) to G(-187):C(180)), and PK1 contains a 4-bp duplex (U(-185):A(-2) to A(-182):U(-5)).

[0026] Stem 2 contains a 16-bp duplex (from G(-171):U(-125)) with two non-canonical base pairs (G(-171):U(-125) to G(-161):U(-135)), and stem 3 consists of three base pairs (G(-118):C(-109) to C(-116):G(-111)). PK2 has two base pairs (G(-122):C(-88) and G(-121):C(-87)) connecting stems 2 and 3. Stem 4 contains a 16-bp duplex with two coaxially stacked duplexes (U(-102):A(-80) to C(-96):G(-86)). As shown in Figure 1-3D, in particular, C(-103) located between stems 3 and 4 flips out and is inserted into stem 1 and PK1, forming a base pair with G(-181), thereby forming a continuous helix consisting of stem 1 and PK1. As shown in Figure 1-3, the 5' region (A(-202):A(-194)) and stem 5 (U(-70):U(-15)) are disordered in structure, suggesting its flexibility.

[0027] In Figures 3-1 to 3-3, Figure 3-1A shows a structural comparison between AsCas12f and UnCas12f derived from an uncultured archaea (PDB ID: 7C7L). Figure 3-2B shows the dimer interface between AsCas12f.1 and AsCas12f.2 and the recognition site of the sgRNA scaffold. Figure 3-2C shows a superposition of AsCas12f.1 and AsCas12f.2 based on the REC lobe. Figure 3-3D shows the dimer interface between REC.1 and REC.2, and Figure 3-3E shows the dimer interface between RuvC.1 and RuvC.2. Figure 3-3F shows the recognition of Stem2 by AsCas12f.2, and Figure 3-3G shows the recognition of AsCas12f. 1 represents the recognition of Stem2 by

[0028] AsCas12f (422 residues) is over 100 residues smaller than UnCas12f (529 residues), while its cognate sgRNA is approximately 40 nucleotides longer than UnCas12f. As shown in Figure 3-1A, structural comparison of AsCas12f and UnCas12f revealed that, although their domain organization is similar, AsCas12f lacks the zinc finger (ZF) domain inserted between the WED and REC domains found in UnCas12f. Instead, in the AsCas12f structure, the ZF domain is replaced by PK2 and Stem3, which are not present in UnCas12f. These findings explain the smaller size of AsCas12f, which is compensated for by the longer sgRNA.

[0029] Small AsCas12f Scaffolds Associated with Large sgRNAs As shown in Figures 3-2 to 3-3B-E, AsCas12f.1 and AsCas12f.2 interact with each other primarily through hydrophobic interactions at the REC.1 / REC.2 and RuvC.1 / RuvC.2 interfaces, thereby promoting AsCas12f dimerization, as observed with UnCas12f. As shown in Figures 3-3D and 3-3E, REC.1 and REC.2 form a symmetric interface facing each other, with W43, F48, and I116 playing central roles. Meanwhile, RuvC.1 and RuvC.2 form an asymmetric interface, with each protomer adopting a different conformation. In particular, as shown in Figure 3-3E, RuvC.1 / RuvC.2 interacts with each other primarily through hydrophobic interactions at the REC.1 / REC.2 and RuvC.1 / RuvC.2 interfaces, promoting AsCas12f dimerization, as observed with UnCas12f. The α1 helix and α1-α2 loop of RuvC.1 interact with the α1 helix and α2 helix of RuvC.2. As shown in Figure 3F and 3G, in addition to protein-protein interactions, the lower and upper regions of stem 2 interact extensively with AsCas12f.1 and AsCas12f.2, respectively, suggesting that protein-nucleic acid interactions also contribute to AsCas12f dimerization.

[0030] Figure 4-1A shows the recognition site for the guide RNA and target DNA, and Figure 4-1B shows the electrostatic surface potential of AsCas12f. The sgRNA-target DNA heteroduplex and stem 1, stem 2, and PK1 regions of the sgRNA scaffold are accommodated within the positively charged groove of the AsCas12f dimer. Figure 4-2C–E show the recognition of stem 1 and PK1, stem 3, and the guide RNA-target DNA heteroduplex, respectively. Figure 4-3F shows a structural comparison of the TS, NTS, and RuvC active site of AsCas12f with UnCas12f from an uncultured archaeon (PDB ID: 7C7L). The position of the RuvC.1 active site relative to the target DNA is similar in both structures, suggesting that AsCas12f also cleaves the target DNA using the RuvC.1 domain.

[0031] As shown in Figure 4-1A and 4-1B, assembly of the sgRNA scaffold and AsCas12f is facilitated by both base-specific and non-specific interactions. As shown in Figure 4-2C, the continuous helix consisting of Stem 1 and PK 1 is recognized and accommodated primarily within the groove formed by the WED.1 and RuvC.1 domains. As shown in Figure 4-3F and 4-3G, stem 2 passes through the gap between AsCas12f.1 and AsCas12f.2 and interacts extensively with both protomers, likely enhancing dimerization as described above. As shown in Figure 4-2D, PK2 is recognized by the WED.1 and REC.1 domains through sugar-phosphate backbone interactions and further stabilized by metal ion coordination. In contrast, stems 3 and 4 are solvent-exposed and have minimal interactions with AsCas12f.

[0032] As shown in Figure 4-1B and Figure 4-2E, the heteroduplex of guide RNA and target DNA is accommodated within the positively charged central channel and recognized by AsCas12f through nonspecific base interactions. As shown in Figure 4-2E, the PAM-proximal region of the heteroduplex (G1:dC18 to G12:dC7) is primarily recognized by Cas12f.1, whereas the PAM-distal region of the heteroduplex (C13:dG6 to G18:dC1) is recognized by AsCas12f.2. Notably, as shown in Figure 4-2E, P240.2 of RuvC.2 stacks with the G18:dC1 base pair of the heteroduplex, indicating that the 18 nucleotides of the spacer sequence function as a guide segment. As shown in Figure 4-3F, structural comparison with UnCas12f revealed that the location of RuvC.1 in AsCas12f is similar to that of RuvC.1 in UnCas12f, which cleaves both the target and non-target strands. These structural observations suggest that AsCas12f.1 is responsible for cleaving the target DNA, while AsCas12f.2 plays an important role in recognizing the PAM-distal region of the heteroduplex.

[0033] Engineering AsCas12f to Improve Genome Editing Efficiency in Mammalian Cells: AsCas12f has been shown to exhibit limited genome editing activity in human cells. To expand the utility of this ultra-compact protein, we engineered AsCas12f mutants with enhanced activity. Deep mutational scanning (DMS) was performed to examine the effect of all amino acid substitutions on genome editing efficiency in HEK293T cells. Specifically, DMS was used to evaluate how mutations that replaced all 422 residues of AsCas12f with 20 different amino acids affected the activity of genome editing targeting GFP (Green Fluorescent Protein). (Genome editing efficiency was determined by the loss of GFP fluorescence.)

[0034] In Figure 5-1, A is an illustration of the AsCasl2f DMS library design, and B is a diagram outlining the DMS approach for genomic evaluation of gene editing efficiency in the context of GFP gene deletion. In Figure 5-2, C is an illustration of how every single mutation affects genome editing activity. D is an illustration of a Split-GFP-reconstitution assay to evaluate genome editing activity against the VEGF gene in HEK293T cells. Plasmid-encoded mCherry was co-infected to identify plasmid-infected cells. Also, in Figure 5-2, E is a diagram showing the genome editing activity of AsCas12f mutants carrying editing efficiency-enhancing mutations, as determined by flow cytometry (n = 4), and F is a graph showing the time course of GFP deletion induced by AsCas12f-HKRA and AsCas12f in HEK293T cells expressing d2EGFP (n = 3, mean ± SD). sgRNA_ΔS5 and sgRNA_ΔS3-5_v7 were transduced using single-copy lentiviral infection. Note that in the graph F of Figure 5-2, (a) represents the results for HKRA + ΔS3-5_v7, (b) represents the results for HKRA + ΔS5, (c) represents the results for WT + ΔS3-5_v7, and (d) represents the results for WT + ΔS5.

[0035] As shown in Figure 5-1A, a plasmid expressing both EGFP and AsCas12f was designed, and an AsCas12f library covering all 20 amino acid substitutions in the entire sequence (M1-K422) was created. As shown in Figure 5-1B, sgRNA targeting GFP was introduced into HEK293T cells, and then the AsCas12f library packaged in lentivirus was expressed at a multiplicity of infection (MOI) of less than 0.2, ensuring that the number of mutant AsCas12f1 expressed per cell did not exceed 1. Because full-length sgRNAs did not have optimal genome editing efficiency, a stem 5 deletion mutant (sgRNA_ΔS5), which exhibits enhanced genome editing in human cells, was used for screening.

[0036] Two days after infection, GFP-positive cells were selected to select lentivirus-infected cells, which were then cultured for an additional five days. As shown in Figure 5-1B, on day 7 after infection, mRNA was extracted separately from GFP-positive and -negative cells, and deep sequencing analysis was performed to identify mutations. Figure 5-2C shows the genome editing efficiency as a bar graph and concentration distribution. As genome editing efficiency increases compared to the wild type, the bar graph becomes longer and the color becomes darker in the concentration distribution. Note that in Figure 5-2C, the concentration distribution is a schematic representation, and editing efficiency is primarily represented by the bar graph. As shown in this figure, the editing efficiency of each mutation was defined as the ratio of GFP-negative read counts to total read counts and normalized to the value for wild-type AsCas12f (WT).

[0037] Figure 6-1 (A) is a graph showing the correlation of the effect of single mutants on GFP deletion, determined from replicas of independent mutation libraries. Figure 6-2 (B-1) and Figure 6-3 (B-2) are lists of single amino acid substitutions that increase editing efficiency by 20% or more compared to WT. Figure 6-4 (C) is a graph showing the genome editing activity of AsCasl2f mutants in double, triple, and quadruple configurations.

[0038] As shown in Figure 6-1A, the DMS experiment was performed in duplicate, and similar results were obtained [R2 (coefficient of determination) = ∼0.6]. As a result, as shown in Figures 6-2 and 6-3B-1 and B-2 and Figure 6-4C, more than 200 amino acid substitutions were identified that improved editing efficiency by 20% or more compared to WT. Note that in B-1 and B-2, the horizontal axis shows 208 mutations that increased genome editing efficiency by 1.2 times or more compared to the wild type, and the vertical axis represents the genome editing efficiency of each mutation when the genome editing efficiency of the wild type is set to 1. In Figures 6-2 to 6-4B-1 to E, the mutations contained in the final mutants (F48Y, I123H, S188H, D195K, D208R, V232A, E316M) are highlighted with diagonal lines.

[0039] Figures 6-4C, D, and E and Figure 5-2E and F show the results of cross-mutation and genome editing efficiency measurements. In these measurements, in contrast to Figures 6-2 and 6-3B-1 and B-2, genome editing was performed on a gene that had lost GFP fluorescence, restoring GFP fluorescence. Of these mutations, we focused on S188H and D195K. These are located near the nucleic acid in this structure and may form additional interactions to enhance target DNA binding by AsCasl2f. Therefore, by cross-mutating multiple mutations into S188H and D195K, we ultimately developed two highly active mutants (Figure 5-2E).

[0040] Next, as shown in Figure 5-2D, we developed a split-GFP reporter system linked with a frame-shift linker containing a VEGFA sequence and evaluated the genome editing activity of WT and several mutants by measuring the proportion of frame-shift GFP-positive cells. As shown in Figure 5-2E, consistent with the DMS analysis, the S188H and D195K mutants exhibited higher genome editing activity than WT. To create mutants with higher genome editing activity than the single S188H or D195K mutants, we combined the S188H or D195K substitution with other effective substitutions identified in DMS analysis to create more than 20 mutants.

[0041] As shown in Figure 5-2E and Figure 6-4C, several double mutants showed reduced genome editing activity, while S188H / V232A and D195K / V232A showed improved activity. To further improve activity, we introduced additional mutations that enhanced activity when combined with S188H or D195K to create seven triple mutants. As shown in Figure 5-2E and Figure 6-4D, the triple mutants S188H / V232A / E316M and D195K / D208R / V232A further improved genome editing activity. Finally, we combined the triple mutants S188H / V232A / E316M or D195K / D208R / V232A with other effective substitutions identified by DMS analysis to construct 29 quadruple mutants and evaluated their genome editing activity. As shown in Figures 5E and 6E, most of these mutants exhibited activity comparable to or less than that of the triple mutant, whereas the quadruple mutants F48Y / S188H / V232A / E316M and I123H / D195K / D208R / V232A further enhanced genome editing activity. Herein, the F48Y / S188H / V232A / E316M and I123H / D195K / D208R / V232A mutants are designated AsCas12f-YHAM and AsCas12f-HKRA, respectively.

[0042] Furthermore, as shown in the graph in Figure 5-2E, the proportion of frame-shift GFP-positive cells in S188H, an example in which there is no mutation at amino acid position 195 but a mutation at amino acid position 188 and no mutations at other amino acids, and D195K, an example in which there is no mutation at amino acid position 188 but a mutation at amino acid position 195 and no mutations at other amino acids, is about 18%. Therefore, the mutant in which amino acid position 188 is mutated has a significantly higher proportion of frame-shift GFP-positive cells compared to the value in the WT, which is a comparative example. In particular, S188H, in which amino acid position 188 is mutated to histidine (H), has a significantly higher proportion of frame-shift GFP-positive cells compared to the WT. Furthermore, it was shown that the mutants with a mutation at amino acid position 195 had a higher ratio of frame-shifted GFP-positive cells compared to the WT, and in particular, D195K, in which amino acid position 195 was mutated to lysine (K), had a significantly higher ratio of frame-shifted GFP-positive cells compared to the WT.

[0043] In contrast, S188A / V232A, an example of a double mutant with a mutation at amino acid position 188 and an additional mutation at amino acid position 232, had just under two times as many frame-shift GFP-positive cells as S188H, a significantly improved value, provided that there was no mutation at amino acid position 195. Furthermore, S188H / V232A / E316M, an example of a triple mutant with a mutation at amino acid position 316 in addition to mutations at amino acid positions 188 and 232, had about 2.5 times as many frame-shift GFP-positive cells as S188H, a further improved value.

[0044] Furthermore, in an example of a quadruple mutant, F48Y / S188H / V232A / E316M, which has a mutation at amino acid position 48 in addition to mutations at amino acid positions 188, 232, and 316, the number of frame-shifted GFP-positive cells is slightly three times higher than that of S188H, and this value is further improved. Furthermore, these results strongly suggest that multiple mutants, such as quintuple or sextuple mutants, containing the F48Y / S188H / V232A / E316M mutation will also have a higher proportion of frame-shifted GFP-positive cells compared to WT. These findings indicate that mutants with a mutation at amino acid position 188, particularly mutants containing only the S188H mutation, and any multiple mutants, such as double, triple, or quadruple mutants, containing at least the S188H mutation, are useful. Therefore, an important finding has been obtained that in the case of a mutant in which amino acid position 188 is mutated, the proportion of frame-shift GFP positive cells tends to be significantly improved.

[0045] Also, in Figure 6-4C, the double mutant combining S188H and D208R obtained a value about three times higher than that of WT, but the value was lower compared to other double mutants. Furthermore, as shown in Figure 6-4E, the quadruple mutant combining S188H, V232A, E316M and D208R obtained a value about six times higher than that of WT, but the value was the lowest compared to other quadruple mutants. Therefore, in double mutants, triple mutants, quadruple mutants, quintuple mutants, sextuple mutants... m-fold mutants (m is a natural number greater than or equal to 2) containing the S188H mutation, mutants that do not contain the D208R mutation have a higher proportion of frame-shift GFP-positive cells, and it is strongly predicted that even higher genome editing activity can be obtained.

[0046] Furthermore, in the case of D195K / V232A, an example of a double mutant in which there is a mutation at amino acid position 195 and an additional mutation at amino acid position 232, provided that there is no mutation at amino acid position 188, the number of frame-shift GFP-positive cells was slightly less than twice as high as that of D195K, a significant improvement. Furthermore, in the case of D195K / D208R / V232A, an example of a triple mutant in which there is a mutation at amino acid position 208 in addition to mutations at amino acid positions 195 and 232, the number of frame-shift GFP-positive cells was about 2.5 times as high as that of D195K, a further improvement. Furthermore, in an example of a quadruple mutant, I123H / D195K / D208R / V232A, which has mutations at amino acid positions 195, 232, and 208 as well as a mutation at amino acid position 123, the number of frame-shifted GFP-positive cells was slightly three times higher than in D195K, and this value was further improved. Furthermore, these results strongly suggest that multiple mutants, such as quintuple or sextuple mutants containing the I123H / D195K / D208R / V232A mutation, will also have a higher proportion of frame-shifted GFP-positive cells compared to WT. These findings demonstrate the usefulness of mutants with a mutation at amino acid position 195, particularly mutants containing only the D195K mutation, and any multiple mutants, such as double, triple, or quadruple mutants, containing at least the D195K mutation. Therefore, an important finding has been obtained that the ratio of frame-shifted GFP positive cells tends to be significantly improved in mutants with a mutation at amino acid position 195.

[0047] Thus, mutants containing at least the S188H mutation and mutants containing at least the D195K mutation are particularly useful. Furthermore, as shown in the graph in Figure 6-4C, the proportion of frame-shift GFP-positive cells in the double mutant (S188H / D195K) containing both the S188H and D195K mutations was about 25%, which is higher than the value obtained for mutants containing only the S188H mutation or the D195K mutation.

[0048] On the other hand, as shown in Figure 6-4D, in triple mutants (e.g., S188H / V232A / E316M) that do not have a mutation at amino acid position 195 and contain at least the S188H and V232A mutations, the proportion of frame-shift GFP-positive cells is approximately 40% to 50%. This value is significantly higher than the value for mutants with only the S188H mutation and the value for double mutants containing both the S188H and D195K mutations, demonstrating that these are particularly useful. Furthermore, as shown in Figure 6-4E, in the triple mutant containing S188H, V232A, and E316M mutations (S188H / V232A / E316M) and the quadruple mutant containing at least S188H, V232A, and E316M mutations (S188H / V232A / E316M / F48Y, etc.), the ratio of frame-shift GFP-positive cells was approximately 20% to 50%, indicating that these mutants are useful. In particular, S188H / V232A / E316M / I23K, S188H / V232A / E316M / F48Y, S188H / V232A / E316M / K80R, S188H / V232A / E316M / I123H, S188H / V232A / E316M / E246M, S188H / V232A / E316M / I262 The quadruple mutants S188H / V232A / E316M / E354T, S188H / V232A / E316M / D364Q, and S188H / V232A / E316M / I388N each had higher values ​​than the double mutant containing the S188H and D195K mutations, indicating that these are particularly useful.

[0049] Furthermore, as shown in Figure 6-4D, in triple mutants (e.g., D195K / D208R / V232A) that do not have a mutation at amino acid position 188 and contain at least the D195K and V232A mutations, the proportion of frame-shift GFP-positive cells is approximately 30% to 50% or higher. This value is significantly higher than the value for mutants with only the D195K mutation and the value for double mutants containing the S188H and D195K mutations, demonstrating their particular usefulness. Furthermore, as shown in Figure 6-4E, in the triple mutant containing D195K, V232A, and D208R mutations (D195K / D208R / V232A) and the quadruple mutant containing at least D195K, V232A, and D208R mutations (D195K / D208R / V232A / I123H, etc.), the ratio of frame-shift GFP-positive cells was approximately 25% to 50%, indicating that these mutants are useful.

[0050] In addition, in FIG. 6-4E, the quadruple mutant D195K / D208R / V232A / S188R containing a mutation at amino acid position 188 exhibited a rate of approximately 25% frame-shift GFP-positive cells, a value comparable to that of a double mutant containing both S188H and D195K mutations. However, as shown on the right side of FIG. 6-4E, other quadruple mutants containing the D195K / D208R / V232A mutation exhibited a rate of approximately 35% to 50% frame-shift GFP-positive cells. This indicates that quadruple mutants containing a mutation at amino acid position 195 but not at amino acid position 188, particularly quadruple mutants containing at least the D195K / D208R / V232A mutations but not at amino acid position 188, are useful. In addition to quadruple mutants, quintuple mutants and sextuple mutants containing at least the D195K / D208R / V232A mutations without a mutation at amino acid position 188 are also considered useful. The triple mutants shown in D of Figure 6-4 are all triple mutants containing a mutation at only one of amino acid positions 188 and 195, and the proportion of frame-shift GFP-positive cells in these mutants exceeds 30%. Therefore, both 1) a triple mutant containing at least the S188H / V232A mutations without a mutation at amino acid position 195, and 2) a triple mutant containing at least the D195K / V232A mutations without a mutation at amino acid position 188 are considered useful. In 1) and 2), not only triple mutants but also any multiple mutants, such as quadruple mutants and quintuple mutants, are also considered useful.

[0051] These results demonstrate that the proportion of frame-shifted GFP-positive cells is significantly improved by a protein with a mutation at amino acid position 188 in the amino acid sequence of AsCas12f. Furthermore, the proportion of frame-shifted GFP-positive cells is also significantly improved by a protein with a mutation at amino acid position 195 in the amino acid sequence of AsCas12f.

[0052] More specifically, as shown on the left side of C in Figure 6-4, in each of the double mutants S188H / I2Y to S188H / L337I, the proportion of frame-shifted GFP-positive cells is significantly higher than the value (approximately 3-4%) in the WT, which is the comparative example. In particular, although the proportion of frame-shifted GFP-positive cells in S188 / D208R is the lowest at approximately 6%, it is clearly shown to be significantly higher than that of the WT. From this, it can be said that double mutants that satisfy the condition that amino acid position 188 is mutated to histidine are useful. Furthermore, in the graph for S188H in Figure 6-4C, S188H / K80R, S188H / Y105T, S188H / I123H, S188H / D195K, S188H / V232A, S188H / E246M, S188H / E316M, and S188H / L337I all have a higher proportion of frame-shift GFP-positive cells than when only S188H is mutated. Therefore, these are shown to be particularly useful double mutants.

[0053] Next, as shown on the right side of Figure 6-4C, in each of the double mutants D195K / I22Y to D195K / L337I, the proportion of frame-shifted GFP-positive cells is significantly higher than the value (approximately 3-4%) in the WT, which is the comparative example. In particular, although the proportion of frame-shifted GFP-positive cells in D195K / Y105T is the lowest at approximately 16%, it is clearly shown to be significantly higher than that of the WT. From this, it can be said that double mutants that satisfy the condition that amino acid position 195 is mutated to lysine are useful. Furthermore, in the graph relating to D195K in Figure 6-4C, D195K / I2Y, D195K / N70Y, D195K / K80R, D195K / I123H, D195K / D208R, D195K / V232A, D195K / E246M, D195K / E316M, and D195K / L337I all showed a higher proportion of frame-shift GFP-positive cells than when only S188H was mutated. Therefore, these are shown to be particularly useful double mutants.

[0054] Next, as shown in Figure 6-4D, the proportion of frame-shifted GFP-positive cells in each of the triple mutants S188H / V232A / I123H, S188H / V232A / E246M, S188H / V232A / E316M, and S188H / V232A / L337I was approximately 35% or higher, which is significantly higher than the value in the WT (comparison example) and is comparable to or higher than the S188H / V232A double mutant. In particular, the triple mutants S188H / V232A / E246M and S188H / V232A / E316M obtained higher values ​​than the S188H / V232A double mutant, indicating the usefulness of these triple mutants. Furthermore, because such a high percentage of frame-shifted GFP-positive cells was stably obtained, it can be said that the triple mutant, which satisfies the condition of containing the S188H / V232A mutation, is particularly useful. Furthermore, as shown on the left side of Figure 6-4E, the percentage of frame-shifted GFP-positive cells was also high in all quadruple mutants, which satisfies the condition of containing the S188H / V232A mutation, indicating that not only triple mutants but also n-fold mutants (n is a natural number greater than or equal to 3) containing at least the D195K / V232A mutation are useful.

[0055] Furthermore, the proportion of frame-shifted GFP-positive cells in each of the triple mutants D195K / V232A / N70Y, D195K / V232A / D206R, and D195K / V232A / E316M was approximately 30% or higher, significantly higher than the value in the WT control, and comparable to or higher than the D195K / V232A double mutant. In particular, the triple mutants S188H / V232A / E246M and S188H / V232A / E316M yielded higher values ​​than the S188H / V232A double mutant, demonstrating the usefulness of these triple mutants. Furthermore, since such a high proportion of frame-shifted GFP-positive cells was stably obtained, it can be said that triple mutants that satisfy the condition of including the D195K / V232A mutation are particularly useful. Furthermore, as shown on the right side of Figure 6-4E, the ratio of frame-shifted GFP-positive cells was also high in all quadruple mutants that fulfilled the condition of containing the D195K / V232A mutations. This indicates that not only triple mutants but also n-fold mutants (n is a natural number of 3 or more) containing at least the D195K / V232A mutations are useful.

[0056] As shown on the left side of Figure 6-4E, in the quadruple mutant containing the S188H / V232A / E316M mutations, the proportion of frame-shifted GFP-positive cells was approximately 20% to approximately 50%, a value significantly higher than that of the WT, which is the comparative example. Also, Figure 6-4E clearly shows that, although the proportion of frame-shifted GFP-positive cells in the S188H / V232A / E316M / D206 quadruple mutant was minimal at approximately 20%, this value was significantly higher than that of the WT. The consistently high proportion of frame-shifted GFP-positive cells thus obtained indicates that the quadruple mutant, which satisfies the requirement of containing the S188H / V232A / E316M mutations, is particularly useful. Furthermore, the quadruple mutants S188H / V232A / E316M / F48Y, S188H / V232A / E316M / E246M, S188H / V232A / E316M / I262N, and S188H / V232A / E316M / I388N each showed similar or higher rates of frame-shifted GFP-positive cells than the triple mutant S188H / V232A / E316M, indicating that these quadruple mutants are particularly useful. These findings strongly suggest that not only quadruple mutants containing at least the S188H / V232A / E316M mutation, but also m-mutants (m is a natural number greater than or equal to 4) containing at least the S188H / V232A / E316M mutation are useful.

[0057] As shown on the right side of Figure 6-4E, in the quadruple mutant containing the D195K / D208R / V232A mutations, the proportion of frame-shifted GFP-positive cells was approximately 30% to approximately 50%, a value significantly higher than that of the WT, which is the comparative example. Also, Figure 6-4E clearly shows that, although the proportion of frame-shifted GFP-positive cells in the D195K / D208R / V232A / S188R quadruple mutant was minimal at approximately 30%, this value was significantly higher than that of the WT. The consistently high proportion of frame-shifted GFP-positive cells thus obtained indicates that the quadruple mutant, which satisfies the requirement of containing the D195K / D208R / V232A mutations, is particularly useful. Furthermore, the quadruple mutants D195K / D208R / V232A / I23K, D195K / D208R / V232A / K80R, D195K / D208R / V232A / Q93M, and D195K / D208R / V232A / I123H had a rate of frame-shifted GFP-positive cells comparable to or higher than that of the triple mutant D195K / D208R / V232A, indicating that these quadruple mutants are particularly useful. These findings strongly suggest that not only quadruple mutants containing D195K / D208R / V232A, but also m-mutants (m is a natural number greater than or equal to 4) containing at least the D195K / D208R / V232A mutation are useful.

[0058] In Figure 6-4E, the quadruple mutant with the combination of D195K / D208R / V232A and S188R exhibits a very high cleavage value, approximately 10 times higher than that of WT, but the value is lower compared to other quadruple mutants. Therefore, it is estimated that among quadruple, quintuple, sextuple, heptuple, ... m-fold mutants (where m is a natural number of 4 or greater) containing the D195K / D208R / V232A mutation, mutants lacking the S188R mutation will exhibit high cleavage values. Furthermore, in addition to the double, triple, and quadruple mutants shown in Figure 6-4C to E, it is estimated that triple, quadruple, quintuple, sextuple, heptuple, octuple, ... m-fold mutants (where m is a natural number of 3 or greater) that further contain amino acid substitutions from these mutants will also exhibit high cleavage values. For these mutants, if the sequence identity with the amino acid sequence shown in SEQ ID NO: 1 becomes very low, cleavage activity at the target site may be lost. However, it is presumed that mutants with a relatively high sequence identity will have a high proportion of frame-shift GFP-positive cells, similar to these mutants, and will have high cleavage activity.

[0059] In the example of C in Figure 6-4, for each of the double mutant having a substitution of at least amino acid position 188 with histidine (H) in the amino acid sequence shown in SEQ ID NO: 1 and the double mutant having a substitution of at least amino acid position 195 with lysine (K) in the amino acid sequence shown in SEQ ID NO: 1, a mutant that has cleavage activity at the target site and has a sequence identity of, for example, 90% or more, preferably 95% or more, and more preferably 98% or more with the amino acid sequence shown in SEQ ID NO: 1 is preferred. Similarly, for each of the triple mutant shown in D in Figure 6-4 and the quadruple mutant shown in E in Figure 6-4, a mutant that has cleavage activity at the target site and has a sequence identity of, for example, 90% or more, preferably 95% or more, and more preferably 98% or more with the amino acid sequence shown in SEQ ID NO: 1 and has cleavage activity at the target site is preferred.

[0060] Furthermore, as mentioned above, among double, triple, quadruple, quintuple, sextuple, ... m-fold mutants (where m is a natural number of 2 or more) containing the S188H mutation, mutants that do not contain the D208R mutation stably exhibit higher genome editing activity. Therefore, in these mutants, the cleavage activity at the target site is high even in a range with lower sequence identity with the amino acid sequence shown in SEQ ID NO: 1. Mutants with lower sequence identity than the above-mentioned mutants, for example, mutants with sequence identity of 88% or more with the amino acid sequence shown in SEQ ID NO: 1 and cleavage activity at the target site, are also considered suitable. Similarly, among quadruple, quintuple, quintuple, sextuple, heptad ... m-fold mutants (where m is a natural number of 4 or more) containing the D195K / D208R / V232A mutation, mutants without the S188R mutation are expected to exhibit higher values.

[0061] Optimization of sgRNA for AsCasl2f Figure 7-1A shows the sequence of the sgRNA mutant (sgRNA_ΔS3-5_v1-v7), in which stems 3 and 4 have been removed and the length and sequence of the linker region have been modified to preserve the original architecture of the remaining region. The linker region is enclosed in a red dotted box.

[0062] Figure 7-1B is an illustration showing the genome editing activity of sgRNA variants as determined by flow cytometry. Figure 7-1C is a graph showing sgRNA expression quantified by qPCR for a common sequence site (n = 3, ±SD). Figure 7-1D is a graph showing the time course of GFP deletion induced by various Cas enzymes in HEK293T cells expressing d2EGFP (n = 3, ±SD). Each sgRNA was introduced by single-copy lentiviral infection. GFP deletion induced by AsCas12f-HKRA with sgRNA_ΔS3-5_v7 was slightly slower than that induced by SpCas9, but was faster than the deletion rates observed with AsCas12a, enAsCas12a, and Ultra. In Figure 7-1D, (a) represents the results for WT + ΔS5, (b) represents the results for WT + ΔS3-5_v7, (c) represents the results for HKRA + ΔS5, (d) represents the results for WHKRA + ΔS3-5_v7, (e) represents the results for AsCas12a, (f) represents the results for enAsCas12a, (g) represents the results for Ultra, and (h) represents the results for SpCas9.

[0063] Figure 7-2E shows a schematic diagram of the wild-type sgRNA, and Figure 7-2F shows a schematic diagram of sgRNA_ΔS3-5_v7. Disordered regions are shown in gray. PK1, PK2, stem 1, and stem 2 form extensive interactions with the AsCas12 protein, whereas stems 3 and 4 form little interaction with the protein, and stem 5 is disordered. Therefore, sgRNA_ΔS3-5_v7 was designed by removing stems 3-5, which are shown enclosed by a dotted line. The remaining regions retain the original architecture.

[0064] Figure 7-3G shows the structure of the wild-type sgRNA scaffold, and Figure 7-3H shows the structure of the sgRNA_ΔS3-5_v7 scaffold. Disordered regions are indicated by dotted lines. Figure 7-4I shows an enlarged view of the PK1 region of the wild-type sgRNA, and Figure 7-4J shows an enlarged view of the PK1 region of sgRNA_ΔS3-5_v7. Figure 7-4K shows the PK2 region of the wild-type sgRNA, and Figure 7-4L shows an enlarged view of the stem 3 region of sgRNA_ΔS3-5_v7.

[0065] In this embodiment, stem 5 is completely disordered. It is also shown that PK1, PK2, stem 1, and stem 2 interact extensively with the AsCas12f protein, while stems 3 and 4 are solvent-exposed and barely interact with the protein.

[0066] Therefore, as shown in Figure 7-1A, we further engineered the sgRNA by truncating these domains. Seven sgRNA mutants (sgRNA_ΔS3-5_v1 to sgRNA_ΔS3-5_v7) were designed to remove stems 3 and 4, while the remaining regions retained their original structure. As a result, as shown in Figure 7-1B, all sgRNA mutants showed increased genome editing activity compared to sgRNA_ΔS5, with sgRNA_ΔS3-5_v7 being the most effective.

[0067] As shown in Figure 7-1C, the expression level of sgRNA_ΔS3-5_v7 in HEK293 cells was 4.5-fold higher than that of sgRNA_ΔS5. This indicates that shortening the sgRNA increases expression levels and enhances genome editing activity. To further clarify the effects of sgRNA modification, GFP-targeting sgRNA_ΔS5 and sgRNA_ΔS3-5_v7 were introduced via lentiviral infection with a single copy, and the GFP deletion rate was evaluated. As shown in Figure 5-2F and Figure 7-1D, sgRNA_ΔS3-5_v7 significantly improved the GFP deletion efficiency of both AsCas12f-WT and AsCas12f-HKRA compared to sgRNA_ΔS5, suggesting that sgRNA_ΔS3-5_v7 is effective in situations where expression levels are limited, such as in vivo gene delivery.

[0068] Structures of sgRNA-optimized AsCas12f mutants determined by cryo-electron microscopy. Panel A of Figure 8-1 shows structural models of the AsCas12f-sgRNA-target DNA (left), AsCas12f-YHAMsgRNA_ΔS3-5_v7-target DNA (center), and AsCas12f-HKRAsgRNA_ΔS3-5_v7-target DNA (right) complexes. Mutated residues are shown as CPK models.

[0069] Figure 8-2B shows the dimer interface of AsCas12f-YHAM, Figure 8-2C shows target DNA recognition by AsCas12f-YHAM, and Figure 8-2D shows hydrophobic interactions by AsCas12f-YHAM. Mutated residues are highlighted in red and are represented by dark areas in the figure. Figure 8-2E shows guide RNA-target DNA heteroduplex recognition by AsCas12f-HKRA, Figure 8-2F shows target DNA recognition, and Figure 8G shows sgRNA scaffold recognition. Mutated residues are highlighted in red and are represented by dark areas in the figure.

[0070] To gain mechanistic insight into the enhanced DNA cleavage activity of AsCas12f mutants, we used cryo-electron microscopy to measure the structures of the AsCas12f-YHAM-sgRNA_ΔS3-5_v7-target DNA and AsCas12f-HKRA-sgRNA_ΔS3-5_v7-target DNA complexes at a total resolution of 2.9 Å, as shown in Figure 8-1A, Figures 2-3 to 2-6E-L, and Figures 9 and 10, which are illustrations of the results of data collection, processing, model refinement, and validation. The results revealed that the overall structures of the AsCas12f-YHAM and AsCas12f-HKRA mutants were similar to those of wild-type AsCas12f, and that the mutations introduced by DMS experiments did not significantly affect the overall structure of the complexes. In addition, in FIG. 8A, the introduced mutations are displayed in cpk (displayed in different colors according to atom type).

[0071] Figures 7-2 to 7-4 (E, G, I, and K) show diagrams of the original wild-type sgRNA, while Figures 7-2 to 7-4 (F, H, J, and L) show diagrams of the modified sgRNA variant, sgRNA_ΔS3-5_v7. As shown, sgRNA_ΔS3-5_v7 consists of a 20-nt guide segment (G1 to C20) and a 96-nt sgRNA scaffold (A(-96) to C(-1)). In sgRNA_ΔS3-5_v7, stems 3, 4, and 5 have been deleted while stems 1, 2, and PK1 have been maintained, resulting in extensive interaction with AsCasl2f.

[0072] As shown in Figure 7-4I and J, in particular, C(-6), which corresponds to the inverted C(-103) in the wild-type sgRNA, base-pairs with G(-83) and connects PK1 / stem 1 to form a continuous helix. As shown in Figure 7-4K and L, the stem 3 region, which corresponds to the wild-type PK2 region, maintains G(-24):C(-8) and G(-23):C(-9) base pairs (compared to G(-122):C(-88) and G(-121):C(-87) in the wild-type), while G(-12) is inverted and stacks with Trp17. This is similar to the interaction between G(-90) and Trp17 in the wild-type. Unexpectedly, as shown in Figure 7-4 L, A(-10) forms a non-canonical base pair with U(-22), connecting the base pairs G(-24):C(-8) to G(-23):C(-9) and C(-21):G(-14) to C(-20):G(-15) in stem 3, explaining why sgRNA_ΔS3-5_v7 is the optimal sgRNA variant.

[0073] Stem 1 to stem 5 are all stem loops, and the optimal sgRNA mutants were created by adjusting the sequence and length of the linker region connecting the deleted portions. In this way, it was confirmed that deleting a portion of the sgRNA in different Cass increases genome editing activity. One reason for increased genome editing activity is thought to be that shortening the length of the sgRNA improves the transcription level of the sgRNA in introduced mammalian cells, making it easier to form a complex with Cas. In particular, the structure of AsCas12f showed that stem 4 and stem 5 hardly interact with AsCas12f, so a short sgRNA mutant was created by deleting this portion.

[0074] Figures 8-2B to 8-2G are explanatory diagrams showing the introduced mutations. The F48Y mutation, represented by Y48.1 in Figure 8-2B, is present at the dimer interface, and by becoming Y, it forms a new interaction with another molecule, strengthening dimer assembly. Figure 8-2C, ... 208R, shown as DR208.1 in Figure 8-2F and R208.2 in Figure 8-2G, forms new interactions with DNA / RNA. The formation and increased stability of these new interactions results in increased genome editing activity of the mutants.

[0075] Specifically, in the AsCas12f-YHAM structure, Tyr48.1 (F48Y.1, labeled Y48.1 in the figure), located at the dimer interface, is thought to form a hydrogen bond with the main chain of G57.2, promoting more stable dimerization, in addition to the hydrophobic interaction observed in wild-type AsCas12f. As shown in Figure 8-2C, the side chain of His188.1 (S188H.1, labeled H188.1 in the figure) is stabilized by Ile2 and interacts with the backbone phosphate group between dT19 and dC18, indicating that the S188H mutation promotes target DNA unwinding. As shown in Figure 8-2D, Met316.1 and Met316.2 (E316M.1 and E316M.2, denoted as M316.1 and M316.2 in the figure) may form additional hydrophobic interactions with Thr239 and Ala241, thereby increasing overall stability.

[0076] As shown in Figure 8-2E, in the AsCas12f-HKRA structure, His123.1 (H123.1 in the figure) and Lys195.1 (K195.1 in the figure) provide hydrophobic and electrostatic interactions with the ribose moiety of A5 and the backbone phosphate groups of A3 and A5, respectively, stabilizing the guide RNA-target DNA heteroduplex. As shown in Figure 8-2F, Arg208.1 (R208.1 in the figure) forms hydrogen bonds and hydrophobic interactions with dT19 and Ile2, respectively. This is similar to His188.1 in AsCas12f-YHAM, thereby promoting initial heteroduplex formation. Furthermore, as shown in Figure 8-2 G, Arg208.2 of another subunit (shown as R208.2 in the figure) also forms an electrostatic interaction with the phosphate backbone of G(-43) in the sgRNA scaffold, stabilizing the complex structure.

[0077] Although both AsCas12f-YHAM and AsCas12f-HKRA mutants contain the V232A mutation, our structural analysis did not provide clear insight into the contribution of the V232A mutation to the enhanced DNA cleavage activity of these mutants. These findings suggest that DMS-based engineering approaches have great potential for generating highly active mutants that cannot be predicted by structural information alone.

[0078] <Characteristics evaluation of AsCas12f mutants in human cells> Figure 11-1A is an explanatory diagram of the measurement of PAM specificity, and Figure 11-1B is a graph showing the results. While the wild type (WT) hardly causes genome editing with any PAM, the two mutants according to this embodiment are shown to cause genome editing with TTR PAM. The vertical axis of Figure 11-1B shows the editing frequency (%), and the bar graphs show the values ​​for WT (wild type), AsCas12f-YHAM, and AsCas12f-HKRA, respectively, from the left in the figure. Figure 11-2 C to E are graphs showing the genome editing activity of various Cass, SpCas9 is the most commonly used highly active Cas9, AsCas12a is the most commonly used highly active Cas12a (Cpf1), Ultra, enAsCas12a represent variants of AsCas12a, respectively. From these results, depending on the target gene, the AsCas12f mutants according to this embodiment (AsCas12f-YHAM, AsCas12f-HKRA) exhibit higher activity than these commonly used Cass, or it was found to exhibit equal or greater activity.

[0079] Figure 12-1A is a graph showing the comparative results of target editing efficiency between AsCas12f-YHAM and AsCas12f-HKRA. As shown in Figures 11-1A and 12-1A, to evaluate the PAM specificity of enAsCas12f mutants for a wide range of targets, a self-targeting library containing both sgRNA and its target sequence was developed, and indel formation induced by wild-type AsCas12f (referred to as AsCas12f for simplicity of description), AsCas12f-YHAM, and AsCas12f-HKRA was measured for 750 different spacer sequences with 16 NTTN PAMs. As shown in Figure 12-1B, deep sequencing analysis showed that the three enzymes induced indels in the NTTR PAM but not in the NTTY (Y is T or C) PAM, indicating that they all recognize the NTTR sequence as a PAM. In other words, AsCasl2f can be said to recognize TTR (R is A or G) as a PAM.

[0080] As shown in Figure 12-1B, cryo-EM structures revealed that the nucleobases of dT(-3*) and dT(-2*) in the PAM duplex form hydrophobic interactions with Tyr76.1, and the N6 of dA(-1*) forms a hydrogen bond with His72.1. Modeling suggested that the O6 of dG(-1*) forms a hydrogen bond with His72.1, providing a clear explanation for the NTTR PAM preference. As shown in Figure 11-1B, in NTTR PAMs, AsCas12f induced an average of 3.0% indels, while AsCas12f-YHAM and AsCas12f-HKRA induced averages of 40.8% and 44.7% indels, respectively.

[0081] These results revealed that the enAsCas12f mutant exhibited much higher genome editing activity than AsCas12f at various target sequences using the NTTR PAM. Next, using HEK 293T cells, the genome editing efficiency of AsCas12f and the enAsCas12f mutant (variant) was compared with that of Cas12a and Cas12a mutants (UltraCas12a and enAsCas12a) at five target sites using the NTTG PAM.

[0082] In Figure 11-2C, the bar graphs for each item (VEGFA, HEXA, etc.) represent the indel frequency (%) in AsCas12f, AsCas12f-YHAM, AsCas12f-HKRA, AsCas12a, UltraCas12a, and enAsCas12a, respectively, from left to right in the figure. At these five sites, AsCas12f, AsCas12f-YHAM, AsCas12f-HKRA, AsCas12a, UltraCas12a, and enAsCas12a generated indels at an average frequency of 8.5%, 22.3%, 14.3%, 9.0%, 11.0%, and 11.1%, respectively. These results indicate that the enAsCas12f mutant can induce indels with efficiency equal to or greater than that of Cas12a and Cas12a mutants, while AsCas12f does not induce indels with such efficiency. Also, in Figure 11-2 D and E, the bar graphs for each item show the measurement results for AsCas12f, AsCas12f-YHAM, AsCas12f-HKRA, and SpCas9, respectively, from the left in the figure.

[0083] As shown in Figure 11-2D, genome editing efficiencies were compared with those of SpCas9 at eight target sites in HEK293T cells. These eight target sites share the same spacer sequence: 5'-TTTG PAM for AsCas12f and NGG-3' PAM for SpCas9. As shown, AsCas12f, AsCas12f-YHAM, AsCas12f-HKRA, and SpCas9 generated indels at average frequencies of 1.7%, 14.8%, 16.9%, and 22.9% at the eight target sites. Furthermore, indel formation in three therapeutic targets, namely, PCSK9 and ANGPTL3 for atherosclerosis and TTR for transthyretin amyloidosis, was also measured in HEK293T cells.

[0084] As shown in Figure 11-2E, higher indel frequencies were observed with AsCas12f-YHAM and AsCas12f-HKRA than with SpCas9 at two of the three target sites. As shown in Figures 12-2C and 12-3E, similar results were obtained with other target sites and other human cell lines, such as Huh-7 hepatoma cells and HT-1080 sarcoma cells. This indicates that AsCas12f-YHAM and AsCas12f-HKRA efficiently induce indels regardless of the target site or cell line.

[0085] Next, mismatch tolerance was examined to examine the specificity of the enAsCas12f mutant. Figure 13-1A is a graph showing the effect of mismatches (n = 3) on EGFP activation mediated by AsCas12f-YHAM and AsCas12f-HKRA in a split-GFP-reframing assay using the TTR sequence. The "-" on the left side of the graph represents data obtained using a normal guide RNA that is perfectly complementary to the target DNA. "M01", "M02", etc. represent locations where mismatches were introduced between the guide RNA and the target DNA, i.e., locations where incorrect guide sequences were used to prevent base pairing. Therefore, "M01" represents a mismatch at the first position of the guide RNA, and "M01 / 02" represents a double-mismatched guide RNA with mismatches at the first and second positions of the guide RNA. In Figure 13-1A, the vertical axis represents indel efficiency (Indel efficiency%), and in each bar graph of M01, M02, M03, etc., the bar graph on the left side of the figure shows the value for AsCas12f-YHAM, and the bar graph on the right side shows the value for AsCas12f-HKRA.

[0086] As shown in Figure 13-1A, both enAsCas12f mutants showed broad tolerance to single mismatches, except in the PAM-distal region (positions 16-20), as observed with Cas12a and other Cas12fs, but showed negligible tolerance to double mismatches. Furthermore, we investigated the genome-wide specificity of AsCas12f-HKRA and SpCas9 using GUIDE-seq (genome-wide, unbiased identification of double-strand breaks by sequencing) at nine target sites. Figure 13-1B illustrates the number of off-target sites detected by guide sequencing for AsCas12f-HKRA and SpCas9 at the same spectator site as in Figure 11-2D. Figures 13-2C-1, 13-3C-2, and 13-4C-3 are all diagrams illustrating the off-target sites identified by guide sequencing.

[0087] As shown in Figure 13-1B, Figure 13-2C-1, Figure 13-3C-2, and Figure 13-4C-3, AsCas12f-HKRA and SpCas9 have similar numbers of off-target sites, and AsCas12f-HKRA tolerates mismatches in the PAM-distal 5-nt region, consistent with the mismatch experiments. These results indicate that the enAsCas12f mutant carrying sgRNA_ΔS3-5_v7 exhibits comparable or even superior genome editing activity and specificity compared to SpCas9 and Cas12a, despite its highly compact size.

[0088] <Therapeutic Potential of AsCas12f-Mediated Genome Editing> Figures 14-1 and 14-2 are diagrams illustrating the generation of muscular dystrophy model induced pluripotent stem cells (iPSCs) and their differentiation into cardiomyocytes (iPSC-CMs). Figure 14-1A shows confirmation of complete DMD exon deletion by PCR, and Figure 14-1B shows the genomic sequence of a DMD exon 44 deletion-iPSC line. Figure 14-2C shows flow cytometry analysis of α-actinin-mCherry on day 14 of cardiac differentiation, demonstrating that all DMD exon 44 deletion lines efficiently differentiated into cardiomyocytes without purification. Figure 14-2D shows Western blot analysis of AR and DMD iPSC cardiomyocytes (#2-16, 33, 66, and 84). Arrows in the diagram indicate dystrophin immunoreactive bands, and asterisks indicate nonspecific bands. Figure 14-3E is a graph showing the comparison of luciferase activity between dAsCas12f WT and dAsCas12f-HKRA (n = 3, mean ± SD). Figure 14-3F is a schematic diagram of the dAsCas12f plasmid constructs with VP64 or VPR bound to different sites. HEXA-targeting gRNA and the plasmid were transfected into Huh-7 cells stably expressing luciferase under the control of a minimal CMV promoter with a HEXA gRNA binding site. Figure 14-3G is a graph showing the increase in luciferase activity in cells transfected with the plasmid vector (n = 3, mean ± SD).

[0089] As shown in Figures 14-1 and 14-2 (A-D), induced pluripotent stem cells (iPSCs) lacking Duchenne muscular dystrophy (DMD) exon 44 (hDMDΔEx44), the major mutation in DMD, were generated and differentiated into cardiomyocytes (iPSC-CMs) to evaluate potential therapeutic applications. Dystrophin deficiency causes myopathy and cardiomyopathy, and exon 45 skipping to restore DMD protein is being investigated.

[0090] Figures 15-1 A to 15-2 G to 15-2 are explanatory diagrams of treatment using mutants and knock-in. In detail, Figures 15-1 A to 15-2 C are explanatory diagrams of treatment (iPSCs) using an AsCas12f mutant against dystrophin. Figures 15-1 D to 15-2 F are explanatory diagrams of treatment (administration to mice using AAV) using an AsCas12f mutant against transthyretin. In Figure 15-1 E, the vertical axis represents plasma transthyretin (μg / ml), and the horizontal axis represents the number of days (weeks) elapsed. Also, (a) represents the plasma transthyretin value at 3.0 x 10 vg of AsCas12f, (b) represents the plasma transthyretin value at 1.0 x 10 vg of AsCas12f, (c) represents the plasma transthyretin value at 3.0 x 10 vg of HKRA, and (d) represents the plasma transthyretin value at 1.0 x 10 vg of HKRA. In addition, in Figure 15-1F, the vertical axis represents the indel frequency (%), and the horizontal axis represents (a) 3.0 x 1011vg AsCas12f, (b) 1.0 x 1012vg AsCas12f, (c) 3.0 x 1011vg HKRA, and (d) 1.0 x 1012vg HKR. Figure 15-2G to I represent GFP / FIX knock-in (mammalian cells) at the Alb locus using the AsCas12f mutant. In Figure 15-2I, the vertical axis represents the ratio (%) of EGFP-positive cells, and in Figure 15-2J, the vertical axis represents FIX:C (%), which represents the increase in plasma factor IX activity (FIX:C).

[0091] As shown in Figure 15-1A, we used an all-in-one AAV serotype 6 vector to express AsCas12f-HKRA and an sgRNA targeting the DMD gene in WT iPSC-CM, and found that AsCas12f-HKRA efficiently reduced the amount of dystrophin protein. Furthermore, as shown in Figure 15-1B and C, AsCas12f-HKRA, when paired with an sgRNA designed to skip exon 45, partially restored dystrophin protein in hDMDΔEx44-iPSC-CM. This confirms the potential for AAV gene delivery in treating DMD.

[0092] Next, to investigate the therapeutic potential of AsCas12f in vivo, we applied AsCas12f-HKRA to genome editing in mouse liver. As shown in Figure 15-1D, we constructed an AAV vector encoding AsCas12f (or AsCas12f-HKRA) under the HCRhAAT promoter and a promoter-driven sgRNA targeting the TTR gene for transthyretin amyloidosis under the control of the U6 promoter (Figure 15-1D). Seven-week-old mice were injected with 3 x 10 vg or 1 x 10 vg of liver-tropic AAV serotype 8, and plasma transthyretin expression levels were assessed. As shown in Figure 15-1E, AsCas12f failed to regulate plasma transthyretin levels, whereas AsCas12f-HKRA reduced them in a dose-dependent manner. Eight weeks after AAV-based gene delivery, the target locus was analyzed by targeted amplicon deep sequencing, and a high editing rate (66.3%) was achieved with 1 x 10 vg of AsCas12f-HKRA (Figure 15-1F).

[0093] Next, we evaluated the in vivo knock-in efficiency of the AsCas12f system, which inserts the EGFP gene into the mAlb 3'UTR. Gene knock-in at the mAlb locus is a platform being investigated for ectopic production of therapeutic proteins, including coagulation factors and lysosomal enzymes, from the liver. Two AAV vectors were prepared: one expressing AsCas12f or AsCas12f-HKRA and an sgRNA targeting the mAlb 3'UTR, and the other providing a donor template for EGFP knock-in exclusively at the DSB via homology-directed repair (HDR) (Figure 15-2G). These two AAV vectors were injected intraperitoneally into C57BL / 6 wild-type neonatal mice, and EGFP expression in the liver was assessed by immunofluorescence microscopy 4 weeks after vector injection.

[0094] As shown in Figure 15-2, panels H and I, injection of an AAV vector carrying AsCas12f-HKRA resulted in a significant increase in EGFP-positive hepatocytes. Next, the EGFP gene in the donor vector was replaced with a coagulation factor IX (F9) cDNA carrying the Padua mutation, and the vector was injected into neonatal hemophilia B mice (F9-deficient mice). As shown in Figure 15-2, panel J, knocking in the F9 gene at the Alb locus using AsCas12f-HKRA significantly elevated plasma coagulation factor IX (FIX) activity above the therapeutic range compared to donor-only control, whereas no such increase was observed with AsCas12f. These data indicate that AsCas12f variants with optimal sgRNAQ can be used in in vivo gene therapy, for example, for hemophilia.

[0095] <Applications of compact enAsCas12f> Due to its small size, the AsCas12f gene can be packaged into an all-in-one AAV vector along with multiple sgRNAs and large partner genes, enabling its application to genome editing therapy, which was previously impossible with genome editing tools. Figure 16-1 A to F is an explanatory diagram of FIX knock-in (mouse) at the Alb locus using an AsCas12f mutant. Figure 16-2 G to J is an explanatory diagram of gene transcription activation (CRISPR activation) (mammalian cells) using an AsCas12f mutant. As shown in the figure, a significant increase in transcriptional activity is achieved by simultaneously introducing MS2.

[0096] As shown in Figure 16-1A, a single AAV vector encoding AsCas12f (or AsCas12f-HKRA) under the control of the liver-tropic Ttr promoter, donor sequence, and sgRNA operating under the control of the U6 promoter was designed to insert F9 cDNA carrying the Padua mutation into the Alb 3'UTR locus. As shown in Figure 16-1B and C, when a single AAV serotype 8 vector was injected into newborn hemophilia B mice, both plasma FIX activity (FIX:C) and antigen (FIX:Ag) were significantly increased by the AAV vector carrying AsCas12f-HKRA, but not by the AAV vector carrying AsCas12f. Consistent with these results, clotting time assessed by activated partial thrombin time (APTT) and F9 mRNA expression levels assessed by quantitative RT-PCR were significantly improved, as shown in Figure 16-1 D and E. As shown in Figure 16-1 F, amylose gel analysis of mRNA PCR fragments revealed that cDNA insertion occurred via HDR, although insertion via non-homologous end joining (NHEJ) was also observed.

[0097] Finally, as shown in Figure 16-2G, to investigate the utility of enAsCas12f in epigenome editing, a transcriptional activation assay was performed using Huh-7 cells stably expressing luciferase driven by a minimal CMV promoter with two HEXA gRNA recognition sites. To enhance the transcriptional activity of enAsCas12f, an sgRNA was created with an MS2 aptamer inserted into its stem loop. As shown in Figure 16-2G, a plasmid expressing AsCas12f-HKRA (dead AsCas12f-HKRA, hereafter referred to as dAsCas12f-HKRA), which lacks DNA cleavage activity and is conjugated with VP64 and the MS2 fusion activator (MS2-p65-HSF1), and an engineered sgRNA targeting HEXA were transfected into Huh-7 cells.

[0098] As shown in Figure 16-2 (H) and Figure 14-3 (E), dAsCas12f-HKRA in combination with an MS2 aptamer-inserted sgRNA significantly enhanced luciferase expression. As shown in Figure 14-3 (F) and (G), direct conjugation of VPRs (VP64, p65, and Rta) failed to enhance MS2-p65-HSF1-mediated transcription, whereas binding of VP64 to the terminus of dAsCas12f-HKRA was the most effective transcriptional activation. Furthermore, as shown in Figure 16-2 (I) and (J), transfection of Huh-7 cells with a single AAV serotype 6 vector encoding dAsCas12f-HKRA conjugated with JVP64, MS2-p65-HSF1, and sgRNA significantly increased luciferase expression in a dose-dependent manner. These results indicate that enAsCas12f can be used as a transcriptional activation tool.

[0099] Figure 17A is an explanatory diagram of an example in which an AAV8-type vector expressing luciferase driven by a minimal CMV promoter with two HEXA gRNA recognition sites and an AAV8-type vector having the structure shown in Figure 16-2 I were simultaneously intravenously administered to wild-type mice. This figure is a color heat map converted to grayscale, with count values ​​increasing as the count value increases from blue (50 counts) to red (200 counts). Figure 17A shows that, compared to the four AsCas12fWT mice (left), the first mouse from the left, AsCas12fHKRA (right), has a circular region with count values ​​of approximately 50 to 100, shown in dark gray, a circular region with count values ​​of approximately 100 to 170, shown in light gray within it, and a smaller region with count values ​​of approximately 170 to 200, shown in slightly darker gray within it. The second mouse from the left in the figure shows a scattered dark gray area with count values ​​of about 50 to 100 and a light gray area with count values ​​of about 100 to 170. The third and fourth mice from the left in the figure show a circular dark gray area with count values ​​of about 50 to 100, an inner light gray area with count values ​​of about 100 to 170, and a larger closed area slightly darker gray within that with count values ​​of about 170 to 200.

[0100] FIG. 17B is a graph showing the ROI values ​​(vertical axis) for WT and HKRA. As shown in the figure, luciferase activity increased approximately 10-fold in Cas12f-HKRA compared to wild-type AsCas12f. As shown in the figure, P = 0.0474, indicating statistical significance (p < 0.05). Based on these results, it is believed possible to increase the transcription of endogenous genes by approximately 10-fold. For example, by increasing the transcription of utrophin, a fetal dystrophin, it is believed possible to treat Duchenne muscular dystrophy.

[0101] FIG. 18 shows an explanatory diagram of the nucleic acid sequence used for structural analysis in this embodiment. FIG. 19 shows an explanatory diagram of the nucleic acid sequence used for structural analysis in relation to the STAR Methods. FIGs. 20-1 to 20-4 show explanatory diagrams of primers and constructs used for genome editing in relation to the STAR Methods. FIGs. 21-1 to 21-7 show explanatory diagrams of key resources in the STAR Methods. These are illustrative and do not limit the invention. In addition, in FIGs. 19 to 20-4, 3A to 3B and 3C to 3F correspond to A to B in FIG. 5-1 and C to F in FIG. 5-2, respectively; 5A to 5B and 5C to 5E correspond to A to B in FIG. 11-1 and C to E in FIG. 11-2, respectively; and 6A to 6F and 6G to 6J correspond to A to F in FIG. 15-1 and G to J in FIG. 15-2, respectively. Also, S4A to S4D, S4E to S4F, S4G to S4H, and S4I to S4L correspond to A to D, E to F, G to H, and I to L in FIG. 7-1, respectively.

[0102] As described above, we successfully generated a compact AsCas12f mutant with improved activity by applying deep mutation scanning (DMS) technology to the CRISPR-Cas effector. The DMS approach combines comprehensive protein mutagenesis and functional screening with deep sequencing, enabling the evaluation of the effects of thousands of mutations in a single experiment. One typical application of DMS is the introduction of libraries into a yeast surface display system and the evaluation of the binding affinity of ligands, including antibodies and viral glycoproteins. In this embodiment, we applied the yeast screening system to mammalian cell-based screening to evaluate genome editing activity. We constructed a library covering all 20 single amino acid substitutions at each position in the entire AsCas12f sequence (422 residues) and identified over 200 effective mutations at various positions. To efficiently explore effective combinations of these mutants, it is advantageous to adopt a structural perspective. As shown in this embodiment, in the absence of an experimentally determined structure, structural prediction models such as AlphaFold can be a useful approach for efficiently selecting mutations identified by DMS.

[0103] Next, an example of applying the AsCas12f mutant to plant genome editing is shown. The AsCas12f mutant used was I123Y / D195K / D208R / V232A. Figure 22 shows an outline of genome editing in Nicotiana benthamiana using a viral vector expressing an AsCas12f mutant. As shown in the figure, in this example, the target site was the PDS gene, and the target gene was knocked out using a viral vector (PVX) expressing the AsCas12f mutant. For each of SpCas9 and the AsCas12f mutant, Nicotiana benthamiana leaves were infected with the viral vector using the Agro-infection method, and the genome editing efficiency was compared for the inoculated leaf (IL) and its upper leaf shown in Figure 22. By performing this comparison, when the PDS gene was knocked out in all alleles, the cells turned white, and the genome-edited cells were visualized. In the figure, IL is the inoculated leaf, and the upper leaves are indicated as L1, L2, L3, L4, . . . in order from the closest to the inoculated leaf IL.

[0104] The results of visualization of genome-edited cells as described above are shown in Figures 23(a) and (b). Figure 23(a) shows the results with the PVX-AsCas12f mutant, and Figure 23(b) shows the results with PVX-SpCas9. For the upper leaf L2, the PVX-AsCas12f mutant clearly showed bleaching of the leaf cells, as shown in Figure 23(a). On the other hand, as shown in Figure 23(b), PVX-SpCas9 showed almost no bleaching observed with the naked eye. Furthermore, for the PVX-AsCas12f mutant, bleaching was clearly observed in L3 to L7, although not as pronounced as in L2. Furthermore, bleaching tended to be more pronounced the closer the leaf was to the inoculated leaf IL. Furthermore, bleaching was also observed at the tip of the leaf at L8. On the other hand, no clear bleaching was observed in any of L2 to L8 with PVX-SpCas9. Figure 24(a) shows the results of NGS analysis using the PVX-AsCas12f mutant and PVX-SpCas9, respectively. As shown in the figure, the NGS analysis results using the PVX-AsCas12f mutant showed a mutation rate of approximately 70% in both the inoculated leaf IL and the upper leaf L2, and the upper leaf L2 showed a slightly higher mutation rate than the inoculated leaf IL. On the other hand, the NGS analysis results using PVX-SpCas9 showed a high mutation rate of approximately 80% in the inoculated leaf IL, but a very low mutation rate of 2% in the upper leaf L2.

[0105] Figure 24(b) shows the results of CAPS analysis using the PVX-AsCas12f mutant and PVX-SpCas9. As shown in the figure, when the PVX-AsCas12f mutant was used, black or gray lines appeared in the line indicated by the black triangle in the figure in both the inoculated leaf IL and the upper leaves L2 to L6, indicating the presence of a mutation. On the other hand, when PVX-SpCas9 was used, a black line appeared in the inoculated leaf IL, but no line was observed in the upper leaves L2 to L6, indicating the absence of a mutation.

[0106] Thus, the use of the AsCas12f mutant demonstrates efficient genome editing in plants inoculated with a viral vector. Furthermore, genome editing efficiency was significantly higher compared to when PVX-SpCas9 was used. Next, the results of a test using the AsCas12f mutant for genome editing in tomatoes are shown. While the examples in Figures 22 to 24 used Nicotiana benthamiana, similar experiments were conducted using tomatoes in the examples shown in Figures 25(a) and 25(b) below. In this example, similar to the example using Nicotiana benthamiana in Figure 22, genome editing in tomatoes was performed using a viral vector expressing the AsCas12f mutant, and the target site was the PDS gene, and the target gene was knocked out using a viral vector (PVX) expressing the AsCas12f mutant.

[0107] Figure 25(a) shows an outline of genome editing in tomato using a viral vector expressing an AsCas12f mutant, and Figure 25(b) shows the results of CAPS analysis using a PVX-AsCas12f mutant. As shown in the figure, by using the PVX-AsCas12f mutant, a gray line appears in the line indicated by the black triangle in the figure in the inoculated leaf IL, indicating that a mutation has been introduced. In the upper leaves L1 and L2, a gray line appears, although it is thinner than in the inoculated leaf IL, indicating that a mutation has been introduced.

[0108] Furthermore, there is no significant difference in the line density between the upper leaves L1 and L2. Therefore, the genome editing efficiency is comparable between the upper leaves L1 and L2, and no sudden drop in genome editing efficiency was observed. Furthermore, as in the example of Nicotiana benthamiana shown in FIG. 24, it is believed that leaves even higher than the upper leaf L2 also achieve genome editing efficiencies not significantly different from those of the upper leaves L1 and L2. As described above, the AsCas12f mutant is also useful for improving genome editing efficiency in plant cells. Furthermore, for this reason, the AsCas12f mutant can be applied not only to improving genome editing efficiency in mammalian cells, but also to improving genome editing efficiency in any animal or plant (or organism) in which cleavage activity can be obtained.

[0109] The present invention discloses the following aspects: In a first aspect, the protein has an amino acid sequence represented by SEQ ID NO: 1, which has a substitution with histidine at amino acid position 188, and further has one substitution selected from the group consisting of a substitution with tyrosine at amino acid position 2, a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 80, a substitution with threonine at amino acid position 105, a substitution with histidine at amino acid position 123, a substitution with lysine at amino acid position 195, a substitution with arginine at amino acid position 208, a substitution with alanine at amino acid position 232, a substitution with methionine at amino acid position 246, a substitution with methionine at amino acid position 316, and a substitution with isoleucine at amino acid position 337, which has 90% or more sequence identity with the amino acid sequence represented by SEQ ID NO: 1 and has DNA cleavage activity at a target site. In the first embodiment, the protein is more preferably a double mutant containing a substitution with histidine at amino acid position 188 and one substitution selected as described above. In the second embodiment, the protein is the protein of the first embodiment in which no substitution is made at amino acid position 208.

[0110] A third aspect is a protein having, in the amino acid sequence represented by SEQ ID NO: 1, a substitution with histidine at amino acid position 188 and a substitution with alanine at amino acid position 232, and further having one substitution selected from a substitution with histidine at amino acid position 123, a substitution with methionine at amino acid position 246, a substitution with methionine at amino acid position 316, and a substitution with isoleucine at amino acid position 337, the protein having 90% or more sequence identity with the amino acid sequence represented by SEQ ID NO: 1 and having cleavage activity against a target site in DNA. In the third aspect, the protein is more preferably a triple mutant having a substitution with histidine at amino acid position 188, a substitution with alanine at amino acid position 232, and one substitution selected as described above.

[0111] The fourth embodiment is an amino acid sequence represented by SEQ ID NO: 1, which has a substitution with histidine at amino acid position 188, a substitution with alanine at amino acid position 232, and a substitution with methionine at amino acid position 316, and further has a substitution with tyrosine at amino acid position 2, a substitution with lysine at amino acid position 23, a substitution with tyrosine at amino acid position 48, a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 80, a substitution with methionine at amino acid position 93, a substitution with threonine at amino acid position 105, and a substitution with arginine at amino acid position 110. a substitution with histidine at amino acid position 123, an substitution with arginine at amino acid position 208, a substitution with methionine at amino acid position 246, an substitution with asparagine at amino acid position 262, an substitution with threonine at amino acid position 354, an substitution with glutamine at amino acid position 364, or an substitution with asparagine at amino acid position 388, and the protein has 90% or more sequence identity with the amino acid sequence represented by SEQ ID NO: 1 and has cleavage activity against a target site in DNA. In a fourth aspect, the protein is more preferably a quadruple mutant containing a substitution with histidine at amino acid position 188, an substitution with alanine at amino acid position 232, an substitution with methionine at amino acid position 316, and one of the substitutions selected as described above.

[0112] A fifth aspect is a protein having, in the amino acid sequence represented by SEQ ID NO: 1, a lysine substitution at amino acid position 195, and one substitution selected from the group consisting of a tyrosine substitution at amino acid position 2, a tyrosine substitution at amino acid position 70, an arginine substitution at amino acid position 80, a threonine substitution at amino acid position 105, a histidine substitution at amino acid position 123, an arginine substitution at amino acid position 208, an alanine substitution at amino acid position 232, a methionine substitution at amino acid position 246, a methionine substitution at amino acid position 316, and an isoleucine substitution at amino acid position 337, the protein having 90% or more sequence identity with the amino acid sequence represented by SEQ ID NO: 1 and having DNA cleavage activity at a target site. In the fifth aspect, a double mutant containing a lysine substitution at amino acid position 195 and one of the substitutions selected as described above is more preferred.

[0113] A sixth aspect is a protein having, in the amino acid sequence represented by SEQ ID NO: 1, a substitution with lysine at amino acid position 195 and a substitution with alanine at amino acid position 232, and further having one substitution selected from a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 208, and a substitution with methionine at amino acid position 316, the protein having 90% or more sequence identity with the amino acid sequence represented by SEQ ID NO: 1 and having cleavage activity against a target site in DNA. In the sixth aspect, the protein is more preferably a triple mutant having a substitution with lysine at amino acid position 195, a substitution with alanine at amino acid position 232, and one substitution selected as described above.

[0114] The seventh embodiment is an amino acid sequence represented by SEQ ID NO: 1, which has a substitution with lysine at amino acid position 195, a substitution with arginine at amino acid position 208, and a substitution with alanine at amino acid position 232, and further has a substitution with tyrosine at amino acid position 2, a substitution with lysine at amino acid position 23, a substitution with tyrosine at amino acid position 48, a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 80, a substitution with methionine at amino acid position 93, a substitution with threonine at amino acid position 105, a substitution with amino acid The protein has one substitution selected from the group consisting of a substitution with histidine at amino acid position 123, a substitution with arginine at amino acid position 188, a substitution with methionine at amino acid position 246, a substitution with asparagine at amino acid position 262, a substitution with threonine at amino acid position 354, a substitution with glutamine at amino acid position 364, and a substitution with asparagine at amino acid position 388, and has a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO: 1 and has cleavage activity against a target site in DNA.

[0115] In a seventh aspect, the protein is more preferably a quadruple mutant comprising a substitution with lysine at amino acid position 195, a substitution with arginine at amino acid position 208, a substitution with alanine at amino acid position 232, and one substitution selected as described above. In an eighth aspect, the protein is a protein according to any one of the first aspect, or any one of the third to seventh aspects, wherein the sequence identity is 95% or more. In a ninth aspect, the protein is a protein according to any one of the first aspect, or any one of the third to seventh aspects, wherein the sequence identity is 98% or more.

[0116] The amino acid sequence of AsCasl2f is as shown in SEQ ID NO: 1 in the attached sequence listing, and is abbreviated as follows in single-letter notation: MIKVYRYEIVKPLDLDWKEFGTILRQLQQETRFALNKATQLAWEWMGFSSDYKDNHGEYPKSKDILGYTNVHGYAYHTIKTKAYRLNSGNLSQTIKRATDRFKAYQKEILRGDMSIPSYKRDIPLDLIKENISVNRMNHGDYIASLSLLSNPAKQEMNVKRKISVIIIVRGAGKTIMDRILSGEYQVSASQIIHDDRKNKWYLNISYDFEP QTRVLDLNKIMGIDLGVAVAVYMAFQHTPARYKLEGGEIENFRRQVESRRISMLRQGKYAGGARGGHGRDKRIKPIEQLRDKIANFRDTTNHRYSRYIVDMAIKEG CGTIQMEDLTNIRDIGSRFLQNWTYYDLQQKIIYKAEEAGIKVIKIDPQYTSQRCSECGNIDSGNRIGQAIFKCRACGYEANADYNAARNIAIPNIDKIIAESIK*

Claims

1. A protein having an amino acid sequence represented by SEQ ID NO:1, which has a substitution of histidine at amino acid position 188, and further has one substitution selected from the following: a substitution of tyrosine at amino acid position 2, a substitution of tyrosine at amino acid position 70, a substitution of arginine at amino acid position 80, a substitution of threonine at amino acid position 105, a substitution of histidine at amino acid position 123, a substitution of lysine at amino acid position 195, a substitution of arginine at amino acid position 208, a substitution of alanine at amino acid position 232, a substitution of methionine at amino acid position 246, a substitution of methionine at amino acid position 316, or a substitution of isoleucine at amino acid position 337; said protein having a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO:1, and having cleavage activity against a target site in DNA.

2. The protein according to claim 1, wherein amino acid position 208 is not substituted.

3. A protein having, in the amino acid sequence shown in SEQ ID NO:1, a substitution with histidine at amino acid position 188 and a substitution with alanine at amino acid position 232, and further having one substitution selected from the following: a substitution with histidine at amino acid position 123, a substitution with methionine at amino acid position 246, a substitution with methionine at amino acid position 316, or a substitution with isoleucine at amino acid position 337; said protein having a sequence identity of 90% or more with the amino acid sequence shown in SEQ ID NO:1; and having cleavage activity against a target site in DNA.

4. A peptide having an amino acid sequence represented by SEQ ID NO:1, which has a substitution at amino acid position 188 with histidine, a substitution at amino acid position 232 with alanine, and a substitution at amino acid position 316 with methionine, and further has a substitution at amino acid position 2 with tyrosine, a substitution at amino acid position 23 with lysine, a substitution at amino acid position 48 with tyrosine, a substitution at amino acid position 70 with tyrosine, a substitution at amino acid position 80 with arginine, a substitution at amino acid position 93 with methionine, a substitution at amino acid position 105 with threonine, a substitution at amino acid position 123 with histidine, a substitution at amino acid position 208 with arginine, a substitution at amino acid position 246 with methionine, a substitution at amino acid position 262 with asparagine, a substitution at amino acid position 354 with threonine, a substitution at amino acid position 364 with glutamine, or a substitution at amino acid position 388 with asparagine, a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO:1, and having cleavage activity against a target site in DNA.

5. A protein having an amino acid sequence represented by SEQ ID NO:1, which has a substitution with lysine at amino acid position 195, and further has one substitution selected from the following: a substitution with tyrosine at amino acid position 2, a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 80, a substitution with threonine at amino acid position 105, a substitution with histidine at amino acid position 123, a substitution with arginine at amino acid position 208, a substitution with alanine at amino acid position 232, a substitution with methionine at amino acid position 246, a substitution with methionine at amino acid position 316, or a substitution with isoleucine at amino acid position 337; said protein having a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO:1, and having cleavage activity against a target site in DNA.

6. A protein having, in the amino acid sequence shown in SEQ ID NO:1, a substitution with lysine at amino acid position 195 and a substitution with alanine at amino acid position 232, and further having one substitution selected from the following: a substitution with tyrosine at amino acid position 70, a substitution with arginine at amino acid position 208, or a substitution with methionine at amino acid position 316; said protein having a sequence identity of 90% or more with the amino acid sequence shown in SEQ ID NO:1; and having cleavage activity against a target site in DNA.

7. A peptide having an amino acid sequence represented by SEQ ID NO:1, which has a substitution at amino acid position 195 with lysine, a substitution at amino acid position 208 with arginine, and a substitution at amino acid position 232 with alanine, and further has a substitution at amino acid position 2 with tyrosine, a substitution at amino acid position 23 with lysine, a substitution at amino acid position 48 with tyrosine, a substitution at amino acid position 70 with tyrosine, a substitution at amino acid position 80 with arginine, a substitution at amino acid position 93 with methionine, a substitution at amino acid position 105 with threonine, a substitution at amino acid position 123 with histidine, a substitution at amino acid position 188 with arginine, a substitution at amino acid position 246 with methionine, a substitution at amino acid position 262 with asparagine, a substitution at amino acid position 354 with threonine, a substitution at amino acid position 364 with glutamine, or a substitution at amino acid position 388 with asparagine, a sequence identity of 90% or more with the amino acid sequence represented by SEQ ID NO:1, and having cleavage activity against a target site in DNA.

8. The protein according to any one of claims 1, 3 to 7, wherein the sequence identity is 95% or more.

9. The protein according to any one of claims 1, 3 to 7, wherein the sequence identity is 98% or more.

Images