Compact genome editor and base editor based on IscB system and application of compact genome editor and base editor

Abstract

The invention provides a genome editor enDelIscB and a genome editor enDelIscB-T5E based on an IscB system, and further provides a base editor ICBE and a base editor IABE based on the enDelIscBD60A. According to the invention, a genome editor and a base editor which are more compact in size and higher in cleavage activity and insertion / deletion efficiency are realized by optimizing a single mutation site of the activity of a DelIscB protein cleavage plasmid, an optimized variant of the DelIscB protein and an optimized sgRNA variant of the DelIscB. According to the present invention, it is verified that the albinism model is efficiently constructed by using the enDelIscB editor and the enDelIscB-T5E editor through micro-injection of the mouse embryo, such that the enDelIscB editor and the enDelIscB-T5E editor have great potential in the field of genome editing.

Description

Technical Field

[0001] The present invention relates to genome editors, and particularly to a compact genome editor and base editor based on the IscB system, and their applications in prokaryotes and eukaryotes. Background Art

[0002] Genome editing is a revolutionary biotechnology. By using genome editing technology to modify specific targets of an organism's genome, gene insertion, deletion, or replacement can be achieved efficiently and precisely, thereby changing its genetic information and phenotypic characteristics. There are mainly three types of tool enzymes for genome editing: ZFNs, TALENs, and the CRISPR / Cas system. Among them, compared with ZFNs and TALENs, the CRISPR / Cas system has a simpler design, higher editing efficiency, and lower cost, and is currently the most mainstream genome editing system.

[0003] The CRISPR / Cas system is mainly divided into six types (Ⅰ-Ⅵ). Among them, type Ⅱ CRISPR / Cas9 is the most widely used, and commonly used ones include SpCas9 from Streptococcus pyogenes and SaCas9 from Staphylococcus aureus, etc. However, the sizes of these Cas9 proteins are relatively large. SpCas9 has 1368 amino acids, and SaCas9 has 1053 amino acids, which makes it difficult for them or their derived editing tools to be efficiently delivered by single adeno-associated virus (AAV). AAV is one of the most widely used nucleic acid drug delivery vectors approved by the US Food and Drug Administration (FDA), but the loading limit is only 4.7 kb.

[0004] IscB (insertion sequences Cas9-like OrfB) is a class of RNA-guided nucleases encoded by transposons and is considered an ancestral protein of Cas9. Its size is only about 30% of that of Cas9, so it is easier to achieve the loading and delivery of a single AAV and has development potential. However, currently, only genome editing tools based on OgeuIscB have been reported. OgeuIscB is an IscB protein that can generate insertions / deletions (indels) identified in human cells, but its activity is relatively low, and its corresponding target adjacent motif (TAM) is relatively complex - NWRRNA. Therefore, exploring other types of IscB with more flexible TAM is of great significance for expanding the genome targeting range of compact genome editing tools based on the IscB system. Summary of the Invention

[0005] The object of the present invention is to provide an IscB protein with a more compact size, higher cleavage activity, and the ability to recognize a more flexible TAM, and a genome editing tool based on this protein.

[0006] The idea of the present invention is to utilize a CRISPR - associated IscB system (abbreviated as DelIscB) from the metagenome of aquatic samples in Delaware Bay. First, it was confirmed that DelIscB could efficiently cleave plasmids in Escherichia coli through plasmid interference experiments. Then, after truncating its ncRNA and removing the polyT motif, it was confirmed that it could cleave fluorescent reporter plasmids in eukaryotic cells. After further systematic engineering of the DelIscB protein and sgRNA, its cleavage activity was improved.

[0007] Based on this, the present invention provides single - mutation sites that improve the plasmid - cleavage activity of the DelIscB protein, and the sites are selected from S97R / K, V396R / K, E317R, C12R, A364R, Q340R / K, W117F, A142P, S100A, G104S, and / or S436R.

[0008] On the other hand, the present invention also provides optimized variants of the DelIscB protein, and the optimized variants are selected from DelIscB - V5 / V6 / V7 / V7.1 / V8 / V9 / V10 / V10.1 / V11, and the sequences of the optimized variants are shown as SEQ ID No.1 - 9 respectively.

[0009] Based on this, the present invention also provides optimized variants of the sgRNA of DelIscB, and the optimized variants are selected from sgRNA - V1 / V2 / V2.1 / V2.2 / V3 / V3.1 / V4 / V5, and the sequences of the optimized variants are shown as SEQ ID No.11 - 18 respectively.

[0010] The present invention provides two genome editors, and the genome editors include enDelIscB and enDelIscB - T5E, and their sequences are shown as SEQ ID No.19 - 20 respectively.

[0011] The present invention also provides two D60A base editors based on enDelIscB, and the base editors include ICBE and IABE. ICBE is obtained by fusing the cytosine deaminase hAPOBEC3A W98Y / W104A / Y130F with two uracil glycosylase inhibitors to the C - terminus of enDelIscB D60A , and the sequence of ICBE is shown as SEQ ID No.21; IABE is obtained by fusing TadA8e V106W to the C - terminus of enDelIscB D60A , and the sequence of IABE is shown as SEQ ID No.22.

[0012] The present invention also provides the genome editing of the above - mentioned genome editors in prokaryotic cells and / or eukaryotes.

[0013] The present invention also provides base editing in eukaryotes using the above base editors.

[0014] The present invention also provides gene editing in animal embryos or animals using the above genome editors.

[0015] The genome editors and / or base editors of the present invention can be applied to the development of prime editing, gene regulation, and / or epigenetic editing tools based on enDelIscB.

[0016] During the process of modifying the DelIscB protein, the present invention adopts a multiple sequence alignment (MSA) strategy to select positively charged or other advantageous amino acids for substitution to enhance the overall activity of the RNA-guided nuclease (RGN). For the truncation of the DelIscB sgRNA backbone, the present invention truncates the initially 365-nt-long ncRNA to 267 nt (sgRNA-V1) and further truncates it to 248 nt (sgRNA-V5). If the cryo-EM structures of DelIscB and the sgRNA bound to the target DNA can be resolved in the future, it is believed that the sgRNA can be further truncated to enhance its activity and enable chemical synthesis. At the same time, this also provides a reference basis for structure-based protein engineering.

[0017] Compared with the existing base editors miCBE and miABE, the editing windows of the base editors ICBE and IABE constructed in the present invention are slightly wider and even exceed the range of the guide sequence. This may be due to the characteristics of enDelIscB itself or the fusion of the deaminase at different ends of the IscB nickase. Attempting to fuse the deaminase within IscB may narrow the editing window.

[0018] The present invention efficiently constructed an albinism model by microinjecting mouse embryos using enDelIscB and enDelIscB-T5E editors, indicating that enDelIscB and enDelIscB-T5E have great potential in in vivo therapy. Considering the advantage of the small size of DelIscB, in the future, gene therapy research can be carried out by delivering enDelIscB, enDelIscB-T5E, ICBE, and IABE through adeno-associated virus (AAV) vectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the plasmid interference experiment in Example 1;

[0020] Figure 2 Representative images of Escherichia coli clones in the plasmid interference experiment in Example 1; where T, targeted; NT, non-targeted; 16 nt / 20 nt / 24 nt are the lengths of the guide sequences;

[0021] Figure 3 Predicted secondary structure of DelIscB non-coding RNA (ncRNA), and qualitative analysis of the activities of different ncRNA truncated versions using plasmid interference assays;

[0022] Figure 4 In Example 2, five amino acids of DelIscB were replaced with arginine, and these amino acids corresponded structurally to the highly active mutations of enIscB. Data are presented as the mean ± standard deviation of three independent biological replicates (the same below);

[0023] Figure 5 Multiple sequence alignment of DelIscB with its 10 homologous proteins in Example 2. Representative regions are shown, and mutagenesis candidate sites are highlighted with red or green boxes. The mutation types are shown at the top of the boxes, with red indicating mutation to arginine or lysine, and green indicating mutation to other amino acid types;

[0024] Figure 6 Amino acid residue substitutions of the DelIscB protein in Example 2. Each point represents the activity of a single variant. The dashed line indicates a 1.2-fold change relative to the wild-type activity;

[0025] Figure 7 Fold change in activity of 22 candidate mutants in Example 2 under three different reporter sequences. The dashed line indicates the cleavage activity of WT-DelIscB-T5E, i.e., 1-fold change;

[0026] Figure 8 Fold change in activity of 22 candidate mutants in Example 2 under three different reporter sequences;

[0027] Figure 9 Fold change in activity of 22 candidate mutants in Example 2 under three different reporter sequences;

[0028] Figure 10 In Example 2, the activities of each mutant combination were detected using two reporter sequences, EGFXXFP–target1 and EGFXXFP-target2;

[0029] Figure 11 In Example 3, several predicted stem-loops of sgRNA-V1 were truncated to enhance its activity. The dashed line indicates the cleavage activity of WT-DelIscB-T5E+sgRNA-V1, i.e., 1-fold change;

[0030] Figure 12 In Example 3, to improve the stability of sgRNA-V1, the weak A-U or G-U base pairs in the stem region were replaced with strong G-C base pairs;

[0031] Figure 13To improve the flexibility of sgRNA-V1 in Example 3, the G-U base pair or several other base pairs in the pseudoknot region were replaced with unpaired bases;

[0032] Figure 14 It is a combination of the effectively truncated and mutated versions of sgRNA-V1 in Example 3;

[0033] Figure 15 It is a combination of the DelIscB engineering protein with or without T5E and the sgRNA version in Example 3;

[0034] Figure 16 It is a comparison of the indel frequencies of WT-DelIscB+sgRNA-V1, enDelIscB, enDelIscB-T5E, and enIscb-T5E at 24 endogenous genomic loci in HEK293T cells in Example 4. Data are presented as mean ± standard deviation (n = 3);

[0035] Figure 17 It is a comparison of the activities of WT-DelIscB+sgRNA-V1, enDelIscB, enDelIscB-T5E, and enIscb-T5E in HEK293T cells in Example 4. Each point represents Figure 15 the average indel level measured in three independent biological replicates for each endogenous locus in. P values were determined by unpaired t-test. All data are presented as mean ± standard deviation;

[0036] Figure 18 It is the deletion and insertion cleavage patterns generated by WT-DelIscB+sgRNA-V1, enDelIscB, and enDelIscB-T5E at the EMX1-sg2 locus in HEK293T cells in Example 4. Data are shown for one independent biological replicate;

[0037] Figure 19 It is the off-target effect analysis of enDelIscB and enDelIscB-T5E at three target sites (EMX1-sg1, VEGFA-sg2, and CXCR4-sg2) in Example 4. The first row of each panel is the target sequence, and the following 16 rows represent computationally predicted off-target sites. TAM is shown in blue, and the mismatched bases of the off-target sites relative to the target site are marked in lowercase red. The numbers and color intensities within each box of the heatmap on the right represent the indel level of the editor at that site. Data are presented as the mean of 3 independent biological replicates;

[0038] Figure 20 It is a schematic diagram for verifying the nickase activity of enDelIscB using a fluorescence reporter system in Example 5;

[0039] Figure 21 The cleavage efficiency after mutating the RuvC and HNH active site residues of enDelIscB to alanine in Example 5;

[0040] Figure 22 For hA3A-enDelIscB in Example 5 D60A -2×UGI and enDelIscB D60A -Comparison of the C-to-T conversion efficiencies of hA3A*-2×UGI and enDelIscB at two endogenous loci (CXCR4-sg2 and PCSK9-sg3). Data are mean ± standard deviation (n = 3);

[0041] Figure 23 For TadA-enDelIscB in Example 5 D60A -TadA*, enDelIscB D60A -TadA and enDelIscB D60A -Comparison of the A-to-G conversion efficiencies of -2×TadA and enDelIscB at two endogenous loci (EMX1-sg2 and EMX1-sg5). Data are mean ± standard deviation (n = 3);

[0042] Figure 24 Comparison of the editing windows, maximum editing efficiencies, and indel frequencies of the base editors in Example 5; where: (A) Editing window plots of ICBE and miCBE show the average C-to-T conversion efficiency at each position of all tested endogenous loci. Data are shown as mean (the same below); (B) Editing window plots of IABE and miABE show the average A-to-G conversion efficiency at each position of all tested endogenous loci; (C) Base editing activities of the four base editors in HEK293T cells. Each point represents the average maximum base editing efficiency of each endogenous locus in three independent biological replicates. All data are shown as mean ± standard deviation. P values were determined by unpaired t-test, and ns indicates no significant difference; (D) Comparison of indel frequencies of ICBE and miCBE at 9 endogenous loci. Data are mean ± standard deviation (n = 3) (the same below); (E) Comparison of indel frequencies of IABE and miABE at 9 endogenous loci;

[0043] Figure 25 The indel rates of the Tyr gene of F0 mice edited by enDelIscB (n = 39) and enDelIscB-T5E (n = 46) in Example 6;

[0044] Figure 26Procedure for detecting IscB cleavage activity in HEK293T cells using a fluorescence reporter system. This system includes an editing plasmid with IscB-T5 exonuclease (T5E), sgRNA, and mCherry, and another reporter plasmid with BFP and EGFxxFP. NLS, nuclear localization signal. T2A, a self-cleaving 2A peptide from Autographa californica multiple nucleopolyhedrovirus. Detailed implementation

[0045] The following examples are used to explain the technical solutions of the present invention non-restrictively.

[0046] In the present invention, the following plasmid construction processes are involved:

[0047] For constructing the prokaryotic expression plasmid, the ORFs of SpCas9, OgeuIscB, AwaIscB, and DelIscB were cloned from addgene plasmids (#44758, #176540, #176538, and #176588 respectively) to the downstream of the lac operon on the pACYC184 (p15A replicon, chloramphenicol resistance) vector through Gibson assembly. The corresponding sgRNA / ωRNA / ncRNA with the target or non-target guiding sequence was cloned to the downstream of the J23119 promoter on the pRSF (RSF replicon, kanamycin resistance, and lacI expression) vector through Gibson assembly. Unless otherwise specified in the present invention, the method for constructing plasmids is Gibson assembly. All fragments used in Gibson assembly were obtained by polymerase chain reaction (PCR) using Q5 Hot start High-Fidelity DNA Polymerase (NEB). The assembly reagent used in Gibson assembly is HiFi DNA Assembly Master Mix (NEB) or ClonExpress Ultra OneStep Cloning Kit (Vazyme).

[0048] For constructing the eukaryotic expression plasmid, the human codon-optimized DelIscB sequence (shown in SEQ ID No. 23, synthesized by Tsingke Biotech) was first cloned into the vector of the PCMV-PE2 plasmid (Addgene, #132775). The sgRNA-V1 was cloned to the downstream of the U6 promoter of the pU6-Sp-pegRNA-HEK3_CTT_ins plasmid (Addgene, #132778). Subsequently, DelIscB and its sgRNA were cloned into the pCAG vector of the enIscB plasmid (Addgene, #205410).

[0049] To construct the fluorescent reporter plasmid, the BFP-T2A-EGFxxFP sequence (as shown in SEQ ID No. 24) was cloned downstream of the CMV promoter on the pCDNA3.1 vector (Addgene, 176540). The targeting sequence on the reporter plasmid was replaced by restriction digestion and ligation methods. The restriction sites were HindIII and EcoRI, and the targeting sequence was synthesized in the form of oligonucleotides. After annealing, it formed double-stranded DNA with sticky ends and was ligated by T4 ligase.

[0050] For the mutants of DelIscB protein, two DNA fragments were amplified by combining two complementary primers with the desired mutations and two upstream and downstream primers at the ends of the vector respectively, and were assembled by Gibson assembly with the gel-extracted vector digested with AgeI and HindIII (NEB). For the variants of sgRNA, two DNA fragments were amplified in the same way and assembled with the gel-extracted vector digested with SphI and NotI (NEB). To target different genomic loci, paired oligonucleotides were annealed and ligated into the BsaI- or BbsI-linearized plasmid vector by T4 DNA ligase (GenStar).

[0051] In the present invention, to construct the DelIscB-T5E, ICBE and IABE plasmids, the ORFs of elements such as T5E, hAPOBEC3A W98Y / W104A / Y130F , UGI, TadA8e V106W , NLS and linker were cloned from the enIscB-T5E, miCBE and miABE plasmids (Addgene, #205411, #205413 and #205412).

[0052] In the present invention, the cleavage activities of each protein were detected by flow cytometry:

[0053] To sensitively detect the cleavage activity of IscB, the present invention referred to a novel fluorescent reporter system in the eukaryotic system (D.Y. Han, Q.Q. Xiao, Y.F. Wang, et al. Development of miniature base editors using engineered IscB nickase[J]. Nature methods, 2023, 20(7): 1029-+.). The principle of this fluorescent reporter system is as Figure 26As shown, it contains an editor plasmid labeled with mCherry and a reporter plasmid labeled with BFP. Among them, after the BFP gene sequence on the reporter plasmid, two EGFP genes with homologous sequences separated by the target sequence are connected through the T2A peptide gene. Only when a double-strand break occurs at the target sequence of the reporter plasmid and is repaired by the homologous recombination repair mechanism in the cell, the normal EGFP ORF is restored, can the normal green fluorescent protein be expressed and quantitatively detected by a flow cytometer. When detected by a flow cytometer, the percentage of cells that also express EGFP among the cells that simultaneously express mCherry and BFP is counted. Among them, T5E is fused to the C-terminus of DelIscB.

[0054] Forty-eight hours after transfection, the cells were digested with 0.25% Trypsin-EDTA and collected into flow tubes. After centrifugation to remove the supernatant, the cells were resuspended in 200 μL of PBS solution. Then, the cells were analyzed or sorted using fluorescence-activated cell sorting (FACS) technology, and the instrument used was a BD FACSAria IIIu flow cytometer. For experiments using a fluorescence reporter to reflect the editing efficiency, only flow analysis is required, and the cell ratios corresponding to the channels of mCherry, blue fluorescent protein (BFP), and enhanced green fluorescent protein (EGFP) are counted. The FACS data was analyzed using FlowJo (vX.0.7). For cells that need to be subjected to next-generation sequencing subsequently, the BD FACSAria IIIu flow cytometer was used to sort approximately the top 20% of positive cells with mCherry intensity, and generally 20,000 cells were sorted.

[0055] In the present invention, the next-generation sequencing (NGS) and data analysis methods are as follows:

[0056] Approximately 20,000 sorted cells were lysed in 10 μL of QuickExtract DNA Extraction Solution (Lucigen), and 1 μL of the lysate was used for the PCR reaction. The target genomic region was amplified by the first-round PCR using 2×Phanta Flash MasterMix (Vazyme) to add Illumina adapters, and then the second-round PCR was performed to add index1 / 2 and P5 / P7. Twenty to thirty different PCR products (sub-libraries) could be mixed into one library, and gel extraction was performed to remove primer dimers and oversized non-specific products. Next, sequencing was performed on the Illumina NovaSeq Xplus platform (Majorbio Bio-Pharm Technology Co., Ltd.) with 150-bp paired-end reads. The sequencing data was split and returned by the company, and we used CRISPResso2 to analyze the Indels efficiency or base conversion efficiency of each editor.

[0057] Example 1 verified the activity of DelIscB in achieving the programmed cleavage of plasmid DNA in Escherichia coli and human cells

[0058] 1. Escherichia coli

[0059] First, prokaryotic expression plasmids expressing AwaIscB, OgeuIscB, and DelIscB were constructed.

[0060] Then, plasmid interference experiments were used to evaluate the activity of different proteins in cleaving plasmids in Escherichia coli:

[0061] As Figure 1 shown, two compatible plasmids were co-transfected into Escherichia coli. Each of these two plasmids had a unique resistance. Among them, plasmid P1 expressed Cas9 or IscB protein, and plasmid P2 expressed sgRNA or ωRNA. And a target site or a non-target site was provided as a control at the same time. After transfection, if the Escherichia coli clones could not grow on the agarose plate containing two antibiotics, it indicated that plasmid P2 was cleaved.

[0062] Using AwaIscB, OgeuIscB, and DelIscB as experimental subjects and SpCas9 as a positive control, the activity of each protein in cleaving plasmids in Escherichia coli was tested at its optimal guide sequence length, and the results were as Figure 2 shown.

[0063] The results showed that AwaIscB and OgeuIscB could not effectively cleave plasmids in Escherichia coli, but SpCas9 could (using a 20-nt guide length), and DelIscB was able to cleave plasmids using a 16-nt guide length, and its efficiency was similar to that of SpCas9.

[0064] 2. Human cells

[0065] First, investigate whether the hairpin structure formed by direct repeats (DR) / anti-repeats in DelIscB can be truncated to facilitate its transcription in human cells:

[0066] Using RNAfold for prediction and analysis, a series of truncations of different lengths were performed around the DR / anti-repeat region, and it was found that some of these truncates could still maintain plasmid cleavage function. One of the shorter and viable truncates, Δ20'-121'+GAAA, was named sgRNA-V0.1, as shown in SEQ ID No.10, and used as the basis for further engineering modification.

[0067] When transcribing sgRNA using the U6 promoter in mammalian cells, PolIII will pause at the poly-T signal. Therefore, the poly-T motif was mutated to U136'A in sgRNA-V0.1, and this mutated sgRNA was named sgRNA-V1 (as shown in SEQ ID No.11). By detecting the cleavage activity in mammalian cells, it was confirmed that it still had activity. In addition, to verify whether the stem-loop structures in other regions are functionally redundant, a series of truncations were also performed on the basis of sgRNA-V0.1 or sgRNA-V1, and it was confirmed that these truncates were functionally inactivated, indicating that these regions are important for function or require more detailed truncation. As Figure 3 shown.

[0068] Example 2 Engineering modification of DelIscB protein to improve cleavage activity

[0069] Use AlphaFold2 (AF2) to predict the protein structure of DelIscB and align it with the structure of OgeuIscB (7UTN). Five corresponding amino acid residues were found on DelIscB, namely Y37, Q85, S97, K350, and A404. These five corresponding residues were replaced with arginine respectively. Through the same activity test as in Example 1, it was proved that only one mutation, S97R, changed the activity by about 2-fold, while the other four replacements (Y37R, Q85R, K350R, and A404R) all led to a decrease in activity, as Figure 4 shown.

[0070] In addition, phylogenetic tree analysis was performed on known IscB proteins, and 10 IscB proteins closely related to DelIscB were selected for multiple sequence alignment (MSA). Based on the MSA, neutral or negatively charged residues on DelIscB were mainly selected. If these residues contained positively charged residues in the alignment positions among DelIscB homologs, the residue was mutated to arginine, and 122 single mutations were generated. Eight of these mutants (C12R, Q192R, E317R, Q340R, A343R, A364R, V396R, and S436R) showed a change in activity of more than 1.2-fold, as Figure 5 shown.

[0071] Meanwhile, when observing the multiple sequence alignment, it was unexpectedly found that DelIscB had unique residues different from homologous proteins at some alignment positions. These unique residues were mutated to residues with a higher proportion at the alignment positions, and it was found that some mutants had enhanced activity. Using this strategy, 66 mutants were constructed, and 5 mutants with a change in activity of more than 1.2-fold (L90E, S100A, G104S, W117F, and A142P) were screened out, as Figure 6 shown. Among them, the S97R mutant showed the highest activity. Since it was observed that there was also a lysine residue at the position aligned with S97 in its homologs, the activity of S97K was also detected and found to be slightly higher than that of S97R. The activities of the corresponding lysine substitution mutants (C12K, Q192K, E317K, Q340K, A343K, A364K, and V396K) of the other 7 arginine substitution mutants were further detected, as Figure 7 shown.

[0072] In addition, in order to investigate whether the obtained highly active variants were sequence-dependent, the activities of the above mutants were further confirmed using the reporter sequences EGFxxFP-target2 and EGFxxFP-target3, respectively, as Figure 7-9 shown. The results showed that, except for C12K, Q192K, E317K, and L90E, the activity changes of all mutants exceeded 1-fold under the simultaneous verification of the three reporter sequences.

[0073] Table 1 Sequences of the targeted regions of the fluorescent reporter plasmids

[0074]

[0075]

[0076] Table 2 Sequences of the targeted sequences of the fluorescent reporter plasmids

[0077]

[0078] Note: T1, T2, and T3 refer to the EGFxxFP-target1, EGFxxFP-target2, and EGFxxFP-target3 reporter plasmids, respectively.

[0079] After averaging the fold changes in activity of the above mutants under three reporter sequences, 14 mutants (S97R / K, V396R / K, E317R, C12R, A364R, Q340R / K, W117F, A142P, S100A, G104S, S436R) with an average increase of more than 1.2-fold were finally selected for subsequent combination to obtain DelIscB-V5 / V6 / V7 / V7.1 / V8 / V9 / V10 / V10.1 / V11, whose sequences are shown in SEQ ID No. 1-9 respectively. The cleavage activity of the mutant combination was detected, and the results showed that the combination of 10 variants (S97K, V396K, E317R, C12R, Q340K, W117F, A142P, S100A, G104S, S436R) exhibited the highest activity. This combination was named DelIscB-V10.1. Under the characterization of the reporter sequence EGFXXFP-target1, its activity was approximately 5 times that of the wild type and higher than that of enIscB-T5E (with T5E fused at the C-terminus). By confirming that DelIscB-V10.1 also showed the best activity under another reporter sequence, as Figure 10 shown.

[0080] Therefore, by screening a large number of mutants and combining highly active mutants, the currently best-performing protein version, DelIscB-V10.1, was obtained, and its sequence is shown in SEQ ID No. 8.

[0081] Example 3 Modifying the sgRNA backbone of DelIscB to further improve activity

[0082] Based on the predicted secondary structure information of sgRNA-V1 obtained in Example 1, its stem-loop structure was truncated to varying degrees. Among them, the truncations V1+Δ35-41+GAAA and V1+Δ67-71+GAAA improved the activity, and they were named sgRNA-V2 and sgRNA-V2.1, as shown in SEQ ID No. 12-13 respectively. The truncation V1+Δ221-240+GAAA had no significant effect on the activity, but considering that it could significantly shorten the total length of the sgRNA, it was also included in the subsequent research and named sgRNA-V2.2 (as shown in SEQ ID No. 14), as Figure 11 shown.

[0083] To enhance the stability of sgRNA-V1, the weak base pairs A-U or G-U were replaced with the strong base pair G-C, or several unpaired bases were paired at the pseudoknot. It was found that the activities of two mutants, V1+U15C+A28G and V1+U65C+A73G, were more than doubled, and thus they were named sgRNA-V3 and sgRNA-V3.1 (shown in SEQ ID No.15-16), as Figure 12 shown.

[0084] During the process of stabilizing sgRNA, it was found that stability mutations at the pseudoknot would reduce the activity, such as U205C and A206G. In addition, inspired by the enhanced activity due to the increased flexibility of the sgRNA of CasX, a series of mismatch mutations were introduced at the pseudoknot in the hope of increasing the flexibility of sgRNA-V1. It was found that the U205A mutation could increase the activity by more than 1.5-fold, and this variant was named sgRNA-V4 (shown in SEQ ID No.17), as Figure 13 shown.

[0085] When the above different versions of sgRNA were combined, it was found that the combination of sgRNA-V4 / V2 / V2.2 could maximize the activity, and it was named sgRNA-V5 (shown in SEQ ID No.18), as Figure 14 shown. Finally, sgRNA-V5 was combined with DelIscB-V10.1-T5E and DelIscB-V10.1 respectively, and the activities after combination were further improved. These two final combinations were named enDelIscB-T5E (enDelIscB-Linker-T5E-nucleoplasmin NLS, shown in SEQ ID No.20) and enDelIscB (enDelIscB-nucleoplasmin NLS, shown in SEQ ID No.19) respectively. Under the detection of the reporter sequence EGFXXFP-target1, their activities were higher than that of enIscB-T5E, as Figure 15 shown.

[0086] Example 4 Evaluation of the gene editing efficiency and accuracy of enDelIscB and enDelIscB-T5E

[0087] To verify and evaluate the gene editing activities of enDelIscB and enDelIscB-T5E obtained in Example 3 at endogenous gene loci, an integrated plasmid expressing enDelIscB / enDelIscB-T5E, sgRNA, and mCherry was first constructed and transfected into HEK293T cells.

[0088] Forty-eight hours after transfection, cells were collected, and the top 20% of mCherry-positive cells with higher expression levels were sorted by flow cytometry for amplicon sequencing to determine the indel (insertion-deletion) efficiency. A total of 24 sites were finally selected from 6 genes for validation, and WT-DelIscB+sgRNA-V1 and enIscb-T5E were added as controls, as Figure 16 shown.

[0089] Since the TAM (5′-NWRRNA-3′) of OgeuIscB is not compatible with the TAM (5’-NAC-3’) of DelIscB, the target sequences with the closest TAM positions were selected for activity comparison, as shown in Table 3.

[0090] Table 3 Sequences targeting human genomic loci

[0091]

[0092]

[0093] Note: The targeting sequences of enDelIscB and enIscB-T5E were designed to target the same region as much as possible. The targeting sequences of base editors are also in this table.

[0094] At these 24 sites, compared with WT-DelIscB+sgRNA-V1, enDelIscB increased the gene editing activity by about 48.9-fold on average, and its activity after fusion with T5E was comparable to that of enIscb-T5E, which were 54.27±28.07% and 53.70±29.84% respectively, as Figure 16 and 17 shown.

[0095] Meanwhile, the indel patterns formed by WT-DelIscB+sgRNA-V1, enDelIscB, and enDelIscB-T5E in the genome were also compared, as Figure 18 shown. At the EMX1-sg2 locus, WT-DelIscB+sgRNA-V1 produced lower insertion and deletion efficiencies (up to about 1.5%) and a narrower range (about 10 bp). enDelIscB produced a wider deletion range (about 90 bp) and a certain proportion of insertions (about 17%). enDelIscB-T5E produced a narrower deletion range than enDelIscB (about 60 bp), but its deletion efficiency was higher than that of enDelIscB, and it hardly caused insertion events, similar to the cleavage pattern of enIscB-T5E.

[0096] In addition, the off-target effects of enDelIscB and enDelIscB-T5E at three target sites were analyzed using Cas-OFFinder, as Figure 19 shown. The results showed that there were almost no insertions or deletions at 16 predicted off-target sites corresponding to two target sites (EMX1-sg4 and CXCR4-sg2) for enDelIscB and enDelIscB-T5E, while there were 4 significant off-target events at VEGFA-sg2.

[0097] In summary, enDelIscB and enDelIscB-T5E exhibited strong gene editing activities in mammalian cells.

[0098] Example 5 Development of a base editor based on the enDelIscB nickase

[0099] By performing a full-sequence alignment of DelIscB with OgeuIscB, the putative active residues (D60, E171, H226, H324, and D327) corresponding to the RuvC and HNH active residues of OgeuIscB on DelIscB were determined. These putative active residues were mutated to alanine, and a pair of fluorescent reporter plasmids containing single-target sequences or dual-target sequences as shown in Table 1 were constructed, and the nickase activity was verified using this pair of fluorescent reporter plasmids, as Figure 20 shown.

[0100] The results were as Figure 21 shown. The corresponding single mutations (D60A, E171A, H324A, and D327A) on the RuvC domain and the H226A mutation on the HNH domain all lost double-strand cleavage activity but retained nickase activity. It should be noted that when verifying the activity with a single-target reporter plasmid, the five single mutants produced approximately 10% more EGFP+ cells than the negative control (NT), presumably due to the background effect caused by the nicked plasmid being more prone to double-strand breakage. In addition, enDelIscB with the D60A + H226A mutation was inactive and was called inactivated enDelIscB (deadenDelIscB).

[0101] Then, the enDelIscB D60A nickase was selected, and the cytosine deaminase hAPOBEC3A W98Y / W104A / Y130F (a high-fidelity version of hAPOBEC3A, which was called hA3A*) was fused to its N-terminus or C-terminus with two uracil glycosylase inhibitors (2×UGI), thereby obtaining a C-to-T DNA base editor. Through next-generation sequencing, it was confirmed that at two endogenous sites (CXCR4-sg2 and PCSK9-sg3), the base editor with hA3A fused to the C-terminus (enDelIscBD60A Compared with the base editor with hA3A fused at the N-terminus (hA3A-enDelIscB D60A -2×UGI), the one with hA3A-2×UGI showed significantly higher C-to-T conversion efficiency, and the former was named ICBE (i.e., IscB-based cytosine base editor) (enDelIscB D60A -XTEN-hA3A*-nucleoplasmin NLS-UGI-UGI-SV40NLS, as shown in SEQ ID No. 21), as Figure 22 shown. This result is consistent with the previous situation where fusing a nuclear localization signal (NLS) at the N-terminus of DelIscB would reduce its activity.

[0102] Meanwhile, adenine base editors were also constructed by fusing TadA8e V106W (a high-fidelity version of TadA8e, which was named TadA*) with enDelIscB D60A in different forms. It was found that the form of fusing the TadA* dimer to the C-terminus of enDelIscB D60A showed the best A-to-G conversion activity at two target sites, and it was named IABE (enDelIscB D60A -Linker-TadA*-BP-NLS-TadA*-BP-NLS, as shown in SEQ ID No. 22), as Figure 23 shown.

[0103] Next, through next-generation sequencing, enIscB D61A and enDelIscB D60AThe editing windows and efficiencies of the base editors. By summarizing the conversion efficiencies of each C-to-T or A-to-G at all target sites, it was found that for ICBE, its editing window was approximately at positions -1 to 12 (counting TAM as positions 17-19), while compared with the prior art, the editing window of miCBE (Han, Dingyi, et al. "Development of miniature base editors using engineered IscB nickase." Nature methods 20.7 (2023): 1029-1036.) was approximately at positions 2 to 14. For IABE, its editing window was approximately between positions -4 and 12 of the target site, while the editing window of miABE (same as above) was between positions 2 and 12. Statistical analysis was performed on the average highest base editing efficiency at each site. The highest efficiency of ICBE of the present invention was 49.26±23.22%, while the highest efficiency of miCBE was 40.93±33.85%. The highest efficiency of IABE of the present invention was 60.22±22.47%, while the highest efficiency of miABE was 38.89±31.93%. The results of the differential analysis showed that there was no significant difference in the editing efficiency of the base editors based on enIscB D61A and enDelIscB D60A at these endogenous sites. At the same time, the indel situations generated by these four editors at the target sites were also detected, and the results showed that the indel rates induced by these four base editors were all very low, as shown in Figure 24 shown.

[0104] In summary, these results indicate that the base editors (ICBE and IABE) of the present invention based on enDelIscB D60A can be used together with other miniature base editors (miBEs, namely miCBE and miABE) as another effective option to expand the miniature base editing tool library.

[0105] Example 6 Efficient construction of a mouse model using enDelIscB and enDelIscB-T5E

[0106] The tyrosinase (Tyr) gene was used as the target to construct an albino mouse model because the phenotype could be directly evaluated by observing the mouse coat color.

[0107] Eighteen sgRNAs targeting exon 1 (exon1) or exon 4 (exon4) of the Tyr gene with different guide sequences were screened in mouse neurocytes (N2a). The indel detection results showed that 9 out of 18 sgRNAs exhibited detectable activity, indicating that enDelIscB-T5E can also function in mouse cells. Among them, the sgRNA targeting exon 1 of the Tyr gene, namely Tyr-sg12, demonstrated the highest efficiency.

[0108] Table 4 Sequences targeting murine genomic loci

[0109]

[0110] Subsequently, Tyr-sg12 was in vitro transcribed. The mRNA containing enDelIscB or enDelIscB-T5E (100 ng / μL) was mixed with the sgRNA containing Tyr-sg12 (200 ng / μL) and co-injected into the cytoplasm of fertilized eggs of C57BL / 6J mice. The phenotypes of the newborn mice showed that among the 39 F0 mice injected with enDelIscB, 35 had pure white fur, 2 had black and white fur, only 1 had black fur, and 1 died prematurely. Among the 47 F0 mice injected with enDelIscB-T5E, 44 had pure white fur and 3 had black and white fur. The genotype analysis results showed that among the 39 F0 mice injected with enDelIscB, only 4 had incomplete gene editing, and the average indel level was 96.77%. Among the 47 F0 mice injected with enDelIscB-T5E, only 3 had incomplete gene editing, and the average indel level was 98.67%, as Figure 25 shown. The phenotype of each mouse was consistent with its genotype.

[0111] These results indicate that both enDelIscB and enDelIscB-T5E exhibit strong gene editing activity in mouse embryos and have the ability to efficiently construct mouse disease models through embryo microinjection.

[0112] In summary, in the present invention, a CRISPR-related IscB system with a flexible TAM (NAC), DelIscB, was characterized and modified to obtain two powerful genome editors, enDelIscB and enDelIscB-T5E. In addition, based on the enDelIscB nickase, effective compact base editors, ICBE and IABE, were established. The present invention verified that animal models can be efficiently constructed using enDelIscB and enDelIscB-T5E.

Claims

1. A single mutation site that improves the plasmid cleavage activity of the DelIscB protein, wherein the site is selected from S97R / K, V396R / K, E317R, C12R, A364R, Q340R / K, W117F, A142P, S100A, G104S and / or S436R.

2. An optimized variant of DelIscB protein, wherein the optimized variant is selected from DelIscB-V5 / V6 / V7 / V7.1 / V8 / V9 / V10 / V10.1 / V11, and the sequences of the optimized variants are shown in SEQ ID No.1-9 respectively.

3. An optimized sgRNA variant of DelIscB, wherein the optimized variant is selected from sgRNA-V1 / V2 / V2.1 / V2.2 / V3 / V3.1 / V4 / V5, and the sequences of the optimized variants are shown in SEQ ID No.11-18, respectively.

4. A genome editor based on the IscB system, wherein the genome editor is enDelIscB or enDelIscB-T5E, and the sequence of the genome editor is shown in SEQ ID No.19-20.

5. Based on enDelIscB D60A A base editor, wherein the base editor is an ICBE or an IABE, wherein the ICBE is a base editor that converts the cytosine deaminase hAPOBEC3A W98Y / W104A / Y130F With two uracil glycosylase inhibitors fused to enDelIscB D60A The sequence of ICBE is shown in SEQ ID No.21; the IABE is obtained by converting TadA8e V106W Fusion to enDelIscB D60A The sequence of IABE obtained at the C-terminus is shown in SEQ ID No.

22.

6. Use of the genome editor and / or base editor described in claim 4 or 5 in gene editing and / or base editing in prokaryotic cells.

7. Use of the genome editor and / or base editor described in claim 4 or 5 in gene editing and / or base editing in eukaryotic cells.

8. Use of the genome editor and / or base editor described in claim 4 or 5 in gene editing and / or base editing in animal embryos.

9. Use of the genome editor and / or base editor described in claim 4 or 5 in developing lead editing, gene regulation and / or epigenetic editing tools based on enDelIscB.

Images