WO2022169235A1 - Prime editing composition with improved editing efficiency - Google Patents

Abstract

The present invention relates to a gene editing complex for prime editing and a use thereof. According to an aspect, a gene editing complex for prime editing and a prime editing composition comprising same of the present invention exhibit remarkably higher gene editing efficiency than conventionally known prime editors, such as PE2, etc. Thus, a highly effective, gene edition-based therapeutic agent can be developed with the gene editing complex used as an active ingredient.

Description

Prime editing composition with improved correction efficiency

The present invention relates to gene editing complexes and uses thereof for prime editing.

Prime Editing is an innovative novel genome editing method capable of introducing genetic changes of virtually any size without the need for donor DNA or double-strand breaks (DSBs) (Anzalone, AV et al. Search-and -replace genome editing without double-strand breaks or donor DNA. Nature 576, 149-157 (2019)). These changes include insertions, deletions, and all possible 12 point mutations, as well as combinations of these changes.

Prime editor (PE) basically consists of Cas9 nickase-reverse transcriptase (RT) fusion protein and prime editing guide RNA (pegRNA); The pegRNA contains a guide sequence recognizing a target sequence, a tracrRNA scaffold sequence, a primer binding site (PBS) required for initiation of reverse transcription, and a desired genetic change. An RT template homologous to the target sequence. include Four types of Prime Editor have been developed: PE1, PE2, PE3, and PE3b.

Among them, PE2 is known to be capable of inducing prime editing in cell types of various species including liver cells. However, PE2 is sometimes not efficient enough for gene editing. Therefore, to further improve prime editing efficiency, PE3 and PE3b can be generated by adding a single guide RNA (sgRNA) to PE2. However, there is a risk of unexpected mutations as PE3 and PE3b often lead to unintended indels.

Under this background, the present inventors have improved the existing PE2 as a result of efforts to improve the prime editing efficiency, and confirmed that the gene editing efficiency of the improved PE2 mutant is significantly improved, thereby completing the present invention.

One aspect is 1) Cas nickase (nickase), reverse transcriptase (reverse transcriptase, RT) and a single-stranded DNA-binding domain (single-stranded DNA-binding domain, ssDBD) comprising a fusion protein; And 2) to provide a gene editing complex comprising a prime editing guide RNA (pegRNA).

Another aspect is 1) Cas nickase (nickase), reverse transcriptase (reverse transcriptase, RT) and a fusion protein comprising a single-stranded DNA-binding domain (ssDBD), a poly encoding the fusion protein a nucleotide or a recombinant vector comprising the polynucleotide; And 2) to provide a composition for gene editing, including a recombinant vector comprising a prime editing guide RNA (pegRNA) or a polynucleotide encoding the same.

Another aspect is to provide a method for editing a gene of a subject, comprising introducing the composition for gene editing into isolated eukaryotic cells or eukaryotic organisms other than humans.

One aspect is 1) Cas nickase (nickase), reverse transcriptase (reverse transcriptase, RT) and a single-stranded DNA-binding domain (single-stranded DNA-binding domain, ssDBD) comprising a fusion protein; and 2) a prime editing guide RNA (pegRNA).

As used herein, the term "gene editing" refers to a genetic engineering technique for manipulating genes or DNA using artificially engineered nucleases or gene scissors, and one or more nucleic acid molecules (eg, 1 - 100,000 bp, 1 - 10,000 bp, 1 - 1,000 bp, 1 - 100 bp, 1 - 70 bp, 1 - 50 bp, 1 - 30 bp, 1 - 20 bp, or 1 - 10 bp) deletion, insertion, substitution, etc. It means to lose, alter, and/or restore (modify) gene function. Specifically, it may include DNA or gene knockout, deletion such as knockdown, gene insertion (knock-in), gene correction, gene expression regulation, or chromosome rearrangement.

The gene knockout may refer to the regulation of gene activity by deletion, substitution, and/or insertion of one or more nucleotides, eg, inactivation, of all or part of a gene (eg, one or more nucleotides). The gene inactivation refers to a modification to encode a protein that has lost its original function or suppressed or downregulated the expression of a gene. In addition, gene regulation involves structural modification of proteins obtained by deletion of exon sites due to simultaneous targeting of both intron sites surrounding one or more exons of the target gene, expression of dominant negative form of protein, expression of competitive inhibitor secreted in soluble form, etc. It may mean a change in the function of a gene as a result. The gene insertion (knock-in) refers to inserting an exogenous base sequence that did not exist in another species or originally in the organism into the genome of the organism or a DNA sequence derived from the organism by using genetic recombination technology.

The gene editing complex is a complex for prime editing, and may specifically edit a gene through prime editing.

As used herein, the term “prime editing” refers to a genome editing method capable of introducing a genetic change by cutting only one strand of DNA without DNA double-strand cutting by fourth-generation gene scissors.

The prime editing may be performed by "Prime editor (PE)" or a variant thereof. The prime editor may include a Cas nickase-reverse transcriptase (RT) fusion protein and a prime editing guide RNA (pegRNA), and additional domains or proteins are added to improve the efficiency of the prime editing. It may be further included in the fusion protein. The type of the prime editor includes, but is not limited to, PE1, PE2, PE3, PE3b, and the like or variants thereof. In one embodiment, the prime editor may be one in which a single-stranded DNA-binding domain (ssDBD) is additionally included in Prime editor 2 (PE2), the following Examples and Experiments In the example, it was named hyPE2 as a variant of PE2.

The prime editor includes a prime editor protein (fusion protein) in which a Cas nickcase, a reverse transcriptase, and a single-stranded DNA binding domain are fused, and a prime editing guide RNA (pegRNA). Prime editor as used herein narrowly refers to a prime editor protein (fusion protein) in which Cas nickcase protein, single-stranded DNA binding domain and reverse transcriptase are fused, broadly, the prime editor protein and prime editing guide RNA form a complex It may mean a type of prime editor complex.

In the present specification, the prime editor may mean including only the Cas nickcase-ssDBD-RT fusion protein, or may mean including the Cas nickcase-ssDBD-RT fusion protein and pegRNA together. For example, when pegRNA is separately introduced into the cell, introduction of the prime editor here may mean introducing only the Cas nickcase-ssDBD-RT fusion protein. That is, when the pegRNA has already been introduced, the introduction of the prime editor may mean introducing only the Cas nickcase-ssDBD-RT fusion protein. In one embodiment, the prime editor may refer to a Cas nickcase-ssDBD-RT fusion protein.

The "Cas nickase" used in the prime editor has a nickcase activity to cut (nick) one-stranded DNA, and the Cas protein may be modified to have a nickcase activity, and a target together with pegRNA. Any Cas protein that has been delivered to a site and modified so that only a single strand can be specifically cleaved can be used without limitation. The Cas nickcase includes Cas9 nickcase, SaCas9 nickcase, SpCas9 nickcase, Cpf1 nickcase, Cas3 nickcase, Cas8a-c nickcase, Cas10 nickcase, Cse1 nickcase, Csy1 nickcase, Csn2 nickcase, Cas4 nickcase , Csm2 nickcase, Cm5 nickcase, Csf1 nickcase, C2C2 nickcase, NgAgo nickcase, Cas12e nickcase, Cas12d nickcase, Cas12a nickcase, Cas12b1 nickcase, Cas13a nickcase, Cas12c nickcase and variants thereof It may be one or more selected from the group, and may specifically be a Cas9 nickcase, and more specifically may be a Cas9 H850A nickcase.

As used herein, the term “reverse transcriptase (RT)” is an enzyme that uses RNA as a template and synthesizes new DNA complementary thereto.

As used herein, the term “single-stranded DNA-binding domain (ssDBD)” refers to an independently folded protein domain comprising one or more structural motifs that recognize single-stranded DNA. The DNA-binding domain may recognize a specific DNA sequence or have a general affinity for DNA, and some DNA-binding domains may include a nucleic acid in a folded structure.

The single-stranded DNA binding domain may be Rad51 DBD or RPA70 DBD, specifically Rad51 DBD.

As used herein, "fusion protein" refers to a protein artificially synthesized such that a Cas nickase, a reverse transcriptase, and a single-stranded DNA binding domain are bound.

In the fusion protein, the single-stranded DNA binding domain may be connected to the C-terminus of the Cas nickcase and the N-terminus of the reverse transcriptase. Accordingly, the fusion protein may include a recombinant protein linked from the N-terminus to a Cas nickcase, a single-stranded DNA binding domain, and a reverse transcriptase in order.

The Cas nick case may include the amino acid sequence of SEQ ID NO: 1, and specifically, may include the amino acid sequence of SEQ ID NO: 1.

The single-stranded DNA binding domain may include the amino acid sequence of SEQ ID NO: 2, and specifically, may include the amino acid sequence of SEQ ID NO: 2.

The reverse transcriptase may include the amino acid sequence of SEQ ID NO: 3, specifically, may consist of the amino acid sequence of SEQ ID NO: 3.

In the fusion protein, the Cas nickase, the single-stranded DNA binding domain, and the reverse transcriptase may be directly linked to each other or linked through a linker, but may be specifically linked through a linker.

The linker is not particularly limited as long as it exhibits the activity of the fusion protein, and for example, glycine, alanine, leucine, isoleucine, proline, serine, threonine, asparagine, aspartic acid, cysteine, glutamine, glutamic acid, lysine, It can be linked using amino acids, such as arginic acid, and can be used by linking 1 to 40 amino acids.

In the fusion protein, a linker connecting the Cas nickcase and the single-stranded DNA-binding domain may be referred to as a first linker, and a linker connecting the single-stranded DNA-binding domain and the reverse transcriptase may be referred to as a second linker. The first linker may include the amino acid sequence of SEQ ID NO: 4, specifically, may consist of the amino acid sequence of SEQ ID NO: 4. The second linker may include the amino acid sequence of SEQ ID NO: 5, specifically, may consist of the amino acid sequence of SEQ ID NO: 5.

Accordingly, the fusion protein may include a recombinant protein linked in order from the N-terminus to the Cas nickcase, the first linker, the single-stranded DNA binding domain, the second linker, and the reverse transcriptase.

The fusion protein may include the amino acid sequence of SEQ ID NO: 6, specifically, the amino acid sequence of SEQ ID NO: 6.

The fusion protein may further include a nuclear localization sequence (NLS).

As used herein, the term "nuclear localization sequence or signal (NLS)" refers to an amino acid sequence that serves to transport a specific substance (eg, protein) into a cell nucleus, and is generally a nuclear pore (Nuclear Pore). ) through the cell nucleus (Kalderon D, et al., Cell 39:499509 (1984); Dingwall C, et al., J Cell Biol. 107(3):8419 (1988)). Although such nuclear localization sequences are not required for gene editing activity in eukaryotes, it is believed that the inclusion of such sequences enhances the activity of the system, particularly targeting nucleic acid molecules in the nucleus. The nuclear localization sequence may include the amino acid sequence of SEQ ID NO: 7, specifically, may consist of the amino acid sequence of SEQ ID NO: 7.

The nuclear localization sequence may be added to the N-terminus and/or C-terminus of the fusion protein, and specifically may be added to the N-terminus and C-terminus. Accordingly, the fusion protein is to include a recombinant protein linked in the order of the nuclear localization sequence, the Cas nickase, the first linker, the single-stranded DNA binding domain, the second linker, the reverse transcriptase and the nuclear localization sequence from the N-terminus. can

In the present specification, the fusion protein comprising the nuclear localization sequence, Cas nickase, first linker, single-stranded DNA binding domain, second linker, reverse transcriptase, and some or all of the above amino acid sequences described in SEQ ID NOs. In addition, as an amino acid sequence exhibiting 80% or more, specifically 90% or more, more specifically 95% or more, more specifically 98% or more, and most specifically 99% or more homology with the sequence, substantially If the amino acid sequence expressing a protein exhibiting the same or corresponding biological activity as the respective protein, the case where some sequences have an amino acid sequence deleted, modified, substituted or added is included without limitation.

As used herein, the term "homology" refers to the degree of similarity between the amino acid sequence constituting the protein or the polynucleotide sequence encoding the same. When the homology is sufficiently high, the expression product of the polynucleotide (gene) and the protein are identical or have a similar activity. In addition, homology can be expressed as a percentage according to the degree of correspondence with a given amino acid sequence or polynucleotide sequence. In the present specification, a homologous sequence having the same or similar activity to a given amino acid sequence or nucleotide sequence is expressed as "% homology". For example, standard software that calculates parameters such as score, identity, and similarity, specifically BLAST 2.0, or hybridization written under defined stringent conditions Appropriate hybridization conditions that can be confirmed by comparing the sequences by experimentation are within the technical scope and are well known to those skilled in the art (eg, J. Sambrook et al., Molecular Cloning, A Laboratory Manual, 2nd Edition, Cold Spring). Harbor Laboratory press, Cold Spring Harbor, New York, 1989; F.M. Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, Inc., New York).

As used herein, the term "pegRNA (prime editing guide RNA)" refers to a guide sequence recognizing a target sequence, a tracrRNA scaffold sequence, a primer binding site required for initiation of reverse transcription (PBS), and a desired gene. Includes an RT template that contains the enemy change.

In the pegRNA, the guide sequence refers to a sequence in the guide RNA that specifies a target site, and includes a sequence that is fully or partially complementary to the target sequence. The guide sequence is any polynucleotide sequence having sufficient complementarity with the target polynucleotide sequence to hybridize with the target DNA sequence and induce sequence-specific binding of the gene editing coalesce to the target DNA sequence. In some embodiments, the degree of complementarity between a guide sequence and its corresponding target sequence is about 50%, 60%, 75%, 80%, 85%, 90%, 95 when optimally aligned using an appropriate alignment algorithm. %, 97.5%, 99% or more. Optimal alignment can be determined using any algorithm suitable for aligning sequences, non-limiting examples of which include the Smith-Waterman algorithm, the Needleman-Wunsch algorithm, the Burroughs- Algorithms based on the Burrows-Wheeler Transform (eg Burrows Wheeler Aligner), ClustalW, Clustal X, BLAT, Novoalign (Novocraft Technologies) ), ELAND (Illumina, San Diego, CA), SOAP (available at soap.genomics.org.cn) and Maq (available at maq.sourceforge.net). In some embodiments, a guide The sequence is about 5, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40 , 45, 50, 75 or more nucleotides in length.In some embodiments, the guide sequence is about 75, 50, 45, 40, 35, 30, 25, 20, 15, 12 or less nucleotides in length. The ability of a guide sequence to induce sequence-specific binding of a gene editing complex of a can be assessed by any suitable assay.

As used herein, the term “target sequence” refers to a target nucleotide sequence for which pegRNA is desired. The target sequence may be a sequence expected to be targeted by pegRNA. The target sequence may be a partial sequence among known genomic sequences, or a sequence to be edited by a person skilled in the art using the system of the present invention.

The complex is for editing a target DNA or a target gene, and specifically, the complex of the present invention can edit a gene more effectively than a previously known prime editor, so that the gene editing efficiency or prime editing efficiency is improved/improved. .

According to one embodiment, the prime editing complex (hyPE2) comprising a fusion protein comprising Cas9 H840A, Rad51 DBD and RT of the present invention has a lower or similar frequency of unintended editing than the existing PE2 excluding Rad51 DBD. As it was confirmed that the prime editing editing efficiency was significantly improved while showing , it can be seen that it has superior efficacy than various previously known prime editors including PE2.

The "prime editing efficiency" means gene editing efficiency by the prime editor. Prime editing efficiency can be calculated as a rate at which editing induced by the prime editor and pegRNA occurs without unintentional mutation in the target sequence when prime editing is performed. The prime editing efficiency may be expressed as a percentage.

According to one embodiment, the data on the efficiency of the prime editing comprises the steps of introducing the genome editing composition into a cell or tissue; performing deep sequencing using the DNA obtained from the cell or tissue introduced with the composition; And it may be obtained by performing a method comprising the step of analyzing the prime editing efficiency from the data obtained by the deep sequencing.

The method of obtaining DNA from cells or tissues into which the prime editor is introduced may be performed using various DNA isolation methods known in the art. Since each cell or tissue is expected to have gene editing in the introduced target sequence, the gene editing efficiency can be detected by sequencing the target sequence. The sequencing method is not limited to a specific method as long as prime editing efficiency data can be obtained, but, for example, deep sequencing may be used.

The step of analyzing the prime editing efficiency from the data obtained by the deep sequencing may include calculating the prime editing efficiency. Prime editing efficiency may vary depending on the type and/or length of the pegRNA sequence and the target sequence. The data on the prime editing efficiency may be provided as a data set.

According to an embodiment, the editing efficiency of the gene editing complex may be affected by the melting temperature of PBS, and specifically, the gene editing efficiency may be improved when the melting temperature of PBS is low.

Another aspect is 1) Cas nickase (nickase), reverse transcriptase (reverse transcriptase, RT) and a fusion protein comprising a single-stranded DNA-binding domain (ssDBD), a poly encoding the fusion protein a nucleotide or a recombinant vector comprising the polynucleotide; And 2) it provides a composition for gene editing, comprising a recombinant vector comprising a prime editing guide RNA (pegRNA) or a polynucleotide encoding the same. The same parts as described above also apply to the composition.

The composition is for editing a target DNA or a target gene, and may be for editing a gene through prime editing.

The polynucleotide encoding the Cas nickcase may include the polynucleotide sequence of SEQ ID NO: 8, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 8.

The polynucleotide encoding the single-stranded DNA binding domain may include the polynucleotide sequence of SEQ ID NO: 9, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 9.

The polynucleotide encoding the reverse transcriptase may include the polynucleotide sequence of SEQ ID NO: 10, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 10.

The polynucleotide encoding the first linker may include the polynucleotide sequence of SEQ ID NO: 11, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 11.

The polynucleotide encoding the second linker may include the polynucleotide sequence of SEQ ID NO: 12, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 12.

The polynucleotide encoding the fusion protein may include the polynucleotide sequence of SEQ ID NO: 13, and specifically, may consist of the polynucleotide sequence of SEQ ID NO: 13.

As used herein, the term “polynucleotide” refers to deoxyribonucleotides or polymers of ribonucleotides that exist in single-stranded or double-stranded form. It encompasses RNA genomic sequences, DNA (gDNA and cDNA) and RNA sequences transcribed therefrom, and includes analogs of natural polynucleotides, unless otherwise specified.

In the present specification, the nucleotide sequence encoding the protein/domain, all or part of the fusion protein including the nucleotide sequence, as well as the nucleotide sequence encoding the amino acid described in each SEQ ID NO: 80% or more, specifically 90% of the sequence A base sequence encoding a protein substantially the same as or corresponding to each protein as a nucleotide sequence showing homology of at least 95%, more specifically at least 98%, and most specifically at least 99%. Sequences are included without limitation.

The fusion protein is 80% or more, 85% or more, 90% or more, 91% or more, 92% or more, 93% of the amino acid sequence of SEQ ID NO: It may include a polynucleotide encoding a protein exhibiting at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% homology.

In addition, the polynucleotide encoding the fusion protein is a range that does not change the amino acid sequence of the protein expressed from the coding region, considering codons preferred in the organism in which the protein is to be expressed due to codon degeneracy. Various modifications may be made to the coding region within. Accordingly, the polynucleotide may be included without limitation as long as it is a polynucleotide sequence encoding each protein.

In addition, the polynucleotide includes not only a nucleotide sequence encoding the amino acid sequence of the fusion protein, but also a sequence complementary to the sequence. The complementary sequence includes not only perfectly complementary sequences, but also substantially complementary sequences, which under stringent conditions known in the art, for example, nucleotides encoding the amino acid sequence of the fusion protein. It refers to a sequence capable of hybridizing with the nucleotide sequence of the sequence.

The "stringent conditions" means conditions that allow specific hybridization between polynucleotides. These conditions are specifically described in the literature (eg, J. Sambrook et al., supra). For example, genes having high homology between genes having homology of 40% or more, specifically 90% or more, more specifically 95% or more, still more specifically 97% or more, and particularly specifically 99% or more homology. Conditions that hybridize with each other and do not hybridize with genes with lower homology, or wash conditions of normal Southern hybridization at 60° C. 1XSSC, 0.1% SDS, specifically 60° C. 0.1XSSC, 0.1% SDS, more specifically As examples, the conditions of washing once, specifically 2 to 3 times, at a salt concentration and temperature equivalent to 68° C. 0.1XSSC, 0.1% SDS can be exemplified.

Hybridization requires that two polynucleotides have complementary sequences, although mismatch between bases is possible depending on the stringency of hybridization. The term "complementary" is used to describe the relationship between nucleotide bases capable of hybridizing to each other. For example, with respect to DNA, adenosine is complementary to thymine and cytosine is complementary to guanine. Accordingly, the present application may also include substantially similar polynucleotide sequences as well as isolated polynucleotide fragments complementary to the overall sequence.

Specifically, polynucleotides having homology can be detected using hybridization conditions including a hybridization step at a Tm value of 55° C. and using the conditions described above. In addition, the Tm value may be 60 °C, 63 °C, or 65 °C, but is not limited thereto and may be appropriately adjusted by those skilled in the art according to the purpose. The appropriate stringency for hybridizing polynucleotides depends on the length of the polynucleotides and the degree of complementarity, and the parameters are well known in the art (see Sambrook et al., supra, 9.50-9.51, 11.7-11.8).

As used herein, the term "recombinant vector" may refer to a medium capable of delivering the polynucleotide into a cell and/or expressing a target protein encoding it. In the present specification, the recombinant vector may include a polynucleotide encoding the fusion protein or a polynucleotide comprising a pegRNA encoding sequence. The vector may contain the necessary regulatory elements operably linked to the insert, ie, the polynucleotide, to allow expression of the insert when present in a cell of an individual. The term “operably linked” means that a nucleic acid expression control sequence and a nucleic acid sequence encoding a protein of interest are functionally linked to perform a general function. The operative linkage with the recombinant vector can be prepared and purified using genetic recombination techniques well known in the art, and site-specific DNA cleavage and ligation can be easily performed using enzymes generally known in the art. can do. The vector may include a promoter, a start codon, and a stop codon terminator. In addition, DNA encoding the signal peptide, and/or enhancer sequence, and/or the untranslated region on the 5' side and the 3' side of the desired gene, and/or a selectable marker region, and/or a replicable unit, etc. are appropriately added may include

The promoter may be constitutive or inducible as a general promoter, lac, tac, T3 and T7 promoters for prokaryotic cells, monkey virus 40 (SV40) for eukaryotic cells, mouse mammary tumor virus (MMTV) promoter, human Immunodeficiency virus (HIV), such as the long terminal repeat (LTR) promoter of HIV, Moloney virus, cytomegalovirus (CMV), Epstein Barr virus (EBV), Loose Sacoma virus (RSV) promoter, as well as , β-actin promoter, human heroglobin, human muscle creatine, human metallothionein-derived promoter, and the like.

The selectable marker is for selecting cells transformed by introducing a vector, and markers conferring a selectable phenotype such as drug resistance, auxotrophy, resistance to cytotoxic agents, or surface protein expression may be used. In an environment treated with a selective agent, only the cells expressing the selection marker survive, so that the transformed cells can be selected.

The vector may be a viral, cosmid or plasmid vector, but is not limited thereto. The type of the vector is not particularly limited as long as it allows the expression of the desired gene and the function of producing the desired protein in various desired cells such as prokaryotic and eukaryotic cells, but specifically a promoter showing strong activity and strong expression A vector capable of producing a large amount of a foreign protein in a form similar to that of a natural state while retaining its strength can be used.

As the vector, a combination of various target cells and vectors may be used. Expression vectors suitable for eukaryotic cells or eukaryotic organisms include, but are not limited to, vectors derived from SV40, bovine papillomavirus, adenovirus, adeno-associated virus, cytomegalovirus, lentivirus and retrovirus, etc. may be used. may, but is not limited thereto. Expression vectors that can be used in bacterial hosts include, but are not limited to, pET21a, pET, pRSET, pBluescript, pGEX2T, pUC vectors, col E1, pCR1, pBR322, pMB9 or derivatives thereof. Bacterial plasmids obtained, plasmids having a wider host range such as RP4, phage DNA exemplified by phage lambda derivatives such as λgt10, λgt11 or NM989, and others such as M13 and filamentous single-stranded DNA phage Other DNA phages and the like may be included. For insect cells, pVL941 or the like can be used.

The composition may be for editing target DNA or genes ex vivo or in vivo, and may be used for gene editing in eukaryotic cells, prokaryotic cells, eukaryotic organisms or prokaryotic organisms, specifically It can be used for genome editing of eukaryotic cells or eukaryotic organisms. The eukaryotic cells may be cells of yeast, mold, protozoa, plants, higher plants and insects, amphibians or birds, or mammalian cells such as CHO, HeLa, HEK293, and COS-1, for example , embryonic cells, stem cells, somatic cells, germ cells, cultured cells (in vitro), transplanted cells (graft cells) and primary cultured cells (in vitro and ex vivo), and in vivo, commonly used in the art It may be an (in vivo) cell, a cell of an organism, or a cell of a mammal including a human (mammalian cell). The eukaryotic organism may be a yeast, a fungus, a protozoan, a plant, a higher plant and insect, an amphibian, a bird or a mammal (human, primate such as monkey, dog, pig, cow, sheep, goat, mouse, rat, etc.). Preferably, the composition can be used for priming of eukaryotic cells or eukaryotic organisms.

A method of delivering the polynucleotide or recombinant vector to a cell can be accomplished using various methods known in the art. For example, calcium phosphate-DNA co-precipitation method, DEAE-dextran-mediated transfection method, polybrene-mediated transfection method, electroshock method, microinjection method, liposome fusion method, lipofectamine and protoplast fusion method, etc. It can be carried out by several methods known in In addition, in the case of using a viral vector, a target object, that is, the vector can be delivered into a cell using viral particles by means of infection. In addition, the vector can be introduced into the cell by gene bambadment or the like. The introduced vector may exist as a vector itself in a cell or may be integrated into a chromosome, but is not limited thereto.

Another aspect provides a method of editing a gene of a subject, comprising introducing the composition for gene editing of the present invention into a eukaryotic cell or eukaryotic organism. The same parts as those described above are equally applied to the above method.

The eukaryotic cell may be an individual or a cell isolated from a eukaryotic organism. Also, the eukaryotic organism may be non-human.

1) Cas nickase (nickase), a reverse transcriptase (reverse transcriptase, RT) and a fusion protein comprising a single-stranded DNA-binding domain (ssDBD), a polynucleotide encoding the fusion protein or a recombinant vector comprising the polynucleotide; And 2) the prime editing guide RNA (pegRNA) or a recombinant vector comprising a polynucleotide encoding the same can be introduced simultaneously or sequentially, as long as it does not affect the gene editing efficiency of the composition, it is not limited thereto .

All steps performed in the genome editing method may be performed intracellularly or extracellularly, or in vivo or ex vivo.

The method of editing the genome may be performed by prime editing.

According to one aspect, the gene editing complex for prime editing of the present invention and the composition for gene editing comprising the same show significantly superior gene editing efficiency than conventionally known prime editors such as PE2, using the gene editing complex as an active ingredient Gene editing-based therapeutics with excellent effects can be developed.

1 is a diagram showing the structure of the PE2 variant used in the present invention.

2 is a diagram illustrating the correlation between prime editing efficiency of biological replicates for high-throughput evaluation of prime editing activity. The number of target sequences is 83.

3 is a diagram comparing the prime editing efficiency of hyPE2 and PE2-mid_RPA70, which are PE2 variants, normalized to the efficiency of PE2. Target sequences with less than 1% PE2-induced prime editing efficiency are indicated by white dots. The number of target sequences is 64.

4 is a diagram comparing the prime editing efficiency of PE2 variants, hyPE2 and PE2-mid_RPA70, normalized to that of PE2, and the results for a target sequence having a PE2-induced prime editing efficiency of more than 1%. The number of target sequences is 30.

5 is a diagram comparing the prime editing efficiencies of PE2 variants hyPE2, PE2-N_Rad51 and PE2-C_Rad51 normalized to that of PE2, and results for a target sequence having a PE2-induced prime editing efficiency of more than 1%. The number of target sequences is 32.

6 is a view showing a comparison result of the prime editing efficiency of PE2 and hyPE2 in HEK293T cells. Data points represent the average prime editing efficiency of three biological replicates in each target sequence, with 0.1% added to all efficiency values to allow logarithmic scales for both x- and y-axes. The number of target sequences is 88.

7 is a diagram comparing the prime editing efficiency of hyPE2 and PE2-mid_RPA70, which are PE2 variants, normalized to that of PE2 in HCT116 cells in HCT116 cells. The results are for a target sequence having a PE2-induced prime editing efficiency of more than 1%. The number of target sequences is 43.

8 is a diagram showing the structure of the hyPE2 linker variant and sequence information of the linker.

9 is a diagram comparing the prime editing efficiency of hyPE2 and hyPE2 linker variants normalized to the efficiency of PE2. The fold increase was expressed as an adjusted fold increase, and the number of pegRNAs was 82.

FIG. 10 is a diagram comparing the prime editing efficiency of hyPE2 against the same endogenous target of HEK293T and HCT116 cells normalized to that of PE2. The results for the target sequence having a PE2-induced prime editing efficiency of more than 1%. The number of pegRNAs is 31 for HEK293T and 11 for HCT116.

11 is a diagram showing the results of comparing the prime editing efficiency of PE2 and hyPE2 in the endogenous region of HEK293T cells. The number of pegRNAs is 63.

12 is a view showing a comparison result of the prime editing efficiency of PE2 and hyPE2 in the endogenous region of HCT116 cells. The number of pegRNAs is 51.

13 is a diagram showing the results of comparing the prime editing efficiency of PE2 and hyPE2 in six endogenous regions of primary human dermal fibroblasts. The number of pegRNAs is 51.

14 is a diagram comparing the prime editing efficiency of hyPE2 in the same target sequence of primary human dermal fibroblasts normalized to that of PE2. Adjusted fold increments are plotted on the y-axis and are plotted as mean±standard deviation for three independent biological replicates.

15 is a diagram comparing the prime editing efficiency and unintended editing frequency of hyPE2 in the same endogenous region of HEK293T cells normalized to the result of PE2. The adjusted P-value was calculated by one-way ANOVA followed by a post-hoc analysis by Tukey's multiple comparisons, and the P-value was not recorded when the difference between the two groups was not statistically significant. The number of pegRNAs is 25.

FIG. 16 shows off-target effects of hyPE2 and PE2 at potential off-target sites of pegRNAs 2, 3, 5 and 4 other pegRNAs targeting HEK4 in HEK293T cells. Intended edit positions are highlighted in yellow, and mismatched and overhanging nucleotides are indicated in red and blue lowercase fonts, respectively. Data are expressed as mean ± standard deviation for three independent biological replicates.

FIG. 17 is a diagram comparing the on-target and off-target editing frequencies of hyPE2 normalized to PE2 in the same endogenous region of HEK293T cells. P-values were calculated with a two-tailed, unpaired Student's t-test, where the number of on-target pegRNAs was 7 and the number of off-target pegRNAs was 22.

FIG. 18 is a diagram comparing the prime editing efficiency of hyPE2 normalized to that of PE2 in the same target sequence integrated as a lentivirus into HEK293T cells. Results for target sequences with PE2-induced prime editing efficiency greater than 1%, the number of target sequences being 423.

19 is a diagram illustrating a result of comparing Spearman correlation coefficients between prediction models.

20 is a diagram showing the top 10 functions related to hyPE2 activity compared to PE2 activity determined by Tree SHAP (XGBoost classifier). Dot colors indicate high (red) or low (blue) values of the relevant function for each pegRNA.

21 is a diagram showing the fold increase dependence of hyPE2 prime editing efficiency compared to PE2 on the primer binding site (PBS) melting temperature. Editing efficiencies for hyPE2 and PE2 were determined from the same target sequence (Library B) incorporated lentivirally into HEK293T cells. The number of pegRNAs is 4 (<20 °C), 32 (20-30 °C), 236 (30-40 °C), 348 (40-50 °C) and 11 (≥50 °C), respectively.

22 is a diagram schematically illustrating the structure of hyPE2.

23 is a diagram showing the prediction results of the three-dimensional structure of hyPE2 (left) and PE2 (right) before and after the reverse transcription domain binds to the target ssDNA/pegRNA hybrid with a nick. The hypothetical interaction modeling between Rad51, the target ssDNA with nick and the pegRNA primer binding site is shown in hyPE2.

Hereinafter, it will be described in more detail through Examples and Experimental Examples. However, these Examples and Experimental Examples are for illustrative purposes, and the scope of the present invention is not limited to these Examples and Experimental Examples.

Example 1: Plasmid vector construction

Sequences encoding human RPA70-C and Rad51DBD were synthesized by request of GeneScript. To construct a plasmid encoding ssDBD-PE2-, the sequences were amplified by PCR and cloned into pCMV-PE2 (Addgene, no. 132775) plasmid. The plasmids were named PE2-mid_RPA70, hyPE2, PE2-N_Rad51 and PE2-C_Rad51 (FIG. 1). Linker variants were derived from the hyPE2 plasmid and cloned using Gibson assembly.

Example 2: Plasmid library construction and cell library construction

In a previous study, a plasmid library of 54,836 pairs of pegRNA encoding sequences and target sequences was generated (Nature Biotechnology volume 39, pages 198-206 (2021)). In the present invention, colonies were selected and 107 plasmids were randomly selected from the library, and the selected plasmids were mixed in an equimolar ratio (Library A).

Next, another library containing 665 pairs of pegRNAs and target sequences was further designed for a more extensive evaluation of various types of editing in a larger number of target sequences, which was designated as 'Library B'. To design the library B, 100 deletion-inducing pegRNAs, 100 insertion-inducing pegRNAs and 200 substitution-inducing pegRNAs were selected from the plasmid library of the 54,836 pairs of pegRNA encoding sequences and target sequences published in the previous study. For the selection, divided into 8 layers according to the editing efficiency published in a previous study (Nature Biotechnology volume 39, pages 198-206 (2021)) (<1%, 1-3%, 3-6%, 6- 10%, 10-20%, 20-30%, 30-40%, and >40%), a similar number of pegRNAs from each stratum were randomly selected to ensure that pegRNAs associated with all levels of efficiency were included. A total of 507 pegRNAs were derived by adding 400 pegRNAs selected through the above process and 107 pegRNAs selected from library A. In addition, 158 of the 507 pegRNAs can be modified to induce silent mutations in the NGG PAM sequence, and 158 modified pegRNAs that can induce silent mutations in the PAM sequence in addition to the initially designed editing in library B were added. did. Thus, the total number of pegRNAs in library B was 507 + 158 = 665. Each pegRNA was associated with three barcodes. Therefore, the number of oligonucleotides used to generate library B was 665 Х 3 = 1995.

Next, to prepare lentivirus generation from the 107 plasmid library, 4.0 Х 10 ⁶ HEK293T cells were seeded into 100 mm dishes containing DMEM. After 15 hours, the culture medium was replaced with DMEM containing 25 μM chloroquine diphosphate, and the cells were further cultured for 5 hours. The plasmid library was mixed with psPAX2 (Addgene no. 12260) and pMD2.G (Addgene no. 12259) in a molar ratio of 1.3:0.72:1.64, and then the plasmid was mixed with HEK293T using polyethyleneimine (PEI MAX, Polysciences). The cells were cotransfected. The next day, the culture medium was replaced with fresh medium. 48 hours after transfection, the medium containing the lentivirus was collected and filtered using a Millex-HV 0.45-μm low protein-binding membrane (Millipore), which was aliquoted and stored at -80°C. For titration of lentivirus, serial dilutions of virus aliquots in the presence of 8 μg/ml polybrene (Sigma) were transduced into HEK293T cells cultured in DMEM supplemented with 10% FBS. Both non-transduced and transduced cells were cultured in DMEM supplemented with 10% FBS and 2 μg/ml of puromycin. After all non-transduced cells died, the number of viable cells in the transduced population was counted to estimate virus titer.

Next, for lentiviral transduction, 1.0 Х 10 ⁶ HEK293T or HCT116 cells were inoculated into 100 mm dishes and cultured overnight. To achieve over 3000-fold coverage compared to the number of selected pegRNA-encoding plasmids, lentiviral libraries were transduced at an MOI of 0.3, and the next day, the culture medium was replaced with DMEM supplemented with 10% FBS and 2 μg/ml puromycin. replaced. In order to remove the non-transduced cells, the culture was maintained under the above conditions for 5 days.

Example 3: Delivery of PE2 or PE2 variants into a cell library

In order to transfer the PE2 variant prepared in Example 1 to the cell library A or B, the PE2 variant-encoding plasmid, pcDNA BSD-encoding plasmid, and puro-eGFP-encoding plasmid were mixed in a weight ratio of 10:1:1 for a total A plasmid mixture of 12 μg (for library A) or 24 μg (for library B) was made, and then Lipofectamine 2000 (Invitrogen) was used for a total of 1Х10 ⁶ cells from cell library A or a total of 6Х10 ⁶ cells from cell library B. Cells were transfected. After overnight incubation, the culture medium was replaced with DMEM containing 10% FBS and 40 μg/ml blasticidin S. After 5 days of culture, transfected cells were harvested using 0.25% trypsin for genomic DNA extraction and deep sequencing.

Example 4: Determination of Prime-Editing Activity at Endogenous Sites

To evaluate hyPE2 and PE2 activity at endogenous sites, HEK293T or HCT116 cells were seeded in 24-well plates and transfected at 70-80% cell density. Specifically, 750 ng of PE2-, 250 ng of pegRNA- and 100 ng of eGFP-Puro- (Addgene no. 45561) encoding plasmids were mixed and cells were co-transfected using Lipofectamine 2000 according to the manufacturer's protocol. The next day, the culture medium was replaced with DMEM supplemented with 10% FBS and 2 μg/ml puromycin. After 5 days of culture, transfected cells were harvested using 0.25% trypsin for genomic DNA extraction and deep sequencing.

Next, a skin punch biopsy was performed from the participants by a dermatologist after obtaining written consent from the study participants, who were healthy people. The consent procedure and research were approved by the institutional review committee of Yonsei University Health and Medical Center Severance Hospital (No. 4-2012-0028). Fibroblasts obtained from skin biopsies were cultured in DMEM containing 10% FBS and penicillin/streptomycin. A total of 1Х10 ⁶ human dermal fibroblasts were mixed with 3 μg of PE2-, 1 μg of pegRNA-, and 1 μg of eGFP-Puro encoding plasmid and electroporated using the Neon electroporation kit according to the manufacturer's protocol. After 5 days of culture, transfected cells were harvested using 0.25% trypsin for genomic DNA extraction and deep sequencing.

Example 5: Deep sequencing

The deep sequencing process used the method used in the previously published literature (Nature Biotechnology volume 39, pages 198-206 (2021)). Specifically, genomic DNA was extracted from the pelleted cells using the Wizard Genomic DNA Purification Kit (Promega) according to the manufacturer's protocol. To measure the prime editing efficiency for library experiments, a total of 16 μg (over 16,000x coverage) of genomic DNA was PCR-amplified using 2x pfu PCR Smart mix (Solgent). The resulting PCR product was combined and purified with a MEGAquick-spin total fragment DNA purification kit (iNtRON Biotechnology). Next, 20 ng of the purified product was PCR amplified using the Illumina adapter and primers containing the barcode sequence. To determine prime editing efficiency at endogenous sites, ~200 ng of individual genomic DNA samples were PCR amplified in a 20 μl reaction volume. The resulting PCR products were combined and purified. Next, 100 ng of the purified product was PCR amplified in a 20 μl reaction volume using primers containing the Illumina adapter sequence. The resulting product was purified and sequenced by MiniSeq (Illumina).

Example 6: Analysis of Prime Editing Activity

The prime-editing efficiency (i.e., intended editing frequency) of library experiments was calculated as follows using the Python script disclosed in the previously published literature 'Nature Biotechnology volume 39, pages198-206 (2021)':

[Equation 1]

To identify individual pegRNA and target-sequence pairs, a 22-nt sequence consisting of an 18-nt barcode and a 4-nt sequence upstream of the barcode was used. To improve the accuracy of the analysis, pegRNA and target sequence pairs with fewer than 100 deep sequencing reads were excluded. Reads containing the desired edit but no unintended mutations in the broad target sequence including the PAM were classified as PE2-induced mutations.

Cas-analyzer was used to evaluate the intended editing frequency, unintended editing frequency, and indel frequency in the endogenous region. Calculated like this:

[Equation 2]

[Equation 3]

[Equation 4]

For analysis of unintended substitutions near the target site, a 40 nt region ranging from -10 nucleotides (nts) to +25 nts from the nick site was evaluated for substitutions, and the average value was considered the number of reads for subsequent calculations.

In some cases, to avoid mathematical errors that may occur when PE2 efficiency is 0% and to attenuate insignificant fold increases, an adjusted fold increase with +0.1% added to both hyPE2 and PE2 efficiencies was used, Calculated as:

[Equation 5]

For example, a fold increase adjusted from 0.015% to 0.15% can be calculated as (0.15% + 0.1%)/(0.015% + 0.1%) = 2.2-fold instead of 10-fold. However, an increase from 1.5% to 15% can be calculated as (15% + 0.1%)/(1.5% + 0.1%) = 9.4 times, which is close to 10 times. When using adjusted fold increase instead of fold increase, this is noted in the legend for the relevant figure.

Example 7: Determination of prime editing activity at potential PE2 off-target sites

Potential PE2 off-target sites with up to 2 nucleotide mismatches or 1 nucleotide RNA or DNA overhangs were identified by Cas-OFFinder.

To evaluate the efficiency of prime editing at potential off-target sites, the genomic DNA samples used to measure prime editing activity at endogenous sites described above were used as templates for PCR amplification. The resulting product was purified and sequenced by MiSeq.

Example 8: Conventional Machine Learning-Based Model Training

Data of hyPE2- and PE-induced prime editing efficiencies obtained using library B were partitioned into training and test datasets through random sampling, ensuring that pegRNA and target sequences were not shared between the two datasets. Seven existing machine learning algorithms, XGBoost (extremegradient boosting), gradient-boosted regression tree (Boosted RT), random forest, L1-regularized linear regression, Lasso), L2-regularized linear regression (Ridge), L1L2-regularized linear regression (L1L2-regularized linear regression, ElasticNet), and SVM (support vector machine) were trained respectively. The above models were implemented using the XGBoost Python package (version 1.3.3) and scikit-learn (version 0.23.2). Site-independent and site-dependent nucleotides and dinucleotides, melting temperature, number of GCs, minimum self-folding free energy, DeepSpCas9 score (Kim, HK et al. SpCas9 activity prediction by DeepSpCas9, a deep learning A set of 1820 features including -based model with high generalization performance. Sci Adv 5 , eaax9249 (2019)) were extracted from a wide target sequence and PBS and RT template sequences. The MeltingTemp module (https://biopython.org/docs/1.74/api/Bio.SeqUtils.MeltingTemp.html) was used to calculate the melting temperature using default settings. Five-fold cross-validation was performed to select models from the normalization parameters and hyperparameter configurations of each algorithm. Here are the details for each machine learning algorithm.

For XGBoost and gradient-boosted regression trees, we searched more than 16 models selected from the following hyperparameter constructs: number of base estimators (selected from [50, 100]), maximum depth of individual regression estimators ([5, 10] ]), the minimum number of samples in a leaf node (choose from [1, 2]]), and the learning rate (choose from [0.1, 0.2] ).

For the random forest, we searched more than 16 models selected from the same hyperparameter configuration used in XGBoost except for the learning rate, and searched for the maximum number of features to consider when finding the best split ([all features, all square root of features, binary logarithm of all features]).

For L1-, L2 and L1L2-normalized linear regression, we searched at least 16 points equally spaced between 10 ^-6 and 10 ⁶ in log space to optimize the regularization parameters.

For SVM, more than 16 models were searched in the following hyperparameters: penalty parameter C and kernel parameter γ, 4 points equally spaced between 10 ^-3 and 10 ³ .

Example 9: Three-dimensional structural modeling

The structural model of hyPE2 shown in Fig. 23 was built with the Coot program (version WinCoot 0.9.6.1). From the structure of the SpCas9 DNA adeninebase editor (PDB code: 6VPC), a three-dimensional model of Cas9 complexed with guide RNA and target DNA fragments was obtained. To manually model the 3' extension of the 121-nt pegRNA (residues 83-121), RNAfold WebServer was used to predict the secondary structure of this region and adopted a hairpin structure for the residue. The pegRNA RT-template region hybridized with the 16-nt DNA primer region was modeled manually based on the structure of the XMRV RT complexed with an RNA:DNA hybrid (PDB code: 4HKQ). A three-dimensional model of Rad51 ssDBD (residues 16-85) was obtained from the N-terminal domain structure of Rad51 (PDB code: 1B22). Secondary structures were predicted using RaptorX to find putative α-helices in the flexible N- and C-terminal regions of Rad51 ssDBD, Linker A and Linker B. 23 showing the three-dimensional structure was generated using the UCSF Chimera program, and two linkers were displayed on the three-dimensional structure in consideration of the length and secondary structure. For the schematic structural model of hyPE2 shown in Supplementary Fig. 5a, we used the coordinates of Cas9 (PDB 4OO8), RT (PDB 5DMQ) and Rad51 (PDB 1B22). Structural images were prepared using the CueMol (version 2.2.3.443; http://www.cuemol.org) program.

Example 10: Statistical and Reproducibility Analysis

Data are expressed as mean ± standard deviation from independent experiments. P-values were calculated by post-hoc analysis with Tukey's multiple comparisons after two-tailed paired t-test, unpaired Student's t-test, or one-way ANOVA according to the number of independent variables. For high-throughput experiments, triplicates for library A and duplicates for library B and linker variants were independently performed. All replicates showed similar results. Individual evaluation experiments for HEK293T cells, HCT116 cells and human fibroblasts were independently performed in triplicate, and similar results were obtained.

Example 11: Data and Code Availability

Deep sequencing data of the present invention is in the NCBI Sequence Read Archive (SRA; https://www.ncbi.nlm.nih.gov/sra/) accession no. Submitted as SRP307854. Protein structure data for predicting the structures of PE2 and hyPE2 were obtained from Protein Data Bank (https://www.rcsb.org). Information about the PDB code is as follows:

6VPC - SpCas9 DNA adenine-based editor ( https://doi.org/10.2210/pdb6VPC/pdb ),

4HKQ - XMRV RT complexed with RNA:DNA hybrid ( https://doi.org/10.2210/pdb4HKQ/pdb )

1B22 - N-terminal domain of Rad51 ( https://doi.org/10.2210/pdb1B22/pdb )

4OO8 - Cas9 ( https://doi.org/10.2210/pdb4OO8/pdb )

5DMQ - reverse transcriptase (https://doi.org/10.2210/pdb5DMQ/pdb)

For the prime editing efficiency calculation, a Python script published in the previous literature (Nature Biotechnology volume 39, pages 198-206 (2021)) was used, which is available at https://github.com/hkimlab-PE/PE_SupplementaryCode .

Experimental Example 1: Development of hyPE2

When the Cas9-nikase domain and pegRNA bind to the target sequence during the prime editing process, single-stranded DNA (ssDNA) is generated and used as a primer for reverse transcription of the pegRNA RT-template region. Therefore, it was hypothesized that the addition of ssDBD (ssDNA binding protein domain) could improve the efficiency of prime editing by inducing stabilization of ssDNA.

Therefore, in order to develop hyPE2 with better prime editing efficiency than PE2, the following experiments were performed.

1.1: Evaluation of Prime Editing Efficiency with DBD Addition

In order to evaluate whether the prime editing efficiency is improved by adding DBD (DNA-binding domains) to the previously developed PE2, Rad51 DBD or RPA70-C, an ssDBD, was used between the Cas9 nickelase of PE2 and the RT domain. In addition, the PE2 variants were named PE2 mid_Rad51 (hyperPE2 or hyPE2) and PE2-mid_RPA70, respectively (FIG. 1).

On the other hand, when evaluating the activity of the variant in one or two target sequences, it is difficult to draw a generalized conclusion about the activity of the variant because the prime editing efficiency may vary greatly depending on the target sequence. In this regard, in a previous study, a total of 54,836 pairs of activities were determined using a lentiviral library of pegRNA-encoding sequences and target sequences (Nature Biotechnology volume 39, pages 198-206 (2021)). Therefore, 107 plasmids were randomly selected from library 1 of the 48,000 pairs of plasmids used in the previous study. A lentiviral library transduced into HEK293T cells was generated from the 107 pairs of plasmid libraries, and a cell library named library A was created. The pegRNA encoding sequence corresponding to the target sequence in library A was integrated into the genome with a lentivirus.

After the plasmid encoding hyPE2, PE2-mid_RPA70 or PE2 constructed above was transferred to the cell library, the prime editing efficiency was determined using deep sequencing. First, it was confirmed that a high correlation between biological replicates was shown ( FIG. 2 ), and the average of the prime editing efficiencies in three biological replicates was obtained and used for analysis. In addition, from the first 107 pairs, 24 pairs were excluded due to insufficient deep sequencing reads (< 100 reads), and 19 pairs showed an editing efficiency of 0% for PE2, which was useful for normalizing hyPE2 efficiency to PE2. It was excluded because it may indicate an error.

As a result of checking the PE-normalized prime editing efficiency for the remaining 64 pairs, the median increase was 2.4 times (range, 0-360 times) for hyPE2 and 1.5 times (range, 0-232 times) for PE2-mid_RPA70. (Fig. 3). Additionally, a higher median increase was seen for pairs exhibiting less than 1% editing efficiency in PE2 compared to pairs exhibiting greater than 1% efficiencies in PE2.

If the PE2 efficiency in the target sequence is too low, taking into account that the fold increase result of the variant calculated based on it is prone to error, a PE2 efficiency of less than 1% in the calculation to confirm the fold increase An additional 34 target sequences indicated were removed. As a result, it was confirmed that hyPE2 and PE2-mid_RPA70 showed a median increase of 1.6 times (range, 1.1 to 11 times) and 1.2 times (range, 0.57 to 6.6 times), respectively, compared to PE2 (FIG. 4). Also, for 34 pairs with PE2 efficiencies between 0% and 1%, 15 and 13 pairs, respectively, had hyPE2 (mean efficiency 11%, median 9.1%, range 1.1-29%) and PE2-mid_RPA70 (mean efficiency 7.4%) , with a median value of 5.3%, in the range of 1.1-24%), which showed a prime editing efficiency higher than 1%. Of the 19 pairs associated with 0% PE2 efficiency, 3 and 2 pairs, respectively, showed efficiencies greater than 1% in hyPE2 (1.8, 2.3 and 4.0%) and PE2-mid_RPA70 (1.2 and 1.5%), respectively.

Based on the above results, it can be seen that prime editing efficiency is improved when Rad51 DBD is added to PE2. In the following experiment, an experiment was performed based on adding Rad51 DBD.

1.2: Evaluation of Prime Editing Efficiency by Insertion Position of Rad51 DBD

In order to evaluate the prime editing efficiency according to the position where the Rad51 DBD is inserted, the following experiment was performed.

Specifically, mutants in which Rad51 DBD was inserted into the N or C-terminal region of PE2 were prepared, and these were named PE2-N_Rad51 and PE2-C_Rad51, respectively (FIG. 1). In the target sequence showing PE2 efficiency higher than 1%, as a result of confirming the prime editing efficiency of hyPE2 and the two variants compared to PE2, hyPE2 showed the highest overall activity (median value, 1.8 times higher than PE2 activity) , whereas PE2-N_Rad51 showed lower activity than PE2 (median, 0.85 fold), and PE2 C_Rad51 showed slightly higher overall activity than PE2 (median, 1.1 fold) ( FIG. 5 ). In addition, it was confirmed that the hyPE2 efficiency was higher compared to PE2 in all 33 target sequences showing more than 1% PE2 efficiency. Twenty of the remaining 55 target sequences (20/55 = 36%) with a prime editing efficiency of less than 1% by PE2 showed a prime editing efficiency of greater than 1% by hyPE2, with an average efficiency of 9.1% (median 5.7) %, range 1.1-29%) (Fig. 6).

Based on the above results, when Rad51 DBD is added to PE2, it can be seen that the prime editing efficiency of the mutant inserted between Cas9 nickase and RT is the best. In the following experiment, an experiment was performed based on hyPE2.

1.3: Evaluation of Prime Editing Efficiency in Different Cell Lines

In order to confirm whether the excellent prime editing efficiency of hyPE2 is also shown in other cell lines, the following experiment was performed.

Specifically, to evaluate the activity of hyPE2 in different cell lines, we generated an HCT116 cell library containing 107 pairs of pegRNA encoding sequences and target sequences. Similar to the above, 26 pairs with insufficient read count (<100 read count) and 38 pairs with PE2 efficiency <1% were excluded. As a result of evaluating the prime editing efficiency for the remaining 43 pairs, it was confirmed that hyPE2 exhibited a median value of 1.4 times (range, 0.82-2.9 times) than PE2, indicating higher efficiency than PE2 (FIG. 7).

1.4: Evaluating Prime Editing Efficiency According to the Linker

In order to evaluate the prime editing efficiency according to the type of linker used for Rad51 DBD insertion, the following experiment was performed.

Specifically, to evaluate the effect of the linker on the activity of hyPE2, six linker variants named hyPE2-AA, -AB, -BB, -AY, -AX and -XA were constructed (FIG. 8). As a result of comparing the activity of hyPE2 using cell library A based on the method described above, the mutant showed lower prime editing efficiency than hyPE2 (FIG. 9).

Based on the above results, it can be seen that using Linker B and Linker A for insertion into Rad51 DBD in PE2 can exhibit the best prime editing efficiency, and experiments were performed based on hyPE2 in the following experiments. In addition, amino acid sequence information of the protein/domain and linker included in the hyPE2 is described in Tables 1 and 2 below.

protein/ domain	amino acid sequence	order number
Cas9 H840A	DKKYSIGLDIGTNSVGWAVITDEYKVPSKKFKVLGNTDRHSIKKNLIGALLFDSGETAEATRLKRTARRRYTRRKNRICYLQEIFSNEMAKVDDSFFHMKRTADGSEFESPKKKRKVDKKYSIGLDIGTNSVGWAVITDEYKVPSKKFKVLGNTDRHSIKKNLIGALLFDSGETAEATRLKRTARRRYTRRKNRICYLQEIFSNEMAKVDDSFFHRLEESFLVEEDKKHERHPIFGNIVDEVAYHEKYPTIYHLRKKLVDSTDKADLRLIYLALAHMIKFRGHFLIEGDLNPDNSDVDKLFIQLVQTYNQLFEENPINASGVDAKAILSARLSKSRRLENLIAQLPGEKKNGLFGNLIALSLGLTPNFKSNFDLAEDAKLQLSKDTYDDDLDNLLAQIGDQYADLFLAAKNLSDAILLSDILRVNTEITKAPLSASMIKRYDEHHQDLTLLKALVRQQLPEKYKEIFFDQSKNGYAGYIDGGASQEEFYKFIKPILEKMDGTEELLVKLNREDLLRKQRTFDNGSIPHQIHLGELHAILRRQEDFYPFLKDNREKIEKILTFRIPYYVGPLARGNSRFAWMTRKSEETITPWNFEEVVDKGASAQSFIERMTNFDKNLPNEKVLPKHSLLYEYFTVYNELTKVKYVTEGMRKPAFLSGEQKKAIVDLLFKTNRKVTVKQLKEDY FKKIECFDSVEISGVEDRFNASLGTYHDLLKIIKDKDFLDNEENEDILEDIVLTLTLFEDREMIEERLKTYAHLFDDKVMKQLKRRRYTGWGRLSRKLINGIRDKQSGKTILDFLKSDGFANRNFMQLIHDDSLTFKEDIQKAQVSGQGDSLHEHIANLAGSPAIKKGILQTVKVVDELVKVMGRHKPENIVIEMARENQTTQKGQKNSRERMKRIEEGIKELGSQILKEHPVENTQLQNEKLYLYYLQNGRDMYVDQELDINRLSDYDVDAIVPQSFLKDDSIDNKVLTRSDKNRGKSDNVPSEEVVKKMKNYWRQLLNAKLITQRKFDNLTKAERGGLSELDKAGFIKRQLVETRQITKHVAQILDSRMNTKYDENDKLIREVKVITLKSKLVSDFRKDFQFYKVREINNYHHAHDAYLNAVVGTALIKKYPKLESEFVYGDYKVYDVRKMIAKSEQEIGKATAKYFFYSNIMNFFKTEITLANGEIRKRPLIETNGETGEIVWDKGRDFATVRKVLSMPQVNIVKKTEVQTGGFSKESILPKRNSDKLIARKKDWDPKKYGGFDSPTVAYSVLVVAKVEKGKSKKLKSVKELLGITIMERSSFEKNPIDFLEAKGYKEVKKDLIIKLPKYSLFELENGRKRMLASAGELQKGNELALPSKYVNFLYLASHYEKLKGSPEDNEQKQLFVEQHKHYLDEIIEQISEFSKRVILADANLDKVLSAYNKHRDKPIREQAENIIHLFTLTNLGAPAAFKYFDTTIDRKRYTSTKEVLDATLIHQSITGLYETRIDLSQLGGD	One

protein/ domain	amino acid sequence	order number
Rad51 ssDBD		AMQMQLEANADTSVEEESFGPQPISRLEQCGINANDVKKLEEAGFHTVEAVAYAPKKELINIKGISEAKADKILAEAAKLVPMGFTTATEFHQRRSEIIQITTGSKELDKLLQ	2
Reverse transcriptase	TLNIEDEYRLHETSKEPDVSLGSTWLSDFPQAWAETGGMGLAVRQAPLIIPLKATSTPVSIKQYPMSQEARLGIKPHIQRLLDQGILVPCQSPWNTPLLPVKKPGTNDYRPVQDLREVNKRVEDIHPTVPNPYNLLSGLPPSHQWYTVLDLKDAFFCLRLHPTSQPLFAFEWRDPEMGISGQLTWTRLPQGFKNSPTLFNEALHRDLADFRIQHPDLILLQYVDDLLLAATSELDCQQGTRALLQTLGNLGYRASAKKAQICQKQVKYLGYLLKEGQRWLTEARKETVMGQPTPKTPRQLREFLGKAGFCRLFIPGFAEMAAPLYPLTKPGTLFNWGPDQQKAYQEIKQALLTAPALGLPDLTKPFELFVDEKQGYAKGVLTQKLGPWRRPVAYLSKKLDPVAAGWPPCLRMVAAIAVLTKDAGKLTMGQPLVILAPHAVEALVKQPPDRWLSNARMTHYQALLLDTDRVQFGPVVALNPATLLPLPEEGLQHNCLDILAEAHGTRPDLTDQPLPDADHTWYTDGSSLLQEGQRKAGAAVTTETEVIWAKALPAGTSAQRAELIALTQALKMAEGKKLNVYTDSRYAFATAHIHGEIYRRRGWLTSEGKEIKNKDEILALLKALFLPKRLSIIHCPGHQKGHSAEARGNRMADQAARKAAITETPDTSTLLIENSSPSGGSKRTADGSEFEPKKKRKV*	3
Linker B		SGGSPKKKRKVGSSGS	4
Linker A		SGGSSGGSSGSETPGTSESATPESSGGSSGGSS	5

Experimental Example 2: Evaluation of the activity of hyPE2 in endogenous targets

To evaluate the prime editing efficiency of hyPE2 in endogenous targets, the following experiments were performed.

Specifically, the efficiency of hyPE2 and PE2 at 63 and 51 endogenous target sites in HEK293T and HCT116 cells, respectively, was evaluated. As a result, PE2 efficiency was higher than 1% in 31 and 11 targets for HEK293T and HCT116 cells, respectively, and the efficiency of hyPE2 at the target site was 1.4 times higher than that of PE2 in HEK293T and HCT116 cells, respectively (median, range). 0.89 to 2.2 times) and 1.5 times (median, range 1.0 to 2.6 times) higher ( FIGS. 10 to 12 ). Specifically, the prime editing efficiency by hyPE2 increased from 5.8% to 13% in HEK293T cells and from 1.1% to 2.8% in HCT116 cells.

In addition, for the remaining 32 and 40 target sites in HEK293T and HCT116 cells with PE2 efficiencies less than 1%, 11 target sequences (34% = 11/32) of HEK293T cells were 1.6% (median of 1.4%, range 1.1) The hyPE2-induced prime editing efficiency was higher than 1% with an average efficiency of -2.9%), whereas the hyPE2 efficiencies in HCT116 cells were all less than 1%. Based on the above results, it can be seen that hyPE2 has better overall activity than PE2.

Next, we compared the efficiency of hyPE2 with that of PE2 in six target sequences of a therapeutically relevant cell type, primary human dermal fibroblasts. As a result, compared to PE2, hyPE2 showed higher prime editing efficiency in 5 of 6 targets, and it was confirmed that the mean and median increases in 6 targets were 2.1-fold and 2.0-fold (adjusted fold), respectively ( 13 and 14).

Experimental Example 3: Unintentional editing and off-target effect level evaluation by hyPE2

The following experiment was performed to determine whether hyPE2 induces unintended edits at a higher level than PE2 in the endogenous target sequence, and adjusted fold increase to compare the frequency of unintended edits. has been adopted As a result, when hyPE2 was used instead of PE2, the level of unintentional substitution and editing such as indels slightly increased, but the fold increase level of the above unintentional editing was lower than or similar to the fold increase level of the intended editing. of (Fig. 15).

Next, the effect of off-target editing was evaluated by analyzing potential off-target sites of pegRNAs that showed relatively high on-target prime editing efficiency in HEK293T cells. Specifically, we first identified potential off-target sites with up to two nucleotide mismatches or one nucleotide RNA or DNA overhang per pegRNA, and found a total of six potential off-target sites for three pegRNAs. In addition, in an initial study of prime editing, a total of 16 pairs of pegRNAs and 4 HEK4-targeting pegRNAs associated with off-target sites were used to determine prime editing efficiency ( FIG. 16 ). As a result of evaluation of these 22 (= 6 + 16) pairs of pegRNAs and potential off-target sites in HEK293T cells, it was confirmed that the off-target effect of hyPE2 was similar to that of PE2 (Fig. 16 and 17).

Based on the above results, it can be seen that hyPE2 does not specifically induce unintended editing and off-target effects compared to PE2.

Experimental Example 4: Development of PEselector

Considering that the increase in prime editing efficiency induced by hyPE2 is higher in some pegRNAs through the results of the above experimental examples, the following experiment was performed to predict the fold increase in prime editing efficiency by hyPE2.

Specifically, we compared the hyPE2- and PE2-induced prime editing efficiencies using the high-throughput method described above using 665 pairs of pegRNAs and an integrated target sequence (Library B). First, when the fold increase data derived from libraries A and B were combined, the mean-fold and median-fold increase were 1.3-fold and 1.2-fold, respectively ( FIG. 18 ).

Next, the data in library B was randomly partitioned into a training data set and a test data set, ensuring that neither the pegRNA nor the target sequence was shared between the two data sets. Using the training data set of hyPE2- and PE2-induced prime editing efficiencies determined with 568 pegRNAs, we generated 7 computer models predicting a fold increase in hyPE2-induced prime editing efficiencies compared to PE2, wherein the 7 models It was confirmed that the SVM-based model showed the highest accuracy ( FIG. 19 ). The model name was named PEselector, and the model is provided at http://deepcrispr.info/PEselector .

Experimental Example 5: Confirmation of factors related to hyPE2 efficiency

To check the factors related to the multiple increase of the hyPE2-induced prime editing efficiency, the Tree SHAP method (SHapley Additive explanations integrated with the XGBoost algorithm) was used (Lundberg, SM et al. From local explanations to global understanding with Explainable AI for trees. Nature Machine Intelligence 2 , 56-67 (2020)), it was confirmed that the most important factor among the 1820 characteristics was the melting temperature of the primer-binding site (PBS) ( FIG. 20 ). In addition, when it was confirmed how the fold increase in hyPE2 efficiency depends on the melting temperature of PBS, it was confirmed that the fold increase tends to decrease as the PBS melting temperature increases ( FIG. 21 ), when the PBS melting temperature is low It can be seen that hyPE2 will be particularly useful for

Next, to identify a potential mechanism for the higher prime editing efficiency of hyPE2, we predicted the three-dimensional structures of PE2 and hyPE2 (Figs. 22 and 23). As a result, given that Rad51 DBD can bind to both ssDNA and RNA and promote the formation of a DNA/RNA hybrid, Rad51 promotes the binding of pegRNA to the target ssDNA with nicking and reverse transcription. It can be seen that it can be improved. This potential mechanism is consistent with the above results that the improvement of prime editing efficiency is particularly enhanced when the PBS melting temperature is low, which may be linked to the poor binding of the pegRNA PBS domain to the target ssDNA with the nick site when using PE2. .

The above description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

Claims (13)

1) a fusion protein comprising a Cas nickase, a reverse transcriptase (RT) and a single-stranded DNA-binding domain (ssDBD); and

2) Prime editing guide RNA (pegRNA)

comprising, a gene editing complex.

The complex according to claim 1, wherein the complex is for prime editing.

The complex of claim 1, wherein the single-stranded DNA binding domain is Rad51 DBD.

The complex according to claim 1, wherein the single-stranded DNA binding domain is linked to the C-terminus of the Cas nickase and the N-terminus of the reverse transcriptase.

The complex according to claim 1, wherein the Cas nickcase comprises the amino acid sequence of SEQ ID NO: 1.

The complex of claim 1, wherein the reverse transcriptase comprises the amino acid sequence of SEQ ID NO: 3.

The complex of claim 1, wherein the single-stranded DNA binding domain comprises the amino acid sequence of SEQ ID NO:2.

The complex according to claim 1, wherein the single-stranded DNA binding domain is linked to the C-terminus of the Cas nickase via a first linker comprising the amino acid sequence of SEQ ID NO: 4.

The complex according to claim 1, wherein the single-stranded DNA binding domain is linked to the N-terminus of the reverse transcriptase through a second linker comprising the amino acid sequence of SEQ ID NO: 5.

The complex of claim 1, wherein the fusion protein further comprises a nuclear localization sequence or signal (NLS).

1) a fusion protein comprising a Cas nickase, a reverse transcriptase (RT) and a single-stranded DNA-binding domain (ssDBD), a polynucleotide encoding the fusion protein, or the a recombinant vector comprising a polynucleotide; and

2) Recombinant vector containing prime editing guide RNA (pegRNA) or polynucleotide encoding the same

A composition for gene editing comprising a.

The composition of claim 11 , wherein the composition is for editing a target DNA or gene in ex vivo or in vivo.

A method of editing a gene of a subject, comprising introducing the composition for gene editing of claim 11 into isolated eukaryotic cells or eukaryotic organisms other than humans.

Images