lipid nanoparticles

Abstract

The present invention addresses the problem of providing a lipid nanoparticle in which a nucleic acid, etc. required in genome editing is encapsulated and which can be produced by an alcohol dilution method using flow channels and contributes to high genome editing efficiency. The present invention pertains to a lipid nanoparticle which comprises a lipid component, a DNA nuclease, a guide RNA and a single-stranded oligonucleotide, wherein: the lipid component comprises a pH-sensitive cationic lipid, a neutral phospholipid and a polyalkylene glycol-modified lipid; the ratio of the pH-sensitive cationic lipid relative to the total lipids constituting the lipid nanoparticle is 30-50 mol%; the ratio of the neutral phospholipid relative to the total lipids constituting the lipid nanoparticle is 20-50 mol%; and the ratio of the polyalkylene glycol-modified lipid relative to the total lipids constituting the lipid nanoparticle is 1-4 mol%.

Description

[Technical Field]

[0001] The present invention relates to lipid nanoparticles that are useful as carriers of RNA-protein complexes (ribonucleoproteins; RNPs) used in genome editing using the CRISPR system. This application claims priority based on Japanese Patent Application No. 2019-101203, filed on May 30, 2019, the contents of which are incorporated herein by reference. [Background technology]

[0002] Genome editing is a biotechnology that enables selective introduction of mutations or insertion of any DNA sequence, including genes, into any desired genomic DNA region. Genome editing is highly anticipated for its potential to fundamentally cure various intractable diseases, including hereditary and infectious diseases. The CRISPR (clustered regularly interspaced short palindromic repeats) / Cas9 (CRISPR-associated proteins 9) system, a third-generation genome editing technology, is currently attracting the most attention due to its excellent gene knockout efficiency and simple design. This system functions as an RNP consisting of the Cas9 protein, which has DNA double-strand break (DSB) activity, and a chimeric RNA called a gRNA (guide RNA), which is composed of bacterial crRNA (CRISPR RNA) and tracrRNA (trans-activating CRISPR RNA). Therefore, targeted gene knockout can be induced by expressing the RNP in target cells or by delivering the RNP itself into target cells. Furthermore, gene knock-in can be induced by simultaneously delivering donor DNA.

[0003] Genome editing technology can deliver RNPs into target cells by introducing DNA or RNA to express them intracellularly, or by directly introducing the RNPs themselves into cells. The former method is relatively easy because delivery technologies, such as viral and non-viral vectors, are well established. However, the relatively long expression time of the Cas9 protein makes it prone to introducing mutations into unintended DNA regions (off-target effects). On the other hand, the latter method minimizes off-target effects because the RNPs rapidly degrade and disappear after introducing mutations into the target DNA region. In particular, genome editing involving DSBs has a permanent effect as long as the edited cell survives, making it extremely important to suppress off-target effects.

[0004] Several reports have been published on the development of technologies capable of delivering RNP itself. For example, there are a method using DNA nanoclews (Non-Patent Document 1), a method using bioreducible lipid nanoparticles (Non-Patent Document 2), a method using conjugates with gold nanoparticles (Non-Patent Document 3), a method using lipidoids (Non-Patent Document 4), and a method using lecithin nano-liposomal particles (Non-Patent Document 5). Furthermore, CRISPR-Gold (Non-Patent Document 6) has been reported as an example of inducing gene knock-in. However, all of these methods require high concentrations of Cas9 to induce gene knockdown in cultured cells, and there remain issues with genome editing efficiency.

[0005] On the other hand, some methods for producing lipid nanoparticles encapsulating nucleic acids and other molecules are based on the alcohol dilution method using a flow channel. For example, it has been reported that lipid nanoparticles with a diameter of approximately 30 nm can be reproducibly produced by using a microchannel with a built-in three-dimensional micromixer, which can achieve instantaneous mixing of two liquids (Non-Patent Document 7). It has also been reported that a nanosized lipid particle formation system with higher particle size controllability than conventional flow channel structures using a three-dimensional mixer can be formed by using a simple two-dimensional flow channel structure in which baffles of a fixed width relative to the flow channel width are arranged alternately on both sides of a micro-sized flow channel through which a raw solution flows (Patent Document 1). These nanoparticle formulation production methods have recently been adopted primarily for the production of lipid nanoparticles (LNPs) loaded with lipid-soluble drugs or nucleic acids such as short interfering RNA (siRNA) or mRNA. For example, lipid nanoparticles containing pH-sensitive cationic lipids as constituent lipids have been reported as lipid nanoparticles that serve as carriers for the efficient delivery of nucleic acids such as siRNA into target cells (Patent Document 2). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2018 / 190423 [Patent Document 2] International Publication No. 2018 / 230710 [Non-patent literature]

[0007] [Non-Patent Document 1] Sun et al., Angewandte Chemie International Edition, 2015, vol.54, p.12029-12033. [Non-patent document 2] Wang et al., Proceedings of the National Academy of Sciences of the United States of America, 2016, vol.113, p.2868-2873. [Non-patent document 3] Mout et al., American Chemical Society Nano, 2017, vol.11, p.2452-2458. [Non-patent document 4] Li et al., Biomaterials, 2018, vol.178, p.652-662. [Non-patent document 5] Cho et al., Journal of Nanobiotechnology, 2019, vol.17, p.19. [Non-patent document 6] Lee et al., Nature Biomedical Engineering, 2017, vol.1, p.889-901. [Non-Patent Document 7] Leung et al.,Journal of Physical Chemistry C Nanomater Interfaces,2012,vol.116(34),p.18440-18450. [Non-patent document 8] Stroock et al., Science, 2002, vol.295, p.647-651 Summary of the Invention [Problem to be solved by the invention]

[0008] Compared to previously used small molecule drugs and nucleic acids, proteins such as RNPs are more susceptible to irreversible inactivation due to various physical parameters during particle production, such as organic solvents such as alcohol, buffer pH, salt concentration, and temperature. For this reason, there have been no reports to date on the production of RNP-loaded lipid nanoparticle formulations based on the alcohol dilution method.

[0009] The present invention aims to provide lipid nanoparticles that contain nucleic acids and other substances necessary for genome editing, can be produced by an alcohol dilution method using a flow channel, and have excellent genome editing efficiency. [Means for solving the problem]

[0010] The present inventors discovered that in genome editing using the CRISPR / Cas9 system, RNP, a complex of crRNA, tracrRNA, and Cas9 protein, can be efficiently loaded onto lipid nanoparticles consisting of a lipid membrane structure of a specific composition containing pH-sensitive cationic lipids by further complexing it with a single-stranded oligonucleotide (ssON) containing a base sequence region complementary to the crRNA and making it negatively charged, thereby completing the present invention.

[0011] That is, the present invention provides the following lipid nanoparticles: [1] A method for preparing a nucleic acid sequence comprising: a lipid component; a DNA nuclease; a guide RNA; and a single-stranded oligonucleotide; The lipid component is represented by the following general formula (I):

[0012] [ka]

[0013] [In formula (I), a represents an integer of 3 to 5; b represents 0 or 1; R 1 and R 2 are each independently represented by the following general formula (A):

[0014] [ka]

[0015] (In formula (A), q represents an integer of 1 to 9; r represents 0 or 1; s represents an integer of 1 to 3; t represents 0 or 1; u represents an integer of 1 to 8; c represents 0 or 1; and v represents an integer of 4 to 12; and q+2r+s+2t+u+c+v is an integer of 19 or greater, except that when b and c are simultaneously 0, q is an integer of 3 to 5, r and t are 1, s is 1, and u+v is an integer of 6 to 10.) X represents a group represented by the following general formula (B):

[0016] [ka]

[0017] (In formula (B), d represents an integer of 0 to 3; R 3 and R 4 are each independently C 1-4 Alkyl group or C 2-4 Alkenyl group (C 1-4 Alkyl group or C 2-4 The alkenyl group may have one or two hydrogen atoms substituted by a phenyl group, but R 3 and R 4 are bonded to each other to form a 5- to 7-membered non-aromatic heterocyclic ring (one or two hydrogen atoms of the ring are C 1-4 Alkyl group or C 2-4 may be substituted with an alkenyl group) or a 5- to 7-membered non-aromatic heterocyclic group (wherein the group is bonded to (O-CO)b- via a carbon atom and one or two hydrogen atoms of the ring are 1-4 Alkyl group or C 2-4 (which may be substituted with an alkenyl group) The composition contains a pH-sensitive cationic lipid represented by the formula: The ratio of the pH-sensitive cationic lipid to the total amount of lipids constituting the lipid nanoparticles is 30 to 50 mol %, The ratio of the neutral phospholipid to the total amount of lipids constituting the lipid nanoparticles is 20 to 50 mol%, Lipid nanoparticles, wherein the proportion of the polyalkylene glycol-modified lipid relative to the total amount of lipids constituting the lipid nanoparticles is 1 to 4 mol %. [2] The lipid nanoparticles according to [1], wherein the neutral phospholipid is a neutral glycerophospholipid having a saturated or unsaturated fatty acid residue having 12 to 24 carbon atoms. [3] The lipid nanoparticles according to [1], wherein the neutral phospholipid is a phosphatidylethanolamine having an unsaturated fatty acid residue having 12 to 24 carbon atoms. [4] The lipid nanoparticles according to any one of [1] to [3] above, wherein the pH-sensitive cationic lipid is a polyethylene glycol-modified lipid. [5] The DNA nuclease is a Cas9 protein; The lipid nanoparticle according to any one of [1] to [4] above, wherein the guide RNA consists of crRNA and tracrRNA. [6] The lipid nanoparticle of [5], wherein the Cas9 protein is a protein having only one of RuvC nuclease activity and HNH nuclease activity. [7] The lipid nanoparticle of any one of [1] to [4] above, wherein the DNA nuclease is a Cpf1 protein. [8] A genome editing method, comprising introducing any one of the lipid nanoparticles described in [1] to [7] into a cell. [9] A method for producing the lipid nanoparticles according to any one of [1] to [7] above, using a flow channel structure, the flow path structure includes a first inlet path for introducing a first fluid and a second inlet path for introducing a second fluid, which are independent of each other and have a certain length, and which join together to form a single dilution flow path; the dilution flow path has a flow path portion that is two-dimensionally curved in at least a part thereof, When the axial direction or extension direction of the dilution flow path upstream of the bent flow path portion is defined as the X direction, the width direction of the dilution flow path perpendicularly intersecting the X direction is defined as the Y direction, and the flow path width of the dilution flow path upstream of the bent flow path portion is defined as y0, the bent flow path portion is formed by at least two or more structures that protrude alternately from both side wall surfaces of the dilution flow path opposing each other in the Y direction toward the flow path center in approximately the Y direction (approximately the +Y direction, approximately the -Y direction), have constant heights h1, h2... of at least 1 / 2y0 but less than 1y0, and have constant widths x1, x2... in the X direction, and that regulate the flow path width of the dilution flow path, and are provided at constant intervals d1, d2... A method for producing lipid nanoparticles, wherein a lipid solution in which the lipid components are dissolved in ethanol is introduced through the first introduction path, and an aqueous solution containing the DNA nuclease, the guide RNA, and the single-stranded oligonucleotide and having a pH of 5.0 or higher is introduced through the second introduction path, at a total flow rate of 1 μL / min to 100 mL / min, and the ratio of the flow rate of the aqueous solution to the flow rate of the lipid solution is 7 or more.

[10] The flow path structure further includes a third inlet path for introducing a third fluid, The method for producing lipid nanoparticles according to [9], wherein the first inlet channel, the second inlet channel, and the third inlet channel each have a certain length and merge to form a single dilution channel so that the first fluid introduced from the first inlet channel comes into contact with the third fluid introduced from the third inlet channel before merging with the second fluid introduced from the second inlet channel. [Effects of the Invention]

[0018] The lipid nanoparticles of the present invention can efficiently perform genome editing in target cells because the RNP is loaded onto lipid nanoparticles having a lipid membrane structure of a specific composition containing a pH-sensitive cationic lipid. Furthermore, the lipid nanoparticles of the present invention can be produced by the alcohol dilution method because the RNP loaded onto the lipid membrane structure contains ssON and is negatively charged. [Brief explanation of the drawings]

[0019] [Figure 1A]1 is a diagram schematically illustrating the structure of one embodiment of a flow channel structure used in the production of lipid nanoparticles according to the present invention. [Figure 1B] 1 is a diagram schematically illustrating the structure of one embodiment of a flow channel structure used in the production of lipid nanoparticles according to the present invention. [Figure 2] 1 is a schematic diagram of the structure of a flow path structure used in Reference Example 1 and Example 1. FIG. [Figure 3] FIG. 10 is a schematic diagram of the structure of a flow path structure used in Reference Example 2. [Figure 4] FIG. 1 shows the results of measuring the GFP knockout efficiency (%) of each RNP-loaded lipid nanoparticle in HeLa-GFP cells in Example 1. [Figure 5] This figure shows the results of measuring the ratio of knock-in efficiency (%) to knock-out efficiency (%) ([KI (%)] / [KO (%)]) for each RNP-loaded lipid nanoparticle in HeLa-GFP cells in Example 1. [Figure 6] FIG. 1 shows the results of measuring the GFP knockout efficiency (%) of each RNP-loaded lipid nanoparticle in HeLa-GFP cells in the primary screening in Example 2. [Figure 7] FIG. 10 is a diagram showing predicted profiles of number average particle diameter when factors that significantly affect the number average particle diameter in primary screening are changed in Example 2. [Figure 8] FIG. 1 shows a predicted profile of the encapsulation rate when factors that significantly affect the encapsulation rate in the primary screening are changed in Example 2. [Figure 9] FIG. 1 shows predicted profiles of gene knockout activity when factors that significantly affect gene knockout activity in primary screening are changed in Example 2. [Figure 10] FIG. 10 is a diagram showing predicted profiles of number average particle diameter when factors that significantly affect the number average particle diameter in the secondary screening are changed in Example 2. [Figure 11]FIG. 1 shows a predicted profile of the encapsulation rate when factors that significantly affect the encapsulation rate in the secondary screening are changed in Example 2. [Figure 12] FIG. 1 shows predicted profiles of gene knockout activity when factors that significantly affect gene knockout activity in secondary screening are changed in Example 2. [Figure 13] FIG. 10 shows the results of measuring the cell viability (%) of cells to which RNP-loaded lipid nanoparticles B-4 or B-9 were added in Example 3. [Figure 14] FIG. 10 shows the results of measuring the zeta potential and PdI over time in Example 3, in which cells to which RNP-loaded lipid nanoparticles B-9 had been added were stored at 4° C. [Figure 15] FIG. 10 shows the results of measuring the GFP knockout efficiency (%) of RNP-loaded lipid nanoparticles B-9 in HeLa-GFP cells before and after storage at 4° C. in Example 3. [Figure 16A] In Example 4, this figure shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured in the medium containing RNP-loaded lipid nanoparticles B-4 or B-9 so that the Cas9 protein concentration was 0.1 nM. [Figure 16B] In Example 4, this figure shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured in the medium containing RNP-loaded lipid nanoparticles B-4 or B-9 so that the Cas9 protein concentration was 0.3 nM. [Figure 17] FIG. 10 shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured in Example 5 after adding RNP-loaded lipid nanoparticles to the medium so that the Cpf1 protein concentration was 0.5, 1, or 2 nM. [Figure 18]In Example 6, this figure shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured in a medium containing RNP-loaded lipid nanoparticles containing Cas9n protein or RNP-loaded lipid nanoparticles containing Cas9 protein, with the addition of Cas9n protein or other RNP-loaded lipid nanoparticles to a concentration of 0.1, 0.3, 1, or 2 nM. [Figure 19A] In Example 7, this figure shows the results of measuring the GFP knockout efficiency (%) in HEK-GFP cells cultured in the medium containing RNP-loaded lipid nanoparticles B-4 or B-9 at a Cas9 protein concentration of 0.5, 1, 3, or 5 nM. [Figure 19B] In Example 7, this figure shows the results of measuring the percentage (%) of BFP-positive cells in HEK-GFP cells cultured in the medium containing RNP-loaded lipid nanoparticles B-4 or B-9 at a Cas9 protein concentration of 0.5, 1, 3, or 5 nM. [Figure 20] FIG. 10 shows the results of measuring the cell viability (%) of cells to which RNP-loaded lipid nanoparticles B-4 or B-9 were added in Example 7. [Figure 21] In Example 8, (A) shows the results of flow cytometry of BMDMs constitutively expressing GFP before introduction of RNP-loaded lipid nanoparticles, and (B) shows the results of flow cytometry of BMDMs constitutively expressing GFP after introduction of RNP-loaded lipid nanoparticles. [Figure 22] In Example 9, this figure shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured in the presence of each RNP-loaded lipid nanoparticle B-9 added to the medium so that the Cas9 protein concentration was 0.1, 0.3, or 1 nM. [Figure 23A] FIG. 10 shows the results of measuring the GFP knockout efficiency (%) in HeLa-GFP cells cultured with RNP-loaded lipid nanoparticles containing Cas9 protein in Example 10. [Figure 23B]This is a figure showing the measurement results of the GFP knockout efficiency (%) in HeLa-GFP cells cultured by adding RNP-loaded lipid nanoparticles containing the Cpf1 protein in Example 10.

Mode for Carrying Out the Invention

[0020] The lipid nanoparticles according to the present invention are used as carriers for introducing RNP into target cells in the CRISPR system, and are lipid nanoparticles in which an RNP, which is a complex of a DNA nuclease, a guide RNA (gRNA), and a ssON used for genome editing, is loaded on a lipid membrane structure having a specific composition containing a pH-sensitive cationic lipid. Examples of the RNP loaded on the lipid nanoparticles include a complex of a Cas9 protein, a crRNA, a tracrRNA, and a ssON. Since the RNP itself is directly loaded on the lipid nanoparticles and introduced into target cells, the off-target effect is smaller than the method of introducing and expressing a gene encoding a DNA nuclease such as the Cas9 protein into the target cells.

[0021] [DNA Nuclease] In the present invention and the present specification, the DNA nuclease loaded on the lipid nanoparticles is an enzyme that binds to DNA in a gRNA-dependent manner and recognizes and cleaves double-stranded DNA formed by pairing with a part of the gRNA. Examples of the DNA nuclease include Cas9, Cpf1, and the like.

[0022] [Cas9 Protein] In the present invention and the present specification, the Cas9 protein is a protein that binds to DNA in a gRNA-dependent manner and has at least one of RuvC nuclease activity and HNH nuclease activity. The Cas9 protein having both RuvC nuclease activity and HNH nuclease activity cleaves the double strand of genomic DNA. The Cas9 protein having only one of RuvC nuclease activity and HNH nuclease activity cleaves only one of the two strands of the double strand of genomic DNA.

[0023] The Cas9 protein used in the present invention may be a wild-type Cas9 protein derived from bacteria having a CRISPR system, or may be a mutant protein obtained by modifying the wild-type protein. Examples of bacteria having a CRISPR system include Streptococcus pyogenes, Staphylococcus aureus, Neisseria meningitidis, Campylobacter jejuni, Geobacillus stearothermophilus, Streptococcus thermophilus, Treponema denticola, etc. Examples of mutant proteins obtained by modifying the wild-type Cas9 protein include mutants into which a mutation that inactivates either the RuvC nuclease activity or the HNH nuclease activity has been introduced. Examples of such mutants include a mutant (Cas9(D10A)) in which the 10th aspartic acid of the wild-type Cas9 protein is substituted with alanine. Since Cas9(D10A) functions as a DNA nickase, it is also called Cas9 nickase (Cas9n). In addition, a mutant protein into which a mutation that does not affect the nuclease activity of the wild-type Cas9 protein has been introduced may also be used.

[0024] The Cas9 protein used in the present invention may be a wild-type Cas9 protein or a mutant thereof to which various peptides are added, or may be a chimeric protein fused with another protein. Examples of such peptides include tag peptides such as His tag, Myc tag, and Flag tag, and signal peptides such as nuclear localization signal peptides. Examples of other proteins to be fused with the wild-type or mutant Cas9 protein include GST, fluorescent proteins, etc.

[0025] <Cpf1 protein> In the present invention and the present specification, the Cpf1 protein is a protein that binds to DNA in a gRNA-dependent manner and has only RuvC nuclease activity. While the Cas9 protein, which has both RuvC and HNH nuclease activities, cleaves the double strand of genomic DNA to form blunt ends, the Cpf1 protein forms 5'-overhanging ends.

[0026] The Cpf1 protein used in the present invention may be a wild-type Cpf1 protein derived from bacteria having the CRISPR / Cpf1 system, or a mutant protein obtained by modifying the wild-type protein. Examples of bacteria having the CRISPR / Cpf1 system include Acidaminococcus sp. and Lachnospiraceae bacterium. Furthermore, mutant proteins obtained by modifying wild-type Cpf1 proteins include mutants into which a mutation that enhances nuclease activity has been introduced, and mutant proteins into which a mutation that does not affect nuclease activity has been introduced.

[0027] The Cpf1 protein used in the present invention may be a wild-type Cpf1 protein or a mutant thereof to which various peptides have been added, or may be a chimeric protein fused with another protein, such as a peptide or other protein similar to those listed for the Cas9 protein.

[0028] [gRNA] In the present invention and the present specification, gRNA is an RNA having a base sequence capable of pairing with a target sequence (a base sequence to be edited) on genomic DNA to be cleaved by a DNA nuclease. A "base sequence capable of pairing with a target sequence" is typically a base sequence that is homologous (identical) or complementary to the target sequence in order to suppress recognition of regions other than the target sequence. A gRNA may consist of a single RNA or may be a complex of two or more RNAs.

[0029] <crRNA and tracrRNA> When the DNA nuclease loaded on the lipid nanoparticle is Cas9 protein, as the gRNA, crRNA and tracrRNA can be used. crRNA is derived from bacteria having the CRISPR system and is a single-stranded RNA containing a region (binding region with tracrRNA) consisting of a base sequence complementary to a part of tracrRNA and a region (genomic DNA binding region) consisting of a base sequence homologous or complementary to the target sequence on genomic DNA. tracrRNA is also derived from bacteria having the CRISPR system and has a region (binding region with crRNA) consisting of a base sequence complementary to a part of crRNA, and is a single-stranded RNA that hybridizes with crRNA in this region to form a hairpin structure. The Cas9 protein recognizes the hairpin structure and an RNP is formed. crRNA and tracrRNA may each be an independent single-stranded RNA, or may be a single-stranded RNA in which both are linked via an appropriate RNA linker. As the bacteria having the CRISPR system, those as described above can be mentioned. The Cas9 protein, crRNA and tracrRNA may all be derived from the same kind of bacteria, or may be derived from mutually different kinds of bacteria. Further, crRNA and tracrRNA may consist only of natural RNA as long as the function of the CRISPR / Cas9 system is not impaired, or a modified RNA or an artificial nucleic acid may be contained in a part or all of them.

[0030] <gRNA used with Cpf1 protein> When the DNA nuclease loaded on the lipid nanoparticle is Cpf1 protein, as the gRNA, it may be an RNA containing a target sequence similar to crRNA, and tracrRNA is unnecessary. For this reason, the gRNA can be made shorter than when using Cas9 protein. Further, the gRNA may consist only of natural RNA as long as the function of the CRISPR / Cpf1 system is not impaired, or a modified RNA or an artificial nucleic acid may be contained in a part or all of it.

[0031] <Target sequence> The target sequence is typically selected from a region of the base sequence immediately followed by a PAM sequence. The PAM sequence is a sequence recognized by a DNA nuclease such as the Cas9 protein, and is determined depending on the DNA nuclease, such as the Cas9 protein, used. For example, a PAM sequence recognized by the Cas9 protein is 5'-NGG (N:A, G, C, T), and a PAM sequence recognized by the Cpf1 protein is 5'-TTTV (V:A, G, C) or 5'-TTTN (N:A, G, C, T). The base length of the target sequence in a gRNA such as a crRNA is not particularly limited and can be, for example, approximately 15 to 30 bases long, with 18 to 22 bases being preferred.

[0032] [ssON] The ssON used in the present invention contains a region capable of pairing with a portion of the gRNA. This allows it to contain a region capable of pairing with the gRNA in a state in which it is complexed with the DNA nuclease loaded on the lipid nanoparticle. This results in the formation of an RNP, which is a complex of the DNA nuclease, the gRNA, and the ssON. The ssON may be an oligonucleotide consisting of only DNA, an oligonucleotide consisting of only RNA, or a chimeric oligonucleotide containing both DNA and RNA. Furthermore, the ssON may be composed of only natural RNA or DNA, as long as it does not impair the function of the CRISPR system used, and may contain modified nucleic acids or artificial nucleic acids in part or in whole.

[0033] When the DNA nuclease loaded onto the lipid nanoparticle is a Cas9 protein, it contains a region of the crRNA other than the region that binds to the tracrRNA and a region consisting of a complementary base sequence. The ssON and crRNA hybridize via this region (the crRNA-binding region) that is complementary to the crRNA. In other words, the RNP loaded onto the lipid nanoparticle of the present invention is a complex formed by hybridization of the crRNA with the tracrRNA and the crRNA with the ssON, and the Cas9 protein complexed with the resulting ternary complex. The region in the crRNA that hybridizes with the ssON (the ssON-binding region) is not particularly limited, as long as the crRNA can simultaneously hybridize with the tracrRNA and the ssON to form a ternary complex, and this ternary complex can form a hairpin structure recognized by the Cas9 protein. For example, the ssON-binding region in the crRNA may include part or all of the genomic DNA-binding region, or may be identical to the genomic DNA-binding region. In the double-nicking method, which uses Cas9 nickase (Cas9n) in which either RuvC or HNH nuclease activity has been inactivated, two types of guide RNA are used. Therefore, in the case of lipid nanoparticles loaded with RNP for the double-nicking method, two types of ssON are used that hybridize to each guide RNA.

[0034] When knocking in a gene fragment of interest into cleaved genomic DNA using the CRISPR system, ssONs can also be used as donor DNA for the knock-in. For example, ssONs that contain regions for binding to gRNAs such as crRNA and 40-60 bp of homologous sequences (homology arms) at both ends of the gene fragment to be knocked in function as donor DNA.

[0035] Since Cas9 protein and Cpf1 protein are positively charged unlike nucleic acids, their loading efficiency into lipid nanoparticles containing pH-sensitive cationic lipids in the lipid components constituting the lipid membrane is low. In contrast, in the lipid nanoparticles according to the present invention, the nucleic acid that complexes with a DNA nuclease such as Cas9 protein includes not only gRNAs such as crRNA and tracrRNA, but also ssON. Thus, since the amount of nucleic acid contained in the RNP is large, the positive charge of the RNP is suppressed and the negative charge becomes stronger, enabling efficient loading into lipid nanoparticles containing pH-sensitive cationic lipids. The base length of the ssON used in the present invention may be of sufficient length to negatively charge the RNP, and can be appropriately determined in consideration of the base length of gRNAs, such as crRNA and tracrRNA, and the types of Cas9 protein and the like. The length of the ssON can be, for example, about 50 to 500 bases long.

[0036] Determination of the target sequence on genomic DNA, design of the base sequences of gRNAs such as crRNA and tracrRNA, design of the base sequence of ssON, etc. can be carried out by conventional methods using generally used molecular biology tools based on the base sequence information of genomic DNA. For example, the design of gRNA can be carried out using design tools such as CRISPR Design Tool (Horizon Discovery), TrueDesign Genome Editor (Invitrogen).

[0037] [Lipid components] The lipid nanoparticles according to the present invention are lipid nanoparticles in which RNP is loaded into a lipid membrane structure. The lipid components constituting the lipid membrane structure contain at least a pH-sensitive cationic lipid, a neutral phospholipid, and a polyalkylene glycol-modified lipid.

[0038] <pH-sensitive cationic lipid> The pH-sensitive cationic lipid contained in the lipid nanoparticles according to the present invention is a cationic lipid represented by the following general formula (I) (hereinafter sometimes referred to as "the pH-sensitive cationic lipid of the present invention").

[0039] [ka]

[0040] In the general formula (I), a represents an integer of 3 to 5, and is preferably 4. b represents 0 or 1. When b is 0, it means that there is no —O—CO— group and it is a single bond.

[0041] In general formula (I), R 1 and R 2 each independently represents a group represented by the following general formula (A): In general formula (A), q represents an integer of 1 to 9; r represents 0 or 1; s represents an integer of 1 to 3; t represents 0 or 1; u represents an integer of 1 to 8; c represents 0 or 1; and v represents an integer of 4 to 12. However, when b and c are simultaneously 0, this does not include the cases where q is an integer of 3 to 5, r and t are 1, s is 1, and u+v is an integer of 6 to 10.

[0042] [ka]

[0043] As the pH-sensitive cationic lipid of the present invention, in terms of the stability of the lipid nanoparticles of the present invention, R 1 and R 2 It is preferable that the hydrocarbon chain of R is relatively long. Specifically, in the pH-sensitive cationic lipid of the present invention, 1 and R 2 is preferably a group having 20 or more carbon atoms, that is, in general formula (A), q+2r+s+2t+u+c+v is preferably an integer of 19 or more. In particular, for the pH-sensitive cationic lipid of the present invention, q+2r+s+2t+u+c+v is preferably an integer of 19 to 33, more preferably an integer of 19 to 31, even more preferably an integer of 21 to 31, and even more preferably an integer of 21 to 27.

[0044] R of the pH-sensitive cationic lipids of the present invention 1 and R 2 Examples of the aryl group include a group in which, in general formula (A), r is 1, t is 0, q is an integer of 5 to 11, preferably an integer of 6 to 10, s+u is an integer of 5 to 11, preferably an integer of 6 to 10, c is 1, v is an integer of 4 to 12, and q+s+u+v is an integer of 16 or more; a group in which r is 0, t is 1, q+s is an integer of 5 to 11, preferably an integer of 6 to 10, u is an integer of 5 to 8, c is 1, and v is an integer of 4 to 12 and q+s+u+v is an integer of 16 or greater; a group where r and t are 0, q+s+u is an integer of 13 to 23, preferably 15 to 21, c is 1, v is an integer of 4 to 12, and q+s+u+v is an integer of 18 or greater; or a group where r and t are 0, q+s+u is an integer of 13 to 23, preferably 15 to 21, c is 1, v is an integer of 6 to 10, and q+s+u+v is an integer of 18 or greater.

[0045] In the pH-sensitive cationic lipids of the present invention, R 1 and R 2 and may be different groups as long as they are groups represented by general formula (A), but are preferably the same group.

[0046] In general formula (I), X represents a group represented by the following general formula (B) or a 5- to 7-membered non-aromatic heterocyclic group. The 5- to 7-membered non-aromatic heterocyclic group represented by X is bonded to (O—CO) b- via a carbon atom.

[0047] [ka]

[0048] In general formula (B), d represents an integer of 0 to 3. When d is 0, it means that there is no —(CH2)— group and it is a single bond.

[0049] In general formula (B), R 3 and R 4 are each independently C 1-4Alkyl group (alkyl group with 1 to 4 carbon atoms) or C 2-4 R represents an alkenyl group (an alkenyl group having 1 to 4 carbon atoms). 3 and R 4 C indicates 1-4 Alkyl group or C 2-4 In the alkenyl group, one or two hydrogen atoms may be replaced by a phenyl group.

[0050] C 1-4 Examples of the alkyl group include a methyl group, an ethyl group, a propyl group, an isopropyl group, an n-butyl group, an isobutyl group, and a t-butyl group. 2-4 Examples of the alkenyl group include a vinyl group, a 1-propenyl group, a 2-propenyl group, a 1-methylvinyl group, a 2-methyl-1-propenyl group, a 1-butenyl group, a 2-butenyl group, and a 3-butenyl group.

[0051] In general formula (B), R 3 and R 4 may be bonded to each other to form a 5- to 7-membered non-aromatic heterocycle. 3 and R 4 Examples of the 5- to 7-membered non-aromatic heterocycle formed by bonding R to each other include a 1-pyrrolidinyl group, a 1-piperidinyl group, a 1-morpholinyl group, and a 1-piperazinyl group. 3 and R 4 In the 5- to 7-membered non-aromatic heterocycle formed by bonding together, one or two hydrogen atoms in the ring are C 1-4 Alkyl group or C 2-4 The two hydrogen atoms in the ring may be substituted with an alkenyl group. 1-4 Alkyl group or C 2-4 When the alkenyl groups are substituted, they may be substituted with the same groups or with different groups.

[0052] In general formula (I), when X is a 5- to 7-membered non-aromatic heterocyclic group, examples of the heteroatom contained in the heterocyclic group include a nitrogen atom, an oxygen atom, and a sulfur atom. The heteroatom constituting the heterocyclic group may be one, or may be two or more heteroatoms that are the same or different. The heterocyclic group may be a saturated heterocyclic ring and may contain one or more double bonds, but the heterocyclic ring is not an aromatic ring.

[0053] The pH-sensitive cationic lipid of the present invention is a cationic lipid having general formula (I), in which a is an integer of 3 to 5, b is 1, and X is a 5- to 7-membered non-aromatic heterocyclic group (bonded to (O-CO)b- by a carbon atom in the heterocyclic group), preferably a 1-pyrrolidinyl group, a 1-piperidinyl group, a 1-morpholinyl group, or a 1-piperazinyl group (bonded to (O-CO)b- by a carbon atom in the ring, and one hydrogen atom is C 1-4 Alkyl group or C 2-4 and R 1 and R 2 are each independently a group of general formula (A) in which r is 1, t is 0, q is an integer of 5 to 11, preferably an integer of 6 to 10, s + u is an integer of 5 to 11, preferably an integer of 6 to 10, c is 1, v is an integer of 4 to 12, and q + s + u + v is an integer of 16 or more; a lipid of general formula (I) in which a is an integer of 3 to 5, b is 1, and X is a 5- to 7-membered non-aromatic heterocyclic group (bonded to (O-CO)b- by a carbon atom in the heterocyclic group), preferably a 1-pyrrolidinyl group, a 1-piperidinyl group, a 1-morpholinyl group, or a 1-piperazinyl group (bonded to (O-CO)b- by a carbon atom in the ring, and one hydrogen atom is C 1-4 Alkyl group or C 2-4 and R 1 and R 2are each independently a group in which, in general formula (A), r is 0, t is 1, q+s is an integer of 5 to 11, preferably an integer of 6 to 10, u is an integer of 5 to 8, c is 1, v is an integer of 4 to 12, and q+s+u+v is an integer of 16 or more; a lipid in which, in general formula (I), a is an integer of 3 to 5, b is 0, and X is a group in which, in general formula (B), d is 0, and R 3 and R 4 are each independently C 1-4 Alkyl group or C 2-4 Alkenyl group (R 3 and R 4 C indicates 1-4 Alkyl group or C 2-4 The alkenyl group may have one or two hydrogen atoms replaced by a phenyl group, and R 1 and R 2 are each independently a group represented by general formula (A) in which r is 1, t is 0, q is an integer of 5 to 11, preferably an integer of 6 to 10, s+u is an integer of 5 to 11, preferably an integer of 6 to 10, c is 1, v is an integer of 4 to 12, and q+s+u+v is an integer of 16 or more; a lipid represented by general formula (I) in which a is an integer of 3 to 5, b is 0, and X is a group represented by general formula (B) in which d is 0 and R 3 and R 4 are each independently C 1-4 Alkyl group or C 2-4 Alkenyl group (R 3 and R 4 C indicates 1-4 Alkyl group or C 2-4 The alkenyl group may have one or two hydrogen atoms replaced by a phenyl group, and R 1 and R 2 are each independently a group represented by general formula (A) in which r is 0, t is 1, q+s is an integer of 5 to 11, preferably an integer of 6 to 10, u is an integer of 5 to 8, c is 1, v is an integer of 4 to 12, and q+s+u+v is an integer of 16 or more. 1 and R 2 are the same group, the pH-sensitive cationic lipid represented by general formula (I) is more preferably a lipid in which R1 and R 2 Lipids in which are the same group and a is 4 are particularly preferred.

[0054] The pKa of the pH-sensitive cationic lipid represented by general formula (I) is not particularly limited, but can be selected, for example, from about 4.0 to 9.0, preferably from about 4.5 to 8.5, and more preferably from about 6 to 8. It is preferable to select the type of each substituent so as to give a pKa within this range.

[0055] The pH-sensitive cationic lipid represented by general formula (I) can be easily produced, for example, by the method specifically shown in the Examples of this specification. By referring to this production method and appropriately selecting raw material compounds, reagents, reaction conditions, etc., a person skilled in the art can easily produce any lipid within the scope of general formula (I).

[0056] <Neutral phospholipids> The neutral phospholipid contained in the lipid component of the lipid nanoparticles according to the present invention (hereinafter sometimes referred to as "neutral phospholipid of the present invention") is a lipid in which the overall group has a neutral charge and in which a phosphate group and a positively charged group are linked by an appropriate linking group. Examples of the positively charged group include an ammonium group and a quaternary ammonium group.

[0057] The neutral phospholipid of the present invention is preferably a glycerophospholipid or a sphingophospholipid. Examples of neutral glycerophospholipids include phosphatidylethanolamine, phosphatidylcholine, cardiolipin, and plasmalogen. Examples of neutral sphingophospholipids include sphingomyelin, ceramide phosphorylglycerol, and ceramide phosphorylethanolamine. The fatty acid residue in these neutral glycerophospholipids or neutral sphingophospholipids is not particularly limited, but examples include saturated or unsaturated fatty acid residues having 12 to 24 carbon atoms, and saturated or unsaturated fatty acid residues having 14 to 20 carbon atoms are preferred. Specific examples include acyl groups derived from fatty acids such as lauric acid, myristic acid, palmitic acid, palmitoleic acid, stearic acid, oleic acid, linoleic acid, linolenic acid, arachidic acid, arachidonic acid, behenic acid, and lignoceric acid. When these glycerolipids or sphingolipids have two or more fatty acid residues, all of the fatty acid residues may be the same group or may be different groups.

[0058] As the neutral phospholipid of the present invention, from the viewpoint of further improving genome editing efficiency, a neutral glycerophospholipid having a saturated or unsaturated fatty acid residue having 12 to 24 carbon atoms is preferred, a neutral glycerophospholipid or neutral sphingophospholipid having an unsaturated fatty acid residue having 12 to 24 carbon atoms is more preferred, and a neutral glycerophospholipid having an unsaturated fatty acid residue having 12 to 24 carbon atoms is even more preferred. Among these, a phosphatidylethanolamine having an unsaturated fatty acid residue having 12 to 24 carbon atoms is preferred, a phosphatidylethanolamine having a fatty acid residue having 14 to 20 carbon atoms is more preferred, and dioleoylphosphatidylethanolamine (DOPE) is particularly preferred.

[0059] <Polyalkylene glycol modified lipid> The polyalkylene glycol-modified lipid contained in the lipid component of the lipid nanoparticles according to the present invention (hereinafter sometimes referred to as "polyalkylene glycol-modified lipid of the present invention") is not particularly limited as long as it is a lipid modified with polyalkylene glycol, but pH-sensitive cationic lipids and neutral phospholipids are excluded. Polyalkylene glycol is a hydrophilic polymer, and by constructing lipid nanoparticles using a polyalkylene glycol-modified lipid as a lipid membrane-constituting lipid, the surface of the lipid nanoparticles can be modified with polyalkylene glycol. Surface modification with polyalkylene glycol may sometimes increase the stability of the lipid nanoparticles, such as their blood retention.

[0060] Examples of polyalkylene glycols that can be used include polyethylene glycol, polypropylene glycol, polytetramethylene glycol, polyhexamethylene glycol, etc. The weight-average molecular weight of the polyalkylene glycol is, for example, about 300 to 10,000, preferably about 500 to 10,000, and more preferably about 1,000 to 5,000.

[0061] For example, stearylated polyethylene glycol (eg, PEG45 stearate (STR-PEG45)) can be used to modify lipids with polyethylene glycol. Other polyethylene glycol derivatives such as N-[carbonyl-methoxypolyethylene glycol-2000]-1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine, N-[carbonyl-methoxypolyethylene glycol-5000]-1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine, N-[carbonyl-methoxypolyethylene glycol-750]-1,2-distearoyl-sn-glycero-3-phosphoethanolamine, N-[carbonyl-methoxypolyethylene glycol-2000]-1,2-distearoyl-sn-glycero-3-phosphoethanolamine, and N-[carbonyl-methoxypolyethylene glycol-5000]-1,2-distearoyl-sn-glycero-3-phosphoethanolamine can also be used, but polyalkylene glycolated lipids are not limited to these.

[0062] <Other lipids> Among the constituent lipids of the lipid nanoparticles of the present invention, lipids other than the pH-sensitive cationic lipid of the present invention, the neutral phospholipid of the present invention, and the polyalkylene glycol-modified lipid of the present invention can be lipids generally used in forming liposomes. Examples of such lipids include positively or negatively charged phospholipids, sterols, saturated or unsaturated fatty acids, etc. These can be used alone or in combination of two or more.

[0063] Examples of positively or negatively charged phospholipids include phosphatidylserine, phosphatidylinositol, phosphatidylglycerol, ceramide phosphorylglycerol phosphate, phosphatidic acid, etc. Examples of sterols include animal-derived sterols such as cholesterol, cholesterol succinate, lanosterol, dihydrolanosterol, desmosterol, and dihydrocholesterol; plant-derived sterols (phytosterols) such as stigmasterol, sitosterol, campesterol, and brassicasterol; and microbial-derived sterols such as zymosterol and ergosterol.

[0064] <Lipid composition> The pH-sensitive cationic lipid, neutral lipid, and polyalkylene glycol-modified lipid of the present invention constituting the lipid nanoparticles of the present invention may each be one type, or two or more types. When the pH-sensitive cationic lipid of the present invention constituting the lipid nanoparticles of the present invention is two or more types, the amount of the pH-sensitive cationic lipid of the present invention means the total amount of lipid molecules corresponding to the pH-sensitive cationic lipid of the present invention among the lipid molecules constituting the lipid nanoparticles. Similarly, when the neutral lipid of the present invention constituting the lipid nanoparticles of the present invention is two or more types, the amount of the neutral lipid of the present invention means the total amount of lipid molecules corresponding to the neutral lipid of the present invention among the lipid molecules constituting the lipid nanoparticles. When the polyalkylene glycol-modified lipid of the present invention constituting the lipid nanoparticles of the present invention is two or more types, the amount of the polyalkylene glycol-modified lipid of the present invention means the total amount of lipid molecules corresponding to the polyalkylene glycol-modified lipid of the present invention among the lipid molecules constituting the lipid nanoparticles.

[0065] The greater the proportion of the pH-sensitive cationic lipid of the present invention in the total amount of lipid molecules (lipid components) constituting the lipid nanoparticles, the smaller the particle diameter of the lipid nanoparticles and the higher the RNP encapsulation efficiency.Furthermore, the greater the proportion of the neutral phospholipid of the present invention in the total amount of lipid molecules constituting the lipid nanoparticles, the higher the RNP encapsulation efficiency tends to be.Furthermore, since the lipid molecules constituting the lipid nanoparticles contain a small amount of the polyalkylene glycol-modified lipid of the present invention, the particle diameter of the lipid nanoparticles tends to be sufficiently small, the RNP encapsulation efficiency tends to be sufficiently high, and the genome editing efficiency tends to be sufficiently high. In order to achieve higher genome editing efficiency, the ratio of the pH-sensitive cationic lipid of the present invention to the total amount of lipids constituting the lipid nanoparticles of the present invention ([amount (mol) of the pH-sensitive cationic lipid of the present invention] / ([amount (mol) of total lipids constituting the lipid nanoparticles])×100%) is 30 to 50 mol%, the ratio of the neutral phospholipid of the present invention to the total amount of lipids constituting the lipid nanoparticles ([amount (mol) of the neutral phospholipid of the present invention] / ([amount (mol) of total lipids constituting the lipid nanoparticles])×100%) is 20 to 50 mol%, and the ratio of the polyalkylene glycol of the present invention to the total amount of lipids constituting the lipid nanoparticles is 20 to 50 mol%. It is preferable that the ratio of the polyalkylene glycol-modified lipid ([amount (mol) of the polyalkylene glycol-modified lipid of the present invention] / ([amount (mol) of all lipids constituting the lipid nanoparticles])×100%) is 1.0 to 4.0 mol %, and it is more preferable that the ratio of the pH-sensitive cationic lipid of the present invention to the total amount of lipids constituting the lipid nanoparticles of the present invention is 40 to 50 mol %, the ratio of the neutral phospholipid of the present invention to the total amount of lipids constituting the lipid nanoparticles is 20 to 50 mol %, and the ratio of the polyalkylene glycol-modified lipid of the present invention to the total amount of lipids constituting the lipid nanoparticles is 1.5 to 2.0.

[0066] <Surface modification> The lipid nanoparticles according to the present invention can be subjected to appropriate surface modification, if necessary. The blood retention of the lipid nanoparticles according to the present invention can be improved by modifying the surface with a hydrophilic polymer, etc. In some cases, surface modification can be achieved by using lipids modified with these modifying groups as constituent lipids of the lipid nanoparticles.

[0067] In producing the lipid nanoparticles according to the present invention, lipid derivatives that can be used to enhance blood retention include, for example, glycophorin, ganglioside GM1, phosphatidylinositol, ganglioside GM3, glucuronic acid derivatives, glutamic acid derivatives, polyglycerin phospholipid derivatives, etc. Furthermore, hydrophilic polymers that can be used to enhance blood retention include, in addition to polyalkylene glycol, dextran, pullulan, ficoll, polyvinyl alcohol, styrene-maleic anhydride alternating copolymer, divinyl ether-maleic anhydride alternating copolymer, amylose, amylopectin, chitosan, mannan, cyclodextrin, pectin, carrageenan, etc. for surface modification.

[0068] Furthermore, to promote nuclear translocation of the lipid nanoparticles according to the present invention, the surface of the lipid nanoparticles can be modified with an oligosaccharide compound of three or more sugars. The type of oligosaccharide compound of three or more sugars is not particularly limited, but for example, an oligosaccharide compound having about 3 to 10 sugar units bonded thereto can be used, and preferably an oligosaccharide compound having about 3 to 6 sugar units bonded thereto can be used. Among these, oligosaccharide compounds that are preferably glucose trimers or hexamers can be used, and more preferably, oligosaccharide compounds that are glucose trimers or tetramers can be used. More specifically, isomaltotriose, isopanose, maltotriose, maltotetraose, maltopentaose, or maltohexaose can be suitably used, and among these, maltotriose, maltotetraose, maltopentaose, or maltohexaose in which glucose is α1-4-linked is more preferred. Maltotriose or maltotetraose is particularly preferred, and maltotriose is the most preferred. The amount of surface modification of lipid nanoparticles with an oligosaccharide compound is not particularly limited, but is, for example, about 1 to 30 mol %, preferably about 2 to 20 mol %, and more preferably about 5 to 10 mol % relative to the total lipid amount.

[0069] The method for surface-modifying lipid nanoparticles with oligosaccharide compounds is not particularly limited, but for example, liposomes in which the surface of lipid nanoparticles is modified with monosaccharides such as galactose or mannose (WO 2007 / 102481) are known, and the surface modification method described in this publication can be adopted. The disclosure of the above publication is incorporated herein by reference in its entirety.

[0070] Furthermore, the lipid nanoparticles according to the present invention can be imparted with one or more of the following functions: temperature sensitivity, membrane permeability, gene expression, and pH sensitivity. By appropriately adding these functions, the retention of the lipid nanoparticles in the blood can be improved, the capture rate by reticuloendothelial tissues such as the liver and spleen can be reduced, and the lipid nanoparticles can be efficiently released from endosomes after endocytosis in target cells and transported into the nucleus, thereby achieving high genome editing activity in the nucleus.

[0071] Furthermore, the lipid nanoparticles of the present invention can be modified with substances such as antibodies that can specifically bind to cell surface receptors or antigens, thereby improving the efficiency of substance delivery into the cell nucleus. For example, it is preferable to place a monoclonal antibody against a biological component specifically expressed in a target tissue or organ on the surface of the lipid nanoparticle. This technique is described, for example, in STEALTH LIPOSOME (pp. 233-244, published by CRC Press, Inc., edited by Danilo Lasic and Frank Martin). By incorporating a lipid derivative capable of reacting with a mercapto group in a monoclonal antibody or a fragment thereof (e.g., Fab fragment, F(ab')2 fragment, or Fab' fragment) as a component of the lipid nanoparticle, such as a lipid derivative having a maleinimide structure, such as poly(ethylene glycol)-α-distearoylphosphatidylethanolamine-ω-maleinimide or α-[N-(1,2-distearoyl-sn-glycero-3-phosphoryl-ethyl)carbamyl)-ω-[3-[2-(2,5-dihydro-2,5-dioxo-1H-pyrrol-1-yl)ethanecarboxamido]propyl}-poly(oxy-1,2-ethanedyl), the monoclonal antibody can be bound to the surface of the lipid nanoparticle membrane.

[0072] The surface of the lipid nanoparticles of the present invention may be modified with a polypeptide containing multiple consecutive arginine residues (hereinafter referred to as "polyarginine"). Polyarginine is preferably a polypeptide containing 4 to 20 consecutive arginine residues, more preferably a polypeptide consisting of only 4 to 20 consecutive arginine residues, and particularly preferably octaarginine. Modifying the surface of lipid nanoparticles such as liposomes with polyarginines such as octaarginine can improve the intracellular delivery efficiency of RNPs encapsulated in the liposomes (Journal of Controlled Release, 98, pp. 317-323, 2004; WO 2005 / 32593). Modification of the lipid nanoparticle surface with polyarginine can be easily achieved by using lipid-modified polyarginine, such as stearylated octaarginine, as a constituent lipid of the lipid nanoparticles, according to the methods described in the above publications. The disclosures of the above publications and all references cited therein are incorporated herein by reference.

[0073] [Other ingredients] The lipid nanoparticles of the present invention may further contain one or more substances selected from the group consisting of antioxidants such as tocopherol, propyl gallate, ascorbyl palmitate, or butylated hydroxytoluene, charged substances, and membrane polypeptides. Examples of charged substances that impart a positive charge include saturated or unsaturated aliphatic amines such as stearylamine and oleylamine; saturated or unsaturated cationic synthetic lipids such as dioleoyltrimethylammonium propane; or cationic polymers. Examples of charged substances that impart a negative charge include dicetyl phosphate, cholesteryl hemisuccinate, phosphatidylserine, phosphatidylinositol, and phosphatidic acid. Examples of membrane polypeptides include surface membrane polypeptides and integral membrane polypeptides. The amounts of these substances to be incorporated are not particularly limited and can be selected appropriately depending on the purpose.

[0074] [Lipid nanoparticles] The lipid nanoparticles according to the present invention are lipid membrane structures composed of lipid components and are loaded with RNP. Since the encapsulation rate of RNP in the lipid nanoparticles is further increased, the ratio of RNP loaded in the lipid nanoparticles to the total amount of lipids constituting the lipid nanoparticles according to the present invention ([amount of RNP (mol)] / ([amount of total lipids constituting the lipid nanoparticles (mol)])×100%) is 1.8 to 3.6×10 -2 It is preferably expressed as mole percent.

[0075] The size of the lipid nanoparticles according to the present invention is such that high delivery efficiency is likely to be achieved even when target cells are located relatively deep within the body, and therefore the average particle diameter is preferably 80 nm or less, more preferably 50 nm or less, even more preferably 40 nm or less, even more preferably 30 nm or less, and particularly preferably 10 to 30 nm. The average particle diameter of the lipid nanoparticles refers to the number-average particle diameter measured by dynamic light scattering (DLS). Measurement by dynamic light scattering can be carried out by standard methods using a commercially available DLS device or the like.

[0076] The polydispersity index (PDI) of the lipid nanoparticles according to the present invention is about 0.05 to 0.1, preferably about 0.06 to 0.08, and more preferably about 0.07. The zeta potential can be in the range of 5.5 mV to 6.0 mV, and preferably about 5.8 mV.

[0077] The form of the lipid nanoparticles according to the present invention is not particularly limited, and examples thereof include unilamellar liposomes, multilamellar liposomes, spherical micelles, irregular layered structures, etc. The lipid nanoparticles according to the present invention are preferably unilamellar liposomes or multilamellar liposomes when dispersed in an aqueous solvent.

[0078] [Manufacturing method] The method for producing lipid nanoparticles according to the present invention is not particularly limited, and any method available to those skilled in the art can be employed. For example, all lipid components are dissolved in an organic solvent such as chloroform, and a lipid membrane is formed by drying under reduced pressure using an evaporator or spray drying using a spray dryer. An aqueous solvent containing RNPs or the like is then added to the dried mixture, and the resulting mixture is emulsified using an emulsifier such as a homogenizer, an ultrasonic emulsifier, or a high-pressure injection emulsifier. Liposomes can also be produced by well-known methods for producing liposomes, such as reverse-phase evaporation. To control the size of lipid nanoparticles, extrusion filtration can be performed under high pressure using a membrane filter with a uniform pore size.

[0079] The lipid nanoparticles according to the present invention can be produced by an alcohol dilution method using a flow channel. The flow channel used for production may be a microflow channel with a built-in three-dimensional micromixer that can achieve instantaneous mixing of two liquids, but because this allows for the formation of a nano-sized lipid particle formation system with high particle size control, it is preferable to use a simple two-dimensional flow channel structure, as described in Patent Document 1, in which baffles (baffle plates) of a fixed width relative to the width of the flow channel are arranged alternately on both sides of the micro-sized flow channel through which the raw material solution flows.

[0080] Specifically, it is preferable to use a flow path structure as shown in Fig. 1A (hereinafter, sometimes referred to as the "flow path structure of the present invention"), in which a first inlet path 10 for introducing a first fluid and a second inlet path 20 for introducing a second fluid, which are independent of each other, each have a certain length, join together and form a single dilution flow path 30 toward the downstream side. The dilution flow path 30 has a flow path portion 50 that is two-dimensionally curved in at least a part thereof, and the curved flow path portion 50 is formed by at least two or more structural elements 40 that protrude alternately from both side wall surfaces of the dilution flow path opposing each other in the Y direction toward the flow path center in approximately the Y direction (approximately the +Y direction, approximately the −Y direction), have constant heights h1, h2... of 1 / 2y0 or more and less than 1y0, and have constant widths x1, x2... in the X direction, and that regulate the flow path width of the dilution flow path, being defined as the X direction, the Y direction, and the flow path width of the dilution flow path upstream of the curved flow path portion 50, respectively. That is, in the region where the structural element 40 is present, the flow path width y1, y2... of the dilution flow path is restricted to 1 / 2y0 or less, particularly 1 / 2y0 or less and 1 / 40y0 or more, over a certain length x1, x2... in the X direction.

[0081] While the flow channel structure of the present invention conceptually has a shape similar to that of a micro-sized flow channel, as illustrated in FIG. 1A and described above, with approximately rectangular baffles alternately arranged on both sides, in practice, it is not limited to a configuration in which separate baffles are arranged on the flow channel. That is, as long as a flow channel of a similar shape is formed corresponding to the flow channel formed by arranging such baffles, the configuration of the structural element 40 is not particularly limited. To form the aforementioned structural element 40, the wall surfaces of the flow channel structure may be integrally formed while being bent into a predetermined shape (while maintaining a substantially constant thickness) to form a two-dimensional flow channel shape that bends and contracts as defined above. Naturally, such embodiments are included in the flow channel structure of the present invention. Such two-dimensional flow channels can be relatively easily formed by, for example, injection molding, cast molding, or molding using a three-dimensional printer using thermoplastic resins, thermosetting resins, ultraviolet-curable resins, metals, or glassy materials.

[0082] The channel width y0 of the dilution channel 30 after the first inlet channel 10 and the second inlet channel 20 join depends to some extent on the particle size of the nano-sized lipid particles to be formed, but is typically about 20 to 1000 μm, more preferably about 100 to 200 μm. To obtain lipid particles of the desired nano size, specifically, for example, within a particle size range of about 10 to 100 nm, it is somewhat necessary to dilute the lipid solution with a diluent medium with a channel width y0 within the above range.

[0083] Furthermore, the height h1, h2... (length in the Y direction) of each structural element 40 is ½y0 or more and less than 1y0, preferably ½y0 or more and 39 / 40y0 or less, more preferably ½y0 or more and ¾y0 or less, of the flow path width y0 of the dilution flow path 30 on the upstream side. The presence of each structural element 40 reduces the flow path width y1, y2... from the flow path width y0 of the dilution flow path 30 on the upstream side to a width less than ½y0 but greater than 0. Note that the heights h1, h2... of the structural elements 40 provided in the curved flow path portion 50 do not necessarily have to be identical and may be different as long as the above-mentioned predetermined conditions are satisfied. The flow path widths y1, y2... formed thereby may also be different. For example, the widths h1, h2... of each structural element 40 may gradually increase in the downstream direction, narrowing the flow path widths y1, y2.... The heights h1, h2... (lengths in the Y direction) of each structural element 40 are predetermined, and the flow path widths y1, y2... at the locations where these elements exist are narrowed to widths less than 1 / 2y0, thereby improving the efficiency of molecular diffusion.

[0084] Although it is not particularly limited and depends on other conditions such as the size of the lipid particles to be obtained, the number of structural elements 40, the width (length in the X direction) x1, x2... of each mixer structural element 40, and the interval d1, d2... between adjacent structural elements 40, specifically, for example, if the flow path width y0 of the upstream dilution flow path is 200 μm, it is desirable that the heights h1, h2... of each structural element 40 be 100 μm to less than 200 μm. Therefore, the flow path width y1, y2... at the position where each structural element 40 is present is less than 1 / 2y0 and greater than 0, that is, approximately less than 100 μm.

[0085] The widths (lengths in the X direction) x1, x2... of each structural element 40 are influenced by other conditions, such as the size of the lipid particles to be obtained, the number of structural elements 40, the heights h1, h2... (lengths in the Y direction) of each structural element 40, and the spacing d1, d2... between adjacent structural elements 40, but are preferably set to approximately 1 / 10y0 or more and 5y0 or less of the channel width y0 of the upstream dilution channel. Specifically, for example, if the channel width y0 of the upstream dilution channel is 20 to 1000 μm, typically 200 μm, then the widths x1, x2... of each structural element 40 are desirably approximately 20 μm to 1000 μm. The widths x1, x2... of each structural element 40 do not necessarily have to be identical and may be different as long as the above-mentioned predetermined conditions are met. For example, the widths x1, x2... may gradually increase in the downstream direction.

[0086] The distances d1, d2... between adjacent structural elements 40 depend on other conditions, such as the size of the lipid particles to be obtained, the number of structural elements 40, the heights h1, h2... (lengths in the Y direction) of each structural element 40, and the widths x1, x2... (lengths in the X direction) of each structural element 40, but are preferably set to approximately 1 / 10y0 or more and 5y0 or less of the channel width y0 of the upstream dilution channel. Specifically, for example, if the channel width y0 of the upstream dilution channel is 20 to 1000 μm, typically 200 μm, the distances d1, d2... between adjacent structural elements 40 are preferably set to approximately 20 μm to 1000 μm. The distances d1, d2... between adjacent structural elements 40 do not necessarily have to be identical and may be different as long as the above-mentioned predetermined conditions are met. For example, the distances d1, d2... may be gradually narrower downstream.

[0087] In the flow channel structure of the present invention, the axial direction or extension direction of the upstream dilution channel is defined as the X direction, and the width direction of the dilution channel perpendicularly intersecting the X direction is defined as the Y direction. As described above, each structural element 40 extends alternately from both side walls toward the center of the channel in the approximate Y direction (approximately +Y direction and approximately −Y direction), and has wall surfaces that are approximately perpendicular to the channel direction (X direction). However, this angle does not necessarily have to be strictly 90°; even a slight inclination can be effective. While not particularly limited, specific examples include angles within the range of approximately 30 to 150°, more preferably 40 to 140°, and particularly preferably 80 to 100°. Furthermore, a certain degree of roundness is also acceptable for the shape of the corners of each structural element 40 on the channel center side. While not particularly limited, for example, a radius of 50 μm or less, more preferably 20 μm or less, may be acceptable. However, in order to obtain more controllable and uniform nano-sized lipid particles, it is desirable to minimize these tolerances. 1A, the axial direction of the upstream dilution flow channel in the flow channel structure, or the X direction, which is its extension direction, is shown as a straight line for convenience, but this X direction merely indicates the axial direction of the dilution flow channel, and in reality, it is not limited to such a straight line and may be curved, for example, with a certain curvature. In such a case, the Y direction, which is the width direction of the dilution flow channel that intersects perpendicularly with the X direction, refers to the direction perpendicular to the X direction at the unit length.

[0088] Furthermore, since the flow path structure of the present invention is a flow path structure having a two-dimensional structure as described above, the size of the flow path in the depth direction (paper thickness direction in FIG. 1A) is not particularly limited, but is preferably, for example, about 10 to 1000 μm, more preferably about 50 to 200 μm.

[0089] The flow channel used in producing lipid nanoparticles according to the present invention by the alcohol dilution method is not particularly limited as long as the dilution flow channel 30 in the flow channel structure shown in Figure 1A is a flow channel that can generate a three-dimensional flow. For example, the flow channel structure of the present invention may be a chaotic micromixer (staggered herringbone mixer) (Non-Patent Document 8) in which a chaotic flow is generated by grooves or microprotrusions formed on the wall surface of the dilution flow channel 30, instead of the two-dimensionally curved flow channel portion 50.

[0090] In the case of the flow channel structure of the present invention, as shown in Figure 1A, a first inlet channel 10 for introducing a first fluid and a second inlet channel 20 for introducing a second fluid are joined together to form a single dilution channel, and a lipid solution containing lipid components dissolved in ethanol is introduced through the first inlet channel, and an aqueous solution containing Cas9 protein, crRNA, tracrRNA, and ssON (RNP-containing aqueous solution) is introduced through the second inlet channel. In the dilution channel, the lipid solution is diluted with the RNP-containing aqueous solution, and in this process, lipid nanoparticles loaded with RNP are produced.

[0091] The flow path structure of the present invention has a plurality of inlet paths that are independent of one another, and these inlet paths each have a certain length and merge to form a single dilution flow path, and may have three inlet paths. When the flow path structure of the present invention has three inlet paths, the first inlet path, the second inlet path, and the third inlet path can each have a certain length and merge to form a single dilution flow path so that the first fluid introduced from the first inlet path comes into contact with the third fluid introduced from the third inlet path before merging with the second fluid introduced from the second inlet path.

[0092] For example, as shown in FIG. 1B , if the inlet channels whose merging points into the dilution channel 30 are the furthest from each other are the first inlet channel 10 and the second inlet channel 20, and the remaining inlet channel is the third inlet channel 60, then a lipid solution in which lipid components are dissolved in ethanol is introduced through the first inlet channel 10, an RNP-containing aqueous solution is introduced through the second inlet channel 20, and the aqueous solvent used to prepare the RNP-containing aqueous solution is introduced through the third inlet channel 60. The lipid solution introduced through the first inlet channel 10 first merges with the aqueous solvent introduced through the third inlet channel 60, and then merges with the RNP-containing aqueous solution introduced through the second inlet channel 20. Direct contact of the RNP-containing aqueous solution with the lipid solution, which is a high-concentration ethanol solution, is avoided, thereby preventing aggregation of proteins in the RNP-containing aqueous solution due to the high concentration of ethanol in the dilution channel 30, particularly near its inlet.

[0093] Dilution in the flow channel structure of the present invention depends on molecular diffusion. The faster the dilution rate of the raw lipid solution, the smaller the size of the lipid particles produced. Therefore, by adjusting the width, length, and arrangement of the structure (baffle), the dilution rate of the raw solution can be controlled, making it possible to form lipid nanoparticles with higher particle size controllability than conventional methods.

[0094] The RNP-containing aqueous solution can be prepared by dissolving the Cas9 protein, crRNA, tracrRNA, and ssON in an aqueous solvent. The aqueous solvent is not particularly limited, as long as it is capable of mixing the RNP with the lipid solution while maintaining genome editing activity and the resulting lipid nanoparticles can be stably dispersed. Examples of aqueous solvents include buffers such as phosphate buffer, citrate buffer, and phosphate-buffered saline, saline, and cell culture media. These aqueous solvents (dispersion media) may further contain monosaccharides such as glucose, galactose, mannose, fructose, inositol, ribose, and xylose; disaccharides such as lactose, sucrose, cellobiose, trehalose, and maltose; trisaccharides such as raffinose and melezinose; polysaccharides such as cyclodextrin; sugar alcohols such as erythritol, xylitol, sorbitol, mannitol, and maltitol; and polyhydric alcohols such as glycerin, diglycerin, polyglycerin, propylene glycol, polypropylene glycol, ethylene glycol, diethylene glycol, triethylene glycol, polyethylene glycol, ethylene glycol monoalkyl ether, diethylene glycol monoalkyl ether, and 1,3-butylene glycol.

[0095] If the pH of the RNP-containing aqueous solution is low, the genome editing activity of the produced lipid nanoparticles loaded with RNP may be reduced. Therefore, the pH of the RNP-containing aqueous solution is preferably 5.0 or higher, more preferably within the range of 5.0 to 8.5, even more preferably within the range of 5.0 to 8.0, and even more preferably within the range of 5.0 to 7.5. The pH of the RNP-containing aqueous solution is also preferably 6 or higher, more preferably within the range of 6 to 8.5, even more preferably within the range of 6 to 8, and even more preferably within the range of 6 to 7.5.

[0096] The ratio of the flow rates of the lipid solution and the RNP-containing aqueous solution affects the dilution rate of the lipid solution and, consequently, the size of the lipid particles produced. In the production of lipid nanoparticles according to the present invention, in order to produce lipid nanoparticles having a sufficiently small average particle size and high uptake efficiency into target cells, the ratio of the flow rate of the RNP-containing aqueous solution to the flow rate of the lipid solution is preferably 7 or more, more preferably 7 to 10, and even more preferably 7 to 9.

[0097] The total flow rate of the lipid solution and the RNP-containing aqueous solution is not particularly limited and can be appropriately adjusted within the range of 1 μL / min to 100 mL / min. In the production of lipid nanoparticles according to the present invention, the total flow rate of the lipid solution and the RNP-containing aqueous solution is preferably within the range of 50 μL / min to 1 mL / min, more preferably within the range of 50 to 800 μL / min, even more preferably within the range of 50 to 600 μL / min, and even more preferably within the range of 50 to 500 μL / min.

[0098] The lipid nanoparticles of the present invention are carriers that deliver RNPs for genome editing to target cells. Genome editing can be achieved by introducing the lipid nanoparticles of the present invention into target cells. When the target cells are cultured cells, the lipid nanoparticles of the present invention are added to the culture medium. When the target cells are cells within the body of an animal, the lipid nanoparticles of the present invention are administered to the animal. The route of administration is not particularly limited, but is preferably parenteral administration such as intravenous administration, enteral administration, intramuscular administration, subcutaneous administration, transdermal administration, nasal administration, or pulmonary administration.

[0099] The animals to which the lipid nanoparticles of the present invention are administered are not particularly limited and may be humans or non-human animals, including mammals such as cows, pigs, horses, sheep, goats, monkeys, dogs, cats, rabbits, mice, rats, hamsters, and guinea pigs, and birds such as chickens, quails, and ducks. [Example]

[0100] The present invention will now be described in more detail with reference to examples, but the present invention is not limited to the following examples.

[0101] [Reference example 1] We investigated the conditions for preparing RNP-loaded lipid nanoparticles using a manufacturing device with a flow channel structure as shown in Figure 1A. Specifically, we investigated the pH of the buffer solution in which RNP is dissolved, the flow rate ratio (FRR) of the RNP solution to ethanol, and the total flow rate (TFR) of the ethanol and RNP solution.

[0102] Figure 2 shows a schematic diagram of the manufacturing apparatus actually used. Figure 2(A) is a schematic diagram of the overall structure, Figure 2(B) is a perspective view of the mixer-integrated microchannel (channel structure), and Figure 2(C) is an enlarged view of a portion of the dilution channel 104c. As shown in Figure 2(A), a syringe 101 loaded with a lipid solution and a syringe 102 loaded with an RNP solution were each installed in a flow control device 103, which were connected to an inlet 104a of the first inlet channel and an inlet 104b of the second inlet channel of the mixer-integrated microchannel 104 via a tube 105. An outlet 104d of the dilution channel 104c was connected via a tube 105 to a collection container 106 for collecting the produced lipid nanoparticles.

[0103] To eliminate the influence of lipids, ethanol was used instead of a lipid solution. The RNP solution used consisted of the Cas9 protein (160 kDa) from Streptococcus pyogenes (product name: "Alt-R Sp Cas9 Nuclease V3" manufactured by Integrated DNA Technologies), crRNA (SEQ ID NO: 2, 36 bases long, with the 1st to 20th bases complementary to the target sequence) in the green fluorescent protein (GFP) gene (SEQ ID NO: 1, 20 bases long), and tracrRNA (SEQ ID NO: 3, 67 bases long) dissolved in buffer at a molar ratio of 1:1:1. The crRNA and racrRNA were amplified by PCR using the genomic DNA of HeLa cells stably expressing GFP (HeLa-GFP) as a template. The pH 6 buffer solution used was an MES buffer solution (20 mM MES, 50 mM NaCl, pH 6.0), and the pH 4.0, 5.0, or 5.5 buffer solution used was a citrate buffer solution (20 mM citric acid, 50 mM NaCl, pH 4.0, 5.0, or 5.5).

[0104] After the RNP solution and ethanol were pumped through the microchannels of the manufacturing device shown in Figure 2, the solution discharged from the dilution channel was collected and placed in a dialysis membrane (MWCO: 12,000-14,000). The solution was dialyzed in PBS(-) at 4°C for at least 2 hours to exchange the buffer and remove the alcohol. The Cas9 protein concentration of the dialyzed RNP solution was quantified using fluorescamine. The RNP solution was then mixed with double-stranded DNA (dsDNA) containing the target sequence (SEQ ID NO: 4, 0.25 pmol) at a molar ratio of 2 or 5 equivalents of Cas9 protein and incubated at 37°C for 1 hour. The reaction solution was subjected to agarose gel electrophoresis to evaluate DNA cleavage efficiency. dsDNA without RNP was used as a negative control, and a sample containing dsDNA to which RNP had not been passed through the microchannel served as a positive control. The relative cleavage activity of each sample was calculated by setting the cleavage activity of the positive control sample containing 5 equivalents of RNP relative to the target dsDNA as 1. Image analysis software Image J was used for quantitative analysis.

[0105] The RNP solution was prepared in a pH 6.0 buffer solution, the FRR was set to 9.0, and experiments were conducted with TFRs ranging from 50 to 500 μL / min to examine the effect of TFR. The results are shown in Table 1. In the table, "NC" indicates the results of the negative control, and "PC" indicates the results of the positive control. As shown in Table 1, the cleavage activity was comparable to that of the positive control at all total flow rates, and no effect of the total flow rate on DNA cleavage activity was observed. Therefore, in subsequent experiments, a faster mixing rate of 500 μL / min was used.

[0106] [Table 1]

[0107] The RNP solution was prepared in a pH 6.0 buffer solution, the TFR was set to 500 μL / min, and experiments were conducted with FRRs ranging from 3.0 to 9.0 to investigate the effect of FRR. The results are shown in Table 2. As shown in Table 2, when the FRR was 5.0 or less, a decrease in DNA cleavage activity was observed as the FRR decreased. On the other hand, when the FRR was 7.0 or more, activity was similar to that of the positive control, and no effect on the flow rate through the microchannel was observed. Based on these results, an FFR of 9.0 was used in subsequent experiments.

[0108] [Table 2]

[0109] Experiments were conducted to examine the effect of pH on the RNP solution, with a TFR of 500 μL / min, an FRR of 9.0, and buffer solutions prepared in a pH range of 4.0 to 6.0. The results are shown in Table 3. As shown in Table 3, at pH 6.0, DNA cleavage activity was maintained at a level comparable to that of the positive control, while a significant decrease in DNA cleavage activity was observed with decreasing pH below pH 5.5. Based on these results, a pH 6.0 buffer solution was used to prepare the RNP solution in subsequent experiments.

[0110] [Table 3]

[0111] [Reference example 2] We investigated whether the lipid nanoparticles according to the present invention can be produced using a production apparatus equipped with a chaotic mixer. Specifically, we investigated the pH of the buffer solution in which the RNP is dissolved, the flow rate ratio (FRR) of the RNP solution to ethanol, and the total flow rate (TFR) of the ethanol and RNP solution. Lipid nanoparticles were produced using the production apparatus used in Reference Example 1, except that the mixer-integrated microchannel 104 is the chaotic mixer-integrated microchannel 107 shown in Figure 3. Figure 3(A) is a perspective view of the chaotic mixer-integrated microchannel (channel structure), and Figure 3(B) is an enlarged view of a portion of the dilution channel 107c.

[0112] In this condition study, ethanol was used instead of the lipid solution to eliminate the influence of lipids, as in Reference Example 1. The RNP solution used was a solution (RNP solution) prepared by dissolving the Cas9 protein (160 kDa) derived from Streptococcus pyogenes, crRNA (SEQ ID NO: 2, 36 bases long) containing a base sequence complementary to the target sequence (SEQ ID NO: 1, 20 bases long) in the GFP gene, and tracrRNA (SEQ ID NO: 3, 67 bases long) in a buffer solution at a molar ratio of 1:1:1. The pH 6-6.6 buffer solution used was MES buffer (20 mM MES, 50 mM NaCl, pH 6.0), and the pH 5.5 buffer solution used was citrate buffer (20 mM citric acid, 50 mM NaCl, pH 5.5).

[0113] The RNP solution and ethanol were pumped through the microchannel of the manufacturing apparatus shown in Figure 3 under conditions of a TFR of 500 μL / min and an FRR of 9. The RNP solution discharged from the dilution channel was collected and dialyzed as in Reference Example 1. The dialyzed RNP solution was mixed with dsDNA at a molar ratio of 5 equivalents (in terms of Cas9 protein amount) as in Reference Example 1, and the mixture was allowed to react at 37°C for 1 hour to evaluate the DNA cleavage efficiency. The relative cleavage activity of each sample was calculated, assuming the cleavage activity of the positive control, in which 5 equivalents of RNP were added to the target dsDNA, as 1. The results are shown in Table 4. As in Reference Example 1, DNA cleavage activity was maintained at a similar level to the positive control at pH 6.0, while a significant decrease in DNA cleavage activity was observed at pH 5.5.

[0114] [Table 4]

[0115] The RNP solution was prepared in a pH 6.0 buffer solution, and a TFR of 500 μL / min was used. The same experiment was performed with FRRs of 5.0, 7.0, or 9.0 to examine the effect of FRR. The results are shown in Table 5. While DNA cleavage activity was slightly lower at an FRR of 5.0, activity comparable to that of the positive control was observed at FRRs of 7.0 or higher, and no effect of fluid flow on the microchannel was observed.

[0116] [Table 5]

[0117] These results demonstrate that lipid nanoparticles according to the present invention can be produced using the production apparatus shown in Figure 3 under the same conditions as when using the production apparatus shown in Figure 2. This suggests that the structure selectivity of the mixer is low in the production of lipid nanoparticles according to the present invention.

[0118] [Example 1] Lipid nanoparticles with different lipid compositions loaded with RNP for knocking out the GFP gene were manufactured, and their GFP knockout activity in HeLa-GFP cells was examined. Lipid nanoparticles with different lipid compositions loaded with RNP for knocking in the GFP gene to the BFP (blue fluorescent protein) gene were also manufactured, and their GFP knockin activity in HeLa-GFP cells was examined. Lipid nanoparticles were manufactured using the manufacturing equipment used in Reference Example 1.

[0119] The constituent lipids of the lipid nanoparticles were a pH-sensitive cationic lipid, CL4H6 (pKa 6.25, Patent Document 2), a neutral phospholipid, 1,2-distearoyl-sn-glycero-3-phosphatidylcholine (DSPC) or 1,2-dioleoyl-sn-glycero-3-phosphatidylethanolamine (DOPE), and other lipids, such as cholesterol (chol) and 1,2-dimyristoyl-rac-glycero-3-methoxypolyethylene glycol-2000 (PEG-DMG).

[0120] [Chemical formula]

[0121] For the production of lipid nanoparticles, lipid components with five types of lipid compositions described in Table 6 were used.

[0122]

Table 6

[0123] The Cas9 protein, crRNA, and tracrRNA used were the same as those used in Reference Example 1. When producing RNP for GFP knockout, as the ssON for GFP knockout, a 132-base DNA consisting of the same base sequence as the partial region of genomic DNA containing the target sequence (SEQ ID NO: 1) (the 68th to 87th regions of SEQ ID NO: 5 are the same base sequence as the target sequence) was used. When producing RNP for knock-in from GFP to BFP, as the ssON for BFP knock-in, DNA (SEQ ID NO: 6) in which mutations were introduced into three bases (the 65th, 67th, and 72nd bases) of the ssON used for knockout was used. When the homology-dependent repair (HDR) pathway functions due to the introduction of the mutant sequence, the amino acids (threonine-tyrosine-glycine) constituting the fluorophore of GFP are replaced with the amino acids (serine-histidine-glycine) constituting the fluorophore of BFP.

[0124] <Measurement of GFP knockout activity> An RNP solution was prepared by dissolving 160 nM of Cas9 protein, 160 nM of crRNA, 160 nM of tracrRNA, and 160 nM of ssON in the pH 6.0 buffer solution used in Reference Example 1. The solution was then loaded into syringe 102 connected to inlet 104b of the second inlet. A lipid ethanol solution (total lipid: 8.20 mM) was loaded into syringe 101 connected to inlet 104a of the first inlet. The solution was pumped into the microchannel under conditions of FRR of 9.0 and TFR of 500 μL / min, producing RNP-loaded lipid nanoparticles. The RNP-loaded lipid nanoparticles were dialyzed as described in Reference Example 1, and then the number-average particle size (nm), polydispersity index (PdI), and ζ potential (mV) were measured by dynamic light scattering using a particle analyzer "Zetasizer Nano ZS ZEN3600" (Marvern) (n = 3, Mean ± SD). The encapsulation rate (%) of gRNA and ssON in the lipid nanoparticles and their concentrations in the lipid nanoparticle solution were measured using a Ribogreen assay (Thermo Fisher Scientific) (n = 3, Mean ± SD). The ratio (%) of the total amount of nucleic acid recovered after dilution and purification by dialysis using the apparatus described in Figure 3 to the total amount of gRNA and ssON used in the production of RNP-loaded lipid nanoparticles was calculated from the measured concentrations of gRNA and ssON in the lipid nanoparticle solution.

[0125] [Table 7]

[0126] The measurement results are shown in Table 7. Lipid nanoparticles with a composition containing 20% DOPE tended to have a small particle size and a high encapsulation rate.

[0127] Each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the previous day so that the Cas9 protein concentration was 0.3, 1, or 5 nM, and the cells were cultured. Two days after the addition of the RNP-loaded lipid nanoparticle, the medium was changed, and one day later, the cells were collected, and the GFP knockout efficiency (%) (the percentage of cells not emitting GFP fluorescence in the whole cells) was measured by flow cytometry. As a comparison, a commercially available transfection reagent, "Lipofectamine RNAiMAX" (manufactured by Thermo Fisher Scientific), which has a proven track record in introducing RNP, was used.

[0128] The measurement results (n = 3, Mean ± SD) of the GFP knockout efficiency (%) of each RNP-loaded lipid nanoparticle are shown in Fig. 4. The higher the content of neutral phospholipid, the higher the knockout activity was shown. Also, the RNP-loaded lipid nanoparticle containing DOPE as the neutral phospholipid showed higher knockout activity than the RNP-loaded lipid nanoparticle containing DSPC. In addition, the RNP-loaded lipid nanoparticle with a high content of neutral phospholipid or containing DOPE as the neutral phospholipid showed higher knockout activity than "Lipofectamine RNAiMAX".

[0129] <Measurement of GFP knock-in activity> RNP-loaded lipid nanoparticles were prepared in the same manner as described above, except that knock-in ssONs were used instead of GFP knock-out ssONs, and the lipid composition was changed to a 20% DSPC content (20% (DSPC) in Table 6) or a 20% DOPE content (20% (DOPE) in Table 6). Each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before, at a Cas9 protein concentration of 0.3, 1, or 5 nM, and the cells were cultured. The medium was changed two days after the addition of the RNP-loaded lipid nanoparticles, and the cells were harvested one day later. The knock-in efficiency (%) (the percentage of cells emitting BFP fluorescence) was measured by flow cytometry. Because BFP has a shorter fluorescence wavelength than GFP, cells in which the GFP gene had been modified with the BFP gene by knock-in can be quantitatively identified by flow cytometry. The ratio of knock-in efficiency (%) to knock-out efficiency (%) ([KI (%)] / [KO (%)]) was calculated for each RNP-loaded lipid nanoparticle, and the results are shown in Figure 5. As a result, when ssONs with mutated genes were used, all RNP-loaded lipid nanoparticles showed a gene knock-in efficiency of 4-5%.

[0130] [Example 2] The effect of lipid nanoparticle formulation on gene knockout activity was investigated.

[0131] <First screening> The items examined were the pH-sensitive cationic lipid (CL) content (30, 40, 50 mol%), neutral phospholipid (PL) content (20, 35, 50 mol%), PEG-DMG content (1, 2.5, 4 mol%), type of pH-sensitive cationic lipid (CL4H6 or CL15H6 (pKa ∼7.25, Patent Document 2)), type of neutral phospholipid (DSPC or DOPE), and RNP / lipid ratio (1.8, 2.7, 3.6 × 10 -4 The six factors were:

[0132] Three levels were selected for continuous variables and two levels for categorical factors. From a total of 324 combinations, 14 formulations listed in Table 8 were selected using a definitive screening design, a type of experimental design (DoE). The lipid composition was pH-sensitive cationic lipid (CL):neutral phospholipid (PL):cholesterol:PEG-DMG = X:Y:(100-XY):Z (mol%) (X: the value in the "CL [%]" column in Table 8, Y: the value in the "PL [%]" column in Table 8).

[0133] [Table 8]

[0134] [ka]

[0135] RNP-loaded lipid nanoparticles for GFP knockout were produced in the same manner as for RNP-loaded lipid nanoparticles for GFP knockout in Example 1, except that the ratio of lipid components, lipid, and RNP was as shown in Table 8. The number-average particle size (nm) of the obtained RNP-loaded lipid nanoparticles, and the encapsulation rates (%) of PdI, gRNA, and ssON in the lipid nanoparticles were measured in the same manner as in Example 1.

[0136] Each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before, so that the Cas9 protein concentration was 0.1, 0.5, or 2 nM, and the cells were cultured. The medium was replaced two days after the addition of the RNP-loaded lipid nanoparticles, and the cells were then harvested three days later. The GFP knockout efficiency (%) was measured by flow cytometry in the same manner as in Example 1. The results are shown in Table 9. Figure 6 shows the measurement results of the GFP knockout efficiency (%).

[0137] [Table 9]

[0138] Statistical analysis was performed on the number-average particle size. Table 10 shows the main effects and two-factor interactions that significantly affected the number-average particle size, and Figure 7 shows the predicted profile of the number-average particle size when factors significantly affected the number-average particle size were varied. The pH-sensitive cationic lipid content, PEG-DMG content, type of pH-sensitive cationic lipid, and neutral phospholipid were detected as factors that significantly affected the particle size. More specifically, it was found that particle size could be reduced by increasing the pH-sensitive cationic lipid content, setting the PEG-DMG content to around 2.5 mol%, using CL4H6 as the pH-sensitive cationic lipid, or using DOPE as the neutral phospholipid. In particular, it was found that the amount of PEG lipid and the type of neutral phospholipid had a significant impact.

[0139] [Table 10]

[0140] Statistical analysis was also performed on the encapsulation rate. Table 11 shows the results of the main effects and two-factor interactions that significantly affected the encapsulation rate, and Figure 8 shows the predicted profile of the encapsulation rate when factors that significantly affected the encapsulation rate were changed. The results revealed that all factors significantly affected the encapsulation rate. More specifically, it was revealed that the encapsulation rate was improved by increasing the pH-sensitive cationic lipid content, increasing the neutral phospholipid content, decreasing the PEG-DMG content, using CL15H6 as the pH-sensitive cationic lipid, using DOPE as the neutral phospholipid, and lowering the RNP / lipid ratio. The type of neutral phospholipid was the factor with the greatest impact.

[0141] [Table 11]

[0142] Statistical analysis was also performed on gene knockout activity (%). Table 12 shows the results of the main effects and two-factor interactions that significantly affected gene knockout activity (%), and Figure 9 shows the predicted profile of gene knockout activity (%) when factors that significantly affected gene knockout activity (%) were varied. PEG-DMG content, type of pH-sensitive cationic lipid, and type of neutral phospholipid were identified as factors that significantly affected knockout activity. More specifically, it was found that gene knockout activity was improved by increasing the PEG-DMG content to approximately 1 mol%, changing the pH-sensitive cationic lipid to CL4H6, and changing the neutral phospholipid to DOPE. In particular, the type of cationic lipid and the type of neutral phospholipid were influential factors.

[0143] [Table 12]

[0144] In fact, formulations A-7 and A-11, which fulfilled the conditions for improving gene knockout activity, showed higher knockout activity than other lipid nanoparticles (Figure 6), and the IC 50 The 50% inhibitory concentration (50%) was less than 0.1 nM (equivalent to the Cas9 protein concentration), and the maximum knockout efficiency reached over 95% (Figure 6).

[0145] <Secondary screening> Following the results of the primary cloning, a secondary screening was performed using the same experimental system as that used in the primary screening. The three factors and three levels of pH-sensitive cationic lipid (CL) content (30, 40, 50 mol%), neutral phospholipid (PL) content (20, 35, 50 mol%), and PEG-DMG content (1.0, 2.0, 3.0 mol%) were investigated. From a total of 27 combinations, a definitive screening design was used to select nine formulations, as shown in Table 13. The lipid composition was pH-sensitive cationic lipid (CL):neutral phospholipid (PL):cholesterol:PEG-DMG = X:Y:(100-XY):Z (mol%) (X: the value in the "CL [%]" column in Table 13, Y: the value in the "PL [%]" column in Table 13). CL4H6 was used as the pH-sensitive cationic lipid, and DOPE was used as the neutral phospholipid.

[0146] [Table 13]

[0147] RNP-loaded lipid nanoparticles for GFP knockout were produced in the same manner as in the primary screening, except that the ratio of lipid components, lipids, and RNP was as shown in Table 13. The number-average particle size (nm) of the obtained RNP-loaded lipid nanoparticles, and the encapsulation rates (%) of PdI, gRNA, and ssON in the lipid nanoparticles were measured. Furthermore, in the same manner as in the primary screening, each RNP-loaded lipid nanoparticle was added to HeLa-GFP cells, and the GFP knockout efficiency (%) was measured. The results are shown in Table 13.

[0148] Statistical analysis was performed on the number average particle size. Table 14 shows the results of the main effects that significantly affect the number average particle size, and Figure 10 shows the predicted profile of the number average particle size when factors that significantly affect the number average particle size are changed. It was revealed that the particle size can be controlled to be small by increasing the amount of PEG lipid.

[0149] [Table 14]

[0150] A similar statistical analysis was performed on the encapsulation rate. Table 15 shows the results of the main effects that significantly affected the encapsulation rate, and Figure 11 shows the predicted profile of the encapsulation rate when factors that significantly affected the encapsulation rate were changed. The results revealed that the encapsulation rate improved by increasing the pH-sensitive cationic lipid content, increasing the neutral phospholipid content, or decreasing the PEG-DMG content.

[0151] [Table 15]

[0152] A similar statistical analysis was performed on gene knockout activity (%). Table 16 shows the results of the main effects that significantly affected gene knockout activity (%), and Figure 12 shows the predicted profile of gene knockout activity (%) when factors that significantly affected gene knockout activity (%) were changed. It was revealed that gene knockout activity was improved by reducing the PEG-DMG content or increasing the pH-sensitive cationic lipid content to 40 mol% or more.

[0153] [Table 16]

[0154] Based on these results, we used CL4H6 as the pH-sensitive cationic lipid and its content was set to 40-50 mol%, DOPE as the neutral phospholipid and its content was set to 20-50 mol%, the amount of PEG lipid was set to 1.5-2.0 mol%, and the RNP / lipid ratio was set to 3.6 × 10 -4 It was found that by setting the concentration to 1 mol or more, good RNP-loaded lipid nanoparticles with a small number-average particle size, high RNP encapsulation efficiency, and high gene knockout activity (%) could be obtained.

[0155] [Example 3] The RNP-loaded lipid nanoparticles B-4 and B-9 produced in Example 2 were examined for cytotoxicity and storage stability.

[0156] <Evaluation of cytotoxicity> Each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before to a Cas9 protein concentration of 0.3, 0.5, or 1 nM, and the cells were cultured. One day after the addition of the RNP-loaded lipid nanoparticles, a WST-8 assay was performed. The WST-8 assay was performed using a cell counting kit (product name: "Cell Counting Kit-8," manufactured by Dojindo Laboratories). A similar WST-8 assay was performed on HeLa-GFP cells cultured in PBS(-) instead of the RNP-loaded lipid nanoparticles.

[0157] The cell viability of cells to which each RNP-loaded lipid nanoparticle was added ([WST-8 assay measurement value of cells to which RNP-loaded lipid nanoparticles were added] / [WST-8 assay measurement value of control cells] × 100:%) was calculated. The results are shown in Figure 13. Regardless of the amount of RNP-loaded lipid nanoparticles, the cell viability of cells to which RNP-loaded lipid nanoparticles B-4 and B-9 were introduced was nearly 100%. In other words, these RNP-loaded lipid nanoparticles did not show significant cytotoxicity in HeLa cells.

[0158] <Evaluation of storage stability> RNP-loaded lipid nanoparticles B-9 prepared as in Example 2 were stored at 4°C, and their physical properties and knockout activity were examined after storage. Specifically, zeta potential and PdI were measured over time. Furthermore, RNP-loaded lipid nanoparticles B-9 before and after 2 weeks of storage at 4°C were added to the culture medium to a Cas9 protein concentration of 0.3 nM, and introduced into HeLa-GFP cells, and GFP knockout activity was examined. Zeta potential, PdI, and GFP knockout activity were measured as in Example 1.

[0159] The measurement results of zeta potential and PdI are shown in Figure 14, and the measurement results of GFP knockout activity are shown in Figure 15. As a result, when RNP-loaded lipid nanoparticle B-9 was stored at 4°C, both the physical properties and knockout activity were maintained at the same level as before storage for at least two weeks after production. These results confirmed that the RNP-loaded lipid nanoparticles of the present invention have sufficient storage stability for practical use.

[0160] [Example 4] We investigated the effect of the base length of the ssON loaded onto the RNP-loaded lipid nanoparticles on knockout activity.

[0161] As ssONs for GFP knockout, the 132-base-long DNA (sequence number 5) also used in Example 2, the 20-base-long DNA (same base sequence as the region from positions 68 to 87 of sequence number 5), the 60-base-long DNA (same base sequence as the region from positions 48 to 107 of sequence number 5), or the 60-base-long DNA converted into RNA (sequence number 7) were used.

[0162] First, RNP-loaded lipid nanoparticles B-9 and B-4 were prepared as in Example 2, except that 132-mer DNA, 20-mer DNA, 60-mer DNA, or 60-mer RNA was used as the ssON. Each RNP-loaded lipid nanoparticle was then added to the medium of HeLa-GFP cells seeded the day before to a Cas9 protein concentration of 0.1 nM or 0.3 nM, and the cells were cultured. Two days after the addition of the RNP-loaded lipid nanoparticles, the medium was replaced, and the cells were cultured for an additional three days. The GFP knockout efficiency (%) (the percentage of cells not emitting GFP fluorescence) was measured by flow cytometry. Figure 16A shows the results for cells transfected with RNP-loaded lipid nanoparticles at a Cas9 protein concentration of 0.1 nM, and Figure 16B shows the results for cells transfected with RNP-loaded lipid nanoparticles at a Cas9 protein concentration of 0.3 nM.

[0163] As a result, for both RNP-loaded lipid nanoparticles B-4 and B-9, the gene knockout activity tended to increase with increasing base length of the ssON. Furthermore, when the base length was the same, the knockout activity tended to be higher when the ssON was RNA than when it was DNA.

[0164] [Example 5] Instead of Cas9, we produced lipid nanoparticles loaded with the RNA-dependent DNA nuclease Cpf1 and examined their gene knockout activity.

[0165] First, instead of Cas9 protein, Cpf1 protein (product name: "Alt-R As Cas12a (Cpf1) Ultra", manufactured by Integrated DNA Technologies) was used, and instead of the crRNA and tracrRNA used in Reference Example 1, a 41-base-long RNA (SEQ ID NO: 8) or a 100-base-long RNA (SEQ ID NO: 9) was used as the gRNA. As the ssON, a 120-base-long RNA (SEQ ID NO: 10) or a 60-base-long RNA (same base sequence as the 27th to 86th region of SEQ ID NO: 10; SEQ ID NO: 11) was used. RNP-loaded lipid nanoparticles B-9 were produced in the same manner as in Example 2, except that each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before so that the Cpf1 protein concentration was 0.5, 1, or 2 nM, and the cells were cultured. Two days after the addition of RNP-loaded lipid nanoparticles, the medium was changed and the cells were cultured for a further three days before being harvested and the GFP knockout efficiency (%) (the percentage of cells that did not emit GFP fluorescence out of the total cells) was measured by flow cytometry.

[0166] The results of measuring GFP knockout efficiency are shown in Figure 17. In the figure, "NT" indicates the results for cells cultured in medium without the addition of RNP-loaded lipid nanoparticles. "gGFP" indicates the results for cells cultured in medium without the addition of RNP-loaded lipid nanoparticles using 41-mer RNA (SEQ ID NO: 8). "gGFP+59" indicates the results for cells cultured in medium without the addition of RNP-loaded lipid nanoparticles using 100-mer RNA (SEQ ID NO: 9). In the "ssON" column, "-" indicates the results for cells cultured in medium without the addition of RNP-loaded lipid nanoparticles containing no ssON. The 0.5 nM, 1 nM, and 2 nM columns indicate the results for cells in which RNP-loaded lipid nanoparticles were added to the medium so that the Cpf1 protein concentration was 0.5, 1, or 2 nM, respectively.

[0167] As shown in Figure 17, adding ssON to RNP improved GFP knockout activity, similar to the case of using Cas9 protein. Furthermore, the longer the base length of the added ssON, the greater the knockout activity tended to be. These results demonstrate that the lipid nanoparticles of the present invention are useful as carriers for introducing RNPs not only in the CRISPER / Cas9 system but also in the CRISPER / Cpf1 system.

[0168] [Example 6] The double-nicking method, which uses two pairs of Cas9 nickases (Cas9n) in which either the RuvC nuclease activity or the HNH nuclease activity is inactivated, is known as a method to reduce off-target effects (introduction of mutations into non-target gRNA regions) by Cas9. We produced lipid nanoparticles loaded with RNP for use in this double-nicking method and examined their gene knockout activity.

[0169] First, instead of Cas9 protein, Cas9n protein (product name: "Alt-R Sp Cas9 D10A Nickase V3", manufactured by Integrated DNA Technologies) was used, and as gRNA, the tracrRNA used in Reference Example 1, crRNA (SEQ ID NO: 12, 36 bases long) containing a base sequence complementary to the first target sequence in the GFP gene, and crRNA (SEQ ID NO: 13, 36 bases long) containing a base sequence complementary to the second target sequence were used. RNP-loaded lipid nanoparticles B-9 were produced in the same manner as in Example 2, except that ssON was not used. For comparison, RNP-loaded lipid nanoparticles B-9 loaded with RNP containing Cas9 protein were produced in the same manner as in Example 2.

[0170] Next, each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before so that the Cas9n protein or Cas9 protein concentration was 0.1, 0.3, 1, or 2 nM, and the cells were cultured. Two days after the addition of the RNP-loaded lipid nanoparticles, the medium was changed, and after culturing for an additional three days, the cells were harvested and the GFP knockout efficiency (%) (the percentage of cells that did not emit GFP fluorescence) was measured by flow cytometry.

[0171] The measurement results are shown in Figure 18. RNP-loaded lipid nanoparticles loaded with Cas9 protein also had high GFP knockout activity, similar to RNP-loaded lipid nanoparticles loaded with Cas9 protein. In particular, in cells to which Cas9n protein was added at a concentration of 2 nM, the GFP knockout efficiency was 98%. These results clearly demonstrate that lipid nanoparticles with a specific lipid composition according to the present invention are useful as carriers for introducing RNPs in the double-nicking method using Cas9n protein. These results also suggest that RNPs containing ssONs hybridizing with the respective gRNAs are also useful as carriers for introducing RNPs in the double-nicking method.

[0172] [Example 7] For the RNP-loaded lipid nanoparticles B-4 and B-9 produced in Example 2, they were introduced into GFP stably expressing HEK (HEK-GFP) cells to examine the knockout activity, knock-in activity, and cytotoxicity.

[0173] <Measurement of GFP knockout activity> First, in the same manner as in Example 2, RNP-loaded lipid nanoparticles B-9 and B-4 containing ssON for GFP knockout were produced. Next, each RNP-loaded lipid nanoparticle was added to the medium of HEK-GFP cells seeded the previous day so that the Cas9 protein concentration was 0.5, 1, 3, or 5 nM, and the cells were cultured. After changing the medium 3 days after the addition of the RNP-loaded lipid nanoparticles, the cells were collected 3 days later, and the knockout efficiency (%) (the ratio (%) of cells not emitting GFP fluorescence in the whole cells) was measured by flow cytometry. The measurement results are shown in Fig. 19A. For the cells transfected with either RNP-loaded lipid nanoparticles B-9 or B-4, the maximum knockout efficiency was approximately 97%.

[0174] <Measurement of GFP knock-in activity> First, RNP-loaded lipid nanoparticles B-9 and B-4 were produced in the same manner as in Example 2, except that the ssON for GFP knock-in used in Example 1 was used instead of the ssON for GFP knockout. Next, each RNP-loaded lipid nanoparticle was introduced into HEK-GFP cells and cultured in the same manner as in the measurement of the above GFP knockout activity, and then collected and subjected to flow cytometry to measure the knock-in efficiency (%) (the ratio (%) of cells emitting BFP fluorescence in the whole cells). The measurement results are shown in Fig. 19B. For the cells transfected with either RNP-loaded lipid nanoparticles B-9 or B-4, the maximum knock-in efficiency was approximately 23%.

[0175] <Cytotoxicity> Each RNP-loaded lipid nanoparticle was added to the medium of HEK-GFP cells seeded the previous day so that the Cas9 protein concentration was 0.5, 1, 2, 3, 4, or 5 nM, and the cells were cultured. One day after the addition of the RNP-loaded lipid nanoparticle, a WST-8 assay was performed in the same manner as in Example 3, and the cell viability (%) was measured. The results are shown in Fig. 20. Regardless of the amount of the RNP-loaded lipid nanoparticle, the cell viability of the cells transfected with the RNP-loaded lipid nanoparticles B-4 and B-9 was almost 100%. That is, these RNP-loaded lipid nanoparticles did not show significant cytotoxicity in HEK cells as well as in HeLa cells.

[0176] [Example 8] For the RNP-loaded lipid nanoparticle B-9 produced in Example 2, it was introduced into bone marrow-derived macrophages (BMDM) that constitutively express GFP, and the GFP knockout activity was examined.

[0177] <GFP constitutively expressing BMDM> BMDM that constitutively express GFP was obtained from GFP constitutively expressing mice (C57BL / 6-Tg(CAG-EGFP) mice) (female, 6 weeks old). Specifically, bone marrow cells were collected from the femurs and tibias of GFP constitutively expressing mice, passed through a 40 μm cell strainer, and then red blood cells were removed using a solubilization buffer (product name: "ACK lysing buffer", manufactured by Gibco). Next, the bone marrow cells were cultured in a medium containing recombinant mouse M-CSF (final concentration: 50 ng / mL, manufactured by BioLegend) and heat-inactivated FBS (final concentration: 10%) for 7 days to obtain BMDM.

[0178] <Measurement of GFP knockout activity> First, in the same manner as in Example 2, the RNP-loaded lipid nanoparticle B-9 containing ssON for GFP knockout was produced. Next, the RNP-loaded lipid nanoparticle was added to the medium in which BMDM was cultured so that the Cas9 protein concentration was 8 nM, and the cells were cultured. After changing the medium 1 day after the addition of the RNP-loaded lipid nanoparticle, the cells were collected 2 days later and flow cytometry was performed.

[0179] The results of flow cytometry of BMDM before introducing RNP-loaded lipid nanoparticles are shown in Fig. 21(A), and the results of flow cytometry of BMDM after introducing RNP-loaded lipid nanoparticles are shown in Fig. 21(B). Almost all cells of BMDM before introducing RNP-loaded lipid nanoparticles emitted GFP fluorescence, but in BMDM after introducing RNP-loaded lipid nanoparticles, the proportion of cells not emitting GFP fluorescence increased. From these results, it is clear that the lipid nanoparticles according to the present invention are useful as carriers for introducing RNPs used for genome editing and the like even for cells collected from a living body.

[0180] [Example 9] Using a device in which the mixer-built microchannel 104 of the manufacturing device described in Fig. 2 used in Reference Example 1 and the like was replaced with a channel 107 having three introduction paths as shown in Fig. 1B, RNP-loaded lipid nanoparticles were manufactured, and their physical properties and GFP knockout activity were examined.

[0181] <Production of RNP-Loaded Lipid Nanoparticles B-9> Similar to Fig. 1B, an ethanol solution of lipid was introduced from the first introduction path (the upper introduction path in the figure), an RNP solution was introduced from the second introduction path (the lower introduction path in the figure), and PBS(-) was introduced from the third introduction path (the central introduction path in the figure). RNP-loaded lipid nanoparticles B-9 were produced in the same manner as in Example 2, except that the respective FRRs (flow rate ratio: [flow rate of ethanol solution of lipid (x)] / [flow rate of PBS(-) (y)] / [flow rate of RNP solution (z)]) were as described in Table 17. In Table 16, when FRR is x / y / z = 9 / 0 / 1, it was produced using the manufacturing device with two introduction paths used in Reference Example 1.

[0182] The number-average particle size (nm) of the obtained RNP-loaded lipid nanoparticles, PdI, and the encapsulation rates (%) of gRNA and ssON in the lipid nanoparticles were measured in the same manner as in Example 1. The measurement results are shown in Table 17. As a result, even when a production device with three inlet paths was used, RNP-loaded lipid nanoparticles with physical properties equivalent to or better than those of the RNP-loaded lipid nanoparticles obtained in Example 2 using a production device with two inlet paths were able to be produced.

[0183] [Table 17]

[0184] When the manufacturing device after production was checked, when the manufacturing device with two inlet channels used in Reference Example 1 was used, aggregates were formed on the interface between the lipid ethanol solution and the RNP solution near the entrance of the dilution channel. In contrast, when a manufacturing device with three inlet channels was used and PBS was introduced from the central inlet channel, no aggregates were observed in the dilution channel.

[0185] The obtained RNP-loaded lipid nanoparticles were added to the culture medium to give Cas9 protein concentrations of 0.1, 0.3, or 1 nM, and then introduced into HeLa-GFP cells to examine GFP knockout activity, as in Example 4. The results are shown in Figure 22. All RNP-loaded lipid nanoparticles had GFP knockout activity equivalent to or greater than that of the RNP-loaded lipid nanoparticles obtained in Example 2.

[0186] These results confirmed that by using a microchannel with a built-in mixer having three inlet channels as shown in Figure 1B, RNP-loaded lipid nanoparticles with sufficient physical properties and gene knockout activity can be stably produced while avoiding the formation of aggregates within the microchannel with a built-in mixer.

[0187] [Example 10] The effect of the pH of the RNP solution during the preparation of RNP-loaded lipid nanoparticles on the knockout activity of the prepared RNP-loaded lipid nanoparticles was investigated.

[0188] RNP-loaded lipid nanoparticles containing Cas9 protein were prepared in the same manner as RNP-loaded lipid nanoparticles B-9 produced in Example 2, except that MES buffer (20 mM MES, 50 mM NaCl, pH 6.0 or 6.3) was used for the pH 6.0 or 6.3 buffer, and citrate buffer (20 mM citric acid, 50 mM NaCl, pH 4.0, 5.0, or 5.5) was used for the pH 4.0, 5.0, or 5.5 buffer when preparing the RNP solution. RNP-loaded lipid nanoparticles containing Cpf1 protein were prepared in the same manner as RNP-loaded lipid nanoparticles B-9 produced in Example 2, except that the Cpf1 protein used in Example 5 was used instead of the Cas9 protein.

[0189] Next, each RNP-loaded lipid nanoparticle was added to the medium of HeLa-GFP cells seeded the day before to a Cas9 protein concentration of 0.1 nM or 0.3 nM, or a Cpf1 protein concentration of 0.5 nM, 1.0 nM, or 2.0 nM, and the cells were cultured. After two days of culture, the medium was replaced and the cells were cultured for an additional three days. The GFP knockout efficiency (%) (the percentage of cells not emitting GFP fluorescence) was measured by flow cytometry. Figure 23A shows the results for cells transfected with RNP-loaded lipid nanoparticles containing Cas9 protein, and Figure 23B shows the results for cells transfected with RNP-loaded lipid nanoparticles containing Cpf1 protein.

[0190] As a result, the gene knockout activity was maintained in RNP-loaded lipid nanoparticles containing either Cas9 protein or Cpf1 protein when the pH during preparation was 5.0 or higher. [Explanation of symbols]

[0191] 10...first inlet channel, 20...second inlet channel, 30 dilution channel, 31...junction, 40...structure, 50...bent channel portion, 60...third inlet channel, 101, 102...syringe, 103...flow rate control device, 104...microchannel with built-in mixer, 104a...inlet of first inlet channel, 104b...inlet of second inlet channel, 104c...dilution channel, 104d...outlet of dilution channel, 105...tube, 106...recovery container, 107...microchannel with built-in chaotic mixer, 107a...inlet of first inlet channel, 107b...inlet of second inlet channel, 107c...dilution channel, 107d...outlet of dilution channel.

Claims

1. comprising a lipid component, a DNA nuclease, a guide RNA, and a single-stranded oligonucleotide; The lipid component is represented by the following general formula (I): 【Chemical 1】 [In formula (I), a represents an integer of 3 to 5; b represents 0 or 1; R １ and R ２ are each independently represented by the following general formula (A): 【Chemistry 2】 (In formula (A), q represents an integer of 1 to 9; r represents 0 or 1; s represents an integer of 1 to 3; t represents 0 or 1; u represents an integer of 1 to 8; c represents 0 or 1; v represents an integer of 4 to 12; and q + 2r + s + 2t + u + c + v is an integer of 19 or greater, except when b and c are simultaneously 0, in which case q is an integer of 3 to 5, r and t are 1, s is 1, and u + v is an integer of 6 to 10.) X represents a group represented by the following general formula (B): 【Chemistry 3】 (In formula (B), d represents an integer of 0 to 3; R ３ and R ４ are each independently C １－４ Alkyl group or C ２－４ Alkenyl group (C １－４ Alkyl group or C ２－４ The alkenyl group may have one or two hydrogen atoms substituted by a phenyl group, but R ３ and R ４ are bonded to each other to form a 5- to 7-membered non-aromatic heterocycle (one or two hydrogen atoms of the ring are C １－４ Alkyl group or C ２－４ The alkyl group may be substituted with an alkenyl group. or a 5- to 7-membered non-aromatic heterocyclic group (wherein the carbon atom of the group is (O—CO) b -, and one or two hydrogen atoms of the ring are bonded to C １－４ Alkyl group or C ２－４ (which may be substituted with an alkenyl group) The composition contains a pH-sensitive cationic lipid represented by the formula: The ratio of the pH-sensitive cationic lipid to the total amount of lipids constituting the lipid nanoparticles is 30 to 50 mol %, The ratio of the neutral phospholipid to the total amount of lipids constituting the lipid nanoparticles is 20 to 50 mol%, The ratio of the polyalkylene glycol-modified lipid to the total amount of lipids constituting the lipid nanoparticles is 1 to 4 mol%, A lipid nanoparticle, wherein the single-stranded oligonucleotide comprises a region capable of pairing with the guide RNA in a state complexed with the DNA nuclease.

2. The lipid nanoparticle according to claim 1, wherein the neutral phospholipid is a neutral glycerophospholipid having a saturated or unsaturated fatty acid residue having 12 to 24 carbon atoms.

3. The lipid nanoparticle of claim 1, wherein the neutral phospholipid is a phosphatidylethanolamine having an unsaturated fatty acid residue having 12 to 24 carbon atoms.

4. The lipid nanoparticle according to any one of claims 1 to 3, wherein the pH-sensitive cationic lipid is a polyethylene glycol-modified lipid.

5. the DNA nuclease is a Cas9 protein; The lipid nanoparticle of any one of claims 1 to 4, wherein the guide RNA consists of crRNA and tracrRNA.

6. The lipid nanoparticle of claim 5, wherein the Cas9 protein is a protein having only one of RuvC nuclease activity and HNH nuclease activity.

7. The lipid nanoparticle of any one of claims 1 to 4, wherein the DNA nuclease is a Cpf1 protein.

8. A method for genome editing in a non-human subject, comprising introducing the lipid nanoparticles according to any one of claims 1 to 7 into cells.

9. A method for producing the lipid nanoparticles according to any one of claims 1 to 7 using a flow channel structure, the flow path structure includes a first inlet path for introducing a first fluid and a second inlet path for introducing a second fluid, which are independent of each other and have a certain length, and which join together to form a single dilution flow path; the dilution flow path has a flow path portion that is two-dimensionally curved in at least a part thereof, The bent flow path portion is defined as the axial direction or extension direction of the dilution flow path upstream from this portion as the X direction, and the width direction of the dilution flow path perpendicularly intersecting this X direction as the Y direction. The flow path width of the dilution flow path upstream from this portion is defined as the y direction. ０ In this case, the flow is alternately distributed from both side walls of the dilution flow paths facing each other in the Y direction toward the center of the flow path in the approximately Y direction (approximately +Y direction, approximately −Y direction) by 1 / 2y ０ More than 1y ０ A constant height h less than １ , h ２ ... and has a constant width x in the X direction １ , x ２ ... and the structure that regulates the flow path width of the dilution flow path is １ , d ２ It is formed by at least two or more A method for producing lipid nanoparticles, wherein a lipid solution in which the lipid component is dissolved in ethanol is introduced through the first introduction path, and an aqueous solution containing the DNA nuclease, the guide RNA, and the single-stranded oligonucleotide and having a pH of 5.0 or higher is introduced through the second introduction path, at a total flow rate of 1 μL / min to 100 mL / min, and the ratio of the flow rate of the aqueous solution to the flow rate of the lipid solution is 7 or more.

10. the flow path structure further includes a third inlet path for introducing a third fluid, The first inlet path, the second inlet path, and the third inlet path each have a certain length and merge to form a single dilution flow path so that the first fluid introduced from the first inlet path comes into contact with the third fluid introduced from the third inlet path before merging with the second fluid introduced from the second inlet path. The method for producing lipid nanoparticles according to claim 9.

Images