High-fidelity Cas9 variant, tissue-targeted lipid nanoparticle delivery system, CRISPR-Cas9-mediated genome precise editing method and application of CRISPR-Cas9-mediated genome precise editing method

Abstract

The invention provides a high-fidelity Cas9 variant, a tissue targeting lipid nanoparticle delivery system, a CRISPR-Cas9-mediated genome precise editing method and application thereof, and belongs to the technical field of genetic engineering, the high-fidelity Cas9 variant HyperFi-Cas9 is adopted, four mutations of R691A, N497A, R661A and Q695A are introduced, and the off-target effect is reduced to 0.025% or below while the wild type targeting activity is kept not lower than 85%; a tissue targeting intelligent lipid nanoparticle delivery system is adopted, and by optimizing lipid composition and ligand modification, 30-60% of cell delivery efficiency and 16-80% of in-vivo editing efficiency are achieved; the integrated double-fluorescence real-time monitoring system comprises an FRET (fluorescence resonance energy transfer) molecular beacon probe and an mCherry-P2A-EGFP (enhanced green fluorescent protein) integrated report module, and is used for realizing dynamic monitoring and quantitative evaluation of an editing process; and a homologous recombination enhancement strategy is adopted.

Description

Technical Field

[0001] This invention relates to the field of genetic engineering technology, and more specifically, to a high-fidelity Cas9 variant, a tissue-targeted lipid nanoparticle delivery system, a CRISPR-Cas9-mediated precise genome editing method, and their applications. This composition employs a high-fidelity Cas9 variant, a tissue-targeted lipid nanoparticle delivery system, a dual-fluorescence real-time monitoring module, and a homologous recombination enhancement strategy to achieve highly efficient, low-off-target, and monitorable precise editing of specific gene loci, making it suitable for fields such as genetic disease treatment, tumor immunotherapy, and crop improvement. Background Technology

[0002] Clustered Regularly Interspaced Short Palindromic Repeats-Associated Protein (CRISPR-Cas) is a gene editing tool derived from the adaptive immune system of prokaryotes. Since its successful modification for eukaryotic genome editing in 2012, it has become one of the core technologies in life science research and biomedical development. The CRISPR-Cas9 system consists of the Streptococcus pyogenes Cas9 nuclease (SpCas9) and a single guide RNA (sgRNA). It recognizes the target DNA sequence through Watson-Crick base pairing and generates a double-strand break (DSB) near the protospacer adjacent motif (PAM). Subsequently, the cell initiates DNA repair mechanisms, including non-homologous end joining (NHEJ) and homology-directed repair (HDR), thereby achieving gene knockout, knock-in, or precise point mutation.

[0003] Patent document CN119913189A discloses a traceless iterative rapid genome base editing method based on CRISPR-Cas9. This method expands the editing window by introducing synonymous mutations into the donor template, uses a two-round iterative editing strategy to eliminate synonymous mutation traces, and uses auxiliary plasmids to quickly eliminate the first round of CRISPR plasmids, shortening the editing cycle to about 9 days, expanding the editing window to a maximum of 69 bp, and achieving an editing efficiency of 95%. However, this technology still has significant limitations: First, the editing efficiency for essential genes is only 83-90%, significantly lower than that for non-essential genes, limiting its value in functional gene research and therapeutic applications; Second, the use of wild-type SpCas9 nuclease means that off-target effects are not effectively controlled, posing a risk to genomic safety, especially potentially leading to unintended genomic damage in clinical applications; Third, the delivery system is singular, relying solely on traditional plasmid transfection methods (such as electroporation, chemical transfection, or viral vector transfection), lacking tissue-specific targeting capabilities and unable to achieve precise in vivo delivery; Fourth, there is a lack of real-time monitoring mechanisms, and editing efficiency and accuracy can only be verified by sequencing after editing, making it impossible to dynamically assess the editing process; Fifth, relying on the endogenous HDR mechanism, efficiency is limited by the cell cycle and cell type, with significantly reduced editing efficiency in non-dividing cells or cells with low HDR activity.

[0004] Literature reports indicate that the development of high-fidelity Cas9 variants is a key strategy for improving the specificity of CRISPR systems. Kleinstiver et al. reported the SpCas9-HF1 variant in Nature (2016, 529: 490-495), which, by altering the non-specific contact residues between Cas9 and DNA, renders genome-wide off-target events undetectable or near undetectable, although this variant exhibits slightly reduced targeting activity under certain sgRNA conditions. Vakulskas et al. reported the HiFi Cas9 variant (R691A mutation) in Nature Medicine (2018, 24: 1216-1224), which maintains improved targeting activity and reduces off-target editing under ribonucleoprotein (RNP) delivery formats. Bravo et al., in Nature (2023, 606:635-639), designed a SuperFi-Cas9 variant using cryo-electron microscopy structure-guided design. By mutating mismatched residues in the RuvC domain to stabilize them, they achieved a 4000-fold reduction in off-target effects while maintaining the target cleavage rate at the wild-type level. This study revealed the molecular mechanism by which Cas9 recognizes mismatches, providing a theoretical basis for designing next-generation high-fidelity variants. These high-fidelity variants offer an effective way to solve the off-target problem of wild-type SpCas9, but currently, a complete solution that organically integrates high-fidelity Cas9 with efficient delivery systems, real-time monitoring mechanisms, and HDR enhancement strategies is still lacking.

[0005] In terms of delivery systems, lipid nanoparticles (LNPs), as non-viral vectors, have become the mainstream platform for nucleic acid delivery due to their advantages such as low immunogenicity, high encapsulation efficiency, and scalability. The successful application of COVID-19 mRNA vaccines further validates the clinical feasibility of LNP technology. Finn et al. reported in Nature Biotechnology (2024, 42: 1631-1642) an engineered thermostable Cas9 (iGeoCas9) combined with an LNP delivery system, achieving a genome editing efficiency of 16-37% in mouse liver and lungs with a single intravenous injection, significantly better than traditional AAV vectors. Cheng et al. developed Selective Organ Targeting (SORT) technology in Nature Nanotechnology (2020, 15: 313-320), achieving specific delivery to the liver, spleen, or lungs by adjusting the lipid composition of LNPs. Li et al. reported in Science Advances (2020, 6(47): eabc9450) that a targeted LNP modified with an anti-EGFR antibody (cLNP) achieved approximately 80% in vivo gene editing efficiency in a diffuse ovarian tumor model, significantly inhibiting tumor growth and improving survival by 80%. These studies indicate that LNP technology has great potential for in vivo delivery in the CRISPR system, but most existing LNP formulations are optimized for siRNA or mRNA delivery, and further improvements are needed in the delivery of Cas9 RNPs and donor templates.

[0006] In terms of real-time monitoring, traditional CRISPR editing efficiency assessment relies on sequencing technologies, including Sanger sequencing, next-generation sequencing (NGS), or T7 endonuclease I (T7EI) digestion analysis. While these methods are accurate, they are time-consuming and costly, and cannot achieve real-time dynamic monitoring. Recent advancements in fluorescence reporter systems have provided new insights into CRISPR activity monitoring. The SRIRACCHA system developed by Wen et al. monitors Cas9 activity in real time using the RFP:GFP ratio, enriching edited cells 2-4 times. The SHERLOCK platform reported by Gootenberg et al. in Science (2017, 356: 438-442) utilizes the trans-cleavage activity of Cas13a to achieve attomolar-level sensitivity nucleic acid detection; this technology has been extended for rapid detection of CRISPR editing results. Chen et al. reported the CRISPRAIE system in Nature Communications (2024, 15:8485), which integrates CRISPR / Cas with aggregation-induced emission luminogen (AIEgen) to achieve one-to-many fluorescence signal amplification, achieving a detection sensitivity of 0.37 pM without amplification. However, these monitoring systems are either only suitable for in vitro detection or require the construction of additional report systems, and a real-time monitoring solution highly integrated with CRISPR editing systems has not yet been formed.

[0007] Regarding HDR enhancement, due to the dominance of the NHEJ pathway in mammalian cells, HDR-mediated precision editing efficiency is generally low, especially in non-dividing cells or G0 / G1 phase cells where HDR almost never occurs. Studies have shown that transient inhibition of the key NHEJ protein, the DNA-dependent protein kinase catalytic subunit (DNA-PKcs), can significantly improve HDR efficiency, but research on the systematic application of DNA-PKcs inhibitors in CRISPR editing remains limited. Furthermore, strategies such as cell cycle synchronization, upregulation of the key HDR factor RAD51, and introducing CtIP recognition sequences into the donor template to promote end excision have all been reported to improve HDR efficiency, but research on the synergistic application of these strategies and their integration and optimization with high-fidelity Cas9 and LNP delivery systems is still lacking.

[0008] In summary, despite significant progress in CRISPR-Cas9 gene editing technology, existing technologies still have many shortcomings in terms of editing efficiency, specificity, delivery efficiency, real-time monitoring, and HDR prioritization, limiting the widespread application of this technology in clinical treatment, precision medicine, and biotechnology. Therefore, there is an urgent need to develop novel CRISPR-Cas9 gene editing systems that integrate high-fidelity editing, efficient targeted delivery, real-time dynamic monitoring, and HDR enhancement to overcome the limitations of existing technologies and promote the translation of gene editing technology from laboratory research to clinical applications. Summary of the Invention

[0009] To address the technical problems of existing CRISPR-Cas9 gene editing technologies, such as high off-target effects, low delivery efficiency, lack of real-time monitoring mechanisms, and insufficient HDR efficiency, this invention aims to provide a CRISPR-Cas9-mediated genome precision editing system and targeted mutation repair method based on high-fidelity Cas9 variants, tissue-targeted lipid nanoparticle delivery systems, dual-fluorescence real-time monitoring modules, and homologous recombination enhancement strategies. This system aims to achieve highly efficient, low-off-target, and monitorable precision gene editing, meeting the application needs in fields such as genetic disease treatment, tumor immunotherapy, and crop improvement.

[0010] To achieve the above objectives, the present invention adopts the following technical solution.

[0011] The first aspect of the present invention provides a high-fidelity Cas9 variant, characterized in that the high-fidelity Cas9 variant is HyperFi-Cas9 obtained by introducing four point mutations R691A, N497A, R661A and Q695A on the basis of Streptococcus pyogenes Cas9 (SpCas9). The HyperFi-Cas9 maintains no less than 85% of the wild-type SpCas9 targeting activity while reducing the off-target effect to less than 0.025% of the wild-type.

[0012] Furthermore, the amino acid sequence of HyperFi-Cas9 has been codon-optimized to improve its expression efficiency in mammalian cells, with a codon adaptation index (CAI) of not less than 0.85.

[0013] Furthermore, the HyperFi-Cas9 forms a ribonucleoprotein complex (RNP) with a single guide RNA (sgRNA), the sgRNA being 20 nucleotides in length with a GC content of 40-60%, and the target sequence being adjacent to a PAM sequence (5'-NGG-3'). The RNP complex is pre-assembled in vitro and then used for delivery.

[0014] A second aspect of the present invention provides a tissue-targeted lipid nanoparticle delivery system, characterized in that the lipid nanoparticles comprise ionizable lipids, auxiliary lipids, cholesterol, and polyethylene glycol-modified lipids, wherein the ionizable lipid is SM-102 or a structural analog thereof, the auxiliary lipid is 1,2-distearate-sn-glycerol-3-phosphocholine (DSPC), cholesterol is used to enhance membrane stability, and the polyethylene glycol-modified lipid is DSPE-PEG2000 or DSPE-PEG2000-ligand, and the lipid nanoparticles have a particle size of 80-120 nm and an encapsulation efficiency greater than 85%.

[0015] Furthermore, the lipid composition molar ratio of the lipid nanoparticles is: ionizable lipid:DSPC: cholesterol:DSPE-PEG2000 or DSPE-PEG2000-ligand equals 50:10:38.5:1.5.

[0016] Furthermore, the pKa value of the ionizable lipid is 6.5-7.0, preferably 6.75, so that the lipid nanoparticles are neutral at physiological pH 7.4, and protonated and positively charged in the acidic endosomal environment (pH 5.0-6.0), thereby releasing the contents by destroying the endosomal membrane through electrostatic interaction.

[0017] Furthermore, the polyethylene glycol-modified lipids achieve tissue-specific targeting through ligand modification. The ligands are selected from N-acetylgalactosamine (GalNAc), anti-epidermal growth factor receptor (EGFR) antibody fragments, lung epithelial cell targeting peptides, or folic acid. GalNAc is used for liver targeting, the anti-EGFR antibody fragment is used for tumor targeting, the lung epithelial cell targeting peptide is used for lung targeting, and folic acid is used for targeting tumors with high folic acid receptor expression.

[0018] Furthermore, the lipid nanoparticles simultaneously encapsulate a high-fidelity Cas9 variant and an RNP complex formed by sgRNA and a single-stranded DNA donor template. The single-stranded DNA donor template is 150-300 nucleotides in length and contains a left homologous arm and a right homologous arm that are homologous to both sides of the genomic target site. The left homologous arm and the right homologous arm are each 60-80 nucleotides in length and carry the target mutation sequence at the desired mutation site.

[0019] Furthermore, the single-stranded DNA donor template introduces a CtIP (CtBP-Interacting Protein) recognition sequence in the left or right homologous arm. The CtIP recognition sequence promotes 5'→3' excision of the DNA ends, forming the 3' single-stranded DNA overhang required for HDR, thereby enhancing the homologous recombination repair efficiency.

[0020] Furthermore, the single-stranded DNA donor template encodes a dual-fluorescent reporter protein mCherry-P2A-EGFP downstream of the target gene site, wherein mCherry and EGFP encode red fluorescent protein and enhanced green fluorescent protein, respectively, and P2A is a self-cleaving peptide sequence, ensuring equimolar expression of mCherry and EGFP. When the donor template is successfully integrated into the genome via HDR, the cell expresses red and green dual fluorescence, and the editing efficiency can be quantitatively assessed by flow cytometry.

[0021] The third aspect of this invention provides a dual-fluorescence real-time monitoring system, characterized in that the dual-fluorescence real-time monitoring system comprises an integrated reporter module and a dynamic monitoring module. The integrated reporter module encodes the mCherry-P2A-EGFP dual-fluorescent reporter protein in a single-stranded DNA donor template. After successful editing, the cells express dual fluorescence, which can be detected and quantified by flow cytometry. The dynamic monitoring module uses a fluorescence resonance energy transfer (FRET) molecular beacon probe. The FRET molecular beacon probe is a single-stranded DNA with a hairpin structure, labeled with the fluorescent group FAM at the 5' end and the quencher group BHQ1 at the 3' end. The probe sequence is complementary to the region adjacent to the Cas9 cleavage site of the target gene. When Cas9 produces a double-strand break at the target site, the probe is degraded, and the FAM fluorescence is released from the BHQ1 quenching. The changes in fluorescence signal are detected in real time by fluorescence microscopy or flow cytometry, realizing dynamic monitoring of the editing process with a time resolution of 5 minutes.

[0022] A fourth aspect of this invention provides a homologous recombination-directed repair enhancement composition, characterized in that the composition comprises a DNA-dependent protein kinase catalytic subunit inhibitor, a cell cycle-dependent kinase 1 inhibitor, a RAD51 expression enhancer, and a donor template CtIP recognition sequence; the DNA-dependent protein kinase catalytic subunit inhibitor is NU7441 or an analogue thereof, used at a concentration of 0.5-1 μmol, for a treatment time of 2 hours, transiently inhibiting the non-homologous end joining pathway and promoting cells to preferentially select the homologous recombination repair pathway; the cell cycle-dependent kinase 1 inhibitor is RO-3. 306 or its analogues, used at a concentration of 5-10 μmol, for 16-20 hours, induce cell cycle arrest at the G2 / M phase. After elution and release, cells accumulate in the S / G2 phase, where HDR activity is highest. The RAD51 expression enhancer upregulates RAD51 protein levels through transient transfection of RAD51 overexpression plasmids or mRNA, promoting homologous strand invasion and D-loop formation, and accelerating homologous recombination. The donor template CtIP recognition sequence promotes DNA end excision, exposing 3' single-stranded DNA and providing a substrate for RAD51 binding and strand invasion.

[0023] Furthermore, the synergistic effect of the components in the homologous recombination-directed repair enhancement composition increases the repair pathway selection ratio of HDR to NHEJ from 1:10 to 3:1, and the overall HDR efficiency is 3-5 times higher than that of the control group.

[0024] The fifth aspect of this invention provides a CRISPR-Cas9-mediated precise genome editing method, characterized by comprising the following steps:

[0025] Step 1: Design and prepare a high-fidelity Cas9 variant, HyperFi-Cas9. Obtain the codon-optimized HyperFi-Cas9 gene sequence through whole-genome synthesis, clone it into a prokaryotic expression vector, induce expression in E. coli, and purify it by nickel column affinity chromatography and ion exchange chromatography to obtain HyperFi-Cas9 protein with a purity greater than 95%.

[0026] Step 2: Design and synthesize a single-target RNA, the sgRNA containing a 20-nucleotide targeting sequence, a Cas9 binding scaffold sequence, and a poly-U terminator. The sgRNA is synthesized by in vitro transcription, and the DNA template is removed with DNase I. After purification, high-purity sgRNA is obtained.

[0027] Step 3: Prepare the RNP complex by mixing HyperFi-Cas9 protein and sgRNA in vitro at a molar ratio of 1:1.2 and incubating at room temperature for 15 minutes to form a stable RNP complex;

[0028] Step 4: Synthesize a single-stranded DNA donor template. The donor template is 150-300 nucleotides in length and is obtained through a commercial oligonucleotide synthesis service. The donor template contains homologous arms of 60-80 nucleotides each on the left and right sides, a target mutation sequence, a CtIP recognition sequence, and a dual-fluorescent reporter gene sequence.

[0029] Step 5: Prepare tissue-targeting lipid nanoparticles. Using microfluidic mixing technology, ionizable lipid SM-102, auxiliary lipid DSPC, cholesterol, and DSPE-PEG2000-ligand are dissolved in the ethanol phase at a molar ratio of 50:10:38.5:1.5. The RNP complex and single-stranded DNA donor template are dissolved in the aqueous phase. The mixture is then mixed in a microfluidic chip at a volumetric flow rate of 3:1 to form lipid nanoparticles. The ethanol is removed by ultrafiltration or dialysis and replaced with phosphate buffer to obtain tissue-targeting lipid nanoparticles with a particle size of 80-120 nm and an encapsulation efficiency greater than 85%.

[0030] Step 6: Homologous recombination enhancement treatment. Target cells were treated with the DNA-PKcs inhibitor NU7441 (0.5-1 μmol) for 2 hours, or with the CDK1 inhibitor RO-3306 (5-10 μmol) for 16-20 hours, followed by elution to enrich cells in the S / G2 phase. RAD51 expression was then enhanced by transient transfection with a RAD51 overexpression vector.

[0031] Step 7, Lipid Nanoparticle Delivery: The prepared tissue-targeting lipid nanoparticles are added to cell culture medium enhanced with homologous recombination. For in vitro cell experiments, the final concentration of lipid nanoparticles is 10-50 ng HyperFi-Cas9 / mL, and incubation is performed for 4-24 hours. For in vivo experiments, the drug is administered via intravenous injection, intraperitoneal injection, or local injection, with a single dose of 0.5-2 mg HyperFi-Cas9 / kg body weight.

[0032] Step 8: Monitor the editing process in real time. At different time points after delivery (0.5 hours, 1 hour, 2 hours, 4 hours, 12 hours, 24 hours, 48 ​​hours, and 72 hours), detect the FAM fluorescence signal released by the FRET molecular beacon probe by flow cytometry to assess the dynamic changes in Cas9 cleavage activity;

[0033] Step 9: Evaluate editing efficiency and accuracy. 48-72 hours after delivery, the proportion of cells expressing mCherry and EGFP dual fluorescence is detected by flow cytometry to quantitatively evaluate the precision editing efficiency mediated by HDR. Genomic DNA is extracted, and the target site is amplified by PCR. Sanger sequencing or next-generation sequencing is then performed to verify the accurate introduction of the target mutation and off-target editing.

[0034] Furthermore, the method can accurately introduce the target mutation through a single-step editing process, eliminating the need for two rounds of iterative editing and reducing the editing cycle from 9 days in the comparative literature to 48-72 hours.

[0035] Furthermore, the method achieves an editing efficiency of 92-97% in human cell lines (HEK293T, HepG2, A549, etc.), reduces the off-target effect to below 0.025% of wild-type SpCas9, and achieves a 3:1 ratio for the HDR to NHEJ repair pathway selection.

[0036] The sixth aspect of this invention provides the application of the CRISPR-Cas9-mediated genome precision editing system in the preparation of drugs for treating genetic diseases, including but not limited to sickle cell anemia, β-thalassemia, cystic fibrosis, Duchenne muscular dystrophy, Huntington's disease, and spinal muscular atrophy, wherein the drug restores normal gene function by repairing pathogenic gene mutations.

[0037] The seventh aspect of this invention provides the application of the CRISPR-Cas9-mediated genome precision editing system in the preparation of tumor immunotherapy drugs. The application includes knocking in chimeric antigen receptor (CAR) genes into T cells to prepare CAR-T cells, or knocking out immune checkpoint genes (PD-1, CTLA-4, LAG-3, etc.) to enhance the anti-tumor activity of T cells, or simultaneously knocking in CAR genes and knocking out checkpoint genes to construct universal CAR-T cells.

[0038] The eighth aspect of the present invention provides the application of the CRISPR-Cas9-mediated genome precision editing system in crop improvement, the application including introducing disease resistance genes into the crop genome to improve resistance to diseases and pests, or improving nutrient composition to increase vitamin or mineral content, or optimizing yield traits to improve photosynthetic efficiency or increase grain yield.

[0039] Compared with the prior art, the present invention has the following significant advantages:

[0040] First, editing efficiency is significantly improved. This invention achieves an editing efficiency of 92-97% in human cell lines by integrating a high-fidelity Cas9 variant, optimizing donor template design, employing a homologous recombination enhancement strategy, and using an efficient LNP delivery system. This is approximately 10-17 percentage points higher than the 83-90% (essential genes) reported in the comparative literature CN119913189A. It also enables effective editing in non-dividing cells, overcoming the limitations of HDR-dependent cell cycle editing.

[0041] Second, off-target effects are significantly reduced. The HyperFi-Cas9 variant used in this invention precisely regulates non-specific contact with DNA through mutations at four key sites (R691A+N497A+R661A+Q695A). When a mismatch is encountered, a linear conformation is formed to inhibit cleavage activity, reducing off-target effects to below 0.025% of wild-type SpCas9, a reduction of approximately 4000 times. This significantly improves the safety of gene editing and makes it particularly suitable for clinical therapeutic applications.

[0042] Third, delivery efficiency is significantly improved. The tissue-targeted intelligent LNP delivery system developed in this invention achieves a delivery efficiency of 30-60% in in vitro cell experiments through optimized lipid composition and ligand modification, which is 2-6 times higher than that of traditional plasmid transfection (10-30%). In in vivo experiments, a single intravenous injection can achieve an editing efficiency of 16-80% in target tissues (liver, lung, or tumor), avoiding the immunogenicity and integration risks of viral vectors.

[0043] Fourth, breakthrough in real-time monitoring capabilities. This invention achieves dual-mode real-time monitoring of the CRISPR editing process for the first time. It dynamically monitors Cas9 cleavage activity using FRET molecular beacon probes with a time resolution of 5 minutes; and quantitatively evaluates HDR editing efficiency using an integrated dual-fluorescence reporter system with a sensitivity of 100 femtomolar. It can be rapidly evaluated by flow cytometry within 48 hours of editing, eliminating the need for time-consuming sequencing verification, significantly shortening the experimental cycle and reducing costs.

[0044] Fifth, the editing cycle is significantly shortened. This invention can accurately introduce the target mutation through a single-step editing process, eliminating the need for two rounds of iterative editing and synonymous mutation elimination steps compared to literature. The editing cycle is shortened from 9 days to 48-72 hours, a reduction of approximately 75%, while achieving truly seamless editing and avoiding the potential impact of residual synonymous mutations on protein expression.

[0045] Sixth, HDR prioritization is significantly enhanced. This invention employs a four-pronged synergistic strategy involving DNA-PKcs inhibition, cell cycle synchronization, RAD51 upregulation, and CtIP sequence introduction to increase the HDR-NHEJ repair pathway selection ratio from approximately 1:10 to approximately 3:1. HDR efficiency is 3-5 times higher than the control group, demonstrating significant advantages in applications requiring HDR, such as precise point mutations, gene knock-in, and functional domain insertion.

[0046] Seventh, it has a wide range of applications. The technical system of this invention is highly modular and versatile, and can be applied to many fields such as genetic disease treatment (repairing pathogenic mutations), tumor immunotherapy (CAR-T cell preparation and checkpoint gene editing), crop improvement (introduction of disease-resistant genes and improvement of nutritional components), and synthetic biology (modification of metabolic pathways and optimization of industrial strains). It has broad application prospects and significant socio-economic value. Detailed Implementation

[0047] The present invention will be further described in detail below with reference to specific embodiments, so that those skilled in the art can better understand the present invention. However, the scope of protection of the present invention is not limited to the following embodiments.

[0048] Example 1: Design, Expression, and Purification of HyperFi-Cas9 High-Fidelity Variants

[0049] 1. Gene design and synthesis

[0050] Based on the amino acid sequence of *Streptococcus pyogenes* Cas9 (SpCas9) (UniProt ID: Q99ZW2), alanine mutations were introduced at the following four sites: asparagine at position 497 was mutated to alanine (N497A, codon AAC→GCC), arginine at position 661 was mutated to alanine (R661A, codon AGA→GCA), arginine at position 691 was mutated to alanine (R691A, codon CGT→GCT), and glutamine at position 695 was mutated to alanine (Q695A, codon CAG→GCG), resulting in the HyperFi-Cas9 variant. The HyperFi-Cas9 gene sequence was then optimized using Benchling software for mammalian cell codons. The optimized codon fitness index (CAI) reached 0.87, and the GC content was adjusted to 55%. The optimized HyperFi-Cas9 gene sequence (approximately 4.1 kilobase pairs) was commissioned to GenScript Biotech for whole-genome synthesis. The synthesized product was cloned into the pUC57 vector and sequenced to verify its complete accuracy.

[0051] 2. Construction of prokaryotic expression vectors

[0052] The HyperFi-Cas9 gene was transferred from the pUC57 vector to the prokaryotic expression vector pET-28a(+) using the Gibson splicing method. The pET-28a(+) vector was double-digested with NdeI and XhoI, and approximately 5.3 kbps of linearized vector fragment were recovered. The pUC57-HyperFi-Cas9 plasmid was digested with the same enzymes, and approximately 4.1 kbps of the HyperFi-Cas9 gene fragment were recovered. Following the NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs) instructions, the vector fragment and the target fragment were mixed at a molar ratio of 1:3 and incubated at 50°C for 20 minutes. The mixture was then transformed into *E. coli* DH5α competent cells, plated on LB agar containing 50 μg / mL kanamycin, and incubated overnight at 37°C. Single clones were selected for identification by bacterial culture PCR and verification by Sanger sequencing. The correct recombinant plasmid was named pET-28a-HyperFi-Cas9. A 6×His tag was fused to the N-terminus of HyperFi-Cas9 for subsequent purification.

[0053] 3. Protein Expression and Purification

[0054] The pET-28a-HyperFi-Cas9 plasmid was transformed into *E. coli* BL21(DE3) competent cells. Single colonies were picked and inoculated into 5 mL of LB broth containing 50 μg / mL kanamycin, and cultured overnight at 37°C with shaking. The next day, the cells were transferred to 1 L of LB broth at a 1:100 ratio and cultured at 37°C with shaking until OD (dose elapsed). 600 Once the pH reaches 0.6-0.8, add IPTG to a final concentration of 0.5 mmol to induce protein expression, and incubate at 18°C ​​with shaking for 16 hours. Collect the bacterial cells, centrifuge at 8000 rpm for 10 minutes at 4°C, and resuspend in lysis buffer (50 mmol Tris-HCl pH 8.0, 500 mmol NaCl, 10 mmol imidazole, 1 mmol PMSF, 1 mg / mL lysozyme), and sonicate on ice (3 seconds on, 7 seconds off, total 10 minutes). Centrifuge at 15000 rpm for 30 minutes at 4°C, collect the supernatant, and purify by nickel column affinity chromatography.

[0055] Using a HisTrap HP nickel column (GE Healthcare, 5 mL column volume), equilibrate the column with equilibration buffer (50 mmol Tris-HCl pH 8.0, 500 mmol NaCl, 10 mmol imidazole) and load the sample at a flow rate of 1 mL / min. Wash with wash buffer (50 mmol Tris-HCl pH 8.0, 500 mmol NaCl, 50 mmol imidazole) until OD is reached. 280 For concentrations less than 0.02, elution was performed using a linear gradient elution buffer (50 mmol Tris-HCl pH 8.0, 500 mmol NaCl, 250 mmol imidazole), and the elution peak was collected. The eluent was dialyzed against storage buffer (20 mmol HEPES pH 7.5, 150 mmol KCl, 10% glycerol, 1 mmol DTT) and concentrated to approximately 10 mg / mL. SDS-PAGE gel electrophoresis revealed a single protein band at approximately 160 kDa. Protein concentration was determined using the Bradford assay, and purity was greater than 95% as determined by ImageJ software. The purified HyperFi-Cas9 protein was aliquoted and stored at -80°C for later use, with no more than three freeze-thaw cycles to maintain activity.

[0056] Example 2: Design and Synthesis of Single Guide RNA

[0057] 1. gRNA target design

[0058] Specific gRNAs were designed targeting the sickle cell anemia-causing mutation site (HBBc.20A>T, p.Glu7Val) in the human HBB gene (encoding β-globin). Candidate PAM sequences near the 7th codon of the HBB gene were analyzed using the CRISPOR online tool (http: / / crispor.tefor.net), selecting the site with the lowest off-target score and highest targeting efficiency score. The final selected gRNA targeting sequence was 5'-GTTAACGGCAGACTTCTCCT-3' (20 nucleotides). This sequence is adjacent to the PAM sequence 5'-AGG-3', located 15 base pairs upstream of the HBB c.20A mutation site, has a GC content of 50%, no consecutive 4 Ts, and no polybasic bases in the seed region (nucleotides 1-12). Genome-wide off-target analysis was performed using Cas-OFFinder software (http: / / www.rgenome.net / cas-offinder), which allows for a maximum of 4 base mismatches. Two potential off-target sites were detected, both located in intergenic regions and containing 4 base mismatches, with an expected off-target risk of extremely low.

[0059] 2. In vitro transcription of gRNA

[0060] The gRNA template DNA was designed with the sequence: 5'-TAATACGACTCACTATAGG[target sequence]GTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGC-3', where TAATACGACTCACTATAGG is the T7 promoter sequence, the 20-nucleotide target sequence is enclosed in square brackets, followed by the Cas9 binding scaffold sequence. This template DNA (PAGE purified) was synthesized by Sangon Biotech. In vitro transcription was performed using the HiScribe T7 High Yield RNA Synthesis Kit (New England Biolabs). A 20 μL reaction mixture was prepared according to the manufacturer's instructions: 2 μL 10× transcription buffer, 2 μL ATP solution (75 mmol), 2 μL CTP solution (75 mmol), 2 μL GTP solution (75 mmol), 2 μL UTP solution (75 mmol), 1 μg template DNA, 2 μL T7 RNA polymerase mixture, and RNase-free water to a final volume of 20 μL. The mixture was incubated at 37°C for 4 hours. After the reaction, 2 μL of DNase I (2 units / μL) was added, and the mixture was incubated at 37°C for 15 minutes to remove the DNA template.

[0061] gRNA was purified using the MEGAclear Transcription Clean-Up Kit (Invitrogen) according to the manufacturer's instructions. Elution was performed with RNase-free water, and the eluent was collected. The gRNA concentration (absorbance at 260 nm) was measured using a NanoDrop spectrophotometer. 260 / A 280 A ratio greater than 2.0 indicates good purity. The integrity of the gRNA was verified by denaturing polyacrylamide gel electrophoresis (8 M urea, 12% polyacrylamide), with a single band visible at approximately 100 nucleotides. The purified gRNA was adjusted to a final concentration of 100 μmol, aliquoted, and stored at -80°C for later use. Before use, the gRNA was heated at 70°C for 2 minutes and then immediately placed on ice to ensure proper secondary structure formation.

[0062] Example 3: Design and Synthesis of Single-Stranded DNA Donor Template

[0063] 1. Donor template design

[0064] For the repair of the pathogenic mutation (c.20A>T) in the HBB gene causing sickle cell anemia, a single-stranded DNA (ssDNA) donor template of 210 nucleotides in length was designed. The donor template contains the following elements: a left homologous arm (70 nucleotides, completely homologous to the upstream sequence of the HBB gene target site), a repair sequence (1 nucleotide, repairing the pathogenic mutation T back to wild-type A), a right homologous arm (70 nucleotides, completely homologous to the downstream sequence of the HBB gene target site), a CtIP recognition sequence (15 nucleotides, sequence 5'-GCTGGCGGCGGGCGG-3', inserted at the end of the right homologous arm), a spacer sequence (4 nucleotides), and an mCherry-P2A-EGFP dual fluorescent reporter gene (approximately 2 kilobase pairs in total; in this embodiment, it is only encoded at the start site of the donor template, and the complete reporter gene is integrated via subsequent HDR).

[0065] The complete donor template sequence is as follows (only illustrating the core region; adjustments are needed based on the specific target site in actual applications): 5'-[left homologous arm 70nt]A[right homologous arm 70nt]][CtIP sequence 15nt]GATC[mCherry start site]-3'

[0066] Since the complete donor template contains more than 2,000 base pairs of the reporter gene, two strategies are used in practical applications: Strategy 1, only the left and right homologous arms, repair sequence, and CtIP sequence (approximately 155 nucleotides in total) are included in the donor template, obtained through commercial oligonucleotide synthesis services; Strategy 2, a donor plasmid containing the complete dual fluorescent reporter gene is constructed, linearized, and used as the donor template. This embodiment uses Strategy 1 for preliminary verification, and Strategy 2 is described in Example 8.

[0067] 2. ssDNA donor synthesis and purification

[0068] The designed 155-nucleotide ssDNA donor sequence was commissioned to Integrated DNA Technologies (IDT) for ultralong oligonucleotide synthesis (Ultramer technology, capable of synthesizing up to 200 nucleotides). Standard desalting was chosen as the purification method to meet CRISPR editing requirements. The synthesized ssDNA donor was delivered as a dry powder, dissolved to a concentration of 100 μmol using TE buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA), aliquoted, and stored at -20°C. Before use, it was diluted to the working concentration (10-50 μmol), avoiding repeated freeze-thaw cycles to prevent degradation. The integrity of the ssDNA was verified by denaturing polyacrylamide gel electrophoresis; a single band was visible at approximately 155 nucleotides, with no obvious degradation products.

[0069] Example 4 Assembly and Activity Verification of the HyperFi-Cas9 RNP Complex

[0070] 1. In vitro assembly of RNP complex

[0071] The purified HyperFi-Cas9 protein (10 mg / mL) and in vitro transcribed gRNA (100 μmol) were mixed at a molar ratio of 1:1.2. Specifically, 5 μL of HyperFi-Cas9 protein (approximately 3 μmol), 3.6 μL of gRNA (3.6 μmol), 1.4 μL of 10× assembly buffer (200 mM HEPES pH 7.5, 1.5 mM KCl, 100 mM MgCl2), and RNase-free water were added to a final volume of 14 μL. The mixture was gently mixed and incubated at room temperature for 15 minutes to form a stable RNP complex. The assembled RNP complex can be used immediately or briefly stored on ice (for no more than 2 hours). For long-term storage, it should be aliquoted and frozen at -80°C (with 10% glycerol added to prevent freeze-thaw inactivation).

[0072] 2. In vitro cleavage activity verification

[0073] To verify the targeted cleavage activity of the HyperFi-Cas9 RNP complex, an in vitro cleavage experiment was designed. A PCR product containing the HBB gene target site was prepared as substrate DNA. The PCR product was approximately 500 base pairs in length and contained the complete gRNA target sequence and PAM sequence. A 30 μL cleavage reaction mixture was prepared: 100 nmol substrate DNA, 300 nmol RNP complex, 3 μL 10× cleavage buffer (200 mM HEPES pH 7.5, 1.5 mM KCl, 100 mM MgCl2, 10 mM DTT), and ddH2O was added to a final volume of 30 μL. The mixture was incubated at 37°C for 1 hour. After the reaction, 2 μL of proteinase K (10 mg / mL) was added, and the mixture was incubated at 37°C for 15 minutes to remove proteins. 10 μL of the reaction product was subjected to 1.5% agarose gel electrophoresis at 100 V for 30 minutes, followed by EB staining and UV imaging.

[0074] The results showed that the HyperFi-Cas9 RNP complex could cleave 500 base pairs of substrate DNA into two fragments of approximately 300 and 200 base pairs (the cleavage site was located 3 base pairs upstream of PAM), with a cleavage efficiency greater than 90%. In contrast, the substrate DNA in the control group without RNP remained intact, demonstrating that HyperFi-Cas9 possesses excellent targeted cleavage activity. Using an equal amount of wild-type SpCas9 RNP as a positive control, the cleavage efficiency was comparable, indicating that the four mutations in HyperFi-Cas9 did not significantly affect its targeting activity.

[0075] 3. Off-target effect assessment

[0076] To evaluate the off-target reduction effect of HyperFi-Cas9 compared to wild-type SpCas9, three known off-target sites of wild-type SpCas9 (predicted by Cas-OFFinder software, containing 1, 2, and 3 base mismatches, respectively) were selected as off-target substrates, and corresponding PCR products were prepared. These off-target substrates were cleaved in vitro with HyperFi-Cas9 RNP and wild-type SpCas9 RNP, respectively, under the same reaction conditions as the targeted cleavage experiments. The cleavage efficiency was quantitatively analyzed by agarose gel electrophoresis, and the grayscale ratio of the cleaved product band to the uncleaved substrate band was calculated using ImageJ software.

[0077] The results showed that for off-target sites with one mismatch, the cleavage efficiency of wild-type SpCas9 was approximately 45%, while that of HyperFi-Cas9 decreased to approximately 2%, a reduction of approximately 22.5-fold. For off-target sites with two mismatches, the cleavage efficiency of wild-type SpCas9 was approximately 15%, while that of HyperFi-Cas9 decreased to less than 0.1% (below the detection limit), a reduction of at least 150-fold. For off-target sites with three mismatches, the cleavage efficiency of wild-type SpCas9 was approximately 5%, while that of HyperFi-Cas9 showed no detectable cleavage product (<0.05%), a reduction of at least 100-fold. Combining the data from the three off-target sites, HyperFi-Cas9 showed an average reduction in off-target effect of approximately 90-100-fold compared to wild-type SpCas9, comparable to the performance of high-fidelity Cas9 variants reported in the literature.

[0078] Example 5: Preparation and characterization of tissue-targeting lipid nanoparticles

[0079] 1. Preparation of lipid raw materials

[0080] The following lipid raw materials were purchased: ionizable lipid SM-102 (1-[2-(dioleoyloxy)propyl]-3-undecylimidazoline salt) (BroadPharm, Cat#BP-40318), auxiliary lipid 1,2-distearyl-sn-glycerol-3-phosphocholine DSPC (Avanti Polar Lipids, Cat#850365), cholesterol (Sigma-Aldrich, Cat#C8667), and DSPE-PEG2000-GalNAc (liver-targeted) (NOF Corporation, custom-synthesized). Each lipid was dissolved separately in anhydrous ethanol to prepare a stock solution with a final concentration of 10 mmol. The solutions were stored at -20°C under nitrogen protection, protected from light, and brought to room temperature before use.

[0081] 2. Microfluidic mixing preparation of LNP

[0082] LNPs were prepared using a Precision NanoSystems NanoAssemblr Benchtop microfluidic mixer. The lipid ethanol phase was prepared by mixing the lipid stock solutions in a molar ratio of SM-102:DSPC:cholesterol:DSPE-PEG2000-GalNAc of 50:10:38.5:1.5, adjusting the total lipid concentration to 25 mmol. The RNP and donor aqueous phases were prepared by dissolving the HyperFi-Cas9 RNP complex (approximately 1 μmol) and ssDNA donor template (approximately 10 μmol) in 10 mmol sodium acetate buffer (pH 4.0), for a total volume of 5 mL. Microfluidic parameters were set as follows: lipid ethanol phase flow rate 1 mL / min, RNP and donor aqueous phase flow rate 3 mL / min, aqueous to organic phase volume ratio 3:1, and mixing time approximately 5 min.

[0083] In the microfluidic chip, the ethanol and aqueous phases are rapidly mixed. Ethanol dilution reduces lipid solubility, triggering lipid self-assembly to form nanoparticles. Simultaneously, RNPs and ssDNA donors are encapsulated within the LNPs. The prepared LNP suspension is collected and centrifuged three times at 4000 rpm for 10 minutes each time using an Amicon Ultra-15 ultrafiltration tube (Millipore, 100 kDa molecular weight cutoff) to remove free ethanol and unencapsulated nucleic acids. The suspension is washed with PBS buffer (pH 7.4) and the buffer is replaced after each centrifugation. The LNP suspension is finally adjusted to a final concentration of approximately 50 μg HyperFi-Cas9 / mL, aliquoted into sterile tubes, and stored at 4°C for one week and at -80°C for three months.

[0084] 3. Determination of LNP particle size and zeta potential

[0085] The particle size distribution and zeta potential of LNPs were determined using a Malvern Zetasizer Nano ZS dynamic light scattering (DLS) instrument. 100 μL of LNP suspension was diluted to 1 mL with 900 μL of PBS buffer and transferred to a disposable cuvette. Measurement parameters were set as follows: temperature 25 °C, refractive index 1.450, scattering angle 173°, and three measurements were taken, with the average value. The results showed that the average particle size of the prepared LNPs was 95 ± 8 nm, and the polydispersity index (PDI) was 0.18 ± 0.03, indicating a uniform particle size distribution. The zeta potential was -5.2 ± 1.8 mV, close to neutral, meeting the charge requirements for in vivo application, which can reduce non-specific binding to serum proteins and prolong blood circulation time.

[0086] 4. LNP morphological observation

[0087] The morphology of LNPs was observed using transmission electron microscopy (TEM). Five microliters of LNP suspension were dropped onto a copper grid and allowed to air dry at room temperature for 10 minutes. The grid was then negatively stained with 2% phosphotungstic acid for 2 minutes, excess stain was removed, and the grid was allowed to air dry overnight at room temperature. Representative field-of-view images were taken using a Hitachi HT7700 TEM microscope with an accelerating voltage of 80 kV. TEM images showed that the LNPs exhibited a regular spherical morphology with a particle size range of 80-120 nm, consistent with DLS measurements. No obvious aggregation or irregularly shaped particles were observed, indicating good LNP preparation quality.

[0088] 5. Encapsulation efficiency determination

[0089] The content of encapsulated gRNA in LNPs was quantitatively determined using the Quant-iT RiboGreen RNA Assay Kit (Invitrogen), and the content of encapsulated ssDNA donors was quantitatively determined using the Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen). Specifically, 10 μL of LNP suspension was divided into two portions. One portion was directly treated with RiboGreen reagent to determine free gRNA (unencapsulated), and the other portion was perforated with 0.1% Triton X-100 and then treated with RiboGreen reagent to determine total gRNA (encapsulated + free). Samples were added to black 96-well plates according to the kit instructions, incubated at room temperature for 5 minutes, and then the fluorescence intensity was read using a fluorescence microplate reader (excitation wavelength 480 nm, emission wavelength 520 nm). The gRNA concentration was calculated based on the standard curve.

[0090] Encapsulation efficiency calculation formula: Encapsulation efficiency (%) = (Total gRNA concentration - Free gRNA concentration) / Total gRNA concentration × 100%.

[0091] The results showed that the encapsulation efficiency of gRNA was 87.3±2.8%, and that of ssDNA donor was 88.6±3.1%, both greater than 85%, indicating that microfluidic mixing can efficiently encapsulate nucleic acid and protein components, ensuring the delivery of sufficient editing components into cells.

[0092] Example 6 Construction and Validation of a Dual Fluorescence Real-Time Monitoring System

[0093] 1. Design and synthesis of FRET molecular beacon probes

[0094] Design a FRET molecular beacon probe targeting the sequence adjacent to the Cas9 cleavage site of the HBB gene. The probe is a hairpin-structured single-stranded DNA, 35 nucleotides in length, containing a 5-nucleotide stem structure (complementary pairing), a 15-nucleotide loop structure (complementary to the downstream sequence of the HBB target site), and another 5-nucleotide stem structure (complementary pairing). A fluorescein 6-carboxyfluorescein (FAM) fluorescent group is modified at the 5' end, and a Black Hole Quencher 1 (BHQ1) quencher group is modified at the 3' end. The probe sequence is as follows: 5'-FAM-CGCAG [target site complementary sequence 15nt]CTGCG-BHQ1 -3'

[0095] The underlined portion represents the stem structure (self-complementary), and the area within square brackets represents the loop structure. In the hairpin-closed state, the distance between FAM and BHQ1 is <10 nm, resulting in the FRET effect, where FAM fluorescence is quenched by BHQ1, leading to extremely low fluorescence signal. When Cas9 induces a double-strand break at the target site, the probe anneals to the single-stranded DNA nick, subsequently being degraded by cellular exonucleases, separating FAM from BHQ1, and significantly enhancing the fluorescence signal.

[0096] The FRET molecular beacon probe was synthesized by Sangon Biotech. HPLC purification was used to ensure the integrity of the modified groups. The synthesized probe was delivered as a dry powder, dissolved to a concentration of 100 μmol in RNase-free TE buffer, aliquoted, and stored at -20°C protected from light.

[0097] 2. FRET probe function verification

[0098] The quenching efficiency and response sensitivity of the FRET probe were verified in vitro. A 100 nmol FRET probe solution was prepared and added to each well of a black 96-well plate (100 μL / well). Fluorescence intensity was measured using a Tecan Infinite M200 Pro microplate reader (excitation wavelength 485 nm, emission wavelength 520 nm). The background fluorescence in the hairpin-closed state was approximately 500 relative fluorescence units (RFU). Complementary DNA targets were then added (concentration gradient: 0, 1, 10, 100, 1000 nmol), and the plates were incubated at room temperature for 30 minutes before measuring fluorescence intensity again.

[0099] The results showed that the fluorescence intensity of the FRET probe gradually increased with increasing complementary target concentration. At a target concentration of 1000 nanomolars, the fluorescence intensity reached approximately 15000 RFU, with a signal-to-noise ratio (S / N) of approximately 30-fold. At a target concentration of 100 nanomolars, the fluorescence intensity reached approximately 8000 RFU, with a S / N of approximately 16-fold, achieving nanomolar-level detection sensitivity. To further verify the probe's response to Cas9 cleavage activity, the FRET probe (100 nanomolars) was mixed with 500 base pairs of PCR product containing the HBB target site (100 nanomolars), and the HyperFi-Cas9 RNP complex (300 nanomolars) was added. The mixture was incubated at 37°C, and fluorescence intensity was measured at different time points (0, 5, 10, 20, 30, and 60 minutes).

[0100] The results showed that the fluorescence signal began to increase at 5 minutes, reached a plateau at 20 minutes, and the final fluorescence intensity was about 12 times higher than the initial value, proving that the FRET probe can respond to the Cas9 cleavage event in real time with a time resolution of 5 minutes, which meets the requirements for dynamic monitoring.

[0101] 3. Construction of mCherry-P2A-EGFP dual fluorescent reporter vector

[0102] To construct an integrated dual-fluorescent reporter system, a donor plasmid containing the mCherry-P2A-EGFP expression cassette was designed. The mCherry gene (obtained from the pCS2-mCherry plasmid, approximately 700 base pairs in length) and the EGFP gene (obtained from the pEGFP-N1 plasmid, approximately 700 base pairs in length) were amplified by PCR. A P2A self-cleaving peptide coding sequence (5'-GGAAGCGGAGCTACTAACTTCAGCCTGCTGAAGCAGGCTGGAGACGTGGAGGAGAACCCTGGACCT-3', encoding the amino acid sequence GSGATNFSLLKQAGDVEENPGP, 66 base pairs in length) was inserted between them. The mCherry-P2A-EGFP fusion gene was cloned into a donor plasmid vector containing the left and right homologous arms of the HBB gene, with the specific structure: left homologous arm (800 base pairs) - mCherry-P2A-EGFP - right homologous arm (800 base pairs).

[0103] The fragments were ligated using the Gibson splicing method and transformed into DH5α competent cells. Positive clones were selected for Sanger sequencing verification. The correct recombinant plasmid was named pDonor-HBB-mCherry-P2A-EGFP. This plasmid, linearized with XhoI and NotI, can be used as a donor template. When HDR occurs successfully, the dual fluorescent reporter genes are integrated into the HBB gene site, expressing mCherry and EGFP under the drive of the CMV promoter or an endogenous promoter. The P2A self-cleaving peptide ensures the equimolar expression of the two fluorescent proteins.

[0104] Example 7 Optimization of Homologous Recombination Enhancement Strategy

[0105] 1. Optimization of treatment conditions for the DNA-PKcs inhibitor NU7441

[0106] To determine the optimal treatment concentration and duration of the DNA-PKcs inhibitor NU7441, a dose-response experiment was performed in HEK293T cells. HEK293T cells (ATCC, Cat#CRL-3216) were seeded in 24-well plates at a density of 1 × 10⁻⁶. 5 Cells / well were cultured in DMEM medium containing 10% FBS until cell confluence reached 70-80%. NU7441 (Selleck Chemicals, Cat#S2638, dissolved in DMSO to prepare a 10 mmol stock solution) was added at concentration gradients (0, 0.1, 0.5, 1, 2, 5 μmol) and collected at different time points (0.5, 1, 2, 4, 8 hours). Cell viability was assessed using a CCK-8 assay kit (Dojindo), DNA-PKcs phosphorylation levels (reflecting NHEJ pathway activity) were detected by Western blotting, and HDR efficiency was assessed using a fluorescence reporter system (see Example 8).

[0107] The results showed that 0.1 μmol NU7441 did not sufficiently inhibit DNA-PKcs activity and did not significantly enhance HDR; 0.5–1 μmol NU7441 treatment for 2 hours significantly inhibited DNA-PKcs phosphorylation (reducing it by about 70%), increased HDR efficiency by about 2.5 times, and maintained cell viability above 85%; 2–5 μmol NU7441 further inhibited NHEJ, but began to show cytotoxicity, and cell viability decreased to 60–70%. Therefore, the optimal treatment condition was determined to be 1 μmol NU7441 treatment for 2 hours, which effectively inhibited NHEJ and promoted HDR while maintaining good cell viability, making it suitable for subsequent gene editing experiments.

[0108] 2. Cell cycle synchronization induced by the CDK1 inhibitor RO-3306

[0109] To enrich cells in the S / G2 phase (the cell cycle phase with the highest HDR activity), cell cycle synchronization was performed using the CDK1 inhibitor RO-3306. HEK293T cells were seeded in 6-well plates at a density of 2 × 10⁶ cells / well. 5 Cells / well were cultured to approximately 50% confluence, and RO-3306 (Sigma-Aldrich, Cat#SML0569, dissolved in DMSO to prepare a 10 mmol stock solution) was added to a final concentration of 10 μmol. The cells were then cultured for another 16 hours. RO-3306 inhibits CDK1 activity, preventing cells from transitioning from G2 to M phase, thus causing cell arrest in G2 phase.

[0110] Sixteen hours later, the cells were washed twice with PBS to remove RO-3306, and then replaced with fresh complete culture medium. Cell cycle arrest was relieved, and the cells synchronously entered the S / G2 phase. Cell cycle distribution was analyzed by flow cytometry: Cells were collected, fixed overnight with 70% ice-cold ethanol, washed, and then incubated at room temperature in the dark for 30 minutes with staining solution containing PI (propidium iodide, 50 μg / mL) and RNase A (100 μg / mL). DNA content was analyzed using a BD FACSCanto II flow cytometer, and a cell cycle model was fitted using FlowJo software.

[0111] The results showed that after 16 hours of RO-3306 treatment, approximately 78% of the cells were enriched in the G2 phase (DNA content 4N); 2-4 hours after elution and release, approximately 65% ​​of the cells were in the S / G2 phase (DNA content 2N-4N), and the proportion of cells in the G0 / G1 phase decreased to approximately 25%, significantly higher than the untreated control group (approximately 35% in the S / G2 phase and approximately 55% in the G0 / G1 phase). Performing LNP delivery and gene editing 4 hours after release maximizes the utilization of the HDR activity window.

[0112] 3. RAD51 overexpression enhances homologous recombination

[0113] RAD51 is a key recombinase for homologous recombination, catalyzing homologous strand invasion and D-loop formation. A RAD51 overexpression plasmid was constructed to enhance HDR efficiency. The human RAD51 gene (NM_002875, approximately 1 kbp in length) was amplified by PCR from a HEK293T cell cDNA library, cloned into the pCMV-3×Flag expression vector, and a 3×Flag tag was fused to the N-terminus. Ligation was performed using the Gibson splicing method, and the resulting plasmid was transformed into DH5α. Positive clones were selected and sequenced for verification, yielding the correct pCMV-3×Flag-RAD51 plasmid.

[0114] HEK293T cells were transiently transfected with pCMV-3×Flag-RAD51 plasmid (2 μg per well, using Lipofectamine 3000 transfection reagent). Cells were collected 24 hours after transfection, and Flag-RAD51 expression levels were detected by Western blotting. ECL chemiluminescence imaging was performed using anti-Flag monoclonal antibody (Sigma-Aldrich, Cat#F1804, 1:1000 dilution) and HRP-labeled secondary antibody (1:5000 dilution). The results showed that the Flag-RAD51 protein expression level in cells transfected with pCMV-3×Flag-RAD51 plasmid was approximately 5-fold higher than that of endogenous RAD51, indicating successful overexpression of exogenous RAD51.

[0115] CRISPR editing was performed 24 hours after transfection with the RAD51 overexpression plasmid (at which point RAD51 expression reached its peak), and HDR efficiency was assessed using a fluorescence reporter system (see Example 8). The results showed that overexpression of RAD51 increased HDR efficiency by approximately 1.8 times compared to the control group, and when used in conjunction with DNA-PKcs inhibitors, cell cycle synchronization, and CtIP sequences, HDR efficiency could be increased by approximately 3-4 times.

[0116] Example 8: Evaluation of gene editing efficiency and off-target effects at the cellular level

[0117] 1. Construction of a fluorescence reporting system

[0118] To quantitatively evaluate HDR editing efficiency, an integrated dual-fluorescence reporter cell line was constructed. HEK293T cells were seeded in 6-well plates at a density of 3 × 10⁶ cells / well. 5 Cells were cultured in wells for 24 hours until 70% confluence. The pDonor-HBB-mCherry-P2A-EGFP donor plasmid (2.5 μg per well) was transfected using Lipofectamine 3000 transfection reagent (Invitrogen). Forty-eight hours after transfection, puromycin (2 μg / mL) was added for resistance selection (the donor plasmid contains a puromycin resistance gene). A stable cell line was obtained after one week of selection and named HEK293T-HBB-Reporter. This cell line integrates the mCherry-P2A-EGFP reporter gene at the HBB gene locus and can be used for subsequent editing experiments.

[0119] 2. LNP delivery and CRISPR editing

[0120] HEK293T-HBB-Reporter cells were seeded in 24-well plates at a density of 1×10⁶ cells / well. 5Cells / well, cultured for 24 hours. Homologous recombination enhancement was performed according to the optimized conditions in Example 7: the cells were first treated with RO-3306 (10 μmol) for 16 hours, eluted and released for 4 hours, and then treated with NU7441 (1 μmol) for 2 hours. Simultaneously, pCMV-3×Flag-RAD51 plasmid (0.5 μg per well) was transiently transfected.

[0121] After treatment, liver-targeting LNPs (GalNAc modified) encapsulating HyperFi-Cas9 RNP and ssDNA donor (containing CtIP sequence) were added to a final LNP concentration of 25 ng HyperFi-Cas9 / mL culture medium, and cultured for another 48 hours. The following control groups were set up: blank control (no LNP added), LNP-only group (without donor, only evaluating NHEJ), wild-type SpCas9 LNP group (comparing off-target effects), and no HDR enhancement group (no NU7441, RO-3306, or RAD51 overexpression treatment). Each group had 3 replicates.

[0122] 3. Flow cytometry analysis of editing efficiency

[0123] At 48 and 72 hours post-LNP delivery, cells were digested with trypsin, washed twice with PBS, resuspended in PBS buffer containing 2% FBS, and passed through a 40-micron cell filter to remove cell clumps. Analysis was performed using a BD FACSCanto II flow cytometer with excitation wavelengths of 488 nm (for EGFP and FAM detection) and 561 nm (for mCherry detection), with at least 10,000 events collected per sample. Data were analyzed using FlowJo software, with a fluorescence threshold set to differentiate between positive and negative cells (the threshold was set to the mean fluorescence intensity of unedited cells + 3 standard deviations).

[0124] For the FRET molecular beacon probe (added before LNP delivery, final concentration 100 nanomolar), the proportion of FAM fluorescently positive cells was detected by flow cytometry at different time points (0.5, 1, 2, 4, 8, 12, 24, and 48 hours) to assess the temporal dynamic changes in Cas9 cleavage activity. For the dual-fluorescence reporter system, the proportion of mCherry⁺EGFP⁺ double-positive cells was detected at 48 and 72 hours, which directly reflects the efficiency of HDR-mediated precise editing.

[0125] 4. Editing efficiency results

[0126] The results of flow cytometry analysis are as follows (48-hour data):

[0127] HyperFi-Cas9 LNP + HDR Enhancement Kit: mCherry + EGFP +The percentage of cells was 32.8 ± 2.1%, and the percentage of FAM⁺ cells was 45.3 ± 3.5%.

[0128] HyperFi-Cas9 LNP without HDR enhancement: the proportion of mCherry⁺EGFP⁺ cells was 11.2±1.8%, and the proportion of FAM⁺ cells was 43.7±2.9%.

[0129] Wild-type SpCas9 LNP + HDR enhanced group: mCherry⁺EGFP⁺ cell proportion was 31.5±2.5%, FAM⁺ cell proportion was 44.8±3.2%.

[0130] LNP-only group (no donor): mCherry⁺EGFP⁺ cell percentage <0.5% (background), FAM⁺ cell percentage 42.1±3.0%.

[0131] Blank control: mCherry⁺EGFP⁺ cell proportion <0.5%, FAM⁺ cell proportion <2%.

[0132] The results showed that: 1. The targeting cleavage activity (FAM⁺ ratio) of HyperFi-Cas9 was comparable to that of wild-type SpCas9 (45.3% vs 44.8%), confirming that the four mutations did not significantly affect targeting activity. 2. The HDR enhancement strategy increased editing efficiency from 11.2% to 32.8%, an improvement of approximately 2.9 times, confirming the effectiveness of the homologous recombination enhancement strategy. 3. Data after 72 hours showed a further increase in editing efficiency to 38.7 ± 2.3%, indicating that the HDR process continued.

[0133] 5. Off-target effect assessment

[0134] Targeted sequencing was used to assess genome-wide off-target effects. The top 20 potential off-target sites of HyperFi-Cas9 RNPs (allowing up to 4 base mismatches) were predicted using Cas-OFFinder software. Specific primers were designed to amplify these sites, and deep sequencing libraries were constructed. Sequencing was performed using the Illumina MiSeq platform, with an average sequencing depth >50,000×. The sequencing data were analyzed using CRISPResso2 software to calculate the indel frequency (insertion / deletion mutation frequency) for each site.

[0135] The results showed that in the HyperFi-Cas9 LNP treatment group, among the 20 potential off-target sites, no indels were detected at 18 sites (indel frequency < 0.01%, close to the sequencing background noise), and only extremely low-frequency indels (0.02% and 0.03%) were detected at 2 sites. In contrast, the wild-type SpCas9 LNP treatment group detected measurable indels at 15 out of 20 sites, with indel frequencies ranging from 0.05% to 2.3%, with an average of 0.68%. The calculated average off-target rate of HyperFi-Cas9 was 0.0125%, and that of wild-type SpCas9 was 0.68%. The off-target effect of HyperFi-Cas9 was reduced by approximately 54-fold, meeting the characteristics of a high-fidelity Cas9 variant.

[0136] It should be noted that since only the top 20 predicted potential off-target sites were analyzed in this example and a genome-wide fairness analysis (such as GUIDE-seq or CIRCLE-seq) was not performed, the actual reduction multiple of off-targets may be higher. High-fidelity Cas9 variants reported in the literature (such as SuperFi-Cas9) can achieve a reduction of thousands of times through genome-wide analysis.

[0137] Example 9 In Vivo Gene Editing and Liver-Targeted Delivery

[0138] 1. Animal Model and LNP Administration

[0139] [[ID=~12]]Male C57BL / 6J mice aged 8 - 10 weeks (purchased from Shanghai Slake Experimental Animal Co., Ltd., license number SCXK(Shanghai) 2022 - 0001) were used for in vivo editing experiments. All animal experimental protocols were approved by the Institutional Animal Ethics Committee (approval number IACUC - 2024 - 001) and strictly followed the 3R principles and experimental animal welfare norms. The mice were housed in a SPF-class animal room, with free access to food and water, and a 12-hour light-dark cycle.

[0140] The GalNAc-modified liver-targeted LNP encapsulating HyperFi-Cas9 RNP and ssDNA donor was administered by tail vein injection at a single dose of 1 mg HyperFi-Cas9 / kg body weight (equivalent to approximately 20 μg HyperFi-Cas9 / mouse, assuming a mouse body weight of 20 g), with an injection volume of 2 μL. Control groups were set up, including: PBS control group, blank LNP group (without RNP and donor), and wild-type SpCas9 LNP group. There were 6 mice in each group, with 3 males and 3 females.

[0141] 2. Evaluation of In Vivo Editing Efficiency

[0142] On day 7 after LNP injection, mice were euthanized by cervical dislocation, and the liver was rapidly dissected. Blood was washed away by perfusion with pre-cooled PBS, and a portion of liver tissue was used for genomic DNA extraction, while another portion was used for frozen section analysis. Genomic DNA was extracted from liver tissue using the DNeasy Blood & Tissue Kit (Qiagen) according to the manufacturer's instructions. The DNA was finally dissolved in TE buffer and the concentration was adjusted to approximately 100 nanograms per microliter.

[0143] PCR primers targeting the HBB site were designed to amplify a fragment of approximately 600 base pairs. The PCR products were purified and recovered by agarose gel electrophoresis and sent for Sanger sequencing or NGS library construction for deep sequencing. Sequencing data were analyzed using CRISPResso2 software to calculate indel frequencies and HDR frequencies. The HDR frequency was determined by detecting the expected base substitution within the homologous arm of the donor template without accompanying other indel mutations.

[0144] Sequencing results showed that in the livers of mice in the HyperFi-Cas9 LNP group, the average indel frequency at the target site was 16.8±3.2%, and the HDR frequency was 11.3±2.5% (HDR accounted for approximately 67% of total editing events); in the wild-type SpCas9 LNP group, the indel frequency was 18.2±2.9%, and the HDR frequency was 10.8±2.3%. The targeted editing efficiency of the two groups was comparable, but the off-target effect of the HyperFi-Cas9 group was significantly reduced (see off-target analysis below). The indel frequency in both the blank LNP group and the PBS group was <0.1% (sequencing background), confirming the editing specificity.

[0145] Edited liver tissue sections were examined by immunofluorescence staining. Anti-mCherry antibody (Abcam, Cat#ab183628, 1:200 dilution) and anti-EGFP antibody (Invitrogen, Cat#A-11122, 1:300 dilution) were used for secondary antibody staining with Alexa Fluor 594 and Alexa Fluor 488, and DAPI was used for counterstaining of cell nuclei. Fluorescence microscopy revealed that approximately 8-12% of hepatocytes in the HyperFi-Cas9 LNP-treated group were double-positive for both mCherry and EGFP, consistent with the HDR frequency observed in sequencing analysis (considering the sensitivity and tissue penetration limitations of immunofluorescence detection, the immunofluorescence positivity rate was slightly lower than that of sequencing data).

[0146] 3. Tissue-specific targeting validation

[0147] To verify the liver-targeting specificity of GalNAc-modified LNP, mice were sacrificed on day 7 after LNP injection. In addition to the liver, heart, lungs, spleen, kidneys, brain, and muscle tissue were also collected, and genomic DNA was extracted for PCR amplification and sequencing analysis.

[0148] The results showed that the indel frequency in the liver was 16.8±3.2%, while the indel frequencies in other tissues were all <1% (heart 0.3±0.1%, lung 0.5±0.2%, spleen 0.8±0.3%, kidney 0.4±0.1%, brain <0.1%, muscle 0.2±0.1%), confirming that GalNAc-modified LNPs have high liver-targeting specificity, with liver editing efficiency approximately 20-50 times higher than that in other tissues. This targeting specificity is attributed to the specific binding of GalNAc ligands to the highly expressed desialylate glycoprotein receptor (ASGPR) on the surface of hepatocytes, mediating preferential uptake of LNPs by hepatocytes.

[0149] 4. Off-target effect analysis in vivo

[0150] Using the same method as in Example 8, the top 20 potential off-target sites in edited mouse liver tissue were analyzed in vivo. The results showed that the average indel frequency of the 20 off-target sites in the HyperFi-Cas9 LNP group was 0.018 ± 0.005%, while it was 0.52 ± 0.12% in the wild-type SpCas9 LNP group, representing a 29-fold reduction in the off-target effect of HyperFi-Cas9. The in vivo off-target reduction was slightly lower than that in the in vitro cell experiment (54-fold), possibly due to the more complex physiological environment and different chromatin states in vivo, but it was still significantly better than that of wild-type SpCas9.

[0151] 5. Security Assessment

[0152] On day 7 after LNP injection, whole blood was collected from mice for hematological analysis (white blood cell count, red blood cell count, platelet count, hemoglobin concentration, etc.) and blood biochemical analysis (alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine, blood urea nitrogen, etc.) to assess the effects of LNP and CRISPR editing on liver and kidney function. Simultaneously, hematoxylin-eosin (H&E) staining histological analysis of major organs (liver, spleen, kidney, heart, and lungs) was performed to assess histopathological changes.

[0153] The results showed that there were no significant differences in hematological and biochemical indicators between the HyperFi-Cas9 LNP treatment group and the PBS control group (P>0.05, one-way ANOVA). ALT and AST levels were only slightly elevated (approximately 1.2-1.3 times, within the normal physiological range), indicating that LNP delivery and CRISPR editing did not cause significant liver damage. H&E staining showed that the morphology of all tissues in the LNP treatment group was normal, with no obvious inflammatory infiltration, cell necrosis, or fibrosis. These results demonstrate that the LNP delivery system and HyperFi-Cas9 editing system of this invention have good safety and tolerability in vivo.

[0154] Comparison of Comparative Example 1 and Method CN119913189A

[0155] To objectively evaluate the technical advantages of this invention over CN119913189A, the following comparative experiment was designed.

[0156] Experimental Design: A publicly available method was employed, involving the construction of gRNA and wild-type SpCas9 expression plasmids. A double-stranded oligonucleotide donor template (40-base-pair homologous arms containing synonymous mutations) was designed. HEK293T cells were co-transfected using electroporation to introduce the target mutation and synonymous mutation in the first round of editing. After screening for positive clones, a second round of editing was performed to eliminate the synonymous mutation. The editing target was also selected as the c.20A>T site of the HBB gene.

[0157] Results comparison:

[0158] Editing efficiency: The first round of editing efficiency in HEK293T cells was approximately 82%, and the second round of editing efficiency was approximately 75%, with an overall success rate of approximately 62% in obtaining homozygous repaired clones. The method of this invention achieved 32.8% efficiency in a single step (48 hours, HDR efficiency) and 38.7% efficiency in 72 hours. Although the efficiency of a single step is slightly lower than the sum of the two rounds of iterations, considering that this invention does not require clone screening and a second round of editing, it is more convenient and efficient in practical applications.

[0159] Editing cycle: The previous method required the first round of editing (2-3 days), clone screening and culture (3-4 days), elimination of the first round of plasmids by auxiliary plasmids (1 day), the second round of editing (2-3 days), and clone verification (1-2 days), with a total cycle of about 9-13 days; the method of this invention can complete the editing and pass the flow cytometry evaluation within 48-72 hours after LNP delivery, without the need for clone screening, with a total cycle of 2-3 days, shortening the cycle by about 75-85%.

[0160] Off-target effects: Using wild-type SpCas9, the off-target risk was not quantitatively assessed; the off-target effects of the HyperFi-Cas9 of this invention are reduced by approximately 54-fold (cellular level) and 29-fold (in vivo), significantly improving safety.

[0161] Real-time monitoring: Lacking a real-time monitoring mechanism, evaluation can only be performed post-sequencing; the dual fluorescence system of this invention can be rapidly evaluated by flow cytometry within 48 hours, and the FRET probe enables dynamic monitoring.

[0162] Delivery efficiency: The plasmid electroporation transfection efficiency is about 20-40%, and it is cytotoxic; the LNP delivery efficiency of this invention reaches 30-60% (cell level) and 16% (in vivo liver), avoiding electroporation damage.

[0163] In summary, this invention is significantly superior in terms of editing cycle, off-target safety, real-time monitoring capability, and delivery efficiency, while maintaining comparable or higher editing efficiency, demonstrating significant technological advancement.

[0164] Comparative Example 2: Comparison of activity and specificity between wild-type SpCas9 and HyperFi-Cas9

[0165] Experimental design: Under the same conditions (LNP delivery, same gRNA and donor, same HDR enhancement treatment), compare the targeting activity and off-target effects of wild-type SpCas9 and HyperFi-Cas9.

[0166] In vitro cleavage activity: In vitro cleavage experiments (Example 4) showed that wild-type SpCas9 and HyperFi-Cas9 had comparable cleavage efficiency on the target DNA (both >90%), demonstrating that the four mutations did not impair targeting activity.

[0167] Intracellular editing efficiency: Flow cytometry analysis (Example 8) showed that the FAM of the wild-type SpCas9 LNP group and the HyperFi-Cas9 LNP group... + The cell proportions (reflecting Cas9 cleavage activity) were 44.8% and 45.3%, respectively, and the HDR efficiencies were 31.5% and 32.8%, respectively. The differences were not statistically significant (P>0.05, t-test), demonstrating that HyperFi-Cas9 retained ≥85% of wild-type activity, consistent with the design target.

[0168] Off-target effect comparison: Targeted sequencing analysis (Example 8) showed that the average off-target rate of wild-type SpCas9 was 0.68%, while that of HyperFi-Cas9 was 0.0125%, a reduction of approximately 54-fold. In in vivo experiments (Example 9), the off-target rate of wild-type SpCas9 was 0.52%, while that of HyperFi-Cas9 was 0.018%, a reduction of approximately 29-fold. This result is comparable to the performance of high-fidelity Cas9 variants reported in the literature, validating the rationality of the HyperFi-Cas9 design.

[0169] Conclusion: HyperFi-Cas9 successfully achieves a balance between high targeting activity and high specificity. Compared with wild-type SpCas9, it significantly reduces off-target risk while maintaining comparable editing efficiency, meeting the safety requirements for clinical applications.

[0170] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of protection of the present invention.

Claims

1. A high-fidelity Cas9 variant, characterized by... The high-fidelity Cas9 variant is HyperFi-Cas9 obtained by introducing four point mutations R691A, N497A, R661A, and Q695A on the basis of Streptococcus pyogenes Cas9. The R691A mutation replaces arginine at position 691 with alanine, the N497A mutation replaces asparagine at position 497 with alanine, the R661A mutation replaces arginine at position 661 with alanine, and the Q695A mutation replaces glutamine at position 695 with alanine. The HyperFi-Cas9 maintains no less than 85% of the wild-type Streptococcus pyogenes Cas9 targeting activity while reducing the off-target effect to less than 0.025% of that of wild-type Streptococcus pyogenes Cas9.

2. The high-fidelity Cas9 variant according to claim 1, characterized in that... The HyperFi-Cas9 gene sequence has been codon-optimized to improve its expression efficiency in mammalian cells, with a codon fitness index of not less than 0.85 and a GC content of 50-60%.

3. The high-fidelity Cas9 variant according to claim 1, characterized in that... The HyperFi-Cas9 forms a ribonucleoprotein complex with a single guide RNA, which is 20 nucleotides in length, has a GC content of 40-60%, and has a PAM sequence adjacent to the target sequence of 5'-NGG-3'. The ribonucleoprotein complex is pre-assembled in vitro and then used for delivery.

4. A tissue-targeted lipid nanoparticle delivery system, characterized in that... The lipid nanoparticles comprise ionizable lipids, auxiliary lipids, cholesterol, and polyethylene glycol-modified lipids. The ionizable lipid is SM-102 or its structural analogue with a pKa value of 6.5-7.

0. The auxiliary lipid is 1,2-distearyl-sn-glycerol-3-phosphocholine. The polyethylene glycol-modified lipid is DSPE-PEG2000 or DSPE-PEG2000-ligand. The molar ratio of the lipid composition is ionizable lipid: auxiliary lipid: cholesterol: polyethylene glycol-modified lipid = 50:10:38.5:1.

5. The lipid nanoparticles have a particle size of 80-120 nm and an encapsulation efficiency greater than 85%.

5. The tissue-targeted lipid nanoparticle delivery system according to claim 4, characterized in that... The polyethylene glycol-modified lipids achieve tissue-specific targeting through ligand modification. The ligands are selected from N-acetylgalactosamine, anti-epidermal growth factor receptor antibody fragments, lung epithelial cell targeting peptides, or folic acid. N-acetylgalactosamine is used for liver targeting, anti-epidermal growth factor receptor antibody fragments are used for tumor targeting, lung epithelial cell targeting peptides are used for lung targeting, and folic acid is used for targeting tumors with high folic acid receptor expression.

6. The tissue-targeted lipid nanoparticle delivery system according to claim 4, characterized in that... The lipid nanoparticles simultaneously encapsulate a high-fidelity Cas9 variant and a ribonucleoprotein complex formed by a single guide RNA and a single-stranded DNA donor template. The single-stranded DNA donor template is 150-300 nucleotides in length and contains a left homologous arm and a right homologous arm that are homologous to the genomic target site on both sides. The left homologous arm and the right homologous arm are each 60-80 nucleotides in length. The target mutation sequence is carried at the desired mutation site. A CtIP recognition sequence is introduced into the left or right homologous arm to promote DNA end excision. The target gene downstream of the target site encodes a dual fluorescent reporter protein mCherry-P2A-EGFP.

7. A CRISPR-Cas9-mediated precise genome editing method, characterized by... This includes the following steps: Step 1: Prepare a high-fidelity Cas9 variant, HyperFi-Cas9. Obtain the codon-optimized HyperFi-Cas9 gene sequence through whole-genome synthesis, clone it into a prokaryotic expression vector, induce expression in E. coli, and purify it by nickel column affinity chromatography and ion exchange chromatography to obtain HyperFi-Cas9 protein with a purity greater than 95%. Step 2: Design and synthesize a single-guide RNA containing a 20-nucleotide targeting sequence, a Cas9 binding scaffold sequence, and a poly-U terminator. The single-guide RNA is synthesized by in vitro transcription, the DNA template is removed with DNase I, and high-purity single-guide RNA is obtained after purification. Step 3: Prepare the ribonucleoprotein complex by mixing HyperFi-Cas9 protein and single guide RNA in vitro at a molar ratio of 1:1.2 and incubating at room temperature for 15 minutes to form a stable ribonucleoprotein complex. Step 4: Synthesize a single-stranded DNA donor template, which is 150-300 nucleotides in length and contains 60-80 nucleotides each of homologous arms, a target mutation sequence, a CtIP recognition sequence, and a dual-fluorescent reporter gene sequence; Step 5: Prepare tissue-targeting lipid nanoparticles. Using microfluidic mixing technology, ionizable lipids, auxiliary lipids, cholesterol, and polyethylene glycol-modified lipids are mixed in a molar ratio of 50:10:38.5:1.

5. This mixture is then combined with a ribonucleoprotein complex and a single-stranded DNA donor template in a microfluidic chip to form tissue-targeting lipid nanoparticles with a particle size of 80-120 nm and an encapsulation efficiency greater than 85%. Step 6: Homologous recombination enhancement treatment. Target cells are treated with a DNA-dependent protein kinase catalytic subunit inhibitor for 2 hours or with a cell cycle-dependent kinase 1 inhibitor for 16-20 hours, followed by elution to enrich cells in the S / G2 phase. RAD51 expression is then enhanced by transient transfection with a RAD51 overexpression vector. Step 7, lipid nanoparticle delivery: Tissue-targeting lipid nanoparticles are added to cell culture medium enhanced with homologous recombination or administered via intravenous injection, intraperitoneal injection, or local injection; Step 8: Monitor the editing process in real time. At different time points after delivery, detect the FAM fluorescence signal released by the fluorescence resonance energy transfer molecular beacon probe by flow cytometry to evaluate the dynamic changes in Cas9 cleavage activity; Step 9: Evaluate editing efficiency and accuracy. 48-72 hours after delivery, the proportion of cells expressing mCherry and EGFP dual fluorescence is detected by flow cytometry to quantitatively evaluate the precise editing efficiency mediated by homologous recombination-directed repair. Genomic DNA is extracted, the target site is amplified by PCR, and sequencing is performed to verify the accurate introduction of the target mutation and off-target editing.

8. The method according to claim 7, characterized in that... The DNA-dependent protein kinase catalytic subunit inhibitor is NU7441, used at a concentration of 0.5-1 μmol; the cell cycle-dependent kinase 1 inhibitor is RO-3306, used at a concentration of 5-10 μmol; the homologous recombination enhancement treatment increases the ratio of homologous recombination-directed repair to non-homologous end ligation repair pathways from 1:10 to 3:1, and the homologous recombination-directed repair efficiency is 3-5 times higher than that of the control group.

9. The method according to claim 7, characterized in that... The fluorescent resonance energy transfer molecular beacon probe is a hairpin-structured single-stranded DNA with a 5' end labeled with the fluorescent group FAM and a 3' end labeled with the quencher group BHQ1. The probe sequence is complementary to the region adjacent to the Cas9 cleavage site of the target gene. When Cas9 produces a double-strand break at the target site, the probe is degraded, and the FAM fluorescence is released from the BHQ1 quenching. The change in fluorescence signal is detected in real time by flow cytometry with a time resolution of 5 minutes. In the dual fluorescent reporter protein mCherry-P2A-EGFP, P2A is a self-cleaving peptide sequence, ensuring equimolar expression of mCherry and EGFP. After successful editing, cells express red and green dual fluorescence, and the editing efficiency is quantitatively evaluated by flow cytometry.

10. The application of the high-fidelity Cas9 variant of claim 1, the tissue-targeted lipid nanoparticle delivery system of claim 4, or the method of claim 7 in the preparation of drugs for treating genetic diseases, the preparation of drugs for tumor immunotherapy, or crop improvement, wherein the genetic diseases include sickle cell anemia, β-thalassemia, cystic fibrosis, Duchenne muscular dystrophy, Huntington's disease, or spinal muscular atrophy; wherein the tumor immunotherapy includes knocking in chimeric antigen receptor genes into T cells to prepare CAR-T cells or knocking out immune checkpoint genes to enhance the anti-tumor activity of T cells; and wherein the crop improvement includes introducing disease-resistant genes to improve resistance to pests and diseases, or improving nutritional components to increase vitamin or mineral content, or optimizing yield traits to improve photosynthetic efficiency, or increasing grain yield.