Preparation method and application of optimized lentivirus-like particles

Abstract

The invention discloses a preparation method and application of optimized lentivirus-like particles, belongs to the technical field of gene engineering, and aims to solve the problem that delivery is difficult in an mRNA form of a base editor at present. Virus skeleton protein adopted by a designed delivery system is HIV GAG, elements such as integrase and reverse transcriptase in an HIVpol gene are removed, and therefore the delivery efficiency of the lentivirus-like particles is improved. An expression base editor gene is introduced into a Psi element, and an HDVrz-Psi nucleic acid element combination in an sgRNA expression box is introduced, so that the wrapping efficiency of base editor mRNA and sgRNA is improved. The designed delivery system can efficiently deliver the base editor mRNA only by relying on HIV-GAG capsid protein (without HIV-Pol), the optimized lentivirus-like system removes reverse transcriptase, integrase and other unsafe factors, and the risk of gene integration is removed from the source.

Description

Technical Field

[0001] The present invention belongs to the technical field of genetic engineering, and in particular relates to an optimized preparation method and application of lentivirus-like particles. Background Art

[0002] Gene editing technologies such as CRISPR-Cas9 herald a new era of precise genome modification. These approaches not only hold promise for correcting genetic mutations that underlie a variety of diseases, but they also present challenges that require in-depth and comprehensive research and innovation. CRISPR-Cas9 has revolutionized the potential of genome editing, but the technology typically requires inducing double-stranded DNA breaks (DSBs), which can lead to unintended genomic alterations such as insertions, deletions, and complex gene rearrangements. In contrast, base editors (BEs), including cytosine base editors (CBEs) and adenine base editors (ABEs), can achieve targeted conversion of single nucleotides without inducing DSBs, thereby significantly mitigating these adverse effects. These advanced base editors have been used to ameliorate disease-associated phenotypes in a variety of cells, including humans and non-human primates, by correcting pathogenic mutations, highlighting the potential of base editing as a therapeutic approach.

[0003] Despite these encouraging results, the widespread application of base editing still depends on the development of safe and effective delivery mechanisms. Traditional delivery vectors, such as lentiviral (LV) and adeno-associated viral (AAV) systems, provide a mode of efficient delivery of gene editing components into cells. However, these viral vectors face several limitations. The limited packaging capacity of AAV vectors, which can accommodate approximately 5 kb of DNA, is a key factor to consider in gene delivery applications. This limitation significantly narrows the applicability of AAV for the delivery of gene editing tools, especially since the size of most BEs and prime editors (PEs), SpCas9, exceeds the capacity of a single AAV vector. In addition, the potential risk of long-term exposure to active CRISPR components and the risk of unwanted genomic integration may lead to cytotoxicity and carcinogenicity.

[0004] Recent exploration of non-viral delivery systems, including virus-like particles (VLPs), have shown promise in circumventing the risks associated with viral vectors, providing a more transient but highly efficient means of delivering gene-editing tools. VLPs are non-pathogenic complexes of viral proteins that are used to package and transport specific cargoes, such as mRNAs, proteins, or ribonucleoproteins (RNPs), potentially replacing viral genetic content. These constructs are derived from existing viral frameworks and exploit intrinsic viral properties to facilitate efficient intracellular trafficking. These properties include the ability to encapsulate a variety of cargoes, escape endosomes, and target specific cell types. In contrast to intact viruses, VLPs facilitate transient delivery of gene-editing tools in the form of mRNA or protein rather than DNA, significantly reducing the potential for unintended genetic modifications and integration into the host genome. Due to these attributes, VLPs combine the advantages of both viral and non-viral vector systems and represent a powerful paradigm for gene-editing applications. Recent studies have reported that VLPs can efficiently package gene-editing components, including the CRISPR / Cas9 system. To ensure that VLPs package gene-editing agents, multiple platforms have been developed that can efficiently incorporate CRISPR-Cas9 RNPs. For example, recent studies have fused Cas9 and retroviral gag to create retrovirus-like particles (RVLPs), such as γ-retroviral murine leukemia virus (MLV) (gRVLPs and “enhanced” gRVLPs), as well as LVLP and its variants. Galla et al. designed MLV-derived VLP particles to encapsulate CRISPR / Cas9 RNA in an integrated format for precise gene knockout applications. They successfully promoted the co-delivery of non-viral SpCas9 mRNA and sgRNA transcripts into target cells through the MS2 phage packaging mechanism.

[0005] However, most VLP systems for delivering gene editing tools mainly use HIV-1's Gag and Pol proteins. For example, Philippe et al. designed a VLP called "Nanoblades", which is composed of HIV-Gag and Pol proteins. They fused Cas9 with HIV-Gag protein and used the reverse transcriptase and integrase activities of Gag-Pol protein to promote the efficient encapsulation of Cas9 RNP (Ribonucleoprotein) in VLP, which can achieve efficient gene editing in primary cells and animal models. However, the reverse transcriptase and integrase functions in Pol protein lead to genomic instability and tumorigenic risks. Integrase can randomly insert exogenous genes into the host genome, which may cause the activation of proto-oncogenes or the inactivation of tumor suppressor genes, leading to abnormal cell proliferation. In order to reduce these risks, researchers have recently designed an integrase-deficient lentiviral vector (IDLV) to reduce the risk of integration by mutating the integrase coding sequence and weakening the function of integrase. However, when using IDLV for gene delivery, incompletely inactivated integrases may still exist, and the potential risk of insertion mutations cannot be ignored. Studies have found that even if mutated integrases are used, IDLV vectors still have the risk of genomic integration, which mainly comes from two pathways. One is that the mutant integrase retains a certain catalytic activity and can mediate the integration of exogenous DNA into the host genome; the other is that abnormal recombination in the form of linear and circular vector genomes is integrated into the host genome through NHEJ or circular DNA integration pathways. In addition, the activity of reverse transcriptase can trigger immunogenic epitopes, leading to adverse immune responses. Therefore, when delivering gene editing tools with VLPs, retaining HIV-Pol protein may bring about a non-negligible risk of genomic integration.

[0006] In order to package specific mRNAs within VLPs, a molecular mechanism is required that enables the viral capsid protein to recognize and guide the target mRNA into the particle. Retroviral RNA genomes have a lentiviral packaging signal (Psi) element that promotes the packaging of viral RNA into viral particles. Therefore, engineered mRNA containing Psi should be able to be packaged into viral particles. For example, early studies used Psi to package Cre recombinase mRNA into MLV particles, thereby achieving transient mRNA delivery. Prel et al. used the interaction between MCP and MS2 aptamers to guide mRNA into modified VLPs. In addition, VLPs derived from bacteriophage PP7 can be designed to package target mRNAs and deliver them intracellularly by relying on the self-assembly properties of PP7 capsid protein. In addition, the mechanism of encapsulation of guide sgRNA in virus-like particles reported so far is random encapsulation, which requires high abundance expression of sgRNA in packaging cells, and random packaging of sgRNA into virus-like particles during assembly. The disadvantage of this is that the number of copies of sgRNA in virus-like particles is small, which affects the effect of later gene editing.

[0007] The vast majority of LVLPs are designed for CRISPR-Cas9 system delivery, and fewer LVLPs are used for base editors. In view of the above, this study aims to develop a safer and more efficient lentivirus-like particles (LVLP) system for the delivery of gene editing tools, especially base editors. Our previous work developed an adenine base editor CE8e-SpRY with almost no PAM restrictions, which can edit almost any genomic site without compromising gene editing activity, in order to verify the delivery and gene editing effects of the developed delivery system. Summary of the invention

[0008] The present invention develops an optimized method for preparing lentivirus-like particles and its application. The lentivirus-like particle delivery system can efficiently deliver base editor mRNA by relying solely on HIV-GAG capsid protein (without HIV-Pol). The optimized lentivirus-like system eliminates unsafe factors such as reverse transcriptase and integrase, and eliminates the risk of gene integration from the root. In terms of delivery strategy, the base editor mRNA is packaged into virus-like particles using the packaging signal Psi of the lentivirus itself, and the sgRNA is directionally packaged into virus-like particles using the combination strategy of Psi and HDVrz nucleic acid elements, which can significantly improve the loading capacity and base editing efficiency of base editor mRNA and sgRNA in LVLP.

[0009] In the first aspect, the present invention provides an optimized lentivirus-like particle delivery system, which delivery system includes a virus-like particle backbone protein, a specific membrane protein, a base editor mRNA and a sgRNA, wherein the virus-like particle backbone protein is HIV GAG protein, the base editor mRNA is packaged into the virus-like particles using the packaging signal Psi of the lentivirus itself, and the sgRNA is directionally packaged into the virus-like particles using a combination strategy of Psi and HDVrz nucleic acid elements.

[0010] In a second aspect, the present invention provides a method for preparing an optimized lentivirus-like particle, which is prepared using the above-mentioned lentivirus-like particle delivery system, and the specific process is as follows:

[0011] Step 1: Inoculate 293T cells into a culture flask and culture for 20-24 hours before transfection;

[0012] Step 2: After transfecting the packaging cell line 293T with the plasmid packaging system for 16-18 hours, wash with PBS to remove the remaining plasmid, and then replace with fresh culture medium; collect the supernatant 48h and 72h after transfection, and centrifuge to remove cell debris;

[0013] Step 3: Mix the supernatant containing lentivirus-like particles with Lenti-X concentrate, incubate overnight at 4°C, centrifuge after incubation, resuspend the precipitated lentivirus-like particles with cold PBS, and aliquot and store the purified lentivirus-like particles.

[0014] In one technical solution, the plasmid packaging system includes a specific membrane protein expression plasmid, a packaging plasmid for a virus-like particle backbone protein, a base editor mRNA plasmid, and an sgRNA expression plasmid, with a total amount of 45 μg. Among them, the specific membrane protein expression plasmid is pVSV-G expressing the vesicular stomatitis virus G protein; the packaging plasmid for the virus-like particle backbone protein is pHIV-gag; the base editor mRNA plasmid is pCE8e-SpRY-Psi; and the sgRNA expression plasmid contains the HDVrz and Psi dual nucleic acid elements.

[0015] In some technical solutions, the plasmids of the plasmid packaging system are as follows:

[0016]

[0017]

[0018] In one technical solution, the CE8e-SpRY base editor was used as the delivery object, the PD1 gene was selected as the target gene, a series of sgRNA sequences targeting PD1 were designed, and one sgRNA with good gene editing effect was screened and obtained: PD1-sg7 (see Table 2), which was used for the optimization and preparation of lentivirus-like particles.

[0019] In a third aspect, the present invention also provides an optimized lentivirus-like particle, which is prepared by a method for preparing lentivirus-like particles for packaging base editors.

[0020] Fourthly, the application of an optimized lentivirus-like particle delivery system in targeted gene editing related to tumors and genetic diseases.

[0021] In some technical solutions, the targeted genes include PD1, B2M, CIITA, TIGIT and TRAC. The sequences of sgRNAs screened for PD1, B2M, CIITA, TIGIT and TRAC are shown in the following table:

[0022] Target Gene sgRNA forward sequence information sgRNA reverse sequence information PD1 ctccaggcatgcagatccca tgggatctgcatgcctggag B2M ccgagatgtctcgctccgtg cacggagcgagacatctcgg CIITA ggatgctgcatgctgggtga tcacccagcatgcagcatcc TIGIT cagaagcatgcgctggtgtc gacaccagcgcatgcttctg TRAC tctggatatctgtgggacaa ttgtcccacagatatccaga .

[0023] In one technical solution, the indicator cell line 293T-PD1-Fluc is used to evaluate the targeted PD1 gene editing effect. The indicator cell line 293T-PD1-Fluc can stably express the fusion gene of the PD1 coding region CDS and the firefly luciferase gene. The luciferase value is detected by adding luciferase substrate to reflect the editing efficiency of the targeted PD1 gene.

[0024] Fifthly, the application of an optimized lentivirus-like particle delivery system in the delivery of Crispr / Cas9 gene editing tools.

[0025] Compared with the prior art, the present invention has the following advantages:

[0026] 1. The lentivirus-like particle delivery system developed by the present invention can efficiently deliver base editor mRNA by relying only on HIV-GAG capsid protein (without HIV-Pol). The optimized lentivirus-like system eliminates unsafe factors such as reverse transcriptase and integrase, thus eliminating the risk of gene integration from the root. The existing system contains HIV-GAG and Pol proteins.

[0027] 2. The present invention develops an optimized Psi packaging strategy for the preparation of lentiviral particles, which significantly enhances the packaging and base editing efficiency of mRNA and sgRNA.

[0028] 3. The Psi-LVLP developed by the present invention can efficiently target and edit genes related to immunotherapy response, highlighting its potential for clinical application. It shows that Psi-LVLP is a promising tool for CAR-T cell therapy and other immunotherapy strategies, providing precise and efficient genetic modification with the advantage of reducing off-target effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Several packaging schemes for mRNA generation LVLP. A: Schematic diagram of mRNA generation by LVLP, including MS2-MCP, PP7-PP binding protein, psi-GAG, Psi+pSPAX2; B: Schematic diagram of sgRNA and FX-sgRNA designed to target PD1 CDS A site. CD: A-to-G editing efficiency of CE-8e-SpRY plasmid for (C) 8 sgRNAs and (D) 8 reverse sgRNAs in HEK293T cells after FACS analysis. E: Quantification curve of p24 reagent standard determined by ELISA. F: p24 concentration of LVLP generated by several packaging strategies (corresponding to A).

[0030] Figure 2 Development of PD1-293T reporter cell line. (A) Schematic diagram of the universal third-generation lentiviral packaging process. (B) Schematic diagram of the dual reporter PD1 plasmid (LTR: long terminal repeat; RRE: Rev response element; RFP: red fluorescent protein). (C) PD1 mRNA expression levels in WT-293T cells and PD1-293T reporter cell line. (D) Firefly fluorescence values ​​of WT-293T cells and PD1-293T reporter cell line. (E): Image analysis of RFP in PD1-293T reporter cell line. (F) PD1 protein expression in WT-293T cells and PD1-293T reporter cell line.

[0031] Figure 3 High packaging yield of the Psi packaging solution. (A) Amplification curves of CE-8e-SpRY primers and probes at the plasmid level. (B) Standard curves for CE-8e-SpRY mRNA copy number quantification at different concentrations. (C) CE-8e-SpRY copy number of several LVLP packaging solutions. (D) Percentage of firefly fluorescence of PD1-293T cell lines transfected with different LVLPs compared to the control PD1-293T cell line (normalized using an equal amount of control PD1-293T cell line). (E) A-to-G editing efficiency of several LVLP solutions.

[0032] Figure 4To optimize the plasmid ratio of LVLP. (A) Schematic diagram of LVLP. (B) p24 content of LVLP with different plasmid ratios. (C) mRNA copy number of LVLP with different plasmid ratios. (E) A-to-G editing efficiency of LVLP with different plasmid ratios. (D) Firefly fluorescence percentage of PD1-293T reporter cell line transfected with LVLP with different plasmid ratios. (F) Images of psi-LVLP observed using cryo-EM.

[0033] Figure 5 Psi-LVLPs exhibited strong mRNA packaging potential. (A) Schematic design of the HDVrz-psi-sgRNA plasmid: HDVrz was located at the C-terminus of the PD1 sgRNA sequence to release the sgRNA, while psi was located at the 3' end of HDVrz to facilitate the entry of the sgRNA into the LVLPs. (B) A-to-G editing efficiency of different plasmid variants in 293T cells without screening. (C) p24 concentration of LVLPs generated by WT-sgRNA and HDVrz-psi-sgRNA. (D) sgRNA copy number of LVLPs generated by WT-sgRNA or HDVrz-psi-sgRNA. (E) Firefly fluorescence percentage of LVLPs generated by WT-sgRNA or HDVrz-psi-sgRNA after transfection of PD1-293T reporter cell line. (F) A-to-G editing efficiency of LVLPs generated by WT-sgRNA or HDVrz-psi-sgRNA. (G) PD1 protein expression after transfection of the PD1-293T reporter cell line with LVLPs generated with WT-sgRNA or HDVrz-psi-sgRNA.

[0034] Figure 6 The broad application of Psi-LVLP in several genes related to immunotherapy. (A) A-to-G editing efficiency of Psi-LVLP in Jurkat cells. (B) A-to-G editing efficiency of Jurkat cells transfected with different LVLPs. (C) Schematic diagram of HDVrz-psi-sgRNA targeting B2M, CIITA, TRAC, and TIGIT. (D): A-to-G editing efficiency in 293T cells transfected with different plasmid variants. (E) A-to-G editing efficiency of LVLP targeting B2M, CIITA, TRAC, and TIGIT.

[0035] Figure 7Causal relationship of immunotherapy-related genes in cancer. (A) Cancer types as outcome factors in MR studies. (B) SNP bar graph derived from breast cancer. (C) Breast cancer-specific SNPs distribution ring graph within chromosomes. (D) MR results of significant SNPs matched with cancer-related genes. (E) MR results of significant methylation sites of B2M, PD1, and TIGIT in cancer.

[0036] Figure 8 eQTL-GWAS co-localization analysis. (A) MR results of CIITA significant methylation sites in cancer. (B) eQTL-GWAS results of B2M, CIITA, PD1, and TIGIT significant SNPs in cancer. (C) A representative regional Manhattan plot of major SNPs (TIGIT-prostate cancer). The tagged SNP is the major SNP (in this case for both studies), and other SNPs are tagged based on their linkage disequilibrium with the major SNP.

[0037] Fig. 9 A-to-G or T-to-C (FX-sgRNA) editing efficiency of sgRNAs targeting B2M, CIITA, FX-CIITA, TRAC, FX-TRAC, TIGIT, and FX-TIGIT in HEK293T cells.

[0038] Fig.10 Mass spectra of the plasmid packaging system: (A) pVSV-G; (B) pHIV-gag; (C) pCE8e-SpRY-Psi; (D) pU6-Hdvrz-Psi-sgRNA. DETAILED DESCRIPTION

[0039] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0040] Lentivirus-like particles (LVLP); MS2 coating protein (MCP); lentivirus-packing signal (Psi); red fluorescent protein (RFP); Mendelian Randomization (MR); double-stranded DNA breaks (DSBs); base editing (BE); cytosine base editors (CBEs); adenine base editors (ABEs); lentivirus (LV); adeno-associated virus (AAV)

[0041] 1. Materials and Methods

[0042] 1.1 Cell culture

[0043] Lenti-X 293T cell line (Taraka), HEK 293T cell line (ATCC) and PD1-293T cell line were cultured in DMEM (Gibco) containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (Gibco). Human Jurkat cell line was maintained in RPMI 1640 medium (Gibco). All cells were cultured at 37°C in an environment of 5% CO2.

[0044] 1.2LVLP generation

[0045] LVLPs were produced in T75 flasks by transient transfection of Lenti-X 293T cells.

[0046] 20-24 hours before transfection, cells were plated at 1.5 × 10 ^7 The cells were plated at a density of 10 cells / flask in T75 culture flasks.

[0047] When the total amount of plasmid remains unchanged, a plasmid packaging system is selected for plasmid transfection. The plasmid packaging system and its dosage are shown in Table 1.

[0048] 16-18h after plasmid transfection, wash with PBS to remove residual plasmid and then replace with fresh culture medium; collect supernatant 48h and 72h after plasmid transfection, centrifuge at 4000g for 30 min at 4℃, or filter through a 0.45μm filter to remove cell debris.

[0049] According to the instructions, the supernatant containing LVLP was mixed with Lenti-X concentrate (Taraka) and incubated overnight at 4°C. After incubation, the supernatant containing LVLP was centrifuged at 1500g for 50 minutes at 4°C, and the precipitated LVLP particles were resuspended with 1% or 1‰ volume of cold PBS. The purified LVLP was aliquoted and frozen at -80°C until use.

[0050] Table 1 Plasmid packaging system configuration (each value represents the amount of plasmid required for each T75 culture bottle (μg))

[0051]

[0052] 1.2 Plasmid transfection

[0053] The plasmid was extracted using the EndoFree Plasmid Midi Kit (CW2105, Cwbio), and the concentration was detected using Nanodrop. HEK293T cells were seeded at a density of 50-70% in 24-well plates for transfection. After 20-24 hours, the cells were transfected with 2μl LipoMax and 0.8μg plasmid DNA. Unless otherwise stated, cells were transfected with 0.53μg CE8e-SpRY and 0.27μg sgRNA plasmid per well after incubation at room temperature for 25 minutes. The culture medium was replaced 4-6 hours after transfection. 48-72 hours after transfection, 20,000 eGFP-positive cells were collected by BD Aria II flow cytometer for subsequent base editing analysis.

[0054] LVLP infection

[0055] Before infection, 20,000 HEK293T cells or PD1-293T reporter cells were seeded in a 24-well plate. Then, 20 ng of LVLP (determined by ELISA) was added to the target sample. For luciferase assays, 2,000 PD1-293T reporter cells were seeded in a 96-well plate and transfected with 5 ng of LVLP (determined by ELISA). After 24 hours, the cell culture medium was replaced and cell samples were collected 48-72 hours later. Base editing analysis or Western blot analysis was performed after cell lysis.

[0056] The plasmid maps of pVSV-G, pHIV-gag, pCE8e-SpRY-Psi and pU6-Hdvrz-Psi-sgRNA are as follows: Fig.10 shown.

[0057] 2. Analysis

[0058] 2.1ELISA

[0059] LVLP concentration was determined using a p24 ELISA kit (JL19101, Jonln). LVLP samples were diluted from 1:100 to 1:1000 with universal diluent, and then ELISA was performed according to the manufacturer's instructions. We used a series of standard p24 reagents to establish a quantitative curve. 100 μL of sample or standard reagents at different concentrations were added to the designated wells. Then the biotinylated antibody working solution was added and incubated at 37°C for 1 hour. Next, the liquid was discarded and 300 μL of working wash solution was used to wash three times. Then 100 μL of enzyme conjugate working solution was added to each well and incubated for 30 minutes. After washing five times, 90 μL of substrate solution (TMB) was added to each well and incubated for 15 minutes. Finally, 50 μL of stop solution was added to each well, and the OD value of each sample was tested at a wavelength of 450 nm using SpectraMax iD3 (Molecular Devices). The p24 content of LVLP was determined by comparing with the dilution series of the p24 standard curve.

[0060] 2.2 RNA extraction

[0061] RNA was extracted from concentrated LVLPs using a DOF-9648 purifier (GenMagBio, China) using an RNA isolation kit (GMB-V-F4, GenMagBio). RNA was isolated directly from 140 μL of the pellet-containing supernatant using the QIAamp ViralRNA Mini Kit (QIAGEN). RNA was isolated from cells using Trizol reagent.

[0062] 2.3 In vitro transcription experiments

[0063] To accurately measure the mRNA content of LVLP, we first performed an in vitro transcription experiment (IVT) using HyperScribe TM All in One mRNA Synthesis Kit Plus 1 generated CE8e-SpRYmRNA. In brief, we generated CE8e-SpRYDNA template by PCR and removed the original plasmid by gel extraction. Then the transcription reaction was performed, and the original DNA template was removed using DNase I. Next, the tailing reaction was performed using E-PAP Buffer according to the instructions. Finally, the RNA transcript was purified by phenol-chloroform extraction and ethanol precipitation strategy. The RNA concentration was measured using UV spectrophotometry at a wavelength of 260nm. We designed primers and probes for CE8e-SpRY and verified the specificity at the plasmid level. Then One Step PrimeScript TMRT-PCR Kit (Perfect Real Time) (RR064A, Taraka) was used for RT-qPCR detection, using different concentrations of CE8e-SPRY mRNA as template. The mRNA copy number was calculated by the following formula: RNA copy number = amount (ng) × 6.022 × 10 ^23 / (length (bp)×1×10 ^9 ×330). The established CE8e-SpRY standard copy number curve can be used to quantify the mRNA content in LVLP.

[0064] 2.4ddPCR quantification of sgRNA abundance

[0065] First, the amplification specificity and efficiency of the primers and probes of sgRNA were verified at the plasmid level. Total RNA in LVLP was extracted using an RNA isolation kit (GMB-V-F4, GenMagBio) using a DOF-9648 purifier (GenMagBio, China). DNase I (Beyotime) was used to eliminate any remaining plasmid DNA. Quantitative droplet digital PCR (ddPCR) was performed using the Sniper DQ24 DigitalPCR system, using One Step RT-dPCR Master Mix for Probes (Sniper).

[0066] 2.5 Western blot

[0067] For viral proteins from lentivirus and LVLP, 100 ng of purified lentivirus or LVLP (determined by p24 ELISA) was separated using RIPA buffer and then boiled in loading buffer for 3-5 minutes. Proteins were separated on SDS-PAGE gels and transferred to PVDF membranes. Antibodies used included PD1 (NAT105, ab52587, Abcam, 1:50), β-actin antibody for positive control (AA128, Beyotime, 1:1000). To detect the expression of PD1 protein, PD1-293T reporter cells were transfected with 100 ng of CE8e-SPRY mRNA-encapsulated LVLP or lentivirus. After washing the membrane, the corresponding secondary antibody was mixed with the membrane. Protein signals were visualized using chemiluminescent reagents (Pierce).

[0068] 2.6 Analysis of base editing efficiency

[0069] For cell samples sorted by FACS, 20,000 cells were typically collected and treated with cell lysis buffer (50 mM KCl, 1.5 mM MgCl2, 10 mM Tris pH 8.0, 0.5% Nonidet P-40, 0.5% Tween20, 100 μg / mL proteinase K) under the following conditions: 65°C for 30 minutes, 98°C for 3 minutes. For cells transfected with LVLP, the lysate was used as a DNA template for PCR. Primers are listed in the supplementary materials. PCR products were used for Sanger sequencing, and the A-to-G base editing efficiency was evaluated by the EditR online tool.

[0070] 2.7 Luciferase Assay

[0071] The Bright-Lite Luciferase Assay System (DD1204, Vazyme) is an ultra-sensitive, stable, and uniform reagent designed for the detection of luciferase reporter genes. The assay system contains highly purified luciferin and optimized reaction reagents, which enhance the stability of the reaction and improve environmental tolerance. By adding the mixed Bright-lite assay reagent directly to the cell culture, the cells are lysed, releasing the luciferase and generating a stable light signal. 2000 PD1-293T reporter cells were seeded in a 96-well plate. After 24 hours, the cells were transfected with 5ng LVLP (determined by ELISA), and the cell culture medium was replaced 24 hours after transfection. 48-72 hours after transfection, the cells were mixed with an equal volume of room temperature equilibrated bright-lite assay reagent buffer. The mixture was incubated for at least 2 minutes to ensure complete cell lysis. The luciferase value of each sample was then tested by GloMax (Promega).

[0072] 2.8 Transmission electron microscopy imaging

[0073] In the negative staining process, the LVLP supernatant was concentrated using Lenti-X Concentrator (Taraka). The carbon grid was then immersed in 20 μL of the LVLP sample, after which the particles were stained with phosphotungstic acid. The sample was then dried and examined using a FEI Tecnai 12 electron microscope.

[0074] 2.9 Mendelian randomization analysis

[0075] MR analysis was used to investigate the causal relationship between genetic variants and methylation sites of PD1 (ENSG00000188389), B2M (ENSG00000166710), CIITA (ENSG00000179583), TRAC (ENSG00000277734), and TIGIT (ENSG00000181847) genes and various cancers. We used expression quantitative trait loci (eQTL) and methylation data from https: / / gwas.mrcieu.ac.uk / datasets / as exposure data. Outcome data were also extracted from GWAS datasets. Instrumental variables (IVs) for exposure were selected based on genome-wide significance threshold (p-value < 1e-06). Exposure data were harmonized with outcome data to align effect alleles. To avoid ambiguity, palindromic SNPs with intermediate allele frequencies were excluded. SNPs were further filtered based on F statistics to ensure that they were strong instruments (F > 10). Heterogeneity and pleiotropy between SNPs were assessed by Cochran’s Q test and MR-Egger intercept test. Leave-one-out analysis confirmed that the results were not driven by any single SNP. For MR sensitivity analysis, MR-PRESSO (pleiotropy residuals and outliers) was used to reduce bias caused by correlated levels of pleiotropy, using the TwosampleMR package. Bayesian colocalization analysis was used to assess the likelihood that two traits share the same causal variant, using the default parameters of the coloc package (https: / / github.com / chr1swallace / coloc) according to previous studies. We defined PP.H4.abf>80% as the threshold for colocalization evidence. Visualization was performed using the locuscompare package.

[0076] 2.10 Statistical analysis

[0077] Data are presented as mean ± standard deviation (SD). Unpaired t-test and ANOVA were used to determine significant differences between two and multiple groups. All statistical analyses were performed using GraphPad Prism v8. Significant P values ​​(unpaired t-test and ANOVA) were marked as *P<0.05, **P<0.01, ***P<0.005. All data are presented as triplicate ± SD. MR analyses were performed in Project R.

[0078] 3. Results

[0079] 3.1 Establishment of various LVLPmRNA packaging schemes

[0080] Given the previously mentioned interactions between MS2 and MCP, PP7 and PP7 capsid protein, Psi and HIV-Gag (17,19,21,24,25), we have developed several mRNA packaging schemes ( Figure 1 A). We determined the A-to-G editing efficiency of CE8e-SpRY. Then we designed sgRNAs targeting the A site or the inverted T site (called FX-sgRNA) of the PD1 translation start codon (ATG). Considering the efficient working window (8 bps) of CE8e-SpRY, we designed 8 sgRNAs and 8 FX-sgRNAs from the 1st high activity site to the 8th high activity site ( Figure 1 B). By FACS, sg7 was found to have high editing activity and was selected for subsequent LVLP generation. We tested the packaging activity of several packaging schemes by ELISA, and the positive control (replacing the gag plasmid with pSPAX2, Figure 1 A)( Figure 1 E). The Psi solution showed a higher LVLP generation capacity ( Figure 1 F).

[0081] 3.2 Establishment of PD1-293T reporter cell line

[0082] To evaluate the efficiency of sgRNA knockdown of PD1 expression in LVLP, we established a 293T cell line stably expressing PD1 protein using a third-generation lentiviral packaging system ( Figure 2 A). We created a dual reporter system containing the luciferase and RFP genes. The luciferase reporter gene was cloned at the 3' end of the PD1 CDS region and its expression can be detected by evaluating luciferase activity. RFP was cloned at the 3' end of the luciferase and was used to inhibit the integration of the PD1 gene into the genome ( Figure 2 B). After puromycin selection, we verified the high firefly luciferase activity of the PD1-293T cell line ( Figure 2 C). qRT-PCR assay showed the expression of PD1 in PD1-293T cell line ( Figure 2 D). Image analysis shows the presence of RFP in the PD1-293T cell line ( Figure 2 E). In addition, WB verified the expression of PD1-luciferase fusion protein in PD1-293T cell line ( Figure 2 F). These data indicate that we have developed a PD1-293T reporter cell line that stably expresses PD1 protein and serves as an effective platform for evaluating the effectiveness of LVLP.

[0083] 3.3Psi scheme shows higher packaging efficiency in LVLP

[0084] We developed a system to determine CE8e-SpRY mRNA copy number to accurately assess RNA content in LVLPs ( Figure 3A). In brief, CE8e-SpRY mRNA was first generated by IVT, and then RT-PCR reaction was performed using different concentrations of mRNA as template. The copy number calculation formula is as follows: RNA copy number = amount (ng) × 6.022 × 10 ^23 / (length (bp)×1×10 ^9 ×330) was calculated. The CE8e-SpRYmRNA copy number standard curve was established ( Figure 3 B). In addition, we extracted RNA from LVLP and treated it with DNase I to eliminate the remaining plasmid DNA. Copy number analysis showed that the Psi strategy produced the highest mRNA ( Figure 3 C). Luciferase assay showed that the luminescence index in the PD1-293T cell line decreased, suggesting that PD1 gene expression was disrupted ( Figure 3 D). We tested the base editing efficiency of several packaging schemes by transfecting LVLP into 293T cell lines and found that the Psi strategy showed higher base editing efficiency without FACS (without FACS, Figure 3 E). In summary, the Psi strategy was confirmed to be able to generate high-mRNA LVLPs, and we chose it to be used in subsequent studies.

[0085] 3.4 Optimizing the plasmid dosing ratio of LVLP

[0086] To enhance LVLP production while maintaining efficacy, we investigated the effect of plasmid dosage on LVLP synthesis by fine-tuning the ratio of transfected plasmid DNA (Table 1). This adjustment aims to optimize the balance and potentially increase yield without compromising functional integrity ( Figure 4 A). In the current study, we tested 12 different packaging plasmid ratios (Gag and CE8e-SpRY). We observed that as the amount of Gag / CE8e-SpRY-Psi combination increased, p24 production increased ( Figure 4 B), showing high base editing and mRNA content even at different stoichiometric ratios, indicating a positive effect of Psi on LVLP production. We also examined the attenuation of PD1 gene expression in the PD1-293T reporter cell line by luciferase assay ( Figure 4 D). Consistent with this, Psi-LVLP showed the highest editing without FACS ( Figure 4 E). We chose the V4 ratio because it maintained full activity with only a minor decrease in yield. In addition, we observed virus-like particles generated by the Psi packaging strategy by TME analysis ( Figure 4 F).

[0087] 3.5Psi-LVLP shows high base editing efficiency

[0088] Efficient base editing by LVLP relies heavily on the successful expression and precise processing of sgRNAs. Considering the high packaging efficiency of the Psi scheme in LVLP generation, we thought that the introduction of Psi would induce the target mRNA to enter the LVLP particles. We added Psi to the sgRNA backbone. Previous studies have found that HDVrz is incorporated into the CRISPR-Cas9 system to ensure precise processing of gRNAs. Its self-cleavage activity generates accurate 5' and 3' RNA ends, which is essential for the stability and functionality of sgRNAs. Therefore, to maximize the chances of identifying features and sgRNA backbone entry into LVLP, we placed the HDVrz-Psi element at the C-terminus of the sgRNA backbone, and HDVrz acts as a cleavable linker to promote the release of sgRNA after entering the LVLP ( Figure 5 A). In addition, we determined the effect of HDVrz-Psi on base editing activity by transfecting plasmids. Introduction of HDVrz-Psi did not eliminate gene editing ability, while retaining a certain degree of base editing activity ( Figure 5 B). Next, we used Psi-modified sgRNA to generate LVLP. The sgRNA content in LVLP was detected by ddPCR. The introduction of Psi had a certain effect on LVLP generation ( Figure 5 C), while enhancing the sgRNA copy number in the LVLP content ( Figure 5 D). Interestingly, base editing efficiency was improved ( Figure 5 E-5F). We also observed a decrease in PD1 protein expression in PD1-293T cells ( Figure 5 G). These data suggest that Psi can enhance the base editing activity of LVLP.

[0089] 3.6 Extensive application of Psi-LVLP in multiple immunotherapy-related genes

[0090] To validate the delivery capability of our LVLP system, we tested the gene editing efficiency of LVLP in Jurkat cells. We found that it can produce moderate base editing effects ( Figure 6 A). In addition, in the absence of VSVG protein, the gene editing effect of LVLP was greatly reduced ( Figure 6B), indicating the key role of envelope proteins in LVLP transfection. Similarly, we determined the gene editing characteristics of LVLP by using several known genes related to immunotherapy. B2M (a member of major histocompatibility complex (MHC) class I) and CIITA (a member of MHC class II) are the main causes of graft-versus-host disease (GVHD) associated with CAR-T cell therapy. Targeting the T cell receptor alpha constant (TRAC) site ensures uniform CAR expression and improves T cell potency, which is superior to conventional CAR-T cells. TIGIT is a newly discovered immune checkpoint, and its inhibitors have shown strong anti-tumor immune responses. Therefore, we first designed sgRNAs targeting the A / T sites of the start codons of these genes to produce gene knockout-like effects, which is similar to the previous PD1-sgRNA. We designed 8 forward sgRNAs and 8 reverse sgRNAs (called FX-sgRNAs) targeting ATG for B2M, CIITA, TRAC, and TIGIT (Table 2).

[0091] Table 2 sgRNAs for multiple genes within the editing activity window of CE8e-SpRY

[0092]

[0093]

[0094] By FACS, we detected the efficiency of A-to-G. For B2M, sg4 showed the highest base editing ability; FX-sg8CIITA; FX-sg4 TRAC; FX-sg6 TIGIT( Fig. 9 In addition, we constructed HDVrz-Psi sgRNA based on the above sites ( Figure 6 C). Similarly, when HDVrz-Psi was added, these sgRNAs retained some base editing activity ( Figure 6 D). We then generated LVLPs of these sgRNAs and examined base editing efficiency ( Figure 6 E). B2M-LVLP retained about 20% of base editing activity, while TIGIT-LVLP retained about 10% of base editing capacity. For CIITA and TRAC, LVLP showed mild base editing effects. We speculate that this may be attributed to the limited base editing potential and site heterogeneity of the Psi-modified plasmid in cells.

[0095] 3.7MR analysis validates the causal relationship of target genes in cancer

[0096] MR analysis was introduced to investigate the clinical significance of the selected sgRNA genes. We selected more than 20 cancers for MR analysis ( Figure 7 A). eQTL and methylation sites of PD1, B2M, CIITA, TRAC and TIGIT as exposure factors. GWAS datasets of various cancers, such as breast cancer, as outcome factors ( Figure 7 B-7C). After preliminary selection, the SNPs of each gene and its corresponding methylation site were clustered, generating a set of independent, genome-wide significant variants. The IVW method revealed statistically significant associations (p-value < 0.05) between some B2M, CIITA, PD1, and TIGIT variants and specific cancer risks ( Figure 7 D). MR analysis also elucidated the significant associations of some PD1, B2M, TIGIT, and CIITA methylation sites in cancer ( Figure 7 E-8A). MR-Egger intercept tests indicated no evidence of directional pleiotropy for these significant associations. Cochran's Q test indicated low heterogeneity among the SNPs used as IVs. Leave-one-out analysis confirmed that the results were not driven by any single SNP. MR-PRESSO analysis did not reveal significant outliers, further supporting the robustness of these findings. eQTL-GWAS co-localization analysis revealed shared significant SNPs associated with cancer in B2M, CIITA, PD1, and TIGIT ( Figure 8 B-8C). These findings provide evidence for a potential causal relationship between specific genetic variants and methylation sites in the PDCD1, B2M, CIITA, and TIGIT genes and the risk of certain cancers, suggesting broad clinical applications of Psi-LVLP in cancer immunotherapy.

[0097] 4. Discussion

[0098] This study explored the development and optimization of multiple LVLPmRNA packaging strategies to improve the efficiency and specificity of RNA-guided base editing. The focus was on leveraging the interaction between RNA-binding proteins and their corresponding RNA targets to maximize the packaging and functional delivery of CE8e-SpRYmRNA. The Psi strategy utilizing lentiviral packaging signals performed most effectively, showing superior LVLP generation capacity and efficient mRNA packaging. This finding emphasizes the critical importance of selecting an RNA packaging strategy in improving the delivery and functionality of RNA-guided nucleases. The Psi strategy excels in maximizing mRNA packaging because it improves the efficiency of delivery of editing components to target cells. By ensuring high CE8e-SpRYmRNA and sgRNA transcripts in the LVLP, the Psi strategy enhances base editing activity, as found by attenuating PD1 expression in our reporter cell lines.

[0099] Retaining integrase (IN) and reverse transcriptase (RT) in LVLPs poses several risks due to their roles in viral integration and replication. Integrase is responsible for the introduction of viral DNA into the host genome, which may lead to unintended genetic modifications and undesirable oncogenic effects. Reverse transcriptase plays a key role in converting viral RNA into DNA, a process that is accompanied by immunogenic epitopes that may trigger undesirable immune responses. To reduce these drawbacks, current studies use integrase-deficient lentiviral vectors (IDLVs) to prevent efficient viral DNA integration into the host genome by introducing specific mutations in integrase (e.g., D64, D116, and E152), thereby reducing the risk of insertional mutagenesis. Studies have shown that LVLPs lacking a functional RT are generally safer without significantly reducing their immunogenicity, as demonstrated by the strong immune responses elicited by LVLPs without RT in animal models. Therefore, removing or inactivating integrase and RT in LVLPs can reduce the risk of insertional mutagenesis and undesirable immune responses, making these modified particles more suitable for safe therapeutic use. A key innovation of our study is the complete deletion of integrase and RT in the Psi-LVLP system. Unlike IDLV, we only utilize the self-assembly property of HIV-Gag to produce LVLP without involving integrase and RT. HIV-Gag recognizes Psi-modified CE8e-SpRY and sgRNA, thus encapsulating them into viral particles. In clinical applications, safety issues are critical, especially the removal of integrase and reverse transcriptase.

[0100] Optimizing the packaging plasmid ratio is critical to optimizing LVLP yield while maintaining functional integrity. We systematically explored various plasmid ratios and found that the Psi strategy had a positive impact on LVLP production at higher plasmid ratios. This optimization resulted in increased p24 yield and improved base editing efficiency, demonstrating the importance of balanced plasmid composition for viral particle assembly. The optimized plasmid ratio not only maximized LVLP yield but also ensured the functional integrity of the particles, which is critical for the effectiveness of gene editing applications. Achieving high efficiency in both production and functionality of VLPs is necessary.

[0101] In this study, we focused on the nearly PAM-free CE8e-SpRY to generate LVLP, thereby eliminating the PAM influence of certain target sites. In addition, we selected sgRNA targeting the PD1 start codon, which has the highest base editing activity. Previous studies have also validated the gene knockout effect produced by ABE-induced start codon mutations (ATG to GTG or ACG). To rigorously evaluate the effectiveness of LVLP-mediated sgRNA delivery and PD1 expression, we engineered a dual-reporter 293T cell line that stably expresses PD1. This system provides a reliable readout of gene editing activity, enabling precise quantification of luciferase activity and PD1 expression.

[0102] MR analysis revealed major impacts of PD1, B2M, CIITA, and TIGIT in multiple cancers. Psi-LVLP was able to efficiently target and edit genes associated with immunotherapy response, highlighting its potential for clinical application. These findings suggest that Psi-LVLP may be a promising tool for CAR-T cell therapy and other immunotherapy strategies, providing precise and efficient genetic modification with the advantage of reduced off-target effects.

[0103] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. An optimized lentivirus-like particle delivery system, characterized in that: The delivery system includes a virus-like particle backbone protein, a specific membrane protein, a base editor mRNA and a sgRNA, wherein the virus-like particle backbone protein is HIVGAG protein, the base editor mRNA is packaged into the virus-like particles using the packaging signal Psi of the lentivirus itself, and the sgRNA is directionally packaged into the virus-like particles using a combination strategy of Psi and HDVrz nucleic acid elements.

2. A method for preparing an optimized lentivirus-like particle, characterized in that: The lentivirus-like particle delivery system according to claim 1 is used for preparation, and the specific process is as follows: Step 1: Inoculate 293T cells into a culture flask and culture for 20-24 hours before transfection; Step 2: After transfecting the packaging cell line 293T with the plasmid packaging system for 16-18 hours, wash with PBS to remove the remaining plasmid, and then replace with fresh culture medium; collect the supernatant 48h and 72h after transfection, and centrifuge to remove cell debris; Step 3: Mix the supernatant containing lentivirus-like particles with Lenti-X concentrate, incubate overnight at 4°C, centrifuge after incubation, resuspend the precipitated lentivirus-like particles with cold PBS, and aliquot and store the purified lentivirus-like particles.

3. The method for preparing an optimized lentivirus-like particle according to claim 2, characterized in that: The plasmid packaging system includes a specific membrane protein expression plasmid, a packaging plasmid for virus-like particle backbone protein, a base editor mRNA plasmid, and an sgRNA expression plasmid, with a total amount of 45 μg. Among them, the specific membrane protein expression plasmid is pVSV-G that expresses vesicular stomatitis virus G protein; the packaging plasmid for virus-like particle backbone protein is pHIV-gag; the base editor mRNA plasmid is pCE8e-SpRY-Psi; and the sgRNA expression plasmid contains HDVrz and Psi double nucleic acid elements.

4. The method for preparing an optimized lentivirus-like particle according to claim 3, characterized in that: The plasmids for the plasmid packaging system are as follows:

5. An optimized lentivirus-like particle, characterized in that: The method is prepared by any one of claims 2 to 4.

6. Use of the lentivirus-like particle delivery system according to claim 1 in targeted gene editing related to tumors and genetic diseases.

7. The use according to claim 6, characterized in that: The targeted genes include PD1, B2M, CIITA, TIGIT and TRAC. The sequences of sgRNA screened for PD1, B2M, CIITA, TIGIT and TRAC are shown in the following table:

8. The use according to claim 7, characterized in that: The indicator cell line 293T-PD1-Fluc was used to evaluate the targeted PD1 gene editing effect. The indicator cell line 293T-PD1-Fluc can stably express the fusion gene of the PD1 coding region CDS and the firefly luciferase gene. The luciferase value was detected by adding luciferase substrate to reflect the editing efficiency of the targeted PD1 gene.

9. Use of the lentivirus-like particle delivery system according to claim 1 in delivering Crispr / Cas9 gene editing tools.

Images