A method for screening and verifying liver cancer personalized neoantigens and application thereof

Abstract

This invention discloses a method for screening and validating personalized neoantigens for liver cancer and its application. DNA and RNA are extracted from tumor and adjacent normal tissue samples from patients and subjected to high-throughput sequencing. Mutant peptides are screened, and the patient's HLA typing is determined. The binding affinity of the mutant peptides to HLA alleles is then evaluated, and antigen presentation ability is predicted. Candidate neoantigen peptides are comprehensively screened. The candidate neoantigens undergo coding sequence sequencing, cloning, PCR amplification, in vitro transcription, capping and tailing modifications, and encapsulation to prepare an mRNA-LNP vaccine. The obtained mRNA-LNP vaccine is then transfected into 293T cells in vitro to verify antigen expression efficiency, completing the functional quality control of the mRNA-LNP vaccine. This invention provides a highly efficient, closed-loop system for screening personalized neoantigens for liver cancer and validating vaccines, an integrated approach that directly guides mRNA vaccine construction and in vivo immunogenicity evaluation from multi-omics data screening.

Description

Technical Field

[0001] This invention relates to a method for screening and validating personalized neoantigens for liver cancer and its application, specifically a method for screening and validating personalized neoantigens based on tumor mutation profiles and HLA typing and its application in the preparation of mRNA vaccines, belonging to the fields of biomedicine and immunotherapy technology. Background Technology

[0002] Tumor neoantigens originate from somatic cell mutations and can be specifically recognized by the host's immune system, providing ideal targets for tumor immunotherapy. In recent years, with the widespread adoption of high-throughput sequencing technology, screening effective personalized neoantigens from massive amounts of genomic data has become a core part of cancer vaccine development.

[0003] Currently, this field mainly focuses on bioinformatics prediction and screening. It identifies mutations by performing whole-exome sequencing on tumors and normal tissues, and then uses algorithms to predict the affinity of mutated peptides for human leukocyte antigens (HLA) in patients, and selects candidate neoantigens based on prediction scores.

[0004] However, this computational model-based screening strategy faces a series of challenges in its practical application to vaccine development and application. First, its screening criteria are often too simplistic, focusing excessively on the initial step of HLA binding affinity, failing to systematically integrate and validate the complete biological pathway of neoantigens from transcriptional expression, processing and presentation to ultimately eliciting T-cell immune responses. Therefore, some candidate peptides that score highly in computation may lose activity in subsequent immune recognition stages due to low actual expression levels in tumor cells or inability to be effectively processed and presented. Furthermore, this model typically stops at virtual screening, failing to establish an effective connection with downstream in vivo vaccine efficacy testing. Existing technologies rarely include a coherent process of constructing candidate antigens into vaccine form after screening and conducting comprehensive immunogenicity evaluation in mouse models with humanized immune systems. This leads to a disconnect between screening results and the final vaccine efficacy.

[0005] Therefore, there is an urgent need in this field for a complete technical solution that can connect "bioinformatics screening - rapid vaccine construction - in vivo immune validation" to improve the reliability of personalized neoantigen screening and accelerate its transformation into effective immunotherapy products. Summary of the Invention

[0006] The technical problem to be solved by this invention is to address the shortcomings of existing technologies where bioinformatics screening and in vivo vaccine efficacy evaluation are disconnected. This invention provides a method for screening and validating personalized neoantigens for liver cancer and its application. The method of this invention is an integrated scheme that directly leads from multi-omics data screening to mRNA vaccine construction and in vivo immune effect evaluation, providing an efficient and closed-loop system for screening personalized neoantigens for liver cancer and validating vaccines.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] One objective of this invention is to provide a method for screening and validating personalized neoantigens for liver cancer, comprising the following steps:

[0009] (1) Multi-omics integrated screening and prediction of neoantigens:

[0010] Tumor tissue samples and paired adjacent normal tissue samples from liver cancer patients were obtained. Genomic DNA and total RNA were extracted from these samples. Whole-exome sequencing was performed on the genomic DNA, and reference transcriptome sequencing was performed on the total RNA. HLA typing was performed on the tumor tissue samples to determine the alleles. Mutation identification and bioinformatics analysis were performed on the sequencing results to screen for mutant peptides encoded by genes that exhibit mutations and differential expression. The binding affinity between the mutant peptides and HLA alleles was evaluated, and antigen presentation ability was predicted. Candidate neoantigen peptides that simultaneously meet the criteria of "high expression," "high affinity," and "high presentation score" were screened.

[0011] (2) Construction of mRNA-LNP vaccine:

[0012] The candidate neoantigens obtained in step (1) were encoded by the coding sequence, and a fusion antigen gene TSA was obtained after codon optimization and tandem design. The TSA was cloned into the corresponding site of the expression vector. During cloning, 5′UTR and 3′UTR and His tag coding sequences to enhance translation efficiency were introduced upstream and downstream of the TSA sequence, respectively, to construct the recombinant plasmid pVAX1-M1+2-TSA. Using the plasmid as a template, a linearized DNA transcription template was obtained by PCR amplification using primers with a T7 promoter. The DNA transcription template was transcribed in vitro, capped and tailed in sequence to obtain high-purity mRNA. The high-purity mRNA was then encapsulated with lipid nanoparticles to prepare the mRNA-LNP vaccine.

[0013] (3) In vitro expression validation of mRNA-LNP vaccine:

[0014] The mRNA-LNP vaccine prepared in step (2) was used to verify the antigen expression efficiency by transfecting 293T cells in vitro, which confirmed that the vaccine could express and secrete the target antigen efficiently in vitro, thereby completing the functional quality control of the mRNA-LNP vaccine.

[0015] In the above technical solution, in step (1), when performing mutation identification and bioinformatics analysis on the sample sequencing results, the specific method is as follows: First, compare the genomic DNA sequencing results of tumor tissue and adjacent normal tissue, distinguish and screen out common germline mutations, and retain somatic mutation data that only exist in tumors; then integrate the total RNA transcriptome sequencing results, evaluate the expression level of mutated genes, filter out mutations with low expression levels in tumor tissue, and retain genes with significant differential expression; finally, perform screening, filtering, and intersection analysis on somatic mutation data and differentially expressed gene data, and finally select the peptide segment encoded by the gene with mutation and differential expression.

[0016] In the above technical solution, in step (1), when performing neoantigen prediction and screening, the specific method is as follows: first, perform functional annotation on the identified somatic mutations, and extract the peptide sequence containing the mutation site based on the annotation results; then evaluate the binding affinity of the mutated peptide sequence with the HLA allele, and screen out peptides with high binding affinity; then predict the antigen presentation ability and screen out candidate neoantigens that simultaneously meet the criteria of "high expression", "high affinity" and "high presentation score";

[0017] Furthermore, the HLA allele is HLA-B*40:01;

[0018] Furthermore, the binding affinity of the mutated peptide sequence to the HLA allele was assessed using the NetMHCpan-4.1 algorithm, with the binding affinity screening criterion being IC50 < 500 nM.

[0019] Furthermore, the MHCflurry 2.0 algorithm was used to predict antigen presentation capability;

[0020] Furthermore, the aforementioned "high expression", "high affinity" and "high presentation score" refer to the fact that the source gene of the peptide has a high expression level in tumor tissue, and that the peptide has a high binding affinity and a high presentation score with the patient-specific HLA allele.

[0021] In the above technical solution, in step (2), the complete nucleotide coding sequence of the fusion antigen gene TSA is shown in SEQ ID NO:1; the sequence of the linearized DNA transcription template obtained by PCR amplification is shown in SEQ ID NO:2.

[0022] In the above technical solution, in step (2), the expression vector is pVAX1; the 5′UTR and 3′UTR are preferably selected from known sequences that have the function of enhancing translation efficiency, such as UTR sequences derived from the M gene of influenza virus and subjected to site-directed mutation.

[0023] In the above technical solution, in step (2), the encapsulation of high-purity mRNA using lipid nanoparticles has an encapsulation rate of not less than 90%.

[0024] In the above technical solution, in step (3), the specific method for verifying the in vitro expression of the mRNA-LNP vaccine is as follows: mRNA-LNP vaccine is added to human embryonic kidney 293T cells and then transfected; after 48 hours of transfection, the cells are denatured, gel electrophoresis, membrane transfer, blocking, primary antibody incubation overnight, washing, secondary antibody incubation for 1 hour, washing, and incubation in the dark for a few seconds before image acquisition.

[0025] The second objective of this invention is to provide a personalized anti-liver cancer mRNA-LNP vaccine prepared by the above method.

[0026] A third objective of this invention is to provide the application of the aforementioned mRNA-LNP vaccine in the preparation of a drug for stimulating or evaluating an anti-hepatocellular T-cell immune response.

[0027] The beneficial effects of the technical solution provided by this invention are as follows:

[0028] (1) It realizes the closed-loop connection of the whole process of “screening-construction-validation”, integrates multi-omics prediction, rapid vaccine preparation and in vivo functional evaluation into a coherent system, and significantly improves the efficiency and reliability of personalized vaccine development.

[0029] (2) Immunological validation was performed directly using a humanized HLA-matched mouse model, providing physiologically relevant in vivo evidence for the immunogenicity of candidate neoantigens and overcoming the problem of disconnect between traditional virtual screening and in vivo effects.

[0030] (3) The introduction of in vitro expression verification and LNP quality control in the vaccine construction process ensured the bioactivity and delivery efficiency of the formulation, and provided controllable and reproducible preparation process support for subsequent clinical translation. Attached Figure Description

[0031] Figure 1 This is a flowchart of the screening process for candidate neoantigens in Example 1;

[0032] Figure 2This is a summary diagram of the somatic mutation data and gene information of the patients in Example 1, where: the top left is the variant classification diagram, the top middle is the variant type diagram, the top right is the SNV class diagram, the bottom left is the variant per sample median diagram, the bottom middle is the variant classification summary diagram, and the bottom right is the top 10 mutated genes diagram;

[0033] Figure 3 The physical spectrum of the recombinant plasmid pVAX1-M1+2-TSA in Example 2;

[0034] Figure 4 This is a gel electrophoresis image of the linearized DNA transcription template in Example 2;

[0035] Figure 5 This is a graph showing the mRNA-LNP quality control results from Example 2.

[0036] Figure 6 This is an electron microscopy image of the mRNA-LNP observed in Example 2;

[0037] Figure 7 This is a Western blot (WB) image of 293T cells transfected with mRNA-LNP in Example 3.

[0038] Figure 8 To verify the flowchart of TSA-mRNA-LNP and OVA+FIA vaccine immunization in mice in Example 1;

[0039] Figure 9 To verify the antigen-specific IFN-γ level in mouse spleen tissue in Example 1, ELISPOT results were displayed.

[0040] Figure 10 To verify the statistical chart of antigen-specific IFN-γ level spot count in mouse spleen tissue in Example 1. Detailed Implementation

[0041] The following describes in detail the specific embodiments of the technical solution of the present invention, but the present invention is not limited to the following description:

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, specific embodiments of the invention will be described in detail below with reference to the accompanying drawings. The following examples are for illustrative purposes only and should not be considered as limiting the scope of protection of this invention. Unless otherwise specified, the experimental methods used in this invention are conventional methods in the art, and the reagents and materials used are commercially available.

[0043] This invention provides a complete technical system from bioinformatics screening to in vivo functional verification. Its core lies in the seamless integration and closed-loop verification of personalized neoantigen prediction for liver cancer patients, rapid preparation of multi-antigen mRNA vaccines, and evaluation of immunogenicity in humanized matched animal models. The following specific embodiments will illustrate each step of this system step by step.

[0044] Example 1: Multi-omics integrated screening and prediction of personalized neoantigens in liver cancer patients

[0045] This embodiment demonstrates how to screen for neoantigens from liver cancer patient samples, as shown in the flowchart below. Figure 1 As shown, the specific steps include:

[0046] (1) Sample collection and high-throughput sequencing:

[0047] Establish patient inclusion and exclusion criteria, collect patient clinical data, obtain informed consent from patients and approval from the ethics committee (ethics approval number: 2024-LY-ky013).

[0048] Hepatocellular carcinoma (HCC) tissue and paired adjacent normal tissue samples were collected from patients, and genomic DNA and total RNA were extracted from them. Whole-exome sequencing was performed on the DNA samples using the Illumina NovaSeq 6000 platform, and reference transcriptome sequencing was performed on the RNA samples. After successful library construction, the data were processed and analyzed. Data preprocessing was performed using FastP (Version: 0.19.5), and then clean reads were aligned to the reference genome GRCh38.p13 using BWA (Version: 0.7.12) to determine their positions on the reference genome. The alignment algorithm was BWA Men, and the parameters were set to default. The alignment results were converted and sorted using SAMtools (Version: 1.4), and PCR duplicate reads were removed using Picard (Version: 4.1.0.0). The sample data were evaluated for compliance with standards based on sequencing error rate, data volume, alignment rate, and coverage within the capture interval, and further analysis was performed accordingly.

[0049] (2) Mutation identification and bioinformatics analysis:

[0050] ① Somatic mutation analysis: The sequencing results of the samples were compared with the reference genome GRCh38.p13. The Mutect2 module of GATK4 software was used to detect somatic mutations in normal samples and their matched tumor samples. Single nucleotide variants and insertion / deletion detection results were annotated into databases such as EXAC and gnomAD using Annovar software. In bioinformatics analysis, tumor tissues were compared with normal tissues to screen out shared germline mutations, thus accurately preserving somatic mutations unique to tumors. A summary of some mutation data and their gene information is shown below. Figure 2 As shown.

[0051] ② Differentially expressed gene analysis: RNA-seq data were integrated, quality assessed, and preprocessed. High-quality clean reads were then matched to the reference genome GRCh38.p13 using the hisat2 software to determine the genomic location of each sequence and sample-specific sequence characteristics. Gene expression analysis was performed using DESeq2 software: First, the original counts data were standardized based on BaseMean, and the fold change in differential expression was calculated. Then, a negative binomial distribution test was used for significance analysis. The screening of differentially expressed protein-coding genes was ultimately determined based on both the fold change and statistical significance results. Based on the standardized gene expression data, mutations with low expression levels (TPM < 1) in tumor tissue were filtered out using log2(Fold Change) ≥ 0.58, P < 0.05, and a TPM value ≥ 1 for each mutated gene in tumor tissue as screening criteria, ensuring that candidate neoantigens originated from actively transcribed genes.

[0052] ③ Intersection analysis of mutated and differentially expressed genes: Genes that have been screened and meet the criteria in the somatic mutation and protein coding differential expression data are selected for intersection analysis and the results are compared. DNA mutations are verified by transcriptome analysis and RNA expression information is supplemented. Finally, mutated genes that can be differentially expressed are selected.

[0053] (3) HLA typing of patients: High-resolution HLA typing was performed using tumor tissue samples. Genomic DNA was extracted using a commercial kit (Qiagen) at a concentration ≥50 ng and a purity A260 / A280 between 1.8 and 2.0. Key polymorphic regions of HLA class I genes (including A, B, and C loci) were amplified using high-throughput sequencing on the Illumina platform. After FastQC quality control, the obtained sequencing data were compared with the IMGT / HLA database (version 3.50.0) using OptiType software, and finally, the genotypes of each allele were identified using a four-digit nomenclature.

[0054] (4) Neoantigen prediction and screening:

[0055] A strategy of cross-validation integrating whole-exome and transcriptome sequencing data was adopted to correct for false positives identified solely from somatic mutation lists. Annovar was used to annotate point mutations and insertions / deletions to determine their corresponding amino acid alterations. Human protein reference sequences (FASTA format) were downloaded from NCBI, and peptide sequences containing mutation sites were extracted based on the annotation results. NetMHCpan-4.1 was used to assess the binding affinity of the mutated peptides to the patient's HLA-B*40:01 allele, with an IC50 < 500 nM as the screening criterion. Furthermore, the antigen presentation prediction tool MHCflurry2 was used to score the presentation ability of candidate peptides and prioritize them. Finally, top-tier candidate neoantigens that simultaneously met the criteria of "high expression," "high affinity," and "high presentation score" were selected. The entire screening process is as follows: Figure 1 As shown.

[0056] Example 2: Construction of mRNA vaccine

[0057] This embodiment provides a detailed explanation of how to design a fusion antigen sequence and construct it into an mRNA vaccine:

[0058] (1) Design coding sequences for candidate neoantigens:

[0059] Based on the prediction results of the top candidate neoantigens screened in Example 1, the top 8 predicted HLA-B*40:01 candidate neoantigens (Table 1) were selected, and their coding sequences were optimized using human codons to improve translation efficiency in mammalian cells. Subsequently, the optimized antigen sequences were tandemly linked together using the DNA sequence “GCCGCCGCC” encoding a short flexible linker peptide (whose encoded amino acid sequence is Ala-Ala-Ala) to form a fusion antigen gene, named TSA. The complete nucleotide sequence of the TSA fusion gene (SEQ ID NO:1) is as follows:

[0060] CACGAGATGGTGCAGCAGAAGGTGGTGGCCCTGGCCGCCGCCAAGGAGAGGAGCAACGTGGTGCTGGCCGCCGCCGCCGAGGACAGCCTGGCCGCCCAGGCCCTGGCCGCCGCCGAGGAGGCCAGCAGCCTGAACACCCTGGCCGC CGCCAGGGAGGAGCTGTTCGGCGAGGCCAGCGTGGCCGCCGCCAGGGAGACCACCGCCGTGGCCGTGGCCGCCGCCTTCGAGAACATCGAGAGCCCCCTGATCGCCGCCGCCAGCGAGGCCGAGGGCGGCGTGGTGTGCAGGCTG.

[0061] Table 1: Nucleotide sequences of the top 8 HLA-B*40:01 candidate neoantigen peptides in patients with high phasing

[0062] Gene name sequence PIK3R4 CACGAGATGGTGCAGCAGAAGGTGGTGGCCCTG KTN1 AAGGAGAGGAGCAACGTGGTGCTG ANKRD16 GCCGAGGACAGCCTGCCCGCCCAGGCCCTG ZNF644 GAGGAGGCCAGCAGCCTGAACACCCTG STXBP5L AGGGAGGAGCTGTTCGGCGAGGCCAGCGTG CATSPERG AGGGAGACCACCGCCGTGGCCGTG USP47 TTCGAGAACATCGAGAGCCCCCTGATC LIMCH1 AGCGAGGCCGAGGGCGGCGTGGTGTGCAGGCTG

[0063] (2) Preparation of personalized mRNA-LNP vaccines:

[0064] ① Recombinant plasmid construction and template preparation

[0065] The coding sequence of the TSA fusion gene shown in SEQ ID NO: 1 was cloned into the corresponding site of the eukaryotic expression vector pVAX1. During the cloning process, a 5′UTR sequence (derived from a site-directed mutant of the influenza virus M gene, sequence see SEQ ID NO: 3AGTAAAAGCAGGTAGATATTGAAAG in patent CN119776350A) and a secretory signal peptide sequence were introduced upstream of the TSA sequence to enhance translation efficiency; downstream, the corresponding 3′UTR sequence (sequence see SEQ ID NO: 2AAAACTACCTTGTTTCTACT in the patent) and a 6×His tag coding sequence were introduced. The successfully constructed recombinant plasmid was named pVAX1-M1+2-TSA, and its physical map is shown below. Figure 3 As shown.

[0066] Using pVAX1-M1+2-TSA plasmid as a template, high-fidelity PCR amplification was performed using primers with the T7 promoter (upstream primer: TAATACGACTCACTATAGGGA; downstream primer: AGTAGAAACAAGGTAGTTTTTTA) to obtain a linearized DNA transcription template. The specific PCR amplification reaction system is shown in Table 2, and the reaction procedure is shown in Table 3. The nucleotide sequence of the linearized DNA transcription template (SEQ ID NO:2) is as follows:

[0067] TAATACGACTCACTATAGGGAGTAAAAGCAGGTAGATATTGAAAGATGATTGAGGTGTTATTAGTGACTATTTGTTTAGCTGTGTTCCCTTATCAAGGTAGTCACGAGATGGTGCAGCAGAAGGTGGTGGCCCTGGCCGCCGCCAAGGAGAGGAGCAACGTGGTGCTGGCCGCCGCCGCCGAGGACAGCCTGCCCGCCCAGGCCCTGGCCGCCGCCG AGGAGGCCAGCAGCCTGAACACCCTGGCCGCCGCCAGGGAGGAGCTGTTCGGCGAGGCCAGCGTGGCCGCCGCCAGGGAGACCACCGCCGTGGCCGTGGCCGCCGCCTTCGAGAACATCGAGAGCCCCCTGATCGCCGCCGCCAGCGAGGCCGAGGGCGGCGTGGTGTGCAGGCTGCATCATCATCATCATTAAAAAACTACCTTGTTTCTACT.

[0068] Table 2: PCR amplification reaction system for the target gene

[0069] Components volume Forward primer (10 μM) 8 μL Reverse primer (10 μM) 8 μL 2 × Phanta UniFi Master Mix 100 μL DNA Template 4 μL Water, nuclease-free 80 μL Overall system 200 μL

[0070] Table 3: PCR amplification reaction procedure for the target gene

[0071] step program time 1 Pre-denaturation at 98℃ 30 sec 2 denaturation at 98℃ 10 sec 3 Annealing at 58℃ 10 sec 4 Extended to 72℃ 15 sec 5 GOTO Step 2 30× 6 72℃ final extension 5 min 7 Store at 12℃ ∞

[0072] The amplification product was verified by 1% agarose gel electrophoresis at 140 V, 360 mA, and 30 minutes to be a single bright band, with the size consistent with expectations. Figure 4 The PCR products were then purified using a gel extraction process and used for the next step of transcription.

[0073] ②In vitro transcription of mRNA:

[0074] The purified DNA template was transcribed using the T7 RNA polymerase in vitro transcription system shown in Table 4. After incubating the mixture in a 37°C metal bath for at least 4 hours, 1 μL of DNase was added directly to the reaction tube, and the tube was incubated again at 37°C for 15 minutes to completely degrade the DNA template. Then, 30 μL of nuclease-free water and 30 μL of Lithium Chloride Precipitation Solution (LiCl) were added to the system, and the tube was frozen overnight at -20°C. After centrifugation at 13,200 × g for 15 minutes at 4°C, 1 mL of pre-chilled 75% ethanol solution was added to the tube containing the RNA precipitate. The tube was centrifuged again under the same conditions for 3 minutes, and the ethanol supernatant was discarded. The opened centrifuge tubes were then placed in a clean bench and allowed to air dry at room temperature for 20 minutes. Finally, the precipitate was dissolved in 40 μL of nuclease-free water and stored at -80°C for later use.

[0075] Table 4: In vitro transcription system

[0076] Components volume 10×Transcription Buffer 2μL ATP 1.5μL GTP 1.5μL CTP 1.5μL N'-Me-Pseudo UTP 1.5μL Template DNA 1μg Enzyme Mix 1μL RNase Free Water Up to 20μL

[0077] ③ Capping and tailing modifications of mRNA:

[0078] Take 50 μg of in vitro transcribed RNA (≥100 nt) for the capping reaction, dilute it with nuclease-free water to a volume of 67 μL, incubate at 65°C for 5 minutes, and immediately place on ice for 5 minutes after incubation. Then prepare the capping reaction system according to the order shown in Table 5, and incubate at 37°C for 30 minutes. After capping, prepare the tailing reaction system according to Table 6, and add a Poly(A) tail to the 3′ end of the mRNA using Poly(A) polymerase, and incubate at 37°C for 30 minutes. Purify the reaction product using the LiCl precipitation method described above to obtain high-purity mRNA.

[0079] Table 5: Capped Reaction System

[0080] Components volume Denatured RNA 67μL 10×Capping Reaction Buffer 10μL GTP (10mM) 10μL SAM (20mM) 2.5μL Recombinant RNase Inhibitor (40U / μL) 2.5μL mRNA Cap 2'-O-MethyItransferase(100U / μL) 4μL Vaccinia Capping Enzyme(10U / μL) 4μL

[0081] Table 6: Tailing Reaction System

[0082] Components volume 10×Poly(A) Polymerase Buffer 10μL ATP (10mM) 10μL Capped RNA 100μL Poly(A) Polymerase (5U / μL) 5μL RNase Free Water Up to 125μL

[0083] ④ Encapsulation and characterization of lipid nanoparticles:

[0084] Four lipids—SM102, DSPC, DMG-PEG2000, and cholesterol—were dissolved in anhydrous ethanol to a concentration of 16 mM. These solutions were then mixed in a volume ratio of 50:10:1.5:38.5, and further diluted with an appropriate amount of anhydrous ethanol to prepare a lipid solution. The purified mRNA was diluted with 50 mM citrate buffer to approximately 108 μg / mL, based on an N / P ratio of 6, to obtain the mRNA working solution. Using microfluidic mixing technology at a flow rate ratio of 1:3 and a total flow rate of 16 mL / min, the mRNA and lipids were mixed in the specified proportions, and self-assembled to form mRNA-LNP.

[0085] The encapsulated mRNA-LNP was diluted with 3 volumes of citrate buffer and then added to a 100 kDa ultrafiltration tube. The tube was centrifuged at 3000 × g at 4°C until the ethanol content dropped below 0.5%. Finally, an equal volume of 20% sucrose-PBS solution was added to the concentrate to obtain the final ultrafiltration concentrate, which was stored at 4°C for later use.

[0086] The prepared mRNA-LNP was systematically characterized: dynamic light scattering analysis showed an average particle size of 109.2 nm, a polydispersity index (PI) of 0.1516, and a zeta potential of -3.05 mV. Figure 5 Transmission electron microscopy revealed that the particles were uniformly spherical and regularly shaped. Figure 6 Encapsulation efficiency was assessed using the Quant-iT RiboGreen RNA kit, and results showed that over 90% of the mRNA was successfully encapsulated within the LNP.

[0087] Example 3: In vitro expression validation of mRNA-LNP vaccine:

[0088] A comprehensive in vitro quality assessment was performed on the mRNA-LNP vaccine obtained in Example 2:

[0089] Human embryonic kidney 293T cells were distributed at a rate of 2 × 10⁻⁶ cells per well. 5 Cells were evenly seeded at a density of 1000 cells per well in 6-well plates and incubated at 37°C in a 5% CO2 incubator for 18 hours. The original culture medium in each well was aspirated, and 2 mL of Opti-MEM medium was added. Then, 2 μg of mRNA-LNP was added dropwise to each well. After incubation at 37°C for 6 hours, the medium was replaced with complete culture medium and the cells were cultured for another 6 hours.

[0090] Forty-eight hours after transfection, Western blotting (WB) was performed: cell lysates and supernatants were collected, mixed with 5× protein loading buffer, and heated at 100°C for 10 minutes. The denatured proteins were then subjected to electrophoresis on a 15% SDS-PAGE gel at 80V for 20 minutes, followed by 160V for 50 minutes. A 0.22 μm PVDF membrane was assembled using a "sandwich" structure (sponge-filter paper-gel-PVDF membrane-filter paper-sponge) and transferred at 4°C and 400mA for 15 minutes. After transfer, the membrane was blocked with protein-free rapid blocking buffer on a horizontal shaker for 30 minutes. The specific mouse His-tag primary antibody was diluted 1:10000 with WB-specific primary antibody dilution buffer, and the membrane and working solution were incubated together at 4°C overnight. The PVDF membrane was washed with TBST on a rapid shaker for 10 minutes each time, repeated three times. The HRP-labeled secondary antibody was diluted 1:5000 with WB-specific secondary antibody dilution buffer and incubated on a slow shaker at room temperature for 1 hour. The membrane was washed three times with TBST in the same manner. After incubating the membrane in a chemiluminescent substrate in the dark for a few seconds, images were acquired.

[0091] Image acquisition results as follows Figure 7 As shown, the correct-sized TSA fusion protein bands were detected in the lysate and supernatant of the transfected cells, while no such band was found in the blank control group, confirming that the vaccine can efficiently express and secrete the target antigen in vitro.

[0092] Validation Example 1: Evaluation of the Immunogenic Effect of mRNA-LNP Vaccine in a Humanized Mouse Model

[0093] This validation example evaluates the immunogenicity and ability to elicit T-cell responses of the personalized mRNA-LNP vaccine prepared in Example 2 in vivo. The flowchart is as follows. Figure 8 As shown, the specific steps include:

[0094] 1. Immunization regimen:

[0095] HLA-B*40:01 humanized transgenic mice aged 6-8 weeks were selected. Mice were randomly divided into two groups (n=5 per group): an experimental group (vaccinated with TSA-mRNA-LNP vaccine) and a positive control group (vaccinated with OVA+FIA vaccine). The immunization schedule was as follows: Figure 8 As shown, mice in the experimental group were injected intramuscularly into their legs on days 0 and 14, while mice in the control group were injected subcutaneously. The immunization dose was 30 μg / mouse and the injection volume was 0.1 mL each time.

[0096] 2. OVA+FIA vaccine preparation and quality control:

[0097] Ovalbumin (OVA) was diluted to 0.6 μg / μL with sterile PBS. An equal volume of OVA protein solution was mixed with Freund's incomplete adjuvant (FIA), and the mixture was repeatedly pumped on ice or pushed back and forth with two syringes until completely emulsified. The emulsified product was then dropped into water for testing. If the product was round and did not disperse for a long time, it was the OVA+FIA vaccine.

[0098] 3. Antigen-specific T cell response detection:

[0099] 3.1 Isolation of mouse spleen cells

[0100] Fourteen days after the last immunization (day 28), mice were euthanized and spleen lymphocytes were isolated. Mice euthanized by cervical dislocation were sterilized by immersion in 75% ethanol and dissected in a biosafety cabinet. 5 mL of mouse lymphocyte separation medium was added to a 35 mm culture dish. The spleen was removed, placed in a 70 μm sieve, and immersed in the separation medium. After grinding the spleen with a 1 mL syringe plunger, the separation medium containing the spleen cells was immediately transferred to a 15 mL centrifuge tube. 1 mL of RPMI 1640 medium was added along the tube wall, and the centrifuge was set to 3°C at 800g for 30 minutes. The lymphocyte layer was aspirated, 10 mL of RPMI 1640 medium was added, and the centrifuge was incubated at 250g for 10 minutes. The medium was poured off, 2 mL of erythrocyte lysis buffer was added, and the mixture was allowed to stand for 2 minutes. Then, 10 mL of RPMI 1640 medium was added, and the mixture was incubated at 450g for 5 minutes. The supernatant was discarded, and the lymphocytes were resuspended in 1 mL of medium. Mix 10 μL of cell suspension with an equal volume of trypan blue staining solution, count the cells, and obtain mononuclear cells if the trypan blue rejection rate is ≥95%.

[0101] 3.2 ELISPOT assay to detect the level of IFN-γ secreted by antigen-specific T cells

[0102] The binding sites on the PVDF membrane surface were activated using serum-free culture medium (200 μL / well), and the cells were removed after 10 min. Spleen cells from mice immunized with TSA-mRNA-LNP vaccine and OVA+FIA vaccine were seeded at a density of 1 × 10^5 cells per well in pre-coated ELISPOT plates, with 10 μL of the corresponding stimulant or culture medium added to each well. The experimental setup included the following five groups:

[0103] (1) Specific antigen stimulation group: TSA vaccine group cells were stimulated with a peptide library covering TSA fusion protein to assess TSA-specific immune response; OVA vaccine group cells were stimulated with OVA-specific peptide SIINFEKL to assess OVA-specific immune response.

[0104] (2) Cross-stimulation control group: TSA vaccine group cells were stimulated with OVA peptides, and OVA vaccine group cells were stimulated with TSA peptide library as cross-control to verify the antigen specificity of immune response.

[0105] (3) Negative control group: Each group of cells was set up with wells containing only complete culture medium and no antigen stimulation, in order to detect the background level of spontaneous IFN-γ secretion by cells.

[0106] (4) Background control group: wells containing no cells and only complete culture medium were set up to detect the background signal of the reagents and system.

[0107] (5) Positive control group: Each group of cells was equipped with wells containing PMA (final concentration 500 ng / mL) and Ionomycin (final concentration 10 μg / mL) as positive controls to confirm cell reactivity.

[0108] All wells were incubated at 37°C and 5% CO2 for 30 hours to ensure full activation of antigen-specific T cells and secretion of effector factors. After removing the cell culture, 200 μL of pre-cooled deionized water was added to each well, and the wells were incubated at 4°C for 10 min. Six cycles of washing with 1× washing buffer (260 μL / well) were performed, with each wash lasting 1 min followed by thorough drying. Biotinylated antibody working solution (100 μL / well) was added, and the wells were incubated at room temperature in the dark for 1 hour. The washing process was repeated, with the addition of HRP-labeled avidin (100 μL / well) and incubation at 37°C for 1 hour. The washing process was repeated, and after the final wash, the plate was disassembled, and the back side of the membrane was rinsed three times with deionized water to ensure removal of non-specific adsorbed substances. Tetramethylbenzidine chromogenic substrate (100 μL / well) was used for development in the dark until the spots were clearly visible. Finally, each well was rinsed five times with deionized water to terminate the enzymatic reaction, and the wells were inverted and air-dried for spot counting analysis.

[0109] ELISPOT results of antigen-specific IFN-γ levels in mouse spleen tissue are as follows: Figure 9 As shown, spleen cells of mice in the TSA-mRNA-LNP vaccine group stimulated with the TSA peptide library produced a large number of IFN-γ spots. The spots were counted and statistically analyzed. Figure 10 The results are shown below:

[0110] (1) Under specific stimulation: After correction for the negative control background in the same group, the number of antigen-specific IFN-γ spot-forming cells (SFCs) produced by the spleen cells of mice immunized with TSA-mRNA-LNP vaccine under the stimulation of TSA peptide library was 568.2±39.70 (mean ± SD, n=5); the number of antigen-specific SFCs produced by the spleen cells of mice immunized with OVA+FIA vaccine under the stimulation of SIINFEKL peptide was 152.4±23.31 (mean ± SD, n=5). The specific responses induced by the two vaccines were significantly higher than their respective negative control background levels (both P<0.0001).

[0111] (2) Intergroup comparison: The antigen-specific T cell response intensity induced by the TSA-mRNA-LNP vaccine group was significantly higher than that induced by the OVA+FIA vaccine group (P<0.0001).

[0112] (3) Cross-stimulation verification: The number of spots produced by cells in the TSA-mRNA-LNP vaccine group under SIINFEKL peptide stimulation and the number of spots produced by cells in the OVA+FIA vaccine group under TSA peptide library stimulation were not significantly different from those in the negative control wells. This proves that the T cell responses induced by the two vaccines have strict antigen specificity and there is no cross-reactivity.

[0113] The above results demonstrate that the personalized neoantigen mRNA vaccine (TSA-mRNA-LNP) prepared in this invention can elicit a stronger T-cell immune response than the classic model antigen vaccine (OVA+FIA) in a matched humanized mouse model, and this response exhibits high antigen specificity. This provides crucial functional evidence for the personalized therapeutic potential of the vaccine.

[0114] This invention is not limited to the specific embodiments described above. Any modifications or equivalent substitutions made by those skilled in the art to the technical solutions based on the concept of this invention, without departing from the spirit and scope of the technical solutions of this invention, should be included within the protection scope of this invention.

Claims

1. A method for screening and verifying liver cancer personalized neoantigens, characterized in that, Includes the following steps: (1) Multi-omics integrated screening and prediction of neoantigens: Tumor tissue samples and paired adjacent normal tissue samples from liver cancer patients were obtained. Genomic DNA and total RNA were extracted from the samples. Whole-exome sequencing was performed on the genomic DNA, and reference transcriptome sequencing was performed on the total RNA. HLA typing was performed on the tumor tissue samples to determine the alleles. Mutation identification and bioinformatics analysis were performed on the sequencing results to screen for mutant peptides encoded by genes that have mutations and can be differentially expressed. The binding affinity between mutant peptides and HLA alleles was evaluated, and antigen presentation ability was predicted. Candidate neoantigen peptides that simultaneously meet the criteria of "high expression", "high affinity", and "high presentation score" were screened. (2) Construction of mRNA-LNP vaccine: The candidate neoantigens obtained in step (1) were encoded by the coding sequence, and a fusion antigen gene TSA was obtained after codon optimization and tandem design. The TSA was cloned into the corresponding site of the expression vector. During cloning, 5′UTR and 3′UTR and His tag coding sequences to enhance translation efficiency were introduced upstream and downstream of the TSA sequence, respectively, to construct the recombinant plasmid pVAX1-M1+2-TSA. Using the plasmid as a template, a linearized DNA transcription template was obtained by PCR amplification using primers with a T7 promoter. The DNA transcription template was transcribed in vitro, capped and tailed in sequence to obtain high-purity mRNA. The high-purity mRNA was then encapsulated with lipid nanoparticles to prepare the mRNA-LNP vaccine. (3) In vitro expression validation of mRNA-LNP vaccine: The mRNA-LNP vaccine prepared in step (2) was used to verify the antigen expression efficiency by transfecting 293T cells in vitro, which confirmed that the vaccine could express and secrete the target antigen efficiently in vitro, thereby completing the functional quality control of the mRNA-LNP vaccine.

2. The screening and verification method according to claim 1, characterized in that, In step (1), when performing mutation identification and bioinformatics analysis on the sample sequencing results, the specific method is as follows: First, compare the genomic DNA sequencing results of tumor tissue and adjacent normal tissue, distinguish and screen out common germline mutations, and retain somatic mutation data that only exist in tumors; then integrate the total RNA transcriptome sequencing results, evaluate the expression level of mutated genes, filter out mutations with low expression levels in tumor tissue, and retain genes with significant differential expression; finally, perform screening and cross-analysis on somatic mutation data and differentially expressed gene data, and finally select the peptide segments encoded by genes with mutations and differential expression.

3. The screening and verification method according to claim 1, characterized in that, In step (1), when performing neoantigen prediction and screening, the specific method is as follows: first, perform functional annotation on the identified somatic mutations, and extract the peptide sequence containing the mutation site based on the annotation results; then evaluate the binding affinity of the mutated peptide sequence to the HLA allele, and screen out peptides with high binding affinity; then predict the antigen presentation ability and screen out candidate neoantigens that simultaneously meet the criteria of "high expression", "high affinity" and "high presentation score".

4. The screening and verification method according to claim 3, characterized in that, The HLA allele was HLA-B*40:01; the affinity of the mutated peptide sequence for binding with the HLA allele was assessed using the NetMHCpan-4.1 algorithm, and the affinity screening criterion was IC50 < 500 nM; the antigen presentation ability was predicted using the MHCflurry 2.0 algorithm.

5. The screening and verification method according to claim 1, characterized in that, In step (2), the complete nucleotide coding sequence of the fusion antigen gene TSA is shown in SEQ ID NO:1; the sequence of the linearized DNA transcription template obtained by PCR amplification is shown in SEQ ID NO:

2.

6. The screening and verification method according to claim 1, characterized in that, In step (2), the expression vector is pVAX1; the 5′UTR and 3′UTR are selected from known sequences that enhance translation efficiency; the encapsulation of high-purity mRNA using lipid nanoparticles has an encapsulation rate of not less than 90%.

7. The screening and verification method according to claim 1, characterized in that, In step (3), the specific method for verifying the in vitro expression of the mRNA-LNP vaccine is as follows: mRNA-LNP vaccine is added to human embryonic kidney 293T cells and then transfected; after 48 hours of transfection, the cells are denatured, gel electrophoresis, transferred to a membrane, blocked, incubated with primary antibody overnight, washed, incubated with secondary antibody for 1 hour, washed, and incubated in the dark for a few seconds before image acquisition.

8. A personalized anti-hepatocellular carcinoma mRNA-LNP vaccine prepared by the screening and verification method according to any one of claims 1-7.

9. The use of the mRNA-LNP vaccine of claim 8 in the preparation of a medicament for stimulating or evaluating an anti-hepatocellular T-cell immune response.

Images