Recombinant construct, CAR-T cell and construction and application of CAR-T cell

Abstract

The invention relates to the field of biotechnology and targeted therapy, in particular to a recombinant construct, a CAR-T cell and construction and application of the CAR-T cell. Experiments verify that THEMIS maintains the effect and mechanism of CAR-T for persistently killing tumor cells. The CAR-T cell overexpressing THEMIS is constructed, compared with a control group, the multiplication capacity and the memory cell forming capacity of the THEMIS-7 * 19 CAR-T cell after antigen stimulation are enhanced, cell factors are in a controllable state, cell apoptosis and depletion are reduced, and the more effective long-term tumor killing capacity can be maintained. In-vivo experiments also prove that the THEMIS-7 * 19CAR-T cell has a stronger anti-tumor effect in a lymphoma-bearing mouse model compared with a control group, and the survival time of the mouse is remarkably prolonged.

Description

Technical Field

[0001] The present invention relates to the fields of biotechnology and targeted therapy, and particularly to a recombinant construct, a CAR-T cell, and their construction and application. Background Art

[0002] In recent years, with the emergence of CAR-T therapy, cancer treatment has entered a transformative period. Currently, many large-scale key CAR-T clinical trials, such as ZUMA-1, JULIET, TRANSCEND, and ELIANA, have demonstrated the safety and effectiveness of CAR-T cell products such as axicabtagene (axi-cel), tisagenlecleucel (tisa-cel), lisocabtagene maraleucel (liso-cel) in the treatment of patients with relapsed or refractory B-cell lymphoma, including diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), mantle cell lymphoma (MCL), and primary mediastinal large B-cell lymphoma (PMBCL). Based on these progressions, CAR-T cell therapy has been extended to the treatment of various hematological malignancies, including acute lymphoblastic leukemia (ALL) and chronic lymphocytic leukemia (CLL), and has achieved varying degrees of progress. However, within one year after receiving CAR-T cell therapy, approximately 50% of LBCL patients, 20-30% of FL patients, and 40% of MCL patients will experience disease progression or recurrence. On the one hand, CAR-T drug resistance can be attributed to antigen escape, that is, the emergence of antigen-negative tumor cells under the selection pressure of CAR-T treatment. On the other hand, recurrence also occurs even when the antigen is positive, suggesting intrinsic factors of CAR-T cells such as T cell exhaustion. Although CAR-T cells can initially migrate to the target cells and exhibit activity, continuous antigen stimulation combined with an immunosuppressive microenvironment will induce the dysfunction of CAR-T cells. Functionally exhausted CAR-T cells show a gradual decline in cytokine production, proliferation, and tumor-killing ability, ultimately leading to disease recurrence.

[0003] With the development of emerging technologies such as multi-omics and CRISPR / Cas9 gene editing technology, we have more means to analyze and understand the differences in gene expression profiles, epigenetic modifications, metabolism, etc. of CAR-T cell products between different efficacy groups, identify the key factors among them, and use technologies such as gene editing to improve the efficacy of CAR-T cells.

[0004] Therefore, further exploring the role and mechanism of CAR-T in killing tumor cells and upgrading CAR-T cells to improve their performance of accurately recognizing and effectively eliminating cancer cells in the long term are the current research hotspots in the industry. Summary of the Invention

[0005] The efficacy of chimeric antigen receptor T-cell (CAR-T) in treating hematological malignancies is limited by factors such as the heterogeneity of patients' own T cells and antigen escape. In recent years, the rapid development of multi-omics technology has enabled us to more comprehensively understand the heterogeneity of CAR-T cell products in different patients. Previously, our team (Cell Discovery, www.nature.com / celldisc, Lei et al. Cell Discovery (2024) 10:5,

[0006] https: / / doi.org / 10.1038 / s41421-023-00625-0) demonstrated that targeted CD19 CAR-T carrying interleukin-7 (IL-7) and C-C motif chemokine ligand 19 (CCL19) (named 7×19CAR-T) has good efficacy in treating large B-cell lymphoma, but there are also recurrence phenomena.

[0007] THEMIS is an important regulator of thymocyte positive selection. Previous studies have found that the expression of THEMIS is closely related to aspects such as the development and differentiation of T cells. There are literature reports that mice with THEMIS knockout have defects in T cell selection, resulting in a decrease in the number of single-positive cells and mature peripheral T cells. THEMIS is only expressed in the T cell lineage and belongs to a small gene family conserved during vertebrate evolution. It is phosphorylated by lymphocyte-specific protein tyrosine kinase (LCK) and possibly ZAP70 after cross-linking with the TCR. Some studies have also found that CAR with 4-1BB as the co-stimulatory domain can induce the recruitment of the THEMIS-SHP1 complex, reduce CD3ζ phosphorylation, and to a certain extent reduce the activation degree of T cells, thereby reducing the secretion of their cytokines. However, it is not yet clear whether THEMIS is involved in regulating key phenotypes such as the memory stem cell characteristics, exhaustion differentiation, and apoptosis of CAR-T cells, thus affecting their long-term efficacy. Therefore, it is necessary to conduct further research.

[0008] In this study, through single-cell omics and other technologies, the differences in the subset composition, gene expression patterns, TCR phenotypes, etc. of four generations of CAR-T cell products in different efficacy groups after tumor stimulation were analyzed and compared. A key CAR-T cell subset marked by THEMIS that was significantly upregulated in the complete remission group was identified, and the role and mechanism of THEMIS in maintaining the persistent killing of tumor cells by CAR-T were verified through experiments.

[0009] The purpose of the present invention is to provide a recombinant construct, CAR-T cells, and their construction and application to solve the problems proposed in the above background technology.

[0010] To solve the above technical problems, the present invention provides the following technical solutions:

[0011] A recombinant construct encoding a chimeric antigen receptor that can simultaneously express IL-7, CCL19, and THEMIS, its fragment or variant.

[0012] Preferably, the chimeric antigen receptor is an anti-CD19 chimeric antigen receptor.

[0013] Preferably, the nucleic acid sequence of the recombinant construct is as shown in SEQ ID NO.5, its fragment or variant.

[0014] Preferably, the amino acid sequence encoded by the recombinant construct is as shown in SEQ ID NO.6, its fragment or variant.

[0015] A pharmaceutical composition comprising a recombinant construct as described in any one of the foregoing and at least one pharmaceutically acceptable carrier.

[0016] A CAR-T cell comprising a recombinant construct as described in any one of the foregoing.

[0017] Preferably, the CAR-T cell is autologous, allogeneic or xenogeneic.

[0018] Use of a recombinant construct or a pharmaceutical composition as described in any one of the foregoing or a CAR-T cell as described in any one of the foregoing in the preparation of a kit for diagnosing / treating a malignant tumor.

[0019] Use of a recombinant construct or a pharmaceutical composition as described in any one of the foregoing or a CAR-T cell as described in any one of the foregoing in the preparation of a medicament for diagnosing / treating a malignant tumor.

[0020] Preferably, the malignant tumor is a hematological malignancy.

[0021] The beneficial effects of the present invention are as follows: The proliferation ability and memory cell formation ability of THEMIS-7×19 CAR-T cells are enhanced after antigen stimulation, and the cytokines are in a controllable state, with reduced apoptosis and exhaustion, and can maintain a more effective long-term tumor killing ability. In vivo experiments also prove that THEMIS-7×19 CAR-T cells have a stronger anti-tumor effect than the control group in a lymphoma-bearing mouse model, significantly prolonging the survival time of the mice. Description of the Drawings

[0022] Figure 1-1 Cell subset clustering analysis and differential gene display: (A) The UMAP plot shows 11 cell clusters in the 7×19 CAR-T cell infusion product and the naming of each subgroup; (B) The dot plot shows the expression levels of the top 10 differential genes in different cell types among different cell subsets.

[0023] Figure 1-2 Differences in the distribution of T cell subsets between different efficacy groups: (A) The UMAP plot and bar proportion plot show the distribution preferences of each cell subset in the 7×19 CAR-T cell infusion product; (B) The number distribution of different T cell subsets between different efficacy groups; (C) The Ro / e analysis shows the distribution preferences of 10 T cell subsets between different efficacy groups. Figure 1-3 Functional characteristics among T cell subsets of CAR-T products in different clinical efficacy groups: (A) The heat map shows the cell functional characteristics between different efficacy groups; (B) The heat map shows the functional characteristics between different T cell subsets; (C) The heat map shows the cytokine expression patterns between different T cell subsets.

[0024] Figure 1-4Transcription factor expression levels and transcriptional activities among different T cell subsets: (A) Heatmap showing the transcription factor expression levels among different T cell subsets; (B) Heatmap showing the transcriptional regulatory activities among different T cell subsets;

[0025] Figure 1-5 Gene module analysis and enrichment analysis: (A) Heatmap showing the results of hotspot gene module analysis; (B) GO enrichment analysis results of modules 1, 7, and 11;

[0026] Figure 1-6 Metabolic characteristics among different T cell subsets: (A) Heatmap showing the metabolic pathways (left) and metabolite abundances (right) among different T cell subsets; (B) Dot plot showing the metabolic pathway activities among different T cell subsets;

[0027] Figure 1-7 Interaction analysis between different T cell subsets and tumor cells (B cells): (A) Interaction intensity between cell subsets in different efficacy groups; (B) Interaction pair differences between tumor cells and each T cell subset among different efficacy groups;

[0028] Figure 1-8 Analysis of TCR clonotype diversity among different T cell subsets: (A) Pie chart showing the TCR clonotype types and numbers in different efficacy groups and different T cell subsets; (B) Heatmap and bar chart showing the shared TCR clonotype numbers among T cell subsets in different efficacy groups;

[0029] Figure 1-9 Differentiation trajectory analysis of each T cell subset: (A) Pseudotime trajectory plots of different T cell subsets projected in two-dimensional space. Left: Pseudotime trajectory plot shown by pseudotime values; Middle: Distribution of 7 states on the pseudotime trajectory plot; Right: Trajectory distribution of different T cell subsets on the pseudotime trajectory plot; (B) Distribution of 7 states on the pseudotime trajectory plots corresponding to different T cell subsets; (C) Heatmap of representative genes along the pseudotime axis; (D) RNA velocity analysis plot indicating the differentiation directions of different T cell subsets;

[0030] Figure 1-10 Differences in surface protein expression among different T cell subsets: (A) Violin plot showing the expression levels of 14 surface proteins among different efficacy groups and different T cell subsets; (B) Differences in surface protein expression related to memory and exhaustion phenotypes among different efficacy groups.

[0031] Figure 2-1 Functional characteristic analysis of the CD8+T_THEMIS subset: (A) Volcano plot showing the distribution of upregulated and downregulated genes; (B) Bar chart and bubble chart showing the GO and KEGG enrichment results of upregulated and downregulated genes;

[0032] Figure 2-2Functional analysis of different differentiation stages of CD8+T_THEMIS subsets: (A) Results of differential gene expression and functional enrichment; (B) Heatmap and violin plot showing functional characteristics in the early and late stages of differentiation; (C) Heatmap showing the expression levels of transcription factors in the early and late stages of differentiation;

[0033] Figure 2-3 Identification of prognosis-related genes in CD8+T_THEMIS subsets using public databases: (A) Expression of SOS1, THEMIS, PDE3B, CDK6, and BACH2 genes in CAR-T cell products of different efficacy groups; (B) Expression of SOS1, THEMIS, PDE3B, CDK6, and BACH2 genes in CD8 + T cells of different efficacy groups of CAR-T cell products. Figure 3-1 Changes in the proportion of CAR after THEMIS knockout: (A) Schematic diagram of the 7×19CAR structure; (B) Detection of the transduction efficiency of 7×19CAR cells by flow cytometry; (C) Detection of the knockout efficiency at the transcriptional and translational levels; (D) Detection of the changes in the proportion of T cells expressing CAR before and after THEMIS knockout by flow cytometry;

[0034] Figure 3-2 Flow cytometry detection of CD8 / CD4 flow cytometry plots (A) and proportion changes (B) of 7×19CAR-T cells after THEMIS knockout;

[0035] Figure 3-3 Proliferation ability of 7×19CAR-T cells after THEMIS knockout after antigen stimulation: (A) Proliferation of 7×19CAR-T cells with or without tumor antigen stimulation (CFSE-labeled cells in advance, with the strongest fluorescence intensity at 0h. As the cells divide, the fluorescence intensity of the daughter cells gradually decreases. Therefore, the faster the proliferation, the lower the fluorescence intensity and the more left-shifted the peak); (B) Amplification curve of 7×19CAR-T cells under tumor antigen stimulation;

[0036] Figure 3-4 Luciferase-based cytotoxicity assay and cytokine secretion levels: (A) Cytotoxic effects of 7×19CAR-T cells on Jeko-1 and Raji cell lines before and after knocking out THEMIS at effector-to-target ratios of 20:1, 10:1, 5:1, and 2.5:1; (B) Cytokine levels in the supernatant after killing tumor cells by the two groups at an effector-to-target ratio of 20:1, *P<0.05, **P<0.01, ns P>0.05;

[0037] Figure 3-5Long-term tumor killing ability of 7×19 CAR-T cells before and after THEMIS knockout: (A) Comparison of long-term tumor killing ability of 7×19 CAR-T cells before and after THEMIS knockout when the effector-to-target ratio is 1:2; (B) Comparison of long-term tumor killing ability of 7×19 CAR-T cells before and after THEMIS knockout when the effector-to-target ratio is 1:1;

[0038] Figure 3-6 Flow cytometry detection of changes in memory, exhaustion, and apoptosis phenotypes of 7×19 CAR-T cells after antigen stimulation before and after THEMIS knockout: (A) Flow cytometry results of the memory phenotype ratio in the two groups before and after knockout with or without antigen stimulation; (B) Flow cytometry results of the exhaustion-related phenotype ratio in the two groups before and after knockout under repeated antigen stimulation; (C) Flow cytometry results of the apoptosis ratio in the two groups before and after knockout under repeated antigen stimulation; (D) Comparison of the differences in the memory, exhaustion, and apoptosis phenotype ratios of 7×19 CAR-T cells after antigen stimulation before and after THEMIS knockout, *P<0.05, **P<0.01, ns P>0.05;

[0039] Figure 3-7 Comparison of cytokine secretion levels of 7×19 CAR-T cells before and after THEMIS knockout under repeated antigen stimulation: (A) Cytokine secretion levels in the two groups of cells after the 1st and 3rd antigen stimulations; (B) Degrees of cytokine secretion attenuation in the two groups after repeated antigen stimulation, *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns P>0.05.

[0040] Figure 4-1 Verification of THEMIS expression level: (A) Schematic diagram of the THEMIS-7×19 CAR structure; (B) Flow cytometry detection of the transduction efficiency of 7×19 CAR and THEMIS-7×19 CAR cells; (C) Flow cytometry detection of the intracellular THEMIS protein expression levels in the THEMIS-7×19 CAR-T group and the control group;

[0041] Figure 4-2 Flow cytometry detection of the CD8 / CD4 flow cytometry plot (A) and ratio change (B) of THEMIS-7×19 CAR-T cells;

[0042] Figure 4-3 Proliferation ability of 7×19 CAR-T cells after antigen stimulation after overexpression of THEMIS: (A) Proliferation (CFSE) of 7×19 CAR-T cells in the overexpression group and the control group with or without tumor antigen stimulation; (B) Amplification curves of 7×19 CAR-T cells in the overexpression group and the control group under tumor antigen stimulation;

[0043] Figure 4-4Tumor killing ability of 7×19 CAR-T cells after overexpression of THEMIS: (A) Killing effects of 7×19 CAR-T cells in the overexpression group and the control group on Jeko-1 and Raji cell lines at effector-to-target ratios of 20:1, 10:1, 5:1, and 2.5:1; (B) Analysis of the long-term tumor killing ability of 7×19 CAR-T cells after overexpression of THEMIS by RTCA technology, * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001, **** indicates P<0.0001, ns indicates P>0.05;

[0044] Figure 4-5 Flow cytometry detection of changes in memory, exhaustion, and apoptosis phenotypes of 7×19 CAR-T cells after antigen stimulation following overexpression of THEMIS: (A) Proportions of memory phenotypes of 7×19 CAR-T cells in the overexpression group and the control group with or without antigen stimulation; (B) Proportions of exhaustion-related phenotypes of 7×19 CAR-T cells in the overexpression group and the control group under repeated antigen stimulation; (C) Apoptosis proportions of 7×19 CAR-T cells in the overexpression group and the control group under repeated antigen stimulation; (D) Differences in proportions of memory, exhaustion, and apoptosis phenotypes of 7×19 CAR-T cells in the overexpression group and the control group after repeated antigen stimulation, * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001, **** indicates P<0.0001, ns indicates P>0.05;

[0045] Figure 4-6 Cytokine secretion levels of 7×19 CAR-T cells after overexpression of THEMIS under repeated antigen stimulation: (A) Cytokine secretion levels of cells in the overexpression group and the control group after the first and third antigen stimulations; (B) Degrees of attenuation of cytokine secretion in the overexpression group and the control group after repeated antigen stimulation, * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001, **** indicates P<0.0001, ns indicates P>0.05;

[0046] Figure 4-7 Antitumor effect of 7×19 CAR-T cells after overexpression of THEMIS in vivo: A. Flow chart of animal experiment; B. Fluorescence imaging of mice (7×19 CAR-T as negative control); C. Quantitative results of tumor fluorescence signals and comparison of fluorescence intensities between the two groups on the 26th day; D. Survival analysis, *P<0.05, **P<0.01.

[0047] Figure 5-1 Plasmid map of the target plasmid corresponding to 7×19 CAR-T, pLenti-EF1a-wPRE, 9723 base pairs;

[0048] Figure 5-2 Map of the target gene of THMEIS, THMEIS_CDS, 1926 base pairs;

[0049] Figure 5-3 The plasmid map of the finally synthesized THEMIS-targeted plasmid C2020HGHG0-2, T2A+THEMIS_CAR, has 11,703 base pairs.

[0050] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. Detailed implementation manners

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0052] Table 1 Abbreviations

[0053]

[0054]

[0055]

[0056] Part I Single-cell sequencing reveals key factors for maintaining the efficacy of CAR-T therapy for lymphoma in four generations

[0057] Large B cell lymphoma (LBCL) is a type of aggressive hematological malignancy, accounting for nearly 30% of all non-Hodgkin's lymphoma (NHL) cases. Among them, diffuse large B cell lymphoma (DLBCL) not otherwise specified (NOS) is the most common. The latest World Health Organization classification classifies high-grade B cell lymphoma (with MYC and BCL2 and / or BCL6 rearrangements) or high-grade B cell lymphoma not otherwise specified into the LBCL category. Although most LBCL patients can be cured by standardized chemotherapy regimens, there are still nearly 40% of patients who are refractory or relapsed (R / R). Some of the R / R patients can be cured by second-line treatment, but about 10% to 15% of the patients show primary drug resistance, and most patients eventually die from the disease.

[0058] CAR-T cell therapy is a cell therapy method that targets tumor antigens by expressing antibody-based fusion proteins on the T cell membrane. The initially approved products include axi-cel, tisa-cel, and liso-cel. Updated results from the pivotal single-arm phase II trials ZUMA-1, JULIET, and TRANSFORM showed that this therapy demonstrated good efficacy in patients with refractory / relapsed LBCL. However, within one year after receiving CAR-T cell therapy, approximately 50% of LBCL patients still experience disease progression or relapse. The differences in CAR-T treatment outcomes among different patients have prompted researchers to continuously explore the key factors leading to suboptimal or failed treatment. Currently, the molecular mechanisms underlying the development of acquired resistance to CAR-T cell therapy remain unclear. Although the theory of antigen loss can explain the results of disease relapse caused by CD19 negativity in some patients after receiving CD19 CAR-T therapy, the mechanism of relapse in CD19-positive patients remains unknown.

[0059] Currently, there is a lot of evidence indicating that multiple characteristics of T cells can significantly affect the efficacy of CAR-T cell therapy, especially T cell exhaustion. To deeply explore the relationship between T cell characteristics and CAR-T cell efficacy, we used scRNA-seq, TCR sequencing, and CITE-seq technologies to study the CAR-T cell infusion products of 10 R / R LBCL patients who had received four generations of CAR-T cell therapy, aiming to explore the possible key factors maintaining long-term treatment responses in the infusion products. We found that there was a highly enriched subset of CD8+ T_THEMIS cells in the cell products of CR group patients. This subset had good cell proliferation ability, highly expressed transcription factors related to memory stemness, and had unique differentiation paths and functional characteristics. Additionally, through validation using external databases, it was found that THEMIS, as a key marker gene in this subset, was correlated with the efficacy of CAR-T cell therapy, providing sufficient evidence for our subsequent basic experimental verification.

[0060] 1 Experimental materials

[0061] 1.1 Cell lines and samples

[0062] 1.1.1 Cell lines

[0063] Raji: Human Burkitt's lymphoma cell line, purchased from ATCC.

[0064] 1.1.2 Patient samples

[0065] This study was approved by the Ethics Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine. All samples were from the previous phase I and expanded-phase clinical trials conducted by our team to evaluate the safety and efficacy of 7×19 CAR-T cell therapy.

[0066] 1.2 Main experimental reagents

[0067] Table 2

[0068]

[0069]

[0070] 1.3 Main instruments and equipment

[0071] Table 3

[0072]

[0073] 2 Experimental methods

[0074] 2.1 Cell culture

[0075] Raji cells are cultured in RPMI 1640 complete medium, and the CAR-T cell products after resuscitation are cultured in X-VIVO complete medium. The medium is prepared as follows:

[0076] 1) RPMI 1640 complete medium (1% penicillin-streptomycin + 10% FBS + RPMI 1640 medium);

[0077] 2) X-VIVO complete medium (1% penicillin-streptomycin + 10% Gibco serum + 1% HEPES + 1% sodium pyruvate + 1% glutamine + 1% non-essential amino acids + 300 IU / mL rhIL-2 + 5 ng / mL IL-7 + 5 ng / mL IL-15).

[0078] 2.2 Cell resuscitation

[0079] 1) Pre-warm the medium in a 37 °C water bath. At the same time, prepare a 15 mL centrifuge tube and add 2 - 3 mL of pre-warmed medium;

[0080] 2) Take out the cells to be resuscitated from the -80 °C refrigerator or liquid nitrogen, and quickly thaw the cells in a 37 °C water bath (try to control within 2 min);

[0081] 3) Carefully add the cells to the prepared pre-warmed medium, mix well and centrifuge at 1500 rpm for 5 min;

[0082] 4) Carefully remove the supernatant, add an appropriate amount of pre-warmed medium according to the cell amount, mix well and transfer to a 6-well plate or T25 cell culture flask;

[0083] 5) Observe under a microscope and then place in a 37 °C, 5% CO2 incubator for culture.

[0084] 2.3 Antigen stimulation of CAR-T cell products

[0085] 1) At an effector-to-target ratio of 1:1, take 1×10 6 Raji cells treated with mitomycin and CAR-T cell products with adjusted cell status after resuscitation were co-incubated in an incubator at 37°C and 5% CO2;

[0086] 2) After 24 hours, measure the CAR ratio and cell viability of each group of CAR-T cell products, and perform single-cell sequencing analysis subsequently;

[0087] 2.4 Single-cell sequencing library construction

[0088] 1) Cell preparation: Dilute the cells to an appropriate concentration, generally 2 - 2.5×10 5 cells / mL;

[0089] 2) Single-cell isolation and labeling: Inject the cell suspension into the SCOPE-chip TM microfluidic chip, and complete the isolation of single cells according to the principle of "Poisson distribution". The cells fall into the specially customized chip micro-wells under the action of gravity, ensuring that only 1 cell falls into each micro-well. Then, millions of magnetic beads carrying unique cell barcodes are added to the chip micro-wells, ensuring that only 1 magnetic bead falls into each micro-well. After cell lysis, the magnetic beads with unique molecular identifiers (UMIs) capture mRNA by binding to the poly(A) tail on mRNA, and label the cells and mRNA;

[0090] 3) Reverse transcription and amplification: Collect the magnetic beads in the chip, reverse transcribe the mRNA captured by the magnetic beads into cDNA

[0091] and amplify;

[0092] 4) Single-cell sequencing library construction: After fragmenting the cDNA, dilute each library to 4 nM and pool them for sequencing. Sequencing is performed on the Illumina Novaseq 6000 using 150 bp paired-end reads. The raw reads are processed with fastQC and fastp to remove low-quality reads. The poly(A) tail and adapter sequences are removed by cutadapt. After quality control, the reads are mapped to the reference genome GRCh38 using STAR. Statistical analysis of gene expression levels and UMI counts is performed through the FeatureCounts function. An expression matrix file for subsequent analysis is generated based on gene expression levels and UMI counts.

[0093] 2.5 Quality control, dimensionality reduction, and clustering

[0094] 1) Before analysis, cells were filtered by UMI counts less than 30,000 and gene counts between 200 and 5,000, and then cells with mitochondrial content exceeding 20% were removed;

[0095] 2) After filtering, functions in Seurat (v2.3) were used for dimensionality reduction and clustering. Then, the NormalizeData and ScaleData functions were used to normalize and scale all gene expressions, and the FindVariableFeautres function was used to select the top 2,000 highly variable genes for PCA analysis. For the top 20 principal components, FindClusters was used to divide them into multiple clusters. The batch effects between samples were removed by the Harnomy algorithm;

[0096] 3) Uniform manifold approximation and projection (UMAP) dimensionality reduction analysis was performed for visualization.

[0097] 2.6 Differentially expressed genes (DEGs) analysis and cell type annotation

[0098] To identify differentially expressed genes, we used the FindMarkers function in the Seurat package based on the Wilcox non-parametric test method, and selected genes expressed in more than 10% of the cells in a cluster and with an average log2(FoldChange) > 0.25 as DEGs. For the cell type annotation of each cluster, we combined the expression of classical markers found in the DEGs with the literature, and used the heatmaps, dot plots, and violin plots generated by Seurat to display the expression of markers for each cell type, and manually filtered out doublets.

[0099] 2.7 Pathway enrichment analysis

[0100] To investigate the potential functions of DEGs, gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) analyses using the clusterProfiler package were performed. Pathways with P < 0.05 were considered significantly enriched. The GO gene sets included molecular function (MF), biological process (BP), and cellular component (CC) categories as references. Based on the interactions of known genes with relevant GO terms in StringDB v1.22.0, protein–protein interactions of DEGs in each cluster were predicted.

[0101] 2.8 Gene regulatory network inference (single-cell regulatory network inference and clustering, SCENIC)

[0102] To analyze the transcription factor regulatory network, we performed SCENIC analysis on the scRNA-seq expression matrix through AnimalTFDB. The regulatory network was predicted by the GENIE3 package based on the co-expression of regulators and targets. We used the RcisTarget package to search for transcription factor binding motifs in the data. Genes involved in predicting the regulatory network were defined as a gene set, and the AUC value was calculated by the AUCell package to evaluate the activity of the regulatory network in cells.

[0103] 2.9 Cell developmental trajectory analysis

[0104] 2.9.1 Pseudotime analysis

[0105] 1) We analyzed the developmental trajectory of T cells using the Monocle 2 package. Monocle 2 mainly calculates the gene expression changes between different cells through machine learning based on the expression patterns of selected feature genes, and sorts individual cells to simulate the dynamic changes of cell development;

[0106] 2) The total length of the cell developmental trajectory is defined according to the total amount of transcriptional changes experienced by the cell from the starting state to the ending state. After the cells are arranged in order, the data can be dimensionally reduced by the reverse graph embedding algorithm in Monocle 2, and the function used is reduceDimension;

[0107] 3) Trajectory construction is visualized through the plot_cell_trajectory function, projecting the expression data into a lower-dimensional space;

[0108] 4) Visualization of gene expression along pseudotime: The function plot_genes_in_pseudotime can plot the dynamic changes in the expression levels of individual genes along the pseudotime. To further study genes following similar kinetic trends, the function plot_pseudotime_heatmap was used to generate a heatmap to visualize gene modules that co-vary over different times on the pseudotime axis. Plot_pseudotime_heatmap uses a CellDataSet object (a subset containing important genes) to generate smooth expression curves.

[0109] 2.9.2 RNA velocity analysis

[0110] In RNA velocity analysis, we used BAM files containing T cells and the reference genome GRCh38 (hg38). Analyses were performed in Python with default parameters using velocyto (v0.2.3) and scVelo (v0.17.17) versions. The results were projected onto the UMAP plot generated by Seurat clustering analysis to ensure consistent visualization.

[0111] 2.10 Intercellular interaction analysis

[0112] CellPhoneDB performs cell-cell interaction analysis based on receptor-ligand interactions between two cell types. The cluster labels of all cells were randomly permuted 1000 times to calculate the null distribution of the average ligand-receptor expression levels in the interacting clusters. Single ligand or receptor expression was thresholded using a cutoff value based on the average log gene expression distribution of all genes in all cell types. Significant intercellular interactions were defined as P < 0.05 and finally visualized using the circlize (v0.4.10) package.

[0113] 2.11 Metabolic signature analysis

[0114] 1) Use scMetabolism (v0.2.1) to quantify single-cell metabolic activity: By collecting metabolic-related pathways in the KEGG database and calculating the metabolic pathway enrichment scores based on the VISION algorithm. The scores for specific pathways were visualized using the FeaturePlot / VlnPlot functions and pheatmap in Seurat;

[0115] 2) Single-cell metabolic fluxomes and abundances analysis (scFEA): scFEA (v1.1.2) is a computational method based on a novel probabilistic model and flux balance constraints for inferring cellular metabolic fluxomes and metabolite abundances from scRNA-seq data. In this study, scFEA was run on the Python platform to calculate the metabolic fluxomes and metabolite abundances of T cell types. After estimating the metabolic fluxes and metabolite abundances, 70 metabolic modules and all metabolites were selected and visualized using a heatmap.

[0116] 2.12 Functional gene module analysis

[0117] We used the Hotspot tool to identify functional gene modules and demonstrate the heterogeneity within T cell subsets. Using the "danb" model, the top 500 genes with the highest autocorrelation z-scores were selected for module identification, and then the create_modules function was used to identify modules, setting min_gene_threshold = 15 and fdr_threshold = 0.05. Module scores were calculated using the calculate_module_scores function.

[0118] 2.13 TCR repertoire sequencing and analysis

[0119] The vdj pipeline of Cell Ranger (v4.0.0) was used with the GRCh38 reference genome for TCR sequence alignment and annotation to obtain the TCR repertoire containing clonotype frequencies and barcode information. For TCRs, only cells with one valid TCR alpha chain (TRA) and one valid TCR beta chain (TRB) were retained for further analysis. Each unique TRA(s)-TRB(s) combination was defined as a clonotype. If a clonotype was present in at least two cells, the cells containing that clonotype were considered clonal cells, and the number of cells containing such a clonotype represented the degree of clonality of that clonotype.

[0120] 2.14 CITE-seq analysis

[0121] 1) Library construction: Single-cell suspensions were labeled according to the official protocol of Biolegend (https: / / www.biolegend.com / en-us / protocols / totalseq-a-dual-index-protocol). Subsequently, the cells were diluted to an appropriate concentration using PBS. Then, the cells were loaded onto a microwell chip through the Singeron Matrix single-cell processing system. The barcode beads were removed from the microwell chip, and then the mRNA and oligonucleotides captured by the barcode beads were reverse-transcribed to obtain cDNA for PCR amplification. The amplified cDNA was fragmented and indexed with sequencing adapters to generate a single-cell transcriptome library. The purified cDNA supernatant was used as a template for CITE-seq library construction. Each library was diluted to a concentration of 4 nM, mixed, and sequenced on a Novaseq 6000 (Illumina) with 150 bp paired-end reads;

[0122] 2) Data analysis: The raw data was processed by CeleScope (v1.15.0) to generate gene expression profiles. Barcodes and UMIs were extracted and corrected from the R1 reads. Adapter sequences and polyadenylation tails were removed from the R2 reads, and then the CITE-seq tags were extracted from the R2 reads. Finally, the same reads that matched the corresponding transcriptome cell barcodes were extracted, and the successfully assigned reads were grouped together with the same cell barcodes, UMIs, and CITE-seq tags to generate a CITE-seq expression matrix for further analysis.

[0123] 2.15 Acquisition of public data cohort data

[0124] We downloaded the gene expression data of all samples in the GSE223655 cohort and the corresponding clinical information from the GEO database (https: / / www.ncbi.nlm.nih.gov / gds / ?term=GSE223655) for external database validation.

[0125] 2.16 Statistical analysis

[0126] All data in this study were statistically analyzed using Graphpad Prism (version 10). All data were tested for normality using the Shapiro-Wilk test, and then the corresponding statistical test methods were selected according to whether they conform to normality. For data that conform to a normal distribution, an independent samples t-test was used for comparison between two groups. For data that do not conform to a normal distribution, the Mann-Whitney U test was used. A paired t-test was used for comparison of different conditions in the same group of samples. For comparison among multiple groups, one-way analysis of variance (ANOVA) and Bonferroni post hoc tests were used for normal distribution data. For non-normal distribution data, the Kruskal-Wallis test and Dunn post hoc tests were used. P < 0.05 was considered statistically significant, * for P < 0.05, ** for P < 0.01, *** for P < 0.001, **** for P < 0.0001, and ns for P > 0.05.

[0127] 3 Experimental Results

[0128] 3.1 Single-cell multi-omics analysis of the characteristics of 7×19 CAR-T infusion products from patients with B-cell lymphoma

[0129] 3.1.1 Sample information and pretreatment

[0130] To explore the key factors maintaining the long-term efficacy of fourth-generation CAR-T cells (7×19 CAR-T) in treating CD19-positive LBCL patients, we performed in vitro tumor antigen stimulation on the CAR-T cell infusion products of 10 adult patients (median age: 58.50 years, range: 42 to 72 years) who previously participated in the phase I and expanded-phase clinical trials of 7×19 CAR-T cell therapy for refractory and relapsed LBCL, and carried out single-cell sequencing analysis. The patient information characteristics are shown in Table 1. According to the evaluation results of the patients 3 months after CAR-T treatment, they were divided into two groups: 8 patients achieved complete remission or partial response (PR) after treatment and belonged to responders; while the other 2 patients had disease progression and belonged to non-responders (NR). Responders were further divided into patients who achieved durable complete remission (CR, n = 4) and patients who relapsed during the trial (RL, n = 4). Our goal was to identify the key cell subsets that could maintain the long-term effectiveness of fourth-generation CAR-T cell therapy and analyze the core genes closely related to clinical efficacy. Since the data of the 7×19 CAR-T cell products at baseline alone could not fully reveal the early response mechanism of the cells after activation, we mixed the 7×19 CAR-T cell products with CD19 in a 1:1 effector-to-target ratio in vitro +Raji cells were co-cultured for 24 hours to induce their antigen-specific activation. Subsequently, we performed scRNA-seq analysis on the CAR-T cell infusion products of these patients using the microplate method, and combined with CITE-seq technology, we analyzed 14 surface protein markers, including T cell markers (CD4 and CD8), differentiation and subtype markers (CD45RO, CD45RA, CCR7 and CD62L), activation markers (HLA-DR, CD69, CD38 and 4-1BB), and co-inhibitory markers (PD-1, CTLA-4, LAG-3 and TIGIT).

[0131] Table 4. Basic information and treatment responses of adult patients included

[0132]

[0133] 3.1.2 Single-cell sequencing reveals the main cell types in 7×19 CAR-T infusion products

[0134] After excluding low-quality cells and potential doublets, we analyzed the gene expression profiles of 126,813 cells in total and used UMAP to perform dimensionality reduction clustering on the differentially expressed genes of the cells to cluster all cells. After clustering, we obtained a list of differentially expressed genes in each subgroup, and according to existing literature reports and the definitions of each gene, we analyzed in detail the functions of several genes with the most obvious differences in each subgroup, and then precisely identified and named each cell subgroup.

[0135] First, we removed the batch effects between 10 samples through the Harmony algorithm, and then according to the clustering situation and differentially expressed genes, we determined 11 different cell clusters ( Figure 1-1 A), mainly two main cell types: B cells, namely Raji cells (50,213) in the co-culture system, and T cells (76,600). According to the differential gene expression in each subgroup, we further subdivided the T cells, and a total of 10 subgroups with different transcriptional characteristics were divided, and the subgroup was named according to the specific gene with obvious differences and most related to the T cell function in each subgroup. To evaluate the gene expression patterns of different T cell subgroups in the 7×19 CAR-T infusion products of patients in different efficacy groups after tumor cell stimulation, we plotted dot plots showing the top 10 differentially expressed genes in each cell subgroup ( Figure 1-1 B).

[0136] 3.1.3 Differences in the distribution of T cell subgroups between different clinical efficacy groups

[0137] To further explore the key factors maintaining the long-term efficacy of 7×19 CAR-T therapy, we compared the differences in the distribution ratios of cell subsets between different clinical efficacy groups. The results showed that the proportions and numbers of CD8+T_THEMIS, CD8+T_MKI67, and CD8+T_GZMA subsets were higher in the CR group, while the proportions and numbers of CD4+T_IL2 and CD4+T_CTLA4 subsets were increased in the RL and NR groups ( Figure 1-2 A, B). Due to the limited sample size, traditional statistical methods did not observe the differences in the above subset proportions (P>0.05). We further used the Ro / e analysis to evaluate the distribution characteristics of the target cell population in tissues, aiming to quantify the enrichment or depletion of specific cell clusters in specific groups by comparing the observed cell numbers with the expected cell numbers (i.e., the Ro / e index, Ro / e>1 indicates enrichment of the cell cluster in the subgroup; Ro / e<1 indicates depletion). The results showed that the CD8+T_THEMIS subset was highly enriched in the CR group and might be the most relevant special T cell subset affecting the efficacy of 7×19 CAR-T therapy ( Figure 1-2 C).

[0138] 3.1.4 Evaluation of functional characteristics among different clinical efficacy groups and T cell subsets

[0139] After evaluating the distribution preferences of T cell subsets among different efficacy groups based on the Ro / e index, we further used the Ucell scoring method to systematically analyze the gene expression patterns and functional enrichment characteristics among different efficacy groups and each T cell subset to identify key subsets. The results showed that the 7×19 CAR-T infusion products in the CR group had higher scores in cell proliferation, tissue retention, and lipid metabolism. Although the RL group showed more prominent T cell toxicity and inflammatory effects, its T cells had higher exhaustion scores, which might reveal the mechanism of disease recurrence in patients after cell therapy. In addition, consistent with the clinical efficacy, the NR group had obvious immunosuppressive and T cell exhaustion characteristics ( Figure 1-3 A). To deeply reveal the possible key subsets causing the above results, we evaluated the functional enrichment characteristics of each T cell subset on this basis. The results showed that the CD8+T_TEHMIS subset highly enriched in the CR group showed better cell proliferation ability and significantly enriched the activities of signal pathways related to T cell proliferation and effector functions such as FOXO, IL2-Stat, TCR, and MAPK; while the CD4+T_IL2 and CD4+T_CTLA4 subsets enriched in the RL and NR groups, although having good cell proliferation and antigen response abilities, were more prone to apoptosis at the same time ( Figure 1-3 B).

[0140] To clearly define the functional characteristics of these T cell subsets at the transcriptional level in multiple aspects, we further compared the expression levels of key immune-related molecules in each subset, including effector (GZMB, IFNG, CCL3, PRF1, TNF), regulatory (IL-4, IL-10, IL-13, IL-22), stimulatory (CSF2, IL-2, IL-5, IL-8, IL-9, IL-21), inflammatory (IL-17A, IL-17F), and chemotactic (CCL4, CCL5, CCL20, CXCL10) related factors. The results showed that the CD8+T_THEMIS subset enriched in the CR group had relatively high expression levels of cytokines such as TNF, IL-13, and IL-15, but relatively low gene expression of cytokines such as IFNG and CSF2( Figure 1-3 C). Notably, in the CD4+T_CTLA4 subset enriched in the RL and NR groups, although the expression levels of most effector and regulatory cytokines were upregulated, the excessive or long-term continuous high-level secretion of IL-10 and IFNG would drive the transformation of T cells towards an exhausted phenotype, which might be one of the reasons for the high apoptosis and exhaustion characteristics of this subset.

[0141] From the above, we evaluated the functional characteristics and cytokine expression pattern differences in different clinical efficacy groups and each T cell subset.

[0142] 3.1.5 Transcription Factor Expression Differences among Different T Cell Subsets

[0143] To deeply explore the key "driver genes" affecting the long-term efficacy in the 7×19 CAR-T cell product and lay a foundation for subsequent exploration of molecular mechanisms, we analyzed the expression levels and regulatory activities of transcription factors in each T cell subset using SCENIC analysis, and finally screened out transcription factors with significant regulatory intensity and core roles. The results showed that the expression levels( [[ID= A) and regulatory activities( ​ B) of transcription factors PBX3, FOXO1, BACH1, MEF2A, CERS6, and FOXP1 in the CD8+T_THEMIS subset were higher than those in the other subsets. Notably, except for FOXP1, the above transcription factors were highly expressed and transcribed only in the CD8+T_THEMIS subset. In the CD4+_CTLA4 subset, transcription factors GATA3, MAF, and RORC were relatively highly expressed. Interestingly, GATA3 is a characteristic transcription factor of Th2 cells, and combined with ​ C in which the CD4+_CTLA4 subset highly expressed cytokines such as IL-4, IL-5, IL10, and IL-13, we found that the gene expression pattern of this subset was highly similar to that of Th2 cells.

[0144] In summary, we compared the differences in transcription factor expression and activity among different T cell subsets at the single-cell transcriptome level and found that FOXO1 and FOXP1, which are related to CAR-T cell stemness and expansion, were highly expressed in the CD8+ T_THEMIS subset, while the CD4+_CTLA4 subset highly expressed the Th2 cell characteristic transcription factor GATA3.

[0145] 3.1.6 T cell gene module analysis of 7×19 CAR-T cell products

[0146] To further explore the gene regulatory network among T cells in 7×19 CAR-T cell products, we further used the hotspot method to analyze the co-expression patterns of genes.

[0147] After converting and clustering gene relationships, we obtained 13 functional modules ( ​ A). Among them, 3 modules (module 1, 7, and 11) were enriched in genes related to T cell function. In this regard, we performed GO gene enrichment analysis on these 3 modules for their functions. The results showed that module 1 was closely related to T cell development, activation, and differentiation, while modules 7 and 11 were related to the secretion and response of cytokines and chemokines and the Jak-Stat pathway ( ​ B).

[0148] From the above, we used the hotspot method to find that 3 gene modules were related to T cell development, activation, and effector functions and might play an important role in the tumor killing of 7×19 CAR-T cells.

[0149] 3.1.7 Metabolic characteristics of different T cell subsets

[0150] Different T cell subsets can adapt their metabolism according to their energy requirements and the surrounding microenvironment. For example, T cells rely on mitochondrial pathways with the least nutrient uptake requirements, such as the tricarboxylic acid cycle and oxidative phosphorylation. After being activated, effector T cells will undergo metabolic reprogramming, shifting from mitochondrial reactions to glycolysis and glutaminolysis. Various metabolic pathways involve the uptake of different nutrients to support T cell function and survival. Therefore, we analyzed the changes in metabolic flux and metabolite abundance through scMetabolism and scFEA to further understand the metabolic regulation and microenvironmental factors among different efficacy groups and each T cell subset from a metabolic perspective and reveal their potential metabolic regulatory points. ​In A (left), it can be found that the metabolic pathway of converting adenosine monophosphate (AMP) to adenine in the CD8+T_THEMIS subset is more active than that in other subsets, while the metabolic pathway of converting glutathione (GSH) to glutamate is downregulated compared to other subsets. Additionally, overall analysis reveals that in the CR group, the metabolic pathway of acetyl-CoA converting to farnesyl pyrophosphate (FPP) in each T cell subset is more active compared to the RL and NR groups. Correspondingly, ​ In A (right), it can be seen that the abundances of glucose-6-phosphate (G6P) and glutathione in the CD8+T_THEMIS subset are higher than those in other subsets, while the metabolic abundance of glutamate is relatively lower, which is consistent with the changes in its metabolic pathway. ​ B also reveals enhanced D-glutamine and D-glutamate metabolism in the CD8+T_THEMIS subset.

[0151] In summary, we analyzed the metabolic characteristics among different T cell subsets and found that there are differences in the glutamate metabolic pathway between the CD8+T_THEMIS subset and other subsets.

[0152] 3.1.8 Analysis of interactions among cell subsets

[0153] In CAR-T cell therapy, the function of each T cell subset is often affected by other subsets and tumor cells. Therefore, in-depth analysis of cell communication (based on ligand-receptor pairs) among different T cell subsets and between T cells and tumor cells is of great significance. ​ The results in A show that among different efficacy groups, the interactions between B cells (i.e., tumor cells) and each T cell subset are not very different, but there are differences in the interaction communication among different T cell subsets. In the CR group, the interaction communication between CD8+T_GNLY and other subsets is the most significant; in the RL and NR groups, the interaction communication between CD4+T_CTLA4 and the remaining subsets is the strongest, and these results are consistent with the distribution preference results of each subset among different treatment groups we analyzed before ( ​ C).

[0154] In addition, we also performed ligand-receptor pair analysis between tumor cells and each T cell subset and selected the top 10 most significant ligand-receptor pairs for display. The results show that tumor cells mainly interact with receptors such as CD2, CD27, and HLA-C on the surface of T cells through CD58, CD70, and FAM3C on their surface ( ​B). And we found that compared with the RL and NR groups, the ligand interactions of CCL22_DPP4, CXCL11_DPP4, FAM3C_LAMP1, and CD58_CD2 were more significant in the CR group, and these ligand interactions were related to T cell recruitment, activation, etc.

[0155] Based on the above, we comprehensively analyzed the T cell subsets with the most significant interaction communication among different efficacy groups, and found that in the NR group and RL group, the ligand interactions related to T cell recruitment, activation, etc. decreased, which may reveal the potential mechanism of tumor recurrence after cell therapy and provide an important idea for subsequent basic experimental exploration.

[0156] 3.1.9 TCR Diversity Analysis among Different T Cell Subsets

[0157] Although CAR-T cells mainly bind to tumor antigens through the CAR structure and perform signal transduction functions, the TCRs present on their surfaces also play a role in recognizing antigenic polypeptides and triggering the activation of downstream pathways to activate T cells. To deeply analyze the immune mechanism of four-generation CAR-T cell therapy for B cell lymphoma patients and the possible escape mechanisms leading to recurrence / ineffectiveness, we simultaneously performed TCR sequencing analysis on the CAR-T products to explore the TCR clonotype characteristics and differences of different T cell subsets. ​ As shown in A, both the number of TCR clones and the diversity in each T cell subset of the CR group were higher than those in the RL and NR groups, especially in the CD8+T_GNLY and CD8+T_THEMIS subsets, which may also be related to their more effective and persistent tumor immune clearance ability. In addition, we also noticed that the 7×19 CAR-T cell product of the RL group patients also had more medium and large TCR clone types after being short-term stimulated by tumor cells, which was also consistent with the good immune response results at the initial stage of treatment in this group of patients. In contrast, in the NR group, the number of TCR clones in each T cell subset was less, and a larger proportion of monoclonal TCR types were shown, indicating that the immune response ability of these T cells was in a relatively inefficient state, and the T cell population could not be effectively amplified or activated under antigen stimulation, and ultimately could not effectively clear tumor cells.

[0158] Looking at each T cell subset, we found that the CD8+T_GNLY and CD8+T_MKI67 subsets marked by cytotoxicity and proliferation had more TCR clone numbers and diversity compared to other subsets, especially the former, and these subsets were mainly enriched in the CR group ( ​C). It should be noted that although the CD8+T_THEMIS subset highly enriched in the CR group has a certain degree of clonal diversity, the number of clones is less than that of other subsets, especially in the RL group and the NR group. At the same time, we also analyzed the number of shared TCR clones of different T cell subsets between different treatment efficacy groups. The results showed that the number of shared TCR clones between the CD8+T_GZMA and CD8+T_MKI67 subsets in the CR group was the largest, indicating a strong connection between the two, which may be closely related to the anti-tumor immune response. In the RL group and the NR group, the largest number of shared TCR clones was between the CD8+T_XCL2 and CD8+T_CCL4, and the CD8+T_GZMA and CD4+T_LTB subsets, respectively. It is worth noting that the CD8+T_GNLY with a large number of TCR clones and diversity and the CD8+T_THEMIS subset highly enriched in the CR group shared the fewest TCR clone types between different treatment efficacy groups, showing a certain unique TCR expression pattern ( ​ B).

[0159] From the above, we used TCR sequencing to understand the TCR expression patterns between different T cell subsets and found that the CR group had more TCR clone numbers and diversity. The CD8+T_GNLY and CD8+T_THEMIS subsets had unique TCR expression patterns.

[0160] 3.1.10 Analysis of the differentiation trajectories of each T cell subset

[0161] T cells derived from the same cell often have the same TCR clone type. In the above content, we analyzed the TCR expression patterns between different T cell subsets in different treatment efficacy groups. Although unique TCRs in each subset accounted for the main part, there were still TCR clone types shared to varying degrees among them, suggesting that these two T cells may have a common progenitor cell origin. Therefore, to further study the differentiation and evolution process of T cells in the fourth-generation CAR-T cell product, we used pseudotime analysis and RNA velocity analysis methods to simulate the dynamic evolution process of cells, so as to clarify the direction and state of cell development. ​ The pseudotime trajectory plots in A and B show that at the early stage of cell differentiation, the CD8 + T subset mainly consists of cell subsets marked by cytotoxicity- and proliferation-related genes such as CD8+T_GZMA, CD8+T_MKI67, and CD8+T_GNLY; similarly, the CD4 +The T cell subsets mainly consist of the CD4+T_LTB subset marked by inflammatory response genes. This result indicates that T cells rapidly enter the proliferation and inflammatory response state at the initial stage of immune response, which is closely related to anti-tumor ability and effector function. At the mid-stage of differentiation, T cell subsets related to cytokines and immune effects (such as CD8+T_CCL4, CD8+T_IL5) are mainly distributed, indicating that cytokine secretion mainly drives the immune response during this period. At the end-stage of differentiation, T cells enter the exhaustion period, which is mainly dominated by the CD4+T_CTLA4 and CD8+_THEMIS subsets. The former highly expresses exhaustion marker genes, and the latter was mainly enriched in the CR group in our previous analysis; interestingly, the CD8+T_THEMIS subset exists in each period in the pseudotime trajectory plot, especially in the early and late stages of differentiation. Therefore, we simultaneously performed RNA velocity analysis ( ​ D), and the results showed that the CD8+T_THEMIS subset is different from other subsets and has an independent differentiation path, suggesting that this subset may have a special self-renewing cell pool in CAR-T cell products.

[0162] In addition, we also analyzed the dynamic changes of genes during cell differentiation ( ​ C), and a total of 7 gene modules with different expression change trends were obtained. A total of 8 genes related to T cell function were identified from them. Among them, the 5 genes related to effector and activation, GNLY, STAT1, CD52, IFI6, and LCK, were most significantly expressed in the early stage of differentiation. The chemokines XCL1 and XCL2 were mainly expressed in the mid-stage of differentiation, while the exhaustion-related Treg cell activation marker gene TNFRSF9 was highly expressed in the late stage of differentiation, which is consistent with our previous analysis results.

[0163] From the above, through cell differentiation trajectory analysis, we clarified the evolution process of T cells from the initial activation state to the immune exhaustion state during the immune response process of four generations of CAR-T and the main types of T cell subsets in each differentiation period. In addition, we found that the CD8+T_THEMIS subset has unique differentiation paths and functional characteristics, which may also be one of the reasons for it to be a key subset for maintaining the long-term efficacy of CAR-T cells. Deeply exploring the mechanism behind it may provide ideas and directions for us to improve the continuous tumor-killing ability of CAR-T cells in the future.

[0164] 3.1.11 Analysis of surface proteins of each T cell subset

[0165] To comprehensively understand the characteristics of 7×19 CAR-T cell products in different efficacy groups, we simultaneously performed CITE-seq and analyzed 14 surface proteins by sequencing antibody-derived DNA tags (ADTs), including T cell markers (CD4 and CD8), differentiation and memory-related markers (CD45RO, CD45RA, CCR7 and CD62L), activation markers (HLA-DR, CD69, CD38 and 4-1BB), and immune checkpoints (PD-1, CTLA-4, LAG-3 and TIGIT). The analysis results showed that CD45RO and CD62L related to the memory phenotype were mainly expressed in T cell subsets marked by chemokines and proliferation, such as CD8+T_MKI67 and CD8+T_CCL4. Notably, these two subsets also highly expressed the LAG-3 protein ( ​ A). Consistent with the cell enrichment results, the expression of PD-1 and TIGIT proteins related to cell exhaustion and immunosuppression was upregulated in CD4+T_LTB, CD4+T_IL2, and CD4+T_CTLA4 subsets, which had a higher distribution ratio in the RL and NR groups ( ​ A).

[0166] Previous studies have shown that memory T cells can rapidly proliferate and produce effects after being re-stimulated by antigens, which is the key to maintaining long-term efficacy of CAR-T cell therapy. T cells with immunosuppression and exhaustion are important factors for tumor recurrence and escape. ​ In B, we analyzed the overall protein expression in different efficacy groups. The results showed that the memory-related proteins CD45RO and CD62L were highly expressed in the CR group compared with the NR group (P<0.05). In addition, compared with the CR group, the expression of the immunosuppression and exhaustion-related protein PD-1 was upregulated in the NR group (P<0.05), while the expression of CTLA-4 protein had an upward trend in the RL group (RL vs CR, P>0.05).

[0167] From the above, the CITE-seq results revealed that memory T cell markers were highly expressed in the CR group, while immunosuppression and exhaustion markers were upregulated in the RL and NR groups.

[0168] 3.2 Functional analysis of CD8+T_THEMIS subset and its relationship with the long-term efficacy of 7×19 CAR-T cells

[0169] 3.2.1 Functional characteristic analysis of CD8+T_THEMIS subset

[0170] In the previous section, through Ro / e analysis, we found that the CD8+T_THEMIS subset was highly enriched in the CR group and had unique gene expression characteristics. To further explore the function of this special T cell subset and clarify its possible mechanism for improving the long-term efficacy of CAR-T cell therapy, we performed functional enrichment analysis on 2,977 differentially expressed genes (1,982 upregulated genes and 995 downregulated genes) in this subset ( ​ A). The results showed that the upregulated genes in the CD8+T_THEMIS subset were closely related to T cell differentiation, activation, and proliferation at the functional level, and were also related to epigenetic modifications such as ubiquitination and methylation. Mechanistically, these genes were closely related to the MAPK signaling pathway and also involved signaling pathways such as TCR, Jak-Stat, and FOXO, which have been widely proven to be closely related to T cell activation, effector functions, and other aspects. Analyzing the functions of the downregulated genes, we found that most of them were related to oxidative phosphorylation and mitochondrial respiratory chain. Importantly, we also found that genes related to the apoptosis-promoting pathway were downregulated in this subset ( ​ B).

[0171] Therefore, through gene functional enrichment analysis, we found that genes related to T cell differentiation and proliferation were upregulated in the CD8+_THEMIS subset, while genes related to oxidative phosphorylation and apoptosis were downregulated.

[0172] 3.2.2 Functional differences in different differentiation stages of the CD8+T_THEMIS subset

[0173] In the results of the pseudotime analysis, we noticed that the CD8+T_THEMIS subset had an independent cell differentiation path. To comprehensively understand the changes in its function with cell differentiation, we further compared the gene expression differences between the early differentiation stage (state 2) and the late differentiation stage (state 6) of this subset, and systematically evaluated the functional characteristics and transcription factor expression differences between the two differentiation stages. ​ A The results showed that at the early stage of cell differentiation, genes related to T cell activation and immune response were highly expressed in this subset, while at the late stage of cell differentiation, genes related to protein transcription and translation activity were highly expressed. The Ucell method score showed that the CD8+T_THEMIS subset at the early differentiation stage had a higher cytotoxicity score (P<0.05), indicating that the cells quickly entered the immune response state. Interestingly, the cells of this subset had a higher ferroptosis (P<0.05) and cell exhaustion score (P<0.05) at the late differentiation stage, while also having a higher cell proliferation score (P<0.05) ( ​B). In addition, by comparing the differences in the expression of transcription factors between two time points, it was found that in this subset, the expression of the exhaustion-related transcription factor IRF4 was upregulated at the late stage of differentiation, while the transcription factors MYC, Stat5A, and JUND related to cell proliferation and pre-exhausted T cells were highly expressed. ​ C). These results suggest that a certain proportion of stem cell-like pre-exhausted T cells (Tpex) may be maintained in the CD8+ T_THEMIS subset for a long time at the late stage of differentiation.

[0174] In summary, by comparing the gene expression patterns of the CD8+ T_THEMIS subset in the early and late stages of differentiation, it was found that the early cells had good activation and immune response functions. Although this subset entered an exhausted state at the late stage, a certain proportion of Tpex cells might be maintained. It has been proven in previous studies that this type of T cell is closely related to the long-term efficacy maintenance of CAR-T cell therapy.

[0175] 3.2.3 Identification of prognosis-related genes in the CD8+ T_THEMIS subset

[0176] To further identify the key genes related to clinical efficacy in the CD8+ T_THEMIS subset, we downloaded the gene expression data and clinical information of CAR-T cell products in the GSE223655 cohort from the GEO database and analyzed the top 15 differential genes in the subset. The results showed that compared with the progressive disease (PD) group, the genes SOS1 (P<0.05), THEMIS (P<0.01), PDE3B (P<0.05), CDK6 (P<0.01), and BACH2 (P<0.05) were all upregulated in the CR group. ​ A). Subsequently, we further analyzed the expression of the above genes in CD8 + T cells in different efficacy groups and still found that THEMIS (P<0.01), SOS1 (P<0.05), and CDK6 (P<0.05) were upregulated in the CR group. ​ B).

[0177] In summary, we identified through a public database that the expression levels of the THEMIS, SOS1, and CDK6 genes in CD8 + T cells of CAR-T cell products are related to their efficacy.

[0178] 4 Discussion

[0179] B-cell lymphoma is the most common NHL. Although the current standard regimen represented by rituximab combined with chemotherapy, R-CHOP (cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone), has achieved cure rates of 70% and 40% for the germinal center B-cell type and activated B-cell type of DLBCL, respectively, there are still 30%-45% of patients who will relapse or progress. In recent years, the emergence of CAR-T cell therapy has made it possible to cure R / R B-cell lymphoma. The results of large multi-center clinical studies reported currently show that the CR rates of R / R LBCL, FL, and MCL reach 39-66%, 79-94%, and 67-82%, respectively. However, nearly half of the patients still relapse within 1 year after treatment. Therefore, it is of great significance to further study the differences between CAR-T cell products in patients with different treatment effects and identify the key factors for maintaining long-term efficacy.

[0180] In this study, we utilized scRNA-seq, TCR sequencing, and CITE-seq technologies to analyze the biological characteristics of CAR-T cell infusion products from R / R LBCL patients with different efficacy outcomes after four generations of CAR-T cell therapy, in order to explore the key factors maintaining the long-term efficacy of four generations of CAR-T cells. We deeply analyzed the differences in the biological characteristics of CAR-T cell products between different efficacy groups, including cell subset distribution, functional characteristics, cytokine expression, transcription factor expression profiles, metabolic characteristics, and cell-cell interactions. Through dimensionality reduction clustering methods, we identified 10 T cell subsets with different gene expression patterns, and found the CD8+T_THEMIS subset highly enriched in the CR group using Ro / e analysis. Further, through the Ucell scoring method to systematically evaluate the functional characteristics of different treatment groups and different T cell subsets, it was found that the infusion products in the CR group had better cell proliferation, tissue retention, and lipid metabolism capabilities. Although the RL group had more prominent cytotoxic and inflammatory effects, it also had higher exhaustion characteristics. The functional evaluation results of cell subsets showed that the special subset CD8+T_THEMIS presented better cell proliferation ability consistent with the overall evaluation of the CR group. The latest research found that the killing ability of CAR-T cells with 4-1BB as the co-stimulatory domain is related to their proliferation ability. Although the killing speed of a single cell is slow, its strong proliferation ability and cooperative killing characteristics endow it with more effective continuous tumor killing ability, which may also be one of the important factors for this subset to maintain the long-term efficacy of CAR-T therapy. In further transcriptional factor expression analysis, we found that the expression levels and activities of FOXO1 and FOXP1, etc. were significantly upregulated in the CD8+T_THEMIS subset. Recently, multiple studies have demonstrated that FOXO1 and FOXP1 can promote the expansion of CAR-T cells and maintain their stemness, restrict the excessive differentiation of effector cells, thereby resisting the exhausted state and improving the long-term efficacy of CAR-T cells. On the other hand, we also analyzed the metabolic characteristics of each T cell subset, and the results showed that the CD8+T_THEMIS subset mainly had differences in the glutamate metabolic pathway and had a higher GSH metabolic abundance. Previous studies have reported that GSH can maintain T cell homeostasis in vivo and promote their proliferation after stimulation, and the reduction of GSH will impair the normal glycolysis and glutamine metabolism capabilities of T cells and affect cell proliferation.

[0181] Meanwhile, after analyzing the ligand-receptor interactions between various T cell subsets and tumor cells in different efficacy groups, we found that the ligand-receptor interactions of CCL22_DPP4, CXCL11_DPP4, FAM3C_LAMP1, and CD58_CD2 were more significant in the CR group. Some researchers have found that CCL22, as a factor coupled with B cell antibody affinity, can act as a marker to transmit the affinity information of B cells to T cells. At the same time, it can positively promote the recruitment of more T cells to B cells with high affinity, thereby playing a role in killing tumors. Similarly, CXCL11 (also known as IFN-inducible T cell alpha-chemokine) can chemotax and recruit activated T cells, thus regulating the migration and differentiation of immune cells. In addition, CD58, as a co-stimulatory receptor, its natural ligand CD2 is mainly expressed on the surface of T / NK. Multiple studies have shown that the intact CD58-CD2 axis is necessary for the effective lysis of cancer cells mediated by tumor-infiltrating lymphocytes, and the disruption of this specific ligand-receptor pair will lead to tumor immune escape.

[0182] Interestingly, when performing TCR on each T cell subset, we found that the CD8+T_THEMIS subset showed a unique TCR expression pattern, sharing the fewest TCR clone types among subsets. Cell differentiation trajectory analysis further suggested that it has an independent differentiation path. These results all suggest that this subset may have a special self-renewing cell pool. To further analyze the functional and gene expression evolution of this subset during differentiation, we compared the differences between the early and late stages of differentiation and found that genes related to T cell activation and immune response were highly expressed in the early stage of this subset, showing good effector functions; while in the late stage of differentiation, this cell subset showed high exhaustion characteristics while also showing a certain proliferative ability. Further comparison of the transcription factor expression profiles found that not only the exhaustion-related transcription factor IRF4 was highly expressed in the late stage of its differentiation, but also the transcription factors MYC, Stat5A, and Jun related to Tpex were highly expressed. Recent studies have found that under chronic antigen exposure conditions, the expression of Stat5 in exhausted T cells can directly promote the formation of exhausted intermediate cells and restart some effector biological functions, enhancing anti-tumor potential. In summary, the CD8+T_THEMIS subset may play a key role in maintaining the long-term efficacy of fourth-generation CAR-T cells through its unique functional characteristics and transcription factor expression.

[0183] To further identify the representative genes related to the efficacy of CAR-T cells in the CD8+T_THEMIS subset, we used an external database to verify and found that the expression of SOS1, THEMIS, and CDK6 genes in CAR-T cell infusion products CD8 + T cells was correlated with the efficacy of CAR-T cells, providing a basis for subsequent mechanism exploration.

[0184] In summary, this study identified the key specific subset CD8+T_THEMIS that maintains the long-term efficacy of fourth-generation CAR-T cells. This subset is characterized by active proliferation and high expression of transcription factors related to T cell stemness. In addition, this subset has an independent differentiation pathway and highly expresses transcription factors that can reverse the function of exhausted T cells in the late stage of differentiation. The THEMIS gene, as the marker gene of this subset, has been found to be correlated with the efficacy of CAR-T cells. Moreover, recent studies have found that it is related to the homeostasis of peripheral T cells, and in CAR-T with 4-1BB as the co-stimulatory domain, the THEMIS-SHP1 complex can reduce the phosphorylation of CAR-CD3ζ, thereby weakening the degree of T cell activation and ultimately reducing the secretion of cytokines. However, the current regulatory mechanism of the THEMIS gene on the phenotypes of CAR-T cells such as memory, exhaustion, and apoptosis is not clear, and its impact on the long-term efficacy of CAR-T cells, especially fourth-generation CAR-T cells, is not clear. Therefore, further research on it in the future is necessary.

[0185] 5 Results and Conclusions

[0186] Results: (1) The infused products in the complete remission (CR) group had higher Ucell scores in terms of cell proliferation, tissue retention, and lipid metabolism, while the relapse (RL) group had higher exhaustion scores; (2) Dimensionality reduction clustering and single-cell tissue preference analysis (the ratio of observed to expected cell numbers, Ro / e) found that the CD8+T_TEHMIS subset was highly enriched in the CR group, showing better cell proliferation ability and high expression of memory stemness-related transcription factors; (3) The CD8+T_THEMIS subset had a unique TCR expression pattern and differentiation pathway; (4) Validation in public databases found that the expression of the thymocyte-expressed molecule involved in selection (THEMIS) gene in CAR-T cell products in the complete remission group was higher than that in the disease progression group (P<0.01).

[0187] Conclusions: 1) The CAR-T cell products in the CR group as a whole had better cell proliferation and tissue retention characteristics, while the exhaustion characteristics in the RL group were obvious; 2) The CD8+T_THEMIS subset was highly enriched in the CAR-T cell products in the CR group and was the key subset for maintaining long-term efficacy; 3) The high expression of the THEMIS gene was related to the good efficacy of CAR-T cells.

[0188] Part II: The Killing Effect and Mechanism of THEMIS Gene-Regulated Fourth-Generation CAR-T on B-Cell Lymphoma

[0189] THEMIS is an evolutionarily conserved T-cell-specific gene, and the protein it encodes contains 636 and 641 amino acids in mice and humans, respectively. Multiple studies using gene knockout models and N-ethyl-N-nitrosourea-induced mutations have revealed the key role of THEMIS in thymocyte development. Notably, all mutant models have found significant blockage of thymocytes at the CD4 + CD8 + double-positive stage. These mutations are widely distributed throughout the entire sequence of the THEMIS protein, including frameshift mutations at sites such as Y489X and T512P in the mouse model. THEMIS participates in TCR signaling by constitutively binding to the adaptor protein Grb2 and the tyrosine phosphatase SHP1. Among them, Grb2 mediates the recruitment of the THEMIS-SHP1 complex to phosphorylated LAT molecules. Although multiple studies have confirmed the regulatory effect of THEMIS on SHP1 activity, the specific regulatory direction (activation or inhibition) of TCR signaling during thymocyte development remains controversial. For example, the research results on THEMIS / SHP1 double-knockout mice are divergent: one group observed a complete phenotypic recovery, while the other group did not detect significant changes. Therefore, the impact of the THEMIS gene on T cells and its molecular mechanism still need to be further explored in depth.

[0190] In recent years, studies have found that THEMIS has important functions in peripheral mature T cells. The Gascoigne team revealed that THEMIS promotes the maintenance of CD8 + T cell homeostasis by integrating cytokine signals and low-affinity TCR-ligand interactions. Another research team verified through both mouse models and in vitro experiments that THEMIS is essential for the proliferation of CD8 + T cells driven by IL-2 and IL-15. Similarly, studies have also found that knockout of THEMIS can lead to defects in CD4 + T cell development and dysfunction of regulatory T cells. THEMIS has also been found to be involved in the occurrence and development of various diseases. Genome-wide association studies (GWAS) have found that THEMIS is significantly associated with autoimmune diseases such as celiac disease and inflammatory bowel disease. In addition, THEMIS is also involved in the mechanism of HTLV-1 infection: the viral bZIP factor can induce abnormal proliferation of T cells by binding to THEMIS.

[0191] In CAR-T cell therapy research, some researchers found that CAR with 4-1BB as a co-stimulatory molecule can recruit the THEMIS-SHP1 complex to the CAR signalosome through the intracellular domain of 4-1BB, inhibiting CAR-CD3ζ phosphorylation. Knocking out THEMIS or SHP1 can enhance the basal phosphorylation level, but its regulatory effect on tumor killing function has not been clarified. These findings highlight the potential value of THEMIS as a functional regulatory target in CAR-T cell therapy.

[0192] Based on single-cell sequencing results, we found that the CD8+T_THEMIS subset has unique functional characteristics and roles, and its signature gene THEMIS is related to the prognosis of CAR-T therapy. In this part of the study, we constructed two models to study the regulation of THEMIS on the function of fourth-generation CAR-T: 1) Knock out THEMIS in CAR-T cells using electroporation combined with CRISPR / Cas9 gene editing technology; 2) Connect the CDS sequence of THEMIS to the CAR structure to construct a fourth-generation CAR structure expressing THEMIS, and prepare a new type of fourth-generation CAR-T cell overexpressing THEMIS (named THEMIS-7×19CAR-T) by lentiviral vector transfection. This part aims to study the regulatory effects of THEMIS on the killing function, memory, exhaustion phenotype, and cytokine secretion of fourth-generation CAR-T cells, providing new strategies for enhancing the efficacy of CAR-T cells and modifying the CAR structure.

[0193] Previous literature reports that THEMIS plays an important role in CAR-T cells with 4-1BB as the co-stimulatory domain, and the fourth-generation CAR structure in our current study also uses 4-1BB as the co-stimulatory domain. On this basis, we further completed relevant preclinical studies to explore the effect of THEMIS on the efficacy of fourth-generation CAR-T cells. The results showed that after knocking out THEMIS, the CD8 / CD4 ratio of CAR-T cells decreased, the cell proliferation and memory cell formation ability after antigen stimulation were limited, and the cytokines were in a state of over-secretion, making them more prone to exhaustion and apoptosis. In addition, we found that knocking out THEMIS mainly affected the long-term tumor killing function of CAR-T cells rather than the short-term killing effect. Compared with the control group, THEMIS-7×19CAR-T cells had enhanced proliferation ability and memory cell formation ability after antigen stimulation, and the cytokines were in a controllable state, with reduced cell apoptosis and exhaustion, and could maintain a more effective long-term tumor killing ability. Therefore, these results further revealed the previously undiscovered relationship between THEMIS and the stemness and long-term killing function of CAR-T cells, and also provided new ideas for subsequent modification of CAR-T cells to improve the efficacy.

[0194] 1 Experimental materials

[0195] 1.1 Cell lines and plasmids

[0196] 1.1.1 Cell lines

[0197] 1) Jurkat: a human T lymphocyte leukemia cell line, purchased from ATCC;

[0198] 2) Raji: a human Burkitt's lymphoma cell line, purchased from ATCC;

[0199] 3) Jeko-1: a human mantle cell lymphoma cell line, purchased from ATCC;

[0200] 4) 3T3: a mouse embryonic fibroblast cell line, purchased from ATCC;

[0201] 5) HEK-293T / 17: a human embryonic kidney cell line, purchased from ATCC;

[0202] 1.1.2 Primary cells

[0203] This study was approved by the Ethics Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine. Informed consent was obtained from each healthy volunteer or patient before peripheral blood or bone marrow collection.

[0204] 1) Normal human mononuclear cells: isolated from the peripheral blood of healthy volunteers by gradient density centrifugation.

[0205] 2) T cells from healthy volunteers: obtained from the bone marrow of healthy volunteers by gradient density centrifugation and CD3 + magnetic bead sorting.

[0206] 1.1.3 Plasmids

[0207] The lentiviral packaging plasmids pMDLg / pRRE, pRSV-Rev, and pMD2.G were purchased from Addgene (https: / / www.addgene.org / ), and the (target plasmid) expression vector plasmid pLenti7.3 / V5-DEST TM was purchased from Thermo Fisher Scientific China Co., Ltd. (Shanghai, China).

[0208] 1.2 Main experimental reagents

[0209] Table 5

[0210]

[0211]

[0212] 1.3 Preparation of main reagents:

[0213] 1) Cell cryopreservation solution (serum:DMSO = 9:1): Mix 9 mL of fetal bovine serum and 1 mL of DMSO, and prepare it freshly before use.

[0214] 2) Virus preservation solution (1% HEPES X-VIVO): Mix 9.9 mL of X-VIVO medium and 0.1 mL of HEPES, and store it at 4°C for later use.

[0215] 3) LB liquid medium: Dissolve 5 g of LB broth medium in 200 mL of deionized water, place it in a 300 mL conical flask, seal it with tin foil, sterilize it under high temperature and high pressure, and then store it at 4°C in the refrigerator after returning to room temperature for later use.

[0216] 4) LB solid medium: Dissolve 5 g of LB broth agar in 200 mL of deionized water, place it in a 300 mL conical flask, seal it with tin foil, sterilize it under high temperature and high pressure, and then add 400 μL of 100 mg / mL ampicillin and mix well after cooling to 30 - 50°C. Aliquot 10 mL per petri dish into 10 cm cell culture dishes. After it solidifies, seal it with parafilm and store it inverted at 4°C in the refrigerator for later use.

[0217] 5) Flow cytometry washing solution (2% serum PBS): Add 1 mL of PAN serum to 49 mL of PBS, and store it at 4°C for later use.

[0218] 6) Magnetic bead sorting buffer: Add 1 mL of PAN serum and 2 mM EDTA to 49 mL of PBS, and store it at 4°C for later use.

[0219] 7) 10% ammonium persulphate (APS): Dissolve 1 g of APS in 10 ml of double-distilled water, then aliquot it into EP tubes and store it in the -20°C refrigerator for later use.

[0220] 8) Electrophoresis buffer (1× and 10×): Dissolve 144 g of glycine, 30.3 g of Tris, and 10 g of SDS in double-distilled water, and make the total volume up to 1 L and dissolve it thoroughly. Take 100 mL of 10× electrophoresis buffer, add 200 mL of methanol and 800 mL of deionized water, and mix well to prepare 1× electrophoresis buffer for later use.

[0221] 9) TBST solution (1×): Dilute 20×TBST 20-fold with deionized water.

[0222] 10) Blocking solution (5%): Take 2.5 g of skim milk powder, add 50 mL of 1×TBST solution, and shake it well to dissolve. Prepare it freshly before use.

[0223] 11) Secondary antibody solution: Mix 1 μL of horseradish peroxidase-labeled goat anti-rabbit IgG into 5 mL of 5% skim milk and mix well immediately before use.

[0224] 12) SDS polyacrylamide gel: Prepare gels with different concentrations according to the molecular weight of the protein. The formulation details are shown in Table 6.

[0225] Table 6 Formulation of SDS polyacrylamide gels with different concentrations

[0226]

[0227] 1.4 Main instruments and equipment

[0228] Table 7

[0229]

[0230]

[0231] Experimental methods

[0232] 2.1 Cell culture

[0233] Jurkat, Raji, and Jeko-1 cells are cultured in RPMI 1640 complete medium, HEK-293T / F17 and 3T3 cells are cultured in DMEM complete medium, and CAR-T cells are cultured in X-VIVO complete medium. All cell lines are tested for mycoplasma every 2 weeks. If positive, they are discarded. The media are prepared as follows:

[0234] 1) RPMI 1640 complete medium: 1% penicillin-streptomycin + 10% FBS + RPMI 1640 medium;

[0235] 2) DMEM complete medium: 1% penicillin-streptomycin + 10% FBS + DMEM medium;

[0236] 3) X-VIVO complete medium: 1% penicillin-streptomycin + 10% Gibco serum + 1% HEPES + 1% sodium pyruvate + 1% glutamine + 1% non-essential amino acids + 300 IU / mL rhIL-2 + 5 ng / mL IL-7 + 5 ng / mL IL-15.

[0237] 2.2 Cell cryopreservation

[0238] 1) Collect cells, centrifuge at 1500 rpm for 5 min, discard the supernatant, resuspend with PBS, and count using a flow cytometer.

[0239] 2) After counting, centrifuge at 1500 rpm for 5 min, discard the supernatant, and add cryopreservation solution at 1×10 7 cells / mL. Mix well and transfer to cryotubes.

[0240] 3) First, place the cryotube into a programmable cooling box and put it in a -80°C refrigerator. After 24 hours, take it out and store it in a -80°C refrigerator for short-term storage or transfer it to a liquid nitrogen tank for long-term storage.

[0241] 2.3 Cell Resuscitation

[0242] 1) Preheat the medium in advance. Meanwhile, prepare an appropriate number of 15 mL centrifuge tubes and add 4 - 5 mL of preheated medium.

[0243] 2) Take out the cells to be resuscitated from the -80°C refrigerator or liquid nitrogen tank and quickly thaw them in a 37°C water bath.

[0244] 3) Use a pipette to add the cells to the prepared centrifuge tube, mix well, and centrifuge at 1500 rpm for 5 minutes.

[0245] 4) Carefully remove the supernatant, add an appropriate amount of preheated medium, mix well, and transfer it to a 6-well plate or a T25 cell culture flask.

[0246] 5) Observe the cell status under the microscope and place it in an incubator at 37°C with 5% CO 2 for incubation.

[0247] 2.4 Lentivirus Packaging

[0248] 2.4.1 Transformation

[0249] 1) Take 1 tube of competent cells from the -80°C refrigerator and let it thaw on ice.

[0250] 2) Take 20 μL of competent cells and add them to a 1.5 mL EP tube. Then add 1 μL of the target plasmid, gently flick to mix, incubate on ice for 30 minutes, heat shock in a 42°C metal bath for 90 seconds, and then incubate on ice for 2 minutes again.

[0251] 3) Add 900 μL of liquid LB broth medium without ampicillin, pipette to mix well, and culture it on a shaker at 37°C and 235 rpm for 45 minutes.

[0252] 4) Centrifuge at 6000 rpm for 5 minutes, discard 700 μL of the supernatant, and resuspend the pellet.

[0253] 5) Take 50 μL and drop it onto an LB broth agar plate containing ampicillin, spread it evenly with an L-shaped spatula, place it upright in a 37°C incubator for 1 hour, then invert it and culture for 12 - 16 hours.

[0254] 2.4.2 Plasmid Extraction

[0255] Take the plasmid transformed the previous day, pick a single colony into a test tube containing 5 mL of liquid LB medium, and add 5 μL of ampicillin (100 mg / mL). Incubate in a shaker at 37°C and 235 rpm for 6 - 8 h. After 6 - 8 h, pour the 5 mL of bacteria-containing LB medium into 200 mL of liquid LB medium, and add 200 μL of ampicillin. Incubate in a shaker at 37°C and 235 rpm for 12 - 15 h, and then perform large-scale plasmid extraction:

[0256] 1) Aliquot 200 mL of the bacterial solution into 50 mL centrifuge tubes, centrifuge at 8000 rpm for 3 min, and discard the supernatant;

[0257] 2) Resuspend the cell pellets in the 4 centrifuge tubes with 8 mL of P1 solution and transfer them to a single 50 mL centrifuge tube. Vortex to thoroughly suspend the cells;

[0258] 3) Add 8 mL of P2 solution, invert the tube 6 times to fully mix and lyse the cells, and let it stand at room temperature for 5 min; then add 8 mL of P4 solution, invert the tube 6 times until white flocculent precipitates are observed, and let it stand at room temperature for 10 min;

[0259] 4) Centrifuge at 8000 rpm for 10 min to collect the flocculent matter at the bottom of the tube. Pour all the supernatant into CS1, and push the handle to collect the filtrate;

[0260] 5) Add 0.3 volume of isopropanol to the filtered bacterial solution and invert to mix;

[0261] 6) Prepare adsorption column CP6, add 3 mL of equilibration buffer BL to fully wet the filter membrane, centrifuge at 8000 rpm for 3 min, and discard the waste liquid in the collection tube;

[0262] 7) Add the 8 mL of bacterial solution with isopropanol added to CP6, centrifuge at 8000 rpm for 3 min, discard the waste liquid in the collection tube, and put the adsorption column back into the collection tube;

[0263] 8) Repeat step 7) until all the bacterial solution has passed through the column;

[0264] 9) Add 10 mL of wash buffer (with absolute ethanol added), centrifuge at 8000 rpm for 2 min, discard the waste liquid in the collection tube, and put the adsorption column back into the collection tube;

[0265] 10) Repeat step 9) once;

[0266] 11) Add 3 mL of absolute ethanol to the adsorption column, centrifuge at 8000 rpm for 2 min, discard the waste liquid in the collection tube, and put the adsorption column back into the collection tube;

[0267] 12) Centrifuge at 8000 rpm for 5 min to completely remove the wash buffer in the adsorption column;

[0268] 13) Place the adsorption column CP6 in a new collection tube, open the lid and air dry for 5 min to allow the absolute ethanol to fully evaporate;

[0269] 14) Suspend and rotate to add 500 μL - 1 mL of TB elution buffer to the center of the adsorption column filter membrane, let it stand at room temperature for 10 min, centrifuge at 8000 rpm for 5 min, and the liquid in the collection tube is the plasmid we need;

[0270] 15) Use a Nanodrop micro - spectrophotometer to detect the concentration and OD 260 / 280, and store at - 20 °C for later use.

[0271] 2.4.3 Lentivirus packaging

[0272] 1) One day in advance, culture 1×10 7 293T cells in a 10 - cm dish;

[0273] 2) Two hours in advance, change the cell culture medium to pre - warmed 10% FBS Opti - MEM medium;

[0274] 3) Plasmid preparation: Add 500 μL of Opti - MEM medium to a 15 - mL centrifuge tube. According to the molar ratio of pMDLg / pRRE:pRSV - Rev:pMD2.G:target plasmid = 1:1:0.5:2, add plasmids at 10 μg / dish, calculate the required mass and volume of each packaging plasmid and the target plasmid, add them to the Opti - MEM medium, mix well and incubate at room temperature for 5 min;

[0275] 4) Add 10 μL of Neofectamine, mix well and incubate at room temperature for 17 min. After incubation, add it to the 293T supernatant;

[0276] 5) After 48 h, collect the supernatant, and add 10 mL / dish of 10% FBS Opti - MEM medium again. At 72 h, collect the supernatant again;

[0277] 6) Centrifuge the collected supernatant at 1500 rpm for 5 min and filter it using a 0.45 - μm filter membrane;

[0278] 7) Centrifuge at 4000 g, 4 °C for 12 h;

[0279] 8) After centrifugation, discard the supernatant, resuspend the virus pellet with the virus preservation solution (concentrated at 1:250), aliquot and store at - 80 °C.

[0280] 2.4.4 Virus titer determination

[0281] 1) After cell counting, seed at 2×10 per well 5Cells, 4 gradients, 2 replicates. Take Jurkat cells. Considering cell loss, take Jurkat cells for 10 wells, that is, 2×10 6 cells;

[0282] 2) Centrifuge at 1500 rpm for 5 min. After discarding the supernatant, resuspend with complete 1640 medium and add 2 μL of polybrene. Add 100 μL per well to a 96-well plate;

[0283] 3) Add 90 μL of virus preservation solution to 10 μL of the original virus solution and dilute 10 times;

[0284] 4) Add the original virus solution, preservation solution, and 1640 medium according to the system in Table 8 below to measure the titer:

[0285] Table 8

[0286]

[0287] 5) Centrifuge at 1200 g at 32 °C for 1.5 h;

[0288] 6) After centrifugation, place at 37 °C in a 5% CO 2 incubator and incubate for 4 h;

[0289] 7) After 4 h, centrifuge at 1500 rpm for 5 min. After removing the supernatant, resuspend with 2 mL of complete 1640 medium and place in a 24-well plate;

[0290] 8) After 48 h, detect the positive rate by flow cytometry. Titer = [(2×10 5 ×positive rate) / 0.2]×dilution factor.

[0291] 2.5 Cell line construction

[0292] Raji-luc-GFP and Jeko-1-luc-GFP used in this article are both overexpressed with luciferase-P2A-GFP, and CD19-3T3-GFP is overexpressed with CD19-P2A-GFP. The construction method is as follows:

[0293] 1) Take 1×10 6 wild-type tumor cells / 3T3 cells and centrifuge at 1500 rpm for 5 min;

[0294] 2) Resuspend the cells with 100 μL of the original luciferase-P2A-GFP / CD19-P2A-GFP lentivirus solution (titer: 1×10 7 pfu / ml), supplement with 100 μL of complete 1640 medium, and add 0.2 μL of polybrene. Place in a 96-well plate;

[0295] 3) Centrifuge at 1200 g and 32 °C for 1.5 h;

[0296] 4) Incubate at 37 °C in 5% CO 2 for 4 h;

[0297] 5) Centrifuge at 1500 rpm for 5 min, carefully discard the supernatant, and resuspend in 2 mL of complete 1640 medium in a 24-well plate. Incubate at 37 °C in 5% CO 2 for 48 h;

[0298] 6) Detect GFP expression by flow cytometry;

[0299] 7) After 72 h, sort out GFP-positive cells by flow sorting, culture in 1640 / DMEM medium containing 2% penicillin-streptomycin + 10% FBS

[0300] for 48 h, then replace with complete 1640 / DMEM medium. After determining that the GFP positive rate is 100%, use it for subsequent experiments. Regularly detect the GFP positive rate later, and the GFP positive rate remains 100% for a long time.

[0301] 2.6 Preparation of CAR-T cells

[0302] 2.6.1 Isolation of human peripheral blood mononuclear cells (PBMC)

[0303] 1) Add 10 - 20 mL of healthy human peripheral blood to an equal volume of PBS and mix well;

[0304] 2) Prepare 2 50-mL centrifuge tubes and add 20 mL of human peripheral blood lymphocyte separation medium

[0305] 3) Carefully layer the diluted peripheral blood slowly along the tube wall on the surface of the human peripheral blood lymphocyte separation medium;

[0306] 4) Centrifuge at 800 g, acceleration 1, deceleration 0, at room temperature for 20 min;

[0307] 5) Prepare 2 15-mL centrifuge tubes and add 5 - 8 mL of PBS. Carefully aspirate the white film layer (i.e., PBMC) and add it to the PBS, and mix well;

[0308] 6) Centrifuge at 1500 rpm for 5 min;

[0309] 7) Carefully discard the supernatant, and freeze the cell pellet in cell cryopreservation solution for later use or use directly.

[0310] 2.6.2 Isolation of CD3 + T cells

[0311] 1) Resuspend PBMC in 50 mL of X-VIVO complete medium. After thoroughly pipetting to mix, add it to a T225 flask.

[0312] 2) Incubate in a 37°C, 5% CO 2 incubator for at least 6 h, then carefully turn the culture flask over.

[0313] 3) After 2 h, carefully stand the culture flask on its side. In a biosafety cabinet, pour the cell suspension along the side into a new T225 flask and place it in a 37°C, 5% CO 2 incubator for culture.

[0314] 4) Carefully turn it over every 2 h. When both sides of the culture flask are covered, replace it with a new T225 flask. After filling 2 - 3 T225 flasks, observe the proportion of adherent cells under a microscope. When the proportion of adherent cells < 10%, take a small amount of cells for CD3 detection. If CD3 + > 90%, then T cell activation can be carried out.

[0315] 2.6.3 T cell activation

[0316] 1) Collect the CD3 + > 90% T cells obtained in the previous step and centrifuge at 1500 rpm for 5 min.

[0317] 2) Carefully discard the supernatant, resuspend the cell pellet in an appropriate amount of X-VIVO complete medium, and thoroughly mix before performing cell counting.

[0318] 3) Calculate the required volume of CD3 / CD28 Dynabeads based on the cell number, that is, 30% × total cell number / CD3 / CD28 Dynabeads density. At the same time, centrifuge the cell suspension at 1500 rpm for 5 min.

[0319] 4) Wash the CD3 / CD28 Dynabeads: Prepare 1 15-mL centrifuge tube, add 1 mL of PBS, and then add the CD3 / CD28 Dynabeads calculated in the previous step. Place it on a magnetic stand and let it stand for 5 min.

[0320] 5) Aspirate and discard the PBS in the 15-mL centrifuge tube (keep the 15-mL centrifuge tube on the magnet).

[0321] 6) Remove the 15-mL centrifuge tube from the magnetic stand, and the iron-red CD3 / CD28 Dynabeads on the tube wall can be seen.

[0322] 7) Resuspend the cell pellet in X-VIVO complete medium at 4 - 6×10 6 / mL.

[0323] 8) Add the cell suspension to the washed CD3 / CD28 Dynabeads, pipette to mix well, and then incubate on a rotary shaker for 30 min to allow the cells to come into full contact with the CD3 / CD28 Dynabeads.

[0324] 9) After the rotation ends, place a 15 mL centrifuge tube on a magnetic stand and let it stand for 5 min.

[0325] 10) Carefully discard the supernatant.

[0326] 11) Remove the 15 mL centrifuge tube, and resuspend the pellet with X-VIVO medium at a ratio of 1 - 3×10 6 / mL of CD3 / CD28 Dynabeads, then transfer it to a T25 flask and culture for at least 8 h, which is named D0 at this time.

[0327] 2.6.4 Lentiviral transfection

[0328] 1) Collect the activated T cells obtained in the previous step and centrifuge at 1500 rpm for 5 min.

[0329] 2) Resuspend with an appropriate amount of X-VIVO complete medium and then perform cell counting.

[0330] 3) Take 1×10 6 activated T cells, centrifuge at 1500 rpm for 5 min, and use the remaining cells as the control group (control T cells, CT).

[0331] 4) Carefully discard the supernatant, add the fourth-generation CD19-CAR or CD19-CAR-THEMIS lentivirus at 30 moi (multiplicity of infection), supplement the system to 200 μL with X-VIVO blank medium, add 0.2 μL of polybrene (8 mg / mL), transfer it to one well of a 96-well plate, and carefully seal the 96-well plate with a sealing film.

[0332] 5) Centrifuge at 1200 g and 32 °C for 1.5 h.

[0333] 6) After centrifugation, culture at 37 °C and 5% CO 2 for 4 h.

[0334] 7) Centrifuge at 1500 rpm for 5 min, carefully discard the supernatant, resuspend the cells with 2 mL of X-VIVO complete medium in a 24-well cell culture plate, and name it D1.

[0335] 2.6.5 CAR-T cell culture

[0336] 1) On day D3, flow cytometry was used to detect the transduction efficiency, and 8 mL of X-VIVO medium was supplemented and transferred into a T25 flask for continued culture;

[0337] 2) On day D4, after thoroughly pipetting and mixing the cell suspension, it was transferred into a 15 mL centrifuge tube and placed on a magnetic stand for 5 minutes to remove CD3 / CD28 Dynabeads. The supernatant was collected, which was the required cells; CT cells were processed in the same way.

[0338] 3) During the D3 - 10 culture period, an equal volume of X-VIVO complete medium was supplemented every other day to the existing culture system. According to the culture system, it was successively transferred into T75 and T175 flasks, or part of it was cryopreserved for later use;

[0339] 4) On days D11 - 14, in vitro experiments or animal experiments were carried out.

[0340] 2.7 CRISPR / Cas9 gene editing by electroporation

[0341] 2.7.1 Preparation before electroporation

[0342] 2.7.1.1 Preparation of sgRNA

[0343] After transient centrifugation of a 1.5 nmol sgRNA dry powder tube, 10 μL of DEPC water was added to make the final

[0344] concentration 150 pmol / μL.

[0345] All sgRNAs involved in this project were purchased from Genscript (https: / / www.genscript.com.cn / ), and the specific sequences are as follows:

[0346] Table 9

[0347]

[0348] 2.7.1.2 Cell preparation

[0349] 1) The cells were changed to fresh medium (X-VIVO complete medium) every 2 days;

[0350] 2) After ensuring that the cells were in good growth condition, a part of the cells was taken out into a new culture flask as the cells to be transfected, and the density was adjusted to 1×10^ 5 cells / mL;

[0351] 3) When the cell density reached 4 - 5×10^ 5 cells / mL, electroporation could be carried out;

[0352] 4) On the day of electroporation, prepare a 12-well plate, add 200 μL of blank opti-MEM for each sample, and place it in the incubator for preheating;

[0353] 5) Prepare 1 sterile EP tube, configure the electroporation solution according to the number of samples (20 μL / sample), and add the supplement reagent to the solution reagent at a ratio of 1:4.5;

[0354] 6) Count the cells to be electroporated using a flow cytometer to calculate the cell density;

[0355] 7) Calculate based on the number of 1×10^ 6 cells per sample, and take out the required cell volume;

[0356] 8) Centrifuge at 300 g for 5 min at room temperature, and aspirate the supernatant completely after centrifugation;

[0357] 9) Repeat the steps in 8) with pre-warmed PBS at 37 °C for another wash.

[0358] 2.7.1.3 Ribonucleoprotein (RNP) Complex Preparation

[0359] 1) Slowly add the Cas9 protein to the sgRNA in a ribonuclease-free (RNase-Free) EP tube according to the number of samples to be transfected. The specific dosage is as follows:

[0360] Table 10

[0361]

[0362] 2) Incubate at room temperature for 10 min.

[0363] 2.7.2 Electroporation

[0364] 1) On the day of electroporation, turn on the electroporator and set the type of "nucleocuvette" and "celltypeprogram" to be used;

[0365] 2) Resuspend the cells gently with the prepared electroporation solution according to the amount of 17 μL of electroporation solution required for each sample, taking care not to generate bubbles;

[0366] 3) Add the RNP complex to form the electroporation system according to the following table:

[0367] Table 11

[0368]

[0369] 4) Carefully add the prepared electroporation system to the electrode plate strip and gently tap to make the sample fill the bottom of the container;

[0370] 5) Place the electrode strips into the electroporator. Immediately after electroporation, resuspend the cells in 200 μL preheated blank Opti-MEM in a 24-well plate and place in a 37°C, 5% CO 2 incubator;

[0371] 6) After standing for 20 minutes, add 500 μL of preheated X-VIVO complete medium and place at 37°C with 5% CO 2 The incubator was used for subsequent experimental verification.

[0372] 2.8 Flow cytometry

[0373] 1) Collect 1×10 5 -1×10 6 The cells to be tested were centrifuged at 1500 rpm for 5 min and the supernatant was discarded carefully;

[0374] 2) Wash twice with 500 μL flow cytometry wash buffer;

[0375] 3) Prepare the staining system: 50 μL flow cytometry buffer + 0.25 μL flow cytometry antibody for each sample. If multiple antibodies need to be stained simultaneously, they can be added to the same 50 μL flow cytometry buffer after staggering the fluorescence spectra of the antibodies for simultaneous staining.

[0376] 4) Add 50 μL of the staining system to the cell pellet and mix thoroughly by pipetting to resuspend the cells.

[0377] 5) Incubate at 4°C in the dark for 20 min;

[0378] 6) Add 500 μL of flow cytometry wash solution and centrifuge at 1500 rpm for 5 minutes;

[0379] 7) Repeat step 6) and wash once more;

[0380] 8) Resuspend the cell pellet in 200 μL of flow cytometry wash buffer, add 1 μL of 7-AAD, and analyze on the flow cytometer after 5 minutes.

[0381] 2.9 Western blot (WB)

[0382] 2.9.1 Cell protein extraction and protein quantification

[0383] 1) Collect 1×10 6 Transfer cells to EP tubes, centrifuge at 1000 g for 5 min at room temperature, and discard the supernatant;

[0384] 2) Wash twice with pre-cooled PBS, 1000g, 5min, and remove the supernatant;

[0385] 3) Add RIPA lysis buffer containing 1% phenylmethanesulfonyl fluoride (PMSF) to the cell pellet, mix well by shaking, and place on ice for 30 min for lysis.

[0386] 4) Centrifuge at 13,000 g at 4 °C for 15 min, and discard the supernatant.

[0387] 5) Take 10 μL of the supernatant for measuring protein concentration by the BCA method.

[0388] 6) Add 0.25 volume of 5× protein loading buffer to the remaining protein supernatant in the EP tube, mix well by shaking, heat at 100 °C for 10 min in a metal bath, briefly centrifuge, and then place on ice for standby or store in a -80 °C refrigerator.

[0389] 2.9.2 WB Detection

[0390] 1) Preparation of SDS-polyacrylamide gel: Install the gel plate on a special gel-making rack, check the airtightness, and prepare the separating gel and stacking gel according to the formula in Table 1.3.1.

[0391] 2) Install the gel plate: Place the prepared gel plate into the electrophoresis tank, add 1× electrophoresis buffer to cover the gel plate, and carefully pull out the comb.

[0392] 3) Loading: Accurately and sequentially add the adjusted-volume samples and protein marker into the sample wells.

[0393] 4) Electrophoresis: The initial voltage is a constant 80 V for about 25 min. When the dye enters the separating gel and forms a straight line, or the marker spreads out, increase the voltage to 120 - 140 V until the bromophenol blue reaches the bottom of the separating gel to end the electrophoresis.

[0394] 5) Transfer: Pre-cool the prepared 1× transfer buffer on ice. Immerse the polyvinylidene fluoride (PVDF) membrane in methanol for 10 - 15 s for activation, and then form a "sandwich" structure in the order of the positive electrode of the transfer clamp, one layer of sponge, two layers of filter paper, PVDF membrane, gel, two layers of filter paper, one layer of sponge, and the negative electrode of the transfer clamp from the positive electrode to the negative electrode, paying attention to exhausting the air bubbles. Install the transfer clamp, place the transfer device in an ice bath, then connect the power supply, and transfer at a constant current of 220 mA for 90 - 120 min.

[0395] 6) Blocking: After the transfer is completed, take out the PVDF membrane, immerse it in 1× TBST containing 5% non-fat milk powder, and slowly incubate at room temperature on a shaker for 40 - 50 min. After that, wash it 3 times with 1× TBST solution by shaking, 10 min for each wash.

[0396] 7) Primary antibody incubation: Cut the membrane according to the position of the bands, and then place the cut membranes into 50 mL centrifuge tubes containing the primary antibody solution, close to the tube wall, and rotate and incubate overnight on a shaker at 4°C to allow sufficient binding of the antigen and antibody.

[0397] 8) Secondary antibody incubation: Take out the PVDF membrane and wash it 3 times with 1×TBST for 10 minutes each time. Place it into a 50 mL centrifuge tube containing the secondary antibody solution, gently shake on a shaker, and incubate at room temperature for 40 - 50 minutes. After that, wash it 4 times with 1×TBST for 10 minutes each time.

[0398] 9) Color development: Prepare the color development solution in advance and place it in the dark. Mix the bands and the color development solution well and then expose them. Set the machine to continuous and interval exposure, and save the original high-resolution pictures.

[0399] 2.10 Luciferase cytotoxicity assay based on fluorescent substrate method

[0400] 1) Calculate the number of target cells and effector cells: Seed 1×10 4 target cells per well, set 4 gradients of effector-to-target ratios (20:1, 10:1, 5:1, 2.5:1), calculate according to 5 replicates for each gradient. The number of target cells required is 1×10 4 ×[5 (replicates) × 4 (gradients) × 2 (groups) + 5 (minimum killing wells) + 5 (maximum killing wells)] = 5×10 5 cells. Resuspend the target cells with 100 μL of X-VIVO complete medium per well, then the required volume of the medium is 5 mL; the number of effector cells required is 1×10 4 ×20 (the highest effector-to-target ratio is 20:1) × 5 (replicates) × 2 (gradient dilutions) = 2×10 6 cells, and the required volume of the resuspended medium is 1 mL.

[0401] 2) Take the target cells and effector cells according to the calculated cell amounts, and resuspend the target cells and effector cells with 10% Gibco X-VIVO medium in the calculated volumes.

[0402] 3) Gradient dilution of effector cells: Prepare 3 15 mL centrifuge tubes, add 500 μL of 10% Gibco X-VIVO medium to each tube, and name them 10:1, 5:1, 2.5:1 respectively. Take 500 μL of the cell suspension from the well-resuspended effector cells (20:1) and add it to the 10:1 tube. After thorough mixing, take 500 μL and add it to the 5:1 tube. After thorough mixing, take 500 μL and add it to the 2.5:1 tube, and pipette and mix well to form a concentration gradient.

[0403] 4) Plating: Seed 100 μL of target cells into a 96-well white opaque microplate, and then seed effector cells at each dilution gradient into the corresponding wells at 100 μL / well.

[0404] 5) For the minimum killing well (Kmin), add 100 μL of culture medium. For the maximum killing well (Kmax), add 2.5 μL of 10% Triton-X-100 and 97.5 μL of culture medium to each well.

[0405] 6) Incubate in an incubator at 37 °C and 5% CO 2 for 4 h.

[0406] 7) After incubation, centrifuge at 400 g for 5 min and carefully discard the supernatant.

[0407] 8) Add 100 μL of potassium fluorescein (1.5 mg / mL) to each well, mix well, and incubate at 37 °C in the dark for 10 min.

[0408] 9) Detect the fluorescence intensity using a fluorescence microplate reader.

[0409] 10) Calculation method of killing efficiency: (Kmin - K (fluorescence value of a certain well)) / (Kmin - Kmax).

[0410] 2.11 Cytokine detection

[0411] In this experiment, a human cytokine Flex kit (BD Biosciences) was used to detect supernatant cytokines by flow cytometry bead array (CBA). The specific steps are as follows:

[0412] 1) Prepare the standard curve according to the instructions.

[0413] 2) Prepare the Beads mixture: Calculate the number of samples to be detected, add 5 μL of each type of Beads for each sample, and mix A1 - A6 Beads.

[0414] 3) Prepare several flow tubes, add 25 μL of the mixed Beads mixture to each tube, then add 25 μL of the sample to be detected, and finally add 25 μL of PE detection reagent. After mixing well, incubate in the dark for 3 h.

[0415] 4) Add 500 μL of Wash Buffer to each tube and centrifuge at 300 g for 5 min.

[0416] 5) Carefully discard the supernatant, resuspend the cells and magnetic beads with 200 μL of Wash Buffer, and then load onto the machine under the same conditions as the standard curve, such as the same voltage.

[0417] 6) Use FCAP (v3.0) software to calculate the concentration of each cytokine in each sample.

[0418] 2.12 CFSE Proliferation Assay

[0419] 1) Harvest the target cells. After centrifuging at 1500 rpm for 5 min, resuspend them in PBS at a concentration of 1×10 7 / mL, add 1.2 μL of mitomycin, and incubate in a 37°C incubator for 3 h;

[0420] 2) Harvest the effector cells. After centrifuging at 1500 rpm for 5 min, resuspend the cells in PBS at a concentration of 1×10 6 / mL, add 1 μL of CFSE dye (5 mM), and incubate in the dark at 37°C for 20 min;

[0421] 3) After the treatment of effector cells and target cells is completed, wash the cells twice with PBS, and resuspend the cells in 10%

[0422] Gibco X-VIVO medium at a concentration of 5×10 5 / mL. Pipette 100 μL of effector cells and target cells into each well, that is, at an effector-to-target ratio of 1:1, with 5×10 4 effector and target cells per well, and plate in a 200 μL system;

[0423] 4) At 0 h, 24 h, 48 h, 72 h, and 96 h, collect the cells to detect the CFSE fluorescence intensity. During the culture, supplement 10% Gibco X-VIVO medium according to the cell density, and transfer the cells to a 12-well plate, a 6-well plate, and a T25 cell culture flask for culture in sequence.

[0424] 2.13 In Vitro Repeated Antigen Stimulation Assay

[0425] 1) At an effector-to-target ratio of 1:1, take 5×10 5 tumor cells (pretreated with mitomycin) and effector cells and co-culture them in a 37°C incubator for 24 hours;

[0426] 2) After overnight incubation, collect the supernatant, and re-inoculate the same number of mitomycin-treated tumor cells into the incubation system and culture for another 24 hours;

[0427] 3) Repeat the steps in 2);

[0428] 4) Collect the CAR-T cells every day for flow cytometry analysis, and analyze the supernatant using a human cytokine Flex kit (BD Biosciences) by flow cytometry bead array (CBA) according to the manufacturer's instructions.

[0429] 2.14 Animal Experiment

[0430] This study was approved by the Animal Ethics Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine. On D-3 days, female NSG mice aged 6-8 weeks were injected with 5×10 4 Raji-luc-GFP to establish a tumor model. On D-1 day, after intraperitoneal injection of 3 mg potassium fluorescein for 10 min, the IVIS lumina II small animal in vivo imager was used to confirm successful tumor engraftment, and the mice were randomly divided into 2 groups (7×19 CAR-T, THEMIS-7×19 CAR-T), with 5 mice in each group. On D0 day, 5×10 5 7×19-CAR-T and THEMIS-7×19-CAR-T cells were administered via the tail vein. Subsequently, fluorescence quantitative monitoring of tumor growth in mice was performed twice a week, and the survival time of the mice was observed.

[0431] 2.15 Statistical analysis

[0432] All data in this study were statistically analyzed using Graphpad Prism (version 10). All data were tested for normality by the Shapiro-Wilk test, and then the corresponding statistical test methods were selected according to whether they conformed to normality. For data conforming to a normal distribution, the independent samples t-test was used for comparison between two groups. For data not conforming to a normal distribution, the Mann-Whitney U test was used. The paired t-test was used for comparison of the same group of samples under different conditions. For comparison among multiple groups, one-way analysis of variance (ANOVA) and Bonferroni post hoc test were used for normal distribution data. For non-normal distribution data, the Kruskal-Wallis test and Dunn post hoc test were used. P<0.05 was considered statistically significant, * for P<0.05, ** for P<0.01, *** for P<0.001, **** for P<0.0001, and ns for P>0.05.

[0433] 3 Experimental results

[0434] 3.1 Changes in the function of the fourth-generation CAR-T cells after THEMIS knockout

[0435] 3.1.1 Knockout of THEMIS does not affect the expression of CAR in the fourth-generation CAR-T cells

[0436] In the first part of the study, our multi-omics analysis of four generations of CAR-T cell products suggested that the T cell subset marked by THEMIS is the key subset for maintaining the long-term efficacy of the four generations of CAR-T cells; external databases also proved that THEMIS is highly expressed in the CR group, indicating that THEMIS may be a key factor determining the efficacy of CAR-T. Our team previously constructed a fourth-generation CAR-T cell with the scFv of a CD19 monoclonal antibody as the antigen-binding domain, 4-1BB as the co-stimulatory domain, and the introduction of cytokines and chemokines (IL-7 and CCL19), namely 7×19CAR-T( ​ A), and completed a clinical study. To verify the effect of THEMIS on the function of the fourth-generation CAR-T, in this project, we also constructed 7×19CAR-T cells as the research object. By means of lentiviral transfection, we successfully prepared 7×19CAR-T cells, and the proportion of CAR reached more than 70%( ​ B), and the activated T cells without CAR transfection were used as the control group.

[0437] We used the electroporation combined with CRISPR-Cas9 gene editing technology to knockout the THEMIS gene of 7×19CAR-T cells on the 6th - 8th day, and the AAVS1 safe harbor locus knockout group was used as the control. On the 11th - 14th day, we verified the knockout efficiency at the transcriptional and translational levels respectively( ​ C). To further evaluate the effect of electroporation and the knockout of THEMIS on CAR expression, we measured and compared the proportion of CAR in the AAVS1 KO group and the THEMIS KO group( ​ D), and the results showed that compared with the negative control group, electroporation and the knockout of THEMIS did not affect CAR expression (P>0.05).

[0438] 3.1.2 The CD8 / CD4 ratio of the fourth-generation CAR-T cells decreased after THEMIS knockout

[0439] In addition, we further evaluated the change in the CD8 / CD4 ratio in 7×19CAR-T cells after THEMIS knockout( ​ A). The results showed that compared with the control group, the CD8 / CD4 ratio in the THEMIS knockout group decreased (P<0.05)( ​ B), suggesting that THEMIS may be related to maintaining the CD8 / CD4 homeostasis of T cells.

[0440] 3.1.3 The proliferation and expansion ability of the fourth-generation CAR-T cells with THEMIS knockout was limited after antigen stimulation

[0441] To evaluate the proliferation of 7×19 CAR-T cells after knocking out THEMIS under antigen-unstimulated and stimulated conditions, we labeled the 7×19 CAR-T cells in the AAVS1 KO group and the THEMIS KO group with CFSE respectively. The antigen-stimulated group was co-incubated with CD19-3T3 cells at an effector-to-target ratio of 1:1, and the CFSE fluorescence intensity of the two groups of cells was monitored at different time points. As ​ shown in A, there was no difference in cell proliferation between the two groups under antigen-unstimulated conditions. However, at 96 h after antigen stimulation, the proliferation of 7×19 CAR-T cells in the THEMIS KO group was slower than that in the AAVS1 KO group (the peak was shifted to the right).

[0442] Meanwhile, to further simulate the environment of tumor antigen stimulation, we co-incubated the two groups of cells with mitomycin-treated Raji cells at an effector-to-target ratio of 1:1 and counted the two groups of cells every day. ​ As shown in B, there was no significant difference in the total number of cells between the two groups on the 1st - 2nd day after co-incubation. However, on the 3rd - 5th day, the difference in the total number of cells between the THEMIS KO group and the AAVS1 KO group gradually increased, and the difference was the most significant on the 5th day (P < 0.001).

[0443] Therefore, the proliferation ability of 7×19 CAR-T cells after knocking out THEMIS was restricted after antigen stimulation.

[0444] 3.1.4 Knocking out THEMIS does not affect the short-term tumor killing and cytokine secretion of fourth-generation CAR-T cells

[0445] To further explore the effect of knocking out THEMIS on the killing of B-line lymphoma cells by 7×19 CAR-T cells, we selected the Jeko-1 and Raji cell lines as target cells for evaluation. We overexpressed luciferase-P2A-GFP in the two B-line lymphoma cell lines and constructed cell lines Jeko-1-luc-GFP and Raji-luc-GFP expressing luciferase and GFP genes by flow sorting for luciferase killing experiments. We co-incubated the 7×19 CAR-T cells in the AAVS1 KO group and the THEMIS KO group with different target cells at different gradient effector-to-target ratios for 4 hours and then detected the killing ability. The results showed that there was almost no difference in the killing effect between the two groups. Even at high effector-to-target ratios (20:1 or 10:1), the difference in the killing rate ratio between the two groups was small ( ​ A). Meanwhile, we detected the cytokines in the supernatant of the two groups of cells after killing tumor cells at a high effector-to-target ratio (20:1) using a Multiplex Bead Assay kit based on flow cytometry. ​As shown in B, there were no differences in the secretion levels of IL-2, IL-4, IL-6, TNF-α, and IFN-γ in the supernatants of the two groups after killing tumor cells for 4 hours (P>0.05). Only the secretion level of IL-10 in the THEMIS KO group was higher than that in the AAVS1 KO group (P<0.05).

[0446] From the above, we found that the knockout of THEMIS had basically no obvious effect on the short-term killing of tumors by 7×19 CAR-T cells and the secretion levels of cytokines.

[0447] 3.1.5 Knockout of THEMIS limits the long-term tumor-killing ability of fourth-generation CAR-T cells

[0448] In the above part, we found that the knockout of THEMIS hardly affected the short-term function of 7×19 CAR-T cells in killing B-line lymphoma cells. Considering that the ultimate recurrence of the treated patients is closely related to the long-range killing function of CAR-T cells, we further evaluated the long-range tumor-killing ability of 7×19 CAR-T cells after THEMIS knockout. We selected CD19-3T3 cells as target cells, and co-incubated the 7×19 CAR-T cells of the AAVS1 KO group and the THEMIS KO group with them at a low effector-to-target ratio (1:2 and 1:1), and used the RTCA system to monitor the long-range killing function. ​ As shown in A, under the condition of an effector-to-target ratio of 1:2, within 10 hours of co-incubation (within 20 hours after the start of the experiment), there were no obvious differences in the killing ability of target cells between the two groups. However, as the co-incubation time was extended to 20 hours later, we found that the killing ability of 7×19 CAR-T cells in the THEMIS KO group gradually weakened, while the AAVS1 KO group could maintain a continuous killing effect, and the maximum difference in killing ability could reach more than 30% (P<0.0001). Similarly, under the condition of an effector-to-target ratio of 1:1, we also obtained consistent results ( ​ B, P<0.0001). The above results indicate that although the knockout of THEMIS does not affect the short-term killing ability of 7×19 CAR-T cells, it significantly limits its long-term tumor-killing function.

[0449] 3.1.6 The ability of fourth-generation CAR-T cells to respond to repeated antigen stimulation is weakened after THEMIS knockout

[0450] In the above experiments, we have revealed that the knockout of THEMIS weakens the long-range killing ability of 7×19 CAR-T cells. To further investigate the effects of THEMIS knockout on the memory, exhaustion, and apoptosis phenotypes of 7×19 CAR-T cells under the condition of persistent antigen stimulation, we performed multiple antigen stimulations on the knockout group and the control group, and used flow cytometry analysis to detect the memory (CD45RO, CD62L), exhaustion (TIGIT, TIM3), and apoptosis phenotypes of the two groups. We co-incubated the 7×19 CAR-T cells of the AAVS1 KO group and the THEMIS KO group with mitomycin-treated Raji cells at an effector-to-target ratio of 1:1, and detected the relevant phenotypes of the two groups of cells after 24 hours, and then added the same number of tumor cells again, repeating the stimulation 3 times. The results showed that there was no significant difference in the proportion of memory T cell subsets (CD45RO + CD62L + ) between the two groups under the first antigen stimulation (P>0.05); while after 3 tumor antigen stimulations, the proportion of memory T cell subsets of 7×19 CAR-T cells in the THEMIS KO group was significantly lower than that in the AAVS1 KO group (P<0.05, ​ A). Moreover, after 3 tumor antigen stimulations, the proportion of exhaustion-related T cell subsets (TIGIT + or TIM3 + ) in the THEMIS KO group (P<0.05, ​ B) and the apoptosis rate were significantly higher than those in the AAVS1 KO group (P<0.01, ​ C, D), suggesting that after THEMIS knockout, the ability of 7×19 CAR-T cells to form memory stem T cells is weakened under repeated tumor antigen stimulation, and they are more likely to be exhausted and apoptotic.

[0451] 3.1.7 Knockout of THEMIS causes excessive secretion of cytokines in fourth-generation CAR-T cells and exhaustion

[0452] Previous studies have reported that the THEMIS-SHP1 complex can reduce cytokine secretion to a certain extent. Therefore, we hypothesized whether the easier exhaustion and apoptosis of 7×19 CAR-T cells caused by THEMIS knockout under repeated tumor antigen stimulation are related to excessive cytokine secretion. For this, we collected the supernatants of the two groups after the first (Stim1) and the third antigen stimulation (Stim3), and also detected 6 inflammation-related cytokines (IL-2, IL-4, IL-6, IL-10, TNF, and IFN-γ). By ​As shown in

[0453] A, when first stimulated by tumor antigens, the secretion levels of IL-2 (P<0.0001), IL-4 (P<0.01), IL-6 (P<0.01), IL-10 (P<0.0001), TNF (P<0.01), and IFN-γ (P<0.0001) in 7×19 CAR-T cells of the THEMIS KO group were all higher than those of the AAVS1 KO group. ​ Notably, as we have revealed in ​ D, after 3 antigen stimulations, the exhaustion phenotype and apoptosis ratio of 7×19 CAR-T cells in the THEMIS KO group were higher. Correspondingly, we found that after 3 antigen stimulations, compared with the AAVS1 KO group, the secretion of IL-2 (P<0.001), IL-4 (P<0.0001), IL-6 (P<0.01), IL-10 (P<0.01), TNF (P<0.01), and IFN-γ (P<0.0001) in 7×19 CAR-T cells of the THEMIS KO group significantly attenuated ( ​ B), and the secretion levels of IL-2 (1425.0±40.0 vs 2430.8±433.7, P<0.05) and IL-4 (38.6±5.3 vs 71.4±5.3, P<0.001) were even lower than those of the AAVS1 KO group (

[0454] A).

[0455] 3.2 Functional changes of four generations of CAR-T cells after overexpressing THEMIS

[0456] 3.2.1 Construction and preparation of 7×19 CAR-T cells overexpressing THEMIS

[0457] In the previous section, we verified that the knockout of THEMIS would cause a downregulation of the CD8 / CD4 ratio in 7×19 CAR-T cells and limit the proliferation and expansion of 7×19 CAR-T cells after antigen stimulation. At the same time, we found that although the knockout of THEMIS did not affect the short-term tumor-killing ability of 7×19 CAR-T cells, it significantly restricted their long-range killing effect and reduced their ability to respond to repeated antigen stimulation. Under repeated antigen stimulation, the ability of THEMIS KO 7×19 CAR-T cells to form memory stem T cells was impaired, and they showed excessive cytokine secretion under initial stress, leading to premature exhaustion and apoptosis. As can be seen from the above, THEMIS is closely related to the differentiation, proliferation, and continuous anti-tumor effect of fourth-generation CAR-T cells. Therefore, we plan to construct 7×19 CAR-T cells overexpressing THEMIS, namely THEMIS-7×19 CAR-T, to further explore the synergistic effect of this gene on the anti-B cell lymphoma of fourth-generation CAR-T cells.

[0458] In this study, using 7×19 CAR as a control, we ligated the coding sequence (CDS) of the THEMIS protein downstream of the CD3ζ signal transduction domain of the CAR through the variable spliceosome T2A. The specific schematic diagram of the structure of THEMIS-7×19 CAR is as ​ shown in A. After gradient density centrifugation and CD3 magnetic bead sorting of peripheral blood, we obtained T cells, which were then co-incubated with CD3 / CD28 magnetic beads for 12 - 24 hours for stimulation and activation. On the second day, the activated T cells were separated using a magnetic bead sorting column, and the activated T cells were transfected with the THEMIS-7×19 CAR lentivirus with the VSVG envelope protein. On the fifth day, the magnetic beads were removed, and the transduction efficiency was measured. As ​ shown in B, the transduction efficiencies of THEMIS-7×19 CAR and 7×19 CAR both reached more than 50%. To further clarify the overexpression of THEMIS in THEMIS-7×19 CAR-T cells, we performed fixed permeabilization staining, used the THEMIS-specific antibody to bind to the intracellular THEMIS protein, and labeled it with a specific fluorescent secondary antibody for flow cytometry detection. The results showed that the expression level of the THEMIS protein in the T cells transfected with THEMIS-7×19 CAR was higher than that of the control group (P < 0.05, ​ C).

[0459] 3.2.2 Upregulation of the CD8 / CD4 ratio in fourth-generation CAR-T cells after overexpression of THEMIS

[0460] Similarly, we detected CD8 + T and CD4+ The proportion of T cells ( ​ ). The results showed that compared with the control group, the CD8 / CD4 ratio in THEMIS-7×19 CAR-T cells was upregulated (P<0.05), indicating that THEMIS is related to the differentiation of T cells.

[0461] 3.2.3 Enhanced proliferative ability of four generations of CAR-T cells after antigen stimulation following overexpression of THEMIS

[0462] In 3.1.3, we found that knockout of THEMIS would limit the proliferative ability of 7×19 CAR-T after antigen stimulation. Therefore, to further clarify the relationship between THEMIS and the proliferative ability of 7×19 CAR-T cells, we measured the proliferation of THEMIS-7×19 CAR-T and the control group (7×19 CAR-T) under antigen-unstimulated and stimulated conditions in the same way. As ​ shown in A, there was no difference in cell proliferation between the two groups without the addition of CD19-3T3 antigen stimulation, while the proliferation of THEMIS-7×19 CAR-T cells (i.e., the THMEIS OE group in the figure) was stronger than that of the control group (peak shifted to the left) after 96 h in the antigen-stimulated group.

[0463] Meanwhile, we also co-incubated the two groups of cells with mitomycin-treated Raji cells to simulate a real tumor antigen stimulation environment and counted them every day. As ​ shown in B, there was no significant difference in the total number of cells between the two groups on the first day (P>0.05), but the difference in the total number of cells between the overexpression group and the control group became gradually obvious from the 3rd to the 5th day (P<0.01).

[0464] From the above, we clarified that overexpression of THEMIS can promote the proliferation and expansion of 7×19 CAR-T cells after antigen stimulation, and also proved that the expression of THEMIS is related to the proliferation of four generations of CAR-T cells.

[0465] 3.2.4 Overexpression of THEMIS mainly enhances the long-term tumor killing ability of four generations of CAR-T cells

[0466] In 3.1, we found that the knockout of THEMIS mainly restricted the long-term tumor-killing ability of the fourth-generation CAR-T (7×19 CAR-T) cells. To further verify the relationship between THEMIS and the killing function of 7×19 CAR-T cells, we also evaluated the short-term and long-term killing functions of the THEMIS overexpression group and the control group of 7×19 CAR-T cells. Consistent with 3.1.4, we used the Jeko-1-luc-GFP and Raji-luc-GFP cell lines as target cells, and co-incubated the two groups of cells with different target cells at different gradient effector-to-target ratios for 4 hours, and then detected the killing ability. As ​ shown in A, there was almost no difference in the short-term killing effect between the two groups (P>0.05). Even when there was a difference in the killing ability at a high effector-to-target ratio (20:1) in the Jeko-1-luc-GFP group (P<0.01), the absolute difference in the killing rate between the two groups was also less than 5%.

[0467] Similarly, we used CD19-3T3 cells as target cells, selected an effector-to-target ratio of 1:1, co-incubated THEMIS-7×19 CAR-T and 7×19 CAR-T cells with it, and used the RTCA system to monitor the long-term killing function. ​ As shown in B, within 10 hours of co-incubation (within 20 hours from the start of the experiment), there was no obvious difference in the killing ability of the target cells between the two groups. As the co-incubation time prolonged, the THEMIS-7×19 CAR-T cells still maintained a strong killing ability, while the killing effect of the control group on tumor cells gradually decreased below that of the overexpression group, and the difference was the most obvious after 30 hours of co-incubation (P<0.0001). The above results prove that the overexpression of THEMIS mainly enhanced the long-term tumor-killing ability rather than the short-term killing ability of the fourth-generation CAR-T cells, and also verified the important role of THEMIS in the persistent tumor-killing effect of the fourth-generation CAR-T cells.

[0468] 3.2.5 Enhancement of the ability of the fourth-generation CAR-T cells to respond to continuous antigen stimulation after THEMIS overexpression

[0469] In addition, we also further studied the effects of THEMIS overexpression on the memory, exhaustion, and apoptosis phenotypes of the fourth-generation CAR-T cells under the condition of long-term antigen stimulation. We performed 3 times of tumor antigen stimulation on the two groups of cells, and used flow cytometry analysis technology to detect the memory (CD45RO, CD62L), exhaustion (TIGIT, TIM3), and apoptosis phenotypes of the two groups of cells. ​As can be seen in A and D, there was no significant difference in the proportion of memory T cell subsets between the two groups under the first antigen stimulation (P>0.05); while after receiving three tumor antigen stimulations, the proportion of memory T cell subsets in the overexpression group was significantly higher than that in the control group (P<0.05), and the proportion of exhaustion-related T cell subsets (P<0.05, ​ B) and the apoptosis rate (P<0.001, ​ C) were significantly lower than those in the control group. The above results indicate that the overexpression of THEMIS promotes the formation of memory stem T cells in the fourth-generation CAR-T cells under repeated tumor antigen stimulations and enhances their anti-exhaustion and anti-apoptosis abilities.

[0470] 3.2.6 The overexpression of THEMIS can effectively control the cytokine secretion of the fourth-generation CAR-T

[0471] In part 3.1, we found that the knockout of THEMIS would lead to excessive cytokine secretion in the fourth-generation CAR-T cells under repeated tumor antigen stimulations, resulting in exhaustion. To further verify the relationship between THEMIS and the cytokine secretion of the fourth-generation CAR-T cells, we analyzed the cytokines in the supernatant of the overexpression group and the control group after the first and third antigen stimulations. As ​ shown in A, under the first antigen stimulation, the secretion levels of IL-2 (P<0.01), IL-4 (P<0.001), IL-6 (P<0.05), TNF (P<0.0001) and IFN-γ (P<0.0001) in the THEMIS-7×19 CAR-T cells were lower than those in the control group. However, in 3.2.4, we have demonstrated that there was no significant difference in the short-term tumor killing ability between the two groups. This indicates that the cytokines in the control group were in a state of excessive secretion.

[0472] While in ​ A, we further found that after receiving three antigen stimulations, except for IL-10, the overexpression group was still able to maintain a certain cytokine secretion level, and the degree of cytokine secretion exhaustion was significantly lower than that in the control group ( ​ B), especially for IL-2 (P<0.0001), TNF (P<0.0001) and IFN-γ (P<0.0001). After multiple antigen stimulations, the secretion levels in the control group had significantly decayed and were lower than those in the overexpression group.

[0473] From the above, we have demonstrated that the overexpression of THEMIS can keep the secretion of inflammatory cytokines in the fourth-generation CAR-T during the initial immune response under control, and further indicates that there is a close relationship between the THEMIS gene and the cytokine secretion of CAR-T cells.

[0474] 3.2.7 The overexpression of THEMIS can enhance the in vivo tumor killing ability of the fourth-generation CAR-T cells

[0475] In the above experiments, we have confirmed that 7×19CAR-T cells overexpressing THEMIS have better long-term tumor killing ability in vitro and can secrete cytokines to a limited extent. To further evaluate whether THEMIS-7×19CAR-T cells have better anti-tumor effects in vivo, we injected 5×10 4 Raji cells carrying the luciferase gene (Raji-luc-GFP) were used to construct a lymphoma-bearing mouse model. After the successful construction of the model was confirmed by fluorescence imaging analysis, 5×10 5 7×19CAR-T and THEMIS-7×19CAR-T cells were used to monitor the growth of tumor cells in mice by fluorescence imaging, and the survival time was observed to evaluate the in vivo anti-tumor effect of THEMIS-7×19CAR-T cells ( ​ A). The results showed that the tumor signal of the overexpression group mice was significantly lower than that of the control group. The early tumor signal of the overexpression group mice was significantly suppressed, but recurrence and progression occurred in the later stage ( ​ B / 7C). However, the overall survival time of mice in the overexpression group was significantly longer than that in the control group (P < 0.01). In addition, the median survival time of mice in the overexpression group was 64 days, longer than the 48 days in the control group.

[0476] discuss

[0477] The emergence of CAR-T cell therapy has ushered in a new era of cancer treatment. However, although a number of clinical trials have shown that CAR-T cell therapy offers hope of cure for patients with relapsed and / or refractory lymphoma, some patients still face the problem of relapse. In addition, in CAR-T cell clinical trials, the most common CAR-T cell-specific side effects, namely cytokine release syndrome and neurotoxicity, occurred in 42-100% and 2-4% of patients, respectively. Severe cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) (≥ grade 3) occurred in 46% and 50% of treated patients, respectively. Therefore, it is crucial to further identify the key factors that maintain the long-term efficacy of CAR-T cells and thereby improve their efficacy and reduce their treatment side effects.

[0478] In the single-cell sequencing analysis of the first part of this project, the CD8+T_THEMIS subset was discovered, and the key marker gene THEMIS related to T cell function and CAR-T cell efficacy was identified. To further verify the regulatory effect of this gene on the function of the fourth-generation CAR-T cells, the THEMIS gene in 7×19CAR-T cells was knocked out using the electroporation combined with CRISPR-Cas9 technology in this part of the project. On the other hand, the THEMIS-7×19CAR-T cells overexpressing THEMIS were constructed to explore the mechanism of this gene in improving the long-term efficacy of the fourth-generation CAR-T cells and reducing treatment side effects.

[0479] Previous studies have reported that in CAR-T cells with 4-1BB as the co-stimulatory domain, the THEMIS-SHP1 complex can be recruited to the CAR signalosome through the intracellular domain of 4-1BB, inhibiting CAR-CD3ζ phosphorylation, and knocking out THEMIS or SHP1 can enhance the basal phosphorylation level, leading to increased cytokine secretion. However, its effects and mechanisms on CAR-T cells, especially the fourth-generation CAR-T cells in terms of memory, exhaustion, apoptosis, and cytokines have not been specifically clarified. Our results show that in the fourth-generation CAR-T cells with 4-1BB as the co-stimulatory molecule, THEMIS can regulate T cell differentiation, affect the CD8 / CD4 ratio and the proliferation ability after antigen stimulation. The fourth-generation CAR-T cells (THEMIS-7×19CAR-T) overexpressing this gene can proliferate faster when stimulated by antigens. Literature reports that the killing ability of CAR-T cells with 4-1BB as the co-stimulatory domain is related to their proliferation ability. Although the killing speed of a single cell is slow, its strong proliferation ability and cooperative killing characteristics endow it with more effective continuous tumor killing ability. In addition, our results prove that the effect of THEMIS on the effector function of the fourth-generation CAR-T cells is mainly reflected in the long-range killing effect rather than the short-term killing ability. The CAR-T cells overexpressing THEMIS can more effectively maintain the long-range tumor killing ability. In vivo experiments also prove that the THEMIS-7×19CAR-T cells have a stronger anti-tumor effect than the control group in the lymphoma-bearing mouse model, significantly prolonging the survival time of the mice.

[0480] As the most critical immune cells in anti-tumor immune responses, T cells can produce cytokines with chemotactic, pro-inflammatory, and immunoprotective effects to coordinate the adaptive immune system and directly exert cytotoxic responses against tumor cells. Currently, multiple studies have found that the high-level secretion of IL-2, IL-6, TNF, and IFN-γ in serum during CAR-T therapy is associated with the risk of severe CRS and ICANS, especially IL-6. Our study found that the fourth-generation CAR-T cells with THEMIS knocked out over-secrete IL-2, IL-4, IL-6, TNF, and IFN-γ, but the secretion levels rapidly decline after repeated antigen stimulation. After overexpressing this gene, the fourth-generation CAR-T cells can effectively control the secretion of the above-mentioned inflammatory cytokines, without affecting the short-term tumor killing ability of the fourth-generation CAR-T cells, but instead promoting the long-term tumor killing ability. In addition, multiple characteristics of T cells have been found to significantly affect the efficacy of CAR-T cell therapy, especially the exhaustion of T cells, which is closely related to tumor recurrence. Our research results show that the knockout of THEMIS leads to impaired ability of the fourth-generation CAR-T cells to form memory cells under continuous antigen stimulation, is more likely to express immunosuppressive receptors, and is more prone to apoptosis; after overexpressing THEMIS, its adaptability to continuous antigen stimulation is significantly enhanced, and a higher proportion of memory T cell populations can be formed, and it is not easily exhausted and apoptotic.

[0481] In summary, we found the regulation of the long-term killing ability of the fourth-generation CAR-T cells after knocking out and overexpressing THEMIS, and this gene can effectively control the cytokine secretion of CAR-T cells without affecting the tumor killing activity, which may improve the severe cytokine side effects in the clinical CAR-T cell process.

[0482] Of course, our current results do not reveal the specific molecular mechanisms of the expression regulation of the THEMIS gene in aspects such as the memory stemness and exhaustion degree of CAR-T cells. By deeply analyzing the data of bulk RNA sequencing and single-cell transcriptome sequencing, we are trying to find candidate downstream molecules of the THEMIS protein that may affect the exhaustion and long-term killing of the fourth-generation CAR-T cells.

[0483] 5 Results and Conclusions

[0484] Results: (1) Knockout of the THEMIS gene decreased the CD8 / CD4 ratio in the fourth-generation CAR-T cells, while overexpression upregulated this ratio; (2) The short-term tumor-killing ability of the fourth-generation CAR-T cells overexpressing the THEMIS gene showed no significant change, but the long-term tumor-killing effect was more persistent, the proportion of memory phenotype was upregulated, and the proportions of exhausted phenotype and apoptosis were decreased. Knocking out this gene showed the opposite effect; (3) The secretion of inflammatory cytokines by the fourth-generation CAR-T cells overexpressing the THEMIS gene decreased under single antigen stimulation, but the attenuation degree was lower than that of the control group under repeated stimulation (P<0.01). The effect of knocking out this gene was the opposite. (4) The overall survival time of the mice in the fourth-generation CAR-T cell group overexpressing the THEMIS gene was significantly longer than that of the control group, with statistical difference (P<0.01). The median survival time was 64 days, longer than 48 days of the control group.

[0485] Conclusions: 1) The THEMIS gene is involved in maintaining the CD8 / CD4 ratio in the fourth-generation CAR-T cells; 2) The THEMIS gene can promote the proliferation of the fourth-generation CAR-T cells after antigen stimulation and enhance their ability to respond to repeated antigen stimulation; 3) The THEMIS gene can regulate the long-range tumor-killing ability of the fourth-generation CAR-T cells; 4) The THEMIS gene can regulate the limited secretion of cytokines by the fourth-generation CAR-T cells. 5) The THEMIS-7×19 CAR-T cells have stronger anti-tumor effects than the control group in the lymphoma-bearing mouse model, significantly prolonging the survival time of the mice.

[0486] Part III: Target sequence, destination plasmid, CAR-T cell construction and corresponding sequences

[0487] (I) CAR-T cell preparation process

[0488] 1 Lentivirus packaging

[0489] 1.1 Transformation

[0490] 1) Take 1 tube of competent cells from the -80°C refrigerator and thaw them on ice;

[0491] 2) Take 20 μL of competent cells and add them to a 1.5 mL EP tube, then add 1 μL of the destination plasmid, gently flick to mix, incubate on ice for 30 min, heat shock in a 42°C metal bath for 90 s, and incubate on ice again for 2 min;

[0492] 3) Add 900 μL of liquid LB broth medium without ampicillin, pipette to mix, culture in a shaker at 37°C and 235 rpm for 45 min;

[0493] 4) Centrifuge at 6000 rpm for 5 min, discard 700 μL of the supernatant, and resuspend the pellet;

[0494] 5) Pipette 50 μL and drop it onto an LB broth agar plate containing ampicillin. Spread it evenly with an L-shaped spatula, place it upright in an incubator at 37 °C for 1 h, then invert it and incubate for 12 - 16 h.

[0495] 1.2 Plasmid extraction

[0496] Take the plasmid transformed the previous day, pick a single colony into a test tube containing 5 mL of LB liquid medium, and add 5 μL of ampicillin (100 mg / mL). Incubate it on a shaker at 37 °C and 235 rpm for 6 - 8 h. After 6 - 8 h, pour the 5 mL of bacteria-containing LB medium into 200 mL of LB liquid medium, and add 200 μL of ampicillin. Incubate it on a shaker at 37 °C and 235 rpm for 12 - 15 h, and then perform large-scale plasmid extraction:

[0497] 1) Aliquot 200 mL of the bacterial solution into 50 mL centrifuge tubes, centrifuge at 8000 rpm for 3 min, and discard the supernatant;

[0498] 2) Resuspend the bacterial cells in the 4 centrifuge tubes with 8 mL of P1 solution and transfer them to 1 50 mL centrifuge tube. Vortex to completely suspend the cells;

[0499] 3) Add 8 mL of P2 solution and invert the tube 6 times to fully mix and lyse the cells. Let it stand at room temperature for 5 min; then add 8 mL of P4 solution and invert the tube 6 times until white flocculent precipitate is observed. Let it stand at room temperature for 10 min;

[0500] 4) Centrifuge at 8000 rpm for 10 min to collect the flocculent precipitate at the bottom of the tube. Pour all the supernatant into CS1 and push the handle to collect the filtrate;

[0501] 5) Add 0.3 volume of isopropanol to the filtered bacterial solution and invert to mix;

[0502] 6) Prepare adsorption column CP6, add 3 mL of equilibration solution BL to fully wet the filter membrane, centrifuge at 8000 rpm for 3 min, and discard the waste liquid in the collection tube;

[0503] 7) Add the 8 mL of bacterial solution with isopropanol added to CP6, centrifuge at 8000 rpm for 3 min, discard the waste liquid in the collection tube, and put the adsorption column back into the collection tube;

[0504] 8) Repeat step 7) until all the bacterial solution has passed through the column;

[0505] 9) Add 10 mL of washing solution (with anhydrous ethanol added), centrifuge at 8000 rpm for 2 min, discard the waste liquid in the collection tube, and put the adsorption column back into the collection tube;

[0506] 10) Repeat step 9) once;

[0507] 11) Add 3 mL of absolute ethanol to the adsorption column, centrifuge at 8000 rpm for 2 min, discard the waste liquid in the collection tube, and place the adsorption column back into the collection tube;

[0508] 12) Centrifuge at 8000 rpm for 5 min to completely remove the washing buffer in the adsorption column;

[0509] 13) Place the adsorption column CP6 into a new collection tube, open the lid and air-dry for 5 min to allow the absolute ethanol to fully evaporate;

[0510] 14) Suspend and add 500 μL - 1 mL of TB elution buffer dropwise to the center of the adsorption column filter membrane, let it stand at room temperature for 10 min, centrifuge at 8000 rpm for 5 min, and the liquid in the collection tube is the plasmid we need;

[0511] 15) Use a Nanodrop micro-spectrophotometer to detect the concentration and OD 260 / 280, and store at -20 °C for later use.

[0512] 1.3 Lentivirus packaging

[0513] 1) Culture 1×10 7 293T cells in a 10 cm dish 1 day in advance;

[0514] 2) Replace the cell culture medium with pre-warmed 10% FBS Opti-MEM medium 2 h in advance;

[0515] 3) Plasmid preparation: Add 500 μL of Opti-MEM medium to a 15 mL centrifuge tube. According to the molar ratio of pMDLg / pRRE:pRSV-Rev:pMD2.G:target plasmid = 1:1:0.5:2, add plasmids at 10 μg / dish, calculate the required mass and volume of each packaging plasmid and the target plasmid, add them to the Opti-MEM medium, mix well and incubate at room temperature for 5 min;

[0516] 4) Add 10 μL of Neofectamine, mix well and incubate at room temperature for 17 min. After incubation, add it to the 293T supernatant;

[0517] 5) Collect the supernatant after 48 h, and supplement 10 mL / dish of 10% FBS Opti-MEM medium again. Collect the supernatant again at 72 h;

[0518] 6) Centrifuge the collected supernatant at 1500 rpm for 5 min and filter it using a 0.45 μm filter membrane;

[0519] 7) Centrifuge at 4000 g, 4 °C for 12 h;

[0520] 8) After centrifugation, discard the supernatant, resuspend the virus pellet with the virus preservation solution (concentrated at 1:250), aliquot and store at -80°C.

[0521] 1.4 Virus titer determination

[0522] 1) After cell counting, seed 2×10 5 cells per well, with 4 gradients and 2 replicates. For Jurkat cells, considering cell loss, seed Jurkat cells in 10 wells, i.e., 2×10 6 cells;

[0523] 2) Centrifuge at 1500 rpm for 5 min, discard the supernatant, resuspend with complete 1640 medium, and add 2 μL of polybrene. Add 100 μL per well to a 96-well plate.

[0524] 3) Add 90 μL of virus preservation solution to 10 μL of the original virus solution to dilute 10-fold.

[0525] 4) Add the original virus solution, preservation solution, and 1640 medium according to the system in the following table to determine the titer:

[0526] Table 12

[0527]

[0528] 5) Centrifuge at 1200 g for 1.5 h at 32°C;

[0529] 6) After centrifugation, place at 37°C in a 5% CO 2 incubator and incubate for 4 h;

[0530] 7) After 4 h, centrifuge at 1500 rpm for 5 min, discard the supernatant, resuspend with 2 mL of complete 1640 medium and transfer to a 24-well plate;

[0531] 8) After 48 h, detect the positive rate by flow cytometry. Titer = [(2×10 5 × positive rate) / 0.2] × dilution factor.

[0532] 2 CAR-T cell preparation

[0533] 2.1 Isolation of human peripheral blood mononuclear cells (PBMC)

[0534] 1) Add 10 - 20 mL of healthy human peripheral blood to an equal volume of PBS and mix well;

[0535] 2) Prepare 2 50-mL centrifuge tubes and add 20 mL of human peripheral blood lymphocyte separation medium

[0536] 3) Slowly and carefully layer the diluted peripheral blood along the wall of the tube onto the surface of the human peripheral blood lymphocyte separation medium;

[0537] 4) Centrifuge at 800 g, acceleration 1, deceleration 0, at room temperature for 20 min;

[0538] 5) Prepare 2 15-mL centrifuge tubes, add 5 - 8 mL of PBS, carefully aspirate the buffy coat (i.e., PBMC) and add it to the PBS, and mix well;

[0539] 6) Centrifuge at 1500 rpm for 5 min;

[0540] 7) Carefully discard the supernatant, and freeze the cell pellet in cell cryopreservation medium for later use or use directly.

[0541] 2.2 Isolation of CD3 + T cells

[0542] 1) Resuspend the PBMC in 50 mL of X-VIVO complete medium, mix well by pipetting, and add it to a T225 flask;

[0543] 2) Incubate at 37 °C, 5% CO 2 in an incubator for at least 6 h, then carefully turn the culture flask over;

[0544] 3) After 2 h, carefully stand the culture flask on its side, and pour the cell suspension along the side into a new T225 flask in a biosafety cabinet, and incubate at 37 °C, 5% CO 2 in an incubator;

[0545] 4) Carefully turn the flask over every 2 h. When both sides of the culture flask are fully adhered, replace it with a new T225 flask. After adhering to 2 - 3 T225 flasks, observe the proportion of adherent cells under a microscope. When the proportion of adherent cells < 10%, take a small amount of cells for CD3 detection. If CD3 + > 90%, then T cell activation can be carried out;

[0546] 2.3 T cell activation

[0547] 1) Collect the CD3 + > 90% T cells obtained in the previous step, and centrifuge at 1500 rpm for 5 min;

[0548] 2) Carefully discard the supernatant, resuspend the cell pellet in an appropriate amount of X-VIVO complete medium, mix well, and perform cell counting;

[0549] 3) Calculate the required volume of CD3 / CD28 Dynabeads based on the cell number, that is, 30% × total cell number / CD3 / CD28 Dynabeads density. At the same time, centrifuge the cell suspension at 1500 rpm for 5 min;

[0550] 4) Wash the CD3 / CD28 Dynabeads: Prepare one 15 mL centrifuge tube, add 1 mL of PBS, and then add the calculated amount of CD3 / CD28 Dynabeads from the previous step. Place it on a magnetic stand and let it stand for 5 minutes.

[0551] 5) Aspirate and discard the PBS in the 15 mL centrifuge tube (keep the 15 mL centrifuge tube on the magnet).

[0552] 6) Remove the 15 mL centrifuge tube from the magnetic stand. You can see the iron-red CD3 / CD28 Dynabeads on the tube wall.

[0553] 7) Resuspend the cell clumps with complete X-VIVO medium at a density of 4 - 6×10 6 / mL.

[0554] 8) Add the cell suspension to the washed CD3 / CD28 Dynabeads. After pipetting and mixing evenly, incubate it on a rotary shaker for 30 minutes to allow the cells to come into full contact with the CD3 / CD28 Dynabeads.

[0555] 9) After the rotation ends, place the 15 mL centrifuge tube on the magnetic stand and let it stand for 5 minutes.

[0556] 10) Carefully discard the supernatant.

[0557] 11) Remove the 15 mL centrifuge tube. Resuspend the pellet with X-VIVO medium at a density of 1 - 3×10 6 / mL of CD3 / CD28 Dynabeads, and transfer it to a T25 flask for culturing for at least 8 hours. At this time, it is named D0.

[0558] 2.4 Lentiviral transfection

[0559] 1) Collect the activated T cells obtained in the previous step and centrifuge them at 1500 rpm for 5 minutes.

[0560] 2) Resuspend them with an appropriate amount of complete X-VIVO medium and then perform cell counting.

[0561] 3) Take 1×10 6 activated T cells, centrifuge them at 1500 rpm for 5 minutes. The remaining cells are used as the control group (control T cells, CT).

[0562] 4) Carefully discard the supernatant, add fourth-generation CD19-CAR or CD19-CAR-THEMIS lentivirus at 30 moi (multiplicity of infection), supplement the system to 200 μL with X-VIVO blank medium, add 0.2 μL of polybrene (8 mg / mL), add it to one well of a 96-well plate, and carefully seal the 96-well plate with sealing film;

[0563] 5) Centrifuge at 1200 g and 32 °C for 1.5 h;

[0564] 6) After centrifugation, incubate at 37 °C and 5% CO 2 for 4 h;

[0565] 7) Centrifuge at 1500 rpm for 5 min, carefully discard the supernatant, resuspend the cells in 2 mL of X-VIVO complete medium in a 24-well cell culture plate and culture, named D1.

[0566] 2.5 CAR-T Cell Culture

[0567] 1) On day D3, use flow cytometry to detect the transduction efficiency, supplement 8 mL of X-VIVO medium, transfer it to a T25 flask and continue to culture;

[0568] 2) On day D4, after thoroughly pipetting and mixing the cell suspension, transfer it to a 15 mL centrifuge tube, place it on a magnetic stand and let it stand for 5 min to remove CD3 / CD28 Dynabeads, and collect the supernatant as the required cells; CT cells are processed in the same way;

[0569] 3) During the medium period from D3 to D10, supplement the existing culture system with an equal volume of X-VIVO complete medium every other day, and transfer it to T75 and T175 flasks in sequence according to the culture system, or partially freeze it for later use;

[0570] 4) On days D11 - D14, conduct in vitro experiments or animal experiments.

[0571] (II) Construction of the Target Plasmid

[0572] 1. The construction process specifically includes the following operations based on the already constructed 7×19 target plasmid and the specific restriction enzyme digestion information (as ​ shown): 1). Insert the THEMIS_CDS sequence into the target plasmid behind 7×19 (MluI single restriction enzyme site); 2). Connect the THEMIS_CDS sequence (as ​ shown) to the 7×19 sequence in the CAR plasmid with T2A, and the constructed target plasmid is shown in Figure (5 - 3); 3). Codon optimization.

[0573] 2. The specific corresponding sequences are as follows:

[0574] 2.1: The 7×19 CAR DNA sequence is shown in SEQ ID NO.3:

[0575] gctagccccggggccaccatggccctgcctgtgaccgctctgctgctccctctggccctgctgctgcacgctgccaggcctgatatccagatg

[0576] acccagaccacctccagcctgtccgcctccctcggcgatagggtgaccatcagctgcagggccagccaggacatcagcaagtatctgaactg

[0577] gtaccagcagaagcccgacggcaccgtgaaactgctcatctaccatacctccaggctccacagcggcgtgccttccaggtttagcggctccg

[0578] gctccggcaccgactattccctcaccatcagcaatttagaacaagaggatatcgccacctatttctgccagcagggcaacaccctgccctacac

[0579] ctttggcggcggcaccaagctggagattacaggaggcggcggctccggaggcggaggctccggcggcggaggaagcgaagtgaagctg

[0580] caggaaagcggacctggactcgtggcccctagccagagcctgagcgtgacatgtaccgtgtccggcgtgtccctgcctgactacggagtctc

[0581] ctggatcaggcaaccccctagaaagggtttagaatggctcggcgtgatttggggcagcgagaccacctactacaacagcgccctgaagagca

[0582] gactgaccatcatcaaggacaactccaagtcccaggtgtttctgaagatgaactccctgcagaccgacgataccgccatctactactgcgccaa

[0583] gcactactactatggcggctcctacgccatggactactggggacaaggaacctccgtgacagtgtccagcaccaccacccctgctcctagacc

[0584] ccctacccctgctcccacaattgccagccagcctctgagcctgagacccgaagcctgcagacctgctgccggaggagccgtgcacaccagg

[0585] ggcctggacttcgcctgcgacatctacatttgggctcccctcgctggaacctgtggcgtgctgctgctgtccctggtgattaccctgtactgcaag

[0586] aggggcaggaagaagctgctgtacatcttcaaacagcccttcatgagacccgtgcagaccacccaagaagaggatggctgcagctgcagatt

[0587] ccccgaagaggaggagggaggctgcgagctgagggtgaagtttagcagaagcgccgacgctcccgcttaccagcagggacagaaccagc

[0588] tgtataacgagctgaacctcggcagaagagaggagtacgacgtgctggataagaggaggggcagagaccctgagatgggcggcaagccta

[0589] ggagaaaaaacccccaggagggactgtacaatgagctgcagaaagataagatggccgaggcctatagcgagatcggaatgaagggcgaa

[0590] aggaggaggggcaagggacacgacggcctgtaccagggcctctccacagccaccaaggacacctacgacgccctccatatgcaggccct

[0591] gcctcctaggggcagcggcgccaccaacttttctttactgaagcaagccggtgacgtggaggagaaccccggccccatgttccacgtgtcctt

[0592] cagatacatcttcggtttaccccctctcattttagtgctgctgcccgttgccagcagcgactgcgacatcgaaggcaaagacggcaagcagtatg

[0593] aaagcgtgctgatggtgtccatcgaccagctgctcgactccatgaaggagatcggcagcaactgtttaaacaacgaattcaacttcttcaagag

[0594] gcacatctgtgacgccaacaaggagggcatgtttttattcagagccgctcgtaagctgaggcagtttttaaaaatgaactccaccggcgacttcg

[0595] atttacatctgctgaaggtgtccgagggcaccaccattttactgaattgcaccggccaagttaagggaagaaagcccgctgctttaggcgaagc

[0596] ccagcccacaaagtctttagaggagaacaaatctttaaaggagcagaagaagctgaacgacctctgctttttaaaaaggctgctgcaagaaatc

[0597] aagacatgctggaacaagattttaatgggcaccaaagaacacggctccggcgaaggcagaggctctttactgacttgtggagacgtggaaga

[0598] gaaccccggtcccatggctctgctgctcgctttatctttactggtgctgtggacaagccccgcccctactttaagcggaaccaacgacgccgag

[0599] gactgctgtttaagcgtgacccaaaagcccatccccggttacatcgtgaggaactttcactacttattaatcaaggatggctgtcgtgtgcccgct

[0600] gtggtgttcaccactttaagaggcagacagctgtgcgctcctcccgaccagccttgggtggagagaatcatccagaggctgcagaggaccag

[0601] cgctaagatgaagaggaggagctcctga

[0602] 2.2: The amino acid sequence of 7×19 CAR is shown in SEQ ID NO.4:

[0603] ASPGATMALPVTALLLPLALLLHAARPDIQMTQTTSSLSASLGDRVTISCRASQDISKYLNWYQQKPDGTVKLLIYHTSRLHSGVPSRFSGSGSGTDYSLTISNLEQEDIATYFCQQGNTLPYTFGGGTKLEITGGGGSGGGGSGGGGSEVKLQESGPGLVAPSQSLSVTCTVSGVSLPDYGVSWIRQPPRKGLEWLGVIWGSETTYYNSALKSRLTIIKDNSKSQVFLKMNSLQTDDTAIYYCAKHYYYGGSYAMDYWGQGTSVTVSSTTTPAPRPPTPAPTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYQQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRGSGATNFSLLKQAGDVEENPGPMFHVSFRYIFGLPPLILVLLPVASSDCDIEGKDGKQYESVLMVSIDQLLDSMKEIGSNCLNNEFNFFKRHICDANKEGMFLFRAARKLRQFLKMNSTGDFDLHLLKVSEGTTILLNCTGQVKGRKPAALGEAQPTKSLEENKSLKEQKKLNDLCFLKRLLQEIKTCWNKILMGTKEHGSGEGRGSLLTCGDVEENPGPMALLLALSLLVLWTSPAPTLSGTNDAEDCCLSVTQKPIPGYIVRNFHYLLIKDGCRVPAVVFTTLRGRQLCAPPDQPWVERIIQRLQRTSAKMKRRSS*

[0604] 2.3: THE DNA sequence of THEMIS-7×19 CAR is shown in SEQ ID NO.5:

[0605]

[0606] 2.4: The amino acid sequence of THEMIS-7×19CAR is shown in SEQ ID NO.6:

[0607] ASPGATMALPVTALLLPLALLLHAARPDIQMTQTTSSLSASLGDRVTISCRASQDISKYLNWYQQKPDGT

[0608] VKLLIYHTSRLHSGVPSRFSGSGSGTDYSLTISNLEQEDIATYFCQQGNTLPYTFGGGTKLEITGGGGSGG

[0609] GGSGGGGSEVKLQESGPGLVAPSQSLSVTCTVSGVSLPDYGVSWIRQPPRKGLEWLGVIWGSETTYYNS

[0610] ALKSRLTIIKDNSKSQVFLKMNSLQTDDTAIYYCAKHYYYGGSYAMDYWGQGTSVTVSSTTTPAPRPPTP

[0611] APTIASQPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFKQP

[0612] FMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYQQGQNQLYNELNLGRREEYDVLDKRRGR

[0613] DPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQA

[0614] LPPRGSGATNFSLLKQAGDVEENPGPMFHVSFRYIFGLPPLILVLLPVASSDCDIEGKDGKQYESVLMVSI

[0615] DQLLDSMKEIGSNCLNNEFNFFKRHICDANKEGMFLFRAARKLRQFLKMNSTGDFDLHLLKVSEGTTIL

[0616] LNCTGQVKGRKPAALGEAQPTKSLEENKSLKEQKKLNDLCFLKRLLQEIKTCWNKILMGTKEHGSGEG

[0617] RGSLLTCGDVEENPGPMALLLALSLLVLWTSPAPTLSGTNDAEDCCLSVTQKPIPGYIVRNFHYLLIKDGC

[0618] RVPAVVFTTLRGRQLCAPPDQPWVERIIQRLQRTSAKMKRRSSEGRGSLLTCGDVEENPGPMALSLEEFV

[0619] HSLDLRTLPRVLEIQAGIYLEGSIYEMFGNECCFSTGEVIKITGLKVKKIIAEICEQIEGCESLQPFELPMNFP

[0620] GLFKIVADKTPYLTMEEITRTIHIGPSRLGHPCFYHQKDIKLENLIIKQGEQIMLNSVEEIDGEIMVSCAVAR

[0621] NHQTHSFNLPLSQEGEFYECEDERIYTLKEIVEWKIPKNRTRTVNLTDFSNKWDSTNPFPKDFYGTLILKP

[0622] VYEIQGVMKFRKDIIRILPSLDVEVKDITDSYDANWFLQLLSTEDLFEMTSKEFPIVTEVIEAPEGNHLPQS

[0623] ILQPGKTIVIHKKYQASRILASEIRSNFPKRHFLIPTSYKGKFKRRPREFPTAYDLEIAKSEKEPLHVVATKA

[0624] FHSPHDKLSSVSVGDQFLVHQSETTEVLCEGIKKVVNVLACEKILKKSYEAALLPLYMEGGFVEVIHDKK

[0625] QYPISELCKQFRLPFNVKVSVRDLSIEEDVLAATPGLQLEEDITDSYLLISDFANPTECWEIPVGRLNMTVQ

[0626] LVSNFSRDAEPFLVRTLVEEITEEQYYMMRRYESSASHPPPRPPKHPSVEETKLTLLTLAEERTVDLPKSPKRHHVDITKKLHPNQAGLDSKVLIGSQNDLVDEEKERSNRGATAIAETFKNEKHQK*

[0627] 2.5: The DNA sequence corresponding to the THEMIS-7×19 CAR-T original plasmid is shown in SEQ ID NO.7:

[0628]

[0629] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Finally, it should be stated that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A recombinant construct, characterized in that, A chimeric antigen receptor that can simultaneously express IL-7, CCL19, and THEMIS, its fragments or variants.

2. The recombinant construct according to claim 1, wherein, The chimeric antigen receptor is an anti-CD19 chimeric antigen receptor.

3. The recombinant construct according to claim 2, wherein The nucleic acid sequence of the recombinant construct is as shown in SEQ ID NO.5, its fragments or variants.

4. A recombinant construct according to claim 3, wherein The amino acid sequence encoded by the recombinant construct is as shown in SEQ ID NO.6, its fragments or variants.

5. A pharmaceutical composition, characterized in that: Comprising a recombinant construct according to any one of claims 1 to 4 and at least one pharmaceutically acceptable carrier.

6. A CAR-T cell, characterized in that, Comprising a recombinant construct according to any one of claims 1 to 4.

7. A CAR-T cell according to claim 6, wherein, The CAR-T cells are autologous, allogeneic or xenogeneic.

8. Use of a recombinant construct according to any one of claims 1 to 4, or a pharmaceutical composition according to claim 5, or a CAR-T cell according to any one of claims 6 to 7 in the preparation of a kit for diagnosing / treating a malignant tumor.

9. Use of a recombinant construct according to any one of claims 1 to 4, or a pharmaceutical composition according to claim 5, or a CAR-T cell according to any one of claims 6 to 7 in the preparation of a drug for diagnosing / treating a malignant tumor.

10. Use according to claim 9 in the preparation for the diagnosis / treatment of malignant tumors, characterized in that, The malignant tumor is a hematological malignancy.

Images