Preparation method of natural ligand mediated multi-target recognition regulable gene engineering immune cell

Abstract

The invention discloses a preparation method of a natural ligand mediated multi-target recognition regulable gene engineering immune cell. In order to solve the problems of effectiveness and safety of CAR-T treatment, the inventor separately constructs a recognition part of a growth factor to a target spot, CAR-immune cells and an activation structural domain, in tumor targeted treatment, tag molecule modified growth factors are adopted as regulatory protein, tag specific CAR is constructed on the immune cells, and the CAR-T treatment efficiency is improved. The novel multi-target recognition adjustable gene engineering immune cell is prepared, so that the multi-target stable recognition of tumor-associated antigens is realized, the safety controllability is improved, and the difficulty that the CAR-T cell of a series antibody is high in risk in the treatment process is overcome.

Description

Technical Field

[0001] The present invention relates to adopting natural ligand-mediated genetically engineered immune cells to improve the multi-target recognition activity and safety controllability of anti-tumor immunotherapy. Background Art

[0002] In recent years, cancer immunotherapy has attracted much attention and has become the focus of cancer treatment. In December 2011, Nature and the top clinical oncology journal Journal of Clinical Oncology published a review article with the same title "Cancer immunotherapy comes of age", which evaluated immunotherapy as the only method in existing technology that can completely eliminate cancer cells. This method makes up for the shortcomings of traditional therapies and is considered to be the most active and promising treatment method in the comprehensive treatment model of tumors in the 21st century.

[0003] Chimeric Antigen Receptor-T (CAR-T) cell technology is an immunotherapy technology that uses genetically modified T cells to specifically recognize tumor-specific or related antigens, exert toxic effects on tumor cells, and thus exert anti-tumor effects. The concept of CAR-T immunotherapy was first proposed by Dotti (1) and others. It has been nearly 30 years since the breakthrough progress in the treatment of leukemia in clinical trials. After continuous scientific research and clinical transformation, CAR-T technology has undergone four generations of innovation (2). However, the vast majority of CAR molecular structures currently used in clinical practice are still second-generation CAR molecular structures, and commonly used co-stimulatory molecules include CD28, CD137, OX40, etc. Compared with the classic T cell receptor (T cell receptor, TCR)-human leukocyte antigen (human leukocyte antigen, HLA) complex recognition mode, the recognition of tumor cells by CAR molecules depends on the specificity of antibodies, so it is simpler and clearer, and at the same time avoids the problem of tumor immune escape caused by downregulation of HLA molecule expression.

[0004] CAR is divided into three parts: intracellular signal peptide region, transmembrane region and extracellular antigen binding region: The extracellular antigen binding region binds to antigens to exert specificity. The transmembrane region of CAR is composed of signal molecules, such as CD3, CD3ζ, CD4, CD28, etc. The intracellular signal peptide region includes tyrosine on the T cell receptor CD3ζ chain or FcεRIγ, etc. CAR-modified T cells are T cells selected from the patient's peripheral blood cells, and then CAR is transfected into T cells using genetic engineering technology to form chimeric antigen receptor T cells (3).

[0005] CAR gene-modified T cells are transfected with lentivirus. The modified T cells carry antigens that specifically recognize tumor cells and have the ability to self-replicate and kill tumor cells. Therefore, CAR-T specifically recognizes tumor cells, kills tumor cells in a targeted manner, and reduces the impact of the tumor microenvironment. CAR-T is not only not affected by MHC restriction, but also recognizes diversified antigens, without the risk of TCR mismatch. CAR-T can also directly inhibit the growth of tumor cells. CAR-T secretes cell factors such as interferon-γ (IFN-γ), thereby inducing macrophages to secrete interleukin 6 (IL-6), interleukin 10 (IL-10), and interleukin 12 (IL-12) to inhibit the growth of tumor cells (4). In addition to its own unique killing effect, CAR-T cells can also activate the body's immune system and exert an anti-tumor effect.

[0006] Compared with the rapid development of CAR-T cell therapy in blood tumors, the clinical effect of CAR-T cells in the treatment of solid tumors is far from satisfactory. Compared with blood tumors, solid tumors have complex tissue components and lack ideal specific targets. In order to achieve the ideal therapeutic effect, there are currently as many as 22 different targets (the number even exceeds the number of blood tumors) being tried for CAR-T cell therapy of solid tumors. Among them, the more developed targets include epidermal growth factor receptor (EGFR), human epidermal growth factor receptor 2 (HER2) and mesothelin (MSLN).

[0007] In the studies that have been reported so far, a single target is usually used, but traditional single-target CAR-T has many disadvantages, such as: easy to escape immune system. The difficulty faced by CAR-T cell therapy for tumors is the problem of therapeutic targets. Compared with blood tumors, it is more difficult to select a suitable target for complex solid tumors. There are two main reasons for this. First, in blood tumors, such as ALL, almost all tumor cells express CD19 or CD20, so CAR-T cell therapy targeting these two antigens is expected to eliminate all tumor cells, thereby bringing about the cure of blood tumors. In contrast, the cell composition in solid tumor tissues is complex, and different tumor cells often have different protein expression profiles. It is difficult for a single target to cover all tumor cells. Both theoretical and practical experience show that even if tumor cells expressing a certain antigen can be effectively eliminated, there will still be a large number of antigen-negative tumor cells remaining. This incomplete phenotypic coverage cannot meet the requirements of tumor cure. Secondly, it is difficult to find a tumor-specific antigen target in solid tumors. Most of the existing CAR-T cell treatment experiments for solid tumors select antigens that are highly expressed in tumors, while these antigens are often still expressed at low levels in some normal tissues, such as HER2 and EGFR. This greatly increases the possibility of CAR-T cells attacking normal tissues in the treatment of solid tumors, which is called on-target toxicity. As the anti-tumor function of CAR-T cells increases, the harmfulness of side effects caused by off-target effects will also increase. In response to these problems, the current effective solution is to adopt a multi-target strategy, that is, multi-target CAR-T cell combination. Multi-target CAR-T cell combination can seek to cover a higher proportion of tumor cells. The coverage of multiple targets can increase the proportion of CAR-T cells killing tumor cells. The elimination of a higher proportion of tumor cells is expected to improve the survival benefits of patients. This multi-target (dual-target, or even multi-target) CAR-T cell therapy strategy is also used in the treatment of B-cell leukemia. For example, CAR-T cell therapy that simultaneously targets CD19 and CD20 is believed to be effective in preventing the recurrence of CD19-negative leukemia (5)(6). Multi-target coverage can be achieved by infusing mixed CAR-T cells. In addition, it can also be achieved by integrating two (or more) single-chain antibody (scFv) segments targeting different antigens on the CAR molecule (7), that is, the method of serial antibody. However, during the preparation process of dual-target CAR-immune cells, the mismatch or steric hindrance of the serially expressed antibodies on the cell surface usually reduces the targeting recognition function of CAR-immune cells, and even causes serious off-target toxicity. The use of natural ligands of cell growth factors to recognize multiple tumor-associated antigens has been reported. However, using growth factors as direct recognition sites for CAR-immune cells usually causes serious off-target toxicity because it is unable to distinguish between tumor tissue and normal tissue.Therefore, it is an important idea to select the natural ligands of growth factors that are recognized by appropriate tumor-associated antigens, and on this basis, to construct a regulatable CAR-T treatment strategy that separates tumor targeting recognition and CAR-T cell downstream signal transduction.

[0008] B-cell activating factor (BLyS) is a member of the tumor necrosis factor (TNF) family that regulates B-cell maturation, survival and function. They are overexpressed in a variety of autoimmune diseases. BLyS is a typical type II transmembrane protein with an N-terminus in the cell, lacking a signal peptide sequence, and a C-terminus outside the cell. BLyS is a single glycoprotein with a molecular weight of 31.3kD composed of 285 amino acids, of which amino acids 1 to 46 at the N-terminus are the intracellular region, amino acids 47 to 73 are the hydrophobic transmembrane region, amino acids 74 to 285 are the extracellular region, and amino acids 133 to 285 are the main functional regions. BLyS may have a glycosylation site (N242) and no disulfide bonds exist. The homology between humans and mice is 93%. BLyS has a strong B-cell tropism. As a lymphocyte co-stimulator, BLyS has a strong stimulating effect on activated B cells to promote proliferation and differentiation. Increased expression of BLyS and its receptor has been identified in many B-cell malignancies.

[0009] The biological effects of BLyS are mediated by three cell surface receptors, namely B-cell activating factor receptor (BAFF-R), transmembrane activator and CAML interactor (TACI), and B cell maturation antigen (BCMA). All three are mainly expressed on the surface of B cells and are type III membrane proteins. The extracellular region of the receptor has cysteine-rich regions (CRDs) that can form receptor-ligand structural motifs. Among them, BAFF-R may be a specific receptor that only exists on B cells, while TACI and BCMA are also receptors for APRIL, another member of the TNF superfamily (8).

[0010] Traditional CAR-T cells usually target a single tumor-associated antigen, so during tumor treatment, they usually face the problem of tumor antigen escape. The biology of BLyS used in this study is mediated by three cell surface receptors, which can target and recognize multiple tumor-associated antigen receptor molecules. It has multi-specificity and can reduce the possibility of tumor antigen escape, mismatch and off-target to a certain extent. In addition, compared with existing technologies, the targeting recognition molecule BLyS is a ligand molecule naturally present in the human body, which is not immunogenic, and increases the safety of CAR-T cells during treatment. There are currently no reports on chimeric antigen receptors (CARs) using BLyS as a targeting recognition molecule.

[0011] A proliferation-inducing ligand (APRIL), also known as TNFSF13, is an important member of the tumor necrosis factor family. APRIL is lowly expressed in most normal tissues and is mainly expressed on a variety of immune cells, such as dendritic cells, macrophages, monocytes, and T lymphocytes. APRIL promotes and participates in the proliferation, differentiation, and survival of lymphocytes, as well as inducing the production of immunoglobulins and class switching, by binding to receptors. In recent years, more and more studies have shown that APRIL is involved in the pathogenesis of a variety of autoimmune diseases (9).

[0012] APRIL contains 250 amino acids and is a type II transmembrane protein. It is divided into an intracellular region (N-terminal, without a signal peptide sequence), a transmembrane region (amino acids 29-49) and an extracellular region (amino acids 50-250). It has two forms: a membrane-bound type with a relative molecular mass of about 27,000 and an extracellular soluble type with a relative molecular mass of about 17,000. APRIL can bind to itself to form a homotrimer, or it can form a heterotrimer (one BAFF binds to two APRILs or two BAFFs bind to one APRIL) with the same family member TNFSFl3B, namely B-cell activation factor (BAFF). In addition, APRIL can also bind to TWEAK, a protein of the same family and whose gene is located upstream, to form a type II transmembrane protein.

[0013] As a cytokine with multiple functions, APRIL must bind to receptors on its target cells to exert its biological effects. Currently, there are two clear receptors for APRIL, TACI and BCMA. TACI is mainly expressed on the surface of some B cells and activated T cells, while BCMA is expressed on the surface of mature B cells. Both are common receptors of APRIL and BAFF, and both belong to the tumor necrosis factor receptor family. They can bind to tumor necrosis factor-related protein (TRAF) to promote activation, anti-apoptosis, and pro-proliferation signaling pathways. APRIL can promote and participate in lymphocyte proliferation, differentiation, survival, and induce immunoglobulin production and class switching by binding to receptors. Although there are reports that APRIL is used as the recognition domain of CAR for the preparation of traditional hematological tumor treatment CAR-T cells (10), APRIL-based CAR may cause serious on-target toxicity, such as B cell exhaustion or neurotoxicity. Therefore, optimizing the targeting strategy of APRIL and constructing regulatable CAR-T cells can solve this potential problem.

[0014] Vascular endothelial growth factor (VEGF) is a heparin-binding growth factor specific for vascular endothelial cells that can induce angiogenesis in vivo. It is secreted by certain tumor cells and promotes endothelial cell proliferation by binding to corresponding receptors on the vascular endothelium. It can also increase vascular permeability to allow endothelial cell migration, induce tumor angiogenesis, and maintain continued tumor growth. It is the most potent angiogenic factor discovered to date and is associated with many physiological and pathological processes (11). There are currently seven members of the VEGF family, namely VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E, VEGF-F, and placenta growth factor (PIGF). The human VEGF-A gene is located on chromosome 6p21.3. It is a single gene with a total length of 14kb and consists of 8 exons and 7 introns. The encoded product is a homodimeric glycoprotein of 34 to 45kD. VEGF can produce 5 isomers after transcriptional shearing, which are named VEGF-AA145, VEGF-A165, VEGF-A121, VEGF-A189 and VEGF-A206 according to the length of amino acids. Among them, VEGF-A121 is a weakly acidic polypeptide that does not bind to heparin, and VEGF-A165 is a basic protein with low affinity for heparin. Both are secreted in a soluble, freely diffusible form and are easy to reach target cells; while VEGF-A145, VEGF-A165 and VEGF-A206 have a high affinity for heparin, bind to the cell surface or cell matrix after secretion, and are cell-associated isomers. It has been confirmed that VEGF-A is a selective mitogen for endothelial cells. In addition to increasing the concentration of Ca2+ in the cytoplasm of endothelial cells and increasing the permeability of microvessels (mainly post-capillary veins and venules) to macromolecules, it can also make endothelial cells slender and stimulate their replication in various ways, stimulate glucose transport into endothelial cells, promote the displacement of endothelial cells, mouse monocytes and fetal bovine osteoblasts, change the pattern of endothelial cell gene activation, upregulate the expression of plasminogen activator (including urokinase type and tissue type) and its inhibitor PAI-I, and induce the expression of other endothelial cell proteases, interstitial collagenase and tissue factor. VEGF-A mediates many physiological and pathological angiogenesis, and its expression is also enhanced when tissue blood vessels proliferate. The expression of cells in the differentiated state of embryonic tissue is higher than that of adult and fully differentiated cells. Under physiological conditions, VEGF-A is expressed at high levels in the placenta, many embryonic tissues, and some normal adult tissues with physiological vascular proliferation (such as proliferative endometrium). In addition, it is also expressed at low levels in normal glomerular cells, myocardial cells, prostate epithelium, semen, and certain epithelial cells of the adrenal cortex and lungs of animals and adults.Under pathological conditions, VEGF-A is overexpressed in synovial cells of healing skin wounds, psoriasis, delayed allergic reactions, and rheumatoid arthritis.

[0015] The receptors of the VEGF family discovered so far are divided into tyrosine kinase receptors VEGFR1 (Flt1), VEGFR2 (KDR / Klk1), VEGFR3 (Flt4) and non-tyrosine kinase receptors: neuropilin (NRP1 and NRP2). Flt1, KDR, and Flt4 are all receptor tyrosine kinases (RTKs). The first two are mainly expressed on vascular endothelial cells, and the latter is mainly present in lymphatic endothelial cells. NP1 and NP2 are non-tyrosine protein kinase transmembrane receptors, containing long extracellular segments and short intracellular segments. They are expressed not only in endothelial cells, but also in some tumor cells. The main biological function of VEGF is achieved through VEGFR2. After binding to VEGF, VEGFR2 dimerizes and the intracellular tyrosine residues are phosphorylated. VEGFR2 has 6 confirmed self-phosphorylation sites, Tyr1054, Tyr951, Tyr996, Tyr1059, Tyr1175, and Tyr1214. Most of the functions of these sites are unknown, but it has been confirmed that Tyr951 is related to srchomology 2 and SH2 binding, and Tyr1175 is a phosphatase Cy (PLC-y) binding site. A large number of clinical studies have shown that various diseases in which the VEGF / VEGFR signaling pathway plays a central role in the angiogenesis process, such as cancer, various vascular endothelial growth factor signal inhibitors, including anti-VEGF neutralizing antibodies and VEGFR kinase / multikinase inhibitors, have been successfully developed and are now widely used in clinical practice. Although there are reports on using VEGF-A165 as the recognition domain of CAR to prepare traditional CAR-T cells targeting solid tumor blood vessels (12), traditional VEGF-A-based CAR may cause serious on-target toxicity, such as complications such as cardiovascular failure. In view of this, we decided to optimize the regulatable CAR-T cell strategy with VEGF as the targeting molecule.

[0016] References 1.Dotti G, Savoldo B, & Brenner M (2009) Fifteen years of gene therapy based on chimeric antigen receptors: "are we nearly there yet?".Hum Gene Ther20(11):1229-1239. 2.Chmielewski M,Hombach AA,&Abken H(2014)Of CARs and TRUCKs:chimericantigen receptor(CAR)T cells engineered with an inducible cytokine tomodulate the tumor stroma.Immunol Rev 257(1):83-90. 3.Chen F,et al.(2015)Construction of Anti-CD20 Single-Chain Antibody-CD28-CD137-TCRzeta Recombinant Genetic Modified T Cells and its TreatmentEffect on B Cell Lymphoma.Med Sci Monit 21:2110-2115. 4.Spear P,Barber A,Rynda-Apple A,&Sentman CL(2012)Chimeric antigenreceptor T cells shape myeloid cell function within the tumormicroenvironment through IFN-gamma and GM-CSF.J Immunol 188(12):6389-6398. 5.Zah E,Lin MY,Silva-Benedict A,Jensen MC,&Chen YY(2016)T CellsExpressing CD19 / CD20 Bispecific Chimeric Antigen Receptors Prevent AntigenEscape by Malignant B Cells.Cancer Immunol Res 4(6):498-508. 6.Schneider D,et al.(2017)A tandem CD19 / CD20 CAR lentiviral vectordrives on-target and off-target antigen modulation in leukemia cell lines.JImmunother Cancer 5:42. 7.Feng KC,et al.(2017)Cocktail treatment with EGFR-specific andCD133-specific chimeric antigen receptor-modified T cells in a patient withadvanced cholangiocarcinoma.J Hematol Oncol10(1):4. 8.Bossen C&Schneider P(2006)BAFF,APRIL and their receptors:structure,function and signaling.Semin Immunol 18(5):263-275. 9.Liu XG&Hou M(2013)Immune thrombocytopenia and B-cell-activatingfactor / a proliferation-inducing ligand.Semin Hematol 50Suppl 1:S89-99. 10.Lee L,et al.(2018)An APRIL-based chimeric antigen receptor fordual targeting of BCMA and TACI in multiple myeloma.Blood 131(7):746-758. 11.Lange C,Storkebaum E,de Almodovar CR,Dewerchin M,&Carmeliet P(2016)Vascular endothelial growth factor:a neurovascular target inneurological diseases.Nat Rev Neurol 12(8):439-454. 12. Niederman TM, et al. (2002) Antitumor activity of cytotoxic Tlymphocytes engineered to target vascular endothelial growth factor receptors. Proc Natl Acad Sci USA 99(10):7009-7014. Summary of the invention

[0017] In order to solve the effectiveness and safety problems of CAR-T therapy, the inventors of the present invention separately constructed the tumor targeting recognition part of the natural ligands (BLyS, APRIL and VEGF) and the activation domain of CAR-immune cells. In tumor targeted therapy, label molecules are used to modify the natural ligands of growth factors as regulatory proteins. By constructing label-specific CARs on immune cells, new multi-target recognition and controllable genetically engineered immune cells are prepared, thereby achieving stable recognition of multiple targets of tumor-associated antigens while improving their safety and controllability, so as to overcome the traditional tandem CAR structure that combines the tumor recognition domain and the immune cell activation domain together, thereby causing dangerous problems such as off-target toxicity that genetically engineered immune cells are prone to produce during the treatment process.

[0018] The technical solution adopted by the present invention is: A composition of anti-tumor CAR-immune cells, comprising CAR-immune cells and regulatory proteins constructed based on natural ligands, wherein the natural ligands are naturally occurring ligand molecules that can specifically bind to tumors or related receptors on the surface of their tissues; The CAR domain of the CAR-immune cell is connected to a recognition molecule, and the recognition molecule is used to specifically recognize a label molecule; the regulatory protein has a natural ligand molecule and a label molecule that can recognize a variety of tumors or their tissue-associated antigens, and the label molecule can specifically bind to the recognition molecule of the CAR extracellular recognition domain of the CAR-immune cell.

[0019] In some examples, the regulatory protein has natural ligand molecules that can recognize at least 2, 3 or 4 tumor or tissue-related antigens. The increase in the number of natural ligands on the regulatory protein allows it to bind more tightly to tumor cells, which is beneficial to increase its specificity for tumors and is beneficial to reduce or avoid dangerous problems such as off-target toxicity during use.

[0020] In some examples, the natural ligand of the regulatory protein that recognizes a tumor or tissue-associated antigen is at least one of BLyS, APRIL and VEGF.

[0021] In some examples, the natural ligand that recognizes tumor antigens in the regulatory protein is BLyS, which targets the extracellular binding domains of BAFF-R, BCMA and TACI.

[0022] In some examples, the natural ligand that recognizes the tumor antigen in the regulatory protein is APRIL, which targets the extracellular binding domain of BCMA and TACI.

[0023] In some examples, the natural ligand that recognizes the tumor antigen in the regulatory protein is VEGF, targeting the extracellular binding domains of VEGFR1, VEGFR2 and VEGFR3.

[0024] In some examples, the natural ligand molecule of the regulatory protein and the tag molecule are coupled via a hydrophilic bifunctional linker.

[0025] In some examples, the linker is a linker peptide.

[0026] In some examples, the connecting peptide is selected from a linear polypeptide, a β-fold peptide, or a coiled-coil peptide.

[0027] In some examples, the hydrophilic bifunctional linker is selected from molecules that can be linked via a click reaction.

[0028] In some examples, the tag molecule is selected from FITC, biotin, His tag, Flag tag, Myc tag, GCN4 tag, HA tag, V5 tag, chitosan binding protein (CBP) tag, maltose binding protein (MBP), glutathione-S-transferase (GST) and DNA Aptamer tag. The tag molecule can be selected according to the specific situation. When the regulatory protein is a fusion protein, a polypeptide tag molecule is preferred to facilitate fusion expression. In particular, the regulatory protein is expressed in the form of a fusion protein, and the tag molecule used is a tag molecule for protein separation and purification, which is also conducive to the purification of the fusion protein.

[0029] In some examples, the tag-specific recognition molecule in the CAR-immune cell is selected from a single-chain antibody, a protein ligand, a short peptide molecule or a nucleic acid molecule to specifically recognize and capture the tag molecule on the regulatory protein. In the presence of the regulatory protein, the CAR-immune cell can mediate its specific binding with the tumor cell through the tag molecule on the regulatory protein, thereby achieving the purpose of specifically killing the tumor cell.

[0030] In some examples, the construction method of CAR-immune cells is selected from lentivirus, adenovirus, adeno-associated virus, transposon technology, mRNA transfection, nanogene, Zinc Finger Nuclease, TALEN and CRISPR / Cas9. It can also be constructed using a general method.

[0031] In some examples, the regulatory protein is expressed by fusion through genetic recombination, or is prepared by random chemical coupling and site-specific chemical bonding, or other methods.

[0032] In some embodiments, the immune cell is selected from the group consisting of a T cell, a natural killer cell, a macrophage, a tumor immune cell, and a stem cell-induced immune cell.

[0033] In some examples, the T cell type is selected from CD4, CD8, Treg, Th17, Th9, Tfr, Tfh, Tfc, Tph, Trm.

[0034] In some examples, the hinge region in the CAR structure is selected from at least one of the immunoglobulin family or the tumor necrosis factor superfamily.

[0035] In some examples, the CAR structure of the CAR-immune cell also includes at least one of the costimulatory factors CD28, 4-1BB, OX-40, CD27, and ICOS.

[0036] In some examples, the regulatory protein is selected from: APRIL-Myc protein, Myc-APRIL protein, VEGF121-Myc protein, Myc-VEGF121 protein, VEGF165-Myc protein, Myc-VEGF165 protein, BLyS-FLAG protein, FLAG-BLyS protein, APRIL-FLAG protein, FLAG-APRIL protein, VEGF121-FLAG protein, FLAG-VEGF121 protein, VEGF165-FLAG protein, FLAG-VEGF165 protein.

[0037] The beneficial effects of the present invention are: The genetically engineered immune cells of the present invention connect the label molecule to the CAR domain, and then use the label to target the regulatory protein connected to the label. Preferably, these regulatory proteins have two or more specific recognition receptors. For example, the receptors of BLyS include BCMA, BAFF-R, and TACI. Through the mediation of the regulatory protein, multiple related antigen receptor molecules on tumor cells can be targeted and recognized. It has multi-specificity to achieve the purpose of one protein functional domain recognizing multiple targets, which can reduce the possibility of tumor antigen escape, mismatch and off-target to a certain extent, and improve the safety and effectiveness of CAR-immune cell (such as CAR-T cell) treatment to a certain extent.

[0038] The genetically engineered immune cells of the present invention are prepared by constructing regulatory molecules based on short peptides and / or chemical tag molecules and combining CARs targeting short peptides or chemical tag molecules to prepare regulatable CAR-immune cells, such as CAR-T. By in vitro injection of fusion proteins with such tags, the effect of CAR-T therapy can be controlled by adjusting the injection time and amount, thereby increasing safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of the combination of regulatory protein molecules designed and constructed from natural ligands, CAR-T cells, and tumor surface receptors.

[0040] Figure 2 This is a schematic diagram of the structure of tumor recognition regulatory molecules.

[0041] Figure 3 This is a schematic diagram of the preparation of tumor recognition regulatory molecules by amidation chemical coupling.

[0042] Figure 4 This is a schematic diagram of the preparation of tumor recognition regulatory molecules by click chemistry coupling.

[0043] Figure 5 The results are the results of restriction electrophoresis and immunoblotting verification of different regulatory molecules constructed by BLyS using thrombin.

[0044] Figure 6 It is a schematic diagram of the structure of the short peptide epitope PNE-targeted CAR.

[0045] Figure 7 It is the expression of cell surface receptors in different blood tumors.

[0046] Figure 8 These are the results of flow cytometric analysis of the binding of different regulatory molecules constructed by BLyS to different blood tumor cells.

[0047] Fig. 9 It is the result of different regulatory molecules constructed by BLyS mediating the killing of different blood tumor cells by CAR-T cells.

[0048] Fig.10 It is an analysis of the cytokine release mediated by different regulatory molecules constructed by BLyS to kill tumors by CAR-T cells.

[0049] Fig.11 This is an analysis of the inhibitory activity of CAR-T cells on Nalm6 tumors in mice mediated by regulatory molecules constructed by BLyS.

[0050] Fig.12 This is an analysis of the inhibitory activity of CAR-T cells on the mouse tumor RPMI8226 mediated by the regulatory molecules constructed by BLyS. DETAILED DESCRIPTION

[0051] The technical solution of the present invention is further illustrated below in conjunction with exemplary experiments.

[0052] The present invention uses natural ligands BLyS, APRIL and VEGF as recognition molecules of tumor cells and microenvironment, constructs regulatory molecules that can recognize surface antigen proteins of blood tumors and solid tumors, further introduces label molecules and label-targeted CAR-immune cells, and prepares a natural ligand-mediated regulatable CAR-immune cell strategy for multi-target recognition ( Figure 1 ).

[0053] The technical solution of the present invention is further explained below in conjunction with experiments.

[0054] 1. Construction of regulatory proteins (switch molecules) 1. Construction of fusion protein expression vector The regulatory protein is prepared by biofusion method, and the structure of the expression vector is as follows Figure 1 As shown. Taking BLyS as an example, the target gene of Myc-BlyS or BLyS-Myc is introduced into the N-terminus with a His tag and a thrombin cleavage site by PCR, double digested with restriction endonucleases, cloned into the pET32a(+) vector by T4 DNA ligase, transformed into DH5α competent cells, plated, and single clones are picked for sequencing and identification. In addition, in order to achieve secretory expression in mammalian cells, the target gene of the regulatory protein can also be cloned into the pFuse vector by PCR, double digested with restriction endonucleases, and cloned into the pFuse vector by T4 DNA ligase, transformed into DH5α competent cells, plated, and single clones are picked for sequencing and identification.

[0055] The BLyS-Myc protein sequence is shown in SEQ ID NO.: 1; the Myc-BLyS protein sequence is shown in SEQ ID NO.: 2.

[0056] In addition, the regulatory protein can be prepared by chemical coupling. Taking BLyS as an example, the His tag and thrombin cleavage site are introduced into the N-terminus of the BLyS target gene by PCR, and after double digestion with MscI and XhoI, it is cloned into the pET32a(+) vector using T4 DNA ligase, transformed into DH5α competent cells, plated, and single clones are picked for sequencing identification. The structure of the expression vector is as follows Figure 2 shown.

[0057] Similarly, other regulatory protein sequences were constructed as follows: APRIL-Myc protein sequence As shown in SEQ ID NO.: 3; Myc-APRIL protein sequence SEQ ID NO.: shown in 4; VEGF121-Myc protein sequence SEQ ID NO.: 5; Myc-VEGF121 protein sequence SEQ ID NO.: 6; VEGF165-Myc protein sequence SEQ ID NO.: 7; Myc-VEGF165 protein sequence SEQ ID NO.: 8; BLyS-FLAG protein sequence SEQ ID NO.: shown in 9; FLAG-BLyS protein sequence SEQ ID NO.: shown in 10; APRIL-FLAG protein sequence As shown in SEQ ID NO.: 11; FLAG-APRIL protein sequence SEQ ID NO.: 12; VEGF121-FLAG protein sequence As shown in SEQ ID NO.: 13; FLAG-VEGF121 protein sequence SEQ ID NO.: 14; VEGF165-FLAG protein sequence As shown in SEQ ID NO.: 15; FLAG-VEGF165 protein sequence As shown in SEQ ID NO.: 16.

[0058] 2. Expression and purification of fusion protein The constructed plasmid was transfected into BL21, Rosetta gami(DE3)pLysS or Tuner(DE3)pLysS. After plating, a single clone was picked out and cultured overnight. It was inoculated into 500 mL of fresh TB medium at a ratio of 1:100 until OD 600When the value reaches 2.0-2.4, IPTG is induced for 20 hours. The cells are collected by centrifugation at 4°C, weighed and frozen at -80°C. Lysis Buffer with a volume of 10 times the weight of the cells is added, and then the cells are ultrasonically disrupted until the bacterial solution is no longer viscous. After the cell lysate is centrifuged at 4°C for 1 hour, the supernatant is filtered with a 0.45μm filter membrane. The supernatant of bacterial lysis is purified by nickel column, and Wash Buffer is selected to wash the impurities on the nickel column. The target protein is then eluted with Elution Buffer. The eluted target protein is dialyzed with Tris-HCl / NaCl buffer, and thrombin is added for enzyme digestion at room temperature for 16 hours. After SDS-PAGE electrophoresis verifies that the enzyme digestion is complete, the nickel column is used to remove the enzyme digestion tag. Finally, Endotoxin Removal Agarose Resin resin filler is used to remove endotoxins, and then the endotoxin content is detected with Endotoxin detection system (GenScript, L00350) to ensure that the endotoxin content is below 5EU / kg.

[0059] The fusion protein can also be secreted and expressed in mammalian cells. 293F suspension cells were inoculated into a cell culture shake flask and incubated in a shaker incubator at 37°C, 150 rpm, and 5% carbon dioxide for 24 h until the cell density reached 2×10 6 / mL, transfected with a mixture of the fusion protein expression vector constructed by the pFuse vector and PEI MAX, incubated in a shaking incubator at 37°C, 150 rpm, and 5% carbon dioxide concentration for 72 hours, and the supernatant was collected by centrifugation for separation and purification on a nickel column. Thrombin was also used for enzymatic digestion at room temperature for 16 hours, and the enzymatic digestion effect was verified by SDS-PAGE electrophoresis.

[0060] 3. Preparation of Chemically Coupled Proteins The above protein expression and purification system is used to separate and purify natural BLyS protein, or to construct mutant BLyS protein, that is, non-functional sites on BLyS protein are selected, non-natural amino acids, such as phenylalanine with an azide group, are specifically introduced, and after further protein purification, enzyme cleavage, removal of His tag and endotoxin treatment are performed. For natural purified BLyS protein, small molecules that modify the Myc tag and introduce a succinyl group can be used for random amidation coupling to construct a randomly coupled regulatory protein (such as Figure 4 As shown), other chemical coupling methods, such as mixed anhydride method, diazotization method, carbonyldiimidazole method and other chemical coupling methods can also be used to construct regulatory proteins.

[0061] For mutant BLyS proteins with the introduction of unnatural amino acids, a Myc short peptide tag with a cycloalkyne group can be simultaneously synthesized, and a site-specifically modified regulatory protein (such as Figure 5 Other site-specific modification chemical coupling methods can also be used to construct regulatory proteins, such as engineered cysteine, enzyme-assisted ligation, sugar recombination and sugar conjugation, amino-terminal engineered serine, ligation to Fab nucleotide binding sites, natural cysteine ​​re-bridging, and other site-specific coupling methods.

[0062] 4. SDS-PAGE and Western Blot Verification of Fusion Protein Take an appropriate amount of purified protein molecules, add one-third of the protein sample volume of reducing loading buffer, vortex mix and centrifuge. Boil the sample in a metal bath for 10 minutes and then centrifuge. Use 12% separation gel and 5% concentrated gel, fix the gel plate in the mold, pour in the electrophoresis buffer to observe whether it leaks, unplug the comb, and load the sample in turn; add the loading buffer to the scale line, adjust the voltage to 80V to start electrophoresis, and adjust the voltage to 120V after the sample enters the separation gel, and observe the electrophoresis; soak the membrane, filter paper, and sponge in the electrotransfer buffer before use, and place a piece of gel in the order of white board-sponge-filter paper-membrane-gel-filter paper-sponge-black board (make sure there are no bubbles), place it on ice, and connect the power supply 220mA for 2h. Another piece of gel is stained with Coomassie Brilliant Blue, and the protein electrophoresis is observed after decolorization with water. After the transfer is completed, take out the membrane, cut it into appropriate size, and mark the protein molecular weight standard. Add blocking solution (5% skim milk powder + 0.12% tween20 TBST) and block at room temperature for 1 hour. Wash with TBST for 3 minutes. Incubate the primary antibody at room temperature for 2 hours. Rinse the NC membrane with appropriate amount of TBST for 10 minutes / time, for a total of 3 times. Incubate the secondary antibody at room temperature for 1 hour, and rinse with appropriate amount of TBST for 10 minutes / time, for a total of 3 times. Develop the gel using a gel imaging instrument. The results are shown in Figure 3 As shown, BLyS / Myc and Myc / BLyS showed the correct molecular weight and good purity before and after thrombin digestion. Western blotting further verified the tagged BLyS protein.

[0063] 2. Preparation of CAR-T cells 1. Vector Construction Taking the construction of CAR targeting Myc and FLAG by lentiviral vector pELPS as an example, Myc targeting antibody 9E10 and FLAG targeting antibody FBT are taken as examples, and the variable region of the antibody is introduced into the structure of the second-generation CAR by VL-connector peptide-VH and VH-connector peptide-VL. The conventional second-generation CAR structure includes EF1α promoter, signal peptide molecule, single-chain antibody, hinge region, transmembrane region, co-stimulatory molecule and CD3zeta domain. The constructed CAR is double-digested with BamHI and SalI and then cloned into the pELPS vector using T4 DNA ligase, or the vector is constructed using a homologous recombination kit. The structure of the lentiviral expression vector is as follows: Figure 6 shown.

[0064] The representative CAR structure is as follows: pELPS-CAR-anti-Myc9E10(VL-VH)-BBZ protein sequence is as follows SEQ ID NO.: shown in 17; pELPS-CAR-anti-Myc9E10(VH-VL)-BBZ protein sequence is as follows As shown in SEQ ID NO.: 18; pELPS-CAR-anti-FLAGFBT (VL-VH)-BBZ protein sequence is as follows As shown in SEQ ID NO.: 19; pELPS-CAR-anti-FLAG FBT (VH-VL)-BBZ protein sequence is as follows SEQ ID NO.: 20; pELPS-CAR-anti-FLAGM2(VL-VH)-BBZ protein sequence is as follows SEQ ID NO.: 21; pELPS-CAR-anti-FLAGM2(VH-VL)-BBZ protein sequence is as follows As shown in SEQ ID NO.: 22.

[0065] 2. Preparation of Lentivirus 293FT cells were cultured in DMEM medium (10% FBS, 1% penicillin-streptomycin double antibody, 1% non-essential amino acids, 1% sodium pyruvate, 1% L-glutamine). After the cells adhered to the wall, when the density reached 80%-90%, the cells were replaced with antibiotic-free DMEM medium. The plasmid was mixed in 0.3 mL OPTI-MEM according to the ratio of pELPS-CAR: pMDLgpRRE: pRSV-Rev: pMD2.G = 2.25: 1.8: 1.8: 0.75 (μg), and 20 μL lipofectamine 2000 was added to 0.3 mL OPTI-MEM. After standing for 5 minutes, the two were mixed and vortexed to mix evenly, and then stood at room temperature for 15 minutes and then added to the 293FT cells. After the cells were cultured in a constant temperature incubator for 6 hours, new antibiotic-free DMEM medium was replaced and cultured for another 48 hours at 37°C and 5% CO2. Collect all supernatants, centrifuge at 4°C, 1500 rpm for 10 min, discard the cell pellet, and store the supernatant at -80°C for future use.

[0066] 3. Extraction of Human PBMC Take an appropriate amount of Ficoll-Poque in a centrifuge tube (the ratio of blood to Ficoll-Poque is preferably 2:1). Slowly tilt the fresh blood taken from healthy volunteers into the centrifuge tube containing Ficoll-Poque, so that the cell suspension is resuspended on the layering liquid. Centrifuge at 27°C, 2000rpm for 30min (speed increase is 9, speed decrease is 0). After the centrifugation, take out the centrifuge tube and observe that it is divided into 4 layers: the top layer is plasma, containing some platelets; the second layer is a thin white film layer, mainly mononuclear cells, and a small amount of platelets are mixed; the third layer is the separation liquid layer; the fourth layer is granulocytes and red blood cells, the red blood cells settle at the bottom of the tube, and the granulocytes are closely attached to the red blood cells to form a very thin white film, that is, PBMC. Use a Pasteur pipette to absorb the upper serum (it can be divided and stored after centrifugation), then carefully absorb the white film layer and add it to two centrifuge tubes (the volume should be as equal as possible). Add RPMI-1640 culture medium to 50mL and mix it upside down. Centrifuge horizontally at 27°C, 2000rpm for 10min (9 for ascending speed and 9 for descending speed). Low-speed centrifugation is beneficial for removing residual platelets in the cell suspension. Remove the supernatant after centrifugation. Add AIM-V medium and centrifuge again at 27°C, 2000rpm for 10min. Remove the supernatant after centrifugation. Resuspend with 15mL AIM-V medium and add to T75 culture flask. Culture at 37°C for more than 1h. The purpose of this step is to remove adherent cells. Then aspirate the supernatant into a centrifuge tube, add fresh RPMI-1640 medium and culture at 37°C for standby use.

[0067] 4. PBMC cryopreservation 1×10 8 PBMC cells were resuspended in 700 μL cryopreservation solution, and the resuspended solution was transferred to cryopreservation tubes. The cryopreservation tubes were placed in a programmed cooling box, and the programmed cooling box was transferred to a -80°C ultra-low temperature refrigerator. After 24 hours, the cells were transferred to liquid nitrogen for long-term storage.

[0068] 5. T cell activation Take out the frozen PBMC from liquid nitrogen and quickly thaw them in a 37℃ water bath. Add 10mL AIM-V medium and centrifuge at 24℃, 1500rpm for 5 minutes. After centrifugation, remove the supernatant and resuspend with an appropriate amount of AIM-V medium (if there are particles, transfer the supernatant). 6 Take an appropriate amount of magnetic beads (vortex before use) at a ratio of 1×10 cells to 30 μL of magnetic beads (vortex before use), and wash the magnetic beads with 2-3 mL of PBS containing 5% FBS (the magnetic beads will be adsorbed on the side wall). After aspirating the supernatant of the magnetic beads, wash them again with AIM-V medium. After counting the T cells, dilute them with medium to 1×10 6 / mL and then add magnetic bead suspension. Add IL-2, the initial concentration of IL-2 is 2.6×10 5 IU / mL, and adjust to a final concentration of 300 IU / mL.

[0069] 6. T cell transduction The activated T cells for 24 hours were centrifuged at 1500rpm for 5 minutes, and further resuspended in RPMI-1640 medium. The cells were counted using a hemocytometer and trypan blue. After mixing the appropriate amount of cell suspension with the virus solution, 10 mg / mL of polybrene was added to make the final concentration 8 μg / mL and 300 IU / mL of IL-2. Centrifuge at 33°C, 1000g for 90 minutes, and cultured overnight in a 37°C constant temperature incubator. The cells were centrifuged the next day and supplemented with fresh culture medium and continued to be cultured in a 37°C incubator. After the expression was stable, it was used for anti-tumor activity detection.

[0070] 3. Functional Verification 1. Regulate protein binding activity Blood tumor cells from different sources were collected and the expression of BCMA, BAFFR, TACI and BLyS was verified. Figure 7 ), thus determining that Nalm6, RPMI8226 and IM-9 were positive cells and Jurkat was negative cells. Tumor cells were further collected, incubated with 200 nM regulatory protein at 4°C for 2 hours, washed 3 times with flow cytometry washing solution, incubated with APC-anti-Myc tag secondary antibody at 4°C for 1 hour, and then analyzed using a flow cytometer. The results are shown in Figure 8 As shown, the results show that, on the one hand, BLyS has a broad-spectrum tumor targeting recognition ability, including leukemia, lymphoma and myeloma. It has good specific recognition activity. This recognition ability mainly depends on the multi-target binding function of BLyS protein to BAFF-R, TACI and BCMA; on the other hand, the experimental results can clearly show that modifying the label molecule at the C-terminus of BLyS will cause a certain degree of non-specific binding activity of negative cells, while the N-terminal modified fusion protein of BLyS shows good binding activity and has unexpected effects.

[0071] 2. Cytotoxicity test: Roughly count the cells to determine the approximate amount of cells to be used (mainly to determine the minimum amount, as many cells as possible should be taken to avoid loss of cells due to experimental operations). After the tumor cells are labeled with fluorescent dye CFSE, they are resuspended and accurately counted, and then diluted to determine the cell density, so that the final concentration is 0.2×10 6 / mL; take enough CAR-T cells and resuspend them to make the final concentration 2×10 6 Tumor cells and CAR-T cells were mixed at a ratio of 1:10 (the concentration was 0.1×10 6 Tumor cells / mL, 1×106 T cells / mL, the effector-target ratio is 10:1). Prepare a gradient control protein solution, the highest concentration is 100nM, 10-fold dilution, and the lowest concentration is 1pM. Add protein solutions of different concentrations to each well, 10μL per well. Set up negative and positive controls at the same time. The negative control condition is no fusion protein solution, and the positive control condition is no T cells and fusion protein solution. Place the 96-well plate after the addition of the sample in a 37℃ constant temperature incubator for more than 24h. Half an hour before the end of the culture, 10μL of lysis solution is added to the positive control group in advance. After the incubation, centrifuge the 96-well plate at 2000rpm for 5min in a horizontal centrifuge. Transfer the solution in the centrifuged 96-well plate to another 96-well U-shaped plate (55μL) and centrifuge it at 2000rpm for 5min in a horizontal centrifuge. Take two more 96-well U-shaped plates and transfer 30μL accordingly. Add the test buffer to the substrate MIX and mix well. Add 30μL of the above mixture to each well. Incubate at room temperature in the dark for 20-30 minutes, then add the stop buffer, and then use a microplate reader to detect its absorbance at 490nm; for suspended tumor cells, use CFSE labeling, incubate with CAR-T cells and regulatory proteins in the same way, incubate in a 37℃ constant temperature incubator for 24 hours, use 7-AAD staining, and use a high-throughput flow cytometer to analyze the number of target cells. Graphpad Prism was used to draw graphs to analyze the experimental results. Fig. 9 Although both BLyS / Myc and Myc / BLyS can effectively kill different blood tumor cells, BLyS / Myc shows nonspecific killing activity against negative cells, while the killing activity of Myc / BLyS is more specific. In addition, BLyS / Myc and Myc / BLyS have good killing activity against tumor cells expressing BCMA, BAFFR and TACI, which is due to the high efficiency brought by the high affinity and specificity of natural ligands for different receptors.

[0072] 3. Cytokine Release Assay 1×10 5 Tumor cells were inoculated into 24-well plates, and two replicate wells were set for each target cell. CAR-T cells were washed with PBS and plated at 1×10 6 / mL of cells were diluted into T cell expansion medium without IL-2 and diluted to 1×10 6 T cells were inoculated into the culture wells containing target cells. 10 nM BLyS / Myc and Myc / BLyS were added to the corresponding wells, respectively, and cultured in a 37°C, 5% CO2 incubator for 24 h. The supernatant was collected and the IFN-γ and TNF-α ELISA kits were used to detect the content of IFN-γ and TNF-α in different samples. The results are shown in Figure 2. Fig.10As mentioned above, cytokine secretion is a hallmark of CAR-T cell function. After analyzing the results of cytokine release assays, we found that BLyS / Myc and Myc / BLyS showed significantly enhanced cytokine secretion while mediating CAR-T cell killing of tumors, and had good CAR-T cell-mediated activation effects on tumor cells expressing BCMA, BAFFR, and TACI.

[0073] 4. Animal experiments (determination of the in vivo tumor killing ability of T cells; NSG mice (6-8 weeks old) were raised and treated under specific pathogen-free conditions. To establish a tumor model, mice were injected with Nalm6 and RPMI8226 tumor cells carrying luciferase via the tail vein. After the tumors grew and dispersed well in the NSG mice, the mice were divided into three groups and injected intravenously (iv) with the corresponding CAR-T cells (30×10 6 CAR-T cells / mice were used for tumor model), and 0.5 mg / kg / dose of BLyS / Myc and Myc / BLyS were injected into mice every day, and the distribution of tumor cells in NSG mice was regularly observed using IVIS fluorescence imaging system. Fig.11 As shown in the figure, BLyS / Myc-mediated CAR-T cells can effectively inhibit the proliferation of Nalm6 leukemia tumor cells. In addition, Myc / BLyS-mediated CAR-T cells can also effectively inhibit the proliferation of RPMI8226 myeloma cells ( Fig.12 ), further proving that the regulatable CAR-T cells constructed with natural ligands can be used to construct tumor cells that target different receptors and achieve effective inhibitory activity against different tumors.

[0074] In summary, the functional validation results show that the regulatory switch molecule constructed using BLyS as a targeting molecule can mediate CAR-T cells and achieve effective killing and control effects on blood tumors expressing different tumor-associated antigens. From different perspectives, the safety and effectiveness are greatly improved compared to traditional CAR-T therapy. The combination of regulatable molecules and CAR-T cells has greatly solved the uncontrollability of traditional CAR-T cells in patients. By injecting regulatable molecules to control the killing of tumor cells by CAR-T cells, the safety of CAR-T therapy during use has been improved. At the same time, the multi-target recognition strategy using natural ligand molecules effectively avoids the immune escape problem existing in single-target recognition and increases the effective killing of tumor cells by CAR-T cells.

Claims

1. A composition of anti-tumor CAR-immune cells, comprising CAR-immune cells and a regulatory protein constructed based on a natural ligand, characterized in that: The natural ligand is a naturally occurring ligand molecule that can specifically bind to a tumor or a related receptor on the surface of its tissue, and is selected from at least one of APRIL and VEGF; The CAR domain of the CAR-immune cell is connected to a recognition molecule, and the recognition molecule is used to specifically recognize the tag molecule; The regulatory protein has natural ligand molecules and label molecules that recognize multiple tumors or their tissue-associated antigens, and the label molecules can specifically bind to the recognition molecules of the CAR extracellular recognition domain of the CAR-immune cell.

2. The composition according to claim 1, characterized in that: The tag molecule is selected from one of FITC, biotin, His tag, Flag tag, Myc tag, GCN4 tag, HA tag, V5 tag, chitosan binding protein (CBP) tag, maltose binding protein (MBP), glutathione-S-transferase (GST) and DNA Aptamer tag.

3. The composition according to claim 1 or 2, characterized in that: The natural ligand molecule of the regulatory protein and the tag molecule are coupled via a hydrophilic bifunctional linker; Preferably, the hydrophilic bifunctional linker is a connecting peptide; Preferably, the connecting peptide is selected from a linear polypeptide, a β-folded peptide or a coiled coil peptide.

4. The composition according to any one of claims 1 to 3, characterized in that: The tag-specific recognition molecule in the CAR-immune cell is selected from a single-chain antibody, a protein ligand, a short peptide molecule or a nucleic acid molecule.

5. The composition according to claim 1, characterized in that: The regulatory protein is selected from APRIL-Myc protein, Myc-APRIL protein, VEGF121-Myc protein, Myc-VEGF121 protein, wherein: The APRIL-Myc protein sequence is shown in SEQ ID NO.: 3; The Myc-APRIL protein sequence is shown in SEQ ID NO.: 4; The VEGF121-Myc protein sequence is shown in SEQ ID NO.: 5; The Myc-VEGF121 protein sequence is shown in SEQ ID NO.:

6.

6. The composition according to any one of claims 1 to 5, characterized in that: The CAR structure also includes co-stimulatory factors CD28, 4-1BB, OX-40, CD27, ICOS and at least one of the immunoglobulin family or tumor necrosis factor superfamily.

7. The composition according to any one of claims 1 to 6, characterized in that: The immune cells are selected from T cells, natural killer cells, macrophages, tumor immune cells and stem cell-induced immune cells.

8. The composition according to any one of claims 1 to 7, characterized in that: The regulatory protein has at least two natural ligand molecules that can recognize multiple tumors or their tissue-related antigens.

9. The composition according to any one of claims 1 to 4, characterized in that: The regulatory protein is expressed by fusion through gene recombination; or prepared by random chemical coupling and site-specific chemical bonding.

10. Use of a multi-target recognition controllable genetically engineered immune cell composition in the preparation of an anti-tumor drug, wherein the composition is as described in any one of claims 1 to 9, and the tumor is a B-cell malignant tumor.

Images