Multi-specific antibody targeting BCMA, GPRC5D and T cells and application thereof

Abstract

The present application relates to a multispecific binding molecule that targets BCMA, GPRC5D, and T cell receptors. Specifically, the invention discloses a BCMA-GPRC5D-CD3 multispecific antibody which can activate T cells while being combined with a tumor surface antigen, thereby promoting the specific killing of the T cells on tumor cells, especially BCMA or GPRC5D positive multiple myeloma. The invention also provides a preparation method and application of the multispecific binding molecule.

Description

Multispecific antibodies targeting BCMA, GPRC5D, and T cells and their applications Cross-reference to related applications This application claims priority to Chinese patent application No. 202210631397.9, filed on June 6, 2022, which is incorporated herein by reference in its entirety for all purposes. Technical Field This disclosure relates to the field of biomedicine, and more specifically, to a multispecific antibody targeting BCMA, GPRC5D and T cells and its applications. Background Technology Multiple myeloma (MM) is a malignant plasma cell tumor characterized by the uncontrolled proliferation of plasma cells in the bone marrow, resembling tumor cells, accompanied by monoclonal immunoglobulin secretion. This leads to a range of clinical manifestations, including multiple osteolytic lesions, hypercalcemia, anemia, kidney damage, and recurrent infections. MM accounts for 1% of all cancers and 10-15% of hematologic malignancies. The male-to-female ratio is 1.6:1, and most patients are over 40 years old. Treatment for MM includes chemotherapy and hematopoietic stem cell transplantation. Immunomodulators, such as lenalidomide, and protease inhibitors, such as bortezomib, have shown good efficacy as single agents or in combination, and have become standard treatments for MM patients. However, MM is still considered incurable. Current treatments can only alleviate symptoms and cannot completely eliminate the tumor; almost all patients eventually relapse. Therefore, there is an urgent need for new treatment options. In recent years, CAR-T therapy or targeted ADC therapy has made breakthrough progress in the treatment of multiple myeloma. In particular, CAR-T drugs targeting B cell maturation antigen (BCMA) and antibody ADC drugs targeting BCMA have shown positive clinical results in the treatment of multiple myeloma, and a number of drugs have been approved for marketing. B cell maturation antigen (BCMA), also known as CD269 or TNFRSF17, is a member of the tumor necrosis factor receptor superfamily and was first discovered in the early 1990s. It is primarily expressed on the surface of mature B lymphocytes and plasma cells, serving as a marker protein for B lymphocyte maturation. Structurally, BCMA consists of three main domains: an extracellular domain (aa1-54), a transmembrane domain (aa55-77), and an intracellular domain (aa78-184). B cell activating factor (BAFF) and proliferation-inducing ligand (APRIL) are the main ligands of BCMA, interacting with it to transmit cellular stimulation signals, activating the TRAF-dependent NF-κB and JNK pathways, and increasing B cell proliferation and survival. BCMA is mainly expressed on the surface of mature B lymphocytes and plasma cells, and is almost not expressed in other tissue cells. Furthermore, BCMA is highly expressed in the tumor cells of most multiple myeloma patients, making it an ideal therapeutic target for multiple myeloma. However, for patients with BCMA-negative or low-expressing multiple myeloma, as well as those who have relapsed after BCMA-targeted therapy, there is an urgent need for more effective treatment methods. Therefore, the treatment of multiple myeloma still needs to find other new targets with greater potential. Summary of the Invention To solve the above-mentioned technical problems, this application provides the following technical solution: In a first aspect, this disclosure provides a multispecific binding molecule, said multispecific binding molecule comprising: (a) The first antigen-binding molecule (ABM1) that specifically binds to the first tumor-associated antigen (TAA1); (b) A second antigen-binding molecule (ABM2) that specifically binds to a second tumor-associated antigen (TAA2); and (c) A third antigen-binding molecule (ABM3) that specifically binds to antigens expressed on human immune cells. Secondly, this disclosure provides an isolated nucleic acid that encodes the aforementioned multispecific binding molecule. Thirdly, this disclosure provides a recombinant vector containing the aforementioned isolated nucleic acid. Fourthly, this disclosure provides a host cell containing the aforementioned isolated nucleic acid or the aforementioned recombinant vector. Fifthly, this disclosure provides a method for preparing a multispecific binding molecule, comprising culturing the aforementioned host cells under appropriate conditions and isolating the multispecific binding molecule. In a sixth aspect, this disclosure provides a pharmaceutical composition comprising the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned recombinant vector, the aforementioned host cell, or a product prepared according to the aforementioned method, and one or more pharmaceutically acceptable carriers, diluents, buffers, or excipients. In a seventh aspect, this disclosure provides a method for activating T cells, comprising administering to a subject a therapeutically effective amount of the aforementioned multispecific binding molecule or pharmaceutical composition. Eighthly, this disclosure provides the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned host cell, the aforementioned product prepared by the aforementioned method, or the aforementioned pharmaceutical composition for use in the preparation of a medicament for treating and / or preventing cancer, or for use in treating and / or preventing cancer. Ninthly, this disclosure provides a method for treating and / or preventing cancer, comprising administering to a subject a therapeutically effective amount of the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned host cell, a product prepared by the aforementioned method, or the aforementioned pharmaceutical composition. Beneficial effects: The development of multispecific antibodies and related biopharmaceuticals targeting both GPRC5D and BCMA has the advantages of covering a wider range of cancer patients, potentially leading to more thorough tumor clearance and delaying tumor recurrence, and has enormous potential clinical demand. Attached Figure Description Figure 1 shows a schematic diagram of the molecular structures of the three antibodies and the positive and negative control molecules. Figure 2A shows the FACS results of the CHOK1 stable cell line expressing human GPRC5D protein. Figure 2B shows the FACS results of the CHOK1 stable cell line expressing monkey GPRC5D protein. Figure 3 shows the FACS results of BCMA expression levels in Flp-inCHO-human BCMA cells. Figure 4A shows the ELISA detection of the binding reaction between the triple antibody and human BCMA-his protein. Figure 4B shows the ELISA detection of the binding reaction between the triple antibody and monkey BCMA-his protein. Figure 5A shows the cell-based ELISA detection of the binding reaction between the triple antibody and CHOK1-human GPRC5D cells. Figure 5B shows the cell-based ELISA detection of the binding reaction between the triple antibody and CHOK1-monkey GPRC5D cells. Figure 6A shows the ELISA detection of the binding reaction between the triple antibody and human CD3e / d-his protein. Figure 6B shows the ELISA detection of the binding reaction between the triple antibody and human CD3e-his protein. Figure 7 shows the binding reaction of the triple antibody to Flp-inCHO-human BCMA cells as detected by FACS. Figure 8 shows the binding reaction of the triple antibody to CHOK1-human GPRC5D cells as detected by FACS. Figure 9 shows the binding reaction of the triple antibody to H929 cells as detected by FACS. Figure 10 shows the binding reaction of the triple antibody to Jurkat cells as detected by FACS. Figure 11A shows the binding reaction of the triple antibody to human PBMCs detected by FACS. Figure 11B shows the binding reaction of the triple antibody to monkey PBMCs detected by FACS. Figure 12A shows the activation activity of the triple antibody in Jurkat-NFAT-NanoLuc cells mediated by H929 cells. Figure 12B shows the activation activity of the triple antibody in Jurkat-NFAT-NanoLuc cells mediated by K562 cells. Figure 13 shows the killing effect of T cells on H929-Luc cells mediated by triple antibodies. Figure 14A shows the release of TNFα cytokine during T cell killing of H929-Luc cells mediated by triple antibodies. Figure 14B shows the release of cytokine IFNγ during T cell killing of H929-Luc cells mediated by triple antibodies. Figure 15 shows the killing effect of T cells on RPMI8226-Luc cells mediated by triple antibodies. Figure 16A shows the release of TNFα cytokine during T cell killing of RPMI8226-Luc mediated by triple antibodies. Figure 16B shows the release of cytokine IFNγ during T cell killing of RPMI8226-Luc mediated by triple antibodies. Figure 17 shows the killing effect of T cells on Molp-8-Luc or Molp-8 cells mediated by triple antibodies. Figure 18A shows the release of TNFα cytokine during T cell killing of Molp-8-Luc or Molp-8 mediated by triple antibodies. Figure 18B shows the release of cytokine IFNγ by T cells during triple antibody-mediated killing of Molp-8-Luc or Molp-8. Terminology Definitions and Explanations Unless otherwise defined in this disclosure, scientific and technical terms used in connection with this disclosure shall have the meanings understood by one of ordinary skill in the art. Furthermore, unless otherwise stated herein, singular terms shall include plural terms, and plural terms shall include singular terms. More specifically, as used in this specification and the appended claims, unless otherwise expressly indicated, the singular forms “a” and “this” include plural indicators. The terms “comprising,” “including,” and “having” are used interchangeably in this document to indicate the inclusiveness of a scheme, meaning that the scheme may contain elements other than those listed. It should also be understood that the use of “comprising,” “including,” and “having” in this document also provides for schemes “consisting of…”. When used herein, the term “and / or” includes the meaning of “and,” “or,” and “all or any other combination of elements linked by the term.” The term "BCMA" in this article stands for B cell maturation antigen, a member of the tumor necrosis factor receptor family. BCMA is primarily expressed on the surface of late-stage B cells, short-lived proliferating plasmablasts, and long-lived plasma cells, but is not expressed on naive B cells, CD34-positive hematopoietic stem cells, or other normal tissue cells. However, it is highly expressed in MM cells and plays a crucial role in the survival, proliferation, metastasis, and drug resistance of MM cells by mediating downstream signaling pathways. Therefore, BCMA is an ideal antigen target for the treatment of MM. An exemplary human BCMA sequence can be found in GenBank Protein Accession No: NP_001183.2. The term "GPRC5D" in this article refers to G protein-coupled receptor C5 family subtype D, an orphan receptor and a 7-transmembrane protein. GPRC5D is highly expressed on the surface of primary multiple myeloma cells, while its expression in normal tissues is limited to the hair follicle region. Studies have shown that 65% of multiple myeloma patients have GPRC5D expression exceeding the 50% threshold, making GPRC5D a potential therapeutic target for MM. An exemplary human GPRC5D sequence can be found in GenBank Protein Accession No: NP_061124.1. The term "CD3" in this document refers to an antigen on T cells that serves as a part of the multimolecular T-cell receptor (TCR), and is composed of homodimers or heterodimers formed from two of the following four receptor chains: CD3ε, CD3δ, CD3ζ, and CD3γ. Human CD3-ε (hCD3ε) contains the amino acid sequence described in UniProtKB / Swiss-Prot: P07766.2. Human CD3-δ (hCD3δ) contains the amino acid sequence described in UniProtKB / Swiss-Prot: P04234.1. Therefore, unless explicitly stated to be from a non-human species, such as "mouse CD3," "monkey CD3," etc., the term "CD3" refers to human CD3. The term "tumor-associated antigen" or "TAA" in this article refers to a molecule (typically a protein, carbohydrate, lipid, or some combination thereof) expressed entirely or as a fragment (e.g., MHC / peptide) on the surface of cancerous B cells, and its This can be used to preferentially target pharmacological agents to cancerous B cells. As used herein, the term "cancer B cell" refers to B cells that are undergoing or have undergone uncontrolled proliferation. In some instances, TAAs are markers expressed by both normal and cancer cells, such as lineage markers, such as CD19 on B cells. In some instances, TAAs are B cell surface molecules overexpressed in cancerous B cells compared to normal B cells, for example, 1-fold, 2-fold, 3-fold, or more overexpression compared to normal B cells, such as BCMA and GPRC5D, which are expressed much higher on cancerous B cells than in normal tissues. In some instances, TAAs are cell surface molecules that are inappropriately synthesized in cancerous B cells, for example, molecules containing deletions, additions, or mutations compared to molecules expressed on normal B cells. In some instances, TAAs will be expressed only on the cell surface of cancerous cells, either completely or as fragments (e.g., MHC / peptides), and will not be synthesized or expressed on the surface of normal cells. Therefore, the term "TAA" covers B-cell antigens specific to cancer cells, sometimes referred to in the art as "tumor-specific antigens" (TSA). The term "specific binding" used in this article refers to the ability of antigen-binding molecules (such as antibodies) to specifically bind antigens and substantially the same antigens with high affinity, but not to unrelated antigens with high affinity. Affinity is usually reflected by the equilibrium dissociation constant (KD), where a lower KD indicates higher affinity. For example, high affinity for antibodies typically refers to a KD of approximately 10. -6 M or lower, 10 -7 M or lower, approximately 10 -8 M or lower, approximately 10 -9 M or lower KD. KD is calculated as follows: KD = Kd / Ka, where Kd represents the dissociation rate and Ka represents the binding rate. The equilibrium dissociation constant KD can be measured using methods known in the art, such as surface plasmon resonance (e.g., Biacore) or equilibrium dialysis. The term "antigen-binding molecule" as used in this document, in its broadest sense, refers to a molecule that specifically binds to an antigen. Exemplarily, antigen-binding molecules include, but are not limited to, antibodies or antibody mimics. "Antibody mimic" refers to an organic compound or binding domain capable of specifically binding to an antigen, but independent of the antibody structure. Exemplarily, antibody mimics include, but are not limited to, affibody, affitin, affilin, designed ankylosing spicule repeat protein (DARPin), nucleic acid aptamers, or Kunitz-type domain peptides. The term “antibody” as used in the broadest sense herein refers to a polypeptide or combination of polypeptides containing sufficient sequences from the variable regions of the immunoglobulin heavy chain and / or from the variable regions of the immunoglobulin light chain, thereby enabling specific binding to an antigen. The term “antibody” as used herein encompasses a wide range of forms and structures, provided they exhibit the desired antigen-binding activity. The term “antibody” as used herein includes alternative protein scaffolds or artificial scaffolds having transplanted complementarity-determining regions (CDRs) or CDR derivatives. Such scaffolds include antibody-derived scaffolds (containing mutations introduced to, for example, stabilize the three-dimensional structure of the antibody) and fully synthetic scaffolds containing, for example, biocompatible polymers. See, for example, Korndorfer, IP, Beste, G. & Skerra, A. (2003). Proteins, 53, 121–129.; Roque, ACA, Lowe, CR & Taipa, MA. Antibodies and genetically engineered related molecules: production and purification. Biotechnol. Prog. 20, 639-654 (2004). Such scaffolds may also include non-antibody-derived scaffolds, such as scaffold proteins known in the art for use in transplanting CDRs, including but not limited to tendinins, fibronectins, peptide aptamers, etc. The term "antibody" includes complete antibodies and any antigen-binding fragments (i.e., "antigen-binding portions") or single chains. An "antibody" is a glycoprotein, or its antigen-binding portion, comprising at least two heavy (H) chains and two light (L) chains linked together by disulfide bonds. Each heavy chain consists of a heavy chain variable region (abbreviated here as VH) and a heavy chain constant region. The heavy chain constant region consists of three domains: CH1, CH2, and CH3. Each light chain consists of a light chain variable region (abbreviated here as VL) and a light chain constant region. The light chain constant region consists of one domain: CL. The VH and VL regions can be further subdivided into hypervariable regions called complementarity-determining regions (CDRs), which are scattered within more conserved regions called framework regions (FRs). Each VH and VL consists of three CDRs and four FRs, arranged from the amino terminus to the carboxyl terminus in the following order: FR1, CDR1, FR2, ... CDR2, FR3, CDR3, FR4. The variable regions of the heavy and light chains contain binding domains that can interact with antigens. The constant regions of antibodies mediate the binding of immunoglobulins to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (C1q) of the classical complement system. Due to differences in the amino acid composition and sequence of the constant regions of the immunoglobulin heavy chains, their antigenicity also differs. Accordingly, the "immunoglobulins" in this article can be divided into five classes, or isotypes of immunoglobulins: IgM, IgD, IgG, IgA, and IgE, with their corresponding heavy chains being μ, δ, γ, α, and ε chains, respectively. Within the same class of Ig, based on differences in the amino acid composition of their hinge regions and the number and position of disulfide bonds in their heavy chains, they can be further divided into different subclasses. For example, IgG can be divided into IgG1, IgG2, IgG3, and IgG4, and IgA into IgA1 and IgA2. Light chains are classified into κ chains or λ chains based on differences in their constant regions. Each of the five types of Ig can have either a κ chain or a λ chain. The term "antibody" in this article also includes antibodies that do not contain light chains, such as heavy-chain antibodies (HCAbs) produced by camels such as dromedary camels (Camelus dromedarius), Bactrian camels (Camelus bactrianus), llamas (Lama glama), guanocamels (Lama guanicoe), and alpacas (Vicugna pacos), as well as immunoglobulin new antigen receptors (IgNARs) found in cartilaginous fish such as sharks. The term "antibody" in this article can be derived from any animal, including but not limited to humans and non-human animals, which can be selected from primates, mammals, rodents and vertebrates, such as camels, llamas, guanacos, alpacas, sheep, rabbits, mice, rats or cartilaginous fish (e.g., sharks). The term "heavy chain antibody" in this article refers to an antibody that lacks a light chain, as is commonly used in antibodies. This term specifically includes, but is not limited to, homodimeric antibodies that contain a VH antigen-binding domain and constant CH2 and CH3 domains in the absence of a CH1 domain. The term "nanobody" in this article refers to a naturally occurring heavy chain antibody in animals such as camels that lacks a light chain. Cloning its variable region can yield a single-domain antibody composed only of the heavy chain variable region, also known as VHH (Variable domain of heavy chain antibody), which is the smallest functional antigen-binding fragment. In this article, the terms "VHH domain," "nanobody," and "single domain antibody (sdAb)" have the same meaning and are used interchangeably. They refer to the cloning of the variable region of a heavy chain antibody to construct a single-domain antibody consisting of only one heavy chain variable region. This is the smallest antigen-binding fragment with complete function. Typically, a naturally occurring heavy chain antibody lacking both the light chain and the heavy chain constant region 1 (CH1) is first obtained, and then the variable region of the antibody heavy chain is cloned to construct a single-domain antibody consisting of only one heavy chain variable region. Further descriptions of “heavy chain antibodies,” “single domain antibodies,” “VHH domains,” and “nanobodies” can be found in: Hamers-Casterman et al., Nature. 363, 446-8 (1993). Naturally occurring antibodies devoid of light chains.; Muyldermans’ review article (J Biotechnol. 2001 Jun; 74(4): 277-302. Single domain camel antibodies: current status); and the following patent applications, which are mentioned as general background art: WO 94 / 04678, WO 95 / 04079, and WO 96 / 34103; WO 94 / 25591, WO 99 / 37681, WO 00 / 40968, WO 00 / 43507, WO 00 / 65057, WO 01 / 40310, WO 01 / 44301, EP 1134231, and WO WO 02 / 48193; WO 97 / 49805, WO 01 / 21817, WO 03 / 035694, WO 03 / 054016 and WO 03 / 055527; WO 03 / 050531; WO 01 / 90190; WO 03 / 025020; and WO 04 / 041867, WO 04 / 041862, WO 04 / 041865, WO 04 / 041863, WO 04 / 062551, WO 05 / 044858, WO 06 / 40153, WO 06 / 079372, WO 06 / 122786, WO 06 / 122787 and WO 06 / 122825 and other prior art mentioned in these applications. The term "multispecificity" in this article refers to the ability of an antibody or its antigen-binding fragment to bind to, for example, different antigens or at least two different epitopes on the same antigen. Therefore, terms such as "bispecificity," "trispecificity," and "quadrispecificity" are also relevant. This refers to the number of different epitopes an antibody can bind to. For example, a conventional monospecific IgG antibody has two identical antigen-binding sites (complementary sites), and therefore can only bind to the same epitopes (rather than different epitopes). In contrast, a multispecific antibody has at least two different types of complementary sites / binding sites, and therefore can bind to at least two different epitopes. As described herein, a “complementarity-determining region” refers to an antibody’s antigen-binding site. Furthermore, a single “specificity” can refer to one, two, three, or more than three identical complementarity-determining regions in a single antibody (the actual number of complementarity-determining regions / binding sites in a single antibody molecule is called the “valence”). For example, a single natural IgG antibody is monospecific and bivalent because it has two identical complementary sites. Accordingly, a multispecific antibody contains at least two (different) complementarity-determining regions / binding sites. Therefore, the term “multispecific antibody” refers to an antibody that has more than one complementary site and the ability to bind to two or more different epitopes. The term “multispecific antibody” specifically includes bispecific antibodies as defined above, but generally also includes proteins, such as antibodies that specifically bind to three or more different epitopes, and scaffolds, i.e. antibodies with three or more complementary sites / binding sites. In this article, the term "valence" refers to the presence of a specified number of antigen-binding domains in an antibody / antigen binding molecule. Therefore, the terms "trivalent," "tetravalent," "pentavalent," and "hexavalent" respectively indicate the presence of three, four, five, and six antigen-binding domains in an antibody / antigen binding molecule. As used herein, the term "trivalent" refers to an antigen-binding molecule having three antigen-binding domains. The multispecific antigen-binding molecule of this disclosure is trispecific and specifically binds to TAA1, TAA2, and CD3. Therefore, the trivalent multispecific binding molecule of this disclosure has at least three antigen-binding domains, each binding to a different antigen. Examples of the trivalent multispecific binding molecules of this disclosure are schematically shown in Figures 1N or 1O. As used herein, the term "tetravalent" refers to an antigen-binding molecule having four antigen-binding domains. The multispecific antigen-binding molecules of this disclosure are trispecific and specifically bind to TAA1, TAA2, and CD3. Therefore, the tetravalent multispecific binding molecules of this disclosure typically have two antigen-binding domains binding to the same antigen (e.g., TAA1 or TAA2) and two antigen-binding domains each binding to a separate antigen (e.g., CD3, and TAA1 or TAA2). Examples of the tetravalent multispecific binding molecules of this disclosure are schematically shown in Figures 1H-1J. As used herein, the term "pentavalent" refers to an antigen-binding molecule having five antigen-binding domains. The multispecific binding molecules of this disclosure are trispecific and specifically bind TAA1, TAA2, and CD3. Therefore, the pentavalent multispecific binding molecules of this disclosure typically have two pairs of antigen-binding domains, each binding to the same antigen, and a single antigen-binding domain binding to a third antigen (e.g., CD3). Examples of the pentavalent multispecific binding molecules of this disclosure are schematically shown in Figures 1K-1M. As used herein, the term "hexavalent" refers to an antigen-binding molecule having six antigen-binding domains. The multispecific binding molecules of this disclosure are trispecific and specifically bind to TAA1, TAA2, and CD3. Although different configurations (e.g., three antigen-binding domains bound to TAA1, two antigen-binding domains bound to TAA2, and one antigen-binding domain bound to CD3, or three antigen-binding domains bound to TAA1, two antigen-binding domains bound to CD3, and one antigen-binding domain bound to TAA2) are within the scope of this disclosure, the hexavalent multispecific binding molecules of this disclosure generally have three pairs of two antigen-binding domains, each binding to the same antigen. Examples of the hexavalent multispecific binding molecules of this disclosure are schematically shown in Figures 1A-1G. The terms “full-length antibody,” “intact antibody,” and “complete antibody” are used interchangeably in this article and refer to antibodies with structures that are substantially similar to those of natural antibodies. The terms "antigen-binding fragment" and "antibody fragment" are used interchangeably in this article and do not constitute a complete antibody. The entire structure contains only a portion or a portion of the complete antibody, which has the ability to bind antigens. Exemplarily, the “antigen-binding fragment” or “antibody fragment” herein includes, but is not limited to, Fab, F(ab')2, Fab', Fab'-SH, Fd, Fv, scFv, diabody, and single-domain antibodies. The terms “linker”, “connector”, and “linker” are used interchangeably in this article to refer to the linking polypeptide sequence used to connect protein domains. They are usually flexible and the use of a linker will not result in the loss of the original function of the protein domain. The term "chimeric antibody" as used herein refers to an antibody having a variable sequence of an immunoglobulin derived from one source organism (such as rat, mouse, rabbit, or alpaca) and a constant region of an immunoglobulin derived from a different organism (such as human). Methods for producing chimeric antibodies are known in the art. See, for example, Morrison, 1985, Science 229(4719):1202-1207. Transfectomas Provide Novel Chimeric Antibodies; Gillies et al., J Immunol Methods. 1989 Dec 20; 125(1-2):191-202; the above are incorporated herein by reference. See also patent US5807715A. The term "humanized antibody" in this article refers to a genetically engineered non-human antibody whose amino acid sequence has been modified to increase its homology with that of a human antibody. Typically, all or part of the CDR region of a humanized antibody is derived from a non-human antibody (donor antibody), and all or part of the non-CDR region (e.g., the variable region FR and / or constant region) is derived from a human immunoglobulin (receptor antibody). Humanized antibodies generally retain or partially retain the intended properties of the donor antibody, including but not limited to antigen specificity, affinity, reactivity, ability to enhance immune cell activity, or ability to strengthen the immune response. The term "fully human antibody" as used herein refers to an antibody having variable regions in which both the FR and CDR are derived from human germline immunoglobulin sequences. Furthermore, if the antibody contains a constant region, that constant region is also derived from a human germline immunoglobulin sequence. "Fully human antibody" as used herein may include amino acid residues not encoded by human germline immunoglobulin sequences (e.g., mutations introduced through random or site-specific mutagenesis in vitro or through somatic mutations in vivo). However, "fully human antibody" as used herein does not include antibodies in which a CDR sequence derived from another mammalian species (e.g., mouse) has been grafted onto a human frame sequence. The term "variable region" in this article refers to the region in the antibody heavy or light chain involved in enabling the antibody to bind to the antigen. "Heavy chain variable region" is used interchangeably with "VH" and "HCVR," and "light chain variable region" is used interchangeably with "VL" and "LCVR." The variable regions of the heavy and light chains of natural antibodies generally have similar structures, each containing four conserved frame regions (FRs) and three hypervariable regions (HVRs). See, for example, Kindt et al., Kuby Immunology, 6th ed., WH Freeman and Co., p. 91 (2007). A single VH or VL is sufficient to confer antigen-binding specificity. The terms “complementarity-determining region” and “CDR” are used interchangeably herein and generally refer to a hypervariable region (HVR) found in both the light and heavy chain variable domains. A more conserved portion of the variable domain is called a frame region (FR). As understood in the art, the amino acid position representing the hypervariable region of an antibody can vary depending on the context and various definitions known in the art. Some positions within a variable domain can be considered heterozygous hypervariable positions because these positions can be considered within a set of standards (such as IMGT or KABAT) but outside a set of standards (such as KABAT or IMGT). One or more of these positions can also be found in extended hypervariable regions. This disclosure includes antibodies containing modifications at these heterozygous hypervariable positions. The heavy chain variable region CDR may be abbreviated as HCDR, and the light chain variable region may be abbreviated as LCDR. The variable structural domains of both the natural heavy and light chains each contain four framework regions, primarily employing a lamellar configuration, connected by three CDRs (CDR1, CDR2, and CDR3). These three CDRs form loops connecting the lamellar structures and, in some cases, form part of the lamellar structure. The CDRs in each chain are sequentially linked through FR regions. FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4 remain closely linked and, together with CDRs from other antibody chains, contribute to the formation of the antibody's antigen-binding site. The term "CDR" in this paper can be labeled and defined in a manner known in the art, including but not limited to the Kabat numbering system, the Chothia numbering system, or the IMGT numbering system. The tools and websites used include, but are not limited to, the AbRSA website (http: / / cao.labshare.cn / AbRSA / cdrs.php), the abysis website (www.abysis.org / abysis / sequence_input / key_annotation / key_annotation.cgi), and the IMGT website (http: / / www.imgt.org / 3Dstructure-DB / cgi / DomainGapAlign.cgi#results). The CDR in this paper includes overlaps and subsets of amino acid residues defined in different ways. The term "Kabat numbering system" in this article generally refers to the immunoglobulin alignment and numbering system proposed by Elvin A. Kabat (see, for example, Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md., 1991). The term “Chothia numbering system” in this article generally refers to the immunoglobulin numbering system proposed by Chothia et al., which is a classic rule for identifying the boundaries of CDR regions based on the location of structural loop regions (see, for example, Chothia, C., & Lesk, AM (1987). Canonical structures for the hypervariable regions of immunoglobulins. Journal of molecular biology, 196(4), 901-917.). The term "IMGT numbering system" used in this article generally refers to the numbering system based on the international ImMunoGeneTics information system (IMGT) initiated by Lefranc et al. See Lefranc, MP, Pommié, C., Ruiz, M., Giudicelli, V., Foulquier, E., Truong, L., ... & Lefranc, G. (2003). IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains. Developmental & Comparative Immunology, 27(1), 55-77. The term "heavy chain constant region" in this document refers to the carboxyl-terminal portion of the antibody heavy chain, which does not directly participate in antibody-antigen binding but exhibits effector functions, such as interaction with the Fc receptor, and has a more conserved amino acid sequence relative to the variable domains of the antibody. The "heavy chain constant region" can be selected from the CH1 domain, hinge region, CH2 domain, CH3 domain, or variants or fragments thereof. The term "heavy chain constant region" includes "full-length heavy chain constant region" and "heavy chain constant region fragment," the former having a structure substantially similar to the natural antibody constant region, while the latter includes only a portion of the full-length heavy chain constant region. Exemplarily, a typical "full-length antibody heavy chain constant region" consists of a CH1 domain-hinge region-CH2 domain-CH3 domain; when the antibody is IgE, it also includes a CH4 domain; when the antibody is a heavy chain antibody, it does not include the CH1 domain. Exemplarily, a typical "heavy chain constant region fragment" can be selected from the Fc or CH3 domain. The term "light chain constant region" in this article refers to the carboxyl terminus of the antibody light chain, which does not directly participate in the binding of the antibody to the antigen. The light chain constant region can be selected from the constant κ domain or the constant λ domain. The term "Fc region" used herein is used to define a C-terminal region in an antibody heavy chain that contains at least a portion of a constant region. This term includes both native sequence Fc regions and variant Fc regions. Exemplarily, the human IgG heavy chain Fc region may extend from Cys226 or Pro230 to the C-terminus of the heavy chain. However, antibodies generated by host cells may undergo post-translational cleavage, removing one or more, particularly one or two, amino acids from the C-terminus of the heavy chain. Therefore, antibodies generated by host cells through the expression of a specific nucleic acid molecule encoding the full-length heavy chain may include the full-length heavy chain, or it may include a cleaved variant of the full-length heavy chain. This may be the case when the final two C-terminal amino acids of the heavy chain are glycine (G446) and lysine (K447, numbered according to the Kabat EU index). Therefore, the C-terminal lysine (Lys447), or the C-terminal glycine (Gly446) and lysine (Lys447) of the Fc region may or may not be present. Typically, the IgG Fc region contains IgG CH2 and IgG CH3 domains; optionally, it may also contain complete or partial domains. The hinge region, but does not contain the CH1 domain. The “CH2 domain” of the human IgG Fc region typically extends from amino acid residue approximately at position 231 to amino acid residue approximately at position 340. In one embodiment, a carbohydrate chain is attached to the CH2 domain. The CH2 domain as used herein can be a native sequence CH2 domain or a variant CH2 domain. The “CH3 domain” comprises the residues at the C-terminus of the CH2 domain in the Fc region (i.e., from amino acid residue approximately at position 341 of IgG to amino acid residue approximately at position 447). The CH3 region as used herein can be a native sequence CH3 domain or a variant CH3 domain (e.g., a CH3 domain having a “knob” introduced in one chain and a corresponding “hole” introduced in the other chain; see U.S. Patent No. 5,821,333, which is expressly incorporated herein by reference). As described herein, such variant CH3 domains can be used to facilitate heterodimerization of two dissimilar antibody heavy chains. Unless otherwise specified herein, amino acid residues in the Fc region or constant region are numbered according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, National Institutes of Health, Bethesda, MD, 1991. The term "Fc variant" in this paper refers to a change in the structure or function of an Fc cell caused by one or more amino acid substitutions, insertions, or deletions at appropriate sites. "Inter-Fc variant interactions" refer to the potential for space-filling effects, electrostatic attraction, hydrogen bonding, and hydrophobic interactions between mutant-designed Fc variants. Inter-Fc variant interactions contribute to the formation of stable heterodimeric proteins. A preferred mutant design is a "knob-into-hole (KIH)" type mutant design. The mutation design technology of Fc variants has been widely used in the field to prepare bispecific antibodies or heterodimeric Fc fusion protein forms. Representative examples include the "Knob-into-Hole" form proposed by Cater et al. (Ridgway, JB, Presta, LG, & Carter, P. (1996). 'Knobs-into-holes' engineering of antibody CH3 domains for heavy chain heterodimerization. Protein Engineering, Design and Selection, 9(7), 617-621.); the formation of Fc-containing heterodimers by Amgen engineers using electrostatic steering (US 20100286374 A1); and Jonathan H. Davis et al. (Davis, JH, Aperlo, C., Li, Y., Kurosawa, E., Lan, Y., Lo, KM, & Huston, JS (2010). SEEDbodies: fusion proteins based on strand-exchange engineered domain (SEED)CH3 heterodimers in an Fc analogue platform for asymmetric binders or immunofusions and bispecific antibodies. Protein Engineering, Design & Selection, 23(4), 195-202.) proposed a heterodimer form (SEEDbodies) formed through IgG / IgA chain exchange; Genmab's DuoBody (Gramer, MJ, van den Bremer, ET, van Kampen, MD, Kundu, A., Kopfmann, P., Etter, E., ... & Parren, PW (2013, November). Production of stable bispecific IgG1 by controlled Fab-arm exchange: scalability from bench to large-scale manufacturing by application of standard approaches.The bispecific molecules formed using platform technology (in MAbs (Vol.5, No.6, pp.962-973). Taylor & Francis.); the heterodimeric protein form formed by Xencor's engineers through combined structural calculations and Fc amino acid mutations, and by combining different modes of action (Moore, GL, Bautista, C., Pong, E., Nguyen, DHT, Jacinto, J., Eivazi, A., ... & Lazar, GA (2011, November). A novel bispecific antibody format enables simultaneous bivalent and monovalent co-engagement of distinct target antigens. In MAbs (Vol.3, No.6, pp.546-557). Taylor & Francis.; the Fc modification method based on charge network of Suzhou Corning-Jerry Co., Ltd. (CN201110459100.7) to obtain heterodimeric protein forms; and other genetic engineering methods based on Fc amino acid changes or functional modifications to achieve the formation of heterodimeric functional proteins. The Knob / Hole structure on the Fc variant fragment described in this disclosure refers to the mutation of each of the two Fc fragments, which can then bind together in a "Knob-into-Hole" manner. The "knob-into-hole" method proposed by Cater et al. is preferred. The model is modified by site mutations in the Fc region so that the resulting first and second Fc variants can combine in a "knob-into-hole" manner to form a heterodimer. The selection of specific immunoglobulin Fc regions from specific immunoglobulin classes and subclasses is within the scope of those skilled in the art. Preferably, the Fc regions of human antibodies IgG1, IgG2, IgG3, and IgG4 are used, with the Fc region of human antibody IgG1 being more preferred. A knockout mutation is randomly performed on one of the first or second Fc variants, and a hole mutation is performed on the other. The term "effective function" in this article refers to the activity of an antibody molecule mediated by binding through its domains rather than antigen-binding domains, typically mediated by the binding of effector molecules. Effector functions include complement-mediated effector functions, which are mediated by the binding of, for example, the C1 component of the complement to the antibody. Complement activation is important in opsonization and lysis of cellular pathogens. Complement activation also stimulates inflammatory responses and can participate in autoimmune hypersensitivity responses. Effector functions also include Fc receptor (FcR)-mediated effector functions, which can be triggered by the binding of the antibody's constant domain to the Fc receptor (FcR). Antibody binding to Fc receptors on cell surfaces triggers many important and diverse biological responses, including phagocytosis and destruction of antibody-coated particles, clearance of immune complexes, lysis of antibody-coated target cells by cytotoxic cells (referred to as antibody-dependent cell-mediated cytotoxicity, or ADCC), release of inflammatory mediators, placental transfer, and control of immunoglobulin production. The effector function of an antibody can be altered, for example, by increasing or decreasing the antibody's affinity for effector molecules such as Fc receptors or complement components. Binding affinity is typically altered by modifying the binding site of the effector molecule, and in this case, it is appropriate to target the site of interest and modify at least a portion of the site in a suitable manner. It is also envisioned that altering the binding site on an antibody against an effector molecule does not need to significantly change the overall binding affinity, but can alter the geometry of the interaction, resulting in the disabling of the effector mechanism, as in nonproductive binding. Further envisioning is that effector function can also be altered by modifying sites that do not directly participate in effector molecule binding but otherwise participate in the performance of effector function. The term "conserved amino acid" in this document generally refers to amino acids that belong to the same class or have similar characteristics (e.g., charge, side chain size, hydrophobicity, hydrophilicity, main chain conformation, and rigidity). For example, the amino acids in each of the following groups belong to each other's conserved amino acid residues, and substitutions of amino acid residues within a group constitute substitutions of conserved amino acids: 1) Alanine (A), Serine (S), Threonine (T); 2) Aspartic acid (D), glutamic acid (E); 3) Asparagine (N), glutamine (Q); 4) Arginine (R), Lysine (K), Histidine (H); 5) Isoleucine (I), leucine (L), methionine (M), valine (V); and 6) Phenylalanine (F), Tyrosine (Y), Tryptophan (W). The term "identity" used herein can be calculated as follows: To determine the percentage of "identity" between two amino acid sequences or two nucleic acid sequences, the sequences are aligned for optimal comparison purposes (e.g., vacancies may be introduced in one or both of the first and second amino acid sequences or nucleic acid sequences for optimal alignment, or non-homologous sequences may be discarded for comparison purposes). The amino acid residues or nucleotides at corresponding amino acid or nucleotide positions are then compared. When a position in the first sequence is occupied by the same amino acid residue or nucleotide at the corresponding position in the second sequence, the molecules are identical at that position. Taking into account the number of vacancies that need to be introduced for optimal alignment of the two sequences and the length of each vacancy, the percentage of identity between the two sequences varies with the common positions of the sequences. Mathematical algorithms can be used to compare sequences and calculate the percentage of identity between two sequences. For example, the Needlema and Wunsch algorithms (available at www.gcg.com) from the GAP program, which is integrated into the GCG software package, can be used to perform sequence comparisons and calculate the percentage of identity between two sequences. The percentage of identity between two amino acid sequences is determined using a Blossum 62 matrix or a PAM250 matrix with vacancy weights of 16, 14, 12, 10, 8, 6, or 4 and length weights of 1, 2, 3, 4, 5, or 6. Alternatively, the percentage of identity between two nucleotide sequences can be determined using the GAP program in the GCG software package (available at www.gcg.com) with an NWSgapdna.CMP matrix and vacancy weights of 40, 50, 60, 70, or 80 and length weights of 1, 2, 3, 4, 5, or 6. A particularly preferred set of parameters (and one set of parameters that should be used unless otherwise specified) is a Blossum 62 scoring matrix with a vacancy penalty of 12, a vacancy extension penalty of 4, and a frameshift vacancy penalty of 5. Alternatively, the PAM120 weighted remainder table, a gap length penalty of 12, and a gap penalty of 4 can be used, along with the E. Meyers and W. Miller algorithm, which has been incorporated into the ALIGN program (version 2.0), to determine the percentage of identity between two amino acid sequences or nucleotide sequences. Additionally or alternatively, the nucleic acid and protein sequences described in this disclosure can be further used as "query sequences" to perform searches against public databases to, for example, identify sequences of other family members or related sequences. For example, such searches can be performed using the NBLAST and XBLAST programs (version 2.0) of Altschul et al., (1990) J. Mol. Biol. 215:403-10. BLAST nucleotide searches can be performed using the NBLAST program with a score of 100 and a word length of 12 to obtain nucleotide sequences homologous to the nucleic acid molecules of this disclosure. BLAST protein searches can be performed using the XBLAST program with a score of 50 and a word length of 3 to obtain amino acid sequences homologous to the protein molecules of this disclosure. For comparison purposes, alignment results with gaps can be obtained using methods such as those described by Altschul, SF, Madden, TL. AA, Zhang, J., Zhang, Z., Miller, W., & Lipman, DJ (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic acids research, 25(17), 3389-3402. Use vacancy BLAST as described in [reference needed]. When using BLAST and vacancy BLAST programs, the default parameters of the corresponding programs (e.g., XBLAST and NBLAST) can be used. See www.ncbi.nlm.nih.gov. The term "nucleic acid" in this document includes any compound and / or substance comprising a polymer of nucleotides. Each nucleotide consists of a base, particularly a purine or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T), or uracil (U)), a sugar (i.e., deoxyribose or ribose), and a phosphate group. Typically, nucleic acid molecules are described by the sequence of bases, whereby the bases represent the primary structure (linear structure) of the nucleic acid molecule. The sequence of bases is typically represented as 5' to 3'. In this document, the term nucleic acid molecule encompasses deoxyribonucleic acid (DNA), including, for example, complementary DNA (cDNA) and genomic DNA, ribonucleic acid (RNA), particularly messenger RNA (mRNA), synthetic forms of DNA or RNA, and polymers containing mixtures of two or more of these molecules. Nucleic acid molecules can be linear or circular. Furthermore, the term nucleic acid molecule includes both sense and antisense strands, as well as single-stranded and double-stranded forms. Moreover, the nucleic acid molecules described herein may contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases having residues with derived sugar or phosphate backbones or chemical modifications. Nucleic acid molecules also encompass DNA and RNA molecules suitable as carriers for the direct expression of the antibodies disclosed herein in vitro and / or in vivo, such as in a host or patient. Such DNA (e.g., cDNA) or RNA (e.g., mRNA) carriers can be unmodified or modified. For example, mRNA can be chemically modified to enhance the stability of the RNA carrier and / or the expression of the encoded molecule, thereby allowing the mRNA to be injected into a subject to generate antibodies in vivo (see, for example, Stadler, CR). H., Celik, L., Hebich, B., Roth, AS, Roth, RP, ... & Sahin, U. (2017). Elimination of large tumors in mice by mRNA-encoded bispecific antibodies. Nature medicine, 23(7), 815-817. or EP2101823B1). The term "isolated" nucleic acid in this document refers to a nucleic acid molecule that has been separated from the components of its natural environment. Isolated nucleic acids include nucleic acid molecules found in cells that typically contain such molecules, but which are located outside the chromosome or at a chromosomal location different from their natural chromosomal location. The term "vector" in this article refers to a nucleic acid molecule capable of amplifying another nucleic acid linked to it. This term includes vectors as self-replicating nucleic acid structures as well as vectors integrated into the genome of a host cell into which the vector has been introduced. Some vectors can direct the expression of nucleic acids operatively linked to them. Such vectors are referred to herein as "expression vectors." The term "host cell" in this article refers to a cell in which exogenous nucleic acids have been introduced, including the progeny of such cells. Host cells include "transformers" and "transformed cells," which include primary transformed cells and their progeny, regardless of the number of passages. Progeny may not be identical to parental cells in their nucleic acid contents and may contain mutations. This article includes mutant progeny with the same function or biological activity as those screened or selected in the initially transformed cells. The term "pharmaceutical composition" as used herein refers to a formulation which is present in a form that allows for the effective biological activity of the active ingredient contained therein, and which does not contain any additional ingredients that would have unacceptable toxicity to a subject administering the pharmaceutical composition. The term "pharmaceutically acceptable carrier" as used herein includes any and all solvents, dispersion media, coating materials, surfactants, antioxidants, preservatives (e.g., antibacterial agents, antifungal agents), isotonic agents, absorption delay agents, salts, preservatives, drug stabilizers, binders, excipients, disintegrants, lubricants, sweeteners, flavoring agents, dyes, and combinations thereof, as known to those skilled in the art (see, for example, Remington's Pharmaceutical Sciences, 18th edition. Mack Printing Company, 1990, pp. 1289-1329). The use of any conventional carrier in therapeutic or pharmaceutical compositions is considered, except in cases of incompatibility with the active ingredient. In this article, the term "treatment" refers to surgical or therapeutic treatment aimed at preventing, mitigating (reducing) undesirable physiological changes or lesions in the treated individual, such as cancer and tumors. Beneficial or desired clinical outcomes include, but are not limited to, symptom relief, disease severity reduction, disease stability (i.e., no worsening), delay or slowing of disease progression, improvement or mitigation of disease status, and remission (whether partial or complete), whether detectable or undetectable. Individuals requiring treatment include those already suffering from the condition or disease, those susceptible to the condition or disease, or those intending to prevent the condition or disease. When the terms mitigation, relief, reduction, mitigation, or remission are used, they also include elimination, disappearance, and non-occurrence. The term "subject" in this document refers to an organism that receives treatment for a particular disease or condition as described herein. Exemplarily, "subject" includes mammals such as humans, primates (e.g., monkeys), or non-primate mammals that receive treatment for a disease or condition. The term "effective dose" in this document refers to a therapeutic dose, when administered alone or in combination with another therapeutic agent to cells, tissues, or subjects, that is effective in preventing or alleviating symptoms of a disease or the progression of that disease. "Effective dose" also refers to a dose of a compound sufficient to relieve symptoms, such as treating, curing, preventing, or alleviating an associated medical condition, or increasing the rate at which such symptoms are treated, cured, prevented, or alleviated. When an active ingredient is administered alone to an individual, the therapeutically effective dose refers to that ingredient alone. When a combination is used, the therapeutically effective dose refers to the combined dose of active ingredients that produce a therapeutic effect, regardless of whether they are administered in combination, consecutively, or simultaneously. The term "cancer" in this article refers to or describes a physiological condition in mammals characterized by unregulated cell growth. This definition includes both benign and malignant cancers. The term "tumor" or "tumor" in this article refers to all neoplastic cell growth and proliferation, whether malignant or benign, and all pre-cancerous and cancerous cells and tissues. The terms "cancer" and "tumor" are not mutually exclusive when used in this article. The term "EC50" in this document refers to the half-maximal effective concentration, which includes the antibody concentration that induces a half-range response between baseline and maximum value after a specified exposure time. EC50 essentially represents the antibody concentration in which 50% of its maximum effect is observed, and can be measured by methods known in the art. Detailed Implementation GPRC5D is a subtype D of the C5 family of G protein-coupled receptors, belonging to the orphan receptor family, and is a 7-transmembrane protein. GPRC5D is a novel specific target for multiple myeloma after BCMA. Tissue expression profiling studies have revealed that GPRC5D is specifically highly expressed in plasma cells of multiple myeloma, while its expression is low in normal tissues, limited to immune-forgiving hair follicle regions, and its expression does not overlap with that of BCMA. In tumor relapse models with BCMA loss, GPRC5D CAR-T therapy still shows therapeutic efficacy, overcoming tumor escape. CD3 molecules are important markers of T cells, part of the T cell receptor (TCR), capable of inducing T cell activation and associated with tumor immunity. This disclosure provides a novel multispecific antigen-binding molecule that simultaneously targets tumor antigens BCMA, GPRC5D, and T cell (CD3), combining two different targets of multiple myeloma, thereby inducing T cells to participate in tumor killing, enhancing the tumor killing effect, and thus providing a new approach to tumor treatment. This disclosure also provides, along with the encoding of nucleic acid fragments, vectors, host cells, immune effector cells, methods of preparation, pharmaceutical compositions, pharmaceutical uses, and treatments for tumors or cancers (e.g., B-cell lymphoma or multiple myeloma) that specifically bind to B-cell maturation antigen (BCMA), GPRC5D, and T cells. In a first aspect, this disclosure provides a multispecific binding molecule, said multispecific binding molecule comprising: (a) The first antigen-binding molecule (ABM1) that specifically binds to the first tumor-associated antigen (TAA1); (b) A second antigen-binding molecule (ABM2) that specifically binds to a second tumor-associated antigen (TAA2); and (c) A third antigen-binding molecule (ABM3) that specifically binds to antigens expressed on human immune cells. In some implementations, TAA1 and / or TAA2 are expressed on cancerous B cells. In some implementations, the cancerous B cells are of plasma cell origin or B cell origin. In some implementations, the immune cells are selected from T cells, NK cells, or macrophages. In some embodiments, TAA1 and TAA2 are each independently selected from BCMA and GPRC5D, and / or the antigen expressed on the immune cells is CD3. In some implementations, ABM1, ABM2, and ABM3 may each be independently selected from antibodies, antibody fragments, F(ab')2, Fab', Fab, Fv, scFv, nanobodies, or VHH. In some embodiments, the multispecific binding molecule comprises four polypeptide chains: a first heavy chain, a first light chain, a second heavy chain, and a second light chain. In some specific implementations, (1) the first heavy chain and the second heavy chain are identical and have the following structure: (a)VH a -CH1-L1-VHH b -Fc-L2-VH CD3 -L3-VL CD3 ; (b)VH a -CH1-L1-VH CD3 -L2-VL CD3 -Fc-L3-VHH b ; (c)VHH b -L1-VH a -CH1-L2-VHCD3 -L3-VL CD3 -Fc; (d)VH a -CH1-L1-VHH b -L2-VH CD3 -L3-VL CD3 -Fc; or (e)VH a -CH1-L1-VH b -L2-VL b -Fc-L3-VH CD3 -L4-VL CD3 ; And / or, (2) the first light chain and the second light chain are identical and have the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively; VHH b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA; VH CD3 and VL CD3 These are the heavy chain variable region and light chain variable region that specifically bind to CD3, respectively; Fc is the Fc region of any antibody; and L1, L2, L3, and L4 are the same or different linkers, respectively. In some specific implementations, (1) the first heavy chain and the second heavy chain are different and have the following structure: (a) First heavy chain VH a -CH1-L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (b) First heavy chain VHH b -L1-VH a -CH1-Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (c) First heavy chain VH a -CH1-L1-VH b -L2-VL b -Fc1 and second heavy chain VH a -CH1-L3-VHb -L4-VL b -Fc2-L5-VH CD3 -L6-VL CD3 ; (d) First heavy chain VH a -CH1-L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VHH b -Fc2-L3-VH CD3 -L4-VL CD3 ;or (e) First heavy chain VHH b -L1-VH a -CH1-Fc1 and second heavy chain VHH b -L2-VH a -CH1-Fc2-L3-VH CD3 -L4-VL CD3 ; Optionally, the carboxyl terminus of the first heavy chain in (a) to (e) above is further marked with a Flag tag and / or the carboxyl terminus of the second heavy chain is further marked with a His tag; And, (2) the first light chain and the second light chain are identical and have the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively; VHH b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA; VH CD3 and VL CD3 These are the heavy chain variable region and light chain variable region that specifically bind to CD3, respectively; Fc1 and Fc2 are each the Fc region of any antibody; and L1, L2, L3, L4, L5 and L6 are the same or different linkers, respectively. In some embodiments, the multispecific binding molecule comprises three polypeptide chains: a first heavy chain, a second heavy chain, and a light chain, wherein (1) the first heavy chain and the second heavy chain have the following structures: (a) First heavy chain VHH b -L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (b) First heavy chain VHHb -L1-VH CD3 -L2-VL CD3 -Fc1 and second heavy chain VH a -CH1-Fc2; or (c) First heavy chain VHH b -Fc1 and second heavy chain VH a -CH1-L1-VH CD3 -L2-VL CD3 -Fc2; Optionally, the carboxyl terminus of the first heavy chain in (a) to (c) above is further marked with a Flag tag and / or the carboxyl terminus of the second heavy chain is further marked with a His tag; And, (2) the light chain has the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively; VHH b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA; VH CD3 and VL CD3 These are the heavy chain variable region and the light chain variable region that specifically bind to CD3, respectively; and L1, L2 and L3 are either the same or different linkers. In some embodiments, the Fc1 and Fc2 of the multispecific binding molecule are different, with Fc1 being knock-Fc and Fc2 being hole-Fc; or, Fc1 being hole-Fc and Fc2 being knock-Fc; preferably, Fc1 and Fc2 are the Fc regions of IgG1 or IgG4. In some specific implementations, the hole-Fc contains the T366W mutation, and / or the hole-Fc contains the T366S, L368A, and / or Y407V mutations. In some embodiments, the ABM3 in the multispecific binding molecule is a CD3 antibody or its antigen-binding fragment, comprising a light chain variable region and a heavy chain variable region: (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are shown in SEQ ID NO. 54–56, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 54–56; and (b) the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are shown in SEQ ID NO. 51–53. It may be indicated that, or compared with SEQ ID NO.51-53, it has 1, 2, 3 or more amino acid insertions, deletions and / or substitutions. In some specific embodiments, the light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 10 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 10, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 9 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 9. In some specific implementations, ABM3 is scFv. In some embodiments, TAA1 and TAA2 in the multispecific binding molecule are selected from the group consisting of: (a) TAA1 being BCMA and TAA2 being GPRC5D; or (b) TAA1 being GPRC5D and TAA2 being BCMA. In some specific embodiments, the antigen-binding molecule that specifically binds to GPRC5D in the multispecific binding molecule comprises a light chain variable region and a heavy chain variable region of the antibody, wherein (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 42–44, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 42–44, and the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 39–41, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 39–41; or (b) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 48–50, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 39–41; Compared to SEQ ID NO. 48-50, the heavy chain variable region has 1, 2, 3 or more amino acid insertions, deletions and / or substitutions, and the heavy chain variable region includes HCDR1, HCDR2 and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 45-47, or have 1, 2, 3 or more amino acid insertions, deletions and / or substitutions compared to SEQ ID NO. 45-47. In some specific implementation schemes, where: (a) The light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 6 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 6, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 5 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 5; or (b) The light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 8 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 8, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 7 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 7. In some specific implementations, the antigen-binding molecule that specifically binds GPRC5D in the multispecific binding molecule is Fab. In some specific embodiments, the antigen-binding molecule that specifically binds to BCMA in the multispecific binding molecule includes a heavy chain variable region of the antibody, the heavy chain variable region including HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 27–29 or as shown in SEQ ID NO. 30–32, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 27–29 or SEQ ID NO. 30–32. In some specific embodiments, the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO.1 or SEQ ID NO.2, or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO.1 or SEQ ID NO.2. In some specific implementations, the antigen-binding molecule that specifically binds to BCMA is VHH. In some specific embodiments, the antigen-binding molecule that specifically binds to BCMA in the multispecific binding molecule comprises a light chain variable region and a heavy chain variable region of the antibody: (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 36–38, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 36–38; and (b) the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 33–35, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 33–35. In some specific embodiments, the light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 4 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 4, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 3 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 3. In some specific implementations, the antigen-binding molecule that specifically binds to BCMA among the multispecific binding molecules is scFv. In some implementations, ABM1, ABM2, and / or ABM3 are chimeric, humanized, or fully humanized. In some embodiments, the multispecific binding molecule further comprises an Fc fragment of human IgG1, IgG2, IgG3, or IgG4; preferably, it comprises an Fc fragment of human IgG1 or human IgG4. In some specific embodiments, the Fc fragment of the multispecific binding molecule has an amino acid mutation that alters the effector function; preferably, the amino acid mutation that alters the effector function includes S228P, F234A, L234A and / or L235A mutations; more preferably, the amino acid mutation that alters the effector function includes: (a) L234A and / or L235A mutations in IgG1Fc; and / or (b) S228P, F234A and / or L235A mutations in IgG4Fc. In some embodiments, the Fc fragment of the multispecific binding molecule comprises the same or different Fc1 and Fc2. In some specific embodiments, different Fc1 and Fc2 in the multispecific binding molecule have heterodimerization-promoting amino acid mutations; preferably, the heterodimerization-promoting amino acid mutations include a knob-into-hole (KIH) mutation; more preferably, the knob-into-hole mutation includes: (a) either Fc region contains a T366W mutation; and (b) the other Fc region contains a T366S, L368A, and / or Y407V mutation. In some specific implementations, Fc1 further includes a His tag, and Fc2 further includes a Flag tag; or Fc1 further includes a Flag tag, and Fc2 further includes a His tag. In some embodiments, the multispecific binding molecule is trivalent, tetravalent, pentavalent, or hexavalent. Secondly, this disclosure provides an isolated nucleic acid that encodes the aforementioned multispecific binding molecule. Thirdly, this disclosure provides a recombinant vector containing the aforementioned isolated nucleic acid. Fourthly, this disclosure provides a host cell comprising the aforementioned isolated nucleic acid or the aforementioned recombinant vector; preferably, the host cell is a eukaryotic cell or a prokaryotic cell; more preferably, the host cell is derived from mammalian cells, yeast cells, insect cells, Escherichia coli and / or Bacillus subtilis; even more preferably, the host cell is an Expi293 cell. Fifthly, this disclosure provides a method for preparing a multispecific binding molecule, comprising culturing the aforementioned host cells under appropriate conditions and isolating the multispecific binding molecule. Sixthly, this disclosure provides a pharmaceutical composition comprising the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned recombinant vector, the aforementioned host cell, or a product prepared according to the aforementioned method, and one or more of the above. Pharmaceutically acceptable carriers, diluents, buffers, or excipients; preferably, the pharmaceutical composition further comprises additional cancer therapeutic agents. In a seventh aspect, this disclosure provides a method for activating T cells, comprising administering to a subject a therapeutically effective amount of the aforementioned multispecific binding molecule or pharmaceutical composition. Eighthly, this disclosure provides the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned host cell, the aforementioned product prepared by the aforementioned method, or the aforementioned pharmaceutical composition for use in the preparation of a medicament for treating and / or preventing cancer, or for use in treating and / or preventing cancer. In a preferred embodiment, the cancer is B-cell lymphoma; more preferably, the B-cell lymphoma is multiple myeloma. Ninthly, this disclosure provides a method for treating and / or preventing cancer, comprising administering to a subject a therapeutically effective amount of the aforementioned multispecific binding molecule, the aforementioned isolated nucleic acid, the aforementioned host cell, a product prepared by the aforementioned method, or the aforementioned pharmaceutical composition. In a preferred embodiment, the cancer is B-cell lymphoma; more preferably, the B-cell lymphoma is multiple myeloma. Example The present disclosure will be further described below with reference to specific embodiments, and the advantages and features of the present disclosure will become clearer with the description. Unless otherwise specified in the embodiments, conventional conditions or conditions recommended by the manufacturer shall apply. Reagents or instruments used, unless otherwise specified, are all commercially available products. The embodiments disclosed herein are merely exemplary and do not constitute any limitation on the scope of this disclosure. Those skilled in the art should understand that modifications or substitutions can be made to the details and form of the technical solutions disclosed herein without departing from the spirit and scope of this disclosure, but all such modifications and substitutions fall within the protection scope of this disclosure. Example 1: Construction of the anti-BCMA-GPRC5D-CD3 triple antibody molecule Through a series of experiments including animal immunization, sequencing, and humanization, superior humanized anti-BCMA specific antibodies (BCMA-VHH01, BCMA-VHH02, and BCMA-Hab01), humanized anti-GPRC5D specific antibodies (GPRC5D-Hab01 and GPRC5D-Hab02), and humanized anti-CD3 specific antibody (CD3-Hab01) were obtained. These specific antibodies were then assembled into a trispecific antibody BCMA-GPRC5D-CD3, with an assembly ratio of 2:2:2 (BCMA:GPRC5D). The antibody fragments are scFv, Fab, or VHH, and the heavy chain constant region subtype is IgG1 or IgG4. If the molecule has an asymmetric structure, mismatches are reduced by using a "Knob in Hole". Finally, the fragments are linked together with a linker and constructed into a PTT5 expression vector to obtain the target plasmid. The schematic diagram of the triple antibody molecular structure is shown in Figure 1. The molecular composition of each triple antibody is shown in Table 1. The amino acid sequence of the antibody molecule is shown in Table 2. The CDR determined by the Kabat numbering system of the main antibody variable region is shown in Table 3. The positive control antibodies used in this embodiment were Regeneron's bispecific antibody REGN5459 (WO2020018820A1, targeting BCMA antigen and CD3 receptor expressed on T cell surface) and Johnson & Johnson's bispecific antibody JNJ7564 (WO2019220368A1, targeting GPRC5D antigen and CD3 receptor expressed on T cell surface), and the negative control antibody was antiFITC×CD3-Hab01 (internally prepared). The structures of the positive and negative control antibodies are shown in Figure 1, PR. Both REGN5459 and JNJ7564 contain Fc regions that prevent homodimerization. mutation. Table 1. Structural composition of the three-antibody molecules Table 2. Amino acid sequences of antibody molecules Table 3. CDR sequences determined by the antibody Kabat numbering system. Example 2: Expression and purification of BCMA-GPRC5D-CD3 triple antibody molecule and control antibody 2.1 Expression and purification of the three-antibody molecules The plasmid and transfection reagent PEI (Polysciences, catalog number: 24765-1) were added to OPTI-MEM (Gibco, catalog number: 11058021), mixed well, and incubated for 15 min. This mixture was then added to Expi293F cells (Thermofisher, catalog number: A14527) and cultured in a shaker at 37°C and 120 rpm with 5% CO2. On the second day after transfection, OPM-293ProFeed (Shanghai Aopomai, catalog number: F081918-001) and 6 g / L glucose (Sigma, catalog number: G7528) were added. On the sixth day after transfection, the cell supernatant was collected. The protein was purified using AKTA Pure via Protein A affinity and molecular sieve filtration. The obtained antibody was quantitatively and qualitatively analyzed by SDS-PAGE, SEC-HPLC, and CE-SDS. Depending on the antibody structure, the following purification methods can be used: (A) Purification method of symmetric triple antibodies Tri-antibodies Tri-17, Tri-29, Tri-32, Tri-35, Tri-55, Tri-59, and Tri-60 are suitable for this method. The specific purification method is as follows: 1. Use Protein A column (Mabselect SuRe TM Protein A column (purchased from Cytiva) was initially purified. The column was first equilibrated with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), then the clear culture supernatant was loaded at a flow rate of 8 mL / min. After loading, the column was washed with 3–5 column volumes of high-salt eluent (20 mM phosphate buffer, 1 M NaCl, pH 7.4). Proteins bound to the Protein A column were eluted with elution buffer (20 mM citrate buffer, pH 3.5) and then separated by A. 280 The protein elution was monitored using ultraviolet absorption peaks. The eluted protein was collected and neutralized to pH 5-6 with 1M pH 8.0 Tris-HCl. 2. Purify the target sample using molecular sieves (Ezload 16 / 60 Chromdex 200pg purchased from BorgLone or Superdex 200 Increase 10 / 300 GL purchased from Cytiva), collect the sample, concentrate it, dialyze it into 559 buffer (10mM acetic acid, 9% sucrose, pH 5.5), filter it aseptically through a 0.22μm filter, and store it aseptically to obtain the purified corresponding antibody. (B) Purification method of class I asymmetric triple antibody Tri-antibodies Tri-38, Tri-41, Tri-46, Tri-49, Tri-52, Tri-56, and Tri-62 are suitable for this method. The specific purification method is as follows: 1. Initial purification using a Protein A column. Equilibrate the Protein A column with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), then load the clear culture supernatant at a flow rate of 8 mL / min. After loading, wash with 3–5 column volumes of high-salt eluent. Remove redundant semi-antibodies with elution buffer A (20 mM citrate buffer, pH 3.85), and elute the target protein with elution buffer B (20 mM citrate buffer, pH 3.5). Then, pass the eluent through column A... 280 The protein elution was monitored using ultraviolet absorption peaks. The eluted protein was collected and neutralized to pH 5-6 with 1M pH 8.0 Tris-HCl. 2. Purify the sample using molecular sieves and collect the target sample. After concentration, dialyze the sample into 559 buffer solution, filter it aseptically using a 0.22 μm filter, and preserve it aseptically to obtain the purified corresponding antibody. (C) Purification method for Class II asymmetric triple antibodies Tri-antibodies Tri-63, Tri-64, Tri-65, and Tri-66 are suitable for this method. The specific purification method is as follows: 1. Initial purification using a Protein A column. Equilibrate the Protein A column with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), then load the clear culture supernatant at a flow rate of 8 mL / min. After loading, wash with 3–5 column volumes of high-salt eluent. Elute the protein bound to the Protein A column with elution buffer (20 mM citrate buffer, pH 3.5), monitoring protein elution using the A280 UV absorption peak. Collect the eluted protein and neutralize to pH 5–6 with 1 M pH 8.0 Tris-HCl. 2. Purification was performed using HiTrap KappaSelect (purchased from Cytiva). First, equilibrate with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), then load the purified protein solution at a flow rate of 5 mL / min. After loading, wash with 3–5 column volumes of PBS, followed by 10 column volumes of 10 mM phosphate buffer (pH 7.4). Elution buffer C (50 mM glycine, pH 3.6) was used to elute proteins bound to the KappaSelect column, and elution buffer D (50 mM glycine, pH 3.0) was used to elute other proteins. Protein elution was monitored using the A280 UV absorption peak. The eluted protein was collected and neutralized to pH 5–6 with 1 M pH 8.0 Tris-HCl. 3. Purify the sample using molecular sieves and collect the target sample. After concentration, dialyze the sample into 559 buffer solution, filter it aseptically using a 0.22 μm filter, and preserve it aseptically to obtain the purified corresponding antibody. Triple antibodies purified by different methods were analyzed using SEC-HPLC to characterize the molecular size uniformity and purity of the protein samples. The HPLC system used was an Agilent 1260, with a TSKgel G3000SWXL column from Tosoh Bioscience. The mobile phase was 200 mM phosphate buffer, pH 7.0 / isopropanol (v / v 9:1). The detection temperature was 25℃, the flow rate was 0.5 mL / min, and the detection wavelength was 280 nm. SEC-HPLC data were analyzed using manual integration of chromatograms. Protein purity was calculated using the area normalization method. The main peak was considered the monomer, the peaks preceding the main peak were called aggregates, and the peaks following the main peak were called fragments. The results are shown in Table 4. Table 4: Results of Tri-antibody molecule expression levels and SEC purity 2.2 Expression and purification of control antibodies Anti-hel-hIgG1 (often abbreviated as hIgG1 in this application) is an antibody targeting chicken egg lysozyme (purchased from Baiying, catalog number: B117901), commonly used as a negative control antibody in antibody screening and identification processes. The expression methods for REGN5459, JNJ7564, and antiFITC×CD3-Hab01 are the same as in section 2.1. The expression and purification methods for the three antibody molecules are as follows: (A) Purification method of REGN5459 and antiFITC×CD3-Hab01 1. Initial purification using a Protein A column. Equilibrate the Protein A column with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), then load the clear culture supernatant at a flow rate of 8 mL / min. After loading, wash with 3–5 column volumes of high-salt eluent (20 mM phosphate buffer, 1 M NaCl, pH 7.4). Elute the target protein using a linear gradient of elution buffer A (20 mM citrate buffer, pH 3.85) and elution buffer B (20 mM citrate buffer, pH 3.5), monitoring protein elution using the A280 UV absorption peak. Collect the eluted protein and neutralize to pH 5–6 with 1 M pH 8.0 Tris-HCl. 2. The target sample was purified using molecular sieves and collected. It was then stored in PBS buffer, sterilely filtered through a 0.22 μm filter, and sterilely preserved to obtain purified REGN5459 and antiFITC×CD3-Hab01 antibodies. (B)JNJ7564 Purification Method 1. Purify the corresponding monospecific antibodies JNJ7564-anti-GPRC5D and JNJ7564-anti-CD3 using a Protein A column. The Protein A column was first equilibrated with 3–5 column volumes of equilibration buffer (PBS buffer, pH 7.4), and then the clear culture supernatant was loaded at a flow rate of 8 mL / min. After loading, the column was washed with 3–5 column volumes of high-salt eluent (20 mM phosphate buffer, 1 M NaCl, pH 7.4). The protein bound to the Protein A column was eluted with elution buffer (20 mM citrate buffer, pH 3.5), and the elution was monitored by the A280 UV absorption peak. The eluted protein was collected, neutralized to pH 5–6 with 1 M pH 8.0 Tris-HCl, and dialyzed to PBS buffer. 2. Bispecific antibody generation via in vitro reduction and re-oxidation. Mix 1-20 mg / ml of anti-GPRC5D / anti-CD3 antibody at a molar ratio of 1.08:1, add 75 mM 2-mercaptoethanol (2-MEA), and incubate at 25-37°C for 2-6 hours. Remove 2-mercaptoethanol (2-MEA) by dialysis, ultrafiltration, or desalting. Reassemble at 20°C to generate the bispecific antibody. All triple antibodies purified by different methods were analyzed using SEC-HPLC to characterize the molecular size uniformity and purity of the protein samples. The method was the same as above. The results are shown in Table 5. Table 5: Results of control molecule expression levels and SEC purity Example 3 Construction and Identification of Engineered Cell Lines 3.1 Preparation of CHOK1-human GPRC5D cells and CHOK1-monkey GPRC5D cells Nucleotide sequences encoding the human GPRC5D amino acid sequence (NCBI Gene ID: 55507) and the monkey GPRC5D amino acid sequence (NCBI Gene ID: 102122120) were cloned into the pLVX lentiviral vector, and viral particles were prepared in HEK293T cells. After lentiviral infection of the CHOK1 cell line, cells were selectively cultured for one week in Advanced DMEM / F12 Medium (Gibco, catalog number 12634028) containing 10 μg / ml puromycin (Gibco, catalog number A1113803) and 10% (w / w) fetal bovine serum (ExCell Bio, catalog number FND500). JNJ7564 was used as the primary antibody. Goat anti-human IgG (H+L) antibody (Jackson, catalog number: 109605088) was used as a secondary antibody for detection on a flow cytometer FACS CantoII (purchased from BD Biosciences). Cells with high expression levels and uniform peak shapes were amplified, and the amplified cells were re-analyzed by flow cytometry. Positive cell populations with good growth, high fluorescence intensity, and good uniformity were selected for further expansion culture and cryopreserved in liquid nitrogen. The previously prepared JNJ7564-anti-GPRC5D-hIgG1 (constructed by General Biosystems (Anhui) Co., Ltd., which cloned the GPRC5D-VH and VL sequences from JNJ7564 into expression vectors PTT5-huIgG1 and PTT5-huIgGLC (Kappa) containing the signal peptide MGWSWILLFLLSVTAGVHS (SEQ ID NO.57) and the human heavy / light chain constant region, respectively) was used for identification. This antibody showed cross-reactivity with human and monkey GPRC5D. The FACS results are shown in Figures 2A-2B. The results showed a single peak and can be used to monitor the binding activity of multispecific binding molecules and human and monkey GPRC5D. 3.2 Preparation of Flp-inCHO-human BCMA cell line overexpressing the target antigen BCMA The nucleotide sequence encoding the human BCMA amino acid sequence (NCBI Gene ID: 608) was cloned into the PcDNA5 / FRT vector (purchased from Clontech). After transfection of the Flp-in CHO cell line, cells were selectively cultured for 2 weeks in F12K Medium (purchased from Gibco, catalog number 21127030) containing 10% (w / w) fetal bovine serum (purchased from ExCell Bio, catalog number FND500) and 600 μg / ml hygromycin (purchased from ThermoFisher, catalog number 10687010). Detection was performed on a flow cytometer FACS CantoII (purchased from BD Biosciences) using REGN5459 antibody as the primary antibody and goat anti-human IgG (H+L) antibody (Jackson, catalog number: 109605088) as the secondary antibody. Cells with high expression levels and a single peak shape were amplified, and the amplified cells were re-analyzed by flow cytometry. Select positive cell populations with good growth, high fluorescence intensity, and good uniformity for further expansion culture and cryopreservation in liquid nitrogen. The previously prepared REGN5459-anti-BCMA-hIgG1 (constructed by General Biosystems (Anhui) Co., Ltd., which cloned the BCMA-VH and VL sequences from REGN5459 into expression vectors PTT5-huIgG1 and PTT5-huIgGLC(Kappa) containing the signal peptide MGWSWILLFLLSVTAGVHS and the human heavy / light chain constant region, respectively, to build chimeric monoclonal antibodies) was identified. This antibody exhibits cross-reactivity with human and monkey BCMA. The FACS results, shown in Figure 3, are a single peak and can be used to monitor the binding activity of multispecific binding molecules and human BCMA. 3.3 Preparation of stable cell lines for luciferase reporter gene transfection The pLvx-luciferase-P2A-GFP-RES-hygro vector system was packaged into lentiviruses, and Molp-8 cells were then infected with the virus. After 24 hours, the medium was changed, and 0.125 mg / ml hygro was added for pressure selection. After two weeks of hygro pressure selection, the cell pool showed a GFP expression peak of approximately 40-fold according to FACS, indicating successful pool infection. Cells were seeded using a limiting dilution method, 1 cell / well, in three 96-well plates. After culturing at 37°C for 2-3 weeks, approximately 30 monoclonal cell lines grew. These monoclonal cells were then transfected into 24-well plates and cultured for 3-4 days. Based on cell growth status and GFP intensity, three monoclonal cells were selected for further expansion, and luciferase expression in the cells was further detected. Finally, highly fluorescent Molp-8-Luc monoclonal cells were obtained. The H929-Luc pool and RPMI8226-Luc pool were purchased from Nanjing Kebai, and the preparation method of the monoclonal cells was the same as for Molp-8-Luc. Example 4: Identification of the binding of BCMA-GPRC5D-CD3 triple antibody molecule to target antigen and CD3 (A) Enzyme-linked immunosorbent assay (ELISA) to detect the binding of BCMA-GPRC5D-CD3 triple antibody to human BCMA protein and its cross-binding with monkey BCMA protein. The specific method is as follows: human BCMA protein (purchased from Acro, catalog number: BCA-H522y) and monkey BCMA protein (purchased from Acro, catalog number: BCA-C52H7) were diluted with PBS to a final concentration of 1 μg / mL, and then 50 μl was added to each well to a concentration of 96. Prepare ELISA plates. Seal with plastic film and incubate overnight at 4°C. The next day, wash the plate twice with PBS, add blocking buffer [PBS + 2% (w / v) BSA], and block at room temperature for 2 hours. Discard the blocking buffer, wash the plate three times with PBS, and add 50 μl of serially diluted 100 nM BCMA-GPRC5D-CD3 antibody or control antibody per well. Incubate at 37°C for 2 hours, then wash the plate three times with PBS. Add HRP (horseradish peroxidase)-labeled goat anti-human secondary antibody (purchased from Merck, catalog number: AP113P), incubate at 37°C for 1 hour, and wash the plate five times with PBS. Add 50 μl of TMB substrate per well, incubate at room temperature for 10 minutes, and then add 50 μl of stop buffer (1.0 M HCl) per well. Read the OD value at 450 nm using an ELISA plate reader (Multimode Plate Reader, EnSight, purchased from Perkin Elmer). As shown in Figures 4A and 4B, all three antibody molecules could bind to human BCMA and exhibited cross-binding activity with monkey BCMA. The IgG subtype control was human IgG1 (anti-HEL-IgG1). (B) Cell-based enzyme-linked immunosorbent assay (ELISA) to detect the binding of BCMA-GPRC5D-CD3 triple antibody to CHOK1-human GPRC5D cells and CHOK1-monkey GPRC5D cells. The specific method is as follows: The desired cells are cultured in T-75 cell culture flasks until the logarithmic growth phase. The culture medium is removed, and the cells are washed twice with PBS buffer. The cells are then digested with trypsin, and digestion is terminated with complete culture medium. The cells are then pipetted into a single-cell suspension. After cell counting, the cells are centrifuged, and the cell pellet is resuspended in complete culture medium to a concentration of 4 × 10⁻⁶. 5 Add 100 μl of cells per well to each of a 96-well flat-bottom cell culture plate and incubate overnight at 37°C in a 5% CO2 incubator. The next day, discard the culture medium in the 96-well plate, wash the cells once with PBS, and then fix the cells with immunostaining fixation solution (purchased from Beyotime, catalog number: P0098-500ml), 50 μl per well, at room temperature for 0.5 hours. Add blocking solution [PBS + 5% (w / v) skim milk] and block at room temperature for 2 hours. Discard the blocking solution, wash the plate 3 times with PBS, and add 50 μl of 100 nM serially diluted BCMA-GPRC5D-CD3 antibody or control antibody per well. Incubate at 37°C for 2 hours, then wash the plate 3 times with PBS. Add HRP (horseradish peroxidase)-labeled goat anti-human secondary antibody, incubate at 37°C for 1 hour, and then wash the plate 5 times with PBS. Add 50 μl of TMB substrate to each well, incubate at room temperature for 10 minutes, then add 50 μl of stop solution (1.0 M HCl) to each well. Read the OD value at 450 nm using an ELISA reader. The results are shown in Figures 5A and 5B. All the triple antibody molecules could bind to CHOK1-human GPRC5D cells and showed cross-binding activity with CHOK1-monkey GPRC5D cells. (C) Enzyme-linked immunosorbent assay (ELISA) to detect the binding of BCMA-GPRC5D-CD3 triple antibody to human CD3e / d protein and human CD3e protein. The specific method was the same as in Example 4(A). The proteins coated on the 96-well ELISA plates were human CD3e / d protein (purchased from Sino Biological, catalog number: CT032-H2508) and human CD3e protein (purchased from Sino Biological, catalog number: 10977-H08H). The results are shown in Figures 6A and 6B. The triple antibody molecules bound to both human CD3e / d protein and human CD3e protein to a certain extent. (D) Flow cytometry (FACS) assays were performed to detect the binding of the BCMA-GPRC5D-CD3 triple antibody to cell lines overexpressing the target antigen BCMA and cell lines overexpressing the target antigen GPRC5D, respectively. Flp-inCHO-human BCMA cells and CHOK1-human GPRC5D cells were cultured in T-75 cell culture flasks to the logarithmic growth phase, and the cells were pipetted to a single-cell suspension. After cell counting, the cells were centrifuged, and the cell pellet was resuspended in FACS buffer (PBS + 2% fetal bovine serum) to a concentration of 2 × 10⁻⁶. 6 Cells were added at a rate of 50 μl per well to a 96-well FACS plate. BCMA-GPRC5D-CD3 antibody or control antibody was added, serially diluted 5-fold from 100 nM to 8 spots, 50 μl per well, and incubated at 4°C for 1 hour. After washing three times with PBS buffer, 50 μl of secondary antibody anti-hIgG (Fc) Alexa 647 (Jackson, catalog number: 109-605-098) was added to each well, and the plate was incubated on ice for 1 hour. After washing three times with PBS buffer, 100 μl of the secondary antibody was analyzed using FACS (FACS Canto™, BD). Data analysis was performed using FlowJo software to obtain the mean fluorescence intensity (MFI) of the cells. Further analysis was performed using GraphPad Prism8 software to fit the data and calculate EC50. The results are shown in Tables 6 and 7, and Tables 7 and 8. All triple antibody molecules were respectively... The target antigens BCMA and GPRC5D on overexpressing cells bind well. Table 6: FACS detection of the binding reaction between triple antibody and Flp-inCHO-human BCMA cells Table 7: FACS detection of binding reaction between triple antibody and CHOK1-human GPRC5D cells (E) FACS detection of the endogenous nature of BCMA-GPRC5D-CD3 triple antibody and co-expressed target antigens BCMA and GPRC5D binding of tumor cells The specific method is the same as in Example 4(D). In this experiment, the multiple myeloma cell line NCI-H929 (purchased from ATCC) was selected as the endogenous cell line to detect the binding ability of the three antibodies. NCI-H929 is a cell line that highly expresses both GPRC5D and BCMA. The results are shown in Table 8 and Figure 9. The three antibodies can bind strongly to the endogenous tumor cells NCI-H929 that express antigens BCMA and GPRC5D. Table 8: FACS detection of the binding reaction between the triple antibody and H929 cells (F) FACS detection of the binding of BCMA-GPRC5D-CD3 triple antibody to CD3-expressing endogenous tumor cells and to CD3-expressing activated human PBMCs and monkey PBMCs. The endogenous cells expressing CD3 were Jurkat (purchased from ATCC). The preparation process of human PBMCs and monkey PBMCs expressing CD3 was as follows: PBMCs isolated from human blood were activated and cultured for 3 days using the human T cell activation kit (purchased from Miltenyi Biotec, catalog number: 130-091-441) and the monkey T cell activation kit (purchased from Miltenyi Biotec, catalog number: 130-092-919), respectively. Then, the activated T cells were cultured for 2 weeks in 1640 medium containing 10% (v / v) fetal bovine serum with 100 units of human IL2 protein (purchased from Acro, catalog number: IL2-H4113). The FACS detection method was the same as in Example 4(D). The results are shown in Table 9 and Figure 10, as well as Table 10, Figure 11A, and 11B. The three antibodies (anti-inflammatory, anti-tumor, and anti-inflammatory) molecules all bound to Jurkat cells, human PBMCs, and monkey PBMCs to a certain extent. Table 9. FACS detection of the binding reaction between triple antibody and Jurkat cells Table 10. FACS detection of binding reaction between triple antibodies and human PBMCs and monkey PBMCs Detection of activation activity of (G)BCMA-GPRC5D-CD3 triple antibody against Jurkat-NFAT-NanoLuc cells Preparation of Jurkat-NFAT-NanoLuc cells: Jurkat cells (purchased from ATCC) were seeded at 1E6 cells / well into 6-well culture dishes (Costar: 3516) containing 2 mL of complete culture medium (RPM 1640, 10% premium FBS, penicillin-streptomycin) and cultured at 37°C and 5% CO2 for 24 hours. The next day, 3 μg of pNL [NLucP / NFAT-RE / Hygro] vector was diluted with 3 μl of PLUS reagent (Invitrogen: 11514-015) in 150 μL of Opti-MEM (Gibco: 11058021) and incubated at room temperature for 5 minutes. The pNL[NLucP / NFAT-RE / Hygro] vector contains NFAT, which drives the transcription of the luciferase reporter gene NanoLuc in response to TCR activation. After incubation, the diluted DNA solution was pre-incubated with a 1:1 Lipofectamine LTX solution (Invitrogen: 15338-100) (6 μl Lipofectamine LTX + 144 μl Opti-MEN) at room temperature for 15 minutes to form a DNA-Lipofectamine LTX complex. After incubation, 300 μl of the DNA-Lipofectamine complex was added directly to the cell wells. Jurkat cells were transfected at 37°C and 5% CO2 for 24 hours. After incubation, the cells were washed with 3 mL of PBS and selectively cultured for 2 weeks in 1640 medium (purchased from Gibco, catalog number: 31870082) containing 125 μg / ml hygromycin (Invitrogen: 10687010) and 10% (w / w) fetal bovine serum to obtain the desired Jurkat-NFAT-NanoLuc cell line. The BCMA-GPRC5D-CD3 triple antibody and control antibody were serially diluted 5-fold starting at 100 nM (final concentration) or 3-fold starting at 100 nM (final concentration) in RPMI 1640 medium containing 5% fetal bovine serum. 50 μl of the diluted antibody was added to each well of a 96-well black flat-bottom translucent cell culture plate. H929 cells or K562 negative cells and Jurkat-NFAT-NanoLuc cells were cultured in T-75 cell culture flasks to the logarithmic growth phase, and the cells were pipetted to a single-cell suspension. After cell counting, the cells were centrifuged, and the cell pellet was resuspended in RPMI 1640 medium containing 5% fetal bovine serum at a concentration of 2 × 10⁻⁶ mcg. 6For each milliliter of cells, 25 μl of each of the three cell types was added to each well of the aforementioned 96-well cell culture plate. After mixing, the plates were incubated at 37°C and 5% (v / v) CO2 for 5–6 hours. After incubation, 50 μl of a Nano-Glo luciferase reporter gene assay system (Promega, catalog number: N1110) was added to each well, and the plates were shaken and incubated for 10 minutes. Finally, fluorescence values ​​were read using a multimode plate reader (EnSight, Perkin Elmer). The results are shown in Table 11 and Figures 12A and 12B. The three antibodies effectively activated Jurkat-NFAT-NanoLuc cells mediated by H929 cells expressing the target antigens BCMA and GPRC5D, while they did not activate Jurkat-NFAT-NanoLuc cells or produced very weak nonspecific activation mediated by K562 negative cells that do not express BCMA and GPRC5D. Table 11. Activation activity of triple antibodies on Jurkat-NFAT-NanoLuc cells mediated by H929 and K562 cells, as detected by FACS. Example 5: Detection of the in vitro tumor cell killing activity and cytokine production of BCMA-GPRC5D-CD3 triple antibody-mediated T cells. Three human multiple myeloma cell lines were selected as target cells for the in vitro T-cell killing experiment. NCI-H929 is a cell line that highly expresses both GPRC5D and BCMA; RPMI8226 is a cell line with low GPRC5D expression and moderate BCMA expression; and Molp-8 is a cell line with high GPRC5D expression and low BCMA expression. To facilitate subsequent experimental detection, stable transgenic lines expressing luciferase were constructed for the above three cell lines, hereinafter referred to as H929-Luc, RPMI8226-Luc, and Molp-8-Luc (the construction method is described in Section 3.3 of Example 3). On the first day of the experiment, healthy donor PBMCs (human peripheral mononuclear cells) frozen in liquid nitrogen were thawed, resuspended in RPMI-1640 medium, and cultured overnight in a 5% CO2, 37°C incubator. On the second day, single-cell suspensions were prepared from H929-Luc, RPMI8226-Luc, and Molp-8-Luc cells, respectively. The tumor cell density was adjusted to 1.33 × 10⁻⁶ cells using complete culture medium. 5 The density of PBMCs was adjusted to 1.33 × 10⁻⁶ cells / ml. 6Tumor cells / ml. Take 75 μl of adjusted H929-Luc, RPMI8226-Luc or Molp-8-Luc and 75 μl of PBMCs respectively and mix them evenly (the number of tumor cells and PBMCs per well are 1×10⁻⁶). 4 1×10 5 Cells were added to each well of a 96-well U-bottom plate using a micropipette, with a cell mixture volume of 150 μl / well. Antibodies were serially diluted with complete culture medium (starting concentration 100 nM, 7-fold dilution, 8 gradients), with 100 μl of different concentrations of antibody dilution added to each well, resulting in a final volume of 250 μl per well. After mixing, the cells were incubated in a 5% CO2, 37°C incubator. After 48 hours of culture, 100 μl of the supernatant was used to detect the cytokines IFNγ and TNFα in the cell culture supernatant. The remaining cells were continued to be incubated in a 5% CO2, 37°C incubator. After another 24 hours of culture, cell viability was assessed using the Bright-Glo Luciferase kit (purchased from Promega, catalog number E2620, usage instructions follow). Wells containing only PBMCs were defined as positive controls with 100% cell killing effect, and wells with an antibody concentration of 0 were defined as blank negative controls. Cell viability was calculated as (sample well reading - positive well reading) / (blank well reading - positive well reading) × 100%. The results were calculated and plotted using GraphPad Prism 9.0 software. Some triple antibody in vitro killing assays use Molp-8 cells that do not express luciferase as target cells. The co-incubation method for PBMCs, tumor cells, and antibodies is the same as above. After 24 hours of co-culture, the cell culture plate is centrifuged at 300g, and 100μl of supernatant is transferred to detect the cytokines released from the cell culture supernatant. The remaining 100μl of supernatant is transferred to a flat-bottomed 96-well plate, and the LDH release value per well is detected using an LDH detection kit (purchased from DO JINDO, catalog number CK12; usage instructions follow the product manual). The percentage of target cell killing is calculated according to the kit instructions. There are two methods for detecting cytokines. One method is to use the Cytometric Bead Array human Th1 / Th2 / Th17 cytokine kit (purchased from BD, catalog number 560484, refer to the kit instructions for usage) to detect the amount of cytokines released in the culture medium using flow cytometry. The other method is to use ELISA to detect the cytokines IFNγ and TNFα in the cell culture supernatant (kits purchased from BD, catalog numbers 555142 and 55521 respectively, refer to the kit instructions for usage). The results of T cell killing of tumor cells mediated by triple antibodies are shown in Figures 13, 15, 17, and Table 12. The results indicate that for different tumor cell types, triple antibodies produce better or comparable killing effects compared to the positive control. The killing effect of T cells against Molp-8 mediated by triple antibodies Tri-38 and Tri-41 is weaker than that of the positive control. The results of the secretion of cytokines TNFα and IFNγ during in vitro tumor cell killing are shown in Figures 14A-14B, 16A-16B, and 18A-18B. The results indicate that when triple antibodies Tri-29, Tri-35, Tri-46, and Tri-55 mediate T cell killing of tumor cells, a certain amount of cytokines can be secreted. Table 12. Killing effects of triple antibody-mediated T cells on H929-Luc, RPMI8226-Luc, and Molp-8-Luc or Molp-8 cells, respectively.

Claims

The claims for protection are as follows: A multi-specific binding molecule, wherein, The multispecific binding molecule comprises: (a) The first antigen-binding molecule (ABM1) that specifically binds to the first tumor-associated antigen (TAA1); (b) A second antigen-binding molecule (ABM2) that specifically binds to a second tumor-associated antigen (TAA2); and (c) A third antigen-binding molecule (ABM3) that specifically binds to antigens expressed on human immune cells, optionally, wherein TAA1 and / or TAA2 are expressed on cancerous B cells. According to claim 1, the multispecific binding molecule, wherein, The cancerous B cells are of plasma cell or B cell origin, and optionally, the immune cells are selected from T cells, NK cells, or macrophages. The multispecific binding molecule according to claim 1 or 2, wherein, The TAA1 and TAA2 are each independently selected from BCMA and GPRC5D, and / or the antigen expressed on the human immune cells is CD3. The multispecific binding molecule according to any one of claims 1 to 3, wherein, The ABM1, ABM2, and ABM3 are each independently selected from antibodies, antibody fragments, F(ab')2, Fab', Fab, Fv, scFv, nanobodies, or VHH. The multispecific binding molecule according to any one of claims 1 to 4, wherein, The multispecific binding molecule comprises four polypeptide chains: a first heavy chain, a first light chain, a second heavy chain, and a second light chain. According to claim 5, the multispecific binding molecule, wherein, (1) The first heavy chain and the second heavy chain are identical and have the following structure: (a)VH a -CH1-L1-VHH b -Fc-L2-VH CD3 -L3-VL CD3 ; (b)VH a -CH1-L1-VH CD3 -L2-VL CD3 -Fc-L3-VHH b ; (c)VHH b -L1-VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc; (d)VH a -CH1-L1-VHH b -L2-VH CD3 -L3-VL CD3 -Fc; or (e)VH a -CH1-L1-VH b -L2-VL b -Fc-L3-VH CD3 -L4-VL CD3 ; And / or, (2) the first light chain and the second light chain are identical and have the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively. b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA, respectively. CD3 and VL CD3 These are the heavy chain variable region and light chain variable region that specifically bind to CD3, respectively. Fc is the Fc region of any antibody, and L1, L2, L3, and L4 are the same or different linkers, respectively. According to claim 5, the multispecific molecule, wherein, (1) The first heavy chain and the second heavy chain are different and have the following structure: (a) First heavy chain VH a -CH1-L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (b) First heavy chain VHH b -L1-VH a -CH1-Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (c) First heavy chain VH a -CH1-L1-VH b -L2-VL b -Fc1 and second heavy chain VH a -CH1-L3-VH b -L4-VL b -Fc2-L5-VH CD3 -L6-VL CD3 ; (d) First heavy chain VH a -CH1-L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VHH b -Fc2-L3-VH CD3 -L4-VL CD3 ;or (e) First heavy chain VHH b -L1-VH a -CH1-Fc1 and second heavy chain VHH b -L2-VH a -CH1-Fc2-L3-VH CD3 -L4-VL CD3 ; Optionally, the carboxyl terminus of the first heavy chain in (a) to (e) above may also have a Flag tag and / or the carboxyl terminus of the second heavy chain. It also has a His tag; And, (2) the first light chain and the second light chain are identical and have the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively. b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA, respectively. CD3 and VL CD3 These are the heavy chain variable region and the light chain variable region that specifically bind to CD3, respectively, and L1, L2, L3, L4, L5 and L6 are the same or different linkers. The multispecific binding molecule according to any one of claims 1 to 4, wherein, The multispecific binding molecule comprises three polypeptide chains: a first heavy chain, a second heavy chain, and a light chain, wherein (1) the first heavy chain and the second heavy chain have the following structures: (a) First heavy chain VHH b -L1-VHH b -Fc1 and second heavy chain VH a -CH1-L2-VH CD3 -L3-VL CD3 -Fc2; (b) First heavy chain VHH b -L1-VH CD3 -L2-VL CD3 -Fc1 and second heavy chain VH a -CH1-Fc2; or (c) First heavy chain VHH b -Fc1 and second heavy chain VH a -CH1-L1-VH CD3 -L2-VL CD3 -Fc2; Optionally, the carboxyl terminus of the first heavy chain in (a) to (c) above is further marked with a Flag tag and / or the carboxyl terminus of the second heavy chain is further marked with a His tag; And, (2) the light chain has the following structure: VL a -CL; VH a and VL a These are the heavy chain variable region and light chain variable region that specifically bind to GPRC5D, respectively. b VH is a nanobody that specifically binds to BCMA. b and VL b These are the heavy chain variable region and light chain variable region that specifically bind to BCMA, respectively. CD3 and VL CD3 These are the heavy chain variable region and light chain variable region that specifically bind to CD3, respectively. Fc1 and Fc2 are each the Fc region of any antibody, and L1, L2 and L3 are the same or different linkers, respectively. The multispecific binding molecule according to any one of claims 6 to 8, wherein, The Fc1 and Fc2 of the multispecific binding molecule are different, with Fc1 being knock-Fc and Fc2 being hole-Fc; or, Fc1 being hole-Fc and Fc2 being knock-Fc; preferably, the knock-Fc contains the T366W mutation, and / or the hole-Fc contains the T366S, L368A and / or Y407V mutation; preferably, Fc1 and Fc2 are the Fc regions of IgG1 or IgG4. The multispecific binding molecule according to any one of claims 6 to 8, wherein, The Fc1 and Fc2 of the multispecific molecules are the same or different and have amino acid mutations that alter effector function; preferably, the mutations that alter effector function include S228P, F234A, L234A and / or L235A; more preferably, the mutations that alter effector function include: (a) L234A and / or L235A mutations in IgG1 Fc; and / or (b) S228P, F234A and / or L235A mutations in IgG4 Fc. The multispecific binding molecule according to any one of claims 1 to 3, wherein, The ABM3 is a CD3 antibody or its antigen-binding fragment, comprising a light chain variable region and a heavy chain variable region: (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 54–56, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 54–56; and (b) the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 51–53, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 51–53. The multispecific binding molecule of claim 11, wherein the light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 10 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology to SEQ ID NO: 10, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 9 or The amino acid sequence has 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO:

9. According to claim 12, the multispecific binding molecule, ABM3 is scFv. The multispecific binding molecule according to any one of claims 1 to 3, wherein the TAA1 and the TAA2 are selected from the group consisting of: (a) TAA1 being BCMA and TAA2 being GPRC5D; or (b) TAA1 being GPRC5D and TAA2 being BCMA. According to claim 14, the multispecific binding molecule, wherein, The antigen-binding molecule that specifically binds GPRC5D in the multispecific binding molecule comprises a light chain variable region and a heavy chain variable region of the antibody: (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 42–44, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 42–44, and the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 39–41, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 39–41; or (b) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 48–50, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 39–41; Compared to SEQ ID NO. 48-50, the heavy chain variable region has 1, 2, 3 or more amino acid insertions, deletions and / or substitutions, and the heavy chain variable region includes HCDR1, HCDR2 and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 45-47, or have 1, 2, 3 or more amino acid insertions, deletions and / or substitutions compared to SEQ ID NO. 45-47. The multispecific binding molecule according to claim 15, wherein: (a) The light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 6 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 6, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 5 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 5; or (b) The light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 8 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 8, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 7 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO:

7. According to claim 16, the multispecific binding molecule, wherein the antigen-binding molecule that specifically binds GPRC5D in the multispecific binding molecule is Fab. According to claim 14, the multispecific binding molecule, wherein, The antigen-binding molecule that specifically binds to BCMA in the multispecific binding molecule includes a heavy chain variable region of the antibody, wherein the heavy chain variable region includes HCDR1, HCDR2, and HCDR3, and the amino acid sequences of HCDR1, HCDR2, and HCDR3 are as shown in SEQ ID NO. 27–29 or as shown in SEQ ID NO. 30–32, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 27–29 or SEQ ID NO. 30–32. The multispecific binding molecule according to claim 18, wherein, The heavy chain variable region contains an amino acid sequence as shown in SEQ ID NO.1 or SEQ ID NO.2, or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO.1 or SEQ ID NO.

2. The multispecific binding molecule according to claim 19, wherein the antigen-binding molecule that specifically binds to BCMA is VHH. The multispecific binding molecule according to claim 14, wherein the multispecific binding molecule contains specific binding molecules. The antigen-binding molecule of BCMA comprises a light chain variable region and a heavy chain variable region of the antibody: (a) the light chain variable region comprises LCDR1, LCDR2, and LCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 36–38, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 36–38; and (b) the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3, the amino acid sequences of which are as shown in SEQ ID NO. 33–35, or have 1, 2, 3, or more amino acid insertions, deletions, and / or substitutions compared to SEQ ID NO. 33–35. The multispecific binding molecule of claim 21, wherein the light chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 4 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO: 4, and / or the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 3 or an amino acid sequence having 90%, 95%, 96%, 97%, 98%, 99% or higher homology with SEQ ID NO:

3. The multispecific binding molecule according to claim 22, wherein the antigen-binding molecule that specifically binds to BCMA is scFv. The multispecific binding molecule according to any one of claims 1 to 23, wherein the ABM1, ABM2 and / or ABM3 are chimeric, humanized or fully human. The multispecific binding molecule according to claim 24 further comprises an Fc fragment of human IgG1, IgG2, IgG3 or IgG4; preferably, it comprises an Fc fragment of human IgG1 or human IgG4. According to claim 25, the multispecific binding molecule, wherein, The Fc fragment contains the same or different Fc1 and Fc2. According to claim 25, the multispecific binding molecule, wherein, The Fc fragment has amino acid mutations that alter the effector function; preferably, the amino acid mutations that alter the effector function include S228P, F234A, L234A and / or L235A mutations; more preferably, the amino acid mutations that alter the effector function include: (a) L234A and / or L235A mutations in IgG1 Fc; and / or (b) S228P, F234A and / or L235A mutations in IgG4 Fc. According to claim 27, the multispecific binding molecule, wherein, The different Fc1 and Fc2 have amino acid mutations that promote heterodimerization; preferably, the amino acid mutations that promote heterodimerization include a knob-into-hole (KIH) mutation; more preferably, the knob-into-hole mutation includes: (a) either Fc region contains a T366W mutation; and (b) the other Fc region contains a T366S, L368A and / or Y407V mutation. The multispecific binding molecule according to claim 28, wherein, Fc1 further includes a His tag, and Fc2 further includes a Flag tag; or Fc1 further includes a Flag tag, and Fc2 further includes a His tag. The multispecific binding molecule according to any one of claims 11 to 29, wherein, The multispecific binding molecule is trivalent, tetravalent, pentavalent, or hexavalent. An isolated nucleic acid, wherein, The nucleic acid encodes the multispecific binding molecule as described in any one of claims 1 to 30. A recombinant vector, wherein, The recombinant vector comprises the isolated nucleic acid as described in claim 31. A host cell, in which, The host cell comprises the isolated nucleic acid of claim 31 or the recombinant vector of claim 32; preferably, the host cell is a eukaryotic cell or a prokaryotic cell; more preferably, the host cell is derived from mammalian cells, yeast cells, insect cells, Escherichia coli and / or Bacillus subtilis; more preferably, the host cell is an Expi293 cell. A method for preparing multispecific binding molecules, wherein, The method includes cultivating the right under appropriate conditions. Determine the host cell described in 33, and isolate the multispecific binding molecule. A pharmaceutical composition wherein, The pharmaceutical composition comprises the multispecific binding molecule of any one of claims 1 to 30, the isolated nucleic acid of claim 31, the recombinant vector of claim 32, the host cell of claim 33, or the product prepared by the method of claim 34, and one or more pharmaceutically acceptable carriers, diluents, buffers, or excipients; preferably, the pharmaceutical composition further comprises an additional cancer therapeutic agent; preferably, the pharmaceutical composition is for the treatment and / or prevention of cancer; preferably, the cancer is B-cell lymphoma; more preferably, the B-cell lymphoma is multiple myeloma. A method for activating T cells, wherein, The method comprises administering to a subject a therapeutically effective amount of the multispecific binding molecule of any one of claims 1 to 30 or the pharmaceutical composition of claim 35. The use of the multispecific binding molecule of any one of claims 1 to 30, the isolated nucleic acid of claim 31, the host cell of claim 33, the product prepared by the method of claim 34, or the pharmaceutical composition of claim 35 in the preparation of a medicament for treating and / or preventing cancer, or for the treatment and / or prevention of cancer. Preferably, the cancer is B-cell lymphoma; more preferably, the B-cell lymphoma is multiple myeloma. A method for treating and / or preventing cancer, wherein, The method includes administering to a subject a therapeutically effective amount of any one of claims 1 to 30, the isolated nucleic acid of claim 31, the host cell of claim 33, the product prepared by the method of claim 34, or the pharmaceutical composition of claim 35; Preferably, the cancer is B-cell lymphoma; more preferably, the B-cell lymphoma is multiple myeloma.

Images