N-terminal scfv multispecific binding molecules

Abstract

To provide multispecific binding molecules (MBMs) for new therapeutic strategies.SOLUTION: Provided are MBMs comprising an N-terminal scFv, a first Fab, and a second Fab, MBM conjugates comprising an MBM and a cytotoxic or cytostatic agent, pharmaceutical compositions comprising MBMs and MBM conjugates, methods of using the MBMs, MBM conjugates, and pharmaceutical compositions to treat cancer, nucleic acids encoding the MBMs, cells engineered to express the MBMs, and methods of producing the MBMs.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] 1. CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority to U.S. Provisional Patent Application No. 62 / 930,916, filed November 5, 2019, the contents of which are incorporated herein by reference in their entirety.

[0002] 2. Sequence Listing This application contains a Sequence Listing which has been submitted electronically in ASCII format and is hereby incorporated by reference in its entirety. [Background technology]

[0003] 3.Background technology Most naturally occurring antibody molecules generally contain two so-called light chain polypeptides (light chains) and two so-called heavy chain polypeptides (heavy chains). Each of the heavy and light chain polypeptides contains a variable domain (variable region) (generally the amino terminal portion of the polypeptide chain) that contains a binding region capable of interacting with an antigen. Each of the heavy and light chain polypeptides contains a constant region (generally the carboxyl terminal portion).

[0004] Recombinant monoclonal antibodies, which are produced by a single clone of a cell or cell line, have emerged over the past two decades as a highly successful class of biological drugs for the treatment of a variety of different diseases. Antibody-based therapeutics have been used successfully to treat a variety of diseases, including cancer and autoimmune / inflammatory disorders.

[0005] Due to the biological complexity of some diseases, antibodies targeting two or more antigens or epitopes may be more effective than a single antibody in treating a particular condition. See, for example, Non-Patent Document 1 and Non-Patent Document 2. These antibodies promise better therapeutic control. For example, particularly in the case of antibody-based immunotherapies, there is a need to improve target specificity to reduce the off-target effects associated with many antibody therapies. In addition, multispecific antibodies offer new therapeutic strategies, such as synergistic targeting of multiple cellular receptors, particularly in the context of immunotherapy. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Lindzen et al.,2010,Proc.Natl.Acad.Sci.107(28):12550-12563 [Non-patent document 2] Nagorsen and Baeuerle,2011,Exp.Cell Res.317(9):1255-60 Summary of the Invention

[0007] 4. Overview The present disclosure provides multispecific binding molecules ("MBMs") that contain at least three antigen binding sites ("ABS"), two of which are Fabs and the third of which is an scFv attached to one of the Fabs at the N-terminus of the VH domain. The MBMs of the present disclosure contain an Fc domain made up of two heavy chain Fc domains associated with each other. Each polypeptide chain, including the Fc domain and any associated polypeptide chains, is referred to herein as a "half antibody."

[0008] An exemplary MBM of the present disclosure is shown in FIG. 1, with variations shown in FIGS. A typical MBM of the present disclosure comprises two half antibodies associated through their Fc domains. The half antibodies shown on the left side of Figures 1-3 are, from N-terminus to C-terminus: an scFv consisting of a VH (1) and a VL (3) (in any order) connected by an optional linker (2); an optional linker (4), The first Fab ("Fab1"), consisting of: VH (5) and associated with a constant domain (6), which in the example of Figure 1 is a CH1 domain, but which may be a different type of constant domain such as CL or CH3, which promotes Fab heterodimerization, as shown in Figures 2A and 2C. a constant domain (11), which in the example of Figure 1 is a VL (10) and a CL, but which may be a different type of constant domain, such as a CH1 or CH3, as shown in Figures 2A and 3C, Optional hinge domain (7), As shown in Figure 3, it contains an Fc domain composed of a CH2 domain (8) and a CH3 domain (9), which may contain one or more mutations (such as star-shaped or knob-in-hole mutations) that promote heterodimerization.

[0009] The left half antibody is composed of two polypeptide chains, the first containing the VL (10) and constant domains (11), and the second containing the remaining domains organized as shown in Figures 1-3.

[0010] The half antibody shown on the right side of Figure 1 is, from N-terminus to C-terminus, A second Fab ("Fab2") consisting of: VH (12) and associated constant domain (13), which in the example of Figure 1 is a CH1 domain, but which may be a different type of constant domain such as CL or CH3, which promotes Fab heterodimerization, as shown in Figures 2B and 2D. a constant domain (18), which in the example of Figure 1 is a VL (17) and a CL, but which may be a different type of constant domain, such as a CH1 or CH3, as shown in Figures 2B and 2D, Optional hinge domain (14), As shown in Figure 3, it contains an Fc domain composed of a CH2 domain (15) and a CH3 domain (16), which may contain one or more mutations (such as star mutations or knob-in-hole mutations) that facilitate the purification and / or assembly of the heterodimer.

[0011] The right half antibody is composed of two polypeptide chains, the first containing the VL (17) and constant domains (18) and the second containing the remaining domains organized as shown in Figures 1-3.

[0012] The complete MBM is formed by the association of two half antibodies through the two Fc domains that form the Fc region, resulting in an MBM with an scFv having a first antigen-binding site ("ABS1"), a Fab1 having a second antigen-binding site ("ABS2"), and a Fab3 having a third antigen-binding site ("ABS3").

[0013] The variations of the MBMs of the present disclosure shown in Figures 1-3 are not intended to be limiting, and the MBMs of the present disclosure can include any combination of the modifications shown in Figures 1-3 and in Section 6.2 below, among others. Additionally, reference to a first or second polypeptide chain or left or right half antibody is for convenience only and is not intended to convey that the polypeptide chains or half antibodies are produced or assembled in any particular order.

[0014] In the present disclosure, ABS1, ABS2, and ABS3 of the MBM each bind to a target molecule, e.g., a cell surface-expressed antigen. Preferably, the scFv, Fab1, and Fab2 of the MBM are selected so that ABS1, ABS2, and ABS3 can each simultaneously bind to their respective targets. In some embodiments, ABS1, ABS2, and ABS3 of the MBM can each bind to a different target molecule, or alternatively, two of ABS1, ABS2, and ABS3 can bind to different regions on the same target molecule.

[0015] Advantageously, MBMs that bind to at least two or more different target molecules can be used, for example, to preferentially target a particular tissue type in which such two or more target molecules are expressed.

[0016] scFvs that can be used in the MBMs of the present disclosure are described in Section 6.2.1 below and in specific embodiments 1-12, 19-20, 31-35, and 55-60. Fabs that can be used in the MBMs of the present disclosure are described in Section 6.2.2 below and in specific embodiments 1-12, 19-20, 22-23, 28-30, and 36-54. The scFv can be connected to Fab1, for example, by a peptide linker. Linkers that can be used to connect the scFv to Fab1 are described in Section 6.2.3 below and in specific embodiments 13-18. Fc domains that can be used in the MBMs of the present disclosure are described in Section 6.2.4 below and in specific embodiments 24-26.

[0017] The present disclosure further provides drug conjugates (herein referred to for convenience as "antibody-drug conjugates" or "ADCs") comprising the MBMs of the present disclosure. Exemplary features of ADCs are described in Section 6.3 below and in specific embodiment 61.

[0018] The present disclosure further provides nucleic acids encoding the MBM of the present disclosure. The nucleic acid encoding the MBM can be a single nucleic acid (e.g., a vector encoding all polypeptide chains of the MBM) or multiple nucleic acids (e.g., two or more vectors encoding different polypeptide chains of the MBM). The present disclosure further provides host cells and cell lines engineered to express the nucleic acids and MBM of the present disclosure. The present disclosure further provides methods of producing the MBM of the present disclosure. Exemplary nucleic acids, host cells, cell lines, and methods of producing the MBM of the present disclosure are described in Section 6.5 below and in specific embodiments 78-83.

[0019] The present disclosure further provides pharmaceutical compositions comprising the MBM and ADC of the present disclosure. Exemplary pharmaceutical compositions are described in Section 6.6 below and in specific embodiment 62. Further provided herein are methods of using the MBMs, ADCs, and pharmaceutical compositions of the present disclosure, for example, to treat cancer. Exemplary methods are described in Section 6.7 below and in specific embodiments 63-77. [Brief explanation of the drawings]

[0020] 5. Brief description of the drawings [Figure 1] 1 is a schematic diagram of an exemplary MBM of the present disclosure. [Figure 2A] Schematic of an exemplary MBM of the present disclosure in which the Fab contains modifications that promote proper VH and VL pairing. Additional strategies are described in Section 6.2.2 below. [Figure 2B] Same as above. [Figure 2C] Same as above. [Figure 2D] Same as above. [Figure 2E] Same as above. [Figure 3-1] Figures 3A-3D: Schematic diagrams of exemplary MBMs of the present disclosure in which the Fc domain contains modifications that facilitate heterodimerization or purification of properly paired heterodimers. Figures 3A and 3B show the CH3 with a star mutation (CH3 with H435R Y436F modifications) as CH3(*), while Figures 3C and 3D show knob-and-hole mutations with (K) and (H) designations, respectively. Elsewhere in the figures, star mutations are indicated with an asterisk (*) and knob-in-hole mutations are indicated with a triangle (e.g., [number 1]).

number

[0021] 6. MODE FOR CARRYING OUT THE INVENTION 6.1.Definition As used herein, the following terms are intended to have the following meanings:

[0022] Antigen-binding site or ABS: As used herein, the term "antigen-binding site" or "ABS" refers to a portion of an MBM capable of specific, non-covalent, and reversible binding to a target molecule. The MBM of the present disclosure comprises a first ABS ("ABS1") that is part of an scFv, a second ABS ("ABS2") that is part of an Fab, and a third ABS ("ABS3") that is part of an Fab.

[0023] Association: The term "association" in the context of an MBM refers to a functional relationship between two or more polypeptide chains. In particular, the term "association" means that two or more polypeptides are associated with one another, e.g., noncovalently through molecular interactions or covalently through one or more disulfide or chemical crosslinks, to produce a functional MBM in which ABS1, ABS2, and ABS3 can bind to their respective targets. Examples of associations that may be present in the MBMs of the present disclosure include (but are not limited to) associations between homodimeric or heterodimeric Fc domains in the Fc region, associations between VH and VL regions in a Fab or scFv, associations between CH1 and CL in a Fab, and associations between CH3 and CH3 in domain-substituted Fabs.

[0024] Complementarity-determining region or CDR: As used herein, the term "complementarity-determining region" or "CDR" refers to the sequence of amino acids in an antibody variable region that confers antigen specificity and binding affinity. Generally, each heavy chain variable region has three CDRs (CDR-H1, CDR-H2, HCDR-H3), and each light chain variable region has three CDRs (CDR1-L1, CDR-L2, CDR-L3). Exemplary rules that can be used to identify the boundaries of CDRs include, for example, the Kabat definition, the Chothia definition, the ABS definition, and the IMGT definition. See, e.g., Kabat, 1991, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (Kabat numbering scheme); Al-Lazikani et al., 1997, J. Mol. Biol. 273:927-948 (Chothia numbering scheme); Martin et al., 1989, Proc. Natl. Acad. Sci. USA 86:9268-9272 (ABS numbering scheme); and Lefranc et al., 2003, Dev. Comp. Immunol. 27:55-77 (IMGT numbering scheme). Public databases are also available for identifying CDR sequences within antibodies.

[0025] Origin: As used herein, the term "origin" refers to the relationship between a first molecule and a second molecule. It generally refers to the structural similarity between the first and second molecules and does not imply or imply any process or source restriction on the first molecule being derived from the second molecule.

[0026] EC50: The term "EC50" refers to the half-maximal effective concentration of an antibody or MBM that induces a response midway between baseline and maximum after a specific exposure time. EC50 essentially represents the concentration of an antibody or MBM at which 50% of its maximum effect is observed. In certain embodiments, the EC50 value is equal to the concentration of an antibody or MBM that confers half-maximal binding to cells expressing a target molecule that can be specifically bound by the antibody or MBM, as determined, for example, by a FACS binding assay. Thus, reduced or weaker binding is observed at an increased EC50, or half-maximal effective concentration value. The EC50 values ​​of the MBMs of the present disclosure are, in some embodiments, about 10 -5 M or less (e.g., 10 -5 Under M, 10 -6 Under M, 10 -7 Under M, 10 -8 Less than M or 10 -9 The activity can be characterized by an EC50 value (less than 1 M).

[0027] Epitope: An epitope, or antigenic determinant, is the portion of an antigen (e.g., a target molecule) that is recognized by an antibody or other antigen-binding moiety described herein. Epitopes can be linear or conformational.

[0028] Fab: The term "Fab" in the context of the MBMs of the present disclosure refers to a pair of polypeptide chains, the first of which comprises an antibody variable heavy (VH) domain (referred to herein as C1) N-terminal to the first constant domain, and the second of which comprises an antibody variable light (VL) domain (referred to herein as C2) N-terminal to a second constant domain capable of pairing with the first constant domain. In a native antibody, the VH is N-terminal to the first constant domain (CH1) of the heavy chain, and the VL is N-terminal to the constant domain of the light chain (CL). The Fabs of the present disclosure, particularly when the MBMs of the present disclosure comprise non-identical Fabs, may be oriented according to their native orientation or may contain domain substitutions or swaps that promote correct VH and VL pairing. For example, the pair of CH1 and CL domains in a Fab may be replaced with a pair of CH3 domains to promote correct modified Fab chain pairing in heterodimeric MBMs. It is also possible to reverse the CH1 and CL, with CH1 attached to VL and CL attached to VH, a configuration commonly known as a crossmab. Alternatively, or in addition to the use of substituted or exchanged constant domains, correct chain pairing can be achieved by using a universal light chain that can pair with both variable regions of the heterodimeric MBMs of the present disclosure.

[0029] Fc domain and Fc region: The term "Fc domain" refers to the portion of a heavy chain that pairs with the corresponding portion of another heavy chain. The term "Fc region" refers to the region of an antibody-based binding molecule formed by the association of two heavy chain Fc domains. The two Fc domains within an Fc region can be the same or different from one another. In natural antibodies, the Fc domains are typically identical, but for purposes of producing the MBMs of the present disclosure, one or both Fc domains can be advantageously modified to allow heterodimerization.

[0030] Half antibody: The term "half antibody" refers to a molecule that contains at least one ABS or ABS chain (e.g., one chain of a Fab) and can associate with another molecule that contains an ABS or ABS chain, for example, through a disulfide bridge or molecular interaction (e.g., knob-in-hole interactions between Fc heterodimers). A half antibody can consist of one polypeptide chain or two or more polypeptide chains (e.g., two polypeptide chains of a Fab). In a preferred embodiment, a half antibody contains an Fc domain.

[0031] Host cell: As used herein, the term "host cell" refers to a cell into which a nucleic acid of the present disclosure has been introduced. The terms "host cell" and "recombinant host cell" are used interchangeably herein. It is understood that such terms refer to the particular subject cell and the progeny or potential progeny of such a cell. Because certain modifications may occur in subsequent generations due to either mutation or environmental influences, such progeny may not, in fact, be identical to the parent cell but still fall within the scope of the term as used herein. Typical host cells are eukaryotic host cells, such as mammalian host cells. Exemplary eukaryotic host cells include yeast and mammalian cells, e.g., vertebrate cells such as mouse, rat, monkey, or human cell lines, e.g., HKB11 cells, PER.C6 cells, HEK cells, or CHO cells.

[0032] Multispecific binding molecule or MBM: As used herein, the term "multispecific binding molecule" or "MBM" refers to a molecule (e.g., an assembly of multiple polypeptide chains) that comprises two half antibodies, specifically binds to at least two different epitopes (and in some cases, three or more different epitopes), and comprises ABS1, ABS2, and ABS3.

[0033] Operably linked: The term "operably linked" refers to a functional relationship between two or more regions of a polypeptide chain, where the two or more regions are joined to produce a functional polypeptide.

[0034] Parent antibody: The term "parent antibody" refers to an antibody (as that term is defined herein) from which an MBM of the present disclosure is derived, e.g., a parent monospecific or bispecific antibody lacking the N-terminal scFv domain present in an MBM of the present disclosure. An MBM of the present disclosure may share binding sequences, e.g., CDR, VH, and / or VL sequences, with the "parent" antibody, e.g., in one or more of its ABSs, but need not be prepared by modifying the parent antibody or its coding sequence.

[0035] Single-chain Fv or scFv: As used herein, the term "single-chain Fv" or "scFv" refers to a polypeptide chain comprising the VH and VL domains of an antibody, wherein these domains are present in a single polypeptide chain.

[0036] Specific (or selective) binding: As used herein, the term "specifically (or selectively) binding" means that the MBM or its antigen binding site ("ABS") forms a complex with the target molecule that is relatively stable under physiological conditions. Specific binding is defined as a binding activity of approximately 5x10 -2 M or less (e.g., 5x10 -2 Under M, 10 -2 Under M, 5x10 -3 Under M, 10 -3 Under M, 5x10 -4 Under M, 10 -4 Under M, 5x10 -5 Under M, 10 -5 Under M, 5x10 -6 Under M, 10 -6 Under M, 5x10 -7 Under M, 10 -7 Under M, 5x10 -8 Under M, 10 -8 Under M, 5x10 -9 Under M, 10 -9 Less than M or 10 -10The binding affinity of an antibody or antibody fragment, e.g., an MBM or ABS, to a target molecule can be characterized by a KD of less than M. Methods for determining the binding affinity of an antibody or antibody fragment, e.g., an MBM or ABS, to a target molecule are well known in the art and include, for example, equilibrium dialysis, surface plasmon resonance (e.g., a Biacore assay), fluorescence-activated cell sorting (FACS) binding assays, and the like. An MBM or ABS thereof that specifically binds to a target molecule from one species may, however, have cross-reactivity to target molecules from one or more other species.

[0037] Target molecule: As used herein, the term "target molecule" refers to any biomolecule (e.g., a protein, carbohydrate, lipid, or combination thereof) that can be specifically bound by the antigen-binding site of an MBM.

[0038] Tissue: As used herein, the term "tissue" refers to a collection or aggregate of cells of a particular type. Tissues may be derived from the mesoderm (e.g., bone, muscle, connective tissue, kidney, and related structures), endoderm (e.g., lungs, other respiratory structures, and digestive organs), or ectoderm (e.g., nervous system, sensory organs, skin, and related structures). Examples of tissues useful for targeting with the MBMs of the present disclosure include, but are not limited to, connective tissue (including fibrous connective tissue, skeletal connective tissue, and fluid connective tissue, blood, bone, tendons, ligaments, adipose, and loose tissue, etc.), muscle tissue (including visceral or smooth muscle, skeletal muscle, and cardiac muscle, etc.), nervous or neural tissue (including from the central nervous system (brain, spinal cord, etc.)), peripheral nervous system (cranial nerves, spinal nerves, motor neurons, etc.), epithelial tissue (including skin, respiratory tract, reproductive tract, digestive tract, simple squamous epithelium, stratified squamous epithelium, simple cuboidal epithelium, transitional epithelium, pseudostratified columnar epithelium, columnar epithelium, glandular epithelium, ciliated columnar epithelium, etc.), endothelial tissue (blood vessels, lymphatic vessels, etc.), and any cell or any component thereof (e.g., extracellular matrix). The tissues targeted by the MBMs of the present disclosure can be (a) solid or liquid tissues (e.g., blood or individual blood cell types), (b) normal or diseased tissues (e.g., cancer cells), and / or (c) located in the same organ (e.g., kidney or liver) or other body structure (e.g., tumor) or in different organs or other body structures.

[0039] Tissue expression profile: The term "tissue expression profile" used herein in relation to a target molecule refers to the expression pattern of the target molecule in the human body, as defined, for example, by the Human Protein Atlas (HPA) and / or the Genotype-Tissue Expression (GTEx) project. The tissue expression profile can be a protein expression profile and / or an mRNA expression profile.

[0040] Bivalent: As used herein, the term "bivalent" refers to an MBM having three antigen-binding sites. In some embodiments, two of the antigen-binding sites bind to the same epitope of the same target. In other embodiments, two of the antigen-binding sites specifically bind to different epitopes of the same target molecule.

[0041] Universal light chain: As used herein in the context of MBM, the term "universal light chain" refers to a light chain polypeptide that can pair with the heavy chain region of Fab1 to form Fab1 and with the heavy chain region of Fab2 to form Fab2. A universal light chain is also known as a "common light chain."

[0042] VH: The term "VH" refers to the variable region of an immunoglobulin heavy chain of an antibody, including the heavy chain of an scFv or Fab. VL: The term "VL" refers to the variable region of an immunoglobulin light chain, including the light chain of an scFv or Fab.

[0043] 6.2. Multispecific binding molecules (MBM) The MBMs of the present disclosure comprise two half antibodies, one of which comprises at least one antigen-binding site (e.g., a Fab) and the other of which comprises at least two antigen-binding sites (e.g., a Fab having an scFv operably linked to the N-terminus of its VH).

[0044] The MBMs of the present disclosure specifically bind to at least two different epitopes (and in some cases three or more different epitopes). The at least two different epitopes can be on the same target molecule or on different target molecules. Generally, the MBMs of the present disclosure specifically bind to two or more different target molecules (sometimes referred to herein as "antigens").

[0045] 6.2.1.scFv Single-chain Fv or "scFv" antibody fragments comprise the VH and VL domains of an antibody in a single polypeptide chain, can be expressed as single-chain polypeptides, and retain the specificity of the intact antibody from which they are derived. Generally, scFv polypeptides further comprise a polypeptide linker between the VH and VL domains that enables the scFv to form the desired structure for target binding. Examples of linkers suitable for connecting the VH and VL chains of scFvs are the linkers identified in Section 6.2.3.

[0046] Unless otherwise specified, as used herein, an scFv can have the VL and VH variable regions in either order, e.g., with respect to the N- and C-termini of the polypeptide, an scFv can comprise VL-linker-VH or VH-linker-VL.

[0047] The scFv can comprise VH and VL sequences from any suitable species, such as murine, human, or humanized VH and VL sequences. To generate scFv-encoding nucleic acids, the VH- and VL-encoding DNA fragments are operably linked to another fragment encoding a linker, e.g., any of the linkers described in Section 6.2.3 (typically repeats of the amino acids glycine and serine, e.g., a sequence containing the amino acid sequence (Gly4 to Ser)3 (SEQ ID NO:1)), such that the VH and VL sequences can be expressed as a contiguous single-chain protein, the VL and VH regions joined by the flexible linker (see, e.g., Bird et al., 1988, Science 242:423-426; Huston et al., 1988, Proc. Natl. Acad. Sci. USA 85:5879-5883; McCafferty et al., 1990, Nature 348:552-554).

[0048] 6.2.2.Fab1 and Fab2 The MBMs of the present disclosure comprise at least one Fab domain in each half antibody. Fab domains are traditionally produced by proteolytic cleavage of immunoglobulin molecules using enzymes such as papain. In the MBMs of the present disclosure, the Fab domains are recombinantly expressed as part of a larger molecule.

[0049] The Fab domain can comprise constant domain and variable region sequences from any suitable species, and thus can be murine, chimeric, human, or humanized. The Fab domain typically comprises a CH1 domain attached to a VH domain, which is paired with a CL domain attached to a VL domain. In wild-type immunoglobulins, the VH domain is paired with the VL domain to form the Fv region, and the CH1 domain is paired with the CL domain to further stabilize the binding module. Disulfide bonds between the two constant domains can further stabilize the Fab domain.

[0050] For the MBMs of the present disclosure, particularly when the light chain is not a common or universal light chain, it is advantageous to use a Fab heterodimerization strategy to allow correct association of Fab domains belonging to the same ABS and minimize aberrant pairing of Fab domains belonging to different ABSs. For example, the Fab heterodimerization strategy shown in Table 1 below can be used:

[0051] [Table 1]

[0052] Thus, in certain embodiments, correct association between the two polypeptides of a Fab is facilitated by swapping the VL and VH domains of the Fab with one another, or swapping the CH1 and CL domains with one another, as described, for example, in WO 2009 / 080251.

[0053] Correct Fab pairing can also be promoted by introducing one or more amino acid modifications in the CH1 domain and one or more amino acid modifications in the CL domain of the Fab, and / or one or more amino acid modifications in the VH domain and one or more amino acid modifications in the VL domain. The modified amino acids are typically part of the VH:VL and CH1:CL interfaces, such that the Fab components preferentially pair with each other rather than with other Fab components.

[0054] In one embodiment, the one or more amino acid modifications are limited to conserved framework residues of the variable (VH, VL) and constant (CH1, CL) domains, as indicated by the Kabat numbering of the residues. Almagro, 2008, Frontiers In Bioscience 13:1619-1633 provides definitions of framework residues based on the Kabat, Chothia, and IMGT numbering schemes.

[0055] In one embodiment, the modifications introduced into the VH and CH1 and / or VL and CL domains are complementary to each other. Complementarity at the heavy and light chain interface can be achieved based on steric and hydrophobic contacts, electrostatic / charge interactions, or a combination of various interactions. Complementarity between protein surfaces has been widely described in the literature in terms of lock-and-key fit, knobs-into-holes, protrusions and cavities, donors and acceptors, etc., all of which refer to the nature of the structural and chemical match between two interacting surfaces.

[0056] In one embodiment, one or more of the introduced modifications introduce new hydrogen bonds across the interface of the Fab component. In one embodiment, one or more of the introduced modifications introduce new salt bridges across the interface of the Fab component. Exemplary substitutions are described in WO 2014 / 150973 and WO 2014 / 082179, the contents of which are incorporated herein by reference.

[0057] In some embodiments, the Fab domain comprises a 192E substitution in the CH1 domain and 114A and 137K substitutions in the CL domain, which introduce a salt bridge between the CH1 and CL domains (see, e.g., Golay et al., 2016, J Immunol 196:3199-211).

[0058] In some embodiments, the Fab domain comprises 143Q and 188V substitutions in the CH1 domain and 113T and 176V substitutions in the CL domain, which serve to exchange the hydrophobic and polar regions of contact between the CH1 and CL domains (see, e.g., Golay et al., 2016, J Immunol 196:3199-211).

[0059] In some embodiments, the Fab domain can include modifications in some or all of the VH, CH1, VL, and CL domains to introduce an orthogonal Fab interface that promotes correct assembly of the Fab domain (Lewis et al., 2014 Nature Biotechnology 32:191-198). In one embodiment, a 39K, 62E modification is introduced in the VH domain, an H172A, F174G modification is introduced in the CH1 domain, a 1R, 38D, (36F) modification is introduced in the VL domain, and an L135Y, S176W modification is introduced in the CL domain. In another embodiment, a 39Y modification is introduced in the VH domain and a 38R modification is introduced in the VL domain.

[0060] Fab domains can also be modified to replace the native CH1:CL disulfide bond with an engineered disulfide bond, thereby improving the efficiency of Fab component pairing. For example, an engineered disulfide bond can be introduced by introducing 126C into the CH1 domain and 121C into the CL domain (see, e.g., Mazor et al., 2015, MAbs 7:377-89).

[0061] Fab domains can also be modified by replacing the CH1 and CL domains with alternative domains that promote correct assembly. For example, Wu et al., 2015, MAbs 7:364-76, describe replacing the CH1 domain with a constant domain of a T cell receptor and the CL domain with a b domain of a T cell receptor, and combining these domain replacements with additional charge-charge interactions between the VL and VH domains by introducing a 38D modification in the VL domain and a 39K modification in the VH domain.

[0062] Alternatively, or in addition to using a Fab heterodimerization strategy to promote correct VH-VL pairing, a VL of a common light chain (also called a universal light chain) can be used in each Fab VL region of the MBMs of the present disclosure. In various embodiments, using a common light chain as described herein reduces the number of incorrect species of the MBM compared to using the original cognate VL. In various embodiments, the VL domain of the MBM is identified from a monospecific antibody that contains a common light chain. In various embodiments, the VH region of the MBM comprises human heavy chain variable gene segments that are rearranged in vivo in mouse B cells previously engineered to express a limited human light chain repertoire, or a single human light chain cognate to a human heavy chain, to generate an antibody repertoire containing multiple human VHs cognate to one of two possible human VLs in response to challenge with an antigen of interest, the antibody repertoire being specific for the antigen of interest. The common light chain is derived from a rearranged human Vκ1-39Jκ5 sequence or a rearranged human Vκ3-20Jκ1 sequence, including somatically mutated (e.g., affinity matured) versions. See, e.g., U.S. Patent No. 10,412,940.

[0063] Linker In certain embodiments, the present disclosure provides MBMs in which two or more components of an ABS (e.g., the VH and VL of an scFv), two or more ABS (e.g., the scFv and Fab of a half antibody), or an ABS and a non-ABS component (e.g., the Fc region) are connected to each other by a peptide linker. Such linkers are referred to herein as "ABS linkers," in contrast to the ADC linkers used to attach drugs to MBMs, e.g., as described in Section 6.4.

[0064] Peptide linkers can range from 2 to 60 or more amino acids, and in certain embodiments, peptide linkers range in length from 3 to 50 amino acids, 4 to 30 amino acids, 5 to 25 amino acids, 10 to 25 amino acids, 10 to 60 amino acids, 12 to 20 amino acids, 20 to 50 amino acids, or 25 to 35 amino acids.

[0065] In certain embodiments, the peptide linker, e.g., the peptide linker separating the scFv domain and the heavy chain, such as the scFv domain of ABS1 and the heavy chain variable region of ABS2, is at least 5 amino acids, at least 6 amino acids, or at least 7 amino acids in length, and optionally, is at most 30 amino acids, at most 40 amino acids, at most 50 amino acids, or at most 60 amino acids in length.

[0066] In some of the foregoing embodiments, the linker is between 5 and 50 amino acids in length, e.g., between 5 and 50, 5 and 45, 5 and 40, 5 and 35, 5 and 30, 5 and 25, or 5 and 20 amino acids in length. In other of the foregoing embodiments, the linker is between 6 and 50 amino acids in length, e.g., between 6 and 50, 6 and 45, 6 and 40, 6 and 35, 6 and 30, 6 and 25, or 6 and 20 amino acids in length. In still other of the foregoing embodiments, the linker is between 7 and 50 amino acids in length, e.g., between 7 and 50, 7 and 45, 7 and 40, 7 and 35, 7 and 30, 7 and 25, or 7 and 20 amino acids in length.

[0067] Charged (eg, charged hydrophilic linkers) and / or flexible linkers are particularly preferred. Examples of flexible ABS linkers that can be used in the MBMs of the present disclosure include those disclosed in Chen et al., 2013, Adv Drug Deliv Rev. 65(10):1357-1369 and Klein et al., 2014, Protein Engineering, Design & Selection 27(10):325-330. Particularly useful flexible linkers are those containing repeats of glycine and serine, e.g., G n S (SEQ ID NO: 2) or SG n (SEQ ID NO: 3), where n is an integer from 1 to 10, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. In one embodiment, the linker is a G4S (SEQ ID NO: 4), e.g., (GGGGS) n is or comprises a repeating monomer or multimer of

[0068] Polyglycine linkers are suitable for use in the MBMs of the present disclosure. In some embodiments, the peptide linker, e.g., the peptide linker separating the scFv domain and heavy chain, such as the scFv domain of ABS1 and the heavy chain variable region of ABS2, comprises two consecutive glycines (2 Gly), three consecutive glycines (3 Gly), four consecutive glycines (4 Gly (SEQ ID NO: 5)), five consecutive glycines (5 Gly (SEQ ID NO: 6)), six consecutive glycines (6 Gly (SEQ ID NO: 7)), seven consecutive glycines (7 Gly (SEQ ID NO: 8)), eight consecutive glycines (8 Gly (SEQ ID NO: 9)), or nine consecutive glycines (9 Gly (SEQ ID NO: 10)).

[0069] In certain embodiments, the ABS linker, e.g., a peptide linker separating the scFv domain and heavy chain, such as the scFv domain of ABS1 and the heavy chain variable region of ABS2, is composed of both G4S (SEQ ID NO: 4) or a multimer thereof and one or more additional glycines, e.g., 2Gly, 3Gly, or 4Gly (SEQ ID NO: 5). Examples of such linkers include G4S GG (SEQ ID NO: 63), 4xG4S GG (SEQ ID NO: 64), and 7xG4S GG (SEQ ID NO: 65).

[0070] Hinge Area The MBMs of the present disclosure can also include a hinge region, for example, connecting the ABS module to the Fc region. The hinge region can be a natural or modified hinge region. The hinge region is typically found at the N-terminus of the Fc region.

[0071] A native hinge region is the hinge region typically found between the Fab and Fc domains of naturally occurring antibodies. A modified hinge region is any hinge that differs in length and / or composition from the native hinge region. Such hinges can include hinge regions from other species, such as human, mouse, rat, rabbit, shark, pig, hamster, camel, llama, or goat hinge regions. Other modified hinge regions can include a complete hinge region derived from an antibody of a different class or subclass than that of the heavy chain Fc region. Alternatively, the modified hinge region can include a portion of a native hinge or repeating unit, with each unit in the repeat being derived from a native hinge region. In a further alternative, the native hinge region can be modified by converting one or more cysteine ​​or other residues to neutral residues, such as serine or alanine, or by converting appropriately positioned residues to cysteine ​​residues. By such means, the number of cysteine ​​residues in the hinge region can be increased or decreased. Other modified hinge regions may be entirely synthetic and may be designed to have desired properties such as length, cysteine ​​composition, and flexibility.

[0072] Several modified hinge regions have been previously described, for example, in U.S. Pat. No. 5,677,425, WO 9915549, WO 2005 / 003170, WO 2005 / 003169, WO 2005 / 003170, WO 9825971, and WO 2005 / 003171, which are incorporated herein by reference.

[0073] In one embodiment, the Fc region of one or both half antibodies of the disclosure has an intact hinge region, e.g., hinge domain, at its N-terminus. In some embodiments, "hinge domain" refers to the sequence from about Glu216 or Cys226 to about Pro230 of human IgG1 (Burton, 1985 Molec. Immunol. 22:161-206), or the corresponding sequence of another antibody class or isotype.

[0074] In various embodiments, positions 233-236 in the hinge domain can be G, G, G, and unoccupied; G, G, unoccupied, and unoccupied; G, unoccupied, unoccupied, unoccupied, and unoccupied; or all unoccupied, and the positions are numbered according to EU numbering.

[0075] In some embodiments, the MBMs of the present disclosure comprise a modified hinge domain that reduces binding affinity for an Fcγ receptor compared to a wild-type hinge domain of the same isotype (e.g., human IgG1 or human IgG4).

[0076] In one embodiment, the Fc region of one or both chains of an MBM of this disclosure has an intact hinge domain at its N-terminus. In one embodiment, both the Fc region and hinge region of the MBM of the present disclosure are derived from IgG4, with the hinge region containing the modified sequence CPPC (SEQ ID NO: 11). The core hinge region of human IgG4 contains the sequence CPSC (SEQ ID NO: 12), compared to IgG1, which contains the sequence CPPC (SEQ ID NO: 11). The serine residues present in the IgG4 sequence provide increased flexibility in this region, allowing a portion of the molecule to form disulfide bonds within the same protein chain (intrachain disulfides) rather than cross-linking to other heavy chains in an IgG molecule to form interchain disulfides (Angel et al., 1993, Mol Immunol 30(1):105-108). Changing the serine residues to prolines, resulting in the same core sequence as IgG1, allows for the complete formation of interchain disulfides in the IgG4 hinge region, thus reducing heterogeneity in the purified product. This modified isotype is referred to as IgG4P.

[0077] 6.2.4.1. Chimeric Hinge Sequences The hinge region may be a chimeric hinge region. For example, a chimeric hinge can comprise an "upper hinge" sequence derived from a human IgG1, human IgG2, or human IgG4 hinge region combined with a "lower hinge" sequence derived from a human IgG1, human IgG2, or human IgG4 hinge region.

[0078] In certain embodiments, the chimeric hinge region comprises the amino acid sequence EPKSCDKTHTCPPCPAPPVA (SEQ ID NO: 13) (previously disclosed as SEQ ID NO: 8 of WO 2014 / 121087, which is incorporated by reference in its entirety) or ESKYGPPCPPCPAPPVA (SEQ ID NO: 14) (previously disclosed as SEQ ID NO: 9 of WO 2014 / 121087). Such a chimeric hinge sequence may be suitably linked to an IgG4 CH2 region (e.g., by incorporation into an IgG4 Fc domain, e.g., a human or mouse Fc domain, which may be further modified in the CH2 and / or CH3 domains to reduce effector function, e.g., as described in Section 6.2.5.1).

[0079] 6.2.4.2. Hinge Sequences with Reduced Effector Function In further embodiments, the hinge region can be modified to reduce effector function, for example, as described in International Publication No. 2016 / 161010(A2), the entire contents of which are incorporated herein by reference. In various embodiments, positions 233-236 of the modified hinge region are G, G, G, and unoccupied; G, G, unoccupied, and unoccupied; G, unoccupied, unoccupied, unoccupied, and unoccupied; or all unoccupied, with positions numbered according to EU numbering (as shown in Figure 1 of International Publication No. 2016 / 161010(A2)). These segments can be represented as GGG-, GG-, G---, or ----, where "-" represents an unoccupied position.

[0080] Position 236 is unoccupied in standard human IgG2 but occupied in other standard human IgG isotypes. Positions 233-235 are occupied by residues other than G in all four human isotypes (as shown in Figure 1 of WO 2016 / 161010(A2)).

[0081] Hinge modifications within positions 233-236 can be combined with position 228 being occupied by P. Position 228 is naturally occupied by P in human IgG1 and IgG2, but by S in human IgG4 and by R in human IgG3. The S228P mutation in IgG4 antibodies is advantageous for stabilizing IgG4 antibodies and reducing heavy-light chain pair exchange between exogenous and endogenous antibodies. Preferably, positions 226-229 are occupied by C, P, P, and C, respectively.

[0082] Exemplary hinge regions have residues 226-236, sometimes referred to as the middle (or core) and lower hinge, occupied by modified hinge sequences referred to as GGG-(233-236), GG--(233-236), G---(233-236), and no G(233-236). Optionally, the hinge domain amino acid sequence comprises CPPCPAPGGG-GPSVF (SEQ ID NO: 15) (previously disclosed as SEQ ID NO: 1 in WO 2016 / 161010(A2)), CPPCPAPGG--GPSVF (SEQ ID NO: 16) (previously disclosed as SEQ ID NO: 2 in WO 2016 / 161010(A2)), CPPCPAPG---GPSVF (SEQ ID NO: 17) (previously disclosed as SEQ ID NO: 3 in WO 2016 / 161010(A2)), or CPPCPAP----GPSVF (SEQ ID NO: 18) (previously disclosed as SEQ ID NO: 4 in WO 2016 / 161010(A2)).

[0083] The above-described modified hinge regions can be incorporated into heavy chain constant regions, which typically include CH2 and CH3 domains and may have additional hinge segments (e.g., upper hinges) flanking the designated regions. Such additional constant region segments are typically of the same isotype, preferably a human isotype, but may also be hybrids of different isotypes. The isotype of such additional human constant region segments is preferably human IgG4, but may also be human IgG1, IgG2, or IgG3, or hybrids thereof in which the domains are of different isotypes. Exemplary sequences of human IgG1, IgG2, and IgG4 are shown in Figures 2-4 of WO 2016 / 161010(A2).

[0084] In a specific embodiment, the modified hinge sequence can be linked to an IgG4 CH2 region (e.g., by incorporation into an IgG4 Fc domain, e.g., a human or mouse Fc domain, which can be further modified in the CH2 and / or CH3 domains to reduce effector function, e.g., as described in Section 6.2.5.1).

[0085] Fc Domain The MBMs of the present disclosure can comprise an Fc region derived from any suitable species, hi one embodiment, the Fc region is derived from a human Fc domain.

[0086] The Fc domain can be derived from any suitable class of antibody, including IgA (including subclasses IgA1 and IgA2), IgD, IgE, IgG (including subclasses IgG1, IgG2, IgG3, and IgG4), and IgM. In one embodiment, the Fc domain is derived from IgG1, IgG2, IgG3, or IgG4. In one embodiment, the Fc domain is derived from IgG1. In one embodiment, the Fc domain is derived from IgG4.

[0087] The two Fc domains within an Fc region can be the same as or different from one another. In natural antibodies, the Fc domains are typically identical, but for purposes of producing multispecific binding molecules, such as the MBMs of the present disclosure, the Fc domains can be advantageously different to allow for heterodimerization, as described in Section 6.2.5.2, below.

[0088] In natural antibodies, the heavy chain Fc domains of IgA, IgD, and IgG are composed of two heavy chain constant domains (CH2 and CH3), while the heavy chain Fc domains of IgE and IgM are composed of three heavy chain constant domains (CH2, CH3, and CH4), which dimerize to create the Fc region.

[0089] In the MBMs of the present disclosure, the Fc region, and / or the Fc domains therein, can comprise heavy chain constant domains from one or more different classes of antibodies, for example, from one, two, or three different classes.

[0090] In one embodiment, the Fc region comprises CH2 and CH3 domains derived from IgG1. In one embodiment, the Fc region comprises CH2 and CH3 domains derived from IgG2.

[0091] In one embodiment, the Fc region comprises CH2 and CH3 domains derived from IgG3. In one embodiment, the Fc region comprises CH2 and CH3 domains derived from IgG4.

[0092] In one embodiment, the Fc region comprises a CH4 domain from IgM. The IgM CH4 domain is typically located C-terminal to the CH3 domain. In one embodiment, the Fc region comprises the CH2 and CH3 domains derived from an IgG and the CH4 domain derived from an IgM.

[0093] It will be understood that heavy chain constant domains for use in producing Fc regions for MBMs of the present disclosure can include variants of the above-described naturally occurring constant domains. Such variants can include one or more amino acid variations compared to the wild-type constant domain. In one example, the Fc region of the present disclosure includes at least one constant domain that differs in sequence from the wild-type constant domain. It will be understood that the variant constant domain can be longer or shorter than the wild-type constant domain. Preferably, the variant constant domain is at least 60% identical or similar to the wild-type constant domain. In another example, the variant constant domain is at least 70% identical or similar. In another example, the variant constant domain is at least 80% identical or similar. In another example, the variant constant domain is at least 90% identical or similar. In another example, the variant constant domain is at least 95% identical or similar.

[0094] IgM and IgA naturally occur in humans as covalently linked multimers of a common H2L2 antibody unit. IgM occurs as a pentamer when a J chain is incorporated and as a hexamer when the J chain is lacking. IgA occurs in both monomeric and dimeric forms. The heavy chains of IgM and IgA have an 18-amino acid extension to the C-terminal constant domain known as the tailpiece. The tailpiece contains cysteine ​​residues that form disulfide bonds between heavy chains in the polymer and is thought to play an important role in polymerization. The tailpiece also contains glycosylation sites. In certain embodiments, the MBM of the present disclosure does not contain a tailpiece.

[0095] The Fc domain incorporated into the MBMs of the present disclosure may contain one or more modifications that alter the functional properties of the protein, for example, binding to an Fc receptor such as FcRn or a leukocyte receptor, binding to complement, modified disulfide bond structure, or modified glycosylation pattern. Exemplary Fc modifications that alter effector function are described in Section 6.2.5.1.

[0096] The Fc domain can also be engineered to include modifications that improve the manufacturability of asymmetric MBMs, for example, by enabling heterodimerization, the preferential pairing of non-identical Fc domains with identical Fc domains. Heterodimerization allows for the production of MBMs in which different ABSs are connected to each other by Fc regions containing Fc domains that differ in sequence. Examples of heterodimerization strategies are illustrated in Section 6.2.5.2.

[0097] It will be appreciated that any of the above modifications can be combined in any suitable manner to achieve desired functional properties and / or combined with other modifications to alter the properties of the MBM.

[0098] 6.2.5.1. Fc Domains with Altered Effector Functions In some embodiments, the Fc domain comprises one or more amino acid substitutions that reduce binding to Fc receptors and / or effector functions.

[0099] In a specific embodiment, the Fc receptor is an Fcγ receptor. In one embodiment, the Fc receptor is a human Fc receptor. In one embodiment, the Fc receptor is an activating Fc receptor. In a specific embodiment, the Fc receptor is an activating human Fcγ receptor, more particularly human FcγRIIIa, FcγRI, or FcγRIIa, most particularly human FcγRIIIa. In one embodiment, the effector function is one or more selected from the group consisting of complement-dependent cytotoxicity (CDC), antibody-dependent cell-mediated cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and cytokine secretion. In a specific embodiment, the effector function is ADCC.

[0100] In one embodiment, the Fc region comprises an amino acid substitution at a position selected from the group of E233, L234, L235, N297, P331, and P329 (numbering according to the Kabat EU index). In a more specific embodiment, the Fc region comprises an amino acid substitution at a position selected from the group of L234, L235, and P329 (numbering according to the Kabat EU index). In some embodiments, the Fc region comprises amino acid substitutions L234A and L235A (numbering according to the Kabat EU index). In one such embodiment, the Fc region is an Igd Fc region, particularly a human Igd Fc region. In one embodiment, the Fc region comprises an amino acid substitution at position P329. In a more specific embodiment, the amino acid substitution is P329A or P329G, particularly P329G (numbering according to the Kabat EU index). In one embodiment, the Fc region comprises an amino acid substitution at position P329 and an additional amino acid substitution at a position selected from E233, L234, L235, N297, and P331 (numbering according to the Kabat EU index). In a more specific embodiment, the additional amino acid substitution is E233P, L234A, L235A, L235E, N297A, N297D, or P331S. In a particular embodiment, the Fc region comprises amino acid substitutions at positions P329, L234, and L235 (numbering according to the Kabat EU index). In a more specific embodiment, the Fc region comprises amino acid mutations L234A, L235A, and P329G ("P329G LALA," "PGLALA," or "LALAPG").

[0101] Typically, the same one or more amino acid substitutions are present in each of the two Fc domains of the Fc region. Thus, in certain embodiments, each Fc domain of the Fc region comprises the amino acid substitutions L234A, L235A, and P329G (Kabat EU index numbering), i.e., in each of the first and second Fc domains in the Fc region, the leucine residue at position 234 is replaced with an alanine residue (L234A), the leucine residue at position 235 is replaced with an alanine residue (L235A), and the proline residue at position 329 is replaced with a glycine residue (P329G) (Kabat EU index numbering).

[0102] In one embodiment, the Fc domain is an IgG1 Fc domain, particularly a human IgG1 Fc domain. Typically, the same one or more amino acid substitutions are present in each of the two Fc domains of the Fc region. Thus, in certain embodiments, each Fc domain of the Fc region comprises the amino acid substitutions L234A, L235A, and P329G (Kabat EU index numbering), i.e., in each of the first and second Fc domains in the Fc region, the leucine residue at position 234 is replaced with an alanine residue (L234A), the leucine residue at position 235 is replaced with an alanine residue (L235A), and the proline residue at position 329 is replaced with a glycine residue (P329G) (Kabat EU index numbering).

[0103] In one embodiment, the Fc domain is an IgG1 Fc domain, particularly a human IgG1 Fc domain. In some embodiments, the IgG1 Fc domain is a variant IgG1 containing D265A, N297A mutations (EU numbering) that reduce effector function.

[0104] In another embodiment, the Fc domain is an IgG4 Fc domain with reduced binding to Fc receptors. Exemplary IgG4 Fc domains with reduced binding to Fc receptors may comprise an amino acid sequence selected from Table A below. In some embodiments, the Fc domain comprises only the bolded portion of the sequence shown below:

[0105] [Table 2-1]

[0106] [Table 2-2]

[0107] [Table 2-3]

[0108] In certain embodiments, the IgG4 with reduced effector function comprises the bolded portion of the amino acid sequence of SEQ ID NO: 31 of WO 2014 / 121087, which may be referred to herein as IgG4 or hIgG4.

[0109] For heterodimeric MBMs, it is possible to incorporate a combination of the variant IgG4 Fc sequences described above, for example an Fc region comprising a combination of SEQ ID NO: 30 of WO 2014 / 121087 (or a bolded portion thereof) and SEQ ID NO: 37 of WO 2014 / 121087 (or a bolded portion thereof), or an Fc region comprising a combination of SEQ ID NO: 31 of WO 2014 / 121087 (or a bolded portion thereof) and SEQ ID NO: 38 of WO 2014 / 121087 (or a bolded portion thereof).

[0110] 6.2.5.2. Fc Heterodimerization Variants Many multispecific molecule formats, unlike native immunoglobulins, involve dimerization between two Fc domains operably linked to non-identical antigen-binding domains (or portions thereof, e.g., VH or VH-CH1 of a Fab). Incorrect heterodimerization of the two Fc regions that form an Fc domain can be an obstacle to increasing the yield of the desired multispecific molecule and represents a purification challenge. Various approaches available in the art can be used to enhance dimerization of Fc domains that may be present in the MBMs of the present disclosure, as described, for example, in European Patent Application Publication No. 1870459(A1), U.S. Patent No. 5,582,996, U.S. Patent No. 5,731,168, U.S. Patent No. 5,910,573, U.S. Patent No. 5,932,448, U.S. Patent No. 6,833,441, U.S. Patent No. 7,183,076, U.S. Patent Application Publication No. 2006 / 204493(A1), and PCT Publication No. 2009 / 089004(A1).

[0111] The present disclosure provides MBMs comprising Fc heterodimers, i.e., Fc regions comprising heterologous, non-identical Fc domains. Heterodimerization strategies are used to enhance dimerization of Fc regions operably linked to different ABSs (or portions thereof, e.g., VH or VH-CH1 of Fabs) and reduce dimerization of Fc domains operably linked to the same ABS. Typically, each Fc domain in an Fc heterodimer comprises an antibody CH3 domain. The CH3 domain is derived from the constant region of an antibody of any isotype, class, or subclass, preferably the IgG (lgG1, lgG2, lgG3, and lgG4) class, as described in the previous section.

[0112] Heterodimerization of two different heavy chains at the CH3 domain will result in the desired MBM, whereas homodimerization of the same heavy chain will reduce the yield of the desired MBM. Thus, in a preferred embodiment, the two half antibodies that associate to form the MBM of the present disclosure will contain CH3 domains with modifications that favor heterodimer association compared to unmodified chains.

[0113] In specific embodiments, the modification that promotes Fc heterodimer formation is a so-called "knob-into-hole" or "knob-in-hole" modification, which includes a "knob" modification in one of the Fc domains and a "hole" modification in the other Fc domain. Knob-into-hole technology is described, for example, in U.S. Pat. Nos. 5,731,168 and 7,695,936; Ridgway et al., 1996, Prot Eng 9:617-621; and Carter, 2001, Immunol Meth 248:7-15. Generally, the method involves introducing a protrusion ("knob") into the interface of a first polypeptide and a corresponding cavity ("hole") into the interface of a second polypeptide, where the protrusion can be positioned within the cavity to promote heterodimer formation and prevent homodimer formation. The protrusion is constructed by replacing a small amino acid side chain from the interface of the first polypeptide with a larger side chain (e.g., tyrosine or tryptophan). Compensatory cavities of identical or similar size to the protrusions are created at the interface of the second polypeptide by replacing large amino acid side chains with smaller ones (eg, alanine or threonine).

[0114] Thus, in some embodiments, an amino acid residue in the CH3 domain of a first subunit of an Fc domain is replaced with an amino acid residue having a larger side chain volume, thereby creating a protrusion in the CH3 domain of the first subunit that can be positioned within a cavity in the CH3 domain of a second subunit, and an amino acid residue in the CH3 domain of a second subunit of an Fc domain is replaced with an amino acid residue having a smaller side chain volume, thereby creating a cavity in the CH3 domain of the second subunit into which the protrusion in the CH3 domain of the first subunit can be positioned. Preferably, the amino acid residue having a larger side chain volume is selected from the group consisting of arginine (R), phenylalanine (F), tyrosine (Y), and tryptophan (W). Preferably, the amino acid residue having a smaller side chain volume is selected from the group consisting of alanine (A), serine (S), threonine (T), and valine (V). The protrusion and cavity can be created by altering the nucleic acid encoding the polypeptide, for example, by site-directed mutagenesis or by peptide synthesis. An exemplary substitution is Y470T.

[0115] In certain such embodiments, in the first Fc domain, the threonine residue at position 366 is replaced with a tryptophan residue (T366W), and in the Fc domain, the tyrosine residue at position 407 is replaced with a valine residue (Y407V), and optionally, the threonine residue at position 366 is replaced with a serine residue (T366S), and the leucine residue at position 368 is replaced with an alanine residue (L368A) (numbering according to the Kabat EU index). In further embodiments, in the first Fc domain, additionally, the serine residue at position 354 is replaced with a cysteine ​​residue (S354C) or the glutamic acid residue at position 356 is replaced with a cysteine ​​residue (E356C) (particularly, the serine residue at position 354 is replaced with a cysteine ​​residue), and in the second Fc domain, additionally, the tyrosine residue at position 349 is replaced with a cysteine ​​residue (Y349C) (numbering according to the Kabat EU index). In a specific embodiment, the first Fc domain comprises amino acid substitutions S354C and T366W, and the second Fc domain comprises amino acid substitutions Y349C, T366S, L368A, and Y407V (numbering according to the Kabat EU index).

[0116] In some embodiments, electrostatic steering (e.g., Gunasekaran et al. al., 2010, J Biol Chem 285(25):19637-46) can be used to promote the association of the first and second subunits of the Fc domain.

[0117] As an alternative to, or in addition to, the use of an Fc domain modified to promote heterodimerization, the Fc domain can be modified to enable a purification strategy that allows for the selection of Fc heterodimers. In one such embodiment, one half antibody contains a modified Fc domain that abolishes its binding to Protein A, thus enabling a purification method that results in a heterodimeric protein. See, e.g., U.S. Patent No. 8,586,713. As such, MBMs comprise a first CH3 domain and a second Ig The Ig CH3 domains comprise a first and second Ig CH3 domain, wherein the first and second Ig CH3 domains differ from each other by at least one amino acid, and the at least one amino acid difference reduces binding of the MBM to Protein A compared to a corresponding MBM lacking the amino acid difference. In one embodiment, the first CH3 domain binds to Protein A and the second CH3 domain contains a mutation / modification that reduces or eliminates Protein A binding, such as an H95R modification (according to IMGT exon numbering, H435R in EU numbering). The second CH3 may further comprise a Y96F modification (according to IMGT, Y436F in EU). Accordingly, this class of modifications is referred to herein as a "star" mutation.

[0118] 6.3.Target molecules ABS1, ABS2, and ABS3 of the MBM of the present disclosure each specifically bind to a target molecule, e.g., a cell surface-expressed antigen such as a protein, carbohydrate, or lipid. In some embodiments, the target molecules bound by ABS1, ABS2, and ABS3 are protein molecules. Exemplary target molecules include human Klotho-beta ("KLB"), human fibroblast growth factor receptor 1c isoform ("FGFR1c"), human fibroblast growth factor receptor 3 ("FGFR3"), human CD63, and human amyloid precursor-like protein 2 (APLP2).

[0119] Preferably, the scFv, Fab1, and Fab2 are selected so that ABS1, ABS2, and ABS3 can each specifically bind to their respective targets simultaneously. In some embodiments, ABS1, ABS2, and ABS3 each specifically bind to different target molecules. In other embodiments, two of ABS1, ABS2, and ABS3 can bind to different epitopes on the same target molecule.

[0120] If two of ABS1, ABS2, and ABS3 bind to different epitopes on the same target molecule, binding to the target molecule is preferably non-competitive, i.e., the ABSs do not compete for binding to the target molecule (as might occur, for example, if the epitopes overlap). Assays for measuring binding competition between antibodies and antibody fragments are known in the art and include, for example, enzyme-linked immunosorbent assays (ELISAs), fluorescence-activated cell sorting (FACS) assays, and surface plasmon resonance assays.

[0121] Competition for binding to target molecules can be determined, for example, using a real-time, label-free biolayer interference assay on the Octet HTX biosensor platform (Pall ForteBio Corp.). In a specific embodiment of the assay, the entire assay is performed in a buffer solution of 10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 1 mg / mL BSA, 0.05% v / v surfactant Tween-20, pH 7.4 (HBS-EBT buffer) at 25° C. with the plate shaking at a speed of 1000 rpm. To assess whether two antibodies or antigen-binding fragments thereof can compete with each other for binding to their respective epitopes on their specific target antigens, a penta-His-tagged target antigen ("penta-His" disclosed as SEQ ID NO: 25) is first captured onto an Octet biosensor chip (Fortebio Inc, #18-5122) coated with an anti-penta-His antibody ("penta-His" disclosed as SEQ ID NO: 25) by immersing the biosensor chip into a well containing the penta-His-tagged target antigen ("penta-His" disclosed as SEQ ID NO: 25). The antigen-captured biosensor chip is then saturated with a first antibody or antigen-binding fragment thereof (hereafter referred to as Ab-1) by immersion in a well containing a solution of Ab-1 (e.g., a 50 μg / mL solution). The biosensor chip is then subsequently immersed in a well containing a solution of a second antibody or antigen-binding fragment thereof (hereafter referred to as Ab-2) (e.g., a 50 μg / mL solution). The biosensor chip is washed with HBS-EBT buffer between every step of the assay. Real-time binding responses can be monitored throughout the course of the assay, and the binding responses at the end of every step can be recorded. The responses of Ab-2 binding to the target antigen precomplexed with Ab-1 can be compared, and the competitive / noncompetitive behavior of different antibodies / antigen-binding fragments against the same target antigen can be determined.

[0122] MBMs that bind to at least two or more different target molecules can be used, for example, to preferentially target specific tissue types in which the two or more target molecules are expressed while minimizing binding to other tissue types where binding is undesirable. For example, if one or two of ABS1, ABS2, and ABS3 bind to a first target molecule having a first tissue expression profile, and at least one of ABS1, ABS2, and ABS3 binds to a second, different target molecule having a second expression profile that overlaps but is not identical to the first tissue expression profile, the MBM can target the tissue common to the first and second expression profiles.

[0123] MBMs with two or more ABSs that bind to the same target molecule (whether on the same epitope or on different epitopes) may be useful for cell surface receptor clustering and activation.

[0124] Expression profiles can be empirically determined or obtained from public databases such as the Genotype-Tissue Expression (GTEx) project and the Human Protein Atlas. Target selection can be based on protein expression profiles, mRNA expression profiles, or both. To minimize MBM activity in undesired tissue sites, the overlap in tissue expression between any two or all three of ABS1, ABS2, and ABS3 is preferably less than 10 tissues, for example, 9 tissues, 8 tissues, 7 tissues, 6 tissues, 5 tissues, 4 tissues, 3 tissues, 2 tissues, or 1 tissue. An exemplary tissue profile analysis for an exemplary target pair, KLB and FGFR1c, bound by the MBM of the present disclosure is shown in Figure 4A (based on the GTEx database) and 4B (based on the HPA database).

[0125] In various embodiments, ABS1 and ABS2 bind to the same or different epitopes on the same target molecule, and ABS3 binds to a different target molecule or ABS1 and ABS3 bind to the same or different epitopes on the same target molecule, and ABS2 binds to a different target molecule or ABS2 and ABS3 bind to the same or different epitopes on the same target molecule, and ABS1 binds to a different target molecule or ABS1, ABS2, and ABS3 bind to different target molecules, or ABS1, ABS2, and ABS3 bind to the same or different epitopes on the same target molecule.

[0126] When two or more of ABS1, ABS2, and ABS3 bind to the same epitope on a target molecule, such ABSs can have the same or different VH and / or VL sequences.

[0127] Without being bound by theory, it is believed that the MBMs of the present disclosure have the advantage of binding to target molecules (or cells expressing target molecules) with greater affinity than a parent monospecific or bispecific antibody lacking all three ABSs. Thus, the MBMs of the present disclosure, in some embodiments, can bind to one or more target molecules and / or cells expressing one or more target molecules with greater affinity than a parent monospecific or bispecific antibody lacking all three ABSs, e.g., a bispecific antibody lacking the scFv of ABS1. For example, the MBMs, in some embodiments, may have a lower KD for binding to a target molecule and / or a more potent EC50 value in cell-based binding assays (e.g., as described in Section 7) than the corresponding parent monospecific or bispecific antibody.

[0128] The agonist or antagonist activity of a given antibody or MBM depends on target selection, epitope coverage, and format selection. Identification of agonist and antagonist antibodies can be achieved, for example, through function-based screening. The N-terminal scFv MBMs of the present disclosure typically have agonist or antagonist activity if the parent antibody or parent bispecific antibody prior to scFv fusion has agonist or antagonist activity, respectively. Without being bound by theory, it is believed that the MBMs of the present disclosure are characterized by enhanced activity (e.g., agonist or antagonist activity) compared to conventional bispecific molecules (e.g., parent bispecific molecules containing two Fabs but lacking an N-terminal scFv domain), for example, due to the novel binding stoichiometry conferred by the additional N-terminal scFv domain.

[0129] KLB-FGFR1c binders In some embodiments, the MBM of the present disclosure can contain one or more ABSs that bind to human KLB and one or more ABSs that bind to human FGFR1c (e.g., to loops D2 or D3). Human tissues that express both KLB and FGFR1c include adipose tissue, mammary tissue, liver, lung, pancreas, stomach, and testis (see Figures 4A-4B). Thus, MBM of the present disclosure having one or more ABSs that bind to human KLB and one or more ABSs that bind to human FGFR1c can preferentially target these tissue types. FGF21, a member of the FGF family, functions as an endocrine hormone because it signals through a receptor complex composed of FGFR1c and KLB. In one exemplary configuration of this target pair, ABS1 and ABS3 bind to different epitopes on KLB, and ABS2 binds to FGFR1c. Without being bound by theory, it is believed that binding of an MBM with this configuration perturbs the receptor complex, resulting in the metabolic benefits shown in Figure 5.

[0130] Exemplary anti-KLB antibodies are provided in Table 2A, and the VH and VL sequences of exemplary anti-KLB antibodies that can be used in the MBMs of the present disclosure are provided in Table 2B. The MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-KLB antibodies provided in Table 2A or Table 2B. Exemplary anti-FGFR1c antibodies are provided in Table 3A, and the VH and VL sequences of exemplary anti-FGFR1c antibodies that can be used in the MBMs of the present disclosure are provided in Table 3B. The MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-FGFR1c antibodies provided in Table 3A or Table 3B.

[0131] [Table 3]

[0132] [Table 4]

[0133] [Table 5]

[0134] [Table 6]

[0135] FGFR3-APLP2 Binders In some embodiments, the MBM of the present disclosure can contain one or more ABSs that bind to human FGFR3 and one or more ABSs that bind to human APLP2. FGFR3 is a clinically validated tumor driver in bladder cancer. APLP2 has been identified as a cell surface receptor that can undergo rapid internalization and degradation. To induce both FGFR3 blockade and enhanced downregulation of FGFR3 via APLP2, the MBM of the present disclosure can have one or more ABSs that bind to human FGFR3 and one or more ABSs that bind to human APLP2. In one exemplary configuration of this target pair, ABS2 and ABS3 bind to the same epitope on FGFR3, and ABS1 binds to APLP2. Without being bound by theory, it is believed that binding of an MBM with this configuration results in blockage and / or degradation of the FGFR3-receptor complex.

[0136] Exemplary anti-FGFR3 antibody sequences are provided in Table 4. MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-FGFR3 antibodies provided in Table 4. Exemplary anti-APLP2 antibodies are provided in Table 5. MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-APLP2 antibodies provided in Table 5.

[0137] [Table 7]

[0138] [Table 8]

[0139] FGFR3-CD63 Binding Agents In some embodiments, the MBM of the present disclosure can contain one or more ABSs that bind to human FGFR3 and one or more ABSs that bind to human CD63. FGFR3 is a clinically validated tumor driver in bladder cancer. CD63 is a ubiquitously expressed cell surface tetraspanin that can regulate the trafficking of associated partners. To induce both FGFR3 blockade and enhanced CD63-mediated downregulation or surface retention of FGFR3, the MBM of the present disclosure can have one or more ABSs that bind to human FGFR3 and one or more ABSs that bind to human CD63. In one exemplary configuration of this target pair, ABS2 and ABS3 bind to the same epitope on FGFR3, and ABS1 binds to CD63. Without being bound by theory, it is believed that binding of an MBM with this configuration results in blockage and / or disassembly of the FGFR3-receptor complex.

[0140] Exemplary anti-FGFR3 antibody sequences are provided in Table 4 above. MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-FGFR3 antibodies provided in Table 4. Exemplary anti-CD63 antibody VH and VL sequences that can be used in MBMs of the present disclosure are provided in Table 6. MBMs of the present disclosure can include, for example, the CDR or VH and / or VL sequences of any of the anti-CD3 antibodies provided in Table 6.

[0141] [Table 9]

[0142] 6.4. Antibody Drug Conjugates The MBMs of the present disclosure can be conjugated to a drug moiety, for example, via a linker, particularly when the MBM is intended for use as a cancer therapeutic. For convenience, such conjugates are referred to herein as antibody-drug conjugates (or "ADCs").

[0143] In certain aspects, the drug moiety exerts cytotoxic or cytostatic activity. In one embodiment, the drug moiety is a maytansinoid, a kinesin-like protein KIF11 inhibitor, a V-ATPase (vacuolar H+-ATPase) inhibitor, a pro-apoptotic agent, a Bcl2 (B-cell lymphoma 2) inhibitor, a MCL1 (myeloid cell leukemia 1) inhibitor, a HSP90 (heat shock protein 90) inhibitor, an IAP (inhibitor of apoptosis) inhibitor, a mTOR (mechanistic target of rapamycin) inhibitor, a microtubule stabilizer, a microtubule destabilizer, an auristatin, a dolastatin, a MetAP (methionine aminopeptidase), a CRM1 (chromosome maintenance 1) inhibitor. , DPPIV (dipeptidyl peptidase IV) inhibitors, proteasome inhibitors, inhibitors of mitochondrial phosphoryl transfer reactions, protein synthesis inhibitors, kinase inhibitors, CDK2 (cyclin-dependent kinase 2) inhibitors, CDK9 (cyclin-dependent kinase 9) inhibitors, kinesin inhibitors, HDAC (histone deacetylase) inhibitors, DNA damaging agents, DNA alkylating agents, DNA intercalators, DNA minor groove binders, RNA polymerase inhibitors, topoisomerase inhibitors, or DHFR (dihydrofolate reductase) inhibitors.

[0144] In some embodiments, the cytotoxic agent is a maytansinoid having the following structure:

[0145] [ka]

[0146] In some embodiments, the cytotoxic agent is a maytansinoid having the following structure:

[0147] [ka]

[0148] In some embodiments, the ADC comprises an MBM of the present disclosure,

[0149] [ka]

[0150] [ka]

[0151] is the binding to MBM. In some embodiments, the antibody-drug conjugate comprises an MBM of the present disclosure,

[0152] [ka]

[0153] [ka]

[0154] is the binding to MBM. In some embodiments, the ADC comprises an MBM of the present disclosure,

[0155] [ka]

[0156] or including mixtures thereof,

[0157] [ka]

[0158] is binding to the MBM of the present disclosure. In some embodiments, the bond is linked to the MBM via the sulfur moiety of a cysteine ​​residue.

[0159] In some embodiments, the bond is linked to the MBM via the nitrogen moiety of a lysine residue. In the ADCs of the present disclosure, cytotoxic and / or cytostatic agents are linked to the MBM by an ADC linker. The ADC linker linking the cytotoxic and / or cytostatic agent to the MBM of the ADC can be short, long, hydrophobic, hydrophilic, flexible, or rigid, or can be composed of segments each independently possessing one or more of the above properties, such that the linker can contain segments with different properties. Linkers can be multivalent, covalently linking two or more agents to a single site on the MBM, or monovalent, covalently linking a single agent to a single site on the MBM.

[0160] In certain aspects, the linker is selected from a cleavable linker, a non-cleavable linker, a hydrophilic linker, a pro-charged linker, or a dicarboxylic acid-based linker.

[0161] As will be appreciated by those skilled in the art, the ADC linker links the cytotoxic and / or cytostatic agent to the MBM by forming a covalent linkage to the cytotoxic and / or cytostatic agent at one location and to the MBM at another location, where the covalent linkages are formed by reaction between functional groups on the ADC linker and functional groups on the agent and MBM.

[0162] The ADC linker is preferably chemically stable to conditions outside the cell, but need not be; it can be designed to be cleaved, broken, and / or otherwise specifically degraded inside the cell. Alternatively, an ADC linker that is not designed to be specifically cleaved or degraded inside the cell can be used. The choice of stable versus unstable ADC linker can depend on the toxicity of the cytotoxic and / or cytostatic agent. For drugs that are toxic to normal cells, a stable linker is preferred. Selective or targeted drugs with lower toxicity to normal cells can be utilized, in which the chemical stability of the ADC linker to the extracellular environment is less important. A wide variety of ADC linkers useful for linking drugs to MBMs in the context of ADCs are known in the art. Any of these ADC linkers, as well as other ADC linkers, can be used to link cytotoxic and / or cytostatic agents to the MBMs of the ADCs of the present disclosure.

[0163] Exemplary multivalent ADC linkers that can be used to link multiple cytotoxic and / or cytostatic agents to a single MBM molecule are described, for example, in WO 2009 / 073445, WO 2010 / 068795, WO 2010 / 138719, WO 2011 / 120053, WO 2011 / 171020, WO 2013 / 096901, WO 2014 / 008375, WO 2014 / 093379, WO 2014 / 093394, and WO 2014 / 093640, the contents of which are incorporated herein by reference in their entireties. For example, the Fleximer linker technology developed by Mersana et al. has the potential to enable high-DAR ADCs with favorable physicochemical properties. The Mersana technology is based on incorporating drug molecules into a solubilizing polyacetal backbone via a series of ester linkages. The methodology affords highly loaded ADCs (DARs up to 20) while maintaining good physicochemical properties.

[0164] Exemplary monovalent ADC linkers that can be used are described, for example, in Nolting, 2013, Antibody-Drug Conjugates, Methods in Molecular Biology 1045:71-100; Ducrry et al., 2010, Bioconjugate Chem. 21:5-13; Zhao et al., 2011, J. Med. Chem. 54:3606-3623; U.S. Pat. No. 7,223,837; U.S. Pat. No. 8,568,728; U.S. Pat. No. 8,535,678; and WO 2004 / 010957, each of which is incorporated herein by reference.

[0165] By way of example and not limitation, some cleavable and non-cleavable ADC linkers that can be included in the ADCs of the present disclosure are listed below.

[0166] In certain embodiments, the selected ADC linker is cleavable in vivo. Cleavable ADC linkers may contain chemically or enzymatically unstable or degradable linkages. Cleavable ADC linkers generally rely on intracellular processes to release the drug, such as reduction in the cytoplasm, exposure to acidic conditions in lysosomes, or cleavage by specific proteases or other enzymes within the cell. Cleavable ADC linkers generally incorporate one or more chemical bonds that are cleavable either chemically or enzymatically, while the remainder of the ADC linker is non-cleavable. In certain embodiments, the ADC linker contains a chemically labile group such as a hydrazone and / or disulfide group. Linkers containing chemically labile groups take advantage of the different properties between plasma and some cytoplasmic compartments. The intracellular conditions that promote drug release for hydrazone-containing ADC linkers are the acidic environment of endosomes and lysosomes, while disulfide-containing ADC linkers are reduced in the cytosol, which contains high thiol concentrations, such as glutathione. In certain embodiments, the phenotypic stability of ADC linkers containing chemically labile groups can be increased by introducing steric hindrance using substituents near the chemically labile group.

[0167] A cleavable ADC linker may contain a non-cleavable moiety or segment, and / or a cleavable segment or moiety may be included in an otherwise non-cleavable ADC linker to render it cleavable. By way of example only, polyethylene glycol (PEG) and related polymers may contain cleavable groups in the polymer backbone. For example, a polyethylene glycol or polymer ADC linker may contain one or more cleavable groups, such as a disulfide, hydrazone, or dipeptide.

[0168] Other degradable linkages that can be included in ADC linkers include ester linkages formed by the reaction of PEG carboxylic acid or activated PEG carboxylic acid with an alcohol group on a biologically active agent; such ester groups generally hydrolyze under physiological conditions to release the biologically active agent. Hydrolytically degradable linkages include, but are not limited to, carbonate linkages; imine linkages resulting from the reaction of amines and aldehydes; phosphate ester linkages formed by reacting alcohols with phosphate groups; acetal linkages, which are the reaction products of aldehydes and alcohols; orthoester linkages, which are the reaction products of formates and alcohols; and oligonucleotide linkages formed by phosphoramidite groups, including, but not limited to, the termini of polymers and the 5' hydroxyl group of oligonucleotides.

[0169] In certain embodiments, the ADC linker comprises an enzymatically cleavable peptide moiety, e.g., a tripeptide or dipeptide. In certain embodiments, the dipeptide is selected from Val-Cit, Cit-Val, Ala-Ala, Ala-Cit, Cit-Ala, Asn-Cit, Cit-Asn, Cit-Cit, Val-Glu, Glu-Val, Ser-Cit, Cit-Ser, Lys-Cit, Cit-Lys, Asp-Cit, Cit-Asp, Ala-Val, Val-Ala, Phe-Lys, Val-Lys, Ala-Lys, Phe-Cit, Leu-Cit, 111-Cit, Phe-Arg, and Trp-Cit. In certain embodiments, the dipeptide is selected from Cit-Val and Ala-Val.

[0170] In any of the various embodiments of ADCs described above or herein, the ADC may have a drug:antibody ratio (or, in this case, drug:MBM ratio) ranging from 1 to 20, more typically from 2 to 10.

[0171] 6.5. Nucleic Acids and Host Cells In another aspect, the present disclosure provides nucleic acids encoding the MBM of the present disclosure. In some embodiments, the MBM is encoded by a single nucleic acid. In other embodiments, the MBM is encoded by multiple (e.g., two, three, four, or more) nucleic acids.

[0172] A single nucleic acid can encode an MBM comprising a single polypeptide chain, an MBM comprising two or more polypeptide chains, or a portion of an MBM comprising three or more polypeptide chains (e.g., a single nucleic acid can encode two polypeptide chains of an MBM comprising three, four or more polypeptide chains, or three polypeptide chains of an MBM comprising four or more polypeptide chains). For separate expression control, open reading frames encoding two or more polypeptide chains can be placed under the control of separate transcriptional regulatory elements (e.g., promoters and / or enhancers). Open reading frames encoding two or more polypeptides can also be controlled by the same transcriptional regulatory element and separated by an internal ribosome entry site (IRES) sequence, allowing translation into separate polypeptides.

[0173] In some embodiments, an MBM comprising two or more polypeptide chains is encoded by two or more nucleic acids. The number of nucleic acids encoding the MBM can be equal to or less than the number of polypeptide chains in the MBM (e.g., when two or more polypeptide chains are encoded by a single nucleic acid).

[0174] The nucleic acids of the present disclosure can be DNA or RNA (eg, mRNA). In another aspect, the disclosure provides host cells and vectors containing the nucleic acids of the disclosure. The nucleic acids can be present in a single vector or in separate vectors present in the same host cell or in separate host cells, as described in more detail herein below.

[0175] Vectors The present disclosure provides vectors comprising nucleotide sequences encoding the MBM or MBM components described herein, e.g., one or two of the polypeptide chains of a half antibody. Vectors include, but are not limited to, viruses, plasmids, cosmids, lambda phage, or yeast artificial chromosomes (YACs).

[0176] Numerous vector systems can be used. For example, one class of vectors utilizes DNA elements derived from animal viruses such as bovine papillomavirus, polyomavirus, adenovirus, vaccinia virus, baculovirus, retrovirus (Rous sarcoma virus, MMTV, or MOMLV), or SV40 virus. Another class of vectors utilizes RNA elements derived from RNA viruses such as Semliki Forest virus, eastern equine encephalitis virus, and flaviviruses.

[0177] Additionally, cells that have stably integrated the DNA into their chromosomes can be selected by introducing one or more markers that allow for the selection of transfected host cells. Markers can provide, for example, prototrophy to auxotrophic hosts, resistance to biocides (e.g., antibiotics), or resistance to heavy metals such as copper. The selectable marker gene can be directly linked to the DNA sequence to be expressed or introduced into the same cell by cotransformation. Additional elements may be required for optimal mRNA synthesis. These elements may include splice signals, as well as transcription promoters, enhancers, and termination signals.

[0178] When the expression vector or DNA sequence containing construct is prepared for expression, the expression vector can be transfected or introduced into suitable host cell.To achieve this, various techniques can be used, such as protoplast fusion, calcium phosphate precipitation, electroporation, retroviral transduction, viral transfection, gene gun, lipid-based transfection or other conventional techniques.The method and conditions for culturing the resulting transfected cells and recovering expressed polypeptide are known to those skilled in the art, and can be modified or optimized according to the specific expression vector and mammalian host cell used based on this description.

[0179] 6.5.2.Cells The present disclosure also provides a host cell comprising a nucleic acid of the present disclosure. In one embodiment, the host cell is genetically engineered to contain one or more nucleic acids described herein.

[0180] In one embodiment, host cells are genetically engineered by using an expression cassette. The phrase "expression cassette" refers to a nucleotide sequence that can affect the expression of a gene in a host that is compatible with such a sequence. Such a cassette can include a promoter, an open reading frame with or without introns, and a termination signal. Additional factors that are necessary or useful for producing expression, such as an inducible promoter, can also be used.

[0181] The present disclosure also provides host cells comprising the vectors described herein. The cell can be, but is not limited to, a eukaryotic cell, a bacterial cell, an insect cell, or a human cell. Suitable eukaryotic cells include, but are not limited to, Vero cells, HeLa cells, COS cells, CHO cells, HEK293 cells, BHK cells, and MDCKII cells. Suitable insect cells include, but are not limited to, Sf9 cells.

[0182] Pharmaceutical Compositions The MBM and / or ADC of the present disclosure may be in the form of a composition comprising the MBM and / or ADC and one or more carriers, excipients, and / or diluents. The composition may be formulated for a particular use, such as veterinary use or pharmaceutical use in humans. The form of the composition (e.g., dry powder, liquid formulation, etc.) and the excipients, diluents, and / or carriers used will depend on the intended use of the MBM and / or ADC and the mode of administration for therapeutic use.

[0183] For therapeutic use, the composition may be provided as part of a sterile pharmaceutical composition containing a pharmaceutically acceptable carrier. The composition may be in any suitable form (depending on the desired method of administration to a patient). The pharmaceutical composition may be administered to a patient by a variety of routes, including oral, transdermal, subcutaneous, intranasal, intravenous, intramuscular, intratumoral, intrathecal, topically, or locally. The most suitable route for administration in any given case will depend on the particular antibody and / or ADC, the subject, and the nature and severity of the disease and the subject's physical condition. Typically, the pharmaceutical composition will be administered intravenously or subcutaneously.

[0184] The pharmaceutical composition can be conveniently presented in a unit dosage form containing a predetermined amount of MBM and / or ADC of the present disclosure per dose. The amount of MBM and / or ADC contained in the unit dose will depend on the disease being treated and other factors well known in the art. Such a unit dosage can be in the form of a lyophilized dry powder containing an amount of MBM and / or ADC suitable for a single administration, or in liquid form. The dry powder unit dosage form can be packaged in a kit with a syringe, a suitable amount of diluent, and / or other components useful for administration. The liquid unit dosage can be conveniently supplied in the form of a syringe pre-filled with an amount of MBM and / or ADC suitable for a single administration.

[0185] Pharmaceutical compositions may also be supplied in bulk, as they contain an amount of the ADC suitable for multiple administrations. Pharmaceutical compositions can be prepared for storage as lyophilized formulations or aqueous solutions by mixing MBM and / or ADC having the desired purity with any pharmaceutically acceptable carrier, excipient, or stabilizer (all of which are referred to herein as "carriers") typically used in the art, i.e., buffers, stabilizers, preservatives, isotonicity agents, non-ionic detergents, antioxidants, and various other additives. See Remington's Pharmaceutical Sciences, 16th edition (Osol, ed. 1980). Such additives should be nontoxic to recipients at the dosages and concentrations used.

[0186] Buffering agents help maintain a pH in a range that approximates physiological conditions. They can be present in a wide variety of concentrations, but will typically be present at concentrations ranging from about 2 mM to about 50 mM. Buffering agents suitable for use in the present disclosure include both organic and inorganic acids and their salts, such as citrate buffers (e.g., monosodium citrate-disodium citrate mixtures, citric acid-trisodium citrate mixtures, citric acid-monosodium citrate mixtures, etc.), succinate buffers (e.g., succinic acid-monosodium succinate mixtures, succinic acid-sodium hydroxide mixtures, succinic acid-disodium succinate mixtures, etc.), tartrate buffers (e.g., tartaric acid-sodium tartrate mixtures, tartaric acid-potassium tartrate mixtures, tartaric acid-sodium hydroxide mixtures, etc.), fumarate buffers (e.g., fumaric acid-monosodium fumarate mixtures, fumaric acid-monosodium fumarate mixtures, etc.), and the like. -disodium fumarate mixture, monosodium fumarate-disodium fumarate mixture, etc.), gluconate buffers (e.g., gluconic acid-sodium glyconate mixture, gluconic acid-sodium hydroxide mixture, gluconic acid-potassium glyconate mixture, etc.), oxalate buffers (e.g., oxalic acid-sodium oxalate mixture, oxalic acid-sodium hydroxide mixture, oxalic acid-potassium oxalate mixture, etc.), lactate buffers (e.g., lactic acid-sodium lactate mixture, lactic acid-sodium hydroxide mixture, lactic acid-potassium lactate mixture, etc.), and acetate buffers (e.g., acetic acid-sodium acetate mixture, acetic acid-sodium hydroxide mixture, etc.). Additionally, phosphate buffers, histidine buffers, and trimethylamine salts such as Tris can be used.

[0187] Preservatives may be added to retard microbial growth in amounts ranging from about 0.2% to 1% (w / v). Preservatives suitable for use in the present disclosure include phenol, benzyl alcohol, metacresol, methylparaben, propylparaben, octadecyldimethylbenzylammonium chloride, benzalkonium halides (e.g., chloride, bromide, and iodide), hexamethonium chloride, and alkylparabens such as methyl or propylparaben, catechol, resorcinol, cyclohexanol, and 3-pentanol. Tonicity adjusting agents, also known as "stabilizers," may be added to ensure the isotonicity of the liquid compositions of the present disclosure and include polyhydric sugar alcohols, such as trihydric or higher sugar alcohols, such as glycerin, erythritol, arabitol, xylitol, sorbitol, and mannitol. Stabilizers refer to a broad category of excipients whose functions can range from bulking agents to additives that help solubilize the therapeutic agent or prevent denaturation or adhesion to the container wall. Typical stabilizers include polyhydric sugar alcohols (listed above); amino acids such as arginine, lysine, glycine, glutamine, asparagine, histidine, alanine, ornithine, L-leucine, 2-phenylalanine, glutamic acid, threonine, etc.; organic sugars or sugar alcohols such as lactose, trehalose, stachyose, mannitol, sorbitol, xylitol, ribitol, myo-inisitol, galactitol, glycerol, etc. (including cyclitols such as inositol); polyethylene glycol; amino acid polymers; sulfur-containing reducing agents such as urea, glutamine, and the like. The sugars may be thiones, thioctic acid, sodium thioglycolate, thioglycerol, α-monothioglycerol, and sodium thiosulfate; low molecular weight polypeptides (e.g., peptides of 10 residues or less); proteins, such as human serum albumin, bovine serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; monosaccharides, such as xylose, mannose, fructose, glucose; disaccharides, such as lactose, maltose, sucrose, and trehalose; and trisaccharides, such as raffinose; and polysaccharides, such as dextran.The stabilizer may be present in an amount ranging from 0.5 to 10% by weight per weight of the ADC.

[0188] Nonionic surfactants or detergents (also known as "wetting agents") can be added to help solubilize glycoproteins and protect them against agitation-induced aggregation, which also allows the formulation to be exposed to stressed shear surfaces without causing denaturation of the protein. Suitable nonionic surfactants include polysorbates (e.g., 20, 80), poloxamers (e.g., 184, 188), and Pluronic® polyols. The nonionic surfactant can be present in a range of about 0.05 mg / mL to about 1.0 mg / mL, e.g., about 0.07 mg / mL to about 0.2 mg / mL.

[0189] Additional miscellaneous excipients include bulking agents (eg, starch), chelating agents (eg, EDTA), antioxidants (eg, ascorbic acid, methionine, vitamin E), and cosolvents.

[0190] 6.7. Therapeutic Indications The MBMs, ADCs, and pharmaceutical compositions of the disclosure can be used to treat cancer, for example, cancers associated with expression of target molecules to which ABS1, ABS2, and ABS3 bind.

[0191] Thus, in one aspect, the present disclosure provides a method for treating cancer, comprising administering to a subject suffering from cancer an effective amount of the MBM, conjugate, or pharmaceutical composition of the present disclosure. In certain embodiments, the MBM of the present disclosure that binds to FGFR3 and optionally binds to CD63 or APLP2 can be used to treat bladder cancer, glioblastoma, and squamous cell lung cancer.

[0192] The MBM, ADCs, and pharmaceutical compositions of the present disclosure can also be used for non-cancer indications, such as treating non-alcoholic steatohepatitis ("NASH"), treating metabolic diseases, reducing circulating HDL cholesterol, increasing circulating LDL cholesterol, reducing blood triglycerides, reducing blood glucose, treating obesity, and treating diabetes. For such non-cancer indications, in some embodiments, the MBM targets KLB and / or FGFR1c.

[0193] Thus, in one aspect, the present disclosure provides a method for reducing circulating HDL cholesterol, comprising administering to a subject with elevated HDL levels an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0194] In another aspect, the present disclosure provides a method for increasing circulating LDL cholesterol, comprising administering to a subject having low LDL levels an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0195] In another aspect, the present disclosure provides a method for reducing blood triglycerides, comprising administering to a subject having elevated triglyceride levels an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0196] In another aspect, the present disclosure provides a method for reducing blood glucose, comprising administering to a subject having elevated glucose levels an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0197] In another aspect, the present disclosure provides a method of treating obesity, comprising administering to a subject suffering from obesity an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0198] In another aspect, the present disclosure provides a method of treating diabetes, comprising administering to a subject suffering from diabetes an effective amount of an MBM, conjugate, or pharmaceutical composition of the present disclosure.

[0199] 7. Working Example 7.1. Example 1: FGFR1c x KLB x KLB 2+1 N-scFv MBM 7.1.1 Materials and Methods 7.1.1.1. Generating MBM Various KLB proteins that bind to KLB epitope 1 (ep1) or epitope 2 (ep2) The scFv was fused to the N-terminus of the FGFR1c VH domain from REGN4366, an existing IgG-like bispecific molecule that targets both FGFR1c and KLB (Figure 1 and Table 7).

[0200] DNA fragments encoding (i) various KLB scFvs with VL (having a 100C mutation, Kabat numbering), linker (4xG4S (SEQ ID NO: 57)), and VH (having a 44C mutation, Kabat numbering), followed by linker orientations of various lengths to connect the scFv to the FGFR1c-binding Fab, (ii) FGFR1c-binding Fab, and (iii) IgG1 Fc domains with knob-forming mutations (S354C, T366W, EU numbering), hole-forming mutations (Y349C, T366S, L368A, Y407V, EU numbering), and star-shaped mutations (H435R, Y436F, EU numbering) were synthesized by Integrated DNA Technologies, Inc. (San Diego, California) or GenScript (Piscataway, NJ).

[0201] Mammalian expression vectors for the individual heavy chains were generated either by using the NEBuilder HiFi DNA Assembly Kit (New England BioLabs Inc.) or by restriction digestion followed by ligation according to the standard molecular cloning protocol provided by New England BioLabs Inc. For expression of FGFR1c / KLB / KLB 2+1 N-scFv MBMs (F1K-scFv1-9), the heavy chains ("Hc1-knob" and "Hc2-hole") were cloned into the 100-knob and 100-knob vectors. * ") and universal light chain DNA were co-transfected into Expi293 cells (ThermoFisher Scientific) according to the manufacturer's protocol. 50 ml of cell culture medium was collected and processed for purification via a HiTrap Protein A FF column (GE Healthcare). For functional confirmation, the selected MBM was scaled up to 200 ml and subjected to a series of purification procedures, including size exclusion chromatography as the final step. REGN4366, which has a Fab that binds to KLB and a Fab that binds to FGFR1c, was generated and purified to serve as an IgG-like bispecific control molecule.

[0202] 7.1.1.2.HEK293.SREluc.hFGFR1cHS / hKLB Reporter-Based Assay MBMs were tested for their agonistic activity using HEK293.SREluc.hFGFR1cHS / hKLB cells, which stably expressed human FGFR1c and KLB and a luciferase reporter gene under the control of a promoter containing a serum response element (SRE). Recombinant human FGF21 with a 6xHis tag (SEQ ID NO: 58) was used as a positive control, and the maximum reporter activity obtained from FGF21 was defined as 100% activity. Cells were treated with each MBM or 6xHis-FGF21 for 6 hours and then subjected to luciferase assays. The percent activity induced by each MBM was normalized to the maximum activity induced by FGF21. A dose-response assay was performed to determine the EC50.

[0203] 7.1.1.3. Biacore Analysis of hFGFR1c and hKLB Binding The binding affinity and mechanism of action for F1K-scFv6 were determined by Biacore analysis. To compare the apparent affinity of F1K-scFv6 for KLB with the monovalent affinity of the parent KLB mAb, an antibody capture format was used. Briefly, FGFR1c parent mAb 19842, KLB parent mAbs 22532P2 and 22393P2, FGFR1c / KLB parent bispecific antibody REGN4366, and F1K-scFv6 were immobilized on a CM5 chip with an anti-human Fc antibody (REGN2567). Different concentrations of hFGFR1c_V5_6xHis (800-12.5 nM, 4-fold dilutions) or hKLB.HA.6xHis (100-1.56 nM, 4-fold dilutions) were injected at 50 μL / min for 2 minutes, and the assay was performed at 25°C. Binding kinetic parameters were determined by fitting the real-time data with a 1:1 binding model using Scrubber 2.0c.

[0204] 7.1.1.4. Cell-Based Binding by Flow Cytometry HEK293 / Cas9-hFGFR1 / hKLB(KLB + ), HEK293 / hFGFR1c (hFGFR1c + ), and HEK293 / hFGFR1c / hKLB(hFGFR1c + / hKLB + ) Cells were cultured at 1 × 10 in complete Dulbecco's modified Eagle's medium (DMEM) medium (10% fetal bovine serum (FBS), 1x pen-strep-glutamine). 6 Resuspend and stain at 1 x 10 cells / mL 5This was performed on cells. MBM was added and the cells were stained for 30 minutes at 2-8°C. The cells were washed twice with FACS wash buffer (phosphate-buffered saline (PBS) containing 1% FBS and 1 mM EDTA) and centrifuged at 1800 RPM for 4 minutes at 4°C. Allophycocyanin-conjugated goat anti-human IgG (Jackson Immuno Research, 109-136-098, 1:400) was added and incubated with the cells for 30 minutes at 2-8°C. The cells were washed as before and resuspended in 100 μL of 2% paraformaldehyde. The cells were incubated for 30 minutes at 2-8°C and washed twice. The stained cells were analyzed using a BD LSRFortessa® FACS instrument.

[0205] 7.1.2.Results Generation of FGFR1c / KLB / KLB 2+1 N-scFv MBM Nine FGFR1c / KLB / KLB 2+1 N-scFv MBMs with the characteristics shown in Table 7 were expressed and purified.

[0206] [Table 10]

[0207] 7.1.2.2. Epitope-Dependent Activities of FGFR1c / KLB / KLB 2+1 N-scFv Trispecific Molecules The results of a cell-based reporter assay using HEK293.SREluc.hFGFR1cHS / hKLB to analyze the agonist activity of purified FGFR1c / KLB / KLB 2+1 N-scFv MBMs are shown in Table 8. F1K_scFv3, F1K_scFv6, and F1K_scFv9 were found to be the top activators based on % activity. They all share the same KLB-targeting scFv that binds to the KLB ep2 region and were significantly more active in the assay than MBMs bearing scFvs targeting the same KLB ep1 region as the original KLB Fab. In particular, F1K_scFv6, which has a 30-amino acid-long scFv-Fab linker, was observed to have the highest activity at 54.1% and the highest potency (EC50 = 9.80E-10 M).

[0208] [Table 11]

[0209] 7.1.2.3.2+1 N-scFv trispecific F1K-scFv6 can simultaneously bind to two different epitopes on the same KLB. The results of Biacore analysis on F1K-scFv6 are shown in Tables 9A-9B. F1K-scFv6 had a K of 1.47E-11 M for hKLB. D The F1K-scFv6 antibody was found to have a 55-fold and 1,333-fold increase in affinity compared to the parent KLB mAbs 22532P2 and 22393P2, respectively. Enhanced affinity for hKLB was also observed when F1K-scFv6 was compared to REGN4366 (the 22393P2 arm for hKLB). This observation strongly supports the conclusion that F1K-scFv6, which has different epitope-targeting arms for KLB, can simultaneously engage both binding sites on the same hKLB molecule. For hFGFR1c binding, F1K-scFv6 has a slightly reduced affinity (4-fold) for hFGFR1c compared to its parent FGFR1c mAb 19842 or the parent bispecific antibody REGN4366 (the 19842 arm for hFGFR1c).

[0210] [Table 12]

[0211] [Table 13]

[0212] 7.1.2.4.2+1 N-scFv trispecific F1K-scFv6 binds more tightly to cells overexpressing hKLB and hFGFR1c / hKLB. To further confirm the relevance of the enhanced affinity of F1K-scFv6 for hKLB in a cell-based setting, FACS binding assays were used to compare the binding of trispecific F1K-scFv6 (FGFR1c / KLB / KLB) to the bispecific antibodies REGN4366 (FGFR1c / KLB 22393 arm) and REGN6799 (FGFR1c / KLB 22532 arm), and the parental antibodies 22393 IgG (KLB), 22532 IgG (KLB), and 19842 IgG (FGFR1c). In both HEK293 / Cas9-hFGFR1 / hKLB (hFGFR1 knockout and hKLB overexpression) and HEK293 / hFGFR1c / hKLB (hFGFR1c and hKLB overexpression) cells, F1K-scFv6 consistently demonstrated stronger EC50 binding than all other controls (Figures 6A and 6C). In HEK293 / hFGFR1c cells, due to the bivalency of hFGFR1c, 19842 IgG was observed to have the strongest binding, followed by REGN4366 and F1K-scFv6 (Figure 6B). The FACS binding data are in good agreement with Biacore analysis. Without being bound by theory, the data in this example suggest that dual epitope engagement of hKLB via the 2+1 N-scFv trispecific design improves antibody-mediated KLB / FGFR1c receptor complex interaction and potential cell surface clustering.

[0213] 7.1.2.5.2+1 N-scFv MBM F1K-scFv6 exhibits superior agonist activity to the bispecific REGN4366 To assess whether the enhanced affinity of F1K-scFv6 could translate into improved efficacy over the bispecific REGN4366, we performed an in vitro reporter-based cell assay using HEK293.SREluc.hFGFR1cHS / hKLB. Both F1K-scFv6 and REGN4366 were purified via size-exclusion chromatography as a final step after affinity purification to remove any aggregation that could confound data interpretation. Compared to the parent bispecific antibody REGN4366, which has only 19% activation when normalized to soluble FGF21, F1K-scFv6 not only enhanced the potency (EC50) of luciferase gene expression by 4-5-fold, but also significantly increased the % activation to 77% (Figure 8).

[0214] A similar trivalent molecule using the same 22532P2 scFv but fused to the C-terminus of Fc, termed 2+1 C-scFv, was also generated and evaluated. This corresponding trispecific molecule had much lower expression and purification yields with inferior functional activity than F1K-scFv6 (results not shown).

[0215] 7.2. Example 2: Evaluation of Linker Length on Trispecific Activity 7.2.1 Materials and Methods The HEK293.SREluc.hFGFR1c.hKLB stable cell line was generated by sequentially transfecting HEK293 cells with the SRE-luciferase reporter, full-length human FGFR1c, and full-length human KLB plasmids.

[0216] Similar to the control bispecific molecule, linker length variants of the trispecific molecule, designated scFv6, were generated and tested in luciferase reporter assays. The linker variants are designated scFv6 LK7:G4S GG (SEQ ID NO: 63), scFv6 LK15:3xG4S (SEQ ID NO: 1), scFv6 LK22:4xG4S GG (SEQ ID NO: 64), scFv6:6xG4S (SEQ ID NO: 59), scFv6 LK37:7xG4S GG (SEQ ID NO: 65), and scFv6 LK45:9xG4S (SEQ ID NO: 60).

[0217] Cells were seeded into 384-well plates and cultured overnight in complete medium containing 10% fetal bovine serum (FBS). The culture medium was changed to Opti-MEM Reduced Serum Medium (ThermoFisher, USA) supplemented with 0.1% FBS. After approximately 24 hours, the cells were treated with serially diluted ligands for 6 hours and then subjected to luciferase assays using the ONE-Glo™ Luciferase Assay System (Promega, USA) according to the manufacturer's instructions.

[0218] 7.2.2.Results In one study, the agonist activity of bispecific binding molecules (BBMs) called REGN4304 and REGN4366 was compared to that of hFGF21. In this assay, BBMs exhibited approximately 30% of the agonist activity of hFGF21 (Figure 7).

[0219] In another study, the agonist activity of linker length variants (peptide linkers between domains designated 2 and 3 in Figure 9) was compared to that of FGF21 and a BBM called REGN4304. The results are shown in Figure 9 and Table 10 below.

[0220] [Table 14]

[0221] All linker length variants exhibited at least approximately twice the activity of the control bispecific binding molecule, with variants with the shortest linker lengths (7 or 15 amino acids) exhibiting the greatest agonist activity.

[0222] 7.3. Example 3: Activation of FGFR1c signaling in HEK293 cells 7.3.1 Materials and Methods The HEK293.SREluc.hFGFR1c.hKLB stable cell line was generated as described in Example 3. For Western blot analysis, HEK293.SREluc.hFGFR1c.hKLB cells were seeded into 6-well plates and cultured overnight in complete medium containing 10% fetal bovine serum (FBS). The culture medium was changed to Opti-MEM Reduced Serum Medium (ThermoFisher, USA) supplemented with 0.1% FBS. Approximately 24 hours later, diluted ligands were added to the cells to a final concentration of 1 nM or 10 nM. After 15 minutes of treatment, the cells were washed with cold PBS and lysed in RIPA lysis buffer (150 mM Tris / HCl, pH 7.4, 50 mM NaCl, 1% NP-40, and 0.1% Tween 20). Total cell lysates were resolved by SDS-PAGE and transferred to a PVDF membrane. For Western blot analysis, the following primary antibodies were used: total ERK (Cell Signaling, 9102), phospho-ERK (Cell Signaling, 9101), PLC-gamma (Cell Signaling, 5690), phosphor-PLC gamma (Cell Signaling, 2821).

[0223] 7.3.2.Results The agonist activity of the bispecific and trispecific binding molecules was tested in HEK293.SREluc.hFGFR1c.hKLB cells stably expressing human FGFR1c and human KLB, along with the irrelevant antibody REGN1945 as a negative control and FGF21 (REGN1438) as a positive control. After treatment, ERK and PLC-gamma ligase oxidation induced by activated FGFR1c were measured, as well as luciferase activity.

[0224] Both F1K_scFv6LK7 and F1K_scFv6L1 strongly induced ERK and PLC-gamma linoxidation at both 1 nM and 10 nM concentrations. Notably, phospho-ERK and phospho-PLC-gamma levels in F1K_scFv6- or F1K_scFv6LK7-treated cells were significantly higher than those in cells treated with the parent bispecific antibody (REGN4366), the FGFR1 / KLB agonist bispecific antibody (REGN4304), or recombinant human FGF21 (REGN1438) at the corresponding concentrations (Figure 10A).

[0225] To assess the time course of FGFR1c activation, HEK293.SREluc.hFGFR1c.hKLB cells were treated with ligands for varying periods of time and harvested for Western blot analysis. The results are shown in Figure 10B. ERK activation, measured by phospho-ERK levels, was observed as early as 15 min after treatment with REGN1438, REGN4304, or F1K_scFv6 and persisted for up to 6 h. F1K_scFv6 exhibited higher phospho-ERK levels compared with REGN1438 or REGN4304 throughout the treatment time course. F1K_scFv6 strongly induced phospho-PLC gamma at 15 min, which then gradually decreased over time.

[0226] 7.4. Example 4: Size analysis of in vitro complexes formed between KLB, FGFR1c, and binding molecules by asymmetric flow field fractionation combined with multi-angle laser light scattering (A4F-MALLS) Overview In principle, the trispecific binding molecule of the present disclosure can form different types of complexes with FGFR1c and KLB.To determine the type of complex formed, size analysis of the in vitro complex formed between 2+1 N-scFv and 2+1 N-Fab trispecific binding molecules was performed using asymmetric flow field fractionation (A4F-MALS) combined with multi-angle light scattering.A4F-MALS was also used to analyze the complexes formed by a control bispecific binding molecule (REGN4304) and a monospecific KLB binding molecule (REGN4661).

[0227] 7.4.2 Materials and Methods 7.4.2.1.A4F-MALLS mobile phase buffer Mobile phase buffer (10 mM sodium phosphate, 500 mM sodium chloride, pH 7.0±0.1) was prepared by combining 1.4 g of sodium phosphate monobasic monohydrate, 10.7 g of sodium phosphate dibasic heptahydrate, and 500 mL of 5 M sodium chloride, then bringing the solution to a volume of 5.0 L with HPLC-grade water. The final measured pH of the buffer was 7.0. The mobile phase buffer was filtered (0.2 μm) before use.

[0228] 7.4.2.2.A4F-MALLS The A4F-MALLS system uses an ultraviolet (UV) diode array detector, Wyatt The instrument consisted of an Eclipse™ 3+A4F separation system coupled to an Agilent 1200 Series HPLC system equipped with a Wyatt Technology Dawn HELEOS® II laser light scattering (LS) instrument and an Optilab® T-rEX differential refractometer (RI) detector. The detectors were connected in series in the following order: UV-LS-RI. The LS and RI detectors were calibrated according to the instructions provided by Wyatt Technology.

[0229] Defined amounts of anti-KLB and anti-FGFR1c polyspecific binding molecule candidates were combined with REGN6424 (recombinant KLB) and REGN6152 (recombinant FGFR1c), respectively, and diluted with 1X DPBS, pH 7.4, to obtain equimolar ratios of 0.2 μM polyspecific binding molecule: 0.2 μM REGN6424 or 0.2 μM polyspecific binding molecule: 0.2 μM REGN6424: 0.2 μM REGN6152. All samples were incubated at ambient temperature for 2 hours and maintained unfiltered at 4°C before injection into an Eclipse™ short channel equipped with a W350 spacer foil (spacer thickness 350 μm, spacer width 2.2 cm) using a 10 kDa MWCO regenerated cellulose membrane. Prior to injection of each sample, the channel was pre-equilibrated with mobile phase buffer (10 mM sodium phosphate, 500 mM sodium chloride, pH 7.0 ± 0.1). Bovine serum albumin (BSA, 2 mg / mL, 10 μg sample load) was injected separately and included as a system suitability control.

[0230] The fractionation method consisted of four steps: injection, focusing, elution, and a channel "washout" step. A4F-MALLS mobile phase buffer (10 mM sodium phosphate, 500 mM sodium chloride, pH 7.0 ± 0.1) was used throughout the fractionation method. Each sample (7 μg) was injected for 1 min at a flow rate of 0.2 mL / min, followed by focusing for 3 min at a focusing flow rate of 1.0 mL / min. The sample was eluted for 15 min at a channel flow rate of 1.0 mL / min with a constant crossflow of 3.0 mL / min, followed by a linear gradient crossflow from 3.0 mL / min to 0 mL / min over 5 min. Finally, the crossflow was held at 0 mL / min for an additional 5 min to wash out the channel. BSA was fractionated using the same parameter settings.

[0231] 7.4.2.3.MALLS Data Analysis Data were analyzed using ASTRA V software (version 5.3.4.14, Wyatt Technology). Data were fitted to an equation relating excess scattered light to solute concentration and weight-average molar mass, Mw (Kendrick et al., 2001, Anal Biochem. 299(2):136-46; Wyatt, 1993, Anal. Chim. Acta 272(1):1-40):

[0232]

number

[0233] where c is the solute concentration, R(θ, c) is the excess Raley ratio from the solute as a function of scattering angle and concentration, Mw is the molar mass, P(θ) represents the angular dependence of the scattered light (approximately 1 for particles with a radius of gyration less than 50 nm), A2 is the second virial coefficient for osmotic expansion (which can be neglected since the measurements are performed in dilute solutions), and

[0234]

number

[0235] In the formula, n0 represents the refractive index of the solvent, and N A is Avogadro's number, λ0 is the wavelength of the incident light in a vacuum, and dn / dc represents the relative refractive index increment of the solute. The molar mass of the BSA monomer served to evaluate the calibration constants of the light scattering and refractive index detectors during data collection (system suitability check). The relative standard deviation (RSD%) of the mean molar mass of BSA determined from the UV and RI detectors was less than 5.0%.

[0236] The light scattering detector normalization factor, interdetector delay, and band broadening term were calculated from the BSA chromatogram collected for the A4F-MALLS conditions used, and these values ​​were applied to the data files collected for all other samples to correct for these aspects.

[0237] The dn / dc values ​​and extinction coefficients at 215 nm were experimentally determined using the protein conjugate analysis provided by the Astra software. All protein-protein complex samples were analyzed using the corrected extinction coefficients and dn / dc values.

[0238] 7.4.3.Results A4F-MALLS was used to assess the relative size distribution of complexes formed between recombinant KLB (REGN6424), recombinant FGFR1c (REGN6152), and several monospecific (REGN4661), bispecific (REGN4304), and trispecific (2+1 N-scFv) binding molecules. The results are shown in Figure 11A (for REGN4661), Figure 11B (for REGN4304), and Figure 11C (2+1 N-scFv format). The theoretical molar masses and predicted stoichiometries of potential antibody:antigen complexes are provided as inserts in Figures 11A-11C. As expected, the monospecific KLB binding molecule (REGN4661) formed standard 1:1 (peak 1, approximately 280 kDa) and 1:2 (peak 2, approximately 356 kDa) complexes with KLB when combined in equimolar ratios (Figure 11A). Similarly, when a control bispecific binding molecule (anti-KLBxFGFR1c, REGN4304) was mixed with an equimolar amount of KLB, a discrete, homogeneous peak (Peak 1) with a calculated molar mass of approximately 280 kDa was observed (Figure 11B). Based on the calculated molar masses of the individual components, Peak 1 likely represents a 1:1 bispecific:KLB complex. Further addition of FGFR1c to this mixture resulted in a broad peak (Peak 2) with a calculated molar mass range of approximately 305-444 kDa, which is generally consistent with a 1:1:1 bispecific:KLB:FGFR1c ternary complex (Figure 11B). The upward trend in molar mass at the tail end of Peak 2 indicates that larger complexes, weakly associated via KLB-FGFR1c interactions, may also be present in solution but readily dissociate upon fractionation.

[0239] Compared to control monospecific and bispecific binding molecules, the trispecific binding molecules bound KLB and FGFR1c with a unique, higher-order stoichiometry. When mixed with an equimolar amount of KLB, F1K-scFv6 IgG1 formed a highly discrete, homogeneous peak (Peak 1) with a molar mass of approximately 579 kDa, likely representing a complex containing two molecules of F1K-scFv6 IgG1 bound to two molecules of KLB (2:2 complex, Figure 11C). After the addition of varying amounts of FGFR1c to this mixture, a slightly broader, later-eluting peak (Peak 2) was observed with a calculated molar mass range of approximately 607 to 644 kDa. Peak 2 likely represents a mixture of ternary complexes containing two molecules of F1K-scFv6 IgG1, two molecules of KLB, and one to two molecules of FGFR1c (2:2:1 and 2:2:2 complexes, Figure 11C).

[0240] Figure 12A shows the stoichiometry of FGF21 complexed with FGFR1c and KLB. Figure 12B shows various alternative stoichiometries of FGF21 antibody binding. The data presented herein demonstrate that the trispecific binding molecules of the present disclosure can bind to KLB and FGFR1c to form ternary complexes with unique stoichiometries compared to control monospecific and bispecific binding molecules, which may contribute to their increased agonist activity compared to bispecific binding molecules.

[0241] 7.5. Example 5: FGFR3 / APLP2 and FGFR3 / CD63 2+1 N-scFv MBM 7.5.1 Materials and Methods 7.5.1.1. Generating MBM To construct a 2+1 N-scFv MBM with monovalent scFvs fused to a bivalent FGFR3 mAb, 16 APLP2-targeting scFvs and 3 CD63-targeting scFvs were individually fused to the N-terminus of the heavy chain of the FGFR3 parent antibody 30108, which has a modified human IgG4 Fc backbone with effector silencing substitutions, hole (Y349C, T366S, L368A, Y407V, EU numbering), and star (H435R, Y436F, EU numbering) mutations (Table 11). The second heavy chain contains the same 30108 Fab and a modified IgG4 Fc with the same effector silencing substitution and knob (S354C, T366W, EU numbering) mutation (Table 11). The general format of the 2+1 N-scFv MBM is shown in Figure 1, and the FGFR3 / APLP2 and FGFR3 / CD63 2+1 N-scFv MBMs contain effector silencing substitutions, knob-in-hole mutations, and star mutations. The general format of the 2+1 N-scFv MBM with hole mutations in the scFv-containing chains and knob mutations in the chain lacking the scFv domain is shown in Figure 3C. The general format of the 2+1 N-scFv MBM with star mutations in the scFv-containing chains is shown in Figure 3A.

[0242] DNA fragments encoding various APLP2 or CD63 scFvs (i) in the orientation of VH (VH-44C, with Kabat numbering), linker (4xG4S (SEQ ID NO: 57)), VL (VL-100C mutation, with Kabat numbering), and linker (G4S)3 (SEQ ID NO: 1) connecting the scFv to the Fab, (ii) the Fab region of 30108, and (iii) modified human IgG4 Fc (Table 10) were synthesized by Integrated DNA Technologies, Inc. (San Diego, California). Individual heavy chain mammalian expression vectors were assembled using the NEBuilder HiFi DNA Assembly Kit (New England BioLabs Inc.). For expression of FGFR3 / APLP2 or CD63 2+1 N-scFv molecules, heavy chain 1-hole was used. *The heavy chain 2-knob and ULC 3-20 DNA were co-transfected into Expi293 cells (ThermoFisher Scientific) according to the manufacturer's protocol. 50 ml of cell culture medium was collected and processed for purification via a HiTrap Protein A FF column (GE Healthcare). For functional confirmation, 3ASB-5 and 3CSB-2 were scaled up to 200 ml and subjected to a series of purification steps, including size exclusion chromatography as the final step. Previously generated and purified 30108 IgG targeting FGFR3 was used as a control.

[0243] 7.5.1.2. Biacore Analysis for Target Binding Biacore kinetic analysis was performed to assess the binding affinity of the two MBMs, 3ASB-5 and 3CSB-2, to their respective targets.

[0244] 7.5.1.3. Proliferation Assay The effect of 2+1 N-scFv on bladder cancer cell proliferation was tested in UMUC14 cells expressing the S249C mutation and RT4 cells expressing the FGFR3-TACC3 fusion mutation.

[0245] 7.5.2.Results 7.5.2.1. Generation of FGFR3 / APLP2 and FGFR3 / CD63 2+1N-scFv MBMs The design outline of representative expressed and purified FGFR3 / APLP2 and FGFR3 / CD63 2+1 N-scFv MBMs is shown in Table 11.

[0246] [Table 15]

[0247] 7.5.2.2. FGFR3 / APLP2 or CD63 2+1 N-scFv can bind to FGFR3, APLP2, and CD63. The results of Biacore kinetic analysis for 3ASB-5 and 3CSB-2 against their respective targets are shown in Tables 12A-12C. Both 2+1 N-scFvs exhibited K D For 3ASB-5, binding of APLP2 was found to be similar to that of the parental 30108 IgG control, with a K = 8 nM. D = 8.04E-10 M, which is in the subnanomolar range. D The monovalent binding affinity of 3ASB-5 and 3CSB-2 was found to be 5.88E-10 M, four times weaker than that of the parent H4H12450N IgG. In summary, both 3ASB-5 and 3CSB-2 have the expected binding capacity for their targets, FGFR3, APLP2, and CD63.

[0248] [Table 16]

[0249] [Table 17]

[0250] [Table 18]

[0251] 7.5.2.3. FGFR3 / APLP2 2+1 N-scFv 3ASB-5 and FGFR3 / CD63 2+1 N-scFv 3CSB-2 can provide potent growth block in both UMUC14(S249C) and RT4(FGFR3-TACC3) cells via different mechanisms.

[0252] The effects of 2+1 N-scFv on bladder cancer cell proliferation were tested in UMUC14 cells expressing the S249C mutation and RT4 cells expressing the FGFR3-TACC3 fusion mutation. The parent antibody H4H30108P2 exhibited potent growth inhibition in RT4 cells but suboptimal growth inhibitory activity in UMUC14 cells. 3CSB-2 and 3ASB-5 exhibited significantly enhanced activity in UMUC14 cells, with growth inhibition increasing by approximately 25% at the highest dose tested (100 nM) (Figure 13A). Additionally, the activity of 2+1 N-scFv 3CSB-2 and 3ASB-5 remained at a high level in RT4 cells, comparable to that of the parent antibody H4H30108P2 (Figure 13B). Unlike the parental FGFR3 antibody, 3CSB-2 and 3ASB-5 were found to be able to block the growth of bladder cancer cells driven by different mutations (FIGS. 13A and 13B).

[0253] To further explore the underlying mechanisms of FGFR3 / APLP2 or FGFR3 / CD63 MBM, the effects of antibody-induced receptor degradation were examined in UMUC14 cells. Interestingly, distinct mechanisms were observed for FGFR3 / APLP2 versus FGFR3 / CD63 MBM. Treatment with FGFR3 / APLP2 MBM demonstrated enhanced levels of antibody-induced receptor degradation compared to cells untreated, or cells treated with the isotype control antibody (REGN1945) or the parental antibody H4H30108P2 (Figure 14). Conversely, treatment with FGFR3 / CD63 MBM did not affect FGFR3 receptor levels, indicating that the enhanced growth inhibitory activity of the bispecific 3CSB-2 MBM was due to distinct mechanisms (Figure 14).

[0254] 8. Specific Embodiments The present disclosure is illustrated by the following specific embodiments. 1. A multispecific binding molecule (MBM), comprising: (a) a first polypeptide chain comprising, in N-terminal to C-terminal orientation, an scFv comprising (i) a first antigen binding site (“ABS1”) operably linked to (ii) a first heavy chain region of a first Fab (“Fab1”), (iii) operably linked to an Fc domain; (b) a second polypeptide chain comprising, in N-terminal to C-terminal orientation, (i) a second heavy chain region of a second Fab (“Fab2”) operably linked to (ii) an Fc domain; and (c) a third polypeptide chain comprising a first light chain that pairs with the first heavy chain region to form Fab1, wherein Fab1 comprises a second antigen binding site (“ABS2”); and (d) a fourth polypeptide chain comprising a second light chain that pairs with a second heavy chain region to form Fab2, wherein Fab2 comprises a third antigen binding site (“ABS3”).

[0255] 2. The MBM of embodiment 1, wherein each antigen-binding site ("ABS") binds to a different epitope. 3. The MBM of embodiment 1 or embodiment 2, wherein two of ABS1, ABS2, and ABS3 specifically bind to different epitopes of the same target molecule.

[0256] 4. The MBM according to any one of embodiments 1 to 3, wherein the scFv, Fab1, and Fab2 are capable of simultaneously specifically binding to their respective targets. 5. The MBM of any one of embodiments 1-4, wherein at least one of ABS1, ABS2, and ABS3 specifically binds to a target molecule having a first tissue expression profile, and at least one of ABS1, ABS2, and ABS3 specifically binds to a target molecule having a second tissue expression profile that overlaps, but is not identical to, the first tissue expression profile.

[0257] 6. The MBM of embodiment 5, wherein the first tissue expression profile and the second tissue expression profile overlap by no more than 10 tissues. 7. The MBM of embodiment 6, wherein the first tissue expression profile and the second tissue expression profile overlap by no more than five tissues.

[0258] 8. The MBM of embodiment 7, wherein the first tissue expression profile and the second tissue expression profile overlap by no more than three tissues. 9. The MBM of any one of embodiments 6 to 8, wherein the first and second tissue expression profiles are defined by the Human Protein Atlas (HPA) and / or by the Genotype-Tissue Expression (GTEx) project.

[0259] 10. The MBM of any one of embodiments 6 to 9, wherein the first and second tissue expression profiles are protein expression profiles. 11. The MBM of any one of embodiments 6 to 9, wherein the first and second tissue expression profiles are mRNA expression profiles.

[0260] 12. An MBM according to any one of embodiments 1 to 11, wherein the affinity of the MBM for the target molecule of Fab1 is lower than the affinity of a second MBM for the target molecule of Fab1, which lacks the scFv of the MBM but is otherwise identical to the MBM in amino acid sequence.

[0261] 13. The MBM according to any one of embodiments 1 to 12, wherein the scFv is linked to the first heavy chain region via a linker. 14. The linker is at least 5 amino acids, at least 6 amino acids, or at least 7 amino acids in length, and optionally up to 30 amino acids, up to 40 amino acids, up to 50 amino acids, or up to 60 amino acids in length, and in some particular embodiments, the linker is (a) a length of 5 to 50 amino acids; (b) a length of 5 amino acids to 45 amino acids; (c) a length of 5 amino acids to 40 amino acids; (d) a length of 5 amino acids to 35 amino acids; (e) a length of 5 amino acids to 30 amino acids; (f) a length of 5 amino acids to 25 amino acids; (g) a length of 5 amino acids to 20 amino acids; (h) a length of 6 amino acids to 50 amino acids; (i) a length of 6 amino acids to 45 amino acids; (j) a length of 6 amino acids to 40 amino acids; (k) a length of 6 amino acids to 35 amino acids; (l) a length of 6 amino acids to 30 amino acids; (m) length of 6 amino acids to 25 amino acids; (n) length of 6 amino acids to 20 amino acids; (o) a length of 7 amino acids to 40 amino acids; (p) length of 7 amino acids to 35 amino acids; (q) length of 7 amino acids to 30 amino acids; (r) length of 7 amino acids to 25 amino acids; (s) 7 to 20 amino acids in length, or (t) The MBM of embodiment 13, which is between 10 amino acids and 60 amino acids in length.

[0262] 15. The MBM of embodiment 14 or 14(m), wherein the linker is 20 to 50 amino acids in length, and optionally the linker is 25 to 35 amino acids in length. 16. The linker is G n S (SEQ ID NO: 61) or SG n (SEQ ID NO: 62), wherein n is an integer from 1 to 7, and optionally the linker is or comprises a multimer of G4S (SEQ ID NO: 4).

[0263] 17. The MBM of any one of embodiments 13 to 16, wherein the linker is or comprises two consecutive glycines (2Gly), three consecutive glycines (3Gly), four consecutive glycines (4Gly (SEQ ID NO: 5)), five consecutive glycines (5Gly (SEQ ID NO: 6)), six consecutive glycines (6Gly (SEQ ID NO: 7)), seven consecutive glycines (7Gly (SEQ ID NO: 8)), eight consecutive glycines (8Gly (SEQ ID NO: 9)), or nine consecutive glycines (9Gly (SEQ ID NO: 10)).

[0264] 18. (a)G n S (SEQ ID NO: 61) or SG n (SEQ ID NO: 62) (e.g., multimers of G4S (SEQ ID NO: 4) where n is an integer from 1 to 7, including, for example, multimers of G4S (SEQ ID NO: 4)); (b) The MBM of any one of embodiments 13 to 17, comprising one or more additional glycines, such as two consecutive glycines (2Gly), three consecutive glycines (3Gly), four consecutive glycines (4Gly (SEQ ID NO: 5)), five consecutive glycines (5Gly (SEQ ID NO: 6)), six consecutive glycines (6Gly (SEQ ID NO: 7)), seven consecutive glycines (7Gly (SEQ ID NO: 8)), eight consecutive glycines (8Gly (SEQ ID NO: 9)), or nine consecutive glycines (9Gly (SEQ ID NO: 10)).

[0265] 19. The MBM of any one of embodiments 1 to 18, wherein at least one of ABS1, ABS2, and ABS3 specifically binds to a membrane-bound antigen. 20. The MBM of any one of embodiments 1-19, wherein ABS1 and ABS3 specifically bind to membrane-bound antigens on the same cell.

[0266] 21. The MBM of any one of embodiments 1-20, which is a trispecific binding molecule ("TBM"). 22. The MBM of any one of embodiments 1 to 21, wherein the first light chain and the second light chain are universal light chains.

[0267] 23. The MBM of any one of embodiments 1-22, wherein the light chain constant region and the first heavy chain constant region (CH1) of the first Fab or the second Fab are in a crossmab configuration.

[0268] 24. The MBM of any one of embodiments 1 to 23, comprising an Fc heterodimer. 25. The MBM of embodiment 24, wherein the Fc domain in the Fc heterodimer comprises knobs-in-holes mutations compared to the wild-type Fc domain.

[0269] 26. The MBM of embodiment 24 or embodiment 25, wherein at least one Fc domain in the Fc heterodimer comprises a star mutation compared to the wild-type Fc domain. 27. The MBM according to any one of embodiments 1 to 26, which is a trivalent MBM.

[0270] 28. The MBM of any one of embodiments 1-27, wherein ABS3 specifically binds to human Klotho beta ("KLB"). 29. The MBM of embodiment 28, wherein ABS3 comprises the CDR sequences of an anti-KLB antibody, for example the CDR sequences of any one of the KLB-binding agents shown in Tables 2A-2B.

[0271] 30.ABS3 is the V of anti-KLB antibody H and / or V L Sequences, e.g., V of any one of the KLB binders shown in Tables 2A-2B H and / or V L 30. The MBM of embodiment 29, comprising the sequence

[0272] 31. The MBM of any one of embodiments 1 to 30, wherein ABS1 specifically binds to human KLB. 32. The MBM of embodiment 31, wherein ABS1 comprises the CDR sequences of an anti-KLB antibody, such as the CDR sequences of any one of the KLB-binding agents shown in Tables 2A-2B.

[0273] 33.ABS1 is the V of anti-KLB antibody H and / or V L Sequences, e.g., V of any one of the KLB binders shown in Tables 2A-2B H and / or V L 33. The MBM of embodiment 32, comprising the sequence

[0274] 34. The MBM of any one of embodiments 1 to 33, wherein ABS1 and ABS3 specifically bind to different epitopes on human KLB. 35. The MBM of embodiment 34, wherein ABS1 and ABS3 are capable of simultaneously specifically binding to their respective epitopes on human KLB.

[0275] 36. The MBM of any one of embodiments 1-35, wherein ABS2 specifically binds to the human fibroblast growth factor receptor 1c isoform ("FGFR1c"). 37. The MBM of embodiment 36, wherein ABS2 comprises the CDR sequences of an anti-FGFR1c antibody, for example the CDR sequences of any one of the FGFR1c binding agents shown in Tables 3A-3B.

[0276] 38.ABS2 is the V of anti-FGFR1c antibody H and / or V L a sequence, e.g., V of any one of the FGFR1c binding agents shown in Tables 3A-3B H and / or V L 38. The MBM of embodiment 37, comprising the sequence

[0277] 39. The MBM according to any one of embodiments 36-38, wherein ABS2 binds to loop D3 of FGFR1c. 40. The MBM according to any one of embodiments 36-38, wherein ABS2 binds to loop D2 of FGFR1c.

[0278] 41. The MBM of any one of embodiments 1-27, wherein ABS2 specifically binds to human fibroblast growth factor receptor 3 ("FGFR3"). 42. The MBM of embodiment 41, wherein ABS2 comprises the CDR sequences of an anti-FGFR3 antibody, such as the CDR sequences of any one of the FGFR3-binding agents shown in Table 4.

[0279] 43.ABS2 is the V of the anti-FGFR3 antibody H and / or V L a sequence, e.g., the V of any one of the FGFR3-binding agents shown in Table 4 H and / or V L 43. The MBM of embodiment 42, comprising the sequence

[0280] 44. The MBM of any one of embodiments 1-27 or 41-43, wherein ABS3 specifically binds to human FGFR3. 45. The MBM of embodiment 44, wherein ABS3 comprises the CDR sequences of an anti-FGFR3 antibody, such as the CDR sequences of any one of the FGFR3-binding agents shown in Table 4.

[0281] 46.ABS3 is the V of the anti-FGFR3 antibody H and / or V L a sequence, e.g., the V of any one of the FGFR3-binding agents shown in Table 4 H and / or V L 46. ​​The MBM of embodiment 45, comprising the sequence

[0282] 47. The MBM of any one of embodiments 1-27, wherein ABS2 and ABS3 specifically bind to human FGFR3. 48. The MBM of embodiment 47, wherein ABS2 comprises the CDR sequences of an anti-FGFR3 antibody, such as the CDR sequences of any one of the FGFR3-binding agents shown in Table 4.

[0283] 49.ABS2 is the V of the anti-FGFR3 antibody H and / or V La sequence, e.g., the V of any one of the FGFR3-binding agents shown in Table 4 H and / or V L 49. The MBM of embodiment 48, comprising the sequence

[0284] 50. The MBM of any one of embodiments 47-49, wherein ABS3 comprises the CDR sequences of an anti-FGFR3 antibody, for example the CDR sequences of any one of the FGFR3 binding agents shown in Table 4.

[0285] 51.ABS3 is the V of the anti-FGFR3 antibody H and / or V L a sequence, e.g., the V of any one of the FGFR3-binding agents shown in Table 4 H and / or V L 51. The MBM of embodiment 50, comprising the sequence:

[0286] 52. The MBM of any one of embodiments 47-51, wherein ABS2 and ABS3 specifically bind to the same epitope on FGFR3. 53. The MBM of embodiment 52, wherein ABS2 and ABS3 comprise the same CDR sequences.

[0287] 54. ABS2 and ABS3 are the same V H and V L 54. The MBM of embodiment 53, comprising the sequence 55. An MBM according to any one of embodiments 41 to 54, wherein ABS1 specifically binds to human CD63.

[0288] 56. The MBM of embodiment 55, wherein ABS1 comprises the CDR sequences of an anti-CD63 antibody, such as the CDR sequences of any one of the CD63-binding agents shown in Table 6. 57.ABS1 is the V of anti-CD63 antibody H and / or V L a sequence, e.g., the V of any one of the CD63-binding agents shown in Table 6 H and / or V L 57. The MBM of embodiment 56, comprising the sequence

[0289] 58. The MBM of any one of embodiments 41-54, wherein ABS1 specifically binds to human amyloid precursor-like protein 2 (APLP2). 59. The MBM of embodiment 58, wherein ABS1 comprises the CDR sequences of an anti-APLP2 antibody, such as the CDR sequences of any one of the APLP2-binding agents shown in Table 5.

[0290] 60.ABS1 is the V of the anti-APLP2 antibody H and / or V L A sequence, e.g., the V of any one of the APLP2-binding agents shown in Table 5 H and / or V L 60. The MBM of embodiment 59, comprising the sequence

[0291] 61. A conjugate comprising an MBM according to any one of embodiments 1 to 27 and a cytotoxic or cytostatic agent. 62. A pharmaceutical composition comprising MBM according to any one of embodiments 1 to 27 or a conjugate according to embodiment 61 and an excipient.

[0292] 63. A method for treating cancer, comprising administering to a subject suffering from cancer an effective amount of the MBM described in any one of embodiments 1 to 27, the conjugate described in embodiment 61, or the pharmaceutical composition described in embodiment 62.

[0293] 64. The method of embodiment 63, wherein the cancer is associated with the expression of a target molecule to which ABS1, ABS2, and / or ABS3 specifically binds. 65. The method according to embodiment 63 or embodiment 64, wherein the MBM is an MBM according to any one of embodiments 41 to 60, or the conjugate comprises an MBM according to any one of embodiments 41 to 60, or the pharmaceutical composition comprises an MBM according to any one of embodiments 41 to 60 or a conjugate comprising an MBM according to any one of embodiments 41 to 60.

[0294] 66. The method of any one of embodiments 63 to 65, wherein the cancer is bladder cancer. 67. The method of any one of embodiments 63 to 65, wherein the cancer is glioblastoma. 68. The method of any one of embodiments 63 to 65, wherein the cancer is squamous cell lung cancer.

[0295] 69. A method for treating non-alcoholic steatohepatitis ("NASH"), comprising administering to a subject suffering from NASH an effective amount of the MBM of any one of embodiments 1 to 27, the conjugate of embodiment 61, or the pharmaceutical composition of embodiment 62.

[0296] 70. A method for treating a metabolic disease, comprising administering to a subject suffering from a metabolic disease an effective amount of the MBM described in any one of embodiments 1 to 27, the conjugate described in embodiment 61, or the pharmaceutical composition described in embodiment 62.

[0297] 71. A method for reducing circulating HDL cholesterol, comprising administering to a subject having elevated HDL levels an effective amount of the MBM of any one of embodiments 1 to 27, the conjugate of embodiment 61, or the pharmaceutical composition of embodiment 62.

[0298] 72. A method for increasing circulating LDL cholesterol, comprising administering to a subject suffering from low LDL levels an effective amount of the MBM described in any one of embodiments 1 to 27, the conjugate described in embodiment 61, or the pharmaceutical composition described in embodiment 62.

[0299] 73. A method for reducing blood triglycerides, comprising administering to a subject suffering from elevated triglyceride levels an effective amount of the MBM of any one of embodiments 1 to 27, the conjugate of embodiment 61, or the pharmaceutical composition of embodiment 62.

[0300] 74. A method for reducing blood glucose, comprising administering to a subject suffering from elevated glucose levels an effective amount of the MBM of any one of embodiments 1 to 27, the conjugate of embodiment 61, or the pharmaceutical composition of embodiment 62.

[0301] 75. A method for treating obesity, comprising administering to a subject suffering from obesity an effective amount of the MBM described in any one of embodiments 1 to 27, the conjugate described in embodiment 61, or the pharmaceutical composition described in embodiment 62.

[0302] 76. A method for treating diabetes, comprising administering to a subject suffering from diabetes an effective amount of the MBM described in any one of embodiments 1 to 27, the conjugate described in embodiment 61, or the pharmaceutical composition described in embodiment 62.

[0303] 77. The method according to any one of embodiments 69 to 76, wherein the MBM is an MBM according to any one of embodiments 28 to 40, or the conjugate comprises an MBM according to any one of embodiments 28 to 40, or the pharmaceutical composition comprises an MBM according to any one of embodiments 28 to 40 or a conjugate comprising an MBM according to any one of embodiments 28 to 40.

[0304] 78. A nucleic acid or a plurality of nucleic acids encoding an MBM according to any one of the embodiments according to any one of embodiments 1 to 60. 79. A cell engineered to express the MBM of any one of embodiments 1 to 60.

[0305] 80. A cell transfected with one or more expression vectors comprising one or more nucleic acid sequences encoding an MBM according to any one of embodiments 1 to 60 under the control of one or more promoters.

[0306] 81. A method for producing MBM, comprising: (a) culturing the cells of embodiment 79 or 80 under conditions in which MBM is expressed; (b) recovering the MBM from the cell culture.

[0307] 82. The method of embodiment 81, further comprising enriching the MBM. 83. The method of embodiment 81 or embodiment 82, further comprising purifying the MBM.

[0308] 9. Citation of References All publications, patents, patent applications, and other documents cited in this application are incorporated herein by reference in their entirety for all purposes to the same extent as if each individual publication, patent, patent application, or other document was individually indicated to be incorporated by reference for all purposes. In the event of a conflict between the teachings of this specification and one or more of the references incorporated into this disclosure, the teachings of this specification are intended.

Claims

1. A multispecific binding molecule (MBM), comprising: (a) in an N-terminal to C-terminal orientation, (i) an scFv comprising a first antigen-binding site (“ABS1”); (ii) the first heavy chain region of the first Fab (“Fab1”), and (iii) an Fc domain; a first polypeptide chain, wherein the scFv is operably linked to the first heavy chain region of Fab1, and the first heavy chain region of Fab1 is operably linked to the Fc domain; (b) in an N-terminal to C-terminal orientation, (i) the second heavy chain region of the second Fab (“Fab2”), and (ii) an Fc domain; a second polypeptide chain, wherein the second heavy chain region of Fab2 is operably linked to the Fc domain; (c) a third polypeptide chain comprising a first light chain that pairs with the first heavy chain region to form Fab1, wherein Fab1 comprises a second antigen binding site (“ABS2”); and (d) a fourth polypeptide chain comprising a second light chain that pairs with the second heavy chain region to form Fab2, wherein Fab2 comprises a third antigen binding site (“ABS3”); MBMs in which ABS2 specifically binds to human fibroblast growth factor receptor 1c isoform ("FGFR1c"), and ABS1 and ABS3 specifically bind to different epitopes on human Klotho beta ("KLB").

2. 2. The MBM of claim 1, wherein the affinity of said MBM for FGFRlc is lower than the affinity of a second MBM for FGFRlc that lacks the scFv of said MBM but is otherwise identical to said MBM in amino acid sequence.

3. ABS3 is the CDR sequence of the anti-KLB antibody, V Ｈ sequence, and / or V Ｌ 3. The MBM of claim 1 or 2, comprising the sequence:

4. ABS1 is the CDR sequence of the anti-KLB antibody, V Ｈ sequence, and / or V Ｌ The MBM according to any one of claims 1 to 3, comprising the sequence:

5. The MBM according to any one of claims 1 to 4, wherein ABS1 and ABS3 are capable of simultaneously specifically binding to their respective epitopes on human KLB.

6. ABS2 is the CDR sequence of the anti-FGFR1c antibody, V Ｈ sequence, and / or V Ｌ The MBM of any one of claims 1 to 5, comprising the sequence:

7. The MBM of any one of claims 1 to 6, wherein ABS2 binds to loop D3 of FGFR1c.

8. The MBM of any one of claims 1 to 6, wherein ABS2 binds to loop D2 of FGFR1c.

9. The MBM of any one of claims 1 to 8, wherein the scFv is linked to the first heavy chain region via a linker.

10. The linker is (a) at least 5 amino acids, at least 6 amino acids, or at least 7 amino acids in length; (b) is at most 30 amino acids, at most 40 amino acids, at most 50 amino acids, or at most 60 amino acids in length; (c) a length of 5 to 50 amino acids; (d) a length of 5 to 45 amino acids; (e) a length of 5 to 40 amino acids; (f) a length of 5 amino acids to 35 amino acids; (g) a length of 5 to 30 amino acids; (h) a length of 5 to 25 amino acids; (i) a length of 5 to 20 amino acids; (j) a length of 6 amino acids to 50 amino acids; (k) a length of 6 amino acids to 45 amino acids; (l) a length of 6 amino acids to 40 amino acids; (m) a length of 6 to 35 amino acids; (n) a length of 6 amino acids to 30 amino acids; (o) a length of 6 amino acids to 25 amino acids; (p) a length of 6 to 20 amino acids; (q) a length of 7 amino acids to 40 amino acids; (r) a length of 7 to 35 amino acids; (s) a length of 7 to 30 amino acids; (t) a length of 7 to 25 amino acids, or The MBM of claim 9, wherein (u) the MBM is 7 to 20 amino acids in length.

11. 10. The MBM of claim 9, wherein the linker is 5 to 25 amino acids in length, 10 to 60 amino acids in length, 20 to 50 amino acids in length, or 25 to 35 amino acids in length.

12. The linker is G ｎ S (SEQ ID NO: 61) or SG ｎ 12. The MBM of any one of claims 9 to 11, which is or comprises a multimer of (SEQ ID NO: 62), wherein n is an integer from 1 to 7.

13. The linker is G ４ 13. The MBM of claim 12, which is or comprises a multimer of S (SEQ ID NO: 4).

14. 14. The MBM of any one of claims 9 to 13, wherein the linker is or comprises two consecutive glycines (2Gly), three consecutive glycines (3Gly), four consecutive glycines (4Gly (SEQ ID NO: 5)), five consecutive glycines (5Gly (SEQ ID NO: 6)), six consecutive glycines (6Gly (SEQ ID NO: 7)), seven consecutive glycines (7Gly (SEQ ID NO: 8)), eight consecutive glycines (8Gly (SEQ ID NO: 9)), or nine consecutive glycines (9Gly (SEQ ID NO: 10)).

15. The MBM of any one of claims 1 to 14, wherein the first light chain and the second light chain are universal light chains.

16. The MBM of any one of claims 1 to 14, wherein the light chain constant region and the first heavy chain constant region (CH1) of the first Fab or the second Fab are in a crossmab configuration.

17. The MBM of any one of claims 1 to 16, comprising an Fc heterodimer.

18. 18. The MBM of claim 17, wherein the Fc domain in the Fc heterodimer comprises a knobs-in-holes mutation compared to the wild-type Fc domain.

19. 18. The MBM of claim 17, wherein at least one Fc domain in the Fc heterodimer comprises a star mutation compared to the wild-type Fc domain.

20. A conjugate comprising the MBM of any one of claims 1 to 19 and a cytotoxic or cytostatic agent.

21. A pharmaceutical composition comprising the MBM of any one of claims 1 to 19 or the conjugate of claim 20 and an excipient.

22. 20. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for treating non-alcoholic steatohepatitis ("NASH"), comprising administering an effective amount of MBM to a subject suffering from NASH.

23. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for treating a metabolic disease, comprising administering an effective amount of MBM to a subject suffering from the metabolic disease.

24. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for reducing circulating HDL cholesterol comprising administering an effective amount of MBM to a subject with elevated HDL levels.

25. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for increasing circulating LDL cholesterol, comprising administering an effective amount of MBM to a subject suffering from low LDL levels.

26. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for reducing blood triglycerides, comprising administering an effective amount of MBM to a subject suffering from elevated triglyceride levels.

27. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for reducing blood glucose comprising administering an effective amount of MBM to a subject suffering from elevated glucose levels.

28. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for treating obesity, comprising administering an effective amount of MBM to a subject suffering from obesity.

29. A pharmaceutical composition comprising as an active ingredient the MBM of any one of claims 1 to 19 for use in a method for treating diabetes, comprising administering an effective amount of MBM to a subject suffering from diabetes.

30. A nucleic acid or a plurality of nucleic acids encoding the MBM of any one of claims 1 to 19.

31. A cell engineered to express the MBM of any one of claims 1 to 19.

32. A cell transfected with one or more expression vectors comprising one or more nucleic acid sequences encoding the MBM of any one of claims 1 to 19 under the control of one or more promoters.

33. 1. A method for producing MBM, comprising: (a) culturing the cell of claim 31 or 32 under conditions in which the MBM is expressed; (b) recovering the MBM from the cell culture.

Images