Rational selection of building blocks for constructing multispecific antibodies

Abstract

The ability to generate a single antibody-based construct that can simultaneously recognize multiple targets is paramount for advancing many therapeutic candidates to the clinic. This often means extensive protein design, with varying degrees of success. For multispecific antibody constructs, there are multiple modules to choose from, and often multiple antigen-binding moieties as well. Described herein is the discovery of a new method for optimally pairing antigen-binding moieties in the appropriate configuration, including the selection of a common light chain.

Description

[Technical Field]

[0001] This invention relates to the technology of biopharmaceuticals. In particular, the invention relates to a method for producing a multispecific antibody construct and a method for selecting the optimal multispecificity module for such a construct. [Background technology]

[0002] Bispecific antibodies represent an intriguing new generation of macromolecular therapeutics in a field currently dominated by monoclonal antibodies (mAbs). 1、2 The defining characteristic of bispecific mAbs is their ability to recognize two epitopes located on the same or different targets. This dual recognition capability extends the functionality of conventional mAbs, enabling diverse applications such as mobilizing immune cells to destroy tumor cells, crosslinking distinct cell surface receptors, or enhancing tissue specificity. 1、3 For example, Amgen's bispecific T cell engager (BiTE®) interacts with CD3 epitopes on the T cell surface and tumor-associated antigens. 4、5 It binds to both T cells and effectively acts as a bridge connecting immunologically active T cells and target tumor cells. To date, more than 100 bispecific forms have been reported, more than 85 are in development, and 3 have received FDA approval. 1、6、7 Generally, bispecific molecules can be classified into three categories: i) fragment fusions (e.g., tandem scFv and DART), ii) IgG fusions (e.g., IgG-scFv and DVD-Ig), and iii) IgG-like molecules (e.g., hetero-Fc). 7、8Fragment fusions and IgG fusions exhibit simple engineered constructs (only one or two polypeptide chains) that are advantageous for purification and stable cell line generation, but these forms often have low yields and undesirable stability profiles. In contrast, IgG-like bispecifics (e.g., heterologous Fc) that mimic the native structure of the IgG molecule have higher stability and better cell production. Furthermore, they are likely one of the most representative bispecific formats in clinical trials due to their good half-life profile in serum and low potential immunogenicity. 1 However, in these formats, due to the large number of chains (3 - 4), mispaired species can become significant contaminants and require multiple purification steps to remove them, which can greatly reduce the final purification yield.

[0003] In engineered IgG-like bispecific antibodies, multiple heavy chains (HCs) and light chains (LCs) are assembled into a single molecule to enable the recognition of two different epitopes. Thus, the challenge in developing IgG-like bispecifics is to ensure correct chain pairing. In the case of four-chain heterologous Fc, co-expression of these chains in the same cell can result in nine possible combinations of mispaired IgG species. 9、10 In the past few years, several strategies including knob-into-hole 11 , chain exchange engineering domains 12 and charge pair mutagenesis (CPM) 13、14 have been developed to address the HC / HC and HC / LC pairing problems. In HC / LC engineering, in most cases, the rationale is to engineer the chain interface to favor homologous over non-homologous HC / LC pairing.However, despite the best engineering efforts, sequence diversity (complementary determining regions (CDRs), frameworks, and LC isotypes) often limits the success when applying these engineering tools to strict platforms for HC / LC pairing.

[0004] To overcome these difficulties, the use of common light chains (cLCs) is appealing because it avoids the need to drive pairing between specific HCs and LCs. However, identifying cLCs that maintain the desired binding profile to different epitopes when paired with different HCs is challenging and often requires significant investment early in the drug development process. 15 Generally, two methods are most commonly used to discover antibodies containing cLCs. The first involves screening display libraries consisting of diverse HC sequences but containing only one or a few LCs. Alternatively, mice expressing universal LCs are immunized for each of the desired targets. 16、17 Since both methods limit the available LCs, these cLC antibodies often exhibit suboptimal binding affinities, requiring extensive manipulation, primarily in the HC, to optimize target affinity. In contrast, with typical mAbs, both HC and LC can be targeted for optimization. Furthermore, the structure of the antibody-antigen complex reveals that many antibody / epitope interactions are HC-driven, and rarely, LC may not engage in any productive interactions. 18 Therefore, some LCs may be suitable for pairing with non-congenital HCs while maintaining target binding affinity. cLC heteroFc assembled with such LCs, along with congenital and non-congenital HCs, retains the innate affinity in the congenital HC / LC arm, thus requiring less optimization.

[0005] The development of therapeutic antibodies typically begins with immunization of humanized animal models with selected antigens. 19、20 This leads to the identification and isolation of lead mAbs. These mAbs are selected to meet design objectives such as target specificity, binding affinity, interspecies reactivity, yield, stability, and non-immunogenicity, but little is known about the properties required to serve as building blocks for bispecifics. Often, this requires hundreds of bispecific empirical tests to evaluate all parent mAb combinations, leading to increased timelines and resources.

[0006] Therefore, two high-throughput screening methods are needed to facilitate the rational selection of lead mAbs for the fabrication of bispecific and multispecific compounds. This application describes competitive and non-competitive chain selectivity evaluations (CSAs). Competitive CSA (cCSA) is used to select mAbs in which HCs and LCs are effectively assembled into hetero-Fc molecules with little to no cross-pairing between the two Fabs. Non-competitive CSA (ncCSA) is a powerful tool for identifying cLCs in a cost-effective manner. [Prior art documents] [Non-patent literature]

[0007] [Non-Patent Document 1] Labrijn, AF, Janmaat, ML, Reichert, JM & Parren, P. Bispecific antibodies: a mechanistic review of the pipeline.Nat Rev Drug Discov 18,585-608(2019) [Non-Patent Document 2] Lu,RMet al.Development of therapeutic antibodies for the treatment of diseases.J Biomed Sci 27,1(2020) [Non-Patent Document 3] Fan, G., Wang, Z., Hao, M. & Li, J. Bispecific antibodies and their applications.J Hematol Oncol 8,130(2015) [Non-Patent Document 4] Wolf, E., Hofmeister, R., Kufer, P., Schlereth, B. & Baeuerle, PABiTEs: bispecific antibody constructs with unique anti-tumor activity. Drug Discov Today 10, 1237-1244 (2005) [Non-Patent Document 5] Kantarjian,H.et al.Blinatumomab versus Chemotherapy for Advanced Acute Lymphoblastic Leukemia.N Engl J Med 376,836-847(2017) [Non-Patent Document 6] Brinkmann, U. & Kontermann, REThe making of bispecific antibodies.MAbs 9,182-212(2017) [Non-Patent Document 7] Wang,Q.et al.Design and Production of Bispecific Antibodies.Antibodies(Basel)8(2019) [Non-Patent Document 8] Spiess, C., Zhai, Q. & Carter, PJAlternative molecular formats and therapeutic applications for bispecific antibodies.Mol Immunol 67,95-106(2015) [Non-Patent Document 9] Suresh, MR, Cuello, AC & Milstein, C. Bispecific monoclonal antibodies from hybrid hybridomas.Methods Enzymol 121,210-228(1986) [Non-Patent Document 10] Carter,P.Bispecific human IgG by design.J Immunol Methods 248,7-15(2001) [Non-Patent Document 11] Ridgway,JB,Presta,LG& Carter,P.'Knobs-into-holes'engineering of antibody CH3 domains for heavy chain heterodimerization.Protein Eng 9,617-621(1996) [Non-Patent Document 12] Davis,JHet al.SEEDbodies:fusion proteins based on strand-exchange engineered domain(SEED)CH3 heterodimers in an Fc analogue platform for asymmetric binders or immunofusions and bispecific antibodies.Protein Eng Des Sel 23,195-202(2010) [Non-Patent Document 13] Gunasekaran,K.et al.Enhancing antibody Fc heterodimer formation through electrostatic steering effects:applications to bispecific molecules and monovalent IgG.J Biol Chem 285,19637-19646(2010) [Non-Patent Document 14] Dillon,M.et al.Efficient production of bispecific IgG of different isotypes and species of origin in single mammalian cells.MAbs 9,213-230(2017) [Non-Patent Document 15] Shiraiwa,H.et al.Engineering a bispecific antibody with a common light chain:Identification and optimization of an anti-CD3 epsilon and anti-GPC3 bispecific antibody,ERY974.Methods 154,10-20(2019) [Non-Patent Document 16] Merchant,AMet al.An efficient route to human bispecific IgG.Nat Biotechnol 16,677-681(1998) [Non-Patent Document 17] Krah,S.et al.Generation of human bispecific common light chain antibodies by combining animal immunization and yeast display.Protein Eng Des Sel 30,291-301(2017) [Non-Patent Document 18] Garces,F.et al.Molecular Insight into Recognition of the CGRPR Complex by Migraine Prevention Therapy Aimovig(Erenumab).Cell Rep 30,1714-1723 e1716(2020) [Non-Patent Document 19] Bruggemann,M.et al.Human antibody production in transgenic animals.Arch Immunol Ther Exp(Warsz)63,101-108(2015) [Non-Patent Document 20] Foltz,IN,Gunasekaran,K.& King,CTDiscovery and bio-optimization of human antibody therapeutics using the XenoMouse(R)transgenic mouse platform.Immunol Rev 270,51-64(2016) [Overview of the project] [Means for solving the problem]

[0008] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0009] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0010] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0011] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0012] In one embodiment, the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, SV40-transformed monkey kidney CV1 cell line ("COS-7"), human fetal kidney cell line 293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

[0013] In one embodiment, the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

[0014] In one embodiment, the binding affinity of each multispecific antibody construct to the first and second antigens is measured using Octet, Forte Bio, Carterra LSA, SPR, and flow cytometry.

[0015] In one embodiment, each multispecific antibody construct is purified by protein A, lambda and kappa resin, as well as affinity tag purification.

[0016] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody construct modules, each containing a heavy chain CDR and a light chain CDR that binds to a specific antigen (c)(iii); (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f)(i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen using Octet. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0017] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0018] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) A step of cloning two or more CDRs into a vector encoding a multispecific antibody construct module, wherein multiple vectors are generated that encode multiple multispecific antibody constructs, each containing a CDR that binds to a different antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species generated. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) For each multispecific antibody construct, compare the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) in order to identify the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen and the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen. This applies to methods that include [specific methods].

[0019] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0020] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0021] In one embodiment, the multispecific antibody construct module is selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0022] In one embodiment, the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, SV40-transformed monkey kidney CV1 cell line ("COS-7"), human fetal kidney cell line 293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

[0023] In one embodiment, the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

[0024] In one embodiment, the proportion of correct and incorrect multispecific antibody construct module species is determined by a method selected from the group consisting of liquid chromatography-mass spectrometry ("LC-MS"), Caliper, HPLC SEC, SDS-PAGE, and microchip capillary electrophoresis ("MCE").

[0025] In one embodiment, each multispecific antibody construct is purified by protein A, lambda and kappa resin, as well as affinity tag purification.

[0026] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) A process in which two or more CDRs are cloned into vectors encoding multispecific antibody construct modules, thereby generating multiple vectors encoding multiple multispecific antibody constructs that bind to each antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species using liquid chromatography-mass spectrometry ("LC-MS"). (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) For each multispecific antibody construct, compare the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) in order to identify the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen and the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen. This applies to methods that include [specific methods].

[0027] In one embodiment, the multispecific antibody construct module is selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0028] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs of different modules, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct of different modules in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal module for pairing three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal module for pairing three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0029] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0030] In one embodiment, the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, SV40-transformed monkey kidney CV1 cell line ("COS-7"), human fetal kidney cell line 293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

[0031] In one embodiment, the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

[0032] In one embodiment, the binding affinity of each multispecific antibody construct to the first and second antigens is measured using Octet, Forte Bio, Carterra LSA, SPR, and flow cytometry.

[0033] In one embodiment, each multispecific antibody construct is purified by protein A, lambda and kappa resin, as well as affinity tag purification.

[0034] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs of different modules, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f)(i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen using Octet. (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal module for pairing three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal module for pairing three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0035] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0036] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) A step of cloning the CDR of the first antibody Fab fragment or scFv and the second antibody Fab fragment or scFv into a vector, The process involves a vector encoding two or more types of multispecific antibody construct modules, resulting in the generation of multiple vectors encoding multiple multispecific antibody constructs of different modules, each containing a CDR that binds to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species produced. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to identify the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen, and the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen, by comparing the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) for each multispecific antibody construct. This applies to methods that include [specific methods].

[0037] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0038] In one embodiment, the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, SV40-transformed monkey kidney CV1 cell line ("COS-7"), human fetal kidney cell line 293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

[0039] In one embodiment, the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

[0040] In one embodiment, the proportion of correct and incorrect multispecific antibody construct module species is determined by a method selected from the group consisting of liquid chromatography-mass spectrometry ("LC-MS"), Caliper, HPLC SEC, SDS-PAGE, and microchip capillary electrophoresis ("MCE").

[0041] In one embodiment, each multispecific antibody construct is purified by protein A, lambda and kappa resin, as well as affinity tag purification.

[0042] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) A step of cloning the CDR of the first antibody Fab fragment or scFv and the second antibody Fab fragment or scFv into a vector, The process involves a vector encoding two or more types of multispecific antibody construct modules, resulting in the generation of multiple vectors encoding multiple multispecific antibody constructs of different modules, each containing a CDR that binds to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species using liquid chromatography-mass spectrometry ("LC-MS"). (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to identify the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen, and the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen, by comparing the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) for each multispecific antibody construct. This applies to methods that include [specific methods].

[0043] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc. [Brief explanation of the drawing]

[0044] [Figure 1A] Figure 1 shows the structural analysis of the HC-LC interface. A) Schematic diagram of the IgG structural arrangement and surface representation of the Fv region. In the blue-colored Fab arm, the Fv(VH / VL) interface, CH1 / CL interface, and CDR region are enclosed by dashed lines. The surfaces of the VH and VL domains of IgG (PDB:1HZH) were prepared using PyMol with color codes for each CDR. The VH and VL interface is highlighted with a dashed line. B) Calculation of HC / LC interface residues using six structures performed by PDBePISA29. The number of interface residues in the corresponding regions is listed. C) Sequence alignment of 508 human antibodies with structures deposited in the Protein Data Bank (PDB) highlights the diversity of CDRs. [Figure 1B]Figure 1 shows the structural analysis of the HC-LC interface. A) Schematic diagram of the IgG structural arrangement and surface representation of the Fv region. In the blue-colored Fab arm, the Fv(VH / VL) interface, CH1 / CL interface, and CDR region are enclosed by dashed lines. The surfaces of the VH and VL domains of IgG (PDB:1HZH) were prepared using PyMol with color codes for each CDR. The VH and VL interface is highlighted with a dashed line. B) Calculation of HC / LC interface residues using six structures performed by PDBePISA29. The number of interface residues in the corresponding regions is listed. C) Sequence alignment of 508 human antibodies with structures deposited in the Protein Data Bank (PDB) highlights the diversity of CDRs. [Figure 1C] Figure 1 shows the structural analysis of the HC-LC interface. A) Schematic diagram of the IgG structural arrangement and surface representation of the Fv region. In the blue-colored Fab arm, the Fv(VH / VL) interface, CH1 / CL interface, and CDR region are enclosed by dashed lines. The surfaces of the VH and VL domains of IgG (PDB:1HZH) were prepared using PyMol with color codes for each CDR. The VH and VL interface is highlighted with a dashed line. B) Calculation of HC / LC interface residues using six structures performed by PDBePISA29. The number of interface residues in the corresponding regions is listed. C) Sequence alignment of 508 human antibodies with structures deposited in the Protein Data Bank (PDB) highlights the diversity of CDRs. [Figure 2A]Figure 2 shows a schematic diagram of the design of the high-throughput chain selection (CSA) method. A) Schematic diagram of high-throughput CSA. Two panels of parental mAbs against target A (blue) and target B (green) are subjected to high-throughput screening using two possible products, quadruple-stranded hetero-Fc and cLC hetero-Fc. B) Outline and timeline of the two high-throughput CSA methods. In the competitive CSA (cCSA) experiment, each anti-target A mAb is combined with all anti-target B mAbs. To ensure HC heterodimerization, CPMs, indicated by red and blue dots, were manipulated in the CH3 domain. Thus, four HC / LC pairing scenarios remain as innate properties within the chain interface. All combinations of two HCs and two LCs are co-transfected into HEK293-6E cells, followed by ProA purification and LC-MS quantification of each IgG species. Samples with a high proportion of correct MW IgG species indicate that the corresponding antibody combination exhibits high preference for congenital HC / LC pairings while minimizing cross-pairing. In non-competitive CSA (ncCSA) experiments, the expression of non-congenital HC / LC pairs in HEK293-6E cells is tested, followed by ProA purification and determination of protein concentration (A280). High levels of hybrid IgG expression suggest that indiscriminate LCs, which should be selected for cLC heterozygotes, assemble together with congenital and non-congenital HCs. To identify positive binders, purified hybrid IgG is further analyzed by ForteBio Octet HTX. [Figure 2B]Figure 2 shows a schematic diagram of the design of the high-throughput chain selection (CSA) method. A) Schematic diagram of high-throughput CSA. Two panels of parental mAbs against target A (blue) and target B (green) are subjected to high-throughput screening using two possible products, quadruple-stranded hetero-Fc and cLC hetero-Fc. B) Outline and timeline of the two high-throughput CSA methods. In the competitive CSA (cCSA) experiment, each anti-target A mAb is combined with all anti-target B mAbs. To ensure HC heterodimerization, CPMs, indicated by red and blue dots, were manipulated in the CH3 domain. Thus, four HC / LC pairing scenarios remain as innate properties within the chain interface. All combinations of two HCs and two LCs are co-transfected into HEK293-6E cells, followed by ProA purification and LC-MS quantification of each IgG species. Samples with a high proportion of correct MW IgG species indicate that the corresponding antibody combination exhibits high preference for congenital HC / LC pairings while minimizing cross-pairing. In non-competitive CSA (ncCSA) experiments, the expression of non-congenital HC / LC pairs in HEK293-6E cells is tested, followed by ProA purification and determination of protein concentration (A280). High levels of hybrid IgG expression suggest that indiscriminate LCs, which should be selected for cLC heterozygotes, assemble together with congenital and non-congenital HCs. To identify positive binders, purified hybrid IgG is further analyzed by ForteBio Octet HTX. [Figure 3A]Figure 3 shows high-throughput screening of low cross-pairing antibody combinations in cCSA experiments. A) Analysis of protein expression levels. Twelve parental mAbs (eight anti-target-A and four anti-target-B) and the resulting 32 hetero-Fc combinations were transiently expressed in HEK293-6E cells. After purifying the molecules in the conditioned medium using a high-throughput KingFisher Flex system, protein concentrations were measured (A280). Expression levels are shown by ProA yield calculated in milligrams (mg) per liter of conditioned medium. B) Representative high-resolution LC-MS analysis of two highlighted ProA purified samples (A3×B4 and A4×B2) (including schematic diagrams showing correct and mispaired IgG species). C) The percentage of IgG species with hetero-HC was calculated from LC-MS data and plotted. The percentage of IgG species with correct MW (1×HC1+1×LC1+1×HC2+1×LC2) reflects correct HC / LC pairing. * The MW difference is small, making it impossible to identify IgG species by LC-MS. [Figure 3B] Figure 3 shows high-throughput screening of low cross-pairing antibody combinations in cCSA experiments. A) Analysis of protein expression levels. Twelve parental mAbs (eight anti-target-A and four anti-target-B) and the resulting 32 hetero-Fc combinations were transiently expressed in HEK293-6E cells. After purifying the molecules in the conditioned medium using a high-throughput KingFisher Flex system, protein concentrations were measured (A280). Expression levels are shown by ProA yield calculated in milligrams (mg) per liter of conditioned medium. B) Representative high-resolution LC-MS analysis of two highlighted ProA purified samples (A3×B4 and A4×B2) (including schematic diagrams showing correct and mispaired IgG species). C) The percentage of IgG species with hetero-HC was calculated from LC-MS data and plotted. The percentage of IgG species with correct MW (1×HC1+1×LC1+1×HC2+1×LC2) reflects correct HC / LC pairing. * The MW difference is small, making it impossible to identify IgG species by LC-MS. [Figure 3C]Figure 3 shows high-throughput screening of low cross-pairing antibody combinations in cCSA experiments. A) Analysis of protein expression levels. Twelve parental mAbs (eight anti-target-A and four anti-target-B) and the resulting 32 hetero-Fc combinations were transiently expressed in HEK293-6E cells. After purifying the molecules in the conditioned medium using a high-throughput KingFisher Flex system, protein concentrations were measured (A280). Expression levels are shown by ProA yield calculated in milligrams (mg) per liter of conditioned medium. B) Representative high-resolution LC-MS analysis of two highlighted ProA purified samples (A3×B4 and A4×B2) (including schematic diagrams showing correct and mispaired IgG species). C) The percentage of IgG species with hetero-HC was calculated from LC-MS data and plotted. The percentage of IgG species with correct MW (1×HC1+1×LC1+1×HC2+1×LC2) reflects correct HC / LC pairing. * The MW difference is small, making it impossible to identify IgG species by LC-MS. [Figure 4A] Figure 4 shows the predictability of HC / LC pairing in quadruple-stranded hetero-Fc molecules. A) Final yield observed after CIEX purification with a purity target of over 90%. Yields of hetero-Fc molecules and their corresponding parent mAbs are plotted side by side. B) Receiver operating characteristic curve (ROC) plot of CIEX yield and percentage of correct IgG species. C) Correlation analysis of CIEX yield and HC / LC pairing. The dashed line shows the 50% benchmark for correct IgG species determined by LC-MS. D) CIEX chromatographs of two representative molecules. E) Breakdown of the number of molecules at each stage of the cCSA experiment. [Figure 4B] Figure 4 shows the predictability of HC / LC pairing in quadruple-stranded hetero-Fc molecules. A) Final yield observed after CIEX purification with a purity target of over 90%. Yields of hetero-Fc molecules and their corresponding parent mAbs are plotted side by side. B) Receiver operating characteristic curve (ROC) plot of CIEX yield and percentage of correct IgG species. C) Correlation analysis of CIEX yield and HC / LC pairing. The dashed line shows the 50% benchmark for correct IgG species determined by LC-MS. D) CIEX chromatographs of two representative molecules. E) Breakdown of the number of molecules at each stage of the cCSA experiment. [Figure 4C] Figure 4 shows the predictability of HC / LC pairing in quadruple-stranded hetero-Fc molecules. A) Final yield observed after CIEX purification with a purity target of over 90%. Yields of hetero-Fc molecules and their corresponding parent mAbs are plotted side by side. B) Receiver operating characteristic curve (ROC) plot of CIEX yield and percentage of correct IgG species. C) Correlation analysis of CIEX yield and HC / LC pairing. The dashed line shows the 50% benchmark for correct IgG species determined by LC-MS. D) CIEX chromatographs of two representative molecules. E) Breakdown of the number of molecules at each stage of the cCSA experiment. [Figure 4D] Figure 4 shows the predictability of HC / LC pairing in quadruple-stranded hetero-Fc molecules. A) Final yield observed after CIEX purification with a purity target of over 90%. Yields of hetero-Fc molecules and their corresponding parent mAbs are plotted side by side. B) Receiver operating characteristic curve (ROC) plot of CIEX yield and percentage of correct IgG species. C) Correlation analysis of CIEX yield and HC / LC pairing. The dashed line shows the 50% benchmark for correct IgG species determined by LC-MS. D) CIEX chromatographs of two representative molecules. E) Breakdown of the number of molecules at each stage of the cCSA experiment. [Figure 4E] Figure 4 shows the predictability of HC / LC pairing in quadruple-stranded hetero-Fc molecules. A) Final yield observed after CIEX purification with a purity target of over 90%. Yields of hetero-Fc molecules and their corresponding parent mAbs are plotted side by side. B) Receiver operating characteristic curve (ROC) plot of CIEX yield and percentage of correct IgG species. C) Correlation analysis of CIEX yield and HC / LC pairing. The dashed line shows the 50% benchmark for correct IgG species determined by LC-MS. D) CIEX chromatographs of two representative molecules. E) Breakdown of the number of molecules at each stage of the cCSA experiment. [Figure 5A]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5B]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5C]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5D]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5E]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5F]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 5G]Figure 5 shows high-throughput screening of cLC heterozygotes by ncCSA. A) Relative expression of 144 non-homogeneous HC / LC pairs in two bispecificity panels (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control. Non-homogeneous HC / LC pairs with relative expression levels greater than 0.5 suggest that their LCs are indiscriminate towards the non-homogeneous HCs they pair with and are selected to construct cLC heterozygotes together with the homogeneous HCs. B) Selected ProA-purified non-homogeneous HC / LC pairs and mAb controls were analyzed by non-reducing SDS-PAGE gel and yields were determined by measurement with A280. C) Schematic diagrams of two possible cLC heterozygotes. CPMs are shown as red and blue dots in the CH3 domain. D) and E) Expression levels of cLC hetero-Fc in two bispecificity panels (A×B and C×B, respectively) (shown by ProA yield calculated in mg per liter of conditioned medium). The dashed line highlights 60 mg / L. F) Heatmap plot of relative binding affinity of cLC hetero-Fc to their congeneral and non-congeneral antigens. The binding affinity (KD) of 106 cLC hetero-Fc and 22 corresponding mAbs to soluble antigens A, B, or C was measured using ForteBio Octet. The relative binding affinity of cLC hetero-Fc compared to that of the corresponding congeneral and non-congeneral HC mAbs was then calculated and plotted. G) Inverted pyramid diagram showing the number of molecules at each stage of the ncCSA experiment. [Figure 6A]Figure 6 shows the expression, purification, and binding characteristics of two selected cLC hetero-Fc molecules. A) Final CIEX yields of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their corresponding parental mAbs. B) CIEX chromatographs of A2×B4 and C4×B3. C-E) Binding kinetics of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their respective controls (two hybrid IgG molecules (HC-A2 / LC-B4 and HC-C4 / LC-B3) and two parental mAbs (B4 and B3)). All binding kinetics sensorgrams show superimposed data with an overall fit to a 1:1 binding model. Weak binding to antigen C is in rapid equilibrium, and the lack of curvature leads to high variance in repeated measures. Binding affinity (KD) was calculated as mean ± SD from three independent measurements. [Figure 6B] Figure 6 shows the expression, purification, and binding characteristics of two selected cLC hetero-Fc molecules. A) Final CIEX yields of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their corresponding parental mAbs. B) CIEX chromatographs of A2×B4 and C4×B3. C-E) Binding kinetics of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their respective controls (two hybrid IgG molecules (HC-A2 / LC-B4 and HC-C4 / LC-B3) and two parental mAbs (B4 and B3)). All binding kinetics sensorgrams show superimposed data with an overall fit to a 1:1 binding model. Weak binding to antigen C is in rapid equilibrium, and the lack of curvature leads to high variance in repeated measures. Binding affinity (KD) was calculated as mean ± SD from three independent measurements. [Figure 6C]Figure 6 shows the expression, purification, and binding characteristics of two selected cLC hetero-Fc molecules. A) Final CIEX yields of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their corresponding parental mAbs. B) CIEX chromatographs of A2×B4 and C4×B3. C-E) Binding kinetics of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their respective controls (two hybrid IgG molecules (HC-A2 / LC-B4 and HC-C4 / LC-B3) and two parental mAbs (B4 and B3)). All binding kinetics sensorgrams show superimposed data with an overall fit to a 1:1 binding model. Weak binding to antigen C is in rapid equilibrium, and the lack of curvature leads to high variance in repeated measures. Binding affinity (KD) was calculated as mean ± SD from three independent measurements. [Figure 6D] Figure 6 shows the expression, purification, and binding characteristics of two selected cLC hetero-Fc molecules. A) Final CIEX yields of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their corresponding parental mAbs. B) CIEX chromatographs of A2×B4 and C4×B3. C-E) Binding kinetics of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their respective controls (two hybrid IgG molecules (HC-A2 / LC-B4 and HC-C4 / LC-B3) and two parental mAbs (B4 and B3)). All binding kinetics sensorgrams show superimposed data with an overall fit to a 1:1 binding model. Weak binding to antigen C is in rapid equilibrium, and the lack of curvature leads to high variance in repeated measures. Binding affinity (KD) was calculated as mean ± SD from three independent measurements. [Figure 6E]Figure 6 shows the expression, purification, and binding characteristics of two selected cLC hetero-Fc molecules. A) Final CIEX yields of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their corresponding parental mAbs. B) CIEX chromatographs of A2×B4 and C4×B3. C-E) Binding kinetics of the two cLC hetero-Fc molecules (A2×B4 and C4×B3) and their respective controls (two hybrid IgG molecules (HC-A2 / LC-B4 and HC-C4 / LC-B3) and two parental mAbs (B4 and B3)). All binding kinetics sensorgrams show superimposed data with an overall fit to a 1:1 binding model. Weak binding to antigen C is in rapid equilibrium, and the lack of curvature leads to high variance in repeated measures. Binding affinity (KD) was calculated as mean ± SD from three independent measurements. [Figure 7] Figure 7 shows the multispecific antibody construct module. [Figure 8] Figure 8 shows an overview of the purification and analysis of 32 quadruple-stranded heterozygous Fc cells and their corresponding parent mAbs. [Figure 9] Figure 9 shows the relative expression of 144 non-homogeneous HC / LC pairs in two bispecific programs (A×B and C×B). The ProA yield of each non-homogeneous HC / LC pair was normalized against the yield of the corresponding homogeneous HC / LC pair (mAb) control and plotted. [Figure 10A] Figure 10 shows the expression of non-homogeneic HC / LC pairs and parental mAb (highlighted in gray) in non-competitive CSA experiments. [Figure 10B] Figure 10 shows the expression of non-homogeneic HC / LC pairs and parental mAb (highlighted in gray) in non-competitive CSA experiments. [Figure 10C] Figure 10 shows the expression of non-homogeneic HC / LC pairs and parental mAb (highlighted in gray) in non-competitive CSA experiments. [Figure 10D] Figure 10 shows the expression of non-homogeneic HC / LC pairs and parental mAb (highlighted in gray) in non-competitive CSA experiments. [Figure 11] Figure 11 shows the expression of the anti-target C parent mAb. [Figure 12A]Figure 12 shows high-resolution LC-MS analysis of ProA-purified cLC heteroIgG for two bispecificity programs (A×B(A) and C×B(B)). The proportion of IgG species with the correct MW was calculated from the LC-MS data and plotted. The mean values ​​of cLC heteroIgG and quadruple-stranded heteroIgG (Figure 3C) are highlighted with dashed lines. [Figure 12B] Figure 12 shows high-resolution LC-MS analysis of ProA-purified cLC heteroIgG for two bispecificity programs (A×B(A) and C×B(B)). The proportion of IgG species with the correct MW was calculated from the LC-MS data and plotted. The mean values ​​of cLC heteroIgG and quadruple-stranded heteroIgG (Figure 3C) are highlighted with dashed lines. [Figure 13A] Figure 13 shows the binding dynamics of parental mAbs to their corresponding antigens. [Figure 13B] Figure 13 shows the binding dynamics of parental mAbs to their corresponding antigens. [Figure 14] Figure 14 shows an overview of the purification and analysis of two cLC heteroIgGs and their parent mAbs. [Figure 15] Figure 15 shows schematic diagrams of one correctly constructed IgG species and nine mispaired IgG species after co-expression of two different HCs and two different LCs in a single cell. [Figure 16A] Figure 16 shows the expression of parental mAb and quadruple-stranded heteroIgG in competitive CSA experiments. A) Analysis of ProA purified protein by non-reducing SDS-PAGE gel. B) Analysis of protein expression levels in milligrams per liter after the ProA purification process. [Figure 16B] Figure 16 shows the expression of parental mAb and quadruple-stranded heteroIgG in competitive CSA experiments. A) Analysis of ProA purified protein by non-reducing SDS-PAGE gel. B) Analysis of protein expression levels in milligrams per liter after the ProA purification process. [Figure 17]Figure 17 shows the relative binding affinity of quadruple-stranded heteroIgG compared to parent mAbs. The binding affinity (KD) of 11 SP-purified heteroIgGs and their corresponding mAbs to soluble antigens A and / or antigen B was measured using Fortebio Octet. The relative binding affinity of the heteroIgG compared to the corresponding parent mAbs was then calculated and plotted. [Modes for carrying out the invention]

[0045] Unless otherwise indicated, the term “at least” preceding a set of elements should be understood to refer to all elements of that set. Those skilled in the art will be able to recognize or confirm, through mere conventional experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be incorporated herein.

[0046] Wherever the terms “and / or” are used herein, they include the meanings of “and,” “or,” and “all or any other combination of the elements connected by the terms.”

[0047] When used herein, the terms “approximately” or “abbreviated” mean within ±20%, preferably within ±15%, more preferably within ±10%, and most preferably within ±5% of a given value or range.

[0048] Throughout this specification and the subsequent claims, unless the context requires otherwise, the term “comprise,” and variations such as “comprises” and “comprising,” mean to include the integer or process, or group of integers or processes, that is stated, but not to exclude any other integer or process, or group of integers or processes. As used herein, the term “comprise” may be replaced with the term “contains,” “includes,” or, as used herein, sometimes with the term “has.”

[0049] As used herein, "consisting of" excludes all elements, processes, or components not specified in the elements of the claim. As used herein, "substantially consisting of" does not exclude materials or processes that do not substantially affect the basic and novel features of the claim.

[0050] In this specification, in each case, any of the terms “including,” “substantially consisting of,” and “consisting of” may be replaced with any of the other two terms.

[0051] As used herein, the term “antigen-binding protein” refers to a protein that specifically binds to one or more target antigens. Antigen-binding proteins may include antibodies and their functional fragments. A “functional antibody fragment” is a portion of an antibody that lacks at least some of the amino acids present in the full-length heavy and / or light chain, but is still capable of specifically binding to an antigen. Examples of functional antibody fragments include, but are not limited to, Fab fragments, Fab' fragments, F(ab')2 fragments, Fv fragments, Fd fragments, and complementarity-determining region (CDR) fragments, and may originate from any mammalian source, such as humans, mice, rats, rabbits, or camelids. Functional antibody fragments can be comparable to intact antibodies in terms of target antigen binding, and fragments may be produced by modifying intact antibodies (e.g., enzymatic or chemical cleavage) or newly synthesized using recombinant DNA technology or peptide synthesis.

[0052] Antigen-binding proteins may also include proteins containing one or more functional antibody fragments incorporated into a single polypeptide chain or multiple polypeptide chains. For example, antigen-binding proteins include single-chain Fv (scFv), diabodies (see, e.g., European Patent No. 404,097, International Publication No. 93 / 11161, and Hollinger et al., Proc. Natl. Acad. Sci. USA, Vol. 90:6444-6448, 1993); intracellular antibodies; domain antibodies (single VL or VH domain, or two or more VH domains linked by a peptide linker; see Ward et al., Nature, Vol. 341:544-546, 1989); and maxibodies (two scFv fused to an Fc region; see Fredericks et al., Protein Engineering, Design & Selection, Vol. 17:95-106, 2004, and Powers et al., Journal of Immunological Sciences). See Methods, Vol.251:123-135, 2001); triabody; tetrabody; minibody (scFv fused to the CH3 domain; see Olafsen et al., Protein Eng Des Sel., Vol.17:315-23, 2004); peptidebody (one or more peptides attached to the Fc region; see International Publication No. 00 / 24782); linear antibody (a pair of tandem Fd segments (VH-CH1-VH-CH1) that form a pair of antigen-binding regions together with a complementary light chain polypeptide; see Zapata et al., Protein See Eng., Vol.8:1057-1062, 1995; small modular immunotherapies (see U.S. Patent Application Publication No. 20030133939); and immunoglobulin fusion proteins (e.g., IgG-scFv, IgG-Fab, 2scFv-IgG, 4scFv-IgG, VH-IgG, IgG-VH, and Fab-scFv-Fc).

[0053] "Multispecificity" means that an antigen-binding protein can specifically bind to two or more different antigens. "Bispecificity" means that an antigen-binding protein can specifically bind to two different antigens. As used herein, an antigen-binding protein "specifically binds" to a target antigen if, under similar binding assay conditions, it has a significantly higher binding affinity to the target antigen compared to its affinity to other unrelated proteins, and as a result can identify that antigen. An antigen-binding protein that specifically binds to an antigen has an equilibrium dissociation constant (K). D )≦1×10 -6 It may have M. The antigen-binding protein is K D ≤ 1 × 10 -8 When it is M, it binds specifically to the antigen with "high affinity".

[0054] Affinity is determined using various techniques, one example being affinity ELISA assays. In various embodiments, affinity is determined by surface plasmon resonance assays (e.g., assays using BIAcore®). Using this method, the association rate constant (k a Unit: M -1 s -1 ) and dissociation rate constant (k d Unit: s -1 ) can be measured. Next, the equilibrium dissociation constant (K D (Unit: M) is the ratio of the velocity constants (k d / k a The affinity can be calculated from the equilibrium dissociation constant (K). In some embodiments, affinity is determined by kinetic methods such as the binding equilibrium exclusion method (KinExA) as described in Rathanaswami et al. Analytical Biochemistry, Vol.373:52-60, 2008. Using the KinExA assay, the equilibrium dissociation constant (K) can be calculated. D , unit: M) and association rate constant (k a Unit: M -1 s -1 ) can be measured. Dissociation rate constant (k d Unit: s -1) are these values ​​(K D ×k a It can be calculated from ). In other embodiments, affinity is determined by equilibrium / solution method. In certain embodiments, affinity is determined by FACS binding assay.

[0055] In some embodiments, the multispecific antigen-binding protein described herein is k d The bond affinity, measured by the (dissociation rate constant), is approximately 10 -2 , 10 -3 , 10 -4 , 10 -5 , 10 -6 , 10 -7 , 10 -8 , 10 -9 , 10 -10 s -1 The following (lower values ​​indicate higher bond affinity), and / or K D The binding affinity, measured by the equilibrium dissociation constant, is approximately 10 -9 , 10 -10 , 10 -11 , 10 -12 , 10 -13 , 10 -14 , 10 -15 , 10 -16 It exhibits desirable characteristics such as M or less (a lower value indicates higher binding affinity).

[0056] As used herein, the term "antigen-binding domain," used interchangeably with "binding domain," refers to a region of an antigen-binding protein containing amino acid residues that interact with the antigen and confer specificity and affinity to that antigen to the antigen-binding protein.

[0057] As used herein, the term "CDR" refers to the complementarity-determining region (also called the "minimum recognition unit" or "hypervariable region") within an antibody variable sequence. There are three heavy-chain variable region CDRs (CDRH1, CDRH2, and CDRH3) and three light-chain variable region CDRs (CDRL1, CDRL2, and CDRL3). As used herein, the term "CDR region" refers to a group of three CDRs (i.e., three light-chain CDRs or three heavy-chain CDRs) present in a single variable region. The CDRs in each of the two chains are typically aligned by a framework region to form a structure that specifically binds to a specific epitope or domain of the target protein. From the N-terminus to the C-terminus, both naturally occurring light-chain and heavy-chain variable regions typically correspond to the following order of these elements: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. A numbering scheme has been devised to assign numbers to the amino acids that occupy positions in each of these domains. This numbering system is defined in Kabat Sequences of Proteins of Immunological Interest (1987 and 1991, NIH, Bethesda, MD) or Chothia & Lesk, 1987, J.Mol.Biol.196:901-917; Chothia et al., 1989, Nature 342:878-883. The complementarity-determining regions (CDRs) and framework regions (FRs) of a given antibody can be identified using this system.

[0058] In some embodiments of the multispecific antigen-binding protein of the present invention, the binding domain comprises Fab, Fab', F(ab')2, Fv, a single-chain variable fragment (scFv), or a nanobody. In one embodiment, both binding domains are Fab fragments. In another embodiment, one binding domain is a Fab fragment and the other binding domain is an scFv.

[0059] When an antibody is digested with papain, two identical antigen-binding fragments called "Fab" fragments (each possessing a single antigen-binding site) and the remainder being an "Fc" fragment (containing the immunoglobulin constant region) are produced. The Fab fragment contains the variable domain, as well as the constant domain of the light chain and the first constant domain (CH1) of the heavy chain. Thus, the "Fab fragment" consists of one immunoglobulin light chain (variable region (VL) and constant region (CL)) and the CH1 region and variable region (VH) of one immunoglobulin heavy chain. The heavy chain of the Fab molecule cannot form disulfide bonds with other heavy chain molecules. The Fc fragment is a carbohydrate and is involved in many antibody effector functions (such as complement binding and cell receptors) that distinguish one class of antibody from another. The "Fd fragment" contains the VH domain and CH1 domain derived from the immunoglobulin heavy chain. The Fd fragment represents the heavy chain component of the Fab fragment.

[0060] A "Fab fragment" is a Fab fragment having one or more cysteine ​​residues derived from the antibody hinge region at the C-terminus of the CH1 domain.

[0061] The "F(ab')2 fragment" is a divalent fragment containing two Fab' fragments linked by disulfide bridges between heavy chains in the hinge region.

[0062] The "Fv" fragment is the smallest fragment containing a complete antigen recognition and binding site derived from the antibody. This fragment consists of a dimer of one immunoglobulin heavy chain variable region (VH) and one immunoglobulin light chain variable region (VL) in a tight, non-cohesive association. In this configuration, the three CDRs of each variable region interact to define an antigen-binding site on the surface of the VH-VL dimer. A single light or heavy chain variable region (or half of an Fv fragment containing only the three antigen-specific CDRs) has the ability to recognize and bind to the antigen, but its affinity is lower than that of the entire binding site containing both VH and VL.

[0063] A "single-chain variable antibody fragment" or "scFv fragment" comprises a VH region and a VL region of the antibody, which are present on a single polypeptide chain and optionally include a peptide linker between the VH and VL regions, allowing Fv to form a desired structure for antigen binding (see, for example, Bird et al., Science, Vol.242:423-426, 1988; and Huston et al., Proc. Natl. Acad. Sci. USA, Vol.85:5879-5883, 1988).

[0064] In certain embodiments of the multispecific antigen-binding protein of the present invention, the binding domain comprises an immunoglobulin heavy chain variable region (VH) and an immunoglobulin light chain variable region (VL) of an antibody or antibody fragment that specifically binds to a desired antigen.

[0065] In this specification, the term "variable region," used interchangeably with "variable domain" (variable region of the light chain (VL), variable region of the heavy chain (VH)), refers to a region in the immunoglobulin light chain and immunoglobulin heavy chain, respectively, that is directly involved in the binding of the antibody to the antigen. As described above, the variable light chain region and the variable heavy chain region have the same general structure, and each region contains four framework (FR) regions whose sequences are widely conserved and linked by three CDRs. The framework regions employ a β-sheet structure, and the CDRs can form loops linking the β-sheet structures. The CDRs in each chain are held in their three-dimensional structure by the framework regions and, together with the CDRs of the other chain, form an antigen-binding site.

[0066] The binding domain that specifically binds to the target antigen may be derived from a) known antibodies against these antigens, or b) novel antibodies or antibody fragments obtained by novel immunization methods using the antigen protein or a fragment thereof, by phage display, or by other conventional methods. The antibody from which the binding domain of the multispecific antigen-binding protein originates may be a monoclonal antibody, a polyclonal antibody, a recombinant antibody, a human antibody, or a humanized antibody. In certain embodiments, the antibody from which the binding domain originates is a monoclonal antibody. In these and other embodiments, the antibody is a human antibody or a humanized antibody, and may be of type IgG1, IgG2, IgG3, or IgG4.

[0067] The term “monoclonal antibody” (or “mAb”), as used herein, refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies constituting the population are identical except for any innate variations that may be present in small amounts. Monoclonal antibodies are highly specific and are typically directed at individual antigen sites or epitopes, in contrast to polyclonal antibody preparations, which typically contain various antibodies against various epitopes. Monoclonal antibodies can be produced using any technique known in the art, for example, by immortalizing spleen cells isolated from transgenic animals after completion of an immunization schedule. Spleen cells can be immortalized using any technique known in the art, for example, by fusing spleen cells with myeloma cells to produce hybridomas. Myeloma cells for use in the fusion procedure to produce hybridomas are preferably non-antibody-producing, highly fusion-efficient, and enzyme-deficient, making them unable to grow in certain selective media that support the growth of only the desired fusion cells (hybridoms). Examples of cell lines suitable for use in mouse fusion include Sp-20, P3-X63 / Ag8, P3-X63-Ag8.653, NS1 / 1.Ag 4 1, Sp210-Ag14, FO, NSO / U, MPC-11, MPC11-X45-GTG 1.7, and S194 / 5XXO Bul. Examples of cell lines used in rat fusion include R210.RCY3, Y3-Ag 1.2.3, IR983F, and 4B210. Other cell lines useful for cell fusion include U-266, GM1500-GRG2, LICR-LON-HMy2, and UC729-6.

[0068] In some cases, hybridoma cell lines are produced by immunizing animals (e.g., transgenic animals possessing human immunoglobulin sequences) with a target antigen, collecting spleen cells from the immunized animals, fusing the collected spleen cells with myeloma cell lines, thereby generating hybridoma cells, establishing hybridoma cell lines from the hybridoma cells, and identifying hybridoma cell lines that produce antibodies binding to the target antigen.

[0069] Monoclonal antibodies secreted by hybridoma cell lines can be purified using any technique known in the art, such as protein A-sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography. Hybridomas or mAbs can be further screened to identify mAbs with specific properties, such as the ability to bind to cells expressing a target antigen, the ability to block or interfere with the binding of a target antigen ligand to its respective receptor, or the ability to functionally block either receptor, using, for example, a cAMP assay.

[0070] In some embodiments, the binding domain of the multispecific antigen-binding protein of the present invention may be derived from a humanized antibody. A “humanized antibody” refers to an antibody in which a region (e.g., a framework region) has been modified to include a region derived from a corresponding human immunoglobulin. Generally, humanized antibodies can be produced from monoclonal antibodies initially made in non-human animals. Typically, certain amino acid residues of this monoclonal antibody, derived from the non-antigen-recognition portion of the antibody, are modified to be homologous to the corresponding residues in the corresponding isotype of human antibody. Humanization can be performed, for example, by replacing at least a portion of the rodent variable region with the corresponding region of a human antibody using various methods (see, for example, U.S. Patent No. 5,585,089 and U.S. Patent No. 5,693,762, Jones et al., Nature, Vol. 321:522-525, 1986; Riechmann et al., Nature, Vol. 332:323-27, 1988; Verhoeyen et al., Science, Vol. 239:1534-1536, 1988). The CDRs of the light chain and heavy chain variable regions of antibodies produced in another species can be grafted onto consensus human FRs. To create consensus human FRs, a consensus amino acid sequence can be identified by aligning FRs derived from multiple human heavy chain amino acid sequences or human light chain amino acid sequences.

[0071] Novel antibodies produced against target antigens that can be the origin of the binding domain of the multispecific antigen-binding protein of the present invention can be fully human antibodies. A "fully human antibody" is an antibody that contains a variable region and a constant region that are derived from or represent a human germline immunoglobulin sequence. One specific means provided for carrying out the production of fully human antibodies is the "humanization" of the mouse humoral immune system. Introducing a human immunoglobulin (Ig) locus into a mouse in which the endogenous Ig gene has been inactivated is one means of producing fully human monoclonal antibodies (mAbs) in a mouse, an animal that can be immunized with any desired antigen. Using fully human antibodies can minimize immunogenic and allergic responses that may occur when mouse mAbs or mouse-derived mAbs are administered to humans as therapeutic agents.

[0072] Fully human antibodies can be produced by immunizing transgenic animals (usually mice) that lack endogenous immunoglobulin production and are capable of producing a repertoire of human antibodies. Antigens for this purpose typically have six or more consecutive amino acids and are optionally conjugated to a carrier (such as a hapten). See, for example, Jakobovits et al., 1993, Proc. Natl. Acad. Sci. USA 90:2551-2555, Jakobovits et al., 1993, Nature 362:255-258, and Bruggermann et al., 1993, Year in Immunol. 7:33. In one example of such a method, the transgenic animal is created by deactivating the endogenous mouse immunoglobulin loci encoding the mouse immunoglobulin heavy and light chains and inserting a large fragment of human genomic DNA containing loci encoding human heavy and light chain proteins into the mouse genome. Next, partially modified animals having less complement than the complete complement of the human immunoglobulin locus are crossbred to obtain animals possessing all the desired immunosystem modifications. Upon administration of the immunogen, these transgenic animals produce antibodies that are immunospecific to the immunogen but have a human amino acid sequence, including the variable region, rather than a mouse amino acid sequence. For further details of such methods, see, for example, International Publication No. 96 / 33735 and International Publication No. 94 / 02602.Further methods related to transgenic mice for producing human antibodies are described in U.S. Patent Nos. 5,545,807, 6,713,610, 6,673,986, 6,162,963, 5,939,598, 5,545,807, 6,300,129, 6,255,458, and 5,877,397. This is described in the detailed specifications, U.S. Patent No. 5,874,299, and U.S. Patent No. 5,545,806, PCT publications International Publication Brochures No. 91 / 10741, 90 / 04036, 94 / 02602, 96 / 30498, and 98 / 24893, and European Patent No. 546073B1 and European Patent Application Publication No. 546073A1.

[0073] The transgenic mice referred to herein as "HuMab" mice contain human immunoglobulin gene miniloci encoding unrearranged human heavy chain (mu and gamma) and kappa light chain immunoglobulin sequences, along with targeted mutations that inactivate endogenous mu and kappa chain gene loci (Lonberg et al., 1994, Nature 368:856-859). Consequently, the mice exhibit reduced expression of mouse IgM or kappa, and in response to immunization, the introduced human heavy and light chain transgenes undergo class switching and somatic mutation to produce high-affinity human IgG kappa monoclonal antibodies (Lonberg et al., op. cit.; Lonberg and Huszar, 1995, Intern. Rev. Immunol. 13:65-93; Harding and Lonberg, 1995, Ann. NY Acad. Sci. 764:536-546). HuMab mice were created by Taylor et al.,1992, Nucleic Acids Research 20:6287-6295;Chen et al.,1993,International Immunology 5:647-656;Tuaillon et al.,1994,J.Immunol.152:2912-2920;Lonberg et al.,1994,Nature 368:856-859;Lonberg,1994,Handbook of Exp.Pharmacology 113:49-101;Taylor et al.,1994,International Immunology 6:579-591;Lonberg and Huszar,1995,Intern.Rev.Immunol.13:65-93;Harding and Lonberg, 1995, Ann.NY This is described in detail in Acad.Sci.764:536-546; Fishwild et al., 1996, Nature Biotechnology 14:845-851, and these documents are incorporated herein by reference in their entirety for all purposes.Furthermore, see U.S. Patent Nos. 5,545,806, 5,569,825, 5,625,126, 5,633,425, 5,789,650, 5,877,397, 5,661,016, 5,814,318, 5,874,299, and 5,770,429, as well as U.S. Patent No. 5,545,807, International Publication No. 93 / 1227, International Publication No. 92 / 22646, and International Publication No. 92 / 03918 (all of these disclosures are incorporated herein by reference in their entirety for all purposes). The techniques used to generate human antibodies in these transgenic mice are disclosed in International Publication No. 98 / 24893 and in Mendez et al., 1997, Nature Genetics 15:146-156, which are incorporated herein by reference.

[0074] Human-derived antibodies can also be produced using phage display technology. Phage display is described, for example, in Dower et al., International Publication No. 91 / 17271, McCafferty et al., International Publication No. 92 / 01047, and Caton and Koprowski, Proc. Natl. Acad. Sci. USA, 87:6450-6454 (1990), each of which is incorporated herein by reference in its entirety. Antibodies produced by phage technology are typically generated in bacteria as antigen-binding fragments, e.g., Fv or Fab fragments, and therefore lack effector function. Effector function can be introduced by one of two strategies. The fragments can be manipulated to, if necessary, become complete antibodies expressed in mammalian cells, or to become multispecific antibody fragments having a second binding site that can trigger effector function. Typically, antibody Fd fragments (VH-CH1) and light chains (VL-CL) are cloned separately by PCR, randomly recombined in a combinatorial phage display library, and then selected to bind to a specific antigen. The antibody fragments are expressed on the phage surface, and selection by antigen binding of Fv or Fab (and therefore phages containing the DNA encoding the antibody fragment) is achieved by performing several rounds of antigen binding and re-amplification, a procedure called panning. Antigen-specific antibody fragments are enriched and finally isolated. Phage display technology can also be used in an approach for humanizing rodent monoclonal antibodies called "guided selection" (see Jespers, LS, et al., Bio / Technology 12, 899-903 (1994)). For this purpose, the Fd fragment of a mouse monoclonal antibody can be presented in combination with a human light chain library, and the resulting hybrid Fab library can then be selected using an antigen. Thus, the mouse Fd fragment provides a template to induce selection. Next, the selected human light chain is combined with a human Fd fragment library.By selecting the obtained libraries, a complete human Fab can be obtained.

[0075] In certain embodiments, the multispecific antigen-binding protein of the present invention is an antibody. As used herein, the term “antibody” refers to a tetrameric immunoglobulin protein comprising two light-chain polypeptides (each about 25 kDa) and two heavy-chain polypeptides (each about 50–70 kDa). The term “light chain” or “immunoglobulin light chain” refers to a polypeptide comprising a single immunoglobulin light chain variable region (VL) and a single immunoglobulin light chain constant domain (CL) from the amino terminus to the carboxyl terminus. The immunoglobulin light chain constant domain (CL) may be kappa (κ) or lambda (λ). The term "heavy chain" or "immunoglobulin heavy chain" refers to a polypeptide containing a single immunoglobulin heavy chain variable region (VH), immunoglobulin heavy chain constant domain 1 (CH1), immunoglobulin hinge region, immunoglobulin heavy chain constant domain 2 (CH2), immunoglobulin heavy chain constant domain 3 (CH3), and optionally, immunoglobulin heavy chain constant domain 4 (CH4) from the amino terminus to the carboxyl terminus. Heavy chains are classified as mu (μ), delta (Δ), gamma (γ), alpha (α), and epsilon (ε), and the antibody isotypes are defined as IgM, IgD, IgG, IgA, and IgE, respectively. IgG class antibodies and IgA class antibodies are further divided into subclasses, namely IgG1, IgG2, IgG3, and IgG4, and IgA1 and IgA2, respectively. The heavy chains of IgG, IgA, and IgD antibodies have three domains (CH1, CH2, and CH3), while the heavy chains of IgM and IgE antibodies have four domains (CH1, CH2, CH3, and CH4). The constant domains of the immunoglobulin heavy chain can be derived from any immunoglobulin isotype, including subtypes. The antibody chains are linked via interpolypeptide disulfide bonds between the CL domain and the CH1 domain (i.e., between the light chain and the heavy chain), and between the hinge regions of the antibody heavy chain.

[0076] In certain embodiments, the multispecific antigen-binding protein of the present invention is a heterodimer antibody (used herein interchangeably with "heteroimmunoglobulin" or "heteroIg"), which refers to an antibody comprising two different light chains and two different heavy chains.

[0077] Heterodimeric antibodies may contain any immunoglobulin constant region. The term "constant region," as used herein, refers to all domains of the antibody other than the variable region. The constant region does not directly participate in antigen binding but exerts various effector functions. As described above, antibodies are classified into specific isotypes (IgA, IgD, IgE, IgG, and IgM) and subtypes (IgG1, IgG2, IgG3, IgG4, IgA1, IgA2) depending on the amino acid sequence of their heavy chain constant region. The light chain constant region can be, for example, a kappa-type or lambda-type light chain constant region, such as the human kappa-type or lambda-type light chain constant region found in all five antibody isotypes.

[0078] The heavy chain constant region of a heterodimer antibody can be, for example, an alpha, delta, epsilon, gamma, or mu, such as a human alpha, delta, epsilon, gamma, or mu heavy chain constant region. In some embodiments, the heterodimer antibody includes a heavy chain constant region derived from IgG1, IgG2, IgG3, or IgG4 immunoglobulin. In one embodiment, the heterodimer antibody includes a heavy chain constant region derived from human IgG1 immunoglobulin. In another embodiment, the heterodimer antibody includes a heavy chain constant region derived from human IgG2 immunoglobulin.

[0079] The term “antibody construct” refers to a molecule whose structure and / or function are based on the structure and / or function of an antibody, such as a full-length or full-length immunoglobulin molecule. Therefore, an antibody construct can bind to its specific target or antigen. Furthermore, the antibody construct according to the present invention includes the minimum structural requirements of the antibody that enable target binding. These minimum requirements can be defined, for example, by the presence of at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region), preferably all six CDRs. Antibodies on which constructs according to the present invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies. A “multispecific antibody construct” is an antibody construct that can bind to two or more antigens or to multiple epitopes of a single antigen.

[0080] An "antibody construct module" refers to the overall design of an antibody construct and may include fragments of a full-length antibody such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab')2, or "r IgG" ("half-antibody"). The antibody construct module according to the present invention may also be a modified fragment of an antibody, also called an antibody variant, such as a nanobody containing only one variable domain which may be VHH, VH, or VL, which specifically binds to an antigen or epitope independently of other V regions or domains, or a single-domain antibody such as a single variable-domain antibody. A "multispecific antibody construct module" is an antibody construct module that binds to two or more antigens, or to multiple epitopes of a single antigen. A non-specific example is shown in Figure 7.

[0081] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0082] To clarify, "multispecific antibody construct" refers to a single specific molecule, while "multispecific antibody construct module" refers to the overall design of a multispecific antibody construct. For example, each antibody construct element is located relative to each other and to any Fc region. Thus, in certain embodiments, the multispecific antibody construct module of the present invention includes an Fc moiety. Such an Fc moiety may be a homodimer or a heterodimer. Such an Fc moiety may also be a single-chain Fc ("scFc"), or instead, it may be formed by two distinct polypeptides.

[0083] In one embodiment, the multispecific antibody of this disclosure is Duobody®. Duobody can be prepared using the DuoBody® technology platform (Genmab A / S) as described, for example, in International Publication No. 2008 / 119353, International Publication No. 2011 / 131746, International Publication No. 2011 / 147986, and International Publication No. 2013 / 060867, Labrijn AF et al., PNAS, 110(13):5145-5150 (2013), Gramer et al., mAbs, 5(6):962-973 (2013), and Labrijn et al., Nature Protocols, 9(10):2450-2463 (2014). Using this technique, half of a first monospecific antibody containing two heavy chains and two light chains can be combined with half of a second monospecific antibody containing two heavy chains and two light chains. The resulting heterodimer contains a pair of one heavy chain and one light chain from the first antibody and one heavy chain and one light chain from the second antibody. If both monospecific antibodies recognize different epitopes of different antigens, the resulting heterodimer is a bispecific antibody.

[0084] For the DuoBody® platform, each monospecific antibody contains a heavy chain constant region with a single point mutation in the heavy chain. These point mutations enable stronger interactions between the heavy chains of the resulting multispecific antibodies than between any single heavy chain of an unmutated monospecific antibody. The single point mutation in each monospecific antibody can be at residues 366, 368, 370, 399, 405, 407, or 409 (EU numbered) in the heavy chain of the heavy chain constant region (see International Publication No. 2011 / 131746). Furthermore, the single point mutation is located at a different residue within one monospecific antibody relative to another monospecific antibody. For example, one monospecific antibody may contain mutation F405L (EU numbering; a mutation from phenylalanine to leucine at the 405th residue), or one of the mutations F405A, F405D, F405E, F405H, F405I, F405K, F405M, F405N, F405Q, F405S, F405T, F405V, F405W, and F405Y, while the other monospecific antibody may contain mutation K409R (EU numbering; a mutation from lysine to arginine at the 409th residue). The heavy chain constant region of a monospecific antibody can be an isotype of IgG1, IgG2, IgG3, or IgG4 (e.g., human IgG1 isotype), and the multispecific antibodies produced by DuoBody® technology can be modified to alter (e.g., reduce) Fc-mediated effector function and / or improve half-life.One method for producing Duobody® includes: (i) individually expressing two parental IgG1 cells containing a single matching point mutation (i.e., K409R and one of the mutations F405L (or F405A, F405D, F405E, F405H, F405I, F405K, F405M, F405N, F405Q, F405S, F405T, F405V, F405W, and F405Y) (EU numbering)) in their heavy chains; (ii) mixing the parental IgG1 cells in vitro under permissible redox conditions to enable recombination of the halves; (iii) removing the reducing agent to re-oxidize the interchain disulfide bond; and (iv) analyzing the exchange efficiency and final product using chromatography or mass spectrometry (MS) (Labrijn et al., Nature). See Protocols, 9(10):2450-2463 (2014).

[0085] Another exemplary method for producing multispecific antibodies is the knob-into-hole technique (Ridgway et al., Protein Eng., 9:617-621 (1996); International Publication No. 2006 / 028936). The major drawback of producing multispecific antibodies, the problem of mispairing of Ig heavy chains, is mitigated in this technique by mutating selected amino acids that form the interface of the heavy chains in IgG. At a location within the heavy chains where the two heavy chains directly interact, an amino acid with a small side chain (hole) is introduced into the sequence of one heavy chain, and an amino acid with a large side chain (knob) is introduced into the other heavy chain at the location of the interacting residue. In some examples, the antibodies of this disclosure have immunoglobulin chains whose heavy chains are modified by mutating selected amino acids that interact at the interface between the two polypeptides in order to preferentially form multispecific antibodies. Multispecific antibodies may consist of immunoglobulin chains of the same or different subclasses. In one example, a multispecific antibody that binds to gp120 and CD3 contains the T366W (EU numbered) mutation in the "knob chain" and the T366S, L368A, and Y407V (EU numbered) mutations in the "hole chain". In a particular embodiment, additional interchain disulfide crosslinks are introduced between the heavy chains, for example, by introducing the Y349C mutation into the "knob chain" and the E356C or S354C mutation into the "hole chain". In a particular embodiment, the R409D and K370E mutations are introduced into the "knob chain", and the D399K and E357K mutations are introduced into the "hole chain". In another embodiment, the Y349C and T366W mutations are introduced into one chain, and the E356C, T366S, L368A, and Y407V mutations are introduced into the other chain. In some embodiments, the Y349C and T366W mutations are introduced into one strand, and the S354C, T366S, L368A, and Y407V mutations are introduced into the other strand. In some embodiments, the Y349C and T366W mutations are introduced into one strand, and the S354C, T366S, L368A, and Y407V mutations are introduced into the other strand.In yet another embodiment, the mutations Y349C and T366W are introduced into one strand, and the mutations S354C, T366S, L368A, and Y407V are introduced into the other strand (all EU numbered).

[0086] Another method for producing multispecific antibodies is the CrossMab technique. A CrossMab is a chimeric antibody composed of two halves of full-length antibodies. This technique combines two techniques to ensure correct pairing of the chains: (i) a knob-into-hole mechanism that favors correct pairing between the two heavy chains; and (ii) an exchange between the heavy and light chains of one of the two Fabs to introduce asymmetry and avoid mispairing of the light chains. See Ridgway et al., Protein Eng., 9:617-621 (1996); Schaefer et al., PNAS, 108:11187-11192 (2011). CrossMab can combine two or more antigen-binding domains to target two or more targets, or to introduce bivalence to a single target, such as a 2:1 morphology.

[0087] To facilitate the association of a particular heavy chain with a light chain of its cognate, both the heavy and light chains may contain complementary amino acid substitutions. As used herein, “complementary amino acid substitution” refers to a pairing of a positively charged amino acid substitution in one chain with an uncharged amino acid substitution in the other chain. For example, in some embodiments, the heavy chain contains at least one amino acid substitution to introduce a charged amino acid, and the corresponding light chain contains at least one amino acid substitution to introduce a charged amino acid, wherein the charged amino acid introduced into the heavy chain has the opposite charge to the amino acid introduced into the light chain. In certain embodiments, one or more positively charged residues (e.g., lysine, histidine, or arginine) can be introduced into the first light chain (LC1), and one or more uncharged residues (e.g., aspartic acid or glutamic acid) can be introduced into the corresponding heavy chain (HC1) at the LC1 / HC1 binding interface. Conversely, one or more uncharged residues (e.g., aspartic acid or glutamic acid) can be introduced into the second light chain (LC2), and one or more positively charged residues (e.g., lysine, histidine, or arginine) can be introduced into the corresponding heavy chain (HC2) at the LC2 / HC2 binding interface. Electrostatic interactions induce LC1 to pair with HC1 and LC2 to pair with HC2 because oppositely charged residues (polarity) at the interface attract each other. Heavy / light chain pairs with the same charged residue (polarity) at the interface (e.g., LC1 / HC2 and LC2 / HC1) repel each other, and as a result, unwanted HC / LC pairings are suppressed.

[0088] In these and other embodiments, the CH1 domain of the heavy chain or the CL domain of the light chain contains an amino acid sequence different from the wild-type IgG amino acid sequence, in which one or more positively charged amino acids in the wild-type IgG amino acid sequence are replaced with one or more uncharged amino acids. Alternatively, the CH1 domain of the heavy chain or the CL domain of the light chain contains an amino acid sequence different from the wild-type IgG amino acid sequence, in which one or more uncharged amino acids in the wild-type IgG amino acid sequence are replaced with one or more positively charged amino acids. In some embodiments, one or more amino acids in the CH1 domain of the first and / or second heavy chain in the heterodimer antibody at EU positions selected from F126, P127, L128, A141, L145, K147, D148, H168, F170, P171, V173, Q175, S176, S183, V185 and K213 are replaced with charged amino acids. In certain embodiments, the preferred residue to be substituted with an uncharged or positively charged amino acid is S183 (EU numbering scheme). In some embodiments, S183 is substituted with a positively charged amino acid. In alternative embodiments, S183 is substituted with an uncharged amino acid. For example, in one embodiment, S183 is substituted with an uncharged amino acid (e.g., S183E) in the first heavy chain, and S183 is substituted with a positively charged amino acid (e.g., S183K) in the second heavy chain.

[0089] In embodiments where the light chain is a kappa light chain, one or more amino acids in the CL domain of the first and / or second light chain in the heterodimer antibody at positions selected from F116, F118, S121, D122, E123, Q124, S131, V133, L135, N137, N138, Q160, S162, T164, S174, and S176 (EU and Kabat numbering in the kappa light chain) are substituted with charged amino acids. In embodiments where the light chain is a lambda light chain, one or more amino acids in the CL domain of the first and / or second light chain in the heterodimer antibody at positions selected from T116, F118, S121, E123, E124, K129, T131, V133, L135, S137, E160, T162, S165, Q167, A174, S176, and Y178 (Kabat numbering in the lambda chain) are substituted with charged amino acids. In some embodiments, a preferred residue to be substituted with an uncharged or positively charged amino acid is S176 (EU and Kabat numbering scheme) in the CL domain of either the kappa light chain or the lambda light chain. In certain embodiments, S176 of the CL domain is substituted with a positively charged amino acid. In alternative embodiments, S176 of the CL domain is substituted with an uncharged amino acid. In one embodiment, S176 is substituted with a positively charged amino acid (e.g., S176K) in the first light chain, and S176 is substituted with an uncharged amino acid (e.g., S176E) in the second light chain.

[0090] In addition to, or as an alternative to, complementary amino acid substitutions in the CH1 and CL domains, the variable regions of the light and heavy chains in a heterodimer antibody may contain one or more complementary amino acid substitutions to introduce charged amino acids. For example, in some embodiments, the VH region of the heavy chain or the VL region of the light chain in a heterodimer antibody contains an amino acid sequence different from the wild-type IgG amino acid sequence, in which one or more positively charged amino acids in the wild-type IgG amino acid sequence are substituted with one or more uncharged amino acids. Alternatively, the VH region of the heavy chain or the VL region of the light chain contains an amino acid sequence different from the wild-type IgG amino acid sequence, in which one or more uncharged amino acids in the wild-type IgG amino acid sequence are substituted with one or more positively charged amino acids.

[0091] The V-region interface residues within the VH region (i.e., amino acid residues that mediate the assembly of the VH and VL regions) include positions 1, 3, 35, 37, 39, 43, 44, 45, 46, 47, 50, 59, 89, 91, and 93 of Kabat. One or more of these interface residues in the VH region can be substituted with charged (positively charged or uncharged) amino acids. In certain embodiments, the amino acid at position 39 of Kabat in the VH region of the first and / or second heavy chain is substituted with a positively charged amino acid, such as lysine. In alternative embodiments, the amino acid at position 39 of Kabat in the VH region of the first and / or second heavy chain is substituted with an uncharged amino acid, such as glutamic acid. In some embodiments, the amino acid at position 39 of Kabat in the VH region of the first heavy chain is substituted with a positively charged amino acid (e.g., G39E), and the amino acid at position 39 of Kabat in the VH region of the second heavy chain is substituted with a positively charged amino acid (e.g., G39K). In some embodiments, the amino acid at position 44 of Kabat in the VH region of the first and / or second heavy chain is substituted with a positively charged amino acid, e.g., lysine. In alternative embodiments, the amino acid at position 44 of Kabat in the VH region of the first and / or second heavy chain is substituted with a positively charged amino acid, e.g., glutamic acid. In certain embodiments, the amino acid at position 44 of Kabat in the VH region of the first heavy chain is substituted with a positively charged amino acid (e.g., G44E), and the amino acid at position 44 of Kabat in the VH region of the second heavy chain is substituted with a positively charged amino acid (e.g., G44K).

[0092] The V-region interface residues within the VL region (i.e., amino acid residues mediating the assembly of the VH region and the VL region) include positions 32, 34, 35, 36, 38, 41, 42, 43, 44, 45, 46, 48, 49, 50, 51, 53, 54, 55, 56, 57, 58, 85, 87, 89, 90, 91, and 100 of Kabat. One or more interface residues in the VL region can be substituted with a charged amino acid, preferably an amino acid with the opposite charge to that introduced in the VH region of the cognate heavy chain. In some embodiments, the amino acid at position 100 of Kabat in the VL region of the first and / or second light chain is substituted with a positively charged amino acid, such as lysine. In alternative embodiments, the amino acid at position 100 of Kabat in the VL region of the first and / or second light chain is substituted with a positively charged amino acid, such as glutamic acid. In certain embodiments, the amino acid at position 100 of Kabat in the VL region of the first light chain is substituted with a positively charged amino acid (e.g., G100K), and the amino acid at position 100 of Kabat in the VL region of the second light chain is substituted with a negatively charged amino acid (e.g., G100E).

[0093] In one embodiment, the first Fc region contains negatively charged amino acids at the residues corresponding to positions 409 and 392, and the second Fc region contains positively charged amino acids at the residues corresponding to positions 399 and 356 (wherein the numbering of amino acid residues follows the EU index described in Kabat).

[0094] In one embodiment, the first Fc region includes the K / R409D and K392D mutations, and the second Fc region includes the D399K and E356K mutations (where the amino acid residue numbering follows the EU index described in Kabat).

[0095] In one embodiment, the first Fc region contains negatively charged amino acids at the residues corresponding to positions 409, 439, and 392, and the second Fc region contains positively charged amino acids at the residues corresponding to positions 399 and 356 (wherein the numbering of amino acid residues follows the EU index described in Kabat).

[0096] In one embodiment, the first Fc region includes mutations K / R409D, K439D, and K392D, and the second Fc region includes mutations D399K and E356K (where the amino acid residue numbering follows the EU index described in Kabat).

[0097] In one embodiment, one Fc region contains the mutations F405L, F405A, F405D, F405E, F405H, F405I, F405K, F405M, F405N, F405Q, F405S, F405T, F405V, F405W, or F405Y, and the other Fc region contains the K409R mutation (where the amino acid residue numbering follows the EU index described in Kabat). In another embodiment, one Fc region contains the T366W mutation, and the other Fc region contains the mutations T366S, L368A, or Y407V (where the amino acid residue numbering follows the EU index described in Kabat). In one embodiment, one Fc region contains the K / R409D and K370E mutations, and the other Fc region contains the D399K and E357K mutations (where the amino acid residue numbering follows the EU index described in Kabat).

[0098] In certain embodiments, the heterodimer antibody comprises a first Fc region containing charged amino acids (e.g., substitutions of K392D and K409D) at the residues corresponding to positions 392 and 409, and a second Fc region containing positively charged amino acids (e.g., substitutions of E356K and D399K) at the residues corresponding to positions 356 and 399. In other specific embodiments, the heterodimer antibody comprises a first Fc region containing charged amino acids (e.g., substitutions of K392D, K409D and K439D) at the residues corresponding to positions 392, 409 and 439, and a second Fc region containing positively charged amino acids (e.g., substitutions of E356K and D399K) at the residues corresponding to positions 356 and 399. In other specific embodiments, the heterodimer antibody comprises a first Fc region containing charged amino acids (e.g., substitutions of K392D, K409D, and K370D) at the residues corresponding to positions 392, 409, and 370, and a second Fc region containing positively charged amino acids (e.g., substitutions of E356K, D399K, and D357K) at the residues corresponding to positions 356, 399, and 357.

[0099] In one embodiment, one Fc region contains the Y349C mutation, and the other Fc region contains the E356C or S354C mutation (where the amino acid residue numbering follows the EU index described in Kabat). In one embodiment, one Fc region contains the Y349C and T366W mutations, and the other Fc region contains the E356C, T366S, L368A, and Y407V mutations (where the amino acid residue numbering follows the EU index described in Kabat). In one embodiment, one Fc region contains the Y349C and T366W mutations, and the other Fc region contains the S354C, T366S, L368A, and Y407V mutations (where the amino acid residue numbering follows the EU index described in Kabat).

[0100] In certain embodiments, the heterodimer antibody of the present invention comprises a first heavy chain and a second heavy chain, and a first light chain and a second light chain, wherein the first heavy chain contains amino acid substitutions at positions 183 (EU), 392 (EU), and 409 (EU); the second heavy chain contains amino acid substitutions at positions 183 (EU), 356 (EU), and 399 (EU); and the first and second light chains contain an amino acid substitution at position 176 (EU), the amino acid substitutions introducing charged amino acids at each of the aforementioned positions. In related embodiments, the serine at position 176(EU) of the first light chain is substituted with lysine, the serine at position 176(EU) of the second light chain is substituted with glutamic acid, the serine at position 183(EU) of the first heavy chain is substituted with glutamic acid, the lysine at position 392(EU) of the first heavy chain is substituted with aspartic acid, the lysine at position 409(EU) of the first heavy chain is substituted with aspartic acid, the serine at position 183(EU) of the second heavy chain is substituted with lysine, the glutamic acid at position 356(EU) of the second heavy chain is substituted with lysine, and / or the aspartic acid at position 399(EU) of the second heavy chain is substituted with lysine.

[0101] In a particular embodiment, the heterodimer antibody of the present invention comprises a first heavy chain and a second heavy chain, and a first light chain and a second light chain, wherein the first heavy chain contains amino acid substitutions at positions 183 (EU), 392 (EU), 409 (EU), and 439 (EU); the second heavy chain contains amino acid substitutions at positions 183 (EU), 356 (EU), and 399 (EU); and the first and second light chains contain an amino acid substitution at position 176 (EU), the amino acid substitutions introducing charged amino acids at each of the aforementioned positions. In related embodiments, the serine at position 176(EU) of the first light chain is substituted with lysine, the serine at position 176(EU) of the second light chain is substituted with glutamic acid, the serine at position 183(EU) of the first heavy chain is substituted with glutamic acid, the lysine at position 392(EU) of the first heavy chain is substituted with aspartic acid, the lysine at position 409(EU) of the first heavy chain is substituted with aspartic acid, the lysine at position 439(EU) of the first heavy chain is substituted with aspartic acid, the serine at position 183(EU) of the second heavy chain is substituted with lysine, the glutamic acid at position 356(EU) of the second heavy chain is substituted with lysine, and / or the aspartic acid at position 399(EU) of the second heavy chain is substituted with lysine.

[0102] In certain embodiments, the heterodimer antibody of the present invention comprises a first heavy chain and a second heavy chain, and a first light chain and a second light chain, wherein the first heavy chain contains amino acid substitutions at positions 44 (Kabat), 183 (EU), 392 (EU), and 409 (EU), the second heavy chain contains amino acid substitutions at positions 44 (Kabat), 183 (EU), 356 (EU), and 399 (EU), and the first and second light chains contain amino acid substitutions at positions 100 (Kabat) and 176 (EU), the amino acid substitutions introducing charged amino acids at the aforementioned positions. In related embodiments, the glycine at position 44 (Kabat) of the first heavy chain is replaced with glutamic acid, the glycine at position 44 (Kabat) of the second heavy chain is replaced with lysine, the glycine at position 100 (Kabat) of the first light chain is replaced with lysine, the glycine at position 100 (Kabat) of the second light chain is replaced with glutamic acid, the serine at position 176 (EU) of the first light chain is replaced with lysine, and the serine at position 176 (EU) of the second light chain is replaced with glutamic acid. Substitutions occur, with the serine at position 183 (EU) of the first heavy chain being replaced by glutamic acid, the lysine at position 392 (EU) of the first heavy chain being replaced by aspartic acid, the lysine at position 409 (EU) of the first heavy chain being replaced by aspartic acid, the serine at position 183 (EU) of the second heavy chain being replaced by lysine, the glutamic acid at position 356 (EU) of the second heavy chain being replaced by lysine, and / or the aspartic acid at position 399 (EU) of the second heavy chain being replaced by lysine.

[0103] As used herein, the term “Fc region” refers to the C-terminal region of an immunoglobulin heavy chain that may be produced by papain digestion of an intact antibody. The Fc region of an immunoglobulin generally comprises two constant domains, namely the CH2 domain and the CH3 domain, and optionally, the CH4 domain. In certain embodiments, the Fc region is derived from IgG1, IgG2, IgG3, or IgG4 immunoglobulin. In some embodiments, the Fc region comprises the CH2 and CH3 domains derived from human IgG1 or human IgG2 immunoglobulin. The Fc region may retain effector functions such as C1q binding, complement-dependent cell-mediated cytotoxicity (CDC), Fc receptor binding, antibody-dependent cell-mediated cytotoxicity (ADCC), and phagocytosis. In other embodiments, the Fc region may be modified to reduce or eliminate effector functions, as described further in this specification.

[0104] In some embodiments of the antigen-binding protein of the present invention, the binding domain located at the carboxyl terminus of the Fc region (i.e., the carboxyl-terminus binding domain) is scFv. In certain embodiments, scFv includes a heavy chain variable region (VH) and a light chain variable region (VL) linked by a peptide linker. The variable regions may be arranged within scFv in a VH-VL or VL-VH orientation. For example, in one embodiment, scFv includes a VH region, a peptide linker, and a VL region from the N-terminus to the C-terminus. In another embodiment, scFv includes a VL region, a peptide linker, and a VH region from the N-terminus to the C-terminus. The VH and VL regions of scFv may contain one or more cysteine ​​substitutions to enable the formation of a disulfide bond between the VH and VL regions. Such cysteine ​​clamping stabilizes the two variable domains in the antigen-binding configuration. In one embodiment, position 44 (Kabat numbered) in the VH region and position 100 (Kabat numbered) in the VL region are substituted with cysteine ​​residues.

[0105] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0106] As long as multiple vectors are transfected into the same mammalian cell, CDRs can be cloned into the same vector or different vectors. Cloning CDRs into vectors involves inserting CDRH into the VH framework and CDRL into the VL framework. The VH and VL thus formed can then fuse with each other via a linker to form scFv, or fuse with the CH1 and CL constant regions.

[0107] Next, the constructed vector is transfected into mammalian host cells to express one type of multispecific antibody construct module. However, many different types of multispecific antibody constructs of this particular module are produced due to the multiple Fab fragments and / or scFv fragments that form them.

[0108] Using the method described above, a common light chain capable of binding to any of the antigens is identified. The source of these light chains may be one of three light chain CDRs that specifically bind to the first antigen, three light chain CDRs that specifically bind to the second antigen, or CDRs derived from Fab and / or scFv fragments identified by binding to a completely different third antigen. In this case, the heavy chain CDRs are primarily responsible for binding to the antigen, while the role of the light chain CDRs is more passive.

[0109] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) A step of cloning two or more CDRs into a vector encoding a multispecific antibody construct module, wherein multiple vectors are generated that encode multiple multispecific antibody constructs, each containing a CDR that binds to a different antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species generated. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) For each multispecific antibody construct, compare the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) in order to identify the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen and the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen. This applies to methods that include [specific methods].

[0110] As long as multiple vectors are transfected into the same mammalian cell, CDRs can be cloned into the same vector or different vectors. Cloning CDRs into vectors involves inserting CDRH into the VH framework and CDRL into the VL framework. The VH and VL thus formed can then fuse with each other via a linker to form scFv, or fuse with the CH1 and CL constant regions.

[0111] Next, the constructed vector is transfected into mammalian host cells to express one type of multispecific antibody construct module. However, many different types of multispecific antibody constructs of this particular module are produced due to the multiple Fab fragments and / or scFv fragments that form them.

[0112] Using the method described above, the optimal binding pair for each antigen in a specific module is identified.

[0113] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0114] In one embodiment, the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv includes three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen) are selected from the group consisting of at least two antibody Fab fragments, scFvs, or combinations thereof; at least three antibody Fab fragments, scFvs, or combinations thereof; at least four antibody Fab fragments, scFvs, or combinations thereof; at least five antibody Fab fragments, scFvs, or combinations thereof; at least six antibody Fab fragments, scFvs, or combinations thereof; at least seven antibody Fab fragments, scFvs, or combinations thereof; at least eight antibody Fab fragments, scFvs, or combinations thereof; at least nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

[0115] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs of different modules, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct of different modules in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal module for pairing three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal module for pairing three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0116] As long as multiple vectors are transfected into the same mammalian cell, CDRs can be cloned into the same vector or different vectors. Cloning CDRs into vectors involves inserting CDRH into the VH framework and CDRL into the VL framework. The VH and VL thus formed can then fuse with each other via a linker to form scFv, or fuse with the CH1 and CL constant regions.

[0117] Next, the constructed vector is transfected into mammalian host cells, where multiple types of multispecific antibody construct modules are expressed, but the antigen-binding regions have the same CDR (six CDRs for one antigen-binding region, six CDRs for other antigen-binding regions, however, a particular module may have multiple antigen-binding regions; see Figure 7).

[0118] Using the method described above, we identify the optimal multispecific antibody construct module for the antigen-binding portion that utilizes cLC.

[0119] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) A step of cloning the CDR of the first antibody Fab fragment or scFv and the second antibody Fab fragment or scFv into a vector, The process involves a vector encoding two or more types of multispecific antibody construct modules, resulting in the generation of multiple vectors encoding multiple multispecific antibody constructs of different modules, each containing a CDR that binds to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) A step of measuring the expression level of each multispecific antibody construct, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species generated. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to identify the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen, and the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen, by comparing the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) for each multispecific antibody construct. This applies to methods that include [specific methods].

[0120] As long as multiple vectors are transfected into the same mammalian cell, CDRs can be cloned into the same vector or different vectors. Cloning CDRs into vectors involves inserting CDRH into the VH framework and CDRL into the VL framework. The VH and VL thus formed can then fuse with each other via a linker to form scFv, or fuse with the CH1 and CL constant regions.

[0121] Next, the constructed vector is transfected into mammalian host cells, where multiple types of multispecific antibody construct modules are expressed, but the antigen-binding regions have the same CDR (six CDRs for one antigen-binding region, six CDRs for other antigen-binding regions, however, a particular module may have multiple antigen-binding regions; see Figure 7).

[0122] Using the method described above, the optimal multispecific antibody construct module for the antigen-binding portion is identified.

[0123] In one embodiment, the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

[0124] In one embodiment, the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, SV40-transformed monkey kidney CV1 cell line ("COS-7"), human fetal kidney cell line 293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

[0125] In one embodiment, the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

[0126] In one embodiment, the binding affinity of each multispecific antibody construct to the first and second antigens is measured using Octet, Forte Bio, Carterra LSA, SPR, and flow cytometry.

[0127] In one embodiment, the proportion of correct and incorrect multispecific antibody construct module species is determined by a method selected from the group consisting of liquid chromatography-mass spectrometry ("LC-MS"), Caliper, HPLC SEC, SDS-PAGE, and microchip capillary electrophoresis ("MCE").

[0128] In one embodiment, each multispecific antibody construct is purified by purification of protein A, lambda and kappa resins, and affinity tags. In a particular embodiment, the affinity tags are selected from the group consisting of polyHis (such as hexaHis), streptavidin, FLAG, HA (hemagglutinin influenza virus), and myc tags.

[0129] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody construct modules, each containing a heavy chain CDR and a light chain CDR that binds to a specific antigen (c)(iii); (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen using Octet. (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0130] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) A process in which two or more CDRs are cloned into vectors encoding multispecific antibody construct modules, thereby generating multiple vectors encoding multiple multispecific antibody constructs that bind to each antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species using liquid chromatography-mass spectrometry ("LC-MS"). (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) For each multispecific antibody construct, compare the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) in order to identify the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen and the optimal pairing of three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen. This applies to methods that include [specific methods].

[0131] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) Vector (i) CDRs of three heavy chain CDRs that specifically bind to the first antigen, (ii) CDRs of three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) CDRs of three light chain CDRs that specifically bind to the first antigen, (b) CDRs of three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs of different modules, each containing a heavy chain CDR and a light chain CDR of (c)(iii) that bind to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f)(i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen using Octet. (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal module for pairing three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal module for pairing three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). This applies to methods that include [specific methods].

[0132] In one embodiment, the present invention is a method for selecting a multispecific antibody construct, (a) A step of obtaining a first antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen; (b) A step of obtaining a second antibody Fab fragment or scFv, wherein each Fab fragment or scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen; (c) A step of cloning the CDR of the first antibody Fab fragment or scFv and the second antibody Fab fragment or scFv into a vector, The process involves a vector encoding two or more types of multispecific antibody construct modules, resulting in the generation of multiple vectors encoding multiple multispecific antibody constructs of different modules, each containing a CDR that binds to a specific antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293 cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280, and (ii) A step of calculating the proportion of correct and incorrect multispecific antibody construct module species using liquid chromatography-mass spectrometry ("LC-MS"). (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to identify the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the first antigen, and the multispecific antibody construct module best suited for pairing three heavy chain CDRs and three light chain CDRs that specifically bind to the second antigen, by comparing the expression level of (f)(i) and the proportion of correct and incorrect multispecific antibody construct module species of (f)(ii) for each multispecific antibody construct. This applies to methods that include [specific methods].

[0133] In certain embodiments, the scFv is fused at its amino terminus to the carboxyl terminus of the Fc region (e.g., the carboxyl terminus of the CH3 domain) via a peptide linker, or otherwise linked. Thus, in one embodiment, the scFv is fused to the Fc region such that the resulting fusion protein contains a CH2 domain, a CH3 domain, a first peptide linker, a VH region, a second peptide linker, and a VL region from the N-terminus to the C-terminus. In another embodiment, the scFv is fused to the Fc region such that the resulting fusion protein contains a CH2 domain, a CH3 domain, a first peptide linker, a VL region, a second peptide linker, and a VH region from the N-terminus to the C-terminus. A "fusion protein" is a protein containing polypeptide components derived from two or more parent proteins or polypeptides. Typically, a fusion protein is expressed from a fusion gene in which a nucleotide sequence encoding a polypeptide sequence from one protein is added in frame together with a nucleotide sequence encoding a polypeptide sequence from a different protein, and optionally separated from that sequence by a linker. This fusion gene can then be expressed by recombinant host cells to produce a single fusion protein.

[0134] A "peptide linker" refers to an oligopeptide of about 2 to about 50 amino acids that covalently bonds one polypeptide to another. Peptide linkers are used to link the VH domain and the VL domain in scFv. Peptide linkers can also be used to link scFv, Fab fragments, or other functional antibody fragments to the amino or carboxyl terminus of the Fc region to create the multispecific antigen-binding proteins described herein. Preferably, the peptide linker is at least 5 amino acids long. In certain embodiments, the peptide linker is about 5 to about 40 amino acids long. In other embodiments, the peptide linker is about 8 to about 30 amino acids long. In yet another embodiment, the peptide linker is about 10 to about 20 amino acids long. Preferably, but not necessarily, the peptide linker contains amino acids from 20 standard amino acids, particularly cysteine, glycine, alanine, proline, asparagine, glutamine, and / or serine. In certain embodiments, the peptide linker is composed of a large number of sterically unhinged amino acids, such as glycine, serine, and alanine. Therefore, preferred linkers in some embodiments include polyglycine, polyserine, and polyalanine, or any combination thereof. Some exemplary peptide linkers include, but are not limited to, poly(Gly) 2~8 (Sequence codes 22-26, 30 and 51), especially (Gly)3 (Sequence code 22), (Gly)4 (Sequence code 23), (Gly)5 (Sequence code 24), (Gly)6 (Sequence code 25) and (Gly)7 (Sequence code 26), as well as poly(Gly)4Ser (Sequence code 48), poly(Gly-Ala) 2~4 (Sequence numbers 33-35) and poly(Ala) 2~8 Examples include (Sequence IDs 36-42). In certain embodiments, the peptide linker is (Gly x Ser) nHerein, x = 3 or 4 and n = 2, 3, 4, 5 or 6 (SEQ ID NOs: 29, 31, 32 and 43-50). Examples of such peptide linkers include "L5" (GGGGS or "G4S"; SEQ ID NO: 27), "L9" (GGGSGGGGS; or "G3SG4S"; SEQ ID NO: 28), "L10" (GGGGSGGGGS; or "(G4S)2"; SEQ ID NO: 29), "L15" (GGGGSGGGGSGGGGS; or "(G4S)3"; SEQ ID NO: 31), and "L25" (GGGGSGGGGSGGGGSGGGGSGGGGS; or "(G4S)5"; SEQ ID NO: 32). In some embodiments, the peptide linker linking the VH region and the VL region in scFv is the L15 or (G4S)3 linker (SEQ ID NO: 31). In these and other embodiments, the peptide linker that links the carboxyl-terminal binding domain (e.g., scFv or Fab) to the C-terminus of the Fc region is an L9 or G3SG4S linker (SEQ ID NO: 28) or an L10(G4S)2 linker (SEQ ID NO: 29).

[0135] Other specific examples of peptide linkers that can be used in the multispecific antigen-binding proteins of the present invention include (Gly)5Lys(SEQ ID NO: 1); (Gly)5LysArg(SEQ ID NO: 2); (Gly)3Lys(Gly)4(SEQ ID NO: 3); (Gly)3AsnGlySer(Gly)2(SEQ ID NO: 4); (Gly)3Cys(Gly)4(SEQ ID NO: 5); GlyProAsnGlyGly(SEQ ID NO: 6); GGEGGG(SEQ ID NO: 7); GGEEEGGG(SEQ ID NO: 8); GEEEG(SEQ ID NO: 9); GEEE(SEQ ID NO: 10); GGDGGG(SEQ ID NO: 11); GDDDDGG(SEQ ID NO: 12); GDDDG(SEQ ID NO: 13); GDDD(SEQ ID NO: 14); GGGGSDDSDEGSDGEDGGGGS(SEQ ID NO: 15); WEWEW(SEQ ID NO: 16); FEFEF(SEQ ID NO: 17); EEEWWW(SEQ ID NO: 18); EEEFFF(SEQ ID NO: 19); WWEEEWW(SEQ ID NO: 20); and FFEEEFF(SEQ ID NO: 21).

[0136] The heavy chain constant region or Fc region of the multispecific antigen-binding proteins described herein may contain one or more amino acid substitutions that affect the glycosylation and / or effector function of the antigen-binding protein. One function of the Fc region of immunoglobulins is to transmit the immunoglobulin to the immune system when it binds to its target. This is commonly referred to as “effector function.” This transmission leads to antibody-dependent cell-mediated cytotoxicity (ADCC), antibody-dependent cell-mediated phagocytosis (ADCP), and / or complement-dependent cell-mediated cytotoxicity (CDC). ADCC and ADCP are mediated through the binding of the Fc region to Fc receptors on the surface of immune system cells. CDC is mediated through the binding of a complement system protein, e.g., C1q, to ​​Fc. In some embodiments, the multispecific antigen-binding proteins of the present invention contain one or more amino acid substitutions in the constant region to enhance effector functions, including ADCC activity, CDC activity, ADCP activity, and / or the clearance or half-life of the antigen-binding protein. Examples of amino acid substitutions (EU numbering) that can enhance effector functionality include, but are not limited to, E233L, L234I, L234Y, L235S, G236A, S239D, F243L, F243V, P247I, D280H, K290S, K290E, K290N, K290Y, R292P, E294L, Y296W, S298A, and S298D. Examples include S298V, S298G, S298T, T299A, Y300L, V305I, Q311M, K326A, K326E, K326W, A330S, A330L, A330M, A330F, I332E, D333A, E333S, E333A, K334A, K334V, A339D, A339Q, P396L, or any combination of the above.

[0137] In other embodiments, the multispecific antigen-binding proteins of the present invention include one or more amino acid substitutions in the constant region to reduce effector function. Examples of amino acid substitutions (EU numbering) that can reduce effector function include, but are not limited to, C220S, C226S, C229S, E233P, L234A, L234V, V234A, L234F, L235A, L235E, G237A, P238S, S267E, H268Q, N297A, N297G, V309L, E318A, L328F, A330S, A331S, P331S, or any combination thereof.

[0138] Glycosylation can be a factor in the effector function of antibodies, particularly IgG1 antibodies. Therefore, in some embodiments, the multispecific antigen-binding proteins of the present invention may contain one or more amino acid substitutions that affect the level or type of glycosylation of the binding protein. Polypeptide glycosylation is typically either N-linked or O-linked. N-linking refers to the binding of a carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid other than proline) are recognition sequences for the enzymatic binding of a carbohydrate moiety to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the binding of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine may also be used.

[0139] In certain embodiments, glycosylation of the multispecific antigen-binding proteins described herein is increased by adding one or more glycosylation sites, for example, to the Fc region of the binding protein. Addition of glycosylation sites to antigen-binding proteins can be easily achieved by modifying the amino acid sequence to contain one or more of the above-described tripeptide sequences (in the case of N-linked glycosylation sites). Modification can also be made by adding or substituting one or more serine or threonine residues to the start sequence (in the case of O-linked glycosylation sites). For ease of implementation, the antigen-binding protein amino acid sequence can be modified by changes at the DNA level, particularly by mutating the DNA encoding the target polypeptide at pre-selected base positions so that codons translated to desired amino acids are generated.

[0140] The present invention also encompasses the production of multispecific antigen-binding protein molecules in which the carbohydrate structure is modified to alter effector activity, such as antigen-binding proteins exhibiting enhanced ADCC activity by lacking or reducing fucosylation. Various methods for reducing or eliminating fucosylation are known in the art. For example, ADCC effector activity is mediated by the binding of antibody molecules to the FcγRIII receptor, which has been shown to depend on the carbohydrate structure of N-linked glycosylation at the N297 residue of the CH2 domain. Non-fucosylated antibodies bind to this receptor with high affinity and induce FcγRIII-mediated effector function more efficiently than naturally occurring fucosylated antibodies. For example, recombinant production of non-fucosylated antibodies in CHO cells in which the alpha-1,6-fucosyltransferase enzyme is knocked out produces antibodies with 100-fold increased ADCC activity (see Yamane-Ohnuki et al., Biotechnol Bioeng. 87(5):614-22, 2004). Similar effects can be achieved by reducing the activity of the alpha-1,6-fucosyltransferase enzyme or other enzymes in the fucosylation pathway, for example, by siRNA or antisense RNA treatment, cell line manipulation to knock out the enzyme, or culture using selective glycosylation inhibitors (see Rothman et al., Mol Immunol. 26(12):1113-23, 1989). Some host cell lines, such as Lec13 or the rat hybridoma YB2 / 0 cell line, naturally produce antibodies with lower fucosylation levels (see Shields et al., J Biol Chem. 277(30):26733-40, 2002 and Shinkawa et al., J Biol Chem. 278(5):3466-73, 2003). For example, it has been found that increasing the level of bifurcated glycans by recombinant antibody production in cells overexpressing the GnTIII enzyme also increases ADCC activity (see Umana et al., Nat Biotechnol. 17(2):176-80, 1999).

[0141] In other embodiments, glycosylation of the multispecific antigen-binding proteins described herein is reduced or eliminated by removing one or more glycosylation sites, for example, from the Fc region of the binding protein. N-linked glycosylation of antigen-binding proteins can be reduced or eliminated by amino acid substitutions that eliminate or modify the N-linked glycosylation sites. In certain embodiments, the multispecific antigen-binding proteins described herein include mutations at position N297 (EU numbering), such as N297Q, N297A, or N297G. In a particular embodiment, the multi-bispecific antigen-binding protein of the present invention includes an Fc region derived from a human IgG1 antibody having the N297G mutation. To improve the stability of the molecule containing the N297 mutation, the Fc region of the molecule may be further manipulated. For example, in some embodiments, one or more amino acids in the Fc region are substituted with cysteine ​​to promote disulfide bond formation in a dimeric state. Therefore, residues corresponding to V259, A287, R292, V302, L306, V323, or I332 (EU numbering) in the IgG1 Fc region may be substituted with cysteine. Preferably, specific pairs of residues are substituted with cysteine ​​so as to preferentially form disulfide bonds with each other, thereby limiting or preventing disulfide bond scrambling. Preferred pairs include, but are not limited to, A287C and L306C, V259C and L306C, R292C and V302C, and V323C and I332C. In certain embodiments, the multispecific antigen-binding proteins described herein include an Fc region derived from a human IgG1 antibody having mutations in R292C and V302C. In such embodiments, the Fc region may also include the N297G mutation.

[0142] For example, modification of the multispecific antigen-binding proteins of the invention to extend serum half-life may also be desirable by incorporation or addition of salvage receptor-binding epitopes (e.g., by mutation of an appropriate region or incorporation of the epitope into a peptide tag followed by fusion to the antigen-binding protein at either the terminus or centrally, e.g., by DNA or peptide synthesis; see, e.g., WO 96 / 32478 pamphlet), or by addition of molecules such as PEG or other water-soluble polymers, e.g., polysaccharide polymers. The salvage receptor-binding epitope preferably constitutes a region in which one or more amino acid residues derived from one or two loops of the Fc region are transferred to a similar position in the antigen-binding protein. Even more preferably, three or more residues derived from one or two loops of the Fc region are transferred. Even more preferably, the epitope is taken from the CH2 domain of the Fc region (e.g., IgG Fc region) and transferred to the CH1, CH3, or VH region of the antigen-binding protein, or two or more such regions. Alternatively, the epitope is taken from the CH2 domain of the Fc region and transferred to the CL region or VL region of the antigen-binding protein, or both. For an explanation of Fc variants and their interaction with salvage receptors, see WO 97 / 34631 pamphlet and WO 96 / 32478 pamphlet of the international application.

[0143] The present invention includes one or more isolated nucleic acids encoding the multispecific antigen-binding proteins and their components described herein. The nucleic acid molecules of the invention include DNA and RNA in both single-stranded and double-stranded forms, as well as corresponding complementary sequences. DNA includes, for example, cDNA, genomic DNA, chemically synthesized DNA, DNA amplified by PCR, and combinations thereof. The nucleic acid molecules of the invention include combinations of full-length genes or cDNA molecules and fragments thereof. The nucleic acids of the invention are preferably derived from a human source, but the invention also includes those derived from non-human species.

[0144] The relevant amino acid sequence derived from immunoglobulin or its region (e.g., variable region, Fc region, etc.) or the polypeptide of interest can be determined by direct protein sequencing, and a suitable coding nucleotide sequence can be designed according to a universal codon table. Alternatively, genomic DNA or cDNA encoding a monoclonal antibody, which may be the origin of the binding domain of the multispecific antigen-binding protein of the present invention, can be isolated and sequenced from cells producing such antibodies using conventional procedures (e.g., by using oligonucleotide probes that can specifically bind to the genes encoding the heavy and light chains of the monoclonal antibody).

[0145] As used interchangeably herein with "isolated polynucleotide", an "isolated nucleic acid" is, in the case of a nucleic acid isolated from a naturally occurring source, a nucleic acid that is separated from the adjacent gene sequences present in the genome of the organism from which the nucleic acid is isolated. For example, in the case of a nucleic acid enzymatically or chemically synthesized from a template, such as a PCR product, a cDNA molecule or an oligonucleotide, the nucleic acid obtained from such a process is understood to be an isolated nucleic acid. An isolated nucleic acid molecule refers to a nucleic acid molecule in the form of a separate fragment or as a component of a larger nucleic acid construct. In a preferred embodiment, the nucleic acid is substantially free of endogenous substances that would otherwise contaminate it. The nucleic acid molecule is preferably in substantially pure form and is derived from at least one isolated DNA or RNA in an amount or concentration that permits identification, manipulation, and recovery of its component nucleotide sequences by standard biochemical methods (such as those outlined in Sambrook et al., Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1989)). Such sequences are preferably provided and / or constructed in the form of an open reading frame that is not interrupted by internal untranslated sequences or introns that are typically present in eukaryotic genes. The sequences of non-translated DNA can be present 5' or 3' to the open reading frame, in which case they do not interfere with the manipulation or expression of the coding region. Unless otherwise specified, the left-hand end of any single-stranded polynucleotide sequence described herein is the 5' end, and the left-hand direction of a double-stranded polynucleotide sequence is called the 5' direction. The direction of production of a nascent RNA transcript from 5' to 3' is called the transcription direction, and the sequence region on the DNA strand that has the same sequence as the RNA transcript and is 5' to the 5' end of the RNA transcript is called the "upstream sequence", and the sequence region on the DNA strand that has the same sequence as the RNA transcript and is 3' to the 3' end of the RNA transcript is called the "downstream sequence".

[0146] The present invention also includes nucleic acids that hybridize to nucleic acids encoding polypeptides described herein under moderately stringent conditions, more preferably highly stringent conditions. Basic parameters influencing the selection of hybridization conditions and guidance for devising preferred conditions are shown by Sambrook, Fritsch, and Maniatis (1989, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, chapters 9 and 11; and Current Protocols in Molecular Biology, 1995, Ausubel et al., eds., John Wiley & Sons, Inc., sections 2.10 and 6.3-6.4), which can be readily determined by those skilled in the art, for example, based on the length and / or base composition of the DNA. One method for achieving moderately stringent conditions involves using a pre-wash solution containing 5×SSC, 0.5% SDS, and 1.0 mM EDTA (pH 8.0), a hybridization buffer of about 50% formamide and 6×SSC, a hybridization temperature of about 55°C (or other similar hybridization solutions, e.g., one containing about 50% formamide at a hybridization temperature of about 42°C), and washing conditions of about 60°C in 0.5×SSC and 0.1% SDS. Generally, highly stringent conditions are defined as hybridization conditions similar to those described above, except for washing at about 68°C with 0.2×SSC and 0.1% SDS. SSPE (1 × SSPE is 0.15 M NaCl, 10 mM NaH2PO4, and 1.25 mM EDTA, pH 7.4) can be replaced with SSC (1 × SSC is 0.15 M NaCl and 15 mM sodium citrate) in the hybridization buffer and wash buffer, and after hybridization is complete, wash for 15 minutes.Naturally, as is known to those skilled in the art, and as will be further described below, the washing temperature and washing salt concentration can be adjusted as necessary to achieve the desired degree of stringency by applying the basic principles that determine the hybridization reaction and the stability of the double strand (see, e.g., Sambrook et al., 1989). When hybridizing a nucleic acid to a target nucleic acid of an unknown sequence, the hybrid length is assumed to be the length of the nucleic acid being hybridized. When a nucleic acid with a known sequence is hybridized, the hybrid length can be determined by aligning the sequences of both nucleic acids and identifying the region or group of regions with optimal sequence complementarity. The hybridization temperature of a hybrid expected to be less than 50 base pairs long must be 5–10°C lower than the melting temperature (Tm) of the hybrid, where Tm is determined by the following formula: For hybrids less than 18 base pairs long, Tm(°C) = 2(number of A+T bases) + 4(number of G+C bases). For hybrids longer than 18 base pairs, Tm(°C) = 81.5 + 16.6(log10[Na+]) + 0.41(%G+C) - (600 / N), (wherein N is the number of bases in the hybrid and [Na+] is the concentration of sodium ions in the hybridization buffer) ([Na+] of 1 × SSC = 0.165M).Preferably, each of the nucleic acids that hybridize is at least 15 nucleotides in length (or more preferably at least 18 nucleotides, or at least 20 nucleotides, or at least 25 nucleotides, or at least 30 nucleotides, or at least 40 nucleotides, or most preferably at least 50 nucleotides), or at least 25% (more preferably at least 50%, or at least 60%, or at least 70%, and most preferably at least 80%) of the length of the nucleic acid of the present invention that it hybridizes, and The nucleic acid of the present invention that hybridizes with this has at least 60% sequence identity (more preferably at least 70%, at least 75%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99%, and most preferably at least 99.5%). Here, sequence identity is determined by comparing the sequences of the nucleic acids that hybridize with this nucleic acid when the sequences of the nucleic acids are aligned to maximize overlap and identity and minimize sequence gaps, as described in more detail above.

[0147] The antigen-binding protein variants described herein can be prepared by producing mutant-encoding DNA using site-directed mutagenesis of nucleotides in the polypeptide-encoding DNA, by cassette or PCR mutagenesis, or by other techniques known in the art, and then expressing the recombinant DNA in a cell culture as outlined herein. However, antigen-binding proteins, including variant CDRs having up to approximately 100–150 residues, can be prepared by in vitro synthesis using established techniques. Variants typically exhibit qualitative biological activity similar to that of their native analogs, e.g., binding to antigens. Such variants include, for example, deletions and / or insertions and / or substitutions of residues in the amino acid sequence of the antigen-binding protein. Any combination of deletions, insertions, and substitutions is performed to arrive at the final construct, provided that the final construct retains the desired properties. Amino acid changes may also alter the post-translational processes of the antigen-binding protein, such as changing the number or location of glycosylation sites. In certain embodiments, antigen-binding protein variants are prepared with the aim of modifying amino acid residues directly involved in epitope binding. In other embodiments, for the purposes described herein, modification of residues that are not directly involved in epitope binding, or residues that are not involved in epitope binding at all, is desirable. Mutagenesis in either the CDR region and / or the framework region is intended. Those skilled in the art can use analysis of covariance techniques to design useful modifications to the amino acid sequence of the antigen-binding protein.For example, see Choulier, et al., Proteins 41:475-484, 2000; Demarest et al., J.Mol.Biol.335:41-48, 2004; Hugo et al., Protein Engineering 16(5):381-86, 2003; Aurora et al., U.S. Patent Application Publication No. 2008 / 0318207A1; Glaser et al., U.S. Patent Application Publication No. 2009 / 0048122A1; Urech et al., International Publication No. 2008 / 110348A1; Borras et al., International Publication No. 2009 / 000099A2. Such modifications, determined by analysis of covariance, can improve the potency, pharmacokinetic, pharmacodynamic and / or manufacturability properties of antigen-binding proteins.

[0148] The present invention also includes vectors comprising one or more nucleic acids encoding one or more components (e.g., variable regions, light chains, heavy chains, modified heavy chains, and Fd fragments) of the multispecific antigen-binding protein of the present invention. The term “vector” refers to any molecule or entity (e.g., nucleic acids, plasmids, bacteriophages, or viruses) used to transfer protein-coding information into a host cell. Examples of vectors include, but are not limited to, plasmids, viral vectors, non-episome mammalian vectors, and expression vectors, e.g., recombinant expression vectors. The terms “expression vector” or “expression construct,” as used herein, refer to a recombinant DNA molecule containing a desired coding sequence and appropriate nucleic acid regulatory sequences necessary for the expression of a coding sequence operably ligated in a particular host cell. An expression vector may, but is not limited to, include sequences that affect or control transcription, translation, and, if introns are present, sequences that affect the RNA splicing of the coding region operably ligated to them. Nucleic acid sequences necessary for expression in prokaryotes include promoters, optionally operator sequences, ribosome binding sites, and possibly other sequences. Eukaryotic cells are known to utilize promoters, enhancers, and termination and polyadenylation signals. Secretory signal peptide sequences can also be optionally encoded by expression vectors and operably ligated to the desired coding sequence, thereby allowing recombinant host cells to secrete the expressed polypeptide, which, if desired, can be more easily isolated from the cell.In certain embodiments, the signal peptide is selected from the group consisting of MDMRVPAQLLGLLLLWLRGARC (SEQ ID NO: 52), MAWALLLLTLLTQGTGSWA (SEQ ID NO: 53), MTCSPLLLTLLIHCTGSWA (SEQ ID NO: 54), MEAPAQLLFLLLLWLPDTTG (SEQ ID NO: 55), MEWTWRVLFLVAAATGAHS (SEQ ID NO: 56), METPAQLLFLLLLWLPDTTG (SEQ ID NO: 57), METPAQLLFLLLLWLPDTTG (SEQ ID NO: 58), MKHLWFFLLLVAAPRWVLS (SEQ ID NO: 59), and MEWSWVFLFFLSVTTGVHS (SEQ ID NO: 60).

[0149] Typically, the expression vector used in host cells to produce the multispecific antigen protein of the present invention contains a sequence for plasmid maintenance, as well as a sequence for cloning and expression of exogenous nucleotide sequences encoding components of the multispecific antigen-binding protein. Such sequences, collectively referred to as “adjacent sequences,” typically in certain embodiments include the following nucleotide sequences: a promoter, one or more enhancer sequences, an origin of replication, a transcription termination sequence, a complete intron sequence containing donor and acceptor splice sites, a sequence encoding a leader sequence for polypeptide secretion, a ribosome binding site, a polyadenylation sequence, a polylinker region for inserting the nucleic acid encoding the polypeptide to be expressed, and one or more of the selection marker elements. Each of these sequences is described below.

[0150] Optionally, the vector may contain a “tag” coding sequence, i.e., an oligonucleotide molecule located at the 5' or 3' end of the polypeptide coding sequence, which encodes polyHis (e.g., hexaHis), streptavidin, FLAG, HA (hemagglutinin influenza virus), myc, or another “tag” molecule for which a commercially available antibody exists. This tag is typically fused to the polypeptide upon expression and can serve as a means for affinity purification or detection of the polypeptide from host cells. Affinity purification can be achieved, for example, by column chromatography using an antibody against the tag as an affinity matrix. Optionally, the tag can then be removed from the purified polypeptide by various means, such as using a specific cleavage peptidase.

[0151] Adjacent sequences can be congeneral (i.e., derived from the same species and / or strain as the host cell), heterogeneous (i.e., derived from a species other than the host cell species or strain), hybrid (i.e., a combination of adjacent sequences from two or more sources), synthetic, or native. Therefore, the sources of adjacent sequences can be any prokaryote or eukaryote, any vertebrate or invertebrate, or any plant, provided that the adjacent sequence functions within and can be activated by the host cell mechanism.

[0152] Flanking sequences useful for the vector of the present invention can be obtained by any of several methods well known in the art. Typically, flanking sequences useful herein will have been identified in advance by mapping and / or restriction endonuclease digestion, and can be isolated from a suitable tissue source using a suitable restriction endonuclease. In some cases, the entire nucleotide sequence of the flanking sequence may be known. In this case, the flanking sequence can be synthesized using conventional methods of nucleic acid synthesis or cloning.

[0153] Whether all or only some of the adjacent sequences are known, adjacent sequences can be obtained by screening a genomic library using polymerase chain reaction (PCR) and / or suitable probes such as oligonucleotides and / or adjacent sequence fragments derived from the same or a different species. If the adjacent sequences are unknown, DNA fragments containing adjacent sequences can be isolated, for example, from large DNA fragments that may contain coding sequences or one or more other genes. Isolation can be achieved by producing suitable DNA fragments by restriction endonuclease digestion, followed by isolation using agarose gel purification, Qiagen® column chromatography (Chatsworth, CA), or other methods known to those skilled in the art. The selection of suitable enzymes for achieving this objective will be readily apparent to those skilled in the art.

[0154] The origin of replication is typically a component of a commercially available prokaryotic expression vector, and this origin is useful for amplifying the vector in host cells. If the selected vector does not contain an origin of replication site, it may be chemically synthesized based on a known sequence and ligated into the vector. For example, the origin of replication derived from plasmid pBR322 (New England Biolabs, Beverly, MA) is suitable for most Gram-negative bacteria, and various viral origins (e.g., SV40, polyoma, adenovirus, varicella-stomatitis virus (VSV), or papillomavirus, e.g., HPV or BPV) are useful for cloning vectors in mammalian cells. In general, the origin of replication component is not necessary for mammalian expression vectors (for example, the SV40 origin is often used only because it also contains the initial viral promoter).

[0155] Transcription termination sequences are typically located at the 3' end of the polypeptide coding region and function to terminate transcription. In prokaryotic cells, the transcription termination sequence is usually a GC-rich fragment followed by a poly-T sequence. This sequence can be readily cloned from libraries, or even purchased commercially as part of a vector, but it can also be readily synthesized using known nucleic acid synthesis methods.

[0156] Selection marker genes encode proteins necessary for the survival and proliferation of host cells grown in a selective culture medium. Typical selection marker genes encode (a) proteins that confer resistance to antibiotics or other toxins, such as ampicillin, tetracycline, or kanamycin, to prokaryotic host cells; (b) proteins that compensate for deficiencies in the cellular nutritional requirements; or (c) proteins that supply essential nutrients unavailable from complex or limited media. Specific selection markers include kanamycin resistance genes, ampicillin resistance genes, and tetracycline resistance genes. Advantageously, neomycin resistance genes can also be used for selection in both prokaryotic and eukaryotic host cells.

[0157] Other select genes may be used to amplify the expressed genes. Amplification is the process by which genes required for the production of proteins important for proliferation or cell survival are repeated in tandem within the chromosomes of recombinant cells over generations. Examples of suitable select markers for mammalian cells include the dihydrofolate reductase (DHFR) and promoter resthymidine kinase genes. Mammalian cell transformants are placed under selective pressure, in which only these transformants are adapted to survive by the select gene present in the vector. Selective pressure is imposed by culturing the transformed cells under conditions in which the concentration of the selector in the culture medium is continuously increased, thereby leading to amplification of both the select gene and the DNA encoding another gene, such as one or more components of the multispecific antigen-binding proteins described herein. As a result, large amounts of polypeptides are synthesized from the amplified DNA.

[0158] The ribosome binding site is typically required for mRNA translation initiation and is characterized by a Shine-Dalgarno sequence (prokaryotes) or a Kozak sequence (eukaryotes). This element is typically located at the 3' end of the promoter and at the 5' end of the coding sequence of the polypeptide to be expressed. In certain embodiments, one or more coding regions may be operably ligated to an internal ribosome binding site (IRES), enabling translation of two open reading frames from a single RNA transcript.

[0159] In cases where glycosylation is desired in eukaryotic host cell expression systems, various pre-sequences or pro-sequences can be manipulated to improve glycosylation or yield. For example, the peptidase cleavage site of a particular signal peptide can be modified, or a pro-sequence can be added, which can also affect glycosylation. The final protein product may have one or more additional amino acids associated with expression at position -1 (relative to the first amino acid of the mature protein), which do not necessarily have to be completely removed. For example, the final protein product may have one or two amino acid residues found at the peptidase cleavage site, attached to the amino terminus. Alternatively, using several enzyme cleavage sites may result in a slightly cleaved form of the desired polypeptide if the enzyme cleaves at such regions within the mature polypeptide.

[0160] The expression vectors and cloning vectors of the present invention typically contain a promoter that is recognized by a host organism and operably ligated to a molecule encoding a polypeptide. The term "operably ligated," as used herein, refers to the ligation of two or more nucleic acid sequences such that a nucleic acid molecule is produced that can direct the transcription of a given gene and / or the synthesis of a desired protein molecule. For example, a regulatory sequence in a vector "operably ligated" to a protein-coding sequence is ligated to the protein-coding sequence such that the expression of the protein-coding sequence occurs under conditions compatible with the transcriptional activity of the regulatory sequence. More specifically, a promoter and / or enhancer sequence (including any combination of cis-acting transcriptional regulatory elements) is operably ligated to a coding sequence if it stimulates or modulates the transcription of the coding sequence in a suitable host cell or other expression system.

[0161] A promoter is a non-transcriptional sequence located upstream (i.e., at the 5' end) of the start codon of a structural gene (generally within approximately 100–1000 bp) and controls the transcription of that structural gene. Promoters are typically classified into two classes: inductive promoters and constitutive promoters. Inductive promoters initiate an increase in the transcription level from DNA under their control in response to any change in culture conditions, such as the presence or absence of nutrients or changes in temperature. Constitutive promoters, on the other hand, transcribe the gene to which they are operably ligated uniformly, i.e., with little or no control over gene expression. Numerous promoters recognized by various potential host cells are well known. A suitable promoter is operably ligated to the DNA encoding, for example, the heavy chain, light chain, modified heavy chain, or other components of the multispecific antigen-binding protein of the present invention, by removing the promoter from the source DNA by restriction enzyme digestion and inserting the desired promoter sequence into a vector.

[0162] Promoterians suitable for use with yeast hosts are also well known in the art. Yeast enhancers are advantageous when used with yeast promoters. Promoterians suitable for use with mammalian host cells are well known and include, but are not limited to, those derived from the genomes of viruses such as polyomaviruses, fowlpox virus, adenoviruses (such as adenovirus type 2), bovine papillomavirus, aerosarcoma virus, cytomegalovirus, retroviruses, hepatitis B virus, and most preferably Simian virus 40 (SV40). Other suitable mammalian promoters include heteromammalian promoters, such as heat shock promoters and actin promoters.

[0163] Further potential promoters include, but are not limited to, the SV40 initial promoter (Benoist and Chambon, 1981, Nature 290:304-310); the CMV promoter (Thornsen et al., 1984, Proc. Natl. Acad. USA 81:659-663); the promoter included in the long 3' terminal repeat of Rous sarcoma virus (Yamamoto et al., 1980, Cell 22:787-797); the herpesthymidine kinase promoter (Wagner et al., 1981, Proc. Natl. Acad. Sci. USA 78:1444-1445); promoters and regulatory sequences derived from the metallothionein gene (Prinster et al., 1982, Nature 296:39-42); and prokaryotic promoters such as the beta-lactamase promoter (Villa-Kamaroff et al.) Examples include al., 1978, Proc. Natl. Acad. Sci. USA 75:3727-3731; or the tac promoter (DeBoer et al., 1983, Proc. Natl. Acad. Sci. USA 80:21-25). The following animal transcriptional regulatory regions, which exhibit tissue specificity and are utilized in transgenic animals, are also included: the elastase I gene regulatory region, which is active in pancreatic acinar cells (Swift et al., 1984, Cell 38:639-646; Ornitz et al., 1986, Cold Spring Harbor Symp. Quant. Biol. 50:399-409; MacDonald, 1987, Hepatology 7:425-515); the insulin gene regulatory region, which is active in pancreatic beta cells (Hanahan, 1985, Nature 315:115-122); and the immunoglobulin gene regulatory region, which is active in lymphocytes (Grosschedl et al., 1984, Cell 38:647-658; Adames et al., 1985, Nature 318:533-538; Alexander et al., 1987, Mol. Cell. Biol.7:1436-1444); Mouse mammary tumor virus regulatory region active in testes, mammary glands, lymphocytes and mast cells (Leder et al., 1986, Cell 45:485-495); Albumin gene regulatory region active in the liver (Pinkert et al., 1987, Genes and Devel. 1:268-276); Alpha-fetoprotein gene regulatory region active in the liver (Krumlauf et al., 1985, Mol.Cell.Biol. 5:1639-1648; Hammer et al., 1987, Science 253:53-58); Alpha-1 antitrypsin gene regulatory region active in the liver (Kelsey et al., 1987, Genes and Devel. 1:161-171); Beta-globin gene regulatory region active in bone marrow cells (Mogram et al., 1985, Nature 315:338-340; Kollias et al., 1986, Cell 46:89-94); myelin basic protein gene regulatory region active in oligodendrocyte cells in the brain (Readhead et al., 1987, Cell 48:703-712); myosin light chain-2 gene regulatory region active in skeletal muscle (Sani, 1985, Nature 314:283-286); and gonadotropin-releasing hormone gene regulatory region active in the hypothalamus (Mason et al., 1986, Science 234:1372-1378).

[0164] Enhancer sequences can be inserted into vectors to increase transcription of DNA encoding components of multispecific antigen-binding proteins (e.g., light chain, heavy chain, modified heavy chain, Fd fragment) by higher eukaryotes. Enhancers are typically cis-acting elements of DNA, approximately 10–300 bp in length, that act on promoters to increase transcription. Enhancers are relatively directional and position-independent and are found at both 5' and 3' positions relative to the transcription unit. Several enhancer sequences are known to be available from mammalian genes (e.g., globin, elastase, albumin, alpha-fetoprotein, and insulin). However, viral enhancers are typically used. The SV40 enhancer, cytomegalovirus initial promoter enhancer, polyoma enhancer, and adenovirus enhancer, known in the art, are exemplary enhancing elements for eukaryotic promoter activation. Enhancers may be located either 5' or 3' of the coding sequence in the vector, but are typically located 5' from the promoter. The secretion of antibodies into extracellular space can be promoted by incorporating sequences encoding appropriate native or heterologous signal sequences (leader sequences or signal peptides) into expression vectors. The choice of signal peptide or leader depends on the type of host cell in which the antibody will be produced, and heterologous signal sequences can replace native signal sequences. Examples of signal peptides are listed above. Other signal peptides that function in mammalian host cells include the interleukin-7 (IL-7) signal sequence described in U.S. Patent No. 4,965,195; the interleukin-2 receptor signal sequence described in Cosman et al., 1984, Nature 312:768; the interleukin-4 receptor signal peptide described in European Patent No. 0367 566; the type I interleukin-1 receptor signal peptide described in U.S. Patent No. 4,968,607; and the type II interleukin-1 receptor signal peptide described in European Patent No. 0460 846.

[0165] The expression vector provided may be constructed from a starting vector such as a commercially available vector. Such a vector may or may not contain all of the desired flanking sequences. If one or more of the flanking sequences described herein are not initially present in the vector, they may be individually obtained and ligated into the vector. The methods used to obtain each of the flanking sequences are well known to those skilled in the art. The expression vector can be introduced into a host cell, whereby a protein can be produced that contains the fusion protein encoded by the nucleic acid described herein.

[0166] In certain embodiments, nucleic acids encoding different components of the multispecific antigen-binding protein of the invention may be inserted into the same expression vector. In such embodiments, the two nucleic acids may be separated under the control of a single promoter by an internal ribosome entry site (IRES) such that the light and heavy chains are expressed from the same mRNA transcript. Alternatively, the two nucleic acids may be under the control of two separate promoters such that the light and heavy chains are expressed from two separate mRNA transcripts.

[0167] Similarly, for the IgG-scFv multispecific antigen-binding protein, the nucleic acid encoding the light chain may be cloned into the same expression vector as the nucleic acid encoding the modified heavy chain (a fusion protein containing the heavy chain and scFv), in which case the two nucleic acids are separated by IRES under the control of a single promoter, or the two nucleic acids are under the control of two separate promoters. For the IgG-Fab multispecific antigen-binding protein, the nucleic acids encoding each of the three components may be cloned into the same expression vector. In some embodiments, the nucleic acid encoding the light chain of the IgG-Fab molecule and the nucleic acid encoding the second polypeptide (containing the other half of the C-terminal Fab domain) are cloned into one expression vector, while the nucleic acid encoding the modified heavy chain (a fusion protein containing the heavy chain and half of the Fab domain) is cloned into a second expression vector. In certain embodiments, all components of the multispecific antigen-binding proteins described herein are expressed from the same host cell population. For example, even if one or more components are cloned into separate expression vectors, both expression vectors can be simultaneously transfected into host cells so that a single cell produces all components of a polybispecific antigen-binding protein.

[0168] After a vector is constructed and one or more nucleic acid molecules encoding components of the multispecific antigen-binding protein described herein are inserted into appropriate sites in one or more vectors, the completed vector can be inserted into a host cell suitable for amplification and / or polypeptide expression. Accordingly, the present invention encompasses isolated host cells containing one or more expression vectors encoding components of the multispecific antigen-binding protein. The term “host cell,” as used herein, means a cell that is transformed with nucleic acid or can be transformed to express the gene of interest. This term includes offspring of parental cells, regardless of whether the morphology or genetic structure of the offspring is identical to that of the original parental cell, as long as the gene of interest is present. A host cell containing the isolated nucleic acid of the present invention, preferably operably ligated to at least one expression regulatory sequence (e.g., a promoter or enhancer), is a “recombinant host cell.”

[0169] Transformation of selected host cells with antigen-binding protein expression vectors can be achieved by well-known methods, including transfection, infection, calcium phosphate coprecipitation, electroporation, microinjection, lipofection, DEAE-dextran mediated transfection, or other known techniques. The method selected will, to some extent, depend on the type of host cell used. These methods and other suitable methods are well known to those skilled in the art and are shown, for example, in Sambrook et al., 2001.

[0170] When cultured under appropriate conditions, host cells synthesize antigen-binding proteins, which can then be harvested from the culture medium (if the host cell secretes it into the medium) or directly from the host cell producing it (if it is not secreted). The selection of a suitable host cell will depend on various factors, including the desired expression level, polypeptide modifications desirable or required for activity (such as glycosylation or phosphorylation), and the ease of folding into a biologically active molecule.

[0171] Exemplary host cells include prokaryotes, yeasts, or higher eukaryotic cells. Prokaryotic host cells include eubacteria such as Gram-negative or Gram-positive microorganisms, such as Enterobacteriaceae, for example Escherichia, for example E. coli, Enterobacter, Erwinia, Klebsiella, Proteus, Salmonella, for example Salmonella typhimurium, and Serratia, for example Serratia marcescens. Examples include the genera *Marcescens*, *Shigella*, and *Bacillus*, such as *B. subtilis* and *B. licheniformis*, *Pseudomonas*, and *Streptomyces*. Eukaryotic microorganisms, such as filamentous fungi or yeasts, are suitable cloning or expression hosts for recombinant polypeptides. *Saccharomyces cerevisiae*, or common baker's yeast, are among the most commonly used lower eukaryotic host microorganisms.However, genera such as Pichia (e.g., P. pastoris), Schizosaccharomyces pombe, Kluyveromyces, Yarrowia, Candida, Trichoderma reesia, Neurospora crassa, and Schwanniomyces (e.g., Schwanniomyces occidentalis) are not included. Aspergillus (occidentalis), as well as filamentous fungi, such as the genera Neurospora, Penicillium, Tolypocladium, and Aspergillus, several other genera, species, and strains such as A. nidulans and A. niger are generally available and useful here.

[0172] Host cells for the expression of glycosylated antigen-binding proteins can be derived from multicellular organisms. Examples of invertebrate cells include plant cells and insect cells. Numerous baculovirus strains and mutants, as well as corresponding insect-acceptable host cells derived from hosts such as the fall armyworm (Spodoptera frugiperda), Aedes aegypti (Aedes aegypti), Asian tiger mosquito (Aedes albopictus), fruit fly (Drosophila melanogaster), and silkworm (Bombyx mori), have been identified. Various virus strains for transfection of such cells, e.g., the L-1 mutant of Autographa californica NPV and the Bm-5 strain of silkworm (Bombyx mori) NPV, are publicly available.

[0173] Vertebrate host cells are also suitable hosts, and recombinant production of antigen-binding proteins from such cells is a common practice. Mammalian cell lines available as hosts for expression are well known in the art and include, but are not limited to, immortalized cell lines available from the American Type Culture Collection (ATCC), such as, but are not limited to, Chinese hamster ovary (CHO) cells, e.g., CHOK1 cells (ATCC CCL61), DXB-11, DG-44, and Chinese hamster ovary cell / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216, 1980); monkey kidney CV1 cell line transformed with SV40 (COS-7, ATCC CRL 1651); human embryonic kidney cell line (293 cells or 293 cells subcloned for growth in suspension culture, (Graham et al., J. Gen Virol. 36:59, 1977)); baby hamster kidney cells (BHK, ATCC CCL 10); Mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251, 1980); Monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); Human cervical cancer cells (HELA, ATCC CCL 2); Canine kidney cells (MDCK, ATCC CCL 34); Buffalo rat liver cells (BRL 3A, ATCC CRL 1442); Human lung cells (W138, ATCC CCL 75); Human hepatocellular carcinoma cells (Hep G2, HB 8065); Mouse mammary tumor cells (MMT 060562, ATCC CCL 51); TRI cells (Mather et al., Annals NY Acad. Sci. 383:44-68, 1982); MRC 5 cells or FS4 cells; Mammalian myeloma cells, and many other cell lines are mentioned. In another embodiment, a cell line derived from a B cell lineage that does not produce its own antibodies but has the ability to produce and secrete heterologous antibodies can be selected. In some embodiments, CHO cells are preferred host cells for expressing the multispecific antigen-binding protein of the present invention.

[0174] For the production of multispecific antigen-binding proteins, host cells are transformed or transfected with the nucleic acids or vectors described above and cultured in conventional nutrient media modified to be suitable for promoter induction, transformant selection, or amplification of genes encoding a desired sequence. In addition, novel vectors and transfected cell lines having multiple copies of transcription units separated by a selection marker are particularly useful for the expression of antigen-binding proteins. Accordingly, the present invention also provides a method for preparing a multispecific antigen-binding protein as described herein, comprising the steps of culturing host cells containing one or more expression vectors as described herein in a culture medium under conditions that enable the expression of a multispecific antigen-binding protein encoded by the one or more expression vectors, and recovering the multispecific antigen-binding protein from the culture medium.

[0175] The host cells used to produce the antigen-binding protein of the present invention can be cultured in a variety of media. Commercial media such as Ham F10 (Sigma), Minimum Essential Medium (MEM, Sigma), RPMI-1640 (Sigma), and Dulbecco's Modified Eagle Medium (DMEM, Sigma) are suitable for culturing host cells. Furthermore, any of the media described in Ham et al., Meth. Enz. 58:44, 1979; Barnes et al., Anal. Biochem. 102:255, 1980; U.S. Patent No. 4,767,704; U.S. Patent No. 4,657,866; U.S. Patent No. 4,927,762; U.S. Patent No. 4,560,655; or U.S. Patent No. 5,122,469; International Publication No. 90103430; International Publication No. 87 / 00195; or U.S. Reissue Patent No. 30,985 can be used as the culture medium for host cells. Any of these culture media may be supplemented as needed with hormones and / or other growth factors (such as insulin, transferrin, or epidermal growth factor), salts (such as sodium chloride, calcium, magnesium, and phosphates), buffers (such as HEPES), nucleotides (such as adenosine and thymidine), antibiotics (such as gentamicin®), trace elements (usually defined as inorganic compounds present at final concentrations in the micromolar range), and glucose or equivalent energy sources. Any other necessary nutritional supplements may also be included in appropriate concentrations known to those skilled in the art. Culture conditions such as temperature and pH are those previously used with host cells selected for expression and will be obvious to those skilled in the art.

[0176] When host cells are cultured, bispecific antigen-binding proteins can be produced intracellularly, in the perimembrane space, or directly secreted into the culture medium. If the antigen-binding protein is produced intracellularly, the first step is to remove granular host cells or lysed fragments, for example, by centrifugation or ultrafiltration. The bispecific antigen-binding protein can be purified, for example, by hydroxyapatite chromatography, cation or anion exchange chromatography, or preferably by affinity chromatography using the target antigen or protein A or protein G as the affinity ligand. Protein A can be used to purify proteins containing polypeptides based on human γ1, γ2, or γ4 heavy chains (Lindmark et al., J.Immunol.Meth.62:1-13, 1983). Protein G is recommended for all mouse isotypes and human γ3 (Guss et al., EMBO J.5:15671575, 1986). The matrix to which the affinity ligand binds is most often agarose, but other matrices are also available. Mechanically stable matrices such as controlled-pore glass or poly(styrenedivinyl)benzene allow for faster flow rates and shorter processing times than can be achieved using agarose. When the protein contains a CH3 domain, Bakerbond ABX® resin (JTBaker, Phillipsburg, NJ) is useful for purification. Depending on the specific multispecific antigen-binding protein to be recovered, other techniques for protein purification, such as ethanol precipitation, reverse-phase HPLC, chromatographic focusing, SDS-PAGE, and ammonium sulfate precipitation, are also possible. [Examples]

[0177] Materials and methods Plasmid construction The antibody HC and LC genes were synthesized using Twist Bioscience, and then individually cloned into mammalian transient expression vectors using the Golden Gate assembly method. 30To reduce protein heterogeneity, all HCs were constructed using human IgG1 scaffolds (IgG1-SEFL2) with non-glycosylation mutations and novel, manipulated disulfide bonds. 31 For hetero-Fc molecules requiring HC heterodimerization, charge-pair mutations (CPMs) were introduced into the Fc region. After Sanger sequencing confirmation, transfection-grade DNA was prepared using the Qiagen Maxi plasmid purification kit, and then mixed with monoclonal antibodies and hybrid IgG (non-homogeneous HC / LC pairs) in a 1:1 (HC:LC) ratio, quadruple-stranded hetero-Fc in a 1:1:1 (HC1:LC1:HC2:LC2) ratio, and cLC hetero-Fc (triple-stranded hetero-Fc) in a 2:1:1 (cLC:HC1:HC2) ratio.

[0178] Cell culture and protein expression All proteins were transiently expressed in a suspension of human embryonic kidney HEK293-6E cells (NRC-BRI) using an improved in-house protocol (Ref. Grace's paper). Briefly, the cells were maintained in FreeStyle F-17 medium (Thermo Fisher) containing 0.1% Kolliphor P188 (Sigma), 25 μg / ml G418 (Gibco), and 6 mM L-glutamine (Invitrogen). For optimal transfection, subculture the cells 26 hours before transfection, and apply 2 × 10⁶ cells per ml. 6 A living cell density of 1 was achieved. 0.5 μg of DNA per 1 ml of cells was conjugated with 1.5 μl of PEImax reagent (Polysciences) in 100 μl of FreeStyle F-17 medium for 10 minutes, and then added to the cell culture. One day after transfection, tryptone N1 solution (Organotechnie) and glucose (Thermo Fisher) were added to the cell culture to final concentrations of 2.5 g / L and 4.5 g / L, respectively. Three days later, 3.75 mM sodium valproate (MP Biomedicals) was added to enhance protein expression. Six days after transfection, the acclimatization medium was collected for purification.

[0179] High-throughput protein purification using ProA magnetic beads The KingFisher® Flex system (Thermo Fisher) was used for high-throughput protein purification using magnetic ProA beads (GE Life Sciences). Briefly, 4 ml of HEK293-6E cells in a 24-well deep block were transfected for protein expression, followed by the addition of 100 μl of magnetic ProA beads, and harvested after 1 day. The beads were then collected and purified using a 24-well magnetic head with KingFisher. After washing three times with PBS and twice with Milli-Q water, the proteins were eluted with 500 μl of 100 mM sodium acetate (pH 3.6) for 10 minutes, and then immediately neutralized by adding 10 μl of 3 M Tris (pH 11.0).

[0180] Two-step purification using ProA and cation exchange chromatography. Proteins expressed in 40 ml of HEK293-6E cells were purified using ProA affinity capture (1 ml of HiTrap MabSelect SuRe, GE Life Sciences), eluted with 100 mM sodium acetate (pH 3.6), and immediately afterward, buffer exchange was performed using a 5 ml HiTrap desalting column (GE Life Sciences) to 10 mM sodium acetate and 150 mM NaCl (pH 5.2), as described above.32 .

[0181] For ion-exchange chromatography, ProA eluate (1.5–1.8 ml) was diluted with 20 ml of 20 mM MES (pH 6.2) and loaded onto a 1 ml cation exchange column (SP-HP HiTrap, GE Life Sciences) at 1 ml / min. The column was washed with 8 column volumes of the same buffer at 1 ml / min, and the protein was eluted over 40 column volumes at 0.4 ml / min with a linear gradient of 0–400 mM NaCl. Fractions with purity of 90% or higher (measured by SEC and MS) were pooled, and their concentrations were measured using a Multiscan GO microplate reader (Thermo Fisher). 33 .

[0182] Non-reducing SDS-PAGE To analyze the purified ProA samples, 1–2 μg of protein was loaded onto Novex 4–20% Tris-glycine gel (Invitrogen) in the absence of a reducing agent. Each gel contained a PageRuler-stained protein ladder of 10–250 kD (Thermo Fisher). After electrophoresis at 150 V for 1 hour, the gels were stained with InstantBlue Coomassie protein stain (Expedeon), briefly washed with Milli-Q water, and then imaged using a Gel Doc system (Bio-Rad).

[0183] Protein mass spectrometry by liquid chromatography-mass spectrometry (LC-MS) The quantification of protein species was performed using a high-resolution LC-MS system, with some modifications, as described above. 34In short, approximately 15 μg of each purified sample was analyzed by non-reducing LC-MS to maintain sample integrity and preserve chain pairing information. The LC-MS system consisted of an Agilent 1290 Infinity II UPLC connected to an Agilent 6224 ESI-TOF mass spectrometer. Chromatographic separation was performed using a Zorbax RRHD 300SB-C8 2.1×50 mm, 1.8 μm UPLC column heated to 70°C at a flow rate of 0.5 ml / min. The MS method scanned m / z [1000-7000] acquiring 0.7 spectra / second. The obtained spectra were summed and then deconvolved using Agilent Mass Hunter Qualitative Analysis software (version B.07.00) or the Protein Metrics Intact Program module. An Excel-based tool was used to calculate the intact mass and intensity of various correct and mispaired species. In the case of 4-stranded heterozygote Fc, the following species are calculated: IgG species with four distinct strands, IgG species with an HC heterodimer but with one LC 2×, IgG species with an HC homodimer, and 1 / 2 Ab species.

[0184] Characterization of binding affinity Binding affinity (K) of purified antibody (bispecific or monoclonal) against soluble antigen D To measure the equilibrium dissociation constant (K), antibodies were first captured on a streptavidin-SAX biosensor chip using a biotinylated polyclonal capture antibody (Jackson Immuno Research), and then incubated with a series of dilutions of each soluble antigen. This assay format was chosen so that the bivalent antibody was immobilized and present on the biosensor chip, and tested against the same serial dilution of each soluble antigen. Thus, the measured quantitative K was obtained. DAffinity exhibits a monovalent 1:1 binding interaction and can be directly compared. Experiments were performed on a ForteBio Octet HTX instrument using 96-chip mode with a standard data acquisition rate of 5 Hz at 27°C and 1000 RPM. Using Genedata Screener V16 software, the raw Octet binding data was processed with the installed SPR rate curve fitting package and globally fitted to a 1:1 binding model to determine the association rate constant (k a ) and dissociation rate constant (k d ) was determined. Next, the equilibrium dissociation constant (K D ) to k d / k a It was calculated as a ratio.

[0185] result The diversity of antibody variable regions provides a theoretical basis for evaluating linkage selectivity. The IgG molecule consists of two fragment antigen-binding (Fab) regions that recognize epitopes and a fragment crystallizable (Fc) region that interacts with receptors on the cell surface (Figure 1A). In the Fab region, two major contact points mediate HC / LC interactions: the VH / VL domain and the CH1 / CL domain. While antibodies largely conserve the CH1 / CL interface (only LC kappa / lambda diversity is considered), the VH-VL interface is highly variable and unique to each antibody. In fact, both the framework (FW) and complementarity-determining region (CDR) directly engage across the interface here (Figure 1A). To further evaluate the intrinsic properties of the VH / VL interface, six X-ray structures of antibodies generated against various relevant therapeutic targets were analyzed. Across these six structures, approximately 100 (94–112) residues are located at the HC / LC interface (1,620–1,880 Å). 2CDRs intersect the embedded surface area of ​​the HCl, with over 40% located at the VH / VL interface, and CDRs account for approximately 20% of the total (Figure 1B). Of all six CDRs, CDR-H2 and CDR-H3 in HC and CDR-L3 in LC have the most contacting residues, accounting for over 15% of the overall HC / LC interactions. Since CDR-H3 and CDR-L3 are major determinants of antibody target specificity, they exhibit high variability and low sequence identity (11.1% and 23.3%, respectively (Figure 1C)). Due to the fact that highly diverse CDRs can contribute to HC / LC interactions, it was hypothesized that there may be more favorable HC / LC pairs and interfaces than others. Furthermore, since VH / VL interactions are reported to be the first step in the assembly of the quaternary IgG structure, 21、22 A favorable VH / VL pair can be key to priming congenital HC / LC pairings. Therefore, since correct HC / LC pairings are crucial for the assembly of quadruple heterozygotes, it is reasonable to explain why a parent mAb that shows higher priority for congenital HC / LC pairings than non-congenital ones can function as an ideal building block. At the same time, strands shown to be nonspecific (i.e., indiscriminate LCs) provide an opportunity to explore cLCs. Thus, a deeper understanding of the inherent properties of antibody sequences can facilitate the creation of bispecific antibodies better suited to design objectives.

[0186] Explanation of the linkage selectivity screening method To facilitate the development of IgG-like bispecific compounds, a high-throughput screening process (CSA) to evaluate HC / LC selectivity is envisioned (Figure 2A). Antibody HC is secreted only when it binds to LC. 21、22The expression level of a given HC / LC pair may correlate with the HC / LC pairing efficiency. Starting with two parental antibody panels (anti-target A and anti-target B), high-throughput CSA was performed in two different scenarios (competitive and non-competitive) to evaluate the specificity in the assembly of hetero-Fc (four-strand) and identify indiscriminate LCs of cLC hetero-Fc (three-strand) respectively (Figures 2A and 2B). The cCSA experiments mimicked co-expression scenarios of four different strands (2×HC and 2×LC), resulting in all possible combinations between anti-target A and anti-target B parental antibodies. The expressed hetero-Fc was purified from conditioned medium containing ProA beads, and the correct HC / LC pairing rate was quantified by high-resolution liquid chromatography-mass spectrometry (LC-MS) (Figures 2B and 2B). It is noteworthy that two of the four heterozygous Fc species may exhibit the same MW, one of which represents a scenario in which both LCs (LCA and LCB) have complete cross-pairing with the opposite non-congenital HC (Figure 2B). Although LC-MS cannot distinguish such species, this scramble of LCs is rarely seen even after limited proteolysis, strongly suggesting that accurate MWs detected by LC-MS indicate correct HC / LC pairing. A high proportion of correct species is an indicator of the preference for congenital HC / LC pairing over non-congenital. Antibody combinations with high levels of correct species are selected as optimal building blocks for bispecifics requiring correct / specific HC / LC pairing (Figure 2B). For ncCSA experiments, each single HC from anti-target A pairs with each LC from anti-target B in a 1:1 ratio, and vice versa. High expression levels measured by ProA capture (comparable to or greater than HC / LC congeneral pairs) indicate that LCs pair well and non-congeneral HCs fold and are secreted (Figure 2B). In contrast, low expression levels suggest that the VH / VL interfaces of these chains are specific to their congeneral interfaces and consequently incompatible with others. Next, using the same ProA purified material, a second high-throughput step was incorporated to characterize binding affinity using ForteBio Octet.This protocol enables the rapid (less than 5 weeks) and efficient screening and identification of rare and valuable LCs that can be used as building blocks for cLC bispecific antibodies.

[0187] Identification of low cross-pairing antibody combinations using competitive CSA method To validate the cCSA method, two parental antibody panels were selected (eight anti-target AAb (A1-A8) and four anti-target BAb (B1-B4)). Mutations were introduced to drive heavy-heavy chain pairing, and antibodies from each target were combined to obtain 32 combinations for evaluation. After purifying secreted IgG from conditioned medium, the expression levels of the combinations and parental mAbs were measured by A280 (Figures 3A and 16). In particular, the parental mAbs showed large variability in expression levels ranging from 50 to 250 mg / L. Surprisingly, most antibody combinations were expressed at levels above 100 mg / L, with a few exceptions (A3×B1, A3×B3, A3×B4, and A6×B1) (which may be related to low-expression parental A3 and B1). To identify whether the constructed and secreted species had correct HC / LC pairing, purified ProA proteins were analyzed using non-reducing LC-MS (illustrated in Figure 3B). By using the Fc region of the molecule for purification with ProA beads, only HC-containing molecules are selected, and all other species (e.g., LC dimers) are discarded. LC-MS analysis cannot confirm whether two pairs of congeneral HC / LCs are correctly paired, but it can determine whether each species has a copy of each of the four chains. While not conclusive, this is a necessary condition for the construction and generation of heterozygous Fc. Therefore, we calculated the proportion of each of these three possible HC / LC scenarios: 1 × HC A +1 × HC B +1 × LC A +1 × LC B , 1×HC A +1 × HC B +2×LC A +0 × LC B and 1×HC A +1 × HC B +0 × LC A+2×LC B (All containing HC heterodimers) (Figure 3C). Two combinations (A2×B3 and A4×B3) failed in LC-MS analysis because the MW difference between the two LCs was small (<60 Da), but quantification of IgG species was successful in all of the other 30 combinations. Furthermore, Fc CPM in the Ab combinations was arranged to enhance HC heterodimerization, but small amounts of homodimers and / or 1 / 2 Ab were still present (Figure 8). However, since the focus of this study was on HC / LC pairing, these species were excluded from further analysis. Interestingly, the data showed that in approximately half (17 / 32) of the molecules tested, the proportion of the desired species (1×HC A +1×LC A +1×HC B +1×LC B ) was more than 50% of all species present in the ProA-purified sample (Figure 3C). Surprisingly, in two combinations (A6×B3 and A6×B4), the proportion of the correct product was over 75%. In contrast, the ProA-purified samples of the remaining 13 combinations contained less than 50% of the desired species, and in 5 of these combinations (A1×B1, A2×B1, A3×B2, A4×B1, and A7×B1), the proportion of the desired species was less than 25% (Figure 3C). In most cases (23 / 32), in the undesired products (1×HC A +1×HC B +2×LC A +0×LC B and 1×HC A +1×HC B +0×LC A +2×LC B ), only 1 / 2 of the LC appeared, either one of the LCs was overexpressed, or LC A and LC BIt was suggested that the two compounds may compete with each other during expression and molecular construction, potentially leading to the complete or partial suppression of the other (Figure 3C). Therefore, using this information, the cCSA method can efficiently screen and identify preferred combinations of parental mAbs that possess the intrinsic properties enabling the correct assembly of heterozygous Fc during expression in single cells. This allows for the elimination of candidates while directing efforts towards candidates with more promising properties, saving time and resources.

[0188] Low cross-pairing antibody combinations are good candidates for generating quadruple-stranded heterozygous FCs. In the cCSA method described above, more than half of the combinations (17 / 32) showed a high level (≥50%) of the desired species (1×HC). A +1 × LC A +1 × HC B +1 × LC B) was shown (Figure 3C). However, to evaluate whether this high-throughput method truly predicts correct congeneral HC / LC pairing, we decided to scale up 32 hetero-Fc molecules and perform a two-step purification following ProA followed by cation exchange chromatography (CIEX). The objective was not only to quantify the final yield of the desired species but also to study the separation profile in the CIEX step. From the 32 hetero-Fc molecules, 11 were successfully purified to a final purity of over 90% as determined by SEC and LC-MS (Figures 4A and 8). Interestingly, the final yield of the parental mAbs, which are the building blocks of these hetero-Fc molecules, appears to correlate with the yield of the resulting bispecific molecules. In fact, the hetero-Fc molecules A1×B3, A1×B4, A4×B3, and A8×B3, which showed the highest protein yields, all contained at least one of the highly expressed parental mAbs A1, A4, and A8 (Figure 4A). For final validation, the final pool with LC-MS purity exceeding 90% was evaluated for binding to target A and target B, which are important for assessing congeneral HC / LC pairing. The affinity of each arm in the bispecific molecule for each target was equivalent to the affinity of the parent mAb (Figure 17), confirming that all 11 hetero-Fc molecules possessed the correct HC / LC pairings in both arms.

[0189] Overall, these data are in good agreement with cCSA experiments (Figure 4E). From 11 successfully purified heterozygous Fc molecules, the proportion of correct HC / LC pairing predicted by LC-MS was high for 8 molecules (over 50% correct HC / LC species: A1×B3, A1×B4, A3×B3, A4×B2, A4×B4, A6×B2, A6×B4, and A8×B3), good for A7×B4 (47.4% correct species), and unclear for A2×B3 and A4×B3 (undetermined due to small MW difference between chains). Most importantly, none of the molecules showing low HC / LC pairing by CSA could be purified by CIEX (Figures 3C and 4A). In fact, the proportion of correct species determined by cCSA predicted the Ab combinations that yielded good heterozygous Fc molecules (ROC AUC 0.80, Figure 4B). Furthermore, we attempted to construct a correlation between the HC / LC pairing rate predicted by cCSA and the final yield. As shown in Figure 4C, the majority of combinations (12 / 13) with less than 50% correct HC / LC pairing by cCSA failed during subsequent CIEX purification. One exception was the A7×B4 molecule, which showed a CIEX profile with adequate separation while exhibiting 47.4% pairing (Figure 4D). Interestingly, CIEX failed to purify 9 out of 17 molecules that had 50% or more correct HC / LC pairing as predicted by cCSA (Figure 4C). In fact, to explain this phenomenon, we selected an example of A8×B4 heteroFc that showed 69.4% correct pairing in the cCSA experiment. This molecule consists of two different species (1×HC A +1 × HC B +1 × LC A +1 × LC B and 1×HC A +1 × HC B +2×LC A +0 × LC BThe peaks exhibit a good shape that obscures mispaired species, preventing the observation of resolution between them (Figure 4D). Therefore, while cCSA can predict high-performance hetero-Fc molecules based on the intrinsic VH / VL interface, properties not screened by cCSA (e.g., properties that affect the separation profile of ion-exchange chromatography columns) also play an important role in selecting hetero-Fc molecules.

[0190] Screening of common LCs using non-competitive CSA method As shown above, some LCs preferably pair with their congeneral HCs, but other LCs can also efficiently bind to non-congeneral HCs. If the resulting non-congeneral HC / LC pair retains the binding, such LCs can function as common LCs (cLCs). The ncCSA method is envisioned as an opportunistic approach to evaluate whether a parent mAb provides such LCs (Figure 2B). To demonstrate the effectiveness of this method, two bispecificity programs (A×B and C×B) were selected, containing eight anti-target A parent mAbs, four anti-target B parent mAbs, and ten anti-target C parent mAbs. As shown in Figure 9, each LC of anti-target B paired with an individual HC derived from either anti-target A or anti-target C mAb. On the other hand, each HC of anti-target B was paired with an individual LC derived from a different anti-target A or anti-target C mAb. The 144 non-homogeneous HC / LC pairs obtained were expressed in 293 6E along with 22 parental mAbs (controls), purified with ProA beads, and subsequently analyzed using non-reducing SDS-PAGE gel and A280 quantitative analysis (Figure 10). Further analysis of ProA yield for these 144 non-homogeneous HC / LC pairs (Figures 5A and 9) showed that only 38 of the 144 combinations (26.4%) exhibited a significant decrease in expression levels (less than 50% compared to the control), suggesting broad indifferent behavior within the HC / LC pairs. Of the remaining 106 combinations, 68 (47.2%) of the HC / LC non-homogeneous pair molecules showed higher expression levels than their corresponding parental mAb controls (Figure 5A). However, many LCs were not broadly indifferent. A closer look at two examples will help explain this phenomenon. When LC-B1~4 were paired with HC-A7, protein expression was significantly lower compared to the congeneral LC-A7 (control) (Figure 5B). In contrast, the expression level of the same anti-target B LC paired with HC-A8 was equivalent to or higher than that of the congeneral LC-A8 (control). Therefore, this suggests that anti-target B LCs are indiscriminate towards HC-A8 but not towards HC-A7, highlighting the role that HC plays in determining LC cross-pairing.

[0191] Coupling analysis to identify common LC heterozygous FC candidates While combinations of well-expressed non-homogeneic Abs are promising candidates for constructing cLC bispecifics, expression levels alone do not provide insight into their function. To select functional bispecifics, selected indiscriminate LCs (showing expression levels of ≥50% compared to homogeneic controls in ncCSA assays (Figure 5A)) were assembled into hetero-Fc morphologies containing both homogeneic and non-homogeneic HCs with Fc CPM to promote HC dimerization (Figure 5C). These cLC hetero-Fc morphologies were evaluated using high-throughput binding assays to identify candidates capable of binding to both targets. After high-throughput expression in 4 mL deep-well blocks (DWBs) using HEK293-6E cells, the cLC hetero-Fc morphologies were purified with ProA. The yield of 92 out of 106 molecules (86.8%) was approximately 100 mg / L, which is comparable to the parent mAb (Figures 3A and 11), while only 14 cLC heterozygous Fc molecules (13.2%) showed a ProA yield of less than 60 mg / L (Figures 5D and 5E). All molecules containing cLC B1 showed remarkably low expression, suggesting that this LC may act as a limiting factor in the overall expression of these bispecifics (Figure 5D). In fact, the ProA yield of the 41.6 mg / L B1 antibody was the lowest among all parent mAbs used as building blocks for the generation of these bispecifics (Figure 3A). While the A3 and A6 parental mAbs also showed relatively low ProA recovery (60.3 and 69.3 mg / L, respectively), the cLC hetero-Fc containing these two building blocks exhibited acceptable protein yields when combined with any B parent other than B1, suggesting that non-homogeneic HCs may have rescued the expression levels in this case (Figure 5D). Another important observation is that the cLC hetero-Fc (A×B and C×B) also showed approximately a twofold increase in correct pairing overall compared to the quadruple-stranded hetero-Fc immediately after ProA purification, highlighting the effect of HC / LC pairing on the production levels of these molecules (Figure 12).

[0192] Next, for rapid binding screening, these one-step purified samples were evaluated by ForteBio Octet. To minimize interference from residual impurities, cLC heteroFc molecules were first captured via the Fc region into a streptavidin optical fiber biosensor containing a biotinylated anti-human IgG Fc polyclonal antibody, and then loaded with soluble antigens A, B, or C for incubation. As expected, all cLC heteroFc molecules showed binding to their respective targets via congenital HC / LC arms with affinity comparable to the parent mAb (Figure 5F). Two cLC heteroFc molecules (A2×B4 and C4×B3) also showed detectable binding via non-congenital HC / LC arms recognizing target A or target C (Figure 5F). In the case of A2×B4, the B4 LC paired with both HCs (A2 and B4), while in the case of C4×B3, both HCs (C4 and B3) paired with the B3 LC. Notably, both of these cLCs were generated against target B. While this non-standard binding is lower than the single-digit nM binding typically observed with parent mAbs (Figure 13), it demonstrates how ncCSA offers a new opportunity to identify LCs with unique structural features that enable highly efficient pairing with non-homogeneous HCs (Figure 5G). Furthermore, rapid binding analysis can reveal rare cLCs that also retain binding to novel epitopes. The manufacturability of IgG-like bispecifics is often challenging, as production levels below those of mAbs. 23The expression and purification characteristics of these cLC heteroFc molecules were investigated. To better mimic the scale and purification process required for therapeutic candidates, these two molecules were expressed in 250 mL of HEK293-6E cells and subjected to a two-step purification using ProA followed by CIEX, meeting a purity target of over 95%. In particular, the level of protein secretion by ProA was more than twice as high for these two cLC heteroFc molecules compared to the parent mAb (Figure 14). More importantly, these cLC heteroFc molecules showed final yields comparable to or higher than the parent mAb (Figure 6A), and all showed purity exceeding 97% of the desired species (Figure 14). Furthermore, these bispecific molecules showed a favorable CIEX profile with the correct species readily separated from impurities (Figure 6B). Subsequently, binding assays were repeated using well-purified cLC heteroFc molecules to confirm affinity for each antigen. As initially observed (Figure 5F), these two molecules exhibited binding affinity to antigen A or antigen C via their non-homogeneous HC / LC arms, while retaining binding properties in their homogeneous arms for antigen B (Figures 6C and S7). To validate the affinities measured for these cLC hetero-Fc molecules, two hybrid IgGs (HC-A2 / LC-B4 and HC-C4 / LC-B3, respectively), composed of non-homogeneous HC and LC, were expressed and purified. The comparable affinities of the hybrid molecules to antigens A and C via their non-homogeneous arms (Figure 6D) further confirmed cLC hetero-Fc binding. The binding signal of the hybrid IgG was approximately twice as high as the signal observed in the non-homogeneous arms of the cLC hetero-Fc, consistent with the number of binding sites present in these molecules (2:1, respectively). Furthermore, since neither appears to retain binding to antigen B, it is suggested that the binding ability of the hybrid IgG is primarily driven by the HC CDR and not by the LC. Conversely, to rule out the possibility of nonspecific binding to antigen A or antigen C by the congeneral arms in cLC heterozygous Fc, the binding of B4 and B3 to the parent mAb was tested.As shown in Figure 6E, neither mAb B4 nor B3 bound to these antigens, further demonstrating that the binding detected for non-homogeneous arms did not originate from nonspecific interactions between homogeneous arms and antigen A or antigen C, nor was it a result of cLC alone.

[0193] In summary, the ncCSA method successfully identified two rare LCs from a pool of mAbs paired with non-homogeneous HCs exhibiting mAb-like productivity, while also allowing the HCs in this non-homogeneous HC / LC arm to retain their binding activity.

[0194] All references cited herein are incorporated herein by reference. 1. Labrijn, AF, Janmaat, ML, Reichert, JM & Parren, P. Bispecific antibodies: a mechanistic review of the pipeline. Nat Rev Drug Discov 18,585-608(2019). 2.Lu,RMet al.Development of therapeutic antibodies for the treatment of diseases.J Biomed Sci 27,1(2020). 3.Fan, G., Wang, Z., Hao, M. & Li, J. Bispecific antibodies and their applications.J Hematol Oncol 8,130(2015). 4. Wolf, E., Hofmeister, R., Kufer, P., Schlereth, B. & Baeuerle, PABiTEs: bispecific antibody constructs with unique anti-tumor activity. Drug Discov Today 10, 1237-1244 (2005). 5.Kantarjian,H.et al.Blinatumomab versus Chemotherapy for Advanced Acute Lymphoblastic Leukemia.N Engl J Med 376,836-847(2017). 6.Brinkmann,U.& Kontermann,R.E.The making of bispecific antibodies.MAbs 9,182-212(2017). 7.Wang,Q.et al.Design and Production of Bispecific Antibodies.Antibodies(Basel)8(2019). 8.Spiess,C.,Zhai,Q.& Carter,P.J.Alternative molecular formats and therapeutic applications for bispecific antibodies.Mol Immunol 67,95-106(2015). 9.Suresh,M.R.,Cuello,A.C.& Milstein,C.Bispecific monoclonal antibodies from hybrid hybridomas.Methods Enzymol 121,210-228(1986). 10.Carter,P.Bispecific human IgG by design.J Immunol Methods 248,7-15(2001). 11.Ridgway,J.B.,Presta,L.G.& Carter,P.‘Knobs-into-holes’ engineering of antibody CH3 domains for heavy chain heterodimerization.Protein Eng 9,617-621(1996). 12.Davis,J.H.et al.SEEDbodies:fusion proteins based on strand-exchange engineered domain(SEED)CH3 heterodimers in an Fc analogue platform for asymmetric binders or immunofusions and bispecific antibodies.Protein Eng Des Sel 23,195-202(2010). 13.Gunasekaran,K.et al.Enhancing antibody Fc heterodimer formation through electrostatic steering effects:applications to bispecific molecules and monovalent IgG.J Biol Chem 285,19637-19646(2010). 14.Dillon,M.et al.Efficient production of bispecific IgG of different isotypes and species of origin in single mammalian cells.MAbs 9,213-230(2017). 15.Shiraiwa,H.et al.Engineering a bispecific antibody with a common light chain:Identification and optimization of an anti-CD3 epsilon and anti-GPC3 bispecific antibody,ERY974.Methods 154,10-20(2019). 16.Merchant,A.M.et al.An efficient route to human bispecific IgG.Nat Biotechnol 16,677-681(1998). 17.Krah,S.et al.Generation of human bispecific common light chain antibodies by combining animal immunization and yeast display.Protein Eng Des Sel 30,291-301(2017). 18.Garces,F.et al.Molecular Insight into Recognition of the CGRPR Complex by Migraine Prevention Therapy Aimovig(Erenumab).Cell Rep 30,1714-1723 e1716(2020). 19.Bruggemann,M.et al.Human antibody production in transgenic animals.Arch Immunol Ther Exp(Warsz)63,101-108(2015). 20.Foltz,I.N.,Gunasekaran,K.& King,C.T.Discovery and bio-optimization of human antibody therapeutics using the XenoMouse(R)transgenic mouse platform.Immunol Rev 270,51-64(2016). 21.Feige,M.J.et al.An unfolded CH1 domain controls the assembly and secretion of IgG antibodies.Mol Cell 34,569-579(2009). 22.Feige,M.J.& Buchner,J.Principles and engineering of antibody folding and assembly.Biochim Biophys Acta 1844,2024-2031(2014). 23.Qiong Wang,Y.C.,Jaeyoung Park ,Xiao Liu,Yifeng Hu,Tiexin Wang,& Betenbaugh,K.M.a.M.J.Design and Production of Bispecific Antibodies.Antibodies(2019). 24.Schaefer,W.et al.Immunoglobulin domain crossover as a generic approach for the production of bispecific IgG antibodies.Proc Natl Acad Sci U S A 108,11187-11192(2011). 25.Van Blarcom,T.et al.Productive common light chain libraries yield diverse panels of high affinity bispecific antibodies.MAbs 10,256-268(2018). 26.Feige,M.J.,Hendershot,L.M.& Buchner,J.How antibodies fold.Trends Biochem Sci 35,189-198(2010). 27.Saphire,E.O.et al.Crystal structure of a neutralizing human IGG against HIV-1:a template for vaccine design.Science 293,1155-1159(2001). 28.Scapin,G.et al.Structure of full-length human anti-PD1 therapeutic IgG4 antibody pembrolizumab.Nat Struct Mol Biol 22,953-958(2015). 29.Krissinel,E.&Henrick,K.Inference of macromolecular assemblies from crystalline state.J Mol Biol 372,774-797(2007). 30.Engler,C.,Kandzia,R.& Marillonnet,S.A one pot,one step,precision cloning method with high throughput capability.PLoS One 3,e3647(2008). 31.Jacobsen,F.W.et al.Engineering an IgG Scaffold Lacking Effector Function with Optimized Developability.J Biol Chem 292,1865-1875(2017). 32.Yoo,D.,Provchy,J.,Park,C.,Schulz,C.& Walker,K.Automated high-throughput protein purification using an AKTApurifier and a CETAC autosampler.J Chromatogr A 1344,23-30(2014). 33.Winters,D.,Chu,C.& Walker,K.Automated two-step chromatography using an AKTA equipped with in-line dilution capability.J Chromatogr A 1424,51-58(2015). 34.Spahr,C.S.et al.Discovery,characterization,and remediation of a C-terminal Fc-extension in proteins expressed in CHO cells.MAbs 10,1291-1300(2018).

Claims

1. A method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) The CDRs of the three heavy chain CDRs that specifically bind to the first antigen, (ii) The CDRs of the three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) The CDRs of the three light chain CDRs that specifically bind to the first antigen, (b) The CDRs of the three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody constructs, each including the heavy chain CDR and the light chain CDR of (c)(iii) that bind to each antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells; (e) A step of purifying each multispecific antibody construct; (f) (i) a step of measuring the expression level of each multispecific antibody construct, and (ii) a step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen, (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of the three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of the three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). A method that includes this.

2. The method according to claim 1, wherein the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen) are selected from the group consisting of two antibody Fab fragments, scFvs, or combinations thereof; three antibody Fab fragments, scFvs, or combinations thereof; four antibody Fab fragments, scFvs, or combinations thereof; five antibody Fab fragments, scFvs, or combinations thereof; six antibody Fab fragments, scFvs, or combinations thereof; seven antibody Fab fragments, scFvs, or combinations thereof; eight antibody Fab fragments, scFvs, or combinations thereof; nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

3. The method according to claim 1 or 2, wherein the plurality of antibody Fab fragments, scFvs, or combinations thereof (where each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen) are selected from the group consisting of two antibody Fab fragments, scFvs, or combinations thereof; three antibody Fab fragments, scFvs, or combinations thereof; four antibody Fab fragments, scFvs, or combinations thereof; five antibody Fab fragments, scFvs, or combinations thereof; six antibody Fab fragments, scFvs, or combinations thereof; seven antibody Fab fragments, scFvs, or combinations thereof; eight antibody Fab fragments, scFvs, or combinations thereof; nine antibody Fab fragments, scFvs, or combinations thereof; and at least ten antibody Fab fragments, scFvs, or combinations thereof.

4. The method according to any one of claims 1 to 3, wherein the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

5. The method according to any one of claims 1 to 4, wherein the mammalian host cells are selected from the group consisting of Chinese hamster ovary ("CHO") cells, monkey kidney CV1 strain transformed with SV40 ("COS-7"), human fetal kidney strain HEK293-6E ("HEK293-6E"), human fetal kidney strain HEK293 ("HEK293"), baby hamster kidney cells ("BHK"), mouse Sertoli cells ("TM4"), monkey kidney cells ("CV1"), African green monkey kidney cells ("VERO-76"), human cervical cancer cells ("HELA"), canine kidney cells ("MDCK"), buffalo rat liver cells ("BRL"), human embryonic cells ("W138"), human hepatoma cells ("Hep G2"), mouse mammary cancer cells ("MMT"), TRI cells, MRC 5 cells, and FS4 cells.

6. The method according to any one of claims 1 to 5, wherein the expression level is determined by a method selected from the group consisting of A280 measurement, SDS-PAGE, microchip capillary electrophoresis (MCE), Bradford assay, and bicinchoninic acid (BCA) assay.

7. The method according to any one of claims 1 to 6, wherein the binding affinity of each multispecific antibody construct to the first antigen and the second antigen is measured using Octet, Forte Bio, Carterra LSA, SPR, and flow cytometry.

8. The method according to any one of claims 1 to 7, wherein each multispecific antibody construct is purified by purification of protein A, lambda and kappa resin, and affinity tag.

9. A method for selecting a multispecific antibody construct, (a) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a first antigen; (b) A step of obtaining a plurality of at least two antibody Fab fragments, scFvs, or combinations thereof, wherein each Fab fragment and scFv comprises three heavy chain CDRs and three light chain CDRs that specifically bind to a second antigen; (c) Vector (i) The CDRs of the three heavy chain CDRs that specifically bind to the first antigen, (ii) The CDRs of the three heavy chain CDRs that specifically bind to the second antigen, and (iii) (a) The CDRs of the three light chain CDRs that specifically bind to the first antigen, (b) The CDRs of the three light chain CDRs that specifically bind to the second antigen, and (c) CDRs of three light chain CDRs that do not specifically bind to either the first or second antigen. Three light chain CDRs selected from the group consisting of The process of cloning, The process involves generating multiple vectors that encode multiple multispecific antibody construct modules, each containing the heavy chain CDR and the light chain CDR of (c)(iii) that bind to each antigen; (d) A step of expressing each multispecific antibody construct in mammalian host cells, wherein the mammalian host cells are selected from the group consisting of HEK293-6E cells and CHO cells; (e) A step of purifying each multispecific antibody construct using protein A chromatography; (f) (i) A step of measuring the expression level of each multispecific antibody construct using A280 measurement, and (ii) A step of measuring the binding affinity of each multispecific antibody construct to the first antigen and the second antigen using Octet, (Here, steps (f)(i) and (f)(ii) may be performed simultaneously or in any order); and (g) A step to compare the expression level of (f)(i) and the binding affinity of (f)(ii) for each multispecific antibody construct in order to identify the optimal pairing of the three heavy chain CDRs that specifically bind to the first antigen with the light chain CDR of (c)(iii), and the optimal pairing of the three heavy chain CDRs that specifically bind to the second antigen with the same light chain CDR of (c)(iii). A method that includes this.

10. The method according to claim 9, wherein the multispecific antibody construct module includes at least two modules selected from the group consisting of Fab / Fab heteroFc, scFab / scFab heteroFc, Fab / scFv heteroFc, Fab / Fab-scFv heteroFc, Fab / scFv-Fab heteroFc, Fab / Fab heteroFc-scFv, IgG-Fab, scFab-Fc-Fab, IgG-scFv, scFv-IgG, and Fab-scFv-Fc.

Images