WO2025134341A1 - Method for producing antibodies - Google Patents

Abstract

A method for producing cells that produce multispecific binding molecules, said method including a step for introducing a first vector and a second vector into cells, wherein the first vector includes the genes of all polypeptide chains constituting the multispecific binding molecules, the second vector includes the gene of a polypeptide chain having low expression, and the multispecific binding molecules may be multispecific antibodies.

Description

Methods for producing antibodies

The present invention relates to a method for producing an antibody.

When using recombinant gene technology to produce recombinant antibodies useful as medicines, mammalian cells are often used as host cells for producing recombinant antibodies because they allow for complex post-translational modifications and folding that cannot be performed by prokaryotic cells. Gene introduction into host cells is performed by incorporating an antibody-producing gene into a plasmid to create an expression vector, and then introducing the expression vector into mammalian cells.

Antibody drugs are made using the immunoglobulin scaffold. There are five major classes of immunoglobulins: IgG, IgM, IgD, IgE, and IgA, but almost all antibody drugs approved to date have sequences derived from IgG. An IgG antibody molecule contains four polypeptide chains, two heavy and two light chains. These polypeptide chains form a symmetric Y-shaped structure through a combination of non-covalent interactions and covalent disulfide bonds between the chains. IgG antibody molecules have heavy chains known as gamma chains. Light chains can be one of two types: kappa chains and lambda chains.

When preparing cells for producing a recombinant antibody, one copy each of DNA encoding the antibody's H chain and L chain is typically introduced into a host cell. In contrast, WO 2009/051108 (Patent Document 1) describes a method for producing an antibody using a host cell into which an expression vector containing one copy of DNA encoding the recombinant antibody's H chain and two copies of DNA encoding the L chain is introduced.

WO 2006/060769 (Patent Document 2) describes a method for producing a heterodimeric or heteromultimeric protein using two expression vectors, each encoding a different polypeptide chain. When the protein is an immunoglobulin, the first vector encodes the heavy chain (H chain) and the second vector encodes the light chain (L chain). It describes the investigation into the optimal ratio of heavy chain plasmid to light chain plasmid for maximum antibody productivity.

The basic structure of an IgG antibody molecule is a monospecific molecule consisting of two identical H chains and two identical L chains, with the same binding specificity on both arms (the tips of the F(ab) region). In contrast, multispecific antibodies have been created in recent years, allowing a single antibody molecule to bind to multiple different targets (for example, WO2006/109592 (Patent Document 3)). Multispecific antibodies include bispecific antibodies (BsAb), which bind to two antigens, and multispecific antibodies, which have even more antigen-binding sites.

Bispecific antibodies have two different antigen specificities. Typically, IgG-type bispecific antibodies contain two different H chains and two different L chains. When these four types of polypeptide chains are expressed in cells, random association of the H chains and L chains produces a variety of different antibody species, and nine types of impure antibodies can be formed in addition to the desired combination. The formation of impure antibodies not only reduces the yield of the desired bispecific antibody, but also poses a major problem in industrial production, as the properties of the 10 types of antibodies formed are similar, making it difficult to separate and remove the impure antibodies using conventional methods used in the production of antibody drugs.

When the heavy or light chain of a bispecific antibody is made common, there are three possible combinations of expression products (two types of monospecific antibodies and one type of bispecific antibody), which makes it possible to improve the yield. There is a demand for technology that can further improve the yield of the desired bispecific antibody and purify it to a high purity.

WO 2016/146594 (Patent Document 4) describes a method for purifying a bispecific antibody with a common heavy chain from a mixture of antibodies. This bispecific antibody is constructed from a single heavy chain and two different light chains. A host cell for producing the antibody is transformed with an expression vector (a three-gene expression vector) containing one DNA encoding the common heavy chain, one DNA encoding the first light chain, and one DNA encoding the second light chain.

Transient and Stable CHO Expression, Purification and Characterization of Novel Hetero-Dimeric Bispecific IgG Antibodies: Biotechnol. Prog., 2017, Vol. 33, No. 2, pp. 469-477 (Non-Patent Document 1) describes that the optimal ratio for producing a specific bispecific antibody in CHO cells using a single plasmid vector is 60% common L chain and 20% each of the H chains.

Optimizing assembly and production of native bispecific antibodies by codon de-optimization MABS 2017, VOL. 9, NO. 2, pp. 231-2397 (Non-Patent Document 2) describes that in the production of bispecific antibodies using a common H chain, the yield of the desired bispecific antibody did not improve even when the expression level of a poorly expressed arm was increased by optimizing codons.

Tuning Relative Polypeptide Expression to Optimize Assembly, Yield and Downstream Processing of Bispecific Antibodies; Antibodies 2018, 7, 29 (Non-Patent Document 3) describes the experiment of varying the ratio of L chains in the production of bispecific antibodies using a common H chain.

WO 2009/051108 (Patent No. 4976502) WO 2006/060769 (Patent Publication No. 2008-522589) WO 2006/109592 (Patent No. 4917024) WO 2016/146594 (Patent No. 6894843)

Transient and Stable CHO Expression, Purification and Characterization of Novel Hetero-Dimeric Bispecific IgG Antibodies: Biotechnol. Prog., 2017, Vol. 33, No. 2, P469-477 Optimizing assembly and production of native bispecific antibodies by codon de-optimization MABS 2017, VOL. 9, NO. 2, P231-P2397 Tuning Relative Polypeptide Expression to Optimize Assembly, Yield and Downstream Processing of Bispecific Antibodies; Antibodies 2018, 7, 29

The objective of the present invention is to provide a method for producing bispecific antibodies with high production efficiency.

The inventors discovered that the use of two types of expression vectors in the production of bispecific antibodies with a common L chain improves the production yield and selectivity of the desired heterodimeric bispecific antibodies, and thus completed the present invention.

In other words, the present invention relates to constructing a cell line by transforming a host cell with two types of recombinant vectors in order to produce a multispecific binding molecule such as a bispecific antibody, and to using the resulting cell line to produce a multispecific binding molecule composed of a desired combination of polypeptide chains. More specifically, the present invention can be expressed as follows:

(1) A method for producing a cell that produces a multispecific binding molecule, comprising the steps of introducing a first vector and a second vector into a cell, the first vector comprising genes for all polypeptide chains that constitute the multispecific binding molecule, and the second vector comprising a gene for a polypeptide chain that is less expressed.

(2) The method of producing (1), wherein the multispecific binding molecule is an antigen-binding molecule or an antibody, the multispecific antigen-binding molecule or antibody may be a bispecific antibody having a common L chain or a common H chain, and the bispecific antibody may be an IgG-type antibody having a common H chain or L chain.

(3) A method for producing cells that produce a bispecific antibody having a common L chain, comprising the steps of introducing a first vector and a second vector into a cell, the first vector comprising genes for all polypeptide chains that constitute the bispecific antibody, and the second vector comprising genes for a polypeptide chain that is poorly expressed in the bispecific antibody.

(4) A method for producing a polypeptide chain according to (3), in which the polypeptide chain with low expression is one of two different types of H chains, and the second vector contains a nucleic acid sequence encoding the H chain and a nucleic acid sequence encoding a common L chain.

(5) The method of (4), characterized in that the second vector contains more nucleic acid sequences encoding the L chain polypeptide than the nucleic acid sequences encoding the H chain polypeptide.

(6) The method of (5), in which the first vector contains a nucleic acid sequence encoding a first H chain, a nucleic acid sequence encoding a second H chain, and a nucleic acid sequence encoding a common L chain in a ratio of 1:1:1 or 1:1:2, and the second vector contains a nucleic acid sequence encoding a lower-expressing H chain and a nucleic acid sequence encoding a common L chain in a ratio of 1:2.

(7) A method for producing a nucleic acid sequence according to (4), in which the first vector contains one copy of a nucleic acid sequence encoding a first H chain, one copy of a nucleic acid sequence encoding a second H chain, and one or two copies of a nucleic acid sequence encoding a common L chain, and the second vector contains one copy of a nucleic acid sequence encoding a less-expressed H chain (either the first or second H chain) and two or more copies of a nucleic acid sequence encoding a common L chain.

(8) A cell that produces an antigen-binding molecule, the cell comprising a first vector and a second vector, the first vector comprising genes for all polypeptide chains that constitute the antigen-binding molecule, the second vector comprising genes for a polypeptide chain with low expression, and the antigen-binding molecule may be multispecific or may be a bispecific antibody in which the H chain or L chain is shared.

(9) A cell that produces a bispecific antibody having a common L chain, the cell comprising a first vector and a second vector, the first vector comprising genes for all polypeptide chains that constitute the bispecific antibody, and the second vector comprising genes for a polypeptide chain that is poorly expressed in the bispecific antibody.

(10) The cell of (9), wherein the second vector contains a gene for the H chain with lower expression and further contains a gene for a common L chain, and the second vector may contain a greater number of copies of the gene for the common L chain than the gene for the H chain.

(11) The cell described in (9), in which the first vector contains one copy of the first H chain gene, one copy of the second H chain gene, and one or two copies of the common L chain gene, and the second vector contains one copy of the H chain gene with lower expression and two or more copies of the common L chain gene.

(12) A method for producing an antigen-binding molecule or a bispecific antibody using any of the cells described above in (1) to (11).

(13) A method for producing the above-mentioned cells, cells, or manufacturing method, in which the cells produce an anti-FIX(a)/FX bispecific antibody, for example, emicizumab.

(14) The method for producing a cell, cell, or manufacturing method described above, wherein the cell is an animal cell, for example a mammalian cell, preferably a CHO cell.

(15) A method for producing a pharmaceutical containing an antigen-binding molecule or a bispecific antibody produced by the above-mentioned method.

In the above-mentioned cell production method, when the method is a method for producing cells that produce bispecific antibodies having a common L chain, the "lowly expressed polypeptide" is a polypeptide that constitutes a monospecific antibody that is expressed in a low amount among two types of monospecific antibodies (i.e., impure antibodies) whose H chains are homozygous.

The method for producing the above-mentioned cells may further include a selection step of selecting clones that highly express the target gene after introducing the first and second vectors into the cells.

The above-mentioned method for producing an antigen-binding molecule or bispecific antibody includes a culture step of culturing clone cells (i.e., a strain producing the antigen-binding molecule or bispecific antibody of interest) obtained by the above-mentioned cell production method. Furthermore, the method for producing an antigen-binding molecule or bispecific antibody of the present invention may be a method including a purification step of purifying the antigen-binding molecule or bispecific antibody of interest from the culture obtained by the clone cell culture step.

The present invention also encompasses transformed cells obtained by the above-mentioned cell production method. The transformed cells contain the above-mentioned first and second vectors. For example, a cell transformed to produce a bispecific antibody having a common L chain contains a first and second vector, where the first vector is an expression vector containing foreign DNA encoding all of the polypeptide chains constituting the bispecific antibody, the second vector is an expression vector containing foreign DNA encoding a polypeptide chain that is less expressed in the bispecific antibody, and the second vector may further contain foreign DNA encoding another polypeptide chain.

The present invention increases the expression level of bispecific antibodies while balancing the expression levels of two monospecific antibodies, which are impure antibodies, making them easier to handle during purification. Using the present invention, cells suitable for producing bispecific antibodies can be created and established as cell lines with high production efficiency for the production of antibody pharmaceuticals.

A schematic diagram of the strand arrangement in each of the three plasmids is shown. The percentage of heterozygotes produced in cell pools into which each of the three types of plasmids was introduced is shown.

1. Multispecific binding molecules Multispecific binding molecules are molecules that can bind to two or more different targets. Multispecific binding molecules contain different polypeptide chains, each of which contains a target binding site that recognizes a different target. Examples of multispecific binding molecules include protein molecules that contain two, three, four, or more types of polypeptide chains.

Examples of multispecific binding molecules include antigen-binding molecules. One antigen-binding molecule is composed of multiple polypeptide chains. The multispecific binding molecule can be a multispecific antigen-binding molecule having at least a first antigen-binding site that recognizes a first antigen and a second antigen-binding site that recognizes a second antigen. A preferred multispecific antigen-binding molecule is a multispecific antibody that can specifically bind to at least two different antigens. The multispecific antibody can include a first antigen-binding site, a second antigen-binding site, and optionally a third antigen-binding site.

In the present invention, "different antigens" do not necessarily mean that the antigens themselves are different, but also include cases where the epitopes are different. Therefore, for example, different epitopes within a single molecule are also included in "different antigens".

Preferred multispecific antibodies in the present invention include bispecific antibodies (sometimes called bispecific antibodies) that can specifically bind to two different antigens.

A bispecific antibody can be, for example, an IgG-type antibody molecule composed of two different H chains (heavy chains) and two different L chains (light chains), i.e., four types of polypeptide chains. Alternatively, a bispecific antibody can be, for example, an IgG-type antibody molecule in which the H chain or L chain is shared.

In one embodiment of the present invention, the L chains of the bispecific antibody may be different, but preferably have a common L chain (referred to as a "common L chain"). A "common L chain" can associate with two or more different H chains and exhibit binding ability to each antigen. Here, "different H chains" preferably refers to, but is not limited to, H chains of antibodies against different antigens, and means H chains whose amino acid sequences are different from each other. The common L chain can be obtained, for example, according to the method described in WO2006/109592 (Patent Document 3).

A bispecific antibody is an immunoglobulin molecule that has specificity for two different antigens. The origin of the polypeptide chains that constitute the antibody is not particularly limited, and they can be derived from humans, mice, rats, etc., and H chains and L chains derived from different animal species may coexist. Each polypeptide chain may be modified to a chimeric state using genetic recombination technology, or may be a synthetic immunoglobulin chain. In one embodiment, the bispecific antibody of the present invention is an antibody modified using genetic recombination technology. In one embodiment, the bispecific antibody of the present invention is a natural antibody. The bispecific antibody of the present invention is preferably a full-length IgG antibody.

"Full length antibody," "complete antibody," and "whole antibody" are used interchangeably herein and refer to an antibody having a structure substantially similar to a native antibody structure. A full length antibody has a heavy chain that includes an Fc region.

Native antibodies refer to immunoglobulin molecules with a variety of naturally occurring structures. As a common example, native IgG antibodies are heterotetrameric glycoproteins of about 150,000 daltons composed of two identical light chains (L chains) and two identical heavy chains (H chains) that are disulfide-bonded. From the N-terminus to the C-terminus, each heavy chain has a variable region (VH), also called the variable heavy domain or heavy chain variable domain, followed by three constant domains (CH1, CH2, and CH3). Similarly, from the N-terminus to the C-terminus, each light chain has a variable region (VL), also called the variable light domain or light chain variable domain, followed by a constant light (CL) domain. The light chains of an antibody may be assigned to one of two types, called kappa (κ) and lambda (λ), based on the amino acid sequence of its constant domains. Natural IgG antibodies, which are composed of two identical light chains (L chains) and two identical heavy chains (H chains), are monospecific molecules and are not "multispecific binding molecules" as defined in the present invention.

2. Construction of a cell line producing a multispecific binding molecule The present invention provides a method for producing a cell producing a multispecific binding molecule using two vectors, the method comprising the steps of introducing a first vector and a second vector into a cell, the first vector comprising genes for all polypeptide chains constituting the multispecific binding molecule, and the second vector comprising genes for a polypeptide chain with low expression.

In the present invention, a "lowly expressed polypeptide chain" is a polypeptide chain that is expressed less than other polypeptide chains when the genes of multiple polypeptide chains that constitute a multispecific binding molecule are introduced into the same cell. For example, in the case of a bispecific antibody having a common L chain, a lowly expressed polypeptide chain can be identified by comparing the amount of homoantibody formed other than the desired bispecific antibody that is formed by association of identical polypeptide chains (i.e., identical H chains and a common L chain). In the case of a bispecific antibody having a common L chain, the second vector is a vector containing genes encoding the lowly expressed H chain and common L chain.

The term "vector" refers to a nucleic acid molecule capable of propagating another nucleic acid to which it is linked. The term generally includes vectors as self-replicating nucleic acid structures and vectors that integrate into the genome of a host cell into which they are introduced. A vector can effect expression of a genetic nucleic acid to which it is operably linked. Such vectors are also referred to as "expression vectors."

In the present invention, a vector containing a gene for a polypeptide chain is an expression vector containing a nucleic acid sequence that codes for the amino acid sequence of the polypeptide. By contacting such a vector with any suitable host cell, a foreign gene that codes for the polypeptide of interest is introduced, and the cell becomes a transformed cell capable of producing the polypeptide.

Any suitable method can be used to contact the cells with the vectors such that the cells express the polypeptides encoded by the genes contained in each vector. Methods for contacting cells such that the cells are modified to express a particular polypeptide are well known in the art. Suitable methods for contacting cells include, for example, infection with a viral vector, calcium chloride transfection using lipofection reagents, cationic polymers, DEAE, or calcium phosphate, and electroporation. Contacting the host cell with the vector introduces the vector into the host cell.

In the process of contacting a cell with two vectors and introducing the vectors into the cell, the order of contact is arbitrary. To contact a cell with two vectors, the cell may be contacted with the first vector and the second vector in a sequential manner (e.g., contacting the cell with the second vector after the first vector, or contacting the cell with the second vector before the first vector), or the cell may be contacted with the first vector and the second vector simultaneously.

In the present invention, the origin of the cells that are contacted with the vector for the purpose of producing a protein that is a multispecific binding molecule, i.e., the cells into which the genes of the polypeptide chains that constitute the multispecific binding molecule are introduced to become a host cell line that produces the multispecific binding molecule, is not particularly limited. The cells can be adherent cells or suspension cells (i.e., cells that grow in suspension). Both eukaryotic microorganisms such as bacteria, filamentous fungi, or yeast, and cells derived from multicellular organisms (invertebrate and vertebrate organisms) can be used as host cells. Examples of invertebrate cells include plant and insect cells.

In the present invention, vertebrate cells are suitable for use as hosts. For example, mammalian cell lines adapted to grow in suspension would be useful. Mammalian cells suitable for constructing cell lines for protein production are known in the art and include the SV40 transformed monkey kidney CV1 line (COS-7); human embryonic kidney lines (293 cells, as described, e.g., in Graham et al., J. Gen Virol. 36:59 (1977)); baby hamster kidney cells (BHK); mouse Sertoli cells (TM4 cells, as described, e.g., in Mather, Biol. Reprod. 23:243-251 (1980)); monkey kidney cells (C V1); African green monkey kidney cells (VERO-76); human cervical carcinoma cells (HELA); canine kidney cells (MDCK); Buffalo rat hepatocytes (BRL 3A); human lung cells (W138); human hepatocytes (Hep G2); mouse mammary carcinoma (MMT 060562); TRI cells (described, for example, in Mather et al., Annals N.Y. Acad. Sci. 383:44-68 (1982)); MRC5 cells; and FS4 cells. Other useful mammalian host cell lines include Chinese hamster ovary (CHO) cells, including DHFR- CHO cells (Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216 (1980)); and myeloma cell lines such as Y0, NS0, and Sp2/0. For a review of specific mammalian host cell lines suitable for antibody production, see, e.g., Yazaki and Wu, Methods in Molecular Biology, Vol. 248 (B.K.C. Lo, ed., Humana Press, Totowa, NJ), pp. 255-268 (2003).

Examples of suitable host cells for producing antibody pharmaceuticals with a full-length IgG backbone include the above-mentioned human cells or non-human mammalian cells. On the other hand, prokaryotic cells (e.g., E. coli) are preferably used as host cells for amplifying or replicating (cloning) recombinant expression vectors.

The terms "host cell," "host cell line," and "host cell culture" are used interchangeably and refer to cells into which exogenous nucleic acid has been introduced, including the progeny of such cells. Host cells include "transformants" and "transformed cells," which include the originally transformed cell and progeny derived from that cell regardless of the number of passages. The progeny may not be completely identical in nucleic acid content to the parent cell and may contain mutations. Host cells also include mutant progeny that have the same function or biological activity as that for which the original transformed cell was screened or selected.

　Generally, recombinant antibodies can be obtained by cloning the DNA encoding them from antibody-producing cells such as hybridomas or sensitized lymphocytes that produce antibodies, incorporating it into an appropriate vector, and introducing this into host cells for production. IgG-type bispecific antibodies can be produced by introducing into cells the genes for the L chains and H chains that make up the two types of IgG of interest, a total of four genes, and co-expressing them. When bispecific antibodies have a common L chain, bispecific IgG can be expressed by expressing IgG by introducing into cells a total of three genes, the genes for this common L chain and both H chains.

Assuming that the expression levels of the common L chain and both H chains are equal, when a vector containing DNA encoding a first H chain, DNA encoding a second H chain, and DNA encoding a common L chain in a ratio of 1:1:2 is used, the three polypeptide chains expressed in this ratio should associate to produce the desired bispecific IgG (i.e., heterozygous H chains) at 50% and monospecific IgG with the same H chains associated at 25% each. However, there are cases where all of the polypeptide chains that make up a bispecific antibody are not expressed at the same level. For example, the expression level of one of the two H chains may be lower than the expression level of the other H chain, resulting in different production levels of the two homozygotes.

In one aspect, the present invention provides a novel method for producing a cell that produces a bispecific antibody having a common L chain. The method includes a step of introducing a first vector and a second vector into a cell. The first vector is a vector that contains genes for all polypeptide chains that constitute the bispecific antibody. Thus, the first vector contains all three genes that respectively code for the first H chain, the second H chain, and the common L chain of the bispecific antibody. For example, the first vector contains the copy numbers of the three genes that respectively code for the first H chain, the second H chain, and the common L chain of the bispecific antibody, for example, in a ratio of 1:1:1 or 1:1:2. For example, the first vector contains one copy of the first H chain gene, one copy of the second H chain gene, and one or two copies of the common L chain gene. The second vector is a vector that contains genes for a polypeptide that is lowly expressed in the bispecific antibody. Among the polypeptides constituting a bispecific antibody having a common L chain, the polypeptide with low expression may be either the first H chain or the second H chain. Thus, the second vector contains a gene encoding either the first H chain or the second H chain.

If necessary, the second vector may further contain a gene encoding a common L chain. The second vector may contain more genes encoding a common L chain than genes encoding an H chain. For example, the second vector may contain the number of copies of the genes encoding either the first H chain or the second H chain and the gene encoding the common L chain in a ratio of 1:1 or 1:2. For example, the second vector may contain one copy of the gene for the H chain with the lower expression level and one copy, two copies, or two or more copies of the gene for the common L chain.

In one embodiment of the present invention, a cell that produces a bispecific antibody having a common L chain is transformed with a first vector and a second vector, the first vector being a vector that encodes all of the polypeptide chains that constitute the bispecific antibody, and the second vector can contain more nucleic acid sequences encoding L chain polypeptides than nucleic acid sequences encoding H chain polypeptides. For example, the second vector can contain one copy of a gene encoding any of the H chains and two or more copies of a gene encoding the common L chain.

In the present invention, the cells that produce bispecific antibodies having a common L chain are preferably cells that are suitable for producing recombinant proteins to be used as pharmaceuticals. Mammalian cells that are suitable for constructing cell lines for recombinant protein production have already been described.

In the present invention, the two vectors (i.e., the first vector and the second vector) used to generate a cell that produces a multispecific binding molecule or a bispecific antibody can be, for example, a plasmid vector or a viral vector. In the present invention, for example, the vector can include a control sequence (depending on the host) such as a transcription and translation initiation and termination codon, an appropriate promoter, a marker gene that allows the selection of a host into which a foreign gene has been introduced, etc., as necessary. Furthermore, for example, the vector can include an antibody signal sequence that promotes the secretion of the antibody into the extracellular environment. In the present invention, for example, the vector can be designed for the purpose of transient expression or stable expression. A person skilled in the art would be able to appropriately construct a recombinant expression vector using known standard techniques depending on the purpose.

In the present invention, the first vector and the second vector can be independently any type of vector. The first vector and the second vector have the same control sequence, appropriate promoter, marker gene, etc., but may differ only in the recombinant polypeptide coding sequence contained therein.

The vectors that can be used in the present invention will be described in more detail. For example, when E. coli is used as the host, in order to amplify and mass-prepare the vector in E. coli (e.g., JM109, DH5α, HB101, XL1Blue), it is preferable that the vector has an "ori" for amplification in E. coli and further has a selection gene for the transformed E. coli (e.g., a drug resistance gene that can be identified by a certain drug (ampicillin, tetracycline, kanamycin, chloramphenicol)). Examples of vectors include M13-based vectors, pUC-based vectors, pBR322, pBluescript, and pCR-Script. In addition to the above vectors, examples of vectors that can be used for the purpose of subcloning and excision of cDNA include pGEM-T, pDIRECT, and pT7.

When vectors are used for the purpose of producing polyspecific binding molecules or bispecific antibodies, expression vectors are particularly useful. For example, when the purpose is expression in E. coli, the expression vector should have the above-mentioned characteristics that allow the vector to be amplified in E. coli, and when the host is E. coli such as JM109, DH5α, HB101, or XL1-Blue, it is preferable that the vector has a promoter that allows efficient expression in E. coli, such as the lacZ promoter (Ward et al., Nature (1989) 341, 544-546; FASEB J. (1992) 6, 2422-2427), araB promoter (Better et al., Science (1988) 240, 1041-1043), or T7 promoter. In addition to the above vectors, such vectors include pGEX-5X-1 (Pharmacia), "QIAexpress system" (Qiagen), pEGFP, or pET (in this case, the host is preferably BL21, which expresses T7 RNA polymerase).

The vector may contain a signal sequence for polypeptide secretion. When producing a polypeptide in the periplasm of E. coli, the signal sequence for polypeptide secretion may be the pelB signal sequence (Lei, S. P. et al J. Bacteriol. (1987) 169, 4379).

In addition to using E. coli as a host, other vectors that can be used in the method of the present invention include mammalian expression vectors (e.g., pcDNA3 (Invitrogen), pEGF-BOS (Nucleic Acids. Res. 1990, 18(17), p5322), pEF, pCDM8 and INPEP4 (Biogen-IDEC)), insect cell-derived expression vectors (e.g., the "Bac-to-BAC baculovairus expression system" (G Examples of such expression vectors include pBacPAK8), plant-derived expression vectors (e.g., pMH1, pMH2), animal virus-derived expression vectors (e.g., pHSV, pMV, pAdexLcw), retrovirus-derived expression vectors (e.g., pZIpneo), yeast-derived expression vectors (e.g., Pichia Expression Kit (Invitrogen), pNV11, SP-Q01), and Bacillus subtilis-derived expression vectors (e.g., pPL608, pKTH50).

When the aim is to express the vector in animal cells such as CHO cells, COS cells, or NIH3T3 cells, it is preferable that the vector has a promoter necessary for expression in the cells, such as the SV40 promoter (Mulligan et al., Nature (1979) 277, 108), MMLV-LTR promoter, EF1α promoter (Mizushima et al., Nucleic Acids Res. (1990) 18, 5322), CMV promoter (Niwa et al., Gene. (1991) 108, 193), mouse β-globin promoter (mBGP), or CAG promoter (Niwa et al., Gene. (1991) 108, 193; Miyazaki et al., Gene (1989) 79,269), and it is even more preferable that the vector has a gene for selecting for transformation into cells (for example, a drug resistance gene that can be distinguished by a drug (neomycin, G418, etc.)). Examples of vectors with such properties include pMAM, pDR2, pBK-RSV, pBK-CMV, pOPRSV, and pOP13. It is known that mRNA with polyA is stable within cells, and it is preferable that the vector has a polyA signal required for adding polyA to a gene, such as the mouse β-globin polyA signal, the bovine growth hormone polyA signal (rBG-pA), or the SV40 polyA signal.

In one embodiment of the present invention, a neomycin resistance gene, a CAG promoter, rBG-pA, etc. can be suitably used for expression in CHO cells. Recombinant proteins produced using CHO cells have been confirmed to be safe for use as pharmaceuticals, and are currently a common technique. A preferred embodiment of the present invention for producing bispecific antibodies as active ingredients of pharmaceuticals is a recombinant protein expression system using CHO cells.

　Methods for delivering genes into cells are well known in the art and may be selected appropriately depending on the cells used as the host, and commercially available gene transfer systems may be used. Methods for introducing vectors into mammalian cells include methods using transfection reagents such as electroporation and lipofection, and methods using viral vectors. Methods for inserting foreign genes into the host genome of mammalian cells include methods using random integration, targeted integration (site-specific gene insertion using recombinase, a sequence-specific recombination enzyme), transposon vectors, and site-specific nucleases. Site-specific gene transfer methods into specific locations in the host genome are expected to be a method for efficiently obtaining cells with excellent production of the target protein and excellent passage stability.

The cells of the present invention may express antigen-binding molecules (antibodies) in a transient expression system (Transient Expression) or in a stable expression system (Stable Expression), with the latter being preferred.

Transient expression systems are a method in which a circular plasmid is introduced into cells and expressed using techniques such as the calcium phosphate method, electroporation, or lipofection. Circular plasmids are not efficiently inserted into chromosomes, and the target gene often exists outside the chromosome. For this reason, it is difficult to maintain the expression of the target gene from a circular plasmid for a long period of time.

The constitutive expression system is a method in which linear plasmids created by restriction enzyme treatment etc. are introduced into cells and expressed using the calcium phosphate method, electroporation method, lipofection method etc. Linear plasmids are more efficiently inserted into chromosomes than circular plasmids, and the efficiency of maintaining the target gene on the chromosome is also higher. This makes it possible to maintain the expression of the target gene for a long period of time. Furthermore, drug selection becomes possible if a drug resistance gene is introduced into the plasmid, and cells in which the target gene is maintained on the chromosome can be efficiently selected. Examples of animal cells used in the constitutive expression system include CHO cells, NS0 cells, and SP2/0 cells, with CHO cells being preferred.

Furthermore, when the aim is to stably express a gene and to amplify the copy number of the gene in the cell, a method can be used in which a vector (e.g., pCHOI, etc.) having a complementary DHFR gene is introduced into CHO cells lacking a nucleic acid synthesis pathway and amplified with methotrexate (MTX). When the aim is to express a gene transiently, a method can be used in which COS cells having a gene expressing SV40 T antigen on their chromosomes are transformed with a vector (e.g., pcD) having an SV40 replication origin. Replication origins can also be derived from polyoma virus, adenovirus, bovine papilloma virus (BPV), etc. Furthermore, in order to amplify the copy number of the gene in the host cell system, the expression vector can contain a selection marker such as the aminoglycoside transferase (APH) gene, thymidine kinase (TK) gene, Escherichia coli xanthine guanine phosphoribosyltransferase (Ecogpt) gene, or dihydrofolate reductase (dhfr) gene.

3. Production of antigen-binding molecules (culture process)
The present invention further provides methods for producing a desired antigen-binding molecule using the above-mentioned cells that produce multispecific binding molecules or cells that produce bispecific antibodies having a common L chain.

The production of antigen-binding molecules can be carried out by culturing the above-mentioned cells under conditions suitable for the production of antibodies. Methods for cell culture are known in the art. The conditions under which cells are cultured vary depending on the type of cells. Such conditions include the temperature of the environment, the culture vessel containing the cells, the composition of various gases (e.g., CO ₂ ) that constitute the cell culture atmosphere or environment, the medium, the cell density, the schedule for replacing the medium with fresh medium, etc. These parameters are known in the art or can be empirically determined. For example, any method can be used to culture cells in a medium so that the cells express (and, in some cases, secrete) the polypeptide encoded by the vector contacted with the cells.

For cell culture, a medium used for normal cell (preferably animal cell) culture can be used. For example, DMEM, MEM, RPMI1640, and IMDM can be used as culture media for animal cells. Commercially available animal cell culture media such as D-MEM (Dulbecco's Modified Eagle Medium), D-MEM/F-12 1:1 Mixture (Dulbecco's Modified Eagle Medium: Nutrient Mixture F-12), RPMI1640, CHO-S-SFM II (Invitrogen), CHO-SF (Sigma-Aldrich), EX-CELL 301 (JRH biosciences), CD-CHO (Invitrogen), IS CHO-V (Irvine Scientific), and PF-ACF-CHO (Sigma-Aldrich) can also be used. Serum supplements such as fetal calf serum (FCS) can also be used in combination with these media, or serum-free culture can be performed.

The medium typically contains amino acids, vitamins, lipid factors, an energy source, an osmotic regulator, an iron source, a pH buffer, and optionally, for example, trace metal elements, surfactants, growth cofactors, nucleosides, etc. The contents of these components are usually within the ranges of 0.05-1500 mg/L for amino acids, 0.001-10 mg/L for vitamins, 0-200 mg/L for lipid factors, 1-20 g/L for energy sources, 0.1-10,000 mg/L for osmotic pressure regulators, 0.1-500 mg/L for iron sources, 1-10,000 mg/L for pH buffers, 0.00001-200 mg/L for trace metal elements, 0-5,000 mg/L for surfactants, 0.05-10,000 μg/L for growth cofactors, and 0.001-50 mg/L for nucleosides, but are not limited to these and can be appropriately determined depending on the type of cells to be cultured, the type of antigen-binding molecule of interest, etc.

More specifically, for example, L-alanine, L-arginine, L-asparagine, L-aspartic acid, L-cysteine, L-cystine, L-glutamine, L-glutamic acid, glycine, L-histidine, L-isoleucine, L-leucine, L-lysine, L-methionine, L-ornithine, L-phenylalanine, L-proline, L-serine, L-threonine, L-tryptophan, L-tyrosine, L-valine, etc., preferably L-alanine, L-arginine, L -Amino acids such as asparagine, L-aspartic acid, L-cystine, L-glutamine, L-glutamic acid, glycine, L-histidine, L-isoleucine, L-leucine, L-lysine, L-methionine, L-phenylalanine, L-proline, L-serine, L-threonine, L-tryptophan, L-tyrosine, and L-valine; i-inositol, biotin, folic acid, lipoic acid, nicotinamide, nicotinic acid, p-aminobenzoic acid, calcium pantothenate, and pyridinium hydrochloride. Examples of media that contain vitamins such as pyridoxal, pyridoxine hydrochloride, riboflavin, thiamine hydrochloride, vitamin B12, ascorbic acid, etc., preferably biotin, folic acid, lipoic acid, nicotinamide, calcium pantothenate, pyridoxal hydrochloride, riboflavin, thiamine hydrochloride, vitamin B12, ascorbic acid, etc.; lipid factors such as choline chloride, choline tartrate, linoleic acid, oleic acid, cholesterol, etc., preferably choline chloride; energy sources such as glucose, galactose, mannose, fructose, etc., preferably glucose, etc.; osmotic regulators such as sodium chloride, potassium chloride, potassium nitrate, etc., preferably sodium chloride; iron sources such as ferric EDTA, ferric citrate, ferrous chloride, ferric chloride, ferrous sulfate, ferric sulfate, ferric nitrate, etc., preferably ferric chloride, ferric EDTA, ferric citrate, etc.; and pH buffers such as sodium bicarbonate, calcium chloride, sodium dihydrogen phosphate, HEPES, MOPS, etc., preferably sodium bicarbonate, etc.

In addition to the above components, the medium may contain trace metal elements such as copper sulfate, manganese sulfate, zinc sulfate, magnesium sulfate, nickel chloride, tin chloride, magnesium chloride, sodium silicate, etc., preferably copper sulfate, zinc sulfate, magnesium sulfate, etc.; surfactants such as Tween 80 and Pluronic (registered trademark) F68; and growth cofactors such as recombinant insulin, recombinant IGF-1, recombinant EGF, recombinant FGF, recombinant PDGF, recombinant TGF-α, ethanolamine hydrochloride, sodium selenite, retinoic acid, putrescine hydrochloride, etc., preferably sodium selenite, ethanolamine hydrochloride, recombinant IGF-1, putrescine hydrochloride, etc.; nucleosides such as deoxyadenosine, deoxycytidine, deoxyguanosine, adenosine, cytidine, guanosine, uridine, etc. In addition, in a preferred example of the above medium, antibiotics such as streptomycin, penicillin G potassium, and gentamicin, and pH indicators such as phenol red may be added.

The pH of the medium varies depending on the cells being cultured, but is preferably about 6 to 8, generally 6.8 to 7.6, and in many cases a pH of 7.0 to 7.4 is appropriate. Culture is usually carried out at about 30 to 40°C for about 15 to 200 hours, with medium replacement, aeration, and stirring added as necessary.

When the cells are CHO cells, the CHO cells can be cultured using methods known to those skilled in the art. For example, the cells can usually be cultured in an atmosphere with a _CO2 concentration in the gas phase of 0-40%, preferably 2-10%, at 30-39°C, preferably about 37°C. The culture period of cells suitable for producing the desired antigen-binding molecule (antibody or fragment thereof) is usually 1 day to 3 months, preferably 1 day to 2 months, and more preferably 1 day to 1 month.

Various types of culture equipment for culturing animal cells include, for example, fermenter-type tank culture equipment, air lift type culture equipment, culture flask type culture equipment, spinner flask type culture equipment, microcarrier type culture equipment, fluidized bed type culture equipment, hollow fiber type culture equipment, roller bottle type culture equipment, and packed tank type culture equipment.

Cultivation may be performed using any method, such as batch culture, fed-batch culture, or continuous culture, but fed-batch culture or continuous culture is preferred, with fed-batch culture being more preferred.

When producing bispecific antibodies as an active ingredient in a pharmaceutical product, it is desirable that the desired bispecific antibodies are secreted into the medium by culturing the cells. It is also desirable that the association of the L chain and the H chain proceeds naturally in the culture of the antibody-producing cells.

The heavy and light chain polypeptides that make up an antibody molecule assemble with the support of BiP (Immunoglobulin heavy chain binding protein), and then fold to complete the complete antibody structure. This assembly process is dependent on the light chain polypeptide (Molecular Biology of the Cell, 1999, 10, 2209). Therefore, it is thought that by increasing the ratio of the number of light chain genes and the proportion of light chain polypeptides, the assembly of heavy and light chain polypeptides is promoted, and the amount produced is increased.

4. Production of antigen-binding molecules (purification process)
In the culture supernatant of animal cells, impurities such as components secreted from the producing cells other than the target antigen-binding molecule (antibody in the narrow sense) are present. Therefore, purification is carried out to remove the impurities and extract the target protein. The highly pure protein obtained through these culture and purification processes is called a "pharmaceutical substance," which then goes through the formulation process (addition of additives, sterile filtration, etc.) and the filling and packaging process to become a finished pharmaceutical product.

Purification can be performed at any time after culturing the cells. Methods for purifying proteins from culture media or culture supernatants are known in the art. Suitable purification methods include, for example, chromatography, electrophoresis, etc.

Antigen-binding molecules such as antibodies can be separated and purified using methods used for normal polypeptides. For example, antibodies can be separated and purified by appropriately selecting and combining chromatography columns such as affinity chromatography, filters, ultrafiltration, salting out, dialysis, SDS-polyacrylamide gel electrophoresis, isoelectric focusing, etc. (Antibodies: A Laboratory Manual. Ed Harlow and David Lane, Cold Spring Harbor Laboratory, 1988), but these are not limited to these. The concentration of the antibody obtained above can be measured by measuring absorbance or enzyme-linked immunosorbent assay (ELISA), etc.

Columns used for affinity chromatography include Protein A columns and Protein G columns. For example, columns using Protein A columns include Hyper D, POROS, and Sepharose F. F. (Pharmacia).

Examples of chromatography other than affinity chromatography include ion exchange chromatography, hydrophobic chromatography, gel filtration, reverse phase chromatography, and adsorption chromatography (Strategies for Protein Purification and Characterization: A Laboratory Course Manual. Ed Daniel R. Marshak et al., Cold Spring Harbor Laboratory Press, 1996). These types of chromatography can be performed using liquid phase chromatography such as HPLC and FPLC. Purification using these types of chromatography can produce antibody drug substances.

In addition, before or after purification, a polypeptide can be modified as desired or peptides can be partially removed by treating it with an appropriate polypeptide-modifying enzyme. Examples of polypeptide-modifying enzymes that can be used include trypsin, chymotrypsin, lysyl endopeptidase, protein kinase, and glucosidase.

By using the cells of the present invention, the ratio of the amounts of antibody-type molecules produced becomes constant, making it easy to separate and remove impure antibodies, and it is possible to obtain highly pure antigen-binding molecules (antibodies) with the desired multispecificity. As a typical example, the antigen-binding molecule of interest is an IgG-type antibody molecule that has a common L chain and is a molecule that contains three types of polypeptide chains (i.e., a common L chain and two types of H chains that are different). A combination in which the two types of H chains are the same is an impure antibody.

5. Specific bispecific antibodies Emicizumab (or Ace910; or product name: Hemlibra®) is a recombinant humanized bispecific monoclonal antibody against activated blood coagulation factor IX (F IX(a)) and blood coagulation factor X (F X). It is used as a medicine to replace the function of activated blood coagulation factor VIII (FVIIIa), which is deficient or functionally abnormal in hemophilia A, by binding to F IX(a) with one arm of the antibody and to FX with the other arm, and by precisely bridging F IX(a) and F X (Sampei, et al. PLoS ONE 2013; 8(2): e57479; Kitazawa, et al. Nature Medicine 2012; 18(10): 1570).

Emicizumab is an anti-FIX(a)/FX bispecific antibody with a common light chain. Emicizumab contains a heavy chain polypeptide containing an antigen-binding site that recognizes F IX(a), a heavy chain polypeptide containing an antigen-binding site that recognizes FX, and a common light chain polypeptide. The sequences of these polypeptides are publicly known (SEQ ID NO: 20 in WO2012/067176 is the heavy chain on the FIX side, SEQ ID NO: 25 is the heavy chain on the FX side, and SEQ ID NO: 32 is the common light chain).

The inventors have previously succeeded in increasing the production rate of heteroantibodies (anti-FIX(a)/FX bispecific antibodies) by controlling the charge of each antibody chain and adopting a common L chain. However, with a plasmid encoding an H chain that recognizes F IX, an H chain that recognizes F X, and a common L chain, there are more F IX homoantibodies (anti-F IX antibodies) and fewer FX homoantibodies (anti-F X antibodies).

Recently, in order to obtain a cell line with improved heterozygous production in the creation of a WCB (working cell bank) during manufacturing, various investigations were carried out, and as a result, a cell line construction method that achieves both high production and heterozygous selectivity was discovered by constructing the line using a plasmid with FIX H chain: FX H chain: common L chain in a 1:1:2 ratio plus a secondary plasmid with the H chain of the side (anti-FX antibody) with lower homozygous production: common L chain in a 1:2 ratio.

Thus, in a preferred embodiment of the present invention, a method for producing a cell producing emicizumab is provided, comprising the steps of introducing a first vector and a second vector into a cell. In this method, the first vector comprises all of the polypeptide chains constituting emicizumab, i.e., the genes of the FIX heavy chain, the FX heavy chain and the common light chain, in a 1:1:1 or 1:1:2 copy number, respectively, and the second vector may comprise the FX heavy chain gene and, optionally, the common light chain gene. For example, the second vector may comprise the FX heavy chain gene and the common light chain gene in a 1:1 or 1:2 copy number. In a particular embodiment, the first vector comprises one copy of the FIX heavy chain gene, one copy of the FX heavy chain gene and two copies of the common light chain gene, and the second vector comprises one copy of the FX heavy chain gene and two or more copies of the common light chain gene.

6. Production of Pharmaceuticals When an antigen-binding molecule or bispecific antibody produced by the method of the present invention has biological activity that allows it to be used as a pharmaceutical, a pharmaceutical can be produced by mixing the antigen-binding molecule or bispecific antibody with a pharma- ceutical acceptable carrier or additive and formulating the mixture.

Examples of pharma- ceutically acceptable carriers and additives include water, pharma-ceutically acceptable organic solvents, collagen, polyvinyl alcohol, polyvinylpyrrolidone, carboxyvinyl polymers, sodium carboxymethylcellulose, sodium polyacrylate, sodium alginate, water-soluble dextran, sodium carboxymethylstarch, pectin, methylcellulose, ethylcellulose, xanthan gum, gum arabic, casein, agar, polyethylene glycol, diglycerin, glycerin, propylene glycol, petrolatum, paraffin, stearyl alcohol, stearic acid, human serum albumin (HSA), mannitol, sorbitol, lactose, surfactants acceptable as pharmaceutical additives, and the like.

Actual additives are selected from the above, either alone or in appropriate combination, depending on the pharmaceutical dosage form, but are of course not limited to these. For example, when used as an injectable formulation, the purified polypeptide can be dissolved in a solvent, such as physiological saline, buffer solution, glucose solution, etc., to which an adsorption inhibitor, such as Tween 80, Tween 20, gelatin, human serum albumin, etc., can be added. Alternatively, the polypeptide may be lyophilized to form a dosage form that can be dissolved and reconstituted before use, and sugar alcohols and sugars, such as mannitol and glucose, can be used as excipients for lyophilization.

The present invention will be specifically explained below using examples. Note that these examples are for the purpose of explaining the present invention and do not limit the scope of the present invention.

Example 1: Gene transfer using one type of plasmid for each of emicizumab FIX H chain, FX H chain, and common L chain structural genes were linked to a CAG promoter upstream, and rBG-pA was linked downstream to create a FIX H chain expression unit, an FX H chain expression unit, and a common L chain expression unit. The FIX H chain expression unit, the FX H chain expression unit, and the common L chain expression unit were linked to pBluescriptII incorporating a neomycin resistance gene to create the IX1X1L1 plasmid consisting of one copy of FIX H chain, one copy of FX H chain, and one copy of common L chain, the L1IX1X1 plasmid in which the positions of the L chains are swapped, and the IX1X1L2 plasmid consisting of one copy of FIX H chain, one copy of FX H chain, and two copies of common L chain (Figure 1), which were then introduced into CHO cell DXB11-derived host cells by electroporation. Electroporation was performed using Nucleofector (registered trademark). The cells were then cultured in the presence of 15 nmol/L MTX to select cells into which the expression plasmid had been introduced.

IX1X1L1 plasmid-transfected cell pools, L1IX1X1 plasmid-transfected cell pools, and IX1X1L2 plasmid-transfected cell pools were obtained and compared by fed-batch culture in 24-well plates. Culture was performed under the following conditions: culture volume 0.8 mL, culture temperature 37°C, shaking speed 160 rpm. 14 days after the start of culture, antibody concentration in the culture medium was measured, and the percentage of the desired heterodimers produced was evaluated by IEC. As a result, the antibody production amount and the percentage of the desired heterodimers produced in the IX1X1L2 plasmid-transfected cell pool were higher than those in the IX1X1L1 plasmid- or L1IX1X1 plasmid-transfected cell pools (Figure 2). This result suggests that the expression amount of the common L chain gene is insufficient with one copy, and that by increasing the copy number, the expression balance of each antibody chain is adjusted by the gene dosage effect, which may have had an advantageous effect on the antibody production amount and the percentage of the desired heterodimers produced.

Example 2: Gene transfer using secondary plasmid (X1 L2) In Figure 2 of Example 1, almost no homozygous peaks of the FX H chain were observed, suggesting that the expression level of the FX H chain may be insufficient. In addition, there is also the possibility that the expression level of the common L chain may still need to be increased. Therefore, we next created an X1L2 plasmid consisting of one copy of the FX H chain and two copies of the common L chain, and transferred this to cells transfected with the IX1X1L2 plasmid, and compared this with cells transfected with only the IX1X1L2 plasmid.

First, the IX1X1L2 plasmid was introduced into CHO cell DXB11-derived host cells by electroporation and cultured in the presence of 15 nmol/L MTX to obtain a cell pool exhibiting resistance to 15 nmol/L MTX. The newly created X1L2 plasmid was then introduced into this cell pool by electroporation, and the cells were cultured once in the presence of 15 nmol/L MTX and 200 μg/mL hygromycin, and then cultured in the presence of 100 nmol/L MTX to obtain a cell pool exhibiting resistance to 100 nmol/L MTX.

Of the cell pools thus obtained, two cell pools (A01 and A07) in which only the IX1X1L2 plasmid was introduced were compared by fed-batch culture with the cell pools in which the X1L2 plasmid was additionally introduced. Fed-batch culture was performed in 96-well plates with a culture volume of 80 μL, at 37°C and at a shaking speed of 250 rpm. The percentage of desired heterozygote production in the culture medium on day 14 after the start of culture was evaluated by IEC, and it was found that the percentage of desired heterozygote production in the cell pool obtained by introducing the X1L2 plasmid in addition to the IX1X1L2 plasmid was higher than that of the cell pool in which only the IX1X1L2 plasmid was introduced. Specifically, when the X1L2 plasmid was additionally introduced into cell pool A01, 15 of the 17 cell pools obtained by additional introduction had a higher percentage of desired heterozygote production than A01. Similarly, when the X1L2 plasmid was additionally introduced into cell pool A07, the percentage of desired heterozygote production in all 10 cell pools obtained by the additional introduction was higher than that of A07. This result suggests that the expression balance of each antibody gene can be improved by additionally introducing the FX heavy chain and common light chain genes.

All publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety.

Claims (15)

A method for producing a cell that produces a multispecific binding molecule, the method comprising the steps of introducing a first vector and a second vector into the cell, the first vector comprising genes for all polypeptide chains that constitute the multispecific binding molecule, the second vector comprising genes for polypeptide chains that are less expressed, and the multispecific binding molecule may be a multispecific antibody. The method according to claim 1, wherein the multispecific binding molecule is a bispecific antibody, and the bispecific antibody is an IgG-type antibody in which the H chain or the L chain is shared. A method for producing cells that produce bispecific antibodies that have a common L chain, the method comprising the steps of introducing a first vector and a second vector into cells, the first vector comprising genes for all polypeptide chains that constitute the bispecific antibody, and the second vector comprising genes for polypeptide chains that are poorly expressed in the bispecific antibody. The method of claim 3, wherein the polypeptide chain with low expression is one of two different H chains, the second vector contains a nucleic acid sequence encoding the H chain, and the second vector further contains a nucleic acid sequence encoding a common L chain. The method of claim 4, wherein the second vector contains more nucleic acid sequences encoding a common L chain than nucleic acid sequences encoding a H chain. The method of claim 5, wherein the first vector contains one copy of a nucleic acid sequence encoding a first H chain, one copy of a nucleic acid sequence encoding a second H chain, and one or two copies of a nucleic acid sequence encoding a common L chain, and the second vector contains one copy of a nucleic acid sequence encoding a less-expressed H chain, and two or more copies of a nucleic acid sequence encoding a common L chain. A cell that produces an antigen-binding molecule, the cell comprising a first vector and a second vector, the first vector comprising genes for all polypeptide chains that constitute the antigen-binding molecule, the second vector comprising genes for a polypeptide chain with low expression, the antigen-binding molecule may be multispecific, and the multispecific antigen-binding molecule may be a bispecific antibody in which the H chain or L chain is shared. A cell that produces a bispecific antibody having a common L chain, the cell comprising a first vector and a second vector, the first vector comprising genes for all polypeptide chains that constitute the bispecific antibody, and the second vector comprising genes for the less expressed H chain and the common L chain. The cell of claim 8, wherein the second vector contains a greater number of copies of the common light chain gene than the heavy chain gene. The cell according to claim 9, wherein the first vector contains one copy of the first H chain gene, one copy of the second H chain gene, and one or two copies of the common L chain gene, and the second vector contains one copy of the H chain gene with lower expression and two or more copies of the common L chain gene. A method for producing an antigen-binding molecule, comprising producing an antigen-binding molecule using the cells of claim 7, or a method for producing a bispecific antibody, comprising producing a bispecific antibody using the cells of claim 8. The method for producing cells according to any one of claims 1 to 6 or the method for producing cells according to claim 11, wherein the cells are mammalian cells. The method for producing the cells according to any one of claims 3 to 6 or the method for producing the cells according to claim 11, wherein the bispecific antibody is an anti-FIX(a)/FX bispecific antibody. The method of claim 13, wherein the bispecific antibody is emicizumab. A method for producing a pharmaceutical containing an antigen-binding molecule or a bispecific antibody produced by the method according to claim 11.