Method and means for producing Ig-like molecules

Abstract

To provide means and methods for producing one or more Ig-like molecules in a single host cell.SOLUTION: Disclosed is a method for producing from a single host cell at least two different Ig-like molecules each comprising two CH3 domains, which method comprises providing a first to a fourth nucleic acid molecules respectively encoding polypeptide chains comprising having a first to a fourth CH3 domains in the cell, where at least two of the nucleic acid molecules are provided with means for preferential pairing between the polypeptides comprising the first and second CH3 domains and between the polypeptides comprising the third and fourth CH3-domains, where the method further comprises the steps of culturing the host cell and allowing for expression of the at least four nucleic acid molecules and harvesting the at least two different Ig-like molecules from the culture.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to the fields of molecular biology, medicine, and biological therapeutics. The present invention particularly relates to the field of therapeutic antibodies used for the treatment of various diverse diseases.

Background Art

[0002] Currently, many of the commonly used biological agents are isolated human or humanized recombinant monoclonal antibodies. These agents enhance the ability of the body's immune system to neutralize or remove cells and / or molecules involved in the disease process, or to eradicate invading pathogens or infected agents.

[0003] Monoclonal antibodies bind to a specific site of an antigen, i.e., an epitope. And in the case of therapeutic antibodies, the epitope is selected so as to exhibit desirable functions such as, for example, removing tumor cells, inhibiting receptor-ligand interactions, or neutralizing viruses.

[0004] Currently, approximately 30 monoclonal antibodies have been approved by the FDA. These antibodies are usually produced in large quantities, and their biophysical and biochemical properties are examined in detail, so that the quality of each batch is maintained and meets the standards.

[0005] Despite these advantages, monoclonal antibodies have several disadvantages. Some of them are related to the single specificity, which is an inherent feature of monoclonal antibodies, and the complexity of diseases.

[0006] In the course of a disease, it is not uncommon for many elements to be involved. And each The effects of these factors can be redundant or synergistic, leading to increased expression of other receptors and signaling. Crosstalk can occur between transmission networks. Therefore, multiple factors related to the disease may be involved. Inhibiting different elements and reaction pathways can lead to improved treatment efficiency. Due to the inherent single-specificity of monoclonal antibodies, monoclonal antibodies are used in complex disease processes. It may only be possible to interfere with one step at a time, and may not be able to achieve sufficient results. be.

[0007] Apart from the issue of the multifaceted nature of disease processes, a single cell, a water-soluble protein, or a pathogen Targeting only a single epitope may not be sufficient to treat the disease effectively. It has become clear that the target epitope is no longer monochromatic. This is because the Nal antibody may bind and become unable to exert the desired effect. For example, tumor cells reduce the expression of target epitopes on growth factor receptors, or It is often possible to evade treatment with monoclonal antibodies by introducing natural mutations or blocking them. Baal.

[0008] By activating other receptors and / or their ligands, tumor cells can trigger a different response. They activate pathways to continue growing and transmigrate. Similarly, viruses and other pathogens also target epitopes. By introducing mutations, deletions, or shielding, treatment with monoclonal antibodies is possible. Dodge.

[0009] Monoclonal antibodies that bind to a single epitope cause polyclonal antibodies It does not generate all of the effects pedal mechanisms. The prominent effects pedal mechanisms here Examples include the following: opsonization (making the antigen more susceptible to phagocytosis), steric hindrance (inhibition of the antigen enclosed in the antibody from reaching the host cell or mucosal surface), detoxification of toxins aggregation, or precipitation (aggregation of the antibody bound to some water-soluble antigens and subsequent removal) activation of complement and cytotoxicity by antibodies (the antibody causes the target cell to be killed by NK cells and neutrophils) .

[0010] Polyclonal antibodies applicable to treatment can be obtained from pooled human sera. Such therapeutically polyclonal antibodies derived from sera are used for the treatment and prevention of viral infections such as rabies virus, cytomegalovirus, and RS virus, the detoxification of toxins such as tetanus toxin and botulinum toxin, or the prevention of anti-D immunization.

[0011] The more extensive use of serum-derived polyclonal antibody preparations is hindered by the fact that the plasma used as the raw material can only be applied to a limited range of targets such as infectious diseases and toxins. Furthermore, the product depends on the blood supply situation of the donor in terms of both quantity and compatibility, resulting in significant variations from batch to batch. In addition, screening technology cannot keep up with the continuous evolution of the virus, and immunoglobulin preparations are accompanied by the risk of mediating the transmission of infectious diseases. Finally, due to the long processes such as blood collection, screening, and immunoglobulin purification, plasma-derived immunoglobulin is costly to produce." A mixture of monoclonal antibodies can enhance the effectiveness of monoclonal antibodies while

[0012] There are no restrictions related to polyclonal antibodies. In this field, combinations of two human or humanized monoclonal antibodies have been tested in preclinical models and clinical trials ( for example, mixtures of two monoclonal antibodies against the HER2 receptor, mixtures of two antibodies against the receptor EGFR and mixtures of two monoclonal antibodies against rabies virus). .

[0013] In this field, combinations of two monoclonal antibodies can exhibit additional or synergistic effects and have been shown to induce effector mechanisms not associated with either antibody alone. For example, mixtures of two monoclonal antibodies against EGFR or HER2 have been shown to kill tumor cells more potently. This is based on a combination of activities including promotion of cytotoxicity by immune system effectors, promotion of receptor internalization, and enhanced inhibition of signaling pathways downstream of the receptor.

[0014] In combination therapies based on two monoclonal antibodies, the component antibodies are produced separately and mixed at the protein level. The drawback of this method is the enormous cost of conducting clinical trials for each of the two antibodies and (partially) repeating the process for the combination. This can make combination therapies based on antibodies cost-prohibitive.

[0015] As an alternative, two recombinant cell lines producing the component monoclonal antibodies can be mixed in a fermenter and the resulting mixture of antibodies can be purified as one sample (WO 2 (Issue 004 / 061104). The drawback of this method is that it is difficult to control the composition. As a result, the reproducibility of the recombinant polyclonal antibody preparations obtained is poor. This is particularly difficult when considering the temporal changes in such composition as cells are cultured. .

[0016] Over the past decade, bispecific antibodies have been developed as an alternative to using combinations of two antibodies. In the case of a combination of two antibodies, the mixture contains two different immunoglobulin molecules. However, it binds to different epitopes located on the same or different targets. This is achieved with bispecific antibodies. This is achieved by a single immunoglobulin molecule.

[0017] By binding to two epitopes of the same or different targets, bispecific antibodies are produced. It can produce the same effect as the combination of two antibodies that bind to the same epitope. Furthermore, bispecific antibodies in IgG form have two different monovalent binding sites on a single molecule. The mixture contains a combination of two types of IgG antibodies, and the two different divalent antibodies are bound to one sample. Because these molecules are combined in various ways, different effects have also been observed depending on their configuration.

[0018] From this point of view, from a technical and regulatory standpoint, developing a single bispecific antibody is... It is less troublesome to do so. The reason is that production, preclinical and clinical trials are performed on a single molecule. Therefore, treatments based on a single bispecific antibody are more A less labor-intensive and cost-effective drug discovery process facilitates more efficient antibody therapy. To provide the law.

[0019] IgG-type bispecific antibodies, consisting of two heavy chains and two light chains, are produced in various ways. For example, bispecific antibodies are produced by fusing or recombinant two cell lines that secrete antibodies. This can be produced by using DNA technology to express two antibodies in a single cell. According to their method, multiple types of antibodies are produced. The reason for this is that the heavy chains corresponding to each antibody... However, a monospecific dimer (also known as a homodimer) contains two identical pairs of heavy chains with the same specificity. (called), and a bispecific dimer containing two different heavy chain pairs with different specificities. This is because they can form a body (also called a heterodimer).

[0020] Furthermore, the light and heavy chains from each antibody are paired randomly, resulting in improperly functioning combinations. This can be a mispair. This problem, known as heavy-chain and light-chain mispairing, is a dual characteristic. This can be solved by selecting antibodies that share a common light chain expressed in the opposite sex. Even when using ordinary light chains, when two heavy chains and one common light chain are expressed in a single cell, This will result in three different types of antibodies. In other words, two monospecific "parent" antibodies. It is a parental antibody and a bispecific antibody. Therefore, the target two Highly specific antibodies need to be generated from the resulting antibody mixture.

[0021] Further increasing the proportion of bispecific antibodies in a mixture of parental antibodies and bispecific antibodies, Several techniques have been employed to reduce the heavy and light chains of mispairing. However, there are forms of bispecific antibodies that eliminate or minimize some of the above-mentioned problems. It is needed. [Overview of the project] [Problems that the invention aims to solve]

[0022] In summary, conventional technology has the potential to be applied to the treatment of patients, and monoclonal antimicrobial Body, bispecific antibody, monoclonal antibody mixture, or single-specific and bispecific antibodies Various techniques and methods are available for producing mixtures of antibodies.

[0023] However, as mentioned above, these existing technologies and methods each have their own drawbacks and It has limitations. Therefore, it is not possible to target multiple molecules that modify the disease, such as mixtures or duals. To produce biological therapeutic agents using a specificity approach, improved, and / or This requires the creation of technologies different from those used before. [Means for solving the problem]

[0024] This invention relates to a mixture or bispecific applo that targets multiple molecules that modify disease. Improved and / or unprecedented methods for producing biological therapeutic agents by the -ch method The present invention provides methods and means of a different technology. This provides the resulting products and their uses.

[0025] In conventional technology, various methods have been used to promote the formation of a specific bispecific antibody of interest. Approaches are described. These methods aim to achieve the desired result in the resulting mixture. This can reduce the composition of undesirable antibodies.

[0026] Regarding antibodies, CH3-CH3 interactions are known to play a major role in Fc dimerization. (Ellerson JR., et al., J. Immunol 1976 (116) 510-517 and Deisenhofer J) Biochemistry 1981 (20) 2361-2370). Furthermore, the two CH3 domains interact with each other. When this happens, the "contact" residue (contact amino acids) interface residues or interface amino acids and It is known that they face each other at a protein-protein contact surface (also called a protein-protein contact surface).

[0027] The contact amino acids of the first CH3 domain are one or more contact amino acids of the second CH3 domain. It interacts with amino acids. Contact amino acids are typically 5.5 Å apart in the three-dimensional structure of antibodies. It is located within a distance of (preferably 4.5 Å). This is different from the contact residue of this one CH3 domain. The interaction between contact residues of the CH3 domain is, for example, van der Waals force, water Elementary bonds, water-mediated hydrogen bonds, salt bridges, or other static bonds Through electric forces, attractive interactions between aromatic side chains, disulfide bonds, or other known forces It is done by doing so.

[0028] Approximately one-third of the contact amino acid side chains at the contact surface of the CH3 domain of human IgG1 are domain It has been shown so far that they have contributed to the majority of the folding and meetings. The (adjacent) amino acid residues can also influence the interaction at this protein-protein contact surface. This is expected.

[0029] Conventional techniques have employed approaches that interfere with the dimerization of the antibody's heavy chain. To promote heterodimerization rather than dimerization, specific manipulation of the CH3 domain (engineer (ing) is applied. Examples of such CH3-CH3 contact surface operations are, for example, international public Described in Publication No. 1998 / 050431, Ridgeway et al., 1996, Merchant et al. 1998 It is also known as the Knob-into-Hole approach. This method introduces mutations that result in complementary protrusions and cavities.

[0030] Generally speaking, this method introduces protuberances to the contact surface of the first polypeptide. As it enters, a cavity corresponding to the second polypeptide is introduced, and the protrusion is in the cavity It is positioned in the cavity. This promotes the formation of heteromultimers, and homomultimers Body formation is hindered.

[0031] The "protrusions," or "knobs," are short amino acid side chains located on the contact surface of the first polypeptide. By replacing it with a long amino acid side chain (e.g., tyrosine or tryptophan) A complementary "cavity" or "hole" of the same or similar size as the projection is created. The longer amino acid side chains at the contact surface of the polypeptide are replaced by shorter amino acid side chains (for example, A It is created by replacing it with ranine or threonine. The protrusions and cavities are Synthetic methods such as modification of nucleic acids encoding lipeptides, or creation by peptide synthesis. can.

[0032] Using only the knob-into-hole technique, two parental antibodies and bispecific antibodies In the mixture, the proportion of the bispecific antibody of interest can be increased to at most 87%. Merchant et al. added a disulfide between two CH3 domains in the CH3-CH3 contact surface. The introduction of rufid bonds increased the proportion of bispecific antibodies in the mixture to 95%. It was successful. However, in order to use such bispecific antibodies as drugs, bispecificity Sex antibodies are isolated and purified from homodimers and prepared as pharmaceutically acceptable diluents or excipients. It must be in the form of homo- The similarity in physicochemical properties between mers and heterodimers poses a significant challenge.

[0033] One of the objectives of the present invention is to further improve the proportion of bispecific antibodies in a mixture, and to provide a single, fine-grained solution. The objective is to provide a method for producing bispecific antibodies from cellular clones. According to the present invention, The B-into-Hole technique is used as one means, alone or in combination with other means, and mixed. This makes it possible to achieve the proportion of the further improved bispecific antibodies in the material.

[0034] Another example of such operations at CH3-CH3 contact surfaces is the technique of heterodimerized Fc. This technology is provided by the strand-exchange engineered domain. The invention of the CH3 heterodimer leads to the development of bispecific and asymmetric fusion proteins. They are involved in designing these SEED·CH3 heterodimers. Human IgG and IgA are composed of regions that alternately contain CH3 sequences. It is a derivative of [the original species]. As a result, a pair of complementary human SEEDs called SEED-bodies • CH3 heterodimers are obtained (Davis JH. et al., Protein Engineering, Design & Selection 2010(23)195-202; International Publication No. 2007 / 110205).

[0035] Further approaches to producing the desired bispecific antibody include, for example, European Patent Application Publication No. 01870459 or U.S. Patent Application Publication No. 2010 / 0015 Issue 133, International Publication No. 2007 / 147901, International Publication No. 2010 / 129304, As described in Gunasekaran et al (2010) and International Publication No. 2009 / 089004. Thus, based on electrostatic manipulation of contact residues that originally have an electric charge at the CH3-CH3 contact surface. It is continuing.

[0036] In the heavy chain CH3 domain mutations described in these publications, the original charged A The contact residue of the amino acid is replaced with an amino acid residue with the opposite charge (charge reversal strategy: charge). (reversal strategy). This changes the polarity of the charge on the opposing contact surfaces of the Fc dimer. Furthermore, co-expressing electrostatically compatible Fc chains can produce favorable attractive interactions. This promotes the formation of desirable Fc heterodimers, while the repulsive charge interaction leads to the desired The formation of undesirable Fc homodimers is inhibited.

[0037] The CH3-CH3 contact surface has four characteristic pairs of charged residues, and the interaction between these domains There are reports that this is related to D356 / K439', E357 / K370', K392 / D399' and D399 / K409' (The residues of the first and second chains are separated by " / ", and a prime symbol (') is used. This follows the numbering system reported by Kabat (1991) to represent the residue numbers of the second chain. CH3- Because the CH3 contact surface is twice as symmetric, each characteristic pair of charged residues is in the original IgG. Appears at a certain degree (i.e., static K439 / D356', K370 / E357', D399 / K392' and K409 / D399') (Electromagnetic interactions also exist at the contact surface.)

[0038] Taking advantage of this twofold symmetry, we can perform a single charge inversion, for example, K409 in the first chain. In the D mutation, or the D399'K mutation of the second chain, homodimerization occurs due to the repulsion of the same charge. It was shown that body formation is reduced. By applying the inversion operation to different charges, Furthermore, this repulsive effect was enhanced. There is a reversal of different complementary charges, different CH3 Domain expression induces heterodimerization, resulting in bispecificity in the mixture. It was shown that the proportion of the species increases.

[0039] The above approach increases the proportion of bispecific antibodies produced from a single cell to approximately 76%. It was possible to achieve a value between approximately 96%. One of the objectives of the present invention is to obtain from a single cell In a method for producing bispecific antibodies, the proportion of the desired bispecific antibody is further improved. The objective is to provide. According to the present invention, electrostatic manipulation technology is used as one of the means, and alone Alternatively, it can be used in conjunction with other methods, such as the knob-into-hole approach, to further improve the aforementioned The desired proportion of (bispecific) antibodies can be achieved.

[0040] In one aspect, the present invention provides at least two different I cells from a single host cell. In the method for producing γ-like molecules, each of the two γ-like molecules forms a contact surface. The method comprises two CH3 domains capable of, and the cell, a. A first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, b. A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain, c. A third nucleic acid molecule encoding a polypeptide chain having a third CH3 domain, and Beauty, d. A fourth nucleic acid molecule encoding a polypeptide chain containing a fourth CH3 domain. The process includes providing, in at least two of the nucleic acid molecules, the first and second CH3 A polypeptide comprising the main and the polypeptide comprising the third and fourth CH3 domains The method comprises means for selective pairing of the host cells, and the method involves culturing the host cells and the small At least four nucleic acid molecules are expressed, and at least two different Ig-like molecules are cultured from the culture medium. Further steps will be taken to recover the materials.

[0041] For example, multiple factors involved in the disease process or the entry, replication, and / or spread of pathogens. To more efficiently inhibit multiple biological pathways, the body produces multiple (bispecific) antibodies. This is often desired.

[0042] Using a mixture of multiple bispecific antibodies is particularly useful in the treatment of specific diseases. For example, during treatment with antibody or small molecule drugs, tumor cells develop resistance and therefore many They employ different strategies. Resistance involves multiple cell surface receptors and water-soluble molecules. This involves addressing multiple molecules related to such diseases and escape simultaneously. Therefore, developing cancer treatments using antibodies is considered beneficial.

[0043] A site involving two or more target molecules or epitopes related to such diseases and escapes. A mixture of bispecific antibodies represents a novel and intriguing therapeutic approach. It is preferable for antibody mixtures to be produced from a single cell, as this simplifies the drug discovery process. As a result, the drug discovery process is less cumbersome from a regulatory standpoint, and is less burdensome for pharmaceutical and clinical development. From a development perspective, it is cost-effective and easy to implement.

[0044] A single-cell-based approach enables the controlled and efficient production of bispecific antibodies. It is desirable to use a method that allows for the mixing of desired bispecificity IgG molecules. To reduce the need to separate the substance from undesirable single-specific IgG molecules, or to completely This can be eliminated. In conventional technology, single-specific and bispecific cells can be extracted from a single cell. Sexual antibodies are being produced (International Publication No. 2004 / 009618). However, these mixtures The compound is a complex mixture of several different types of bispecific and monospecific antibodies. ru.

[0045] A further objective of the present invention is to obtain a mixture of defined bispecific antibodies from a single cell. The objective is to provide a method and means for producing, preferably, a unit amount, as will be described later. Regardless of the amount of bodily by-products, at least 95%, at least 97%, or higher than 99% This method proposes a method that results in a mixture of dimeric IgG molecules (bispecific) antibodies. Generally speaking, in cells that produce multiple, naturally occurring IgG molecules, half of them are produced. (Monomer by-products) are present, but they can be easily removed by known size exclusion chromatography. They are being taken away.

[0046] In one embodiment, the present invention provides a substitute for one (bispecific) antibody of interest. The production of a defined mixture containing at least two different Ig-like molecules in a single cell. The method reduces or eliminates the formation of other undesirable dimeric antibody species. The resulting mixture is clearly defined, and its composition is determined by the design of the CH3 domain mutation. It is controlled. Furthermore, different transfers used for regulating and / or expressing the expression level are also controlled. The ratio of the mixture affects its composition.

[0047] In the method according to the present invention, the CH3 domain encoded in the first nucleic acid molecule is the second It selectively pairs with the CH3 domain encoded in the nucleic acid molecule and encodes a third nucleic acid molecule. The CH3 domain is selectively compared to the CH3 domain encoded in the fourth nucleic acid molecule. They form a pair. Furthermore, the present invention provides for at least two different I obtained by the method of the present invention. We provide a mixture of g-like molecules.

[0048] The selection of polypeptides comprising the first and second CH3 domains used in this book The meaning of "selective pairing" is that the resulting polyparticle having the first CH3 domain A substantial dimer having a polypeptide comprising a peptide and / or a second CH3 domain. Basically, all of them are polypeptides with one first CH3 domain and one second CH3 This means it is a dimer consisting of a pair of polypeptides that each possess a domain.

[0049] Similarly, "Selective pairing of polypeptides comprising the third and fourth CH3 domains" The meaning of "ng" is the polypeptide that is obtained as a result and has a third CH3 domain. substantially all dimers having polypeptides with a fourth CH3 domain However, a polypeptide having one third CH3 domain and one fourth CH3 domain This means that it is a dimer consisting of a pair of polypeptides.

[0050] As a result, a polypeptide comprising four different (A,B,C,D)CH3 domains was obtained. When the encoding nucleic acid molecule is introduced into a single cell, 10 different Ig-like dimers (AA, Instead of a mixture of AB, AC, AD, BB, BC, BD, CC, CD and DD, This results in the production of a mixture of two specific Ig-like molecules.

[0051] As will be detailed below, in one preferred embodiment of the present invention, the first CH3 domain The polypeptide chain comprising has the amino acid substitution T366K, and the second CH3 domain The polypeptide chain containing has the amino acid substitution L351D. The chemical reaction involves selective pairing of polypeptide chains comprising the first and second CH3 domains. This is a suitable means for that purpose.

[0052] The polypeptide chain comprising the first CH3 domain preferably further comprises amino acids It has substitution L351K. Furthermore, the polypeptide chain comprising the second CH3 domain is preferred Furthermore, amino acids selected from the groups Y349E, Y349D, and L368E. It has substitution. Among this group, L368E is the most preferred. Another preferred one In the embodiment, the polypeptide chain comprising the third CH3 domain is an amino acid substitution E35 The polypeptide chain having 6K and D399K and comprising the fourth CH3 domain is ami It contains acid-substituted K392D and K409D.

[0053] In the method according to the present invention, the polypeptide chain comprising each CH3 domain is preferred Furthermore, it has a variable region that recognizes a target epitope. Polypter containing a CH3 domain The variable region, which is part of the ptide chain, preferably shares a common light chain. In that case, the variable region Only the VH differs in the region, while the VL of all variable regions is virtually identical.

[0054] Therefore, in a method according to the present invention provided in a preferred embodiment, the host cell is given The invention further provides a nucleic acid molecule that codes for a common light chain. In one particularly preferred embodiment, Each of the four variable regions of the polypeptide chain comprising four CH3 domains is different It recognizes a target epitope. For example, if the first nucleic acid molecule is specific to antigen A, If the main molecule further encodes a heavy chain, the second nucleic acid molecule is variable specific to antigen B. The heavy chain further contains domains, and the third nucleic acid molecule has a variable domain specific to antigen C. It further encodes a heavy chain containing, and the fourth nucleic acid molecule has a variable domain specific to antigen D. Furthermore, it encodes a heavy chain. In that case, the mixture is Ig-like with AB specific bispecificity. It is produced containing a bispecific Ig-like molecule that is specific to the molecule and CD.

[0055] Formation of single-specific antibodies (AA, BB, CC, or DD specific) or AC, AD, BC Alternatively, the formation of bispecific antibodies specific to BD is reduced or eliminated. The reason is that polypeptides comprising the first and second CH3 domains and the third and By means of selective pairing of polypeptides having 4 CH3 domains. For the production of a defined mixture containing different Ig-like molecules, for example, the 5th and 6th CH Using additional nucleic acid molecules, such as when encoding a polypeptide with three domains. That is certainly possible.

[0056] It is worth noting that the ratio of nucleic acids used in the method according to the present invention is 1:1:1:1. It is not necessary to do so. Also, the ratio of the resulting expressed Ig-like molecules does not necessarily have to be 1:1. It is not necessary. It is possible to prepare an optimized antibody mixture using known methods. For example, the ratio of the expression level of nucleic acid molecules to the resulting Ig-like molecules. The rate uses different genetic elements such as promoters, enhancers, and repressors. Alternatively, by controlling the genomic integration sites of the copy number of the DNA construct encoding the antibody, It can be restricted.

[0057] The means for selective pairing preferably consists of complementary knob-into-holes. Mutations, disulfide bridges, charge mutations including charge inversion mutations, or combinations thereof The means for selective pairing can be selected from a specific range of type variations. Those skilled in the art will understand this. That is, coding a polypeptide chain having a CH3 domain At least four nucleic acid molecules are used as a means of selective pairing due to their charge mutations. It is necessary.

[0058] Furthermore, in certain cases, even unmanipulated wild-type CH3 possesses a wild-type CH3 domain. It is used for the selective pairing of two polypeptide chains. In one particularly preferred embodiment, The means for selective pairing is at least one CH3 variant selected from Table B Differences will be discussed later in this book.

[0059] In a preferred embodiment of the present invention, all four nucleic acid molecules are subjected to the following process: A polypeptide comprising the first and second CH3 domains and the third and fourth CH3 The first and The means for selective pairing of polypeptides comprising a second CH3 domain is, A method for selective pairing of polypeptides having the third and fourth CH3 domains. It is different from a tier.

[0060] In one aspect of the present invention, the method according to the present invention provides the first and second CH3 domains The means for selective pairing of polypeptides comprising the third and fourth This differs from the means for selective pairing of polypeptides comprising a CH3 domain. The meaning of "different" here is that the polypeptides having the first and second CH3 domains For selective pairing, selective pairing of the first and second chains is preferred. It is designed in such a way. In this design, the first, the third and / or There is virtually no interaction with the polypeptide chain containing the fourth CH3 domain. It is done so. In other words, a polypeptide having a first CH3 domain and the aforementioned Dimerization with polypeptide 3 or 4 does not occur in principle, or is very close to it. It is suppressed. Polypeptides with third and fourth CH3 domains are wild-type, A selective pairing method different from the means for selective pairing of the first and second CH3 domains It is equipped with a ring mechanism.

[0061] Recent research, for example, has shown that knob-into-hole technology or the presence of CH3 domains The focus is on producing a single bispecific antibody using mutations (inversions) of charged contact amino acids. The point is being addressed. However, prior to the present invention, other dimer by-products were significantly produced together. Without doing so, the production of a defined mixture of at least two (bispecific) Ig-like molecules is achieved. That has not been achieved.

[0062] This invention relates to a mixture containing a high proportion of bispecific, clearly defined Ig-like molecules. This provides an efficient and controlled method for producing a mixture of two types of compounds. A mixture with two types of dual specificity is desirable. In the system, the proportion of (two) bispecific elements is at least 95%, at least A percentage of 97% or higher can be obtained. This means that at most 5%, at most 3% This means that only a smaller number of single-specific divalent byproducts are obtained. If we leave it as is, the amount of monomer by-products (i.e., half-units) is not very important. The reason for this is... These half-kids can be easily separated by taking advantage of their size difference.

[0063] In another preferred embodiment of the present invention, the first and second CH3 domains are provided. The variable region of the polypeptide chain recognizes different target epitopes, while the third and fourth C The variable region of the polypeptide chain containing the H3 domain recognizes the same target epitope. As a result, mainly one type of bispecific Ig-like molecule and one type of monospecific Ig-like molecule This is the result. For example, if a polypeptide having first and second CH3 domains The variable region of the cytoplasmic chain recognizes different target epitopes, and if the third and fourth CH3 The variable regions of the polypeptide chain containing the domain are both identical, first and second CH When recognizing an epitope different from the target epitope recognized by the 3 domains, AB Alternatively, a mixture of Ig-like molecules with specificity for CC is created.

[0064] A further method provided in accordance with the present invention involves a poly having third and fourth CH3 domains. The target epitopes recognized by the variable region of the peptide chain are identical, but the first or Target epithet recognized by the variable region of the polypeptide chain containing a second CH3 domain It is different from P.

[0065] Alternatively, the variable region of the polypeptide chain comprising the first and second CH3 domains Polypeptides that recognize different target epitopes and possess third and fourth CH3 domains. The variable regions of the chain are both polypeptide chains comprising a first or second CH3 domain and When recognizing the same epitope, specificity for AB and AA, or AB and BB A mixture of Ig-like molecules possessing properties is created.

[0066] The method according to the present invention provides a port comprising third and fourth CH3 domains. The target epitope recognized by the variable region of the lipeptide chain is the first or second CH3 It is identical to the target epitope recognized by the variable region of the polypeptide chain containing the main component. It is.

[0067] Another objective of the present invention is to enable the development of bispecific and monospecific antibodies in single-cell cultures. The objective is to provide a method and means for producing a defined mixture. Non-limiting examples of the regulated mixture include a bispecific antibody specific to AB and a monospecific antibody specific to AA. It is a mixture of monospecific antibodies. Other examples include bispecific antibodies specific to AB and BB. It is a mixture of monospecific antibodies. Another example is bispecific antibodies specific to AB. It is a mixture of antibodies and CC-specific monospecific antibodies. Here again, at least 90%, Preferably 95%, most preferably at least 97%, or even more than 99% A suitable means and method are provided for producing a mixture of antibodies of a target type, which contains antibodies.

[0068] In another embodiment in which the method according to the present invention is provided, the first and second CH3 domains The variable region of the polypeptide chain having the same target epitope recognizes the same target epitope, while the third and The variable region of the polypeptide chain comprising the fourth CH3 domain is the first and second variable It recognizes a second target epitope that is different from the target epitope recognized by the region. This leads to the production of single-specific Ig-like molecules that primarily exhibit AA-specificity or BB-specificity. As a result, the formation of bispecific Ig-like molecules decreases or disappears.

[0069] In some embodiments, a mixture of single-specific antibodies is preferred over a mixture of bispecific antibodies. It is preferred to produce it in a single cell. For example, when crosslinking two identical target molecules is desired. Or, the two targets are too far apart from each other and a single bispecific antibody cannot bind to them. This is when it is not possible. When a single cell produces a mixture of single-specific antibodies, a single therapeutic antibody It can be advantageous because it is considered a product.

[0070] In this field, the therapeutic efficacy and safety of various monospecific antibodies have already been proven. Manufacturing approval has also been obtained. By producing a mixture of single-specific antibodies in single cells, several This facilitates testing of the efficacy and safety of such mixtures, and for regulatory approval and manufacturing. This can reduce the effectiveness and cost. However, it reduces the formation of bispecific byproducts by 5%. A method for producing a specific mixture of monospecific antibodies in a single cell, which can be done in less than [a certain number of] units. Such a bispecific antibody has not yet been obtained. Another objective of the present invention is to obtain such a bispecific antibody. Means for producing a clearly defined mixture of homodimers that form less than 5% The objective is to provide a method of production.

[0071] The method according to the present invention allows for the desired bispecific and / or monospecific Ig-like molecules. It is suitable for the production of mixtures containing more than two different Ig-like molecules. To produce the mixture, for example, the 5th and 6th (and 7th and 8th, etc.) CH It is also possible to use nucleic acid molecules that encode polypeptides having three domains. .

[0072] Preferably, the method according to the present invention includes at least among the mutations provided by the present invention. At least two CH3 domains are used, including one combination of mutations according to the present invention. Through these mutations, a novel, specific interaction is formed between the two CH3 domains. These mutations according to the present invention are described in detail below.

[0073] In this book, the term "Ig-like molecule" refers to a molecule containing at least one immunoglobulin (Ig) domain. This refers to a proteinaceous molecule that possesses at least one immunoglobulin. The sequence comprises a CH3 domain function, preferably the CH3 domain of IgG1. It is equipped with a proteinaceous molecule having at least one CH3 domain. It can be equipped with a joint.

[0074] The CH3 domain of the present invention includes means for selective pairing and a desired heterogeneity To design a combined molecule of a mer or a mixture of combined molecules, two molecules having CH3 domains Used for selective pairing of proteinaceous molecules. Proteinaceous molecules possessing a CH3 domain. The binding site introduced into the molecule is any binding site, including those shown in the following non-limiting examples. It can also be a stepped configuration; single-chain Fvs, single-chain or tandem diabody (TandAb(registered trademark) )), VHH, Anticalins (registered trademark), Nanobodies (registered trademark), BiTE (registered trademark), Fab Ankyrin repeat proteins or DARPINs®, Avimers®, DART, TC R-like antibodies, Adnectins®, Affilins®, Trans-bodies®, Affibodies (registered trademark), TrimerX (registered trademark), MicroProteins, Fynomers (registered trademark) , Centyrins® or KALBITOR®.

[0075] In one preferred embodiment, the binding portion is the variable region of the antibody (i.e., the VH / VL combination) (combined). The variable region, which is part of the polypeptide chain containing the CH3 domain, is preferred. Or they share a common light chain. In such cases, only the VH of the variable region differs. On the other hand, VL is essentially the same across all variable regions.

[0076] In addition to this, or by other means, cytokines, hormones, water-soluble ligands, Ingredients and / or peptides and other molecules can also be introduced into the CH3 domain of the present invention. Cut.

[0077] In a more preferred embodiment, the Ig-like molecule comprises the full length of the Fc backbone. Most preferred In the embodiment, the Ig-like molecule is an antibody. Preferably, the variable regions of these antibodies are They share the same light chain, but may differ in the VH region.

[0078] The term "antibody" as used in this book refers to a protein belonging to the class of immunoglobulins. A molecule having a domain that binds to one or more epitopes on an antigen, anti This refers to molecules that originate from or have sequence homology to the variable region of the body. (Known) The antibodies include IgG1, IgG2, IgG3, IgG4, IgA, IgD, IgE, and I The antibody according to the present invention includes several isotypes such as gM. Alternatively, functional derivatives and / or fragments thereof may also be used. In one preferred embodiment, As an Ig-like molecule, an IgG isotype antibody is produced. The reason for this is that IgG antibodies are This is because, for example, it has a longer half-life compared to other isotype antibodies.

[0079] The antibodies produced by the method according to the present invention contain sequences from murine animals and humans. Antibodies can have sequences of various origins. Antibodies are not all derived from human antibodies, but from a single source. It may consist of sequences derived from the source, or it may be called, for example, a chimeric or humanized antibody. To make it detectable, the sequence may have two or more origins.

[0080] The more closely a therapeutic antibody resembles the antibody naturally present in the target individual, the better (for example, if a human is (In the case of the target, human antibodies). Antibody binding is described in terms of specificity and affinity. Affinity determines which antigen or epitope the binding domain binds to. This is a measure that evaluates the ability to bind to a specific antigen or epitope. Specific binding is Affinity (K D ) at least 1 × 10 -5 M, more preferably 1 × 10 -7 M, yo Suitable for 1 x 10 -9 Defined as a binding with higher affinity than M. Typically used for therapeutic purposes. This is a monoclonal antibody, 1 × 10 -10 A substance with M affinity or higher affinity is used. I can stay.

[0081] In this book, the term "antigen" refers to a substance that, when taken into the body, triggers an antibody reaction in the immune system. It refers to a substance or molecule that causes the production of an antigen. Antigens have various origins, but especially in disease Examples include primordia, tumor cells or other abnormal cells, haptens, or autologous tissues. At the level of antigens, antigens are characterized by their ability to bind to the antigen-binding site of an antibody. Antigen mixture Substances are also considered "antigens." That is, according to those skilled in the art, tumor cell lysates, and While viral particles are often considered "antigens," tumor cell lysates and The compound solution of virus particles also contains many antigenic determinants.

[0082] An antigen has at least one, often two or more, epitopes. The term "antibody" is recognized by the immune system, specifically by antibodies, B cells, or T cells. It refers to a part of an antigen. Epitopes are usually thought to originate from non-self proteins. However, sequences derived from the host can also be classified as epitopes.

[0083] The term "CH3 domain" is well known. The IgG structure consists of two light chains and two heavy chains. It has four chains. Each light chain has two regions: a variable region and a constant region (VL and CL). It has a main chain. Each heavy chain has a variable region (VH) and three constant regions (CH1, It has four domains (CH2,CH3). Regions of the heavy chain CH2 and CH3 domains. Fc (Fragment crystallizable) part, Fc fragment, Fc backbone, or simply Fc It is called [name].

[0084] The IgG molecule consists of two heavy molecules linked by disulfide bonds (-SS-) at the hinge. It is a heterotetramer having a heavy chain and two light chains. The heavy chain has contacts of CH3-CH3 domains. It dimerizes through surface interactions and interactions at the hinge. Disulfide at the hinge The number of bindings varies depending on the immunoglobulin subclass (Papadea and Check 1989).

[0085] The Fc fragment of immunoglobulin is a dimer of two constant regions at the C-terminus. These are the CH2 and CH3 domains of the heavy chain. Part of their physiological function is related to the complement system. This involves interaction with specific receptors on the surface of various cells. It is known that the interaction between the three domains plays a crucial role in inducing heavy chain dimerization. It is being done.

[0086] Therefore, the CH3 domain plays a leading role in the linkage of the antibody's heavy chain. The contact surface between the main chains contains more than 20 contact residues from each chain, CH3-C It plays a role in H3 interactions (Deisenhofer J., Biochemistry 1981(20)2361-2370) ; Miller S., J. Mol. Biol. 1990(216)965-973; Padlan, Advances in Protein Chemist ry 1996(49)57-133).

[0087] The mutant CH3 domain of the present invention is used in conjunction with other antibody domains and has bispecificity. Alternatively, full-length antibodies with single specificity can be generated. The specificity of the antibody being evaluated is generally related to the behavior of heavy chain dimerization induced by the CH3 domain. It does not affect it.

[0088] The terms "contact residue," "contact amino acid," "contact surface residue," and "contact surface amino acid" used in this book are: The term "acid" usually refers to any ammonium compound present in the CH3 domain that may be involved in contact between domains. This refers to the no-acid residue. In this regard, C in the presence and absence of the second chain The solvent-accessible surface area (ASA) of the residues in the H3 domain In the calculation, there was a difference in ASA (>1 Å) under two conditions. 2 ) residues that indicate contact residues By known techniques, including methods for identification (Lee and Richards J. Mol. Biol. 1971(55)379) It can be calculated. The residues identified as contact residues, according to the EU numbering system, are 347, 349,350,351,352,353,354,355,356,357,360, 364,366,368,370,390,392,394,395,397,399, These are located at positions 400, 405, 407, 409, and 439 (Table A).

[0089] [Table A]

[0090] The contact residues at the CH3-CH3 contact surface can be either charged amino acids or neutral amino acid residues. Good. The terms "charged amino acid residue" or "charged residue" used in this book refer to electrically charged amino acid residues. This refers to amino acid residues with charged side chains. These include arginine (Arg,R), Positively charged side, such as that found in histidine (His,H) and lysine (Lys,K) It can be in chain form, or present in aspartic acid (Asp,D) and glutamic acid (Glu,E). A negatively charged side chain like the one described above would also be acceptable.

[0091] In this book, the term "neutral amino acid residue" or "neutral residue" refers to an electrically charged amino acid residue. This refers to all other amino acids that do not have a side chain. These neutral residues include serine ( Ser,S), threonine (Thr,T), asparagine (Asn,N), glutamine ( GLu,Q), cysteine ​​(Cys,C), glycine (Gly,G), proline (Pro ,P), Alanine (Ala,A), Valine (Val,V), Isoleucine (Ile,I) , Leucine (Leu,L), Methionine (Met,M), Phenylalanine (Phe,F It contains tyrosine (Tyr,Y) and tryptophan (Trp,T).

[0092] In this book, "CH3-CH3 domain contact surface" or "CH3 contact surface" or "CH3-C H3 pairing, domain contact surface, or simply "contact surface" is equipped with a CH3 domain. As a result of the interaction of amino acid residues between two CH3 domains of different polypeptides This refers to the relationship between the amino acids of the first CH3 domain and the second CH3 domain. It is at least one interaction between amino acids. Such interactions include, for example, Van der Waals forces, hydrogen bonds, water-mediated hydrogen bonds, salt bridges, or other electrostatic forces, These include attractive interactions between aromatic side chains, the formation of disulfide bonds, or other known forces.

[0093] Polypeptides comprising first and second CH3 domains, as used in this book. and the polypeptide comprising the third and fourth CH3 domains, before selective pairing The means used may be any known means.

[0094] In one embodiment, at least one nucleic acid molecule is, for example, R, F, Y, W, I This refers to a large amino acid residue like L (i.e., a "knob" or "projection") at the position of the contact residue. It encodes the CH3 domain contained in, while at least one other nucleic acid molecule, for example small amino acid residues such as G, A, S, T, or V (i.e., "holes" or "cavities") This encodes a CH3 domain located at the complementary contact residue position. The CH3 domains are paired with each other by the stereoconformation of the contact amino acids. This is likely to happen. The knob-into-hole technology has already been explained in detail here.

[0095] In a further embodiment of the present invention, at least one nucleic acid molecule has an originally charged residue In the position of the contact residue that was originally K, H, R, D, or E, wild Compared to the type, it encodes a CH3 domain containing amino acids that hold the opposite charge, while on the other hand, At the very least, one other nucleic acid molecule will have a complementary contact residue at the location where the originally charged residue was located. It encodes a CH3 domain containing an amino acid that holds the opposite charge compared to the wild type. The manipulated CH3 domains obtained are opposite each other due to the opposite charge of the contact amino acids. While they tend to pair up, CH3 domains of the same type pair up due to electrostatic repulsion. It's annoying.

[0096] According to one embodiment, European Patent Application Publication No. 01870459 (Patent Document 3), International Patent The CH3 mutation described in Publication No. 2009 / 089004, Gunasekaran et al (2010) I am using it.

[0097] In one embodiment, selection of polypeptides comprising the first and second CH3 domains The means of selective pairing are "knob" and "whole" amino acid residues, and the third and A means of selective pairing of polypeptides having a fourth CH3 domain is by manipulating the charge. It is an amino acid. Preferably, a polypeptide comprising the first and second CH3 domains. Selective pairing of a polypeptide comprising the thydore and the third and fourth CH3 domains Both of the aforementioned means are charge-manipulated amino acids.

[0098] In one embodiment, selection of polypeptides comprising the third and fourth CH3 domains Compared to the amino acid residues manipulated for selective pairing, the first and second CH The amino acid residues manipulated for selective pairing of polypeptides with three domains are different.

[0099] In one particularly preferred embodiment, at least the first and second nucleic acid molecules are provided in the present invention. This invention encodes a novel mutation in the CH3 domain. As detailed below, the present invention is remarkable. Without producing an undesirable amount of (dimeric) by-products, the desired specific dimer This invention provides a novel CH3 mutation that enables the production of sex Ig-like molecules. Furthermore, the present invention offers remarkable Without producing an undesirable amount of (dimeric) by-products, the desired specific single unique This invention provides a novel CH3 mutation that enables the production of sexual Ig-like molecules. The use of at least one of these CH3 mutations is preferred.

[0100] The terms "polypeptide," "polypeptide molecule," or "polypeptide chain" as used in this book. The term "protein" refers to a chain of amino acids linked together by covalent bonds, specifically peptide bonds. It is generally composed of one or more polypeptide molecules. One end, called the amino terminus or N-terminus, has a free amino group. The other end containing the ruboxyl group is called the carboxyl terminus or C-terminus. Polypeptides can undergo post-translational modification processes, such as glycosylation. Therefore, the polypeptide chain comprising the CH3 domain of the present invention has at least Ig·CH3 This refers to a polypeptide chain that contains a domain, and may include those that have undergone post-translational modifications.

[0101] In this book, the term "nucleic acid molecule" refers to a chain of nucleotides, more preferably DNA. It is defined as a molecule consisting of nucleotides and / or RNA. In one embodiment, double-stranded RNA is used. In other embodiments, the nucleic acid molecule of the present invention is, for example, a DNA / RNA helix, a peptide Peptide nucleic acid (PNA), locked nucleic acid (LNA) It has other types of nucleic acid structures, such as ribozymes. Thus, the "nucleic acid molecule" The term is identical to non-natural nucleotides, modified nucleotides, and / or natural nucleotides. It also includes chains that contain non-nucleic acid components that exhibit function.

[0102] Furthermore, the present invention relates to a method for creating host cells that produce at least two different Ig-like molecules. In the method, the method involves introducing at least first, second, third, and fourth C into the host cell. The process includes introducing a nucleic acid sequence that encodes a polypeptide chain containing an H3 domain, and a small number of At least two of the nucleic acid sequences comprising the first and second CH3 domains Selective pairing of polypeptides comprising the third and fourth CH3 domains The present invention provides a method comprising steps, wherein the nucleic acid sequences are introduced sequentially or simultaneously.

[0103] In a further aspect of the present invention, a host cell for producing heterodimer Ig-like molecules A method for producing a C, wherein the method involves giving the host cell at least first and second C The process includes introducing a nucleic acid sequence that encodes a polypeptide chain having an H3 domain, The polypeptide chain containing the first CH3 domain was positively charged from the neutral amino acid residues. A poly having at least one substitution to an amino acid residue and comprising the second CH3 domain The peptide chain has at least one negatively charged amino acid residue from a neutral amino acid residue. The nucleic acid sequences are introduced sequentially or simultaneously, before the host cells are prepared. The method preferably involves introducing a nucleic acid sequence that codes for a light chain common to the host cell. It is equipped with.

[0104] In one aspect further provided in this book, at least the first, second, third, and fourth CH3 Recombinant host cells containing nucleic acid sequences encoding polypeptide chains with domains And, in at least two of the nucleic acid molecules, the first and second CH3 domains are provided Selective pairing of a peptide with the third and fourth CH3 domains of the polypeptide It is equipped with means of ing.

[0105] Furthermore, the present invention relates to a polypeptide comprising at least a first and a second CH3 domain. In recombinant host cells having a nucleic acid sequence encoding a chain, the first CH3 domain Polypeptide chains equipped with this feature have a small amount of transition from neutral amino acid residues to positively charged amino acid residues. A polypeptide chain having at least one substitution and comprising the second CH3 domain is neutral. A combination comprising at least one substitution from a neutral amino acid residue to a negatively charged amino acid residue. Replacement host cells are provided.

[0106] The recombinant host cells according to the present invention preferably comprise a nucleic acid sequence encoding a common light chain. ru.

[0107] The "host cell" of the present invention is any host cell capable of expressing recombinant DNA molecules. Cells are also acceptable. For example, Escherichia (e.g., E. coli), Enterobacter, Salmonalla, B Bacteria such as Acillus, Pseudomonas, Streptomyces, S. cerevisiae, K. lactis, P. Yeasts such as pastoris, Candida, or Yarrowia, Neurospora, Aspergillus oryzae, Filamentous fungi such as Aspergillus nidulans and Aspergillus niger, and Spodoptera frugiper Insect cells such as da SF-9 or SF-21 cells, and preferably CHO(Chinese hamster Mau cells including er ovary cells, BHK cells, SP2 / 0 cells and NS-0 myeloma cells Primate cells such as s cells, COS and Vero cells, MDCK cells, BRL 3A cells, Hybridoma, tumor cell, immortalized primary cell, W138, HepG2, HeLa, H Mammalian cells such as embryonic retinal cells like EK293, HT1080, or PER.C6 are included. It can be done.

[0108] In selecting expression systems, it is often the case that antibodies are properly glycosylated, such as in milk. Expression vectors and hosts for various types of cells may be used. Human cell lines, preferably PER.C6 This is advantageous for obtaining antibodies that match the glycosylation pattern in humans. Conditions for growing or proliferating cells (Tissue Culture, Academic Press, Kruse an (See d. Paterson, editors (1973)) and the conditions for the expression of recombinant products are somewhat different. In some cases, the proportion of the product and / or by methods generally known to those skilled in the art may be determined. Alternatively, process optimization is usually performed to increase the growth of cells in relation to each other.

[0109] General guidelines, procedures, and practical techniques for maximizing productivity in mammalian cell cultures. Mammalian Cell Biotechnology: a Practical Approach (M. Butler, ed., IRL Pres. This can be seen in (s, 1991). Antibody expression in recombinant host cells is widely described in publicly available literature. (For example, European Patent Application Publication No. 0120694, European Patent Application Publication No. 03141) Patent No. 61, European Patent Application Publication No. 0481790, European Patent Application Publication No. 0523949, (U.S. Patent No. 4816567, International Publication No. 00 / 63403). Light chain and heavy chain coated The nucleic acid molecules are copied outside the chromosome and / or stably integrated into the chromosomes of the host cell. However, the latter is preferable.

[0110] In a further aspect of the present invention, cultures of recombinant host cells according to the present invention, or obtainable In cultures of recombinant host cells that are possible or obtained by the method according to the present invention, The nutrients consist of at least two different Ig-like molecules, or one of the heterodimeric Ig-like molecules. Products that produce [something] are provided.

[0111] When obtaining expression of the sequence of a nucleic acid molecule encoding a polypeptide containing a CH3 domain, And sequences that can induce such expression include polypeptides containing a CH3 domain. It is known that the sequence of nucleic acid molecules encoding tides is functionally related. The meaning of "having a chain" is that the polypeptide or its precursor possesses a CH3 domain. The nucleic acid sequence to be used is the expression of a polypeptide or its precursor that contains a CH3 domain. In the sense that it can be induced, it is related to sequences that can induce expression. be.

[0112] For example, useful expression vectors such as Invitrogen's pcDNA vector series are available. It is available. The sequence encoding the target polypeptide is the encoded polypeptide. When appropriately inserted into an array that controls the transcription and translation of a peptide, the resulting expression cassette is useful for producing, i.e., expressing, the polypeptide of interest. There is.

[0113] Sequences that induce expression include promoters, enhancers, etc., and combinations thereof. These need to be functional in the host cell and thereby induce the expression of functionally related nucleic acid sequences. The promoter can be constitutive or regulatable and can be obtained from various sources including viruses, prokaryotes or eukaryotes, or can be artificially designed. The promoter can be constitutive or regulatable and can be obtained from various sources including viruses, prokaryotes or eukaryotes, or can be artificially designed. It can be obtained from various sources including viruses, prokaryotes or eukaryotes, or can be artificially designed. .] There is.

[0114] Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. Suitable promoters can also be obtained from eukaryotic cells, such as the metallothionein (MT) promoter, the elongation factor 1α (EF-1α) promoter, the actin promoter, the immunoglobulin promoter, the heat shock promoter, etc. Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. 1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. Expression of the nucleic acid of interest can occur by a natural promoter or a derivative thereof, or a completely heterologous promoter. Well-known and commonly used promoters for expression in eukaryotic cells include virus-derived promoters such as those of adenovirus (e.g., the E1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., the CMV immediate early (IE) promoter), promoters derived from simian virus 40 (SV40), etc. [[ID=No. 34]]There are actin promoters, immunoglobulin promoters, heat shock promoters, etc.

[0115] Any promoter or enhancer / promoter that can induce the expression of the target sequence in the host cell is suitable for the present invention. In one embodiment, the sequence that can induce expression comprises the region of the CMV promoter, preferably the CMV immediate early gene enhancer. Any promoter or enhancer / promoter that can induce the expression of the target sequence in the host cell is suitable for the present invention. In one embodiment, the sequence that can induce expression comprises the region of the CMV promoter, preferably the CMV immediate early gene enhancer. It has the region of the CMV promoter and preferably the CMV immediate early gene enhancer. - / The promoter has a nucleotide region from -735 to +95. Those skilled in the art will notice The expression sequence used in this invention consists of an insulator, a matrix-binding region, Releases such as STAR Elements (International Publication No. 03 / 004704) A combination of elements that stabilize or promote expression would be preferable. This would improve the stability of expression. And / or the level can be raised.

[0116] Protein production in recombinant host cells is, for example, described in Current Protocols in Protein S cience, 1995, Coligan JE, Dunn BM, Ploegh HL, Speicher DW, Wingfield PT, ISBN 0- As described in 471-11184-8, Bendig, 1988, there is a wide range of information available. Cell culture is, To cause those cells to metabolize, and / or grow, and / or divide, and / or This is achieved by producing the target protein. This can be achieved by methods known to those skilled in the art. This process involves not only providing nutrients to the cells, but also attaching to the surface. This includes methods of propagation while growing, methods of propagation in suspension, or combinations thereof.

[0117] Some culture conditions can be optimized by known methods, and the amount of protein produced can be optimized. This can be done. Culturing can be done, for example, in dishes, roller bottles, or reaction vessels, in batch culture or fed-aggregate culture. This is done by cell culture, continuous culture, hollow fiber culture, etc. Large-scale (continuous) culture is performed by cell culture. To produce recombinant proteins, it is necessary to grow them in cells in a suspension. It is known to be preferable. Also, animal or human-derived serum or animal or human It is known that it is preferable to culture cells under conditions where the components of the derived serum are absent. Therefore, since there are no additional animal or human-derived proteins from the culture medium, purification is easy. This increases safety. On the other hand, synthetic culture media are optimal in terms of reproducibility, so the system is also It will also become reliable.

[0118] Ig-like molecules are expressed in host cells and are generally obtained from those cells, or preferably from the cell culture medium. They are recovered by methods known to those skilled in the art. After recovery, these Ig-like molecules are recovered by known methods. It can be purified using methods such as immunoprecipitation, centrifugation, filtration, and size removal. De-chromatography, affinity chromatography, cation and / or anion exchange This includes conversion chromatography, hydrophobic interaction chromatography, etc. IgG molecules For antibody mixtures, protein A or protein G affinity chromatography is preferred. It can be used appropriately (see, for example, U.S. Patent 4801687 and U.S. Patent 5151504). ).

[0119] Ig-like molecules and / or mixtures thereof produced by the method according to the present invention are preferred or have a common light chain. Therefore, what is further provided is the method according to the present invention This further comprises the step of providing the host cell with a nucleic acid molecule that codes for a light chain common to the host cell. A light chain that can pair with at least two different heavy chains, thereby forming a functional antigenic linkage. It forms a binding domain. A functional antigen-binding domain specifically binds to one antigen. It is possible.

[0120] A common light chain that can be paired with all heavy chains produced by the method according to the present invention. It is preferable to use this. Thereby, in the formation of a functional antigen-binding domain, mismatching of heavy and light chains that do not match can be avoided. In one aspect, only one common light chain of the same amino acid sequence can be used. In another method, those skilled in the art will recognize that even if the amino acid sequences are not the same, functionally equivalent light chains can be included in the meaning of "common". Many variants of the light chain exist, and although these have mutations (deletions, substitutions, additions), they do not substantially affect the formation of the functional binding region. Therefore, such variants can also bind to different heavy chains and form a functional antigen-binding domain. The term "common light chain" as used herein refers to a light chain that is either the same or has differences in amino acid sequence, but the resulting antibody after pairing with a heavy chain retains its binding specificity. For example, by introducing and testing conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain, it is possible to create or find light chains that are still functionally equivalent even if they are not the same light chain. The term "common light chain" includes combinations of specific common light chains and such functionally equivalent variants. A detailed description of the use of common light chains can be found in WO 2004 / 009618. Preferably, the common light chain used in the present invention is a light chain derived from a germline-like lineage, more preferably a light chain derived from a germline, preferably a rearranged human germline κ light chain, and most preferably a rearranged human It is recognized that even if the amino acid sequences are not the same, functionally equivalent light chains can be included in the meaning of "common". Many variants of the light chain exist, and although these have mutations (deletions, substitutions, additions), they do not substantially affect the formation of the functional binding region. Therefore, such variants can also bind to different heavy chains and form a functional antigen-binding domain. There are many variants of the light chain, and although these have mutations (deletions, substitutions, additions), they do not substantially affect the formation of the functional binding region. Therefore, such variants can also bind to different heavy chains and form a functional antigen-binding domain. Therefore, such variants can also bind to different heavy chains and form a functional antigen-binding domain.

[0121] The term "common light chain" as used herein refers to a light chain that is either the same or has differences in amino acid sequence, but the resulting antibody after pairing with a heavy chain retains its binding specificity. For example, by introducing and testing conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain, it is possible to create or find light chains that are still functionally equivalent even if they are not the same light chain. The term "common light chain" as used herein refers to a light chain that is either the same or has differences in amino acid sequence, but the resulting antibody after pairing with a heavy chain retains its binding specificity. For example, by introducing and testing conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain, it is possible to create or find light chains that are still functionally equivalent even if they are not the same light chain. For example, by introducing and testing conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain, it is possible to create or find light chains that are still functionally equivalent even if they are not the same light chain. Specifically, conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain are introduced and tested. By introducing and testing conservative amino acid changes and / or amino acid changes in regions that do not contribute or only partially contribute to the binding specificity when paired with a heavy chain, it is possible to create or find light chains that are still functionally equivalent even if they are not the same light chain.

[0122] The term "common light chain" includes combinations of specific common light chains and such functionally equivalent variants. A detailed description of the use of common light chains can be found in WO 2004 / 009618. Preferably, the common light chain used in the present invention is a light chain derived from a germline-like lineage, more preferably a light chain derived from a germline, preferably a rearranged human germline κ light chain, and most preferably a rearranged human Preferably, the common light chain used in the present invention is a light chain derived from a germline-like lineage, more preferably a light chain derived from a germline, preferably a rearranged human germline κ light chain, and most preferably a rearranged human Preferably, the common light chain used in the present invention is a light chain derived from a germline-like lineage, more preferably a light chain derived from a germline, preferably a rearranged human germline κ light chain, and most preferably a rearranged human ​​It is a germline-derived κ light chain, either IgVκ1-39 / Jκ or IgVκ3-20 / Jκ. .

[0123] Alternatively, instead of using a common light chain, mismatches in heavy and light chains can be made. To avoid pairing, a person skilled in the art can, for example, refer to International Publication No. 2009 / 080251, International This is explained in Publication No. 2009 / 080252 and / or International Publication No. 2009 / 080253. One can choose a method for forcing heavy and light chain pairing, as described.

[0124] In this invention, not only are there novel combinations of manipulated CH3 mutations, but also novel manipulated The present invention provides a CH3 domain. Prior to this invention, the invention was related to CH3-CH3 pairing. A known charged contact amino acid of the CH3 domain has the opposite charge (charge reversal). It was substituted with an acid, which affected the CH3-CH3 pairing.

[0125] The mutation according to this invention is a different creation from this approach. The reason is that wild-type CH3 In this case, an uncharged or neutral CH3 amino acid is substituted with a charged residue. In this embodiment of the present invention, charged contact amino acids are converted to amino acids with the opposite charge. Instead of replacing them, uncharged CH3 amino acids are substituted with charged ones.

[0126] The approach of the present invention simply provides a method for efficiently promoting the dimerization of the CH3 domain. Furthermore, at the CH3 contact surface, at least one additional charge-charge interaction is created. This has the advantage of being able to do so. In addition to the charge pairs present on the CH3-CH3 contact surface, Due to additional charge-charge interactions, the dimer according to the present invention is a wild-type dimer (wild-type dimer). The mer is defined as bispecific IgG(AB) that has not undergone CH3 manipulation, and is homozygous for the parent. It is generally more stable compared to the dimer (AA or BB) in contrast.

[0127] Furthermore, surprisingly, the proportion of one or more target Ig-like molecules in the mixture is... It is possible to further increase this. As mentioned above, generally, bispecific antibodies are preferentially produced. Known methods of production produce undesirable dimer byproducts. For example, knob-in Using the two-hole technique, the percentage of the desired bispecific antibody is at best 87%. On the other hand, electrostatic operations in which charged contact amino acids are substituted with amino acids of the opposite charge In the approach, the ratio is 96% (see, for example, Example 11).

[0128] To our great surprise, the inventors of this invention further reduced the proportion of the desired Ig-like molecules in the mixture. We succeeded in introducing a mutation that enhances the effect. For example, in Example 17, the mutation according to the present invention was used Using this method, the desired high proportion of dimer by-products was obtained in the resulting mixture, to the point where no dimer by-products were detected at all. It has been shown that a bispecific antibody can be obtained in which one heavy chain is paired with a common light chain. While unpaired half-units exist to some extent in the mixture, these are heavy chains. As a result of unbalanced expression, it can be easily separated from the mixture by size exclusion chromatography. Cut.

[0129] Therefore, such mutations according to the present invention substantially eliminate the contamination of dimer by-products. Without this, a high proportion of bispecific Ig-like molecules, particularly suitable for pharmaceutical compositions, are produced in single cells. It can be done.

[0130] A preferred embodiment of the present invention provides for the generation of heterodimer Ig-like molecules from a single cell. In the production method, the Ig-like molecule has two CH3 molecules that can form a contact surface. The method comprises a main, and the cell, a. A first nucleic acid molecule encoding a polypeptide chain containing a first CH3 domain. b. A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the step of giving a neutral a The second is comprising at least one substitution from an amino acid residue to a positively charged amino acid residue, Polypeptide chains containing a CH3 domain are formed from neutral amino acid residues to negatively charged amino acids. The method comprises substituting at least one no-acid residue, and the method involves culturing the host cells and the 2 The process of expressing one nucleic acid molecule and recovering the heterodimer Ig-like molecule from the culture is further Prepare. Preferably, the method provides nucleic acid molecules that encode a light chain common to the host cell. Further improvements will be made, and the advantages resulting from these improvements are outlined above.

[0131] The amino acid at position 366 of one CH3 domain and the position of another CH3 domain The amino acid at position 351 forms a contact residue pair at the CH3-CH3 contact surface. It has been reported that these are the three-dimensional conformations of the resulting Ig-like molecules. They are located close enough in the context to interact with each other. Therefore The first CH3 domain is preferentially paired with the second CH3 domain.

[0132] In one embodiment, the threonine (T) at position 366 of the first CH3 domain is the first The charged amino acid is replaced, and the leucine (L) at position 351 of the second CH3 domain is The first and second charged amino acids are replaced by a second charged amino acid, and the first and second charged amino acids are replaced by an opposite charged amino acid. It has a charge. If the poly has a first CH3 domain that holds a charged residue at position 366, The peptide further possesses a variable region that is specific to antigen A, and if opposite A polypeptide comprising a second CH3 domain that retains the charged residue of the charge at position 351 Furthermore, if it has a variable region that is specific to antigen B, then a dual antigen with AB specificity is formed. Specific Ig-like molecules are predominantly formed.

[0133] A method further provided by the present invention comprises the first and second CH3 domains The means for selective pairing of polypeptides, or the third and fourth CH3 domes The means for selective pairing of polypeptides comprising the first or third CH Substitution of the threonine at position 366 of the 3 domain to the first charged amino acid and the second Alternatively, this is a substitution that turns the leucine at position 351 of the fourth CH3 domain into the second charged amino acid. Furthermore, the first and second charged amino acids have opposite charges.

[0134] One preferred combination of mutations according to the present invention comprises a first CH3 domain and a variable region At position 366 of a polypeptide further comprising (for example, specific to A), threonine A substitution of (T) to lysine (K), and a second CH3 domain, with a variable region (for example) If, at position 351 of the polypeptide which is further specific to B, leucine (L) This is a substitution to convert to aspartic acid (D). This is expressed as a mutation of the T366K / L351'D pair. It can be done.

[0135] As mentioned above, the position of the first CH3 domain is 366 and the second CH3 domain is The amino acid at position 351 has been reported to be a pair of contact residues at the CH3-CH3 contact surface. Lysine introduced at position 366 and aspartic acid introduced at position 351 are Having opposite charges, these amino acids electrostatically attract each other. Therefore, the first CH The 3 domains selectively attract the second CH3 domain. Additionally, lysine is added at position 366. It has a first CH3 domain and a second CH3 domain having aspartic acid at position 351. Ig-like molecules paired with nucleotides are predominantly formed.

[0136] If the polypeptide possessing the first CH3 domain has specificity for antigen A, If a polypeptide possessing a second CH3 domain has specificity for antigen B, In total, "AB"-specific bispecific Ig-like molecules are predominantly formed. Note: In that embodiment, the polypeptide chain comprising the first and second CH3 domains The variable regions may be the same in both cases, in which case a single-specific Ig-like molecule (for example, " The formation of "AA" specificity occurs as a result.

[0137] As mentioned above, one of the advantages of the mutation according to the present invention is that the original charged amino acids Instead of replacing the interaction, a novel phase is introduced between the newly introduced pair of charged amino acids. An interaction is generated. This point has not been disclosed or suggested to date.

[0138] One aspect of the present invention provides at least two different Ig-like molecules from a single host cell. In the production method according to the present invention, the polypeptide chain comprising the first CH3 domain The polypeptide chain comprising the amino acid substitution T366K is It is equipped with the amino acid substitution L351D.

[0139] In one embodiment of the method for producing heterodimer Ig-like molecules from a single cell, The Ig-like molecule comprises two CH3 domains capable of forming a contact surface, and the method The cells mentioned above, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the first CH3 domain is provided with amino acid substitution The polypeptide chain comprising T366K and the second CH3 domain is an amino acid substitution L The method comprises 351D, and the host cells are cultured, the nucleic acid molecules are expressed, and the The system further includes a step for recovering telodimer Ig-like molecules from the culture.

[0140] By using the above-described amino acid mutation according to the present invention, a heterodimer I can be obtained from a single cell. This makes it possible to produce g-like molecules. As a result, the inclusion of homodimers is less than 5%. Preferably less than 2%, more preferably less than 1%, or most preferably This method can virtually eliminate the contamination of homodimers.

[0141] One embodiment provides a method for producing heterodimerized Ig-like molecules from a single cell, The aforementioned Ig-like molecule has two CH3 domains that can form a contact surface, and is mixed The presence of homodimers is less than 5%, preferably less than 2%, and more preferably less than 1%. The method involves, most preferably, the presence of homodimers, and the cells, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the step of giving the first CH3 domain is an amino acid The polypeptide chain comprising the substitution T366K and the second CH3 domain is an amino acid The method comprises replacing L351D, and involves culturing the host cells and expressing the two nucleic acid molecules. The method further comprises a step of recovering the Ig-like molecule of the heterodimer from the culture.

[0142] Preferably, a method for producing at least two different Ig-like molecules according to the present invention, or In the method for producing a heterodimer Ig-like molecule according to the present invention, the first CH3 domain A polypeptide chain comprising the above further comprises the amino acid substitution L351K. More preferably, The polypeptide chains comprising the second CH3 domain are Y349E, Y349D, and L The amino acid substitution further comprises an amino acid substitution selected from the group consisting of 368E. Most preferably, the above The polypeptide chain containing the second CH3 domain has the amino acid substitution L368E.

[0143] Therefore, in one preferred embodiment, the above-mentioned T366K / L351'D mutation according to the present invention is The leucine (L) at position 368 of the CH3 domain of 2 is converted to glutamic acid (E). This can be further combined with mutations. For example, with the T366K / L351'D, L368'E mutation. It can be displayed (however, T336K / L351D-L368E or T366K / L351D, L368E or T366K-L351D, L (Other display methods, such as 368E, are also possible.)

[0144] As shown in Example 17, the first CH3 domer specific to antigen A according to the present invention A polypeptide comprising a protein and a polypeptide comprising a second CH3 domain specific to antigen B Introducing this mutation into the lipeptide results in a bispecific Ig-like molecule with dual AB specificity. This can be obtained in a particularly good proportion. This pair of mutations allows for a detectable amount of homodimers. Without the formation of a specific molecule, bispecific antibodies can be obtained.

[0145] A preferred embodiment provides a method for producing heterodimer Ig-like molecules from a single cell. In this case, the Ig-like molecule comprises two CH3 domains capable of forming a contact surface. The presence of homodimers is less than 5%, preferably less than 2%, and more preferably less than 5%. The amount is less than 1%, and most preferably there are substantially no homodimers present, and the method The method is to apply to the aforementioned cells, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The process includes providing a polypeptide chain comprising the first CH3 domain, wherein the polypeptide chain is amino acid substituted The polypeptide chain comprising T366K and the second CH3 domain is an amino acid substitution L The method comprises 351D and L368E, and the method involves culturing the host cells and the two nucleic acids The method further comprises expressing the molecule and recovering the heterodimer Ig-like molecule from the culture.

[0146] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( T) is replaced with lysine (K), and leucine (L) at position 351 of the second CH3 domain is replaced with asparagus. Glycine (D) is converted to glutamine by adding tyrosine (Y) at position 349 of the second CH3 domain. Substitution with acid (E). This is expressed as, for example, the T366K / L351'D,Y349'E mutation, or, for example... It is displayed as T366K-L351D:Y349E or T366K / L351D,Y349E or simply T366K / L351DY349E .

[0147] The Y349 residue is located near the residue at position 351, which can contribute to dimer interactions. According to silico data, Y349E is a single dimer destabilization (higher computational s This not only affects the core, but also the stability of the heterodimer (lower calculated score), and In setting 349, glutamic acid (E) is more preferable than aspartic acid (D). Therefore, a second CH3 domain having an amino acid substitution at position 351 is formed. Introducing a second amino acid substitution into a lipeptide makes heterodimerization more likely.

[0148] A particularly preferred embodiment provides the production of heterodimerized Ig-like molecules from a single cell. In this method, the Ig-like molecule has two CH3 domains that can form a contact surface. It is equipped with a homodimer content of less than 5%, and furthermore, less than 2%. More preferably less than 1%, most preferably substantially none, the method is used to the cells , - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the first CH3 domain is provided with amino acid substitution The polypeptide chain comprising T366K and the second CH3 domain is an amino acid substitution L The method comprises 351D and Y349E, and the method involves culturing the host cells and the two nucleic acids The method further comprises a step of expressing the molecule and recovering the heterodimer Ig-like molecule from the culture. .

[0149] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( T) is replaced with lysine (K), and aspartic acid (D) at position 351 of the second CH3 domain is replaced with Icin (L) is converted to glutamine by adding tyrosine (Y) at position 349 of the second CH3 domain. To the acid (E), leucine (L) at position 368 of the second CH3 domain is converted to glutamic acid ( Substitute with E). This is represented as the T366K / L351'D,Y349'E,L368'E mutation. Two residues Y349 and L368 are residues that can contribute to dimer interactions.

[0150] In silico data indicates that Y349E and L368E destabilize the BB dimer. (Higher in silico score) as well as heterodimer stabilization (lower in silico score) (This leads to the score), and glutamic acid at positions 349 and 368 is aspartic acid (D ) is preferable to. Therefore, a second is added to a B-chain that already has an amino acid substitution at position 351. The introduction of a third amino acid substitution further promotes the formation of heterodimers.

[0151] A particularly preferred embodiment provides a method for producing heterodimer-like Ig molecules from single cells. In this law, the Ig-like molecule has two CH3 domains that can form a contact surface. In preparation, the amount of homodimer contamination should be less than 5%, more preferably less than 2%, and even more Preferably less than 1%, most preferably substantially none, the method involves the cells, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The process includes a step of giving, and the polypeptide chain comprising the first CH3 domain is an amino acid substitution The polypeptide chain comprising T366K and the second CH3 domain is an amino acid substitution L The method comprises 351D, Y349E and L368E, and the method involves culturing the host cells and The process involves expressing the two nucleic acid molecules and recovering the heterodimer Ig-like molecule from the culture. Prepare further.

[0152] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( T) is replaced with lysine (K), and the leucine (L) at position 351 of the first CH3 domain is replaced with lysine. In (K), leucine (L) at position 351 of the second CH3 domain is added to aspartic acid (D ) and convert the leucine (L) at position 368 of the second CH3 domain to glutamic acid (E). Substitute. This is represented as the T366K,L351K / L351'D,L368'E mutation. As shown in the examples... Similarly, this mutation also increases the proportion of the target (bispecific) antibody. Furthermore, this mutation Therefore, no homodimer formation in detectable amounts is achieved, and a bispecific antibody can be obtained.

[0153] Furthermore, in the method provided for producing heterodimer-like Ig molecules from single cells, The Ig-like molecule possesses two CH3 domains that can form a contact surface, and is homodimer. The amount of contamination should be less than 5%, more preferably less than 2%, and even more preferably less than 1%. The method involves, at least, most preferably substantially none, the cells, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the step of giving the first CH3 domain is an amino acid polypeptide comprising substitutions T366K and L351K, and comprising the second CH3 domain. The chain comprises amino acid substitutions L351D and L368E, and the method cultures the host cells. The cells are cultured to express the two nucleic acid molecules, and the heterodimer Ig-like molecule is recovered from the culture. Further processes are included.

[0154] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( T) is replaced with lysine (K), and the leucine (L) at position 351 of the first CH3 domain is replaced with lysine. In (K), leucine (L) at position 351 of the second CH3 domain is added to aspartic acid (D ) and the tyrosine (Y) at position 349 of the second CH3 domain is replaced with aspartic acid (D). Then, the arginine (R) at position 355 of the second CH3 domain is replaced with aspartic acid (D). Replace with this. This is displayed as the T366K,L351K / L351'D,Y349'D,R355'D mutation. This T366K -The L351K / L351'D-Y349'D pair is further improved by the R355'D mutation in the B strand, and BB The in silico score for AB increased, and the in silico score for AB also increased slightly. Yes.

[0155] Furthermore, in the method provided for producing heterodimer-like Ig molecules from single cells, The Ig-like molecule possesses two CH3 domains that can form a contact surface, and is homodimer. The mixture should be less than 5%, more preferably less than 2%, and even more preferably less than 1%. The method involves, at least, most preferably substantially none, the cells, - A first nucleic acid molecule encoding a polypeptide chain having a first CH3 domain, and - A second nucleic acid molecule encoding a polypeptide chain containing a second CH3 domain. The polypeptide chain comprising the step of giving the first CH3 domain is an amino acid polypeptide comprising substitutions T366K and L351K, and comprising the second CH3 domain. The chain comprises amino acid substitutions L351D, Y349D and R355D, and the method is as follows: Host cells are cultured to express the two nucleic acid molecules, and the heterodimer Ig-like molecule is cultured. Further processes will be implemented for harvesting and recovery.

[0156] Table B lists preferred means of selective pairing for the formation of heterodimers or homodimers. This is a list of mutations introduced into the CH3 domain. [Table B]

[0157] A method for producing at least two different Ig-like molecules according to the present invention, A method for producing a heterodimer Ig-like molecule comprising the first and second CH3 domains is described above. The means for selective pairing of lipeptides and / or the third and fourth CH3s The means for selective pairing of polypeptides comprising a main is in at least one Table B The combination of mutations shown is provided, thereby. Preferably, the first and The means for selective pairing of polypeptides comprising a second CH3 domain, and the third The means for selective pairing of polypeptides having a fourth CH3 domain is less At the very least, it possesses the two mutation combinations shown in Table B.

[0158] The novel CH3 mutation combinations provided by the present invention involve at least two mutations in a single cell. It is possible to produce a mixture of single-specific Ig-like molecules, and contamination with bispecific Ig-like molecules is possible. Less than 5%, preferably more than 2%, more preferably less than 1%, most preferably There is virtually no other possibility. These mutations according to the present invention are particularly useful for the production of mixtures of monospecific antibodies. This is suitable for tumors. This is because when a high level of crosslinking of two identical target molecules is desired, tumors To activate specific effector mechanisms such as complement-mediated lysis of cells, in target cells When the antibody density needs to be sufficiently high, or when the two targets are too far apart from each other, When the bispecific antibody does not bind, or when the procedure for obtaining regulatory approval is simpler It is particularly useful for that purpose.

[0159] In such cases, a production platform for such monospecific antibodies is optimal. It is often desirable to convert it. As shown in Example 10 and as can be seen from the present invention, the A polypeptide having a CH3 domain at position 392 (for example, having A specificity) Lysine (K) is substituted with aspartic acid (D), and the poly having the first CH3 domain is obtained. The aspartic acid (D) at position 399 of the peptide is replaced with lysine (K), and the first CH The lysine (K) at position 409 of a polypeptide with three domains is converted to aspartic acid (D). When substituted, it contains at least two different monospecific Ig-like molecules with AA specificity. A mixture of monospecific Ig-like molecules can be produced in a single cell, and a bispecific by-product (two The formation of heavy specificity Ig-like molecules is less than 5%, or less than 3%, or substantially less. It decreases to the point where it is not detectable at all.

[0160] Therefore, the above combination of mutations (represented as K392D, D399K, K409D) is single-specific. It is particularly preferable for the production of mixtures of sexual Ig-like molecules. Those skilled in the art will know that functional variants, that is, It can be recognized that K392E, D399R, and K409E can achieve similar results. In addition, there are double mutants with substitutions of D399K and K409D, or K392D and K40 The 9D, D399R, and K409E models could achieve similar results.

[0161] The glutamic acid (E) at position 356 of the polypeptide having the first CH3 domain is lysine At position 357 of the polypeptide comprising the first CH3 domain, the glutamine (K) Acid (E) is converted to lysine (K) at position 43 of the polypeptide comprising the first CH3 domain. The lysine (K) of 9 is converted to aspartic acid (D), and the polyp comprising the first CH3 domain is converted. Mutation combinations when lysine (K) at position 370 of the plutide is replaced with aspartic acid (D) The same applies to combinations. This combination of mutations (displayed as E356K, E357K, K439D, K370D) It is particularly preferable for the production of mixtures of single-specific Ig-like molecules.

[0162] Those skilled in the art will know that functional variants, namely K356R, E357R, K439E, and K370E, result in It can be recognized that similar effects can be achieved. In addition, E356K and K439D, E357 Triple or double mutants with K and K370D substitutions, or other functional mutants. It could achieve a similar effect.

[0163] A further embodiment provides at least two different single cells from a single host cell. In the method for producing heterozygous Ig-like molecules, each of the two Ig-like molecules forms a contact surface. The method comprises two CH3 domains that can perform the following: -A-specific first nucleus encoding a polypeptide chain comprising a first CH3 domain Acid molecules, and - A second nucleus encoding a polypeptide chain having a second CH3 domain with B-specificity acid molecule The polypeptide chain comprising the step of giving the first CH3 domain is K392D , possessing mutations D399K and K409D, and having the second CH3 domain The ptydos chain either contains a wild-type CH3 domain or E356K, E357K, K4 The method comprises the 39D and K370D mutations, and involves culturing the host cells and the nucleic acid molecules The process further comprises expressing the molecule and recovering at least two different Ig-like molecules from the culture. El.

[0164] Another embodiment provides at least two different single-specific I cells from a single host cell. In the method for producing γ-like molecules, each of the two γ-like molecules can form a contact surface. The method comprises two CH3 domains, and the cell, -A-specific first nucleus encoding a polypeptide chain comprising a first CH3 domain Acid molecules, and - A second nucleus encoding a polypeptide chain having a second CH3 domain with B-specificity acid molecule The polypeptide chain comprising the first CH3 domain is wild-type C It has an H3 domain or has mutations K392D, D399K, K409D, and the second Polypeptide chains containing the CH3 domain include E356K, E357K, K439D, K The method comprises culturing the host cells, expressing the nucleic acid molecule, and The method further comprises a step of recovering at least two different Ig-like molecules from the culture.

[0165] As shown in Example 10, two single-specific Ig-like molecules were produced in a single cell. The formation of bispecific Ig-like molecules is virtually undetectable. Those skilled in the art will know whether it is wild-type or modified. A third nucleic acid molecule encoding a polypeptide chain containing a CH3 domain was selected, and 3 A mixture of single-specific antibodies may be produced and provided to the host cells.

[0166] Provided in one embodiment of the present invention, at least two different Ig-like molecules according to the present invention In the method for producing the child, or the method for producing the heterodimer Ig-like molecule, each CH3 The polypeptide chain containing the domain further contains a variable region that recognizes different target epitopes. Wait, those target epitopes are located on the same molecule.

[0167] This results in the target molecule being ( It can provide more efficient antagonistic effects to biological functions. For example, heterodimers Ig-like molecules are either receptors for growth factors or water-soluble molecules that are important for the proliferation of tumor cells. It can simultaneously bind to two epitopes present in, for example. This allows for the simultaneous binding of several independent cells. It can efficiently inhibit the signaling pathway, leading to uncontrolled proliferation. And at least In any combination of two Ig-like molecules, such growth factor receptors or It can simultaneously bind to 2, 3, or 4 epitopes present in water-soluble molecules.

[0168] In one preferred embodiment, the target molecule is a water-soluble molecule. In another preferred embodiment, In this case, the target molecule is a membrane-bound molecule.

[0169] Production of at least two different Ig-like molecules according to the present invention in another aspect of the present invention. The method or method for producing heterodimer Ig-like molecules comprises each CH3 domain The polypeptide chain further comprises a variable region that recognizes a target epitope, and the target epitope is They are located on different molecules. In this case, each different target molecule is a water-soluble molecule or It can be used as a membrane-binding molecule.

[0170] In one embodiment, the different target molecules are water-soluble molecules. Alternatively, one target The first target molecule is water-soluble, while the second target molecule is membrane-bound. And yet another case... In this case, both target molecules are membrane-bound molecules. In one embodiment, different targets The target molecule is expressed in the same cell, and in other embodiments, different target molecules are expressed in different cells. It will be revealed.

[0171] As a non-restrictive example, any heterodimer Ig-like molecule or at least two Ig-like molecules Any combination of these receptors can simultaneously inhibit multiple membrane-bound receptors or target tumor cells. It can simultaneously neutralize multiple water-soluble molecules such as tokines or growth factors, or different ones. It is suitable for neutralizing virus serotypes or virus strains.

[0172] One preferred embodiment provides at least two Ig-like molecules or heterodimer Ig In the production method of the molecule according to the present invention, at least one of the target epitopes It is located in tumor cells. In another way, or in addition thereto, at least the target e One of the pitopes is located on the surface of effector cells. This is, for example, tumor cells. It is suitable for mobilizing T cells or NK cells for killing. For example, by the method according to the present invention At least one Ig-like molecule produced is specifically targeted to a target molecule located on an immune effector cell. By heterozygously binding, immune effector cells, preferably human immune effectors, are formed. It is possible to mobilize cells.

[0173] In a further embodiment, after the Ig-like molecule binds to the target molecule, the immunoeffect The ter cells are activated. The induction of the effector mechanism can be achieved, for example, by the method according to the present invention. This includes the re-modification of immunomodulatory cytotoxicity by the Ig-like molecules produced. g-like molecules bind to molecules that cause cytotoxicity, such as T cell receptors or Fcγ receptors. This allows for the activation of downstream immune effector pathways.

[0174] In this book, the terms "immune effector cells" or "effector cells" refer to active The natural process of the mammalian immune system, which affects the viability of target cells by transforming them. This refers to the repertoire of cell populations. Immune effector cells include natural killer (NK) cells. Lymphocyte-like cells, including cytotoxic T cells, or B cells only. Instead, myeloid cells such as monocytes, macrophages, dendritic cells, and neutrophils Cells are also considered immune effector cells. Therefore, the effector cells are preferably NK cells, T cells, B cells, monocytes, macrophages, dendritic cells, or neutrophils ru.

[0175] Target antigens present on immune effector cells include CD3, CD16, CD25, and CD28. This includes CD64, CD89, NKG2D and NKp46. Further information is provided. , Method for producing at least two different Ig-like molecules or heterodimer Ig according to the present invention A method for producing a molecule, wherein the target epitope is CD3, CD16, CD25, CD2 It is located in the 8,CD64,CD89,NKG2D, or NKp46 molecule. Target cell survival These abilities include the ability of a cell to survive, proliferate, and / or interact with other cells.

[0176] One aspect of the present invention provides a method for producing heterodimer Ig-like molecules according to the present invention, Each polypeptide chain containing a CH3 domain has a variable region that recognizes the target epitope. It further comprises a region. In one embodiment, the polypeptide chain comprising the CH3 domain Each of the two variable regions recognizes the same target epitope with different affinities. In the application form, each of the two variable regions of the polypeptide chain containing the CH3 domain It recognizes different target epitopes.

[0177] In other embodiments, different target epitopes are located on the same target molecule, and the target molecule The child may be a membrane-bound molecule or a water-soluble molecule. In other embodiments, a different target Epitopes are located on different target molecules, and these target molecules can be located on the same or different cells. It is expressed. Alternatively, the different target molecules are water-soluble molecules, or one target molecule is The second target molecule among water-soluble molecules could be a membrane-binding molecule.

[0178] In one preferred embodiment, at least one target molecule of the heterodimer Ig-like molecule is It is located in tumor cells. In yet another preferred embodiment, a heterodimer Ig-like molecule At least one target molecule is an effector cell (i.e., NK cells, T cells, B cells). Cells, monocytes, macrophages, dendritic cells or neutrophils, and the target epitopes are CD3, CD16, CD25, CD28, CD64, CD89, NKG2D or NKp It is located at the 46th molecule.

[0179] Production of at least two different Ig-like molecules according to the present invention, provided by a preferred embodiment. In the method or method for producing heterodimer Ig-like molecules, as described above, at least two of the above The different Ig-like molecules are antibodies, most preferably IgG isotype antibodies, More preferably, it is an IgG1 isotype antibody.

[0180] The method according to the present invention yields an Ig-like molecule, a heterodimer Ig-like molecule, or at least A mixture of two Ig-like molecules is also provided. The (heterodimer) Ig-like molecule The mixture of Ig-like molecules preferably contains at least one CH3 mutation listed in Table B. The present invention provides at least one Ig-like molecule, or at least two Ig-like molecules. Not only pharmaceutical compositions containing this mixture, but also (heterodimer)Ig-like molecules or at least In a mixture of two Ig-like molecules, having at least one mutation described in Table B It will also be offered.

[0181] In one embodiment, the Ig-like molecule is a bispecific Ig-like molecule such as a bispecific antibody. It is a molecule. In other embodiments, the Ig-like molecule is a single-specific antibody such as a monospecific antibody. It is a specific Ig-like molecule. A preferred embodiment provides a small amount obtained by the method according to the present invention. In a mixture of at least two different Ig-like molecules, the at least two different Ig-like molecules This refers to different epitopes in the same antigen and / or different epitopes in different antigens. Connect to the pu.

[0182] Furthermore, the heterodimer Ig-like molecule obtained by the method according to the present invention is provided, The heterodimer Ig-like molecule has different epitopes and / or different properties on the same antigen. It binds to different epitopes on the antigen. Advantages and suitable uses of this mixture and antibody. The law is as stated above.

[0183] The present invention provides at least two different Ig-like components obtained by the method according to the present invention. In this mixture, the at least two different Ig-like molecules are at least one heterodimer. It comprises a γ-like molecule. In one embodiment, two of the at least two different γ-like molecules The molecule is a heterodimer Ig-like molecule.

[0184] Another preferred embodiment provides a heterodimer antibody with two CH3 domes. It comprises the in, and one of the two CH3 domains is an amino acid substitution L351D and L368 E is present, and one of the other two CH3 domains is an amino acid substitution T366K and L35. It has 1K. These amino acid substitutions are, as described above, the two CH3 domains This is a suitable method for selective pairing.

[0185] The amino acid substitutions L351D and L368E in one of the two CH3 domains mentioned above The other amino acid substitutions T366K and L351K of the two CH3 domains mentioned above are, Also, "DEKK combination mutation", "DEKK variant", "DEKK pair", "D Refer to the "EKK-operated CH3 domain," "DEKK," or simply "DEKK." Other names are used. CH3 domes that retain amino acid substitutions L351D and L368E. The "DE side" is also called the "DE side," and the CH3 holds the amino acid substitutions T366K and L351K. The main group is also referred to as the "KK side."

[0186] (Heterodimer)Ig-like molecules, or less, obtained by any method according to the present invention. A pharmaceutical composition comprising a mixture of two Ig-like molecules is also provided. According to the present invention, the ( A telodimer) Ig-like molecule, or the at least two Ig-like molecules, preferably an antibody. Yes. The pharmaceutical composition contains the (heterodimer) Ig-like molecule, monospecificity or bispecificity. A mixture containing sexual Ig-like molecules or a combination of single-specific and bispecific Ig-like molecules include.

[0187] Furthermore, the pharmaceutical composition according to the present invention comprises a pharmaceutically acceptable carrier. Such "pharmaceutically acceptable carriers" are any and all solvents, salts, and dispersion media. Coatings, antimicrobial and antifungal agents, isotonic agents and absorption-delaying agents, and These include physiologically compatible substances similar to these. The route of administration (e.g., intravenous, dermal) (Below, within the joint, etc.) Ig-like molecules are affected by the action of acids that can inactivate Ig-like molecules and It can be coated with a substance to protect Ig-like molecules from other natural conditions.

[0188] In one embodiment, at least two obtained by any method according to the present invention In a pharmaceutical composition comprising a mixture of Ig-like molecules, the at least two different Ig-like molecules are It is produced by recombinant host cells according to the present invention. Furthermore, in the provided pharmaceutical composition, The present invention comprises a heterodimer Ig-like molecule obtained by any method according to the present invention, and the heterodimer Somatic Ig-like molecules are produced by recombinant host cells according to the present invention.

[0189] Polypeptide having a CH3 domain with at least one mutation listed in Table B CH3 has a nucleic acid molecule that codes for the chain, in addition to at least one mutation listed in Table B. Recombinant comprising at least one nucleic acid molecule encoding a polypeptide chain having a domain Host cells are also provided.

[0190] The present invention is further illustrated by the following embodiments. These embodiments do not limit the present invention. Rather, it is shown simply to clarify the present invention. [Brief explanation of the drawing]

[0191] [Figure 1] Figure 1A) is a schematic diagram of the construct vector MV1057. The stuffer region is the area where the VH region of the antibody is cloned. Figure 1B) is a schematic diagram of the phage display vector MV1043. [Figure 2] This is the amino acid sequence of the Fc of wild-type IgG1 present in the construct vector MV1057 (using the EU numbering system). [Figure 3] This is the nucleotide and amino acid sequence of the VH region used for cloning into various constructs. [Figure 4A] This is the mass spectrum data for transfection A. [Figure 4B] This is the mass spectrum data for transfection G. [Figure 4C]This is the data for the mass spectrum of transfection H. [Figure 5A] This is the mass spectrum data of transfection M. [Figure 5B] This is the mass spectrum data of transfection U. [Figure 6] This is the data for the mass spectrum of transfection O. [Figure 7A] This involves inhibiting homodimerization by substituting neutral amino acids with charged amino acids. [Figure 7B] This involves inhibiting homodimerization by substituting neutral amino acids with charged amino acids. [Figure 8] Figure 8(A) shows the native mass spectrometry (Native MS) spectrum of the transfection sample ZO(T366K / L351'D). Figure 8(B) shows the convolutional mass spectrometry spectrum of the transfection sample ZO(T366K / L351'D). The second / major peak is that of the bispecific molecule. [Figure 9] This is the HADDOCK score for experimentally validated pairs of mutations. [Figure 10] These are illustrations of the interaction at the CH3-CH3 contact surface. Figure 10A) is for K409D:K392D / D399'K:E356'K, Figure 10B) is for D399K:E356K / D399'K:E356'K, and Figure 10C) is for K409D:K392D / K409'D:K392'D. [Figure 11] These are the HADDOCK scores for various 366 / 351' charge mutants. [Figure 12] These are illustrations of the interaction at the CH3-CH3 contact surface. Figure 12A) shows L351D / L351'D, and Figure 12B) shows L351D:S354A:R355D / L351'D:S354'A:R355'D. [Figure 13] This is the HADDOCK score for additional charge variation near position L351. [Figure 14]This is the HADDOCK score for additional charge variations near position T366 of chain A and position L351 of chain B. [Figure 15] This is a diagram illustrating the interaction at the CH3-CH3 contact surface. [Figure 16] This is the HADDOCK score for mutants near T366 / L351. [Figure 17] This is the HADDOCK score for additional mutants near T366 / L351. [Figure 18] These are examples of nMS spectra of bispecific IgG obtained after co-expression of construct T366K,L351K with construct L351D (left figure) or L351D,Y349E (right figure). The spectra are magnified based on the full length of monovalent IgG (half-halves are not shown). [Figure 19A] This is the result of mass spectrometry showing the relative amounts of AA, AB, BB, A, and B (the sum of all types is 100%). [Figure 19B] The data is the same as in Figure 19A, but by excluding A and B, the appearance of the undesirable AA, BB, A, and B is made clearer. [Figure 20] These are the results of the thermal stability assay. Squares: wild type, triangles: charge inversion pairs E356K:D399K / K392D:K409D, circles: mutant CH3 combinations shown above each graph. [Figure 21] These are the results of 10 freeze-thaw experiments. 1122 = antibody BB from the first parent; 1337 = antibody AA from the second parent; wild type = AA, AB, BB; CR = bispecific antibody of charge-reversal pair E356K:D399K / K392D:K409D; 3-6 and 9-12 = bispecific molecule combinations of 3-6 and 9-12 in Table 15. [Figure 22]The stability of serum was measured by ELISA using fibrinogen-coated antigens. Figure 22A) shows the ELISA data when the IgG sample was diluted to 0.5 μg / ml. Figure 22B) shows the ELISA data when the IgG sample was diluted to 0.05 μg / ml. The results were normalized with T=0 (day) as 100%. 1337 = Second parent antibody AA; wild type = AA, AB, BB; CR = Charge inversion pair bispecific antibodies E356K:D399K / K392D:K409D; 3-6 and 9-12 = Bispecific molecule combinations of 3-6 and 9-12 in Table 15. [Figure 23A] These are the nMS results from experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23A shows the DEKK combination mutation, where "A" specificity is on the DE side and "B" specificity is on the KK side. [Figure 23B] These are the nMS results from experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23B shows the DEKK combination mutation, where "C" specificity is on the DE side and "B" specificity is on the KK side. [Figure 23C] These are the nMS results from experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23C shows the charge inversion mutation combinations, where the 'A' specificity is on the E356K:D399K side and the 'B' specificity is on the K392D:K409D side. [Figure 24] Table 20 shows the nMS results for transfections #1-11. [Figure 25] HADDOCK scores of dimers of different CH3-modified vectors. Gray bars: Desired species AB and CD; Black bars: Undesirable species AA, BB, CC, DD, AC, BC, AD, BD. [Figure 26] Table 20 shows SDS-PAGE of transfections #1-11. Control samples DE / KK, DE / DE, and KK / KK are included. [Figure 27A] This is nMS from transfection #9. [Figure 27B] This is nMS from transfection #11. [Figure 28A] This is the nMS of the gel-filtered sample 1516:1516. [Figure 28B] This is the nMS of gel-filtered sample 1337:1337. [Figure 28C] This is the nMS of gel-filtered samples 1516:1337. [Figure 29] This shows the blood concentrations of the DEKK-modified antibody and its two parent antibodies (pK study). [Modes for carrying out the invention]

[0192] (Example 1: Amino acid substitutions that generate various different CH3 domains) Selectively promotes or inhibits the pairing of Ig-like molecules possessing a CH3 domain. To obtain a wide variety of Ig-like molecules with different CH3 domains, we promote the formation of heterodimers. Many amino acid substitutions are known to work, as well as those that have not been reported or tried before. However, many other amino acid substitutions were chosen to promote homodimer formation, It was introduced into a construct vector (construct vector MV1057; Figure 1A).

[0193] The construct vector MV1057 is the F1 of normal wild-type IgG1 as shown in Figure 2. It has a nucleic acid sequence that codes for the c portion. Table 1 shows the amino acid substitutions introduced into this wild-type Fc. This is a list, and it will be a series of seven constructs. All constructs are Geneart Created by [artist name]. Constructs 1, 2, and 3, or their alternatives, have been made to date. There have been reports that this promotes the dimerization of terrorism (European Patent Application Publication No. 01870459, International Publication) (Issue No. 2009 / 089004). There are also reports on Constructs 6 and 7 (International (Published No. 98 / 50431). Constructs 4 and 5 are novel and promote homodimerization. It is designed to do so.

[0194] [Table 1]

[0195] (Example 2: Cloning VH into a construct with a CH3 mutation) Cloning to these constructs involves known specificities and known human IG. Several antibody VH regions capable of binding to the KV1-39 light chain were used. As mentioned above, all CH3 variants, when used in conjunction with other antibody domains, This can be a bispecific or monospecific antibody full length, determined by the VH / VL combination. The specificity of the antibody being tested does not affect the behavior of heavy chain dimerization induced by the CH3 domain. No. The model VH / VL combination, that is, all light chains are human germline IG Based on KV1-39, various combinations of VH will be used throughout this study.

[0196] Figure 3 shows the complete sequence and specificity of the antibody VH region used throughout this study. MF code This is an internal designation for various VHs used by Merus; for example, VH MF1337 is tetanus toxois. Furthermore, MF1025 is specific to porcine thyroglobulin, and MF1122 is specific to bovine fibrinogen. .

[0197] The VH region in the phage display vector MV1043 (Figure 1B) is related to the restriction enzyme SfiI. BstEII (New England Biolabs / cat# R0123L and R0162L / ), follow manufacturer's instructions. It is cut by ) and the VH fragment is extracted from this vector. Standard method ( (Following the manufacturer's instructions) Vector MV1057 is cut by SfiI and BstEII. .

[0198] Fragments and vectors are in gel (Promega / cat# V3125 / follow manufacturer's instructions) The purified and cleaved vector and VH gene insertion are isolated. Both are ly The nucleic acids are bound by ligation, and then the linked nucleic acids are E. coli DH5α (Invitrogen / cat# Transformed into 12297-016) (follow manufacturer's instructions). The next day, a single colony is selected. The vectors that have been extracted and undergone proper insertion are sequenced to identify them.

[0199] (Example 3: Transfection and full-length IgG expression in HEK293T cells) Re-cloned VH variants, and various others encoding the common light chain human IGKV1-39. Transfection of the plasmid into HEK293T cells allows for IgG expression. The process is carried out using standard procedures such as those described in de Kruif et al Biotech Bioeng. 2010. After the test, the IgG expression level in the supernatant is measured using the ForteBIO Octet-QK system. The system is based on Bio-Layer Interferometry (BLI), This enables real-time quantification and dynamic characterization of biomolecular interactions. For details, see below. 1418274598331_0.com See reference. If an expression level exceeding 5 μg / ml is measured, IgG should be purified for protein A affinity. It is purified using [a specific method / tool].

[0200] (Example 4: Purification of IgG) Use a Protein A column (GE Healthcare / cat# 11-0034-95 / follow manufacturer's instructions). After purifying the culture supernatant, it is eluted in 0.1 M citrate buffer at pH 3.0 and immediately... Neutralize with the same volume of 1.0 M Tris-HCl at pH 8.0, or directly desalt using a desalting column. Re-buffer in PBS using this method. Alternatively, use Protein A beads (Sepharose beads) IgG may be purified using CL-4B (GE Healthcare cat#170780-01).

[0201] (Example 5: Antigen-specific ELISA) Antigen-specific ELISA is performed to evaluate the binding activity to the antigen. Antigen capture ELISA is performed to demonstrate the binding activity of bispecific antibodies. The second antigen is used to detect the complex (de Kruif et al Biotech Bioeng. 2010).

[0202] (Example 6: SDS-PAGE) The purified IgG mixture was subjected to standardized nucleotide polymorphism (SD) under reducing and non-reducing conditions. S-PAGE (NuPAGE (registered trademark) 4-12% bis-tris gel / Invitrogen / cat# NP0323BOX) The analysis was performed using a colloidal blue reagent (PageBlue(trademark) prot). Stained with ein staining solution (Fermentas / cat# RO571).

[0203] (Example 7: Deglycosylation of IgG1 by enzyme) Since IgG glycosylation is heterogeneous, to make it suitable for use in mass spectrometry analysis, Deglycosylation was performed so that each product would have a distinct mass. One unit of N-glycosidase F (PNGase F; Roche Diagnosis) is needed for 0 μg of IgG1. The original purified buffer (nostics, Mannheim, Germany) was administered and incubated overnight at 37°C. To remove 0.1M citrate buffer pH 3.0 / 1.0M Tris-HCl pH 8.0, 1 Buffer exchange is performed using a 0kDa·MWCO centrifugal filter column (Millipore). We then re-buffered it in PBS. To remove the separated glycan chains, we used a similar buffer. The replacement procedure was carried out, and the solution was replaced with 150 mM ammonium acetate at pH 7.5. The tester ran 200 μl of 150 mM ammonium acetate at pH 7.5 for 12 minutes at 11000 rpm. It was washed away under conditions of 4°C. After washing, 50 μl of deglycosylated IgG was It was administered to a filter, and 450 μl of 150 mM ammonium acetate at pH 7.5 was added. Then, centrifugation was performed again for 12 minutes at 11,000 rpm and 4°C. Add a new 150 mM pH 7.5 ammonium acetate buffer to a total volume of 500 μl. Adding this, the centrifugation process was repeated a total of five times. After the final centrifugation step, the buff The remaining deglycosylated IgG1, approximately 25 μl, which had been replaced, was recovered. It was transferred to an Eppendorf tube and prepared for mass spectral analysis.

[0204] (Example 8: Native Mass Spectrometry) Identify the different IgG species in the purified IgG mixture and determine the ratio of these IgGs. Mass spectrometry is used to determine whether the gG species is present. In short, 1 2-3 μl of 150 mM ammonium acetate at pH 7.5 containing μM concentration IgG, manufactured in our company Gold-plated borosilicate capillary tube (Sutter P-97 puller [Sutter Instruments Co., No. [VATO, CA, USA] and Edwards Scancoat six sputtering machine [Edwards Laboratories, M Using [ilpitas, CA, USA], it is introduced into a system and is ideal for high-mass detection (Tahallah et al., RCM 2001). The LCT1 mass spectrometer (Waters Corp., Milford, MA, USA), which has been optimized for this purpose, was used to solve the problem. The analysis was performed using a capillary voltage of 1300V and a sampling cone voltage of 200V. However, these settings require a higher resolution signal-to-noise ratio. Adjusted as needed. Source backing pressure was adjusted to promote impact cooling. The ssure was increased to approximately 7.5 mbar. In order to measure denatured IgG1, The protein was sprayed in 5% formic acid at a concentration of 1 μM.

[0205] (Example 9: Data Processing and Quantification) The acquired spectra were processed using MassLynx 4.1 software (Waters C (ORP., Milford, MA, USA). Minimal smoothing is used, and the spectrum is centered. The mass of each type was calculated using a series of charge states. For each state, the corresponding intensity was assigned and added by MassLynx. This approach This allows for relative quantification for all species of the sample. Or, known, the surface under the curve. Peak quantification can also be performed using the area-under-the-curve (AUC) method. All analyses were repeated three times to calculate the mass of IgG and the standard deviation of its relative amount. It was done.

[0206] (Example 10: A mixture of two or three monospecific antibodies from a single cell) Several antibodies possessing known specificity and known human IGKV1-39 light chain binding ability. The VH region (Figure 3) is either the wild-type construct vector MV1057 or the one in Table 1. Used to re-clone into Instruct 4 or Construct 5, resulting in... We obtained ctor I-III (Table 2). As a result, we obtained different CH3 regions and different VH A vector containing a specific Ig heavy chain and a nucleic acid sequence encoding a common human light chain. —I, II, and III were then transfected into cells. Infection is performed with only one antibody to observe the formation of an intact, monospecific antibody, or Alternatively, in order to obtain a mixture of two or three monospecific antibodies, one or two other cons It was performed in combination with a tract vector. Table 3 shows a list of transfections and their results. be.

[0207] [Table 2]

[0208] [Table 3]

[0209] Transfections A, G, and H result in the transfection of vector I, II, or III Cells transfected with either of these only form homodimers, resulting in a bivalent, single-specific compound. AA, BB, or CC are obtained (Figure 4). For transfection A, This point was predicted and had been demonstrated before. This time, for the first time, the Mie net Construct 4 with no acid substitution (i.e., K392D, D399K, K409D) or quadruple amino acids CH3 containing substitution construct 5 (i.e., E356K, E357K, K439D, K370D) is operated Homodimerization of the transfected Ig heavy chain has been reported (transfection G and H).

[0210] Next, co-expression experiments of the two vectors were performed in single cells. What is interesting is that, as shown by transfections M and N, wild type and C When H3-modified Ig heavy chains are co-expressed with a light chain common to a single cell, the desired result is obtained. There is no bispecific antibody present, and only about 4-5% of "other molecules" are mixed into the mixture. Simply by being present, a mixture of two monospecific antibodies was obtained. "The other molecule" is intact. Defined as all molecules that do not possess the mass of IgG, and which consist of a single heavy-chain and light-chain pair or This includes the other half. The important point is that the "other" category does not include bispecific products. That is the case.

[0211] In transfection M, the vector DNA is transfected in equal proportions. The ratio of AA to BB is almost 1:1. However, in transfection N, A The A:CC ratio is approximately 10:1. Therefore, this transfection is DN The ratio of A was adjusted and the process was repeated (transfection U). In fact, vector DNAI: When III was in a 1:5 ratio, the ratio of antibody products AA:CC in the mixture was approximately 1:1. Therefore, as shown by transfections M and U, undesirable byproducts In the absence of material (i.e., the absence of large amounts of AC or half-cell A or C), in a single cell This allows for the expression of two distinct, virtually unadulterated, monospecific antibodies (Figure 5). ). Heterodimer formation between wild-type and construct 4, or wild-type and construct 5. To prevent this from happening, the new CH3 modifications in Constructs 4 and 5 are wild-type CH3 and They are substantially different. This point applies to the mass production of mixtures of single-specific antibodies from single cells. It is advantageous for them.

[0212] Similar to these results, but two different CH3-modified Ig heavy chains (constructs) 4 and 5) There is no presence of further undesirable species and two different monospecific antibodies A mixture of these is expected to be obtained as a result. Modification of CH3 in Construct 4 If the modification of CH3 in construct 5 is substantially different, then heterodimerization will not occur. It is inferred that in that case, the CH3-modified heavy chains of constructs 4 and 5 are used in single cells. When co-expressed with wild-type CH3 heavy chain, only three monospecific antibodies are obtained as a result. It will probably happen.

[0213] In fact, this result was observed. The design was intended to form homodimers rather than heterodimers. When three different Ig heavy chains are expressed together with a common light chain in a single cell, the mixture It was found that a mixture of three pure, monospecific antibodies could be obtained without any other contaminants (Tora). Transfection O) (Figure 6). As can be clearly seen from Table 3, transfection O Even if the vector DNA is made to be in the same proportion, the antibody ratio AA:BB:CC is 1:1 It was not obtained with a :1 ratio. When transfection was performed with a different vector DNA ratio, 1:1:10, transfection V), the ratio of AA:BB:CC in the mixture is desirable. It was shown that the ratio can be manipulated. From the above, as shown in these experiments, In a single cell, two or three substantially pure, monospecific antibodies are produced without undesirable byproducts. This can be done. This offers significant advantages for the mass production of mixtures of monospecific antibodies for therapeutic use.

[0214] (Example 11: A mixture of two bispecific antibodies from a single cell) The production of single bispecific antibodies using CH3-modified heavy chains has been reported elsewhere. Here, this experiment is performed to produce a mixture of two different bispecific antibodies from a single cell. It was designed to investigate whether it was possible. VH region of antibodies with known specificity and known binding ability to human IGKV1-39 light chain The region (Figure 3) is re-cloned into a vector containing constructs 1-3 or 6-7 from Table 1. As a result, vector IV-X (Table 4) was obtained. Different CH3 regions and different VH specifics Vector I contains a sex-positive Ig heavy chain and a nucleic acid sequence encoding a common human light chain, respectively. VX was then transfected into cells. This transfection was performed by To demonstrate interference with the formation of contactless monospecific antibodies, one antibody may be used, or two antibodies may be used. To obtain a heterogeneous antibody or a mixture of two bispecific antibodies, other construct vectors It was performed in combination with a ter. Table 5 shows a list of transfections and their results.

[0215] [Table 4]

[0216] [Table 5]

[0217] The manipulated Ig heavy chains of CH3 encoded in constructs 1 and 2 are used in single cells. When expressed, it has been shown to maintain the ability to form homodimers (International Publication No. 20 (Issue 09 / 089004). However, further reports in International Publication No. 2009 / 089004 Then, it was manipulated to have a triple-charged pair mutation like construct 3. In the CH3 domain, when expressed alone, homodimer formation is no longer possible. In research, these findings were only partially confirmed. In reality, transfects In stages B, C, and D, in addition to a high proportion of unpaired halves, the presence of full-length IgG cells is observed. This was shown. This indicates that the homozygous CH3 domain encoded in constructs 1 and 2 This shows dimerization. Transfections E and F also show, in addition to the unpaired half-pairs. As a result, the production of the full length of IgG was observed. This indicates the triple charge of construct 3. It was shown that even with this mutation, homodimerization is not completely impaired. Construct 6 It was further shown that the "knob" and "hole" CH3 mutants of 7 also form homodimers. (18% were homodimers of "nobu-nobu", and 42% were homodimers of "hole-hole") %).

[0218] When a second CH3 mutant is co-expressed for heterodimer formation, undesirable byproducts ( To inhibit or minimize homodimerization, when expressed alone, homodimerization is completely inhibited. A CH3 mutant that is harmed is preferred.

[0219] Interestingly, this experiment is the first to show that the mixture is virtually free of homodimers and exhibits bispecificity. We demonstrated that a mixture of antibodies can be expressed in a single cell. Transfection K And in L, the predicted bispecific species BC+AB is actually obtained (transformed Transfection K: 38% + 47%, Transfection L: 16% + 60%. A relatively high proportion of undesirable half-pigments were observed in transfection (transfection Transfection K produces 15% half-A+ and half-C, while transfection L produces 24% half-A+. Half-halves (C). Half-halves, which still exist at a relatively high rate, are those in which the heavy chain of the compatible pair is unbalanced. This expression is due to the fact that there is only a small amount of the heavy chain of vector IV. Therefore In transfections S and T, the vector DNA ratio was adjusted to 2:1:1. The cells were then transfected again. As a result, equal amounts of IgG heavy chains consisting of compatible pairs were obtained. Furthermore, there was no presence of IgG half-halves, and only a small 3% homodimer BB was present, indicating a pure bidimer. A mixture of heavily specific IgG was obtained. Ideally, the impurities of this low single-specificity product would be... The proportion should be reduced to virtually zero. Therefore, the presence of contaminating single-specific antibodies should be minimized. It is desirable to find further CH3 variants that can produce a mixture of bispecific antibodies that reduce the size of the CH3. ru.

[0220] This study is the first to minimize the presence of single-specific antibodies in a mixture while simultaneously studying single cells. A virtually pure mixture of two bispecific antibodies that recognize three different target epitopes. It demonstrated that goods can be produced.

[0221] (Example 12: Various mixtures) Production of a mixture of two bispecific antibodies that recognize three epitopes from a single cell, The production of a mixture of two or three monospecific antibodies from a single cell is technically feasible. This was shown. Next, we can demonstrate the feasibility of controlled production of various other mixtures. We explored the possibilities.

[0222] A fourth antibody, VH, possessing known specificity and binding ability to known human IGKV1-39 light chains. The region is used to reclone into a vector containing constructs 1-3 or 7 in Table 1. As a result, vectors I', II', III', or X' (where "'" is the corresponding vector) Compared to the ter number, different specificities were obtained. Different CH3 regions and different V A vector containing an Ig heavy chain with H specificity and a nucleic acid sequence encoding a common human light chain. The ter I'-III',X' and IV-IX were then transfected into the cells. The transfection involved various bispecific and / or monospecific antibodies. To obtain the mixture, it was done by combining it with other construct vectors. Various mixtures have two dual specificities that recognize four epitopes obtained from single cells. A mixture of antibodies, two bispecific antibodies and one single-specific antibody, or one bispecific antibody It contains a mixture of heterogeneous and single-specific antibodies. Table 6 shows one of the transfections. This is a list of the predicted results.

[0223] [Table 6]

[0224] Theoretically, the production of all mixtures is feasible, but other studies to date have shown that The mass production of knob-into-hole mutants has been hindered by stability issues. It is known that transfections ZA, ZB, ZL, ZM and ZN The resulting mixture is expected to present problems when it is moved to mass production.

[0225] In other words, with the set of constructs in Table 1, all can be mass-produced from single cells. It may not be possible to produce the theoretical mixture. The reason is, knob into • Hole variants have been reported to be unstable and have C with either a "knob" or a "hole". The H3 domain dimerizes with both charged mutants, as well as with the wild-type CH3 domain. This cannot be ruled out. Therefore, in order to co-express in a single cell, homodimers are needed. Alternatively, it selectively forms only heterodimers, and homodimers or heterodimers with constructs 1-5 in Table 1. It is desirable to design new CH3 mutants that are engineered to prevent ronimerization.

[0226] (Example 13: Identification of novel charge pair mutations) The purpose of this study is to determine what happens when different IgG heavy chains are expressed in a single cell. The CH3 region of IgG so that only heterodimers or only homodimers are produced as a result. This involves manipulating a domain. Here, the newly manipulated CH3 domain is a known manipulated domain. To prevent homodimerization or heterodimerization with the CH3 domain or the wild-type CH3 domain. Therefore, the most effective way to identify newly manipulated CH3 domains that meet this criterion is... As the first step, one by one, many contact residues on the contact surface of the CH3 domain of IgG i.e., as a group, repulsion between identical heavy chains due to electrostatic interactions - i.e., formation of homodimers. The purpose was to investigate substitutions that would cause a decrease in the IgG heavy chain. When expressed in a mixture, it is used to induce the formation of homodimers and / or heterodimers. To obtain a list of residues that, when substituted with charged residues, cause repulsion between residues on the same chain, This process stabilizes the resulting full-length IgG, leading to its high production rate.

[0227] Further investigations have revealed that matching pairs of CH3 residues in one or more IgG heavy chain-CH3 regions are needed. By performing this operation, a bispecific antibody or a mixture of bispecific or single-specific antibodies is obtained. The identified mutations are used to purify the compound. Furthermore, multiple compounds encoding different heavy chains are used. All nucleic acid molecules have different and complementary CH3 mutations, according to newly identified charge mutations. The mutation pairs are combined with existing pairs and used for expression in cells. This allows for the creation of single specific mutations. A mixture of only heterogeneous antibodies or only bispecific antibodies, or a mixture of specified monospecific and A mixture of bispecific antibodies is selectively obtained. The residues tested in this study are the same as those previously tested. These are defined contact residues (Deisenhofer J., 1981; Miller S., 1990; Padlan, 1996; Gun (asekaran, 2010). The rationale for this approach is that repulsive charges in each effective pair Its purpose is to be introduced into contact residues.

[0228] The sample was then analyzed by SDS-PAGE under non-reducing conditions, revealing approximately 72kD bangs. By examining the dots, we identify pairs in which dimer formation is reduced. Therefore, we screened for single mutations or combinations of single mutations. The reason is that repulsive electrostatic interactions by a single mismatched pair are not sufficiently detectable by this method. This is because it is unclear whether there is enough to obtain a sufficient number of half-kids. These mutations combine It is also used in Waseda.

[0229] According to Table 7, Geneart performed amino acid substitutions on the construct vector MV1057. We introduced it. Furthermore, we transfected HEK293T cells according to standard procedures. Construct expression was achieved as a result. IgG expression levels were measured using Octet. If production fails twice, the mutation will be considered to inhibit expression, and the mutation will be treated as such. No further investigation was conducted on that matter.

[0230] [Table 7]

[0231] The supernatant containing ≥5 μg / ml of IgG was analyzed by SDS-PAGE, and protein A was used. It was purified. The protein was stained with a colloidal blue reagent. The homodimer was approximately 150 It could be identified as a kD band. Smaller bands of approximately 75 kD exist as half of the same frequency. The presence was indicated (see negative controls: K392D, K409D). (Blot) This is shown in Figure 7.

[0232] The results of SDS-PAGE were analyzed and scored, and are shown in Table 7, in the far right column. Q347, Includes S354, Y349, L351, K360, T366, T394, and V397 Many residues show promise and should be combined and tested further. The selection of this component resulted in a high score in inhibiting homodimer formation and in relation to other non-complementary charges. We considered both the availability of modifiable contact residues that would not pose a problem, for example, F4 Residues 05 and Y407 involve interactions with already charged residues, CH3-CH It is known that there are multiple interactions at the 3-contact surface. These interacting residues ( (See Table A) Among these, introducing multiple charge mutations can be problematic.

[0233] To test further combination mutations, a new construct was developed for vector MV105 Created in 7 (Table 8), with known specificity and known binding ability to human IGKV1-39 light chains. The antibody VH region containing these new constructs is re-engineered into vectors containing these new constructs (see Table 9). It was used for cloning. Table 10 shows a list of transfections and their results.

[0234] [Table 8]

[0235] [Table 9]

[0236] [Table 10]

[0237] A combination of CH3 mutants was expressed, and SDS-PAGE (data not shown) and Naybean The results were analyzed by tibacterial mass spectrometry (MS). The results are summarized in Table 10. ZO transfection This method yielded the highest proportion of heterodimers in the mixture (69% AC). Interestingly, in ZO transfection, AA homodimers do not exist, while C The C homodimer is present in a small proportion (7%). Mass spectrometry revealed that the remaining protein in the mixture... It consists of half-cell A, which is thought to be the result of an imbalance in the expression of A and C heavy chains. The raw MS data from the ZO sample is shown in Figure 8. Surprisingly, the transfer A considerable amount of bispecific products were obtained using cation ZO, but the inverted charge pair is Transfection ZP (ZO is T366K / L351'D compared to L351K / T366'D) yields similar results. No fruit was obtained, and only 52% of the bispecific products were observed, along with a considerable amount of two Homodimers were present (30% AA and 13% CC). In contrast to these, negatively charged... Since D is structurally very similar to T, T366D is strong enough to repel itself. Therefore, it was explained and indeed observed that T366D could still form homodimers.

[0238] A small number of newly discovered T366K / L351'D pair variants (e.g., new constructs) (By testing all deformations, including T366R and L351E) It is predicted that a bispecific antibody (BsAbs) of a certain proportion will be obtained as a result.

[0239] (Example 14: Novel C for inducing efficient heterodimerization by HADDOCK) (Design of H3 mutations) As described in Example 13, the newly discovered charge pair T366K / L351'D is used in the mixture. The proportion of heterodimers increases (69%), along with a small proportion of undesirable C A mixture of C homodimer (7%) (L351D / L351'D) and a considerable proportion of half-dimer A (24%). "Contamination" of an object. In this example, using an in silico approach, CH3 Further insights were gained into the amino acid residues involved in the interaction at the contact surface, and the CH3 region was identified. We tested complementary substitutions that inhibit the efficient formation of two heavy chain homodimers, Discover novel CH3 pairs containing complementary substitutions that further increase efficient heterodimerization. .

[0240] HADDOCK (High Ambiguity Driven Protein-Protein Docking): The induced protein-protein docking is used for modeling biomolecular complexes. HADDOCK is a flexible docking approach that utilizes information. Unlike the docking method, ambiguous interaction restraints Encodes information about the contact surfaces of proteins identified or predicted within AIRs, and docking Induce the process (de Vries et al., 2010). Input to the HADDOCK web server The data comes from protein structure files such as crystal structures, NMR structure clusters, or structural models. Yes. After docking or fine-tuning, HADDOCK will have a so-called HADDOCK score. Return. The HADDOCK score is van der Waals energy, electrostatic energy, cover It is a weighted average of surface area and desolvation energy. The HADDOCK score is the actual Direct conversion to experimental data is often difficult, but it can be used as an index of binding energy or affinity. In addition, HADDOCK, based on the results of docking calculations, determines the "top 4" configuration. The structural files of the structure are provided. These structural files can be downloaded and visualized. This allows for detailed analysis of the interactions between individual residues.

[0241] In this example, the interaction between CH3 domains of the IgG monochain is studied. High resolution. The structure is based on the Fc portion (structure 1L6X) of IgG, which has a crystalline structure of a certain degree. (http: / / www.rcsb.org / pdb / explore / explore.do?structureId=1l6x ; Idusogie, EE e t al., JI 2000(164)4178-4184).

[0242] In Example 13, co-expression of vectors XIII and XVI resulted in CC homozygous expression. It was found to lead to the formation of dimer contamination (Table 10). HADDO CK was used to search for further mutations that inhibit homodimerization, in addition to T366K / L351'D. ru.

[0243] The output of the HADDOCK score is calculated energy and the HADDOCK score ( (Weighted average of several energies), and the four minimum energies found by the program These are four structure files corresponding to the energy structure. The HADDOCK score has a different structure. It is used to compare. Other energies indicate what is happening in the structure. Standards (e.g., good electrostatic interaction, less covered surface, high van der Waals) It is used solely to obtain energy. The lower the HADDOCK score, the better. Good. For each mutation pair, scores for the AA, AB, and BB dimers are calculated.

[0244] The set of mutation pairs in Example 12 was run on HADDOCK, and experimental data and calculations were obtained. We investigated whether the energies were correlated. All theoretical energies are shown in Table 11. This is visualized in Figure 9.

[0245] [Table 11]

[0246] For the two wild-type CH3 domains, the CH3 regions of A and B are isomorphic. The HADDOCK score is the same for AA, AB, and BB. In many other cases... As expected, the AB pair had the lowest score. Regarding the T366K / L351D pair... In this case, the BB score was slightly better than the AB score (-210.6 vs. -212.5). However, The difference is within the range of calculation error. Using HADDOCK, the heterodimorphs of these pairs The structure of the body was visualized. For example, construct combinations 1-2, 1-1, and 2-2 are shown in Figure This is shown in 10. From these visualizations, it is clear that salt bridges are formed in the heterodimer. (Figure 10A, left panel), electrostatic repulsion occurs between residues of the same chain. (Figures 10B and C, center and right panel). Higher HADDOCK for homodimers. The score is explained by the electrostatic repulsion of mutated contact residues. These residues are curved relative to each other. Because it avoids interacting with residues from other chains, its affinity decreases.

[0247] Table 11 and Figure 9 confirm the observations in Example 13. AC heterodimers and CC homodimers of L351D / L351'D have similar energies, This explains the existence of both heterodimers and homodimers in the compound. On the other hand, The AA homodimer of T366K / T366'K was hardly detectable in the mixture, but the T366K half-dimer A was present. Table 11 and Figure 9 show that the HADDOCK scores for the AA homodimer T366K / T366'K are This indicates a higher score than that of the AC heterodimer. Therefore, the formation of this homodimer... This is more energetically undesirable.

[0248] (Example 15: 366 / 351 mutation) In Example 13, the T366K / L351'D mutation charge pair was mixed by a method other than the one described above. The hypothesis is that it is possible to design the system to obtain similar results regarding the proportion of bispecific antibodies in a substance. This can be set up. Another method is T366R, T366D, T366E, L351E, L3 This may include substitutions of 51K and L351R. The ratio of CC homodimers of L351D / L351'D is 3 This can be reduced by generating 66 / 351 pairs of mutants. All possible mutant pairs are HAD The tests were run using DOCK, and the resulting scores are shown in Table 12 and visualized in Figure 11.

[0249] [Table 12]

[0250] Looking at the HADDOCK score, some mutations compared to T366K / L351'D Similar "patterns" were observed. In many deformations, AA homodimer The body was found to have a higher HADDOCK score than the AB heterodimer, but B The B homodimer was preferred as much as the AB heterodimer. 351 residues were the same as those in the other strand. It is known that the residues are "adjacent" to each other. That is, the 351 residues of chain A are CH3-CH3 At the contact surface, it pairs with 351 residues of chain B. When a BB dimer is formed, the same charge There are almost no negative effects. Looking at the structure of L351D / L351'D, it explains that asparagine The acids bend and avoid each other, and at least the spontaneously occurring arginine at position 355 This has an effect on stabilization, and furthermore, the negative charge stabilization by spontaneously occurring serine at position 354. There is a chemical reaction (see Figure 12A). Even if these residues are mutated (S354A and R355D), there is almost no change. There is no improvement. From Figure 12B, it can be seen that the hydrogen in the main chain of A354 is responsible for the stabilization of the homodimer. It is clear that this is the case. From this series of events, it is evident that the T366R / L351'E pair is a bispecific molecule. It has the lowest HADDOCK score and is considered the most suitable.

[0251] (Example 16: Mutation around T366K / L351'D) In this embodiment, a series of HADDOCK analyses were performed using T366K / L351'D or T366K / L351 The structure was based on the pair of 'E'. The predicted values ​​of the proportion of the bispecificity of these A and B chains were used. To identify additional mutations that could further enhance the HADDOCK score, mutations were added to the B chain and the HADDOCK score was increased. The energy was calculated. The CH3 domain structure was used for visualization of the protein structure at the molecular level. When examined using a viewer (YASARA, www.yasara.org), the distance between individual residues The separation can be calculated. During this examination, two residues, Y349 and L368, are the phases of the dimer. It was observed that these are adjacent residues that can make a positive or negative contribution to the interaction. These mutations, in addition to the L351 mutation, result in homodimeric and heterodimeric dimers. The effect on the compound was examined (see Figure 13). Both residues increase the stability of the heterodimer ( (Lower HADDOCK score), destabilizes BB dimers (higher HADDOCK score) (Core). Glutamic acid (E) at positions 349 and 368 is preferred over aspartic acid (D). This was suggested. Therefore, the second amino acid in the B chain, which already has an amino acid substitution at position 351. The introduction of acid mutations suggested that heterodimerization would become preferred.

[0252] In the following HADDOCK analysis, we will again use the T366K / L351'D pair as the starting structure. . Substitutions in the B chain (i.e., Y349D / E and An additional mutation was added to the A chain (L368E), which already had the T366K substitution. This is shown in Figure 14. As such, there are several mutations that are considered favorable for the formation of bispecific heterodimers. In the T366K-L351K / L351'D-Y349'D pair, all four mutant residues are heterodimers. It is involved in the binding. This means that K351 is not directly involved in the binding of T366K-L351K / L3. This does not apply to 51'E-L368'E. However, the HADDOCK squaring of the latter heterodimer... A is -228.9, which is significantly lower than T366K / L351'E-L368'E's -214.2. This is explained by the interaction of hydrogen bonds at position 351 K (see Figure 15). In the T366K-L351K / L351'D-Y349'D pair, further improvement may be possible with the R355'D mutation in the B strand. This increases BB's HADDOCK score, but also increases AB's HADDOCK score. It increases slightly. Therefore, compared to a single mutation of T366K in the A chain, an additional L 351K lowers the AB score, but doesn't change the AA and BB scores much. Theoretically, This will result in obtaining a higher quantity of bispecific heterodimers in the material.

[0253] As is clear from Figure 11, making position 366 R instead of K induces heterodimerization. It is presumed to be effective in deriving it. Therefore, this time, in chain A, instead of T366K, use T366R. Several HADDOCK analyses shown in Figure 13 were repeated. R366 of the A chain and the double of the B chain Combining mutations has been shown to be undesirable (Figure 16). This is because of the sub-residue of this residue. Even if the size is large and all salt bridges with R366 exist in the structure, the interaction of other contact surfaces This is because it interferes with the other. Furthermore, the HADDOCK score of the AA homodimer is R366 rather than K366. The value is lower. This also contributes unfavorably to the formation of heterodimers. Therefore, at the contact surface Further HADDOCK analysis using R366 has been discontinued.

[0254] Based on predictions by HADDOCK, a total of 14 optimally performing pairs were selected. (See Table 13 and Figure 17). In some pairs, spontaneous L351 / L351'D interaction To eliminate the stabilization effects of R355, R355D substitution is included.

[0255] [Table 13]

[0256] (Example 17: In vitro analysis of bispecific molecules using CH3 mutations based on HADDOCK prediction) (Expression in ro) As suggested by the analysis in Example 16, the additional mutations around the T366K / L351'D pair Some CH3 mutants have a higher proportion of bispecific components and a lower homodimer. A mixture with a specific ratio of components can be produced. These optimal pairs are used for production. Selected and further analysis was performed. Furthermore, constructs T366R and L351E were also generated. The constructed construct, and known specificity and known human IGKV1-39 mild, were created. This table lists the constructs used to reclone the antibody VH region that has chain binding ability. As shown in 14.

[0257] The expression levels of IgG including each construct were reported in Example 13 above, but Table 1 The same was done for the constructs listed in 4. The purpose was to achieve heterodimerization. This is to evaluate which constructs homodimerize in the absence of a suitable partner. Ideally, a high proportion of half-dimers and a low proportion of homodimers should be formed. The group includes constructs containing charge mutations reported to date, and previously reported Constructs containing the modified knob-in-hole mutation also show overall IgG by recombinant cells. It was used for expression. The supernatant purified with protein A was analyzed by SDS-PAGE, and the results were obtained. The scores are shown in Table 14.

[0258] [Table 14]

[0259] Two different constructs that retain a common light chain and the amino acid substitutions shown in Table 14. Results of co-expression with heavy chains, or heavy chains that retain amino acid substitutions of previous constructs. Table 15 shows the two, each possessing the amino acid substitutions T366K and L351'D:L368'E. Due to the expression of different heavy chains, there is no homodimer AA or BB present, and the mixture contains approximately 87% We were able to obtain the bispecific heterodimer AB (combination number 3 in Table 15). Approximately 12 A half-chain (half-chain A) containing % T366K substitution was observed. Furthermore, an additional A was found in the first heavy chain. When mino acid-substituted L351K is introduced, the proportion of bispecific heterodimer AB increases. For example, those comprising amino acid substitutions T366K:L351K and L351'D:L368'E, respectively. Co-expression of two different heavy chains yields a bispecific heterodimer AB with approximately 92% specificity. On the other hand, AA and BB homodimers were substantially absent from the mixture (combinations in Table 15). (Number 12).

[0260] The combination of 10 and 11 also has a high proportion of heterodimers and virtually no homodimers. This results in a desirable distribution called a mer. The absence of homodimers is due to the intactic The fraction containing the IgG molecule is advantageous because it consists only of heterodimer AB. Purification, And for later therapeutic applications, half-halves are used in standard methods such as size exclusion chromatography. These can be removed by the following approaches: known charge variants and knob-into-hole variants. In the body, the antibodies that "contaminate" the homodimer are not removed. However, these Applying newly identified charge variants to the production process of bispecific antibodies would be advantageous.

[0261] In addition, the charge pairs T366K / L351'D:L368'E and T366K:L351K / L351'D:L368'E are E356K:D399K / K392'D:K409'D and E356K:D399K / K392'D:K409'D:K43 This has even more advantages than the 9'D charge inversion pair. In other words, it is more advantageous than the charge variants described so far. This is based on the charge reversal that is originally present within the CH3-CH3 contact surface, but newly identified electricity The variant adds an additional charge pair (charge-charge interaction) to the CH3-CH3 contact surface. The introduction of additional charge pairs at the CH3-CH3 contact surface improves the stability of the contact surface. This further increases, thereby increasing the stability of the intact antibody. Combination number 4 The same applies to the mutations used in 5, 6, 9, 10, and 11. The mixture contains only very small proportions of homodimers of AA and BB. This leads to a favorable ratio of bispecific heterodimers.

[0262] [Table 15]

[0263] (Native MS) Native MS was performed on all bispecific samples. The resulting graphs were analyzed. The relative ratio of the target species was derived using two methods (peak height and peak area). The method using the area of ​​the circle is a more scientifically correct analytical method, but in other studies to date... Since all analyses are performed on peak height, both methods are included in the analysis for comparison purposes. The difference between the two methods was within the range of measurement error, therefore, for subsequent measurements... Only the peak area value was used.

[0264] Figure 18 shows two typical spectra. A summary of the results is shown in a graph in Figure 19. The numerical values ​​can be found in Table 15. In about half of the samples, the total single-specific IgG The amount of contamination was less than 5%, and only in the case of three samples did it exceed 10%. On the other hand, with wild-type IgG... Approximately 50% of single-specific IgG is expected to be found in the mixture.

[0265] For further analysis, we selected 10 combinations of two different heavy chains from Table 15. These 10 combinations include combinations 1, 2, 3, 4, 5, 6, 9, 10, and 11. This includes 12 (Table 15). These 10 selections are derived from the mixture by nMS. This is not only based on the low proportion of homodimers present, but also on the production yield, SDS-PAGE, This was also based on overall physicochemical characteristics, including the number of mutations in the CH3 domain.

[0266] (Example 18: Analysis of IgG stability) This study found a high proportion of bispecific heterodimers in the intact IgG fraction, and very For a series of CH3 mutation pairs when a small amount (<5%) of parental IgG is obtained, Furthermore, the stability of the Fc portion of the IgG molecule will be analyzed. The mutated CH3 domain has an unexpected destabilizing effect on the Fc region of IgG. This can have an effect. This can lead to a decrease in the in vivo half-life, a decrease in effector function, and / or Alternatively, it could result in undesirable traits such as increased immunogenicity.

[0267] Regarding the newly identified charge pair, wild-type bispecific molecules and previously identified Compared with a bispecific molecule containing a modified charge mutation (chain A with construct 1 and Chain B, which has construct 2. All bispecific molecules in this study have the same heavy chain. Because it includes the variable region of the light chain, the observed effect is due to mutations in the Fc portion of the molecule, and the variable region It is guaranteed that this is not due to differences in region.

[0268] A series of stability studies will be conducted on these bispecific molecules. This provides information on the aggregation state of CH3 mutants, using spectroscopic (UV-) imaging. Vis light absorption, fluorescence, and light scattering) and microscopic (optical and fluorescent with Nile Red staining) This includes observation and analysis using a light microscope.

[0269] UV-Vis absorption spectroscopy was performed using two monochromators with a double beam and a Cary 300 Bio. The photophotometer records the readings at 25°C. The spectrum is measured at 250 to 400 ohms with a path length of 1 cm. It is monitored between m. The amount of absorption at 320 nm and longer wavelengths is used to determine the aggregation of IgG. Status information is provided.

[0270] The intrinsic fluorescence spectrum is monitored at 25°C using a FluoroMax fluorometer. The lighting method is optimized as needed. Fluorescence emission affects conformation and aggregation characteristics. We will provide information on this.

[0271] The 90° light scattering spectrum was obtained using a FluoroMax fluorometer, and the integration time was measured. ) with a 0.01 second interval for a synchronized scan (λ) from 400nm to 750nm. em =λ ex ) execute It is monitored at 25°C. The excitation and emission side slits are optimized as appropriate. For example, Right-angle light scattering can be used to distinguish whether or not 5% of the IgG sample contains dimers.

[0272] In fluorescence microscopy observation with Nile Red staining, immediately before measurement, Nile Red staining is performed using ethanol. Red is added to the sample. The sample is filled into a microscope slide and analyzed by a fluorescence microscope. Particles are counted, but the smallest particle size observed by fluorescence microscopy is approximately 0. It is 0.5 μm.

[0273] Stress on proteins such as temperature, pH, mechanical stress, or denaturants can cause problems. It causes changes in composition (e.g., unfolding) and / or aggregation. To date, charge-manipulated bispecific antibodies have been shown to have a lower melting temperature for modified CH3. This has been reported (Gunasekaran 2010). Therefore, these studies are novel electrical... The aim is to distinguish between charge mutants and charge mutants that are already known to exist.

[0274] Thermal stability of protein A biosensor and IgG using FcRn, measured by Octet. Research is being conducted. A PCR device is being used to investigate the thermal stability of IgG with manipulated CH3. The sample was subjected to a 100 μg / ml solution at 4, 50, 55, 60, 65, 70, and 75°C for 1 hour. Incubate the sample at the concentration of (PBS as solvent). After this, the sample is slowly cooled. The temperature was raised to 25°C in 15 minutes, maintained at this temperature for 2 hours, and then stored at 4°C until the next day. Sedimentation occurred. The antibodies were removed by centrifugation, and the total IgG concentration of water-soluble antibodies was determined by Octet Protein A The results were derived using Octet with Osensor (1 / 10 PBS dilution).

[0275] An assay was developed using Octet to measure the binding of CH3-modified IgG to FcRn. The process is investigated. After binding the IgG light chain to the sensor using a protein L biosensor... Incubate with FcRn in solution. Alternatively, use Anti-Penta-HI. Using an S biosensor, the His-tagged FcRn protein is bound, and the target IgG is... They can be incubated together. These methods are more sensitive than protein A biosensors. It can also be used in thermal stability studies.

[0276] All samples are also subjected to stability analysis under serum. In short, (manipulated IgG samples were incubated in human serum at 37°C, while control samples were incubated at 4°C. It is maintained in this state. After 1, 2, 3, and 4 weeks, the sample is centrifuged and the precipitated IgG is removed. The sample is then added using antigen-specific ELISA to determine the relative amount of functional IgG. The purified control antibody is rapidly administered to human serum and used as a reference group. It is possible.

[0277] (Example 19: Stability Analysis) Previous experiments showed that co-expression of two different heavy chains with CH3 mutations and a common light chain was successful. As a result, a high proportion of bispecific antibodies were obtained (Example 17).

[0278] Eight combinations of two different heavy chains were selected from Table 15 and further analysis was performed. These eight combinations are combinations 3, 4, 5, 6, 9, 10, 11 and 12 This includes (Table 15). This study focused on the stability of the Fc portion of IgG, and these 8 Two combinations were analyzed. The control group included wild-type bispecific molecules (i.e. (including the absence of CH3 mutations) and / or previously reported CH3 charge mutations. Note that wild-type bispecific molecules are selectively induced into heterodimers. Without any means, the two heavy chains and the common light chain were co-expressed. Therefore, these "wild" The term "bispecific molecule of type AA, AB, and BB" represents a mixture of AA, AB, and BB molecules. All of the bispecific molecules in this study are bispecific. Since specific molecules maintain the same heavy and light chain combination, the observed effect is due to the molecule. It is guaranteed that this is due to a mutation in the Fc region and not to a change in the Fab region. .

[0279] As a hypothesis, mutations used to facilitate heterodimer pairing of two different heavy chains The heterogeneous pair is associated with unexpected structural or other destabilizing efficacy of IgG to the Fc region. They can be linked. This later led to the conclusion that the presence of these mutations reduces the in vivo half-life. Further clinical developments such as decreased effector function and / or increased immunogenicity This could result in undesirable problems that could hinder development.

[0280] (thermal stability) When subjected to stress such as a rise or fall in temperature, the conformation of the protein changes. This causes deformation (e.g., unfolding) and / or aggregation of CH3. To investigate the thermal stability of IgG, combinations 3-6 and 9-12 (Table 15) were used, as well as wild-type IgG. Type bispecific molecules and constructs 1 and 2 (E356K:D399K / K392D':K409D' pair) Combination. A bispecific molecule obtained using a "charge reversal" pair is used with PC Using the R apparatus, at 4, 60, 62.5, 65, 67.5, 70, and 72.5°C for 1 hour. The sample was incubated at a concentration of 100 μg / ml (PBS as solvent). After this, the sample was slowly... It was thoroughly cooled, then raised to 25°C over 15 minutes, maintained at this temperature for 2 hours, and then left at 4°C until the next day. The antibodies were stored in [a specific location]. The precipitated antibodies were removed by centrifugation (18000 rpm; 4°C, 20 minutes). The total IgG concentration of water-soluble antibodies was measured by Octet using a protein A biosensor (1 / 1 The result was obtained using (0 PBS dilution).

[0281] The results are shown in Figure 20. The control CH3-modified bispecific antibody (charge inversion) The combination E356K:D399K / K392D':K409D' (triangle) is a wild-type bispecific molecule (square). Compared to ), thermal stability is reduced. From combinations 3-6 and 9-12 (rhombic) The bispecific molecule was also shown to have reduced thermal stability compared to the wild type. However, attention One thing to note is that in the three combinations, the control CH3 operated dual special Improved stability was demonstrated compared to heterozygous antibodies. Combinations 9, 10 and 1 The bispecific molecule 1 is significantly safer than other CH3-operated (charge-reversed) bispecific molecules. It was stable, and at the highest temperature measurement point, it was as stable as the wild-type bispecific molecule.

[0282] (Freeze-thaw stability) To investigate the stability of CH3-modified IgG after repeated freeze-thaw cycles, a combination was used. Bispecific molecules from se3-6 and 9-12 (Table 15), and wild-type bispecific molecules, Constructs 1 and 2 (E356K:D399K / K392D':K409D' combination (charge reversal pair) The bispecific molecule obtained using (a)) was subjected to 10 freeze-thaw cycles. In this process, the sample is left at -80°C for at least 15 minutes until completely frozen, and then thawed at room temperature. It was thawed. Once completely thawed, this freeze-thaw cycle was repeated. 10 freeze-thaw cycles. After the dissolution cycle, the precipitated antibody was centrifuged (18000 rpm; 4°C, 20 minutes). Remove the remaining antibodies and measure the total IgG concentration of the water-soluble antibodies using Octet, and then measure the protein A biosensor (1 The results were derived using a 10 / 10 PBS dilution. The freeze-thaw stability test was repeated three times.

[0283] The results are shown in Figure 21. The control was a charge-reversal CH3-modified bispecific antibody. The body was observed to have slightly reduced stability compared to the wild-type bispecific molecule. In contrast, the bispecific molecules from combinations 3, 4 and 9 were wild-type bispecific molecules. Compared to the child, a slight improvement in stability was observed. From the above, the strict freeze-thaw cycle Under these circumstances, no significant stability issues are raised for CH3-modified mutants. This was the conclusion.

[0284] (in vitro serum stability) To investigate the stability of CH3-modified IgG in serum maintained at 37°C, a combination of factors was used. Bispecific molecules from 3-6 and 9-12 (Table 15), and wild-type bispecific molecules. Furthermore, the charge-reversed bispecific molecules were incubated in 10% human serum at 37°C. The control sample was kept at 4°C. After 1, 2, or 5 days, the precipitated antibody was centrifuged. It was removed by separation. Subsequently, the sample was added using fibrinogen-specific ELISA and function The relative amount of IgG was derived. Purified control antibodies were rapidly administered to human serum. It was then used as a reference group.

[0285] According to fibrinogen ELISA data, in 10% human serum at 37°C for 5 days However, all samples were very stable. Lower Ig from combinations 4 and 5. The bispecific molecule at G concentration is slightly less stable, especially at T=1 and T=2. However, at the end of this experiment, the difference is minimized (Figure 22).

[0286] (Example 20: Further stability testing) A further series of analytical methods were used to evaluate the stability of the mutant IgG. Bispecific molecules from se3-6 and 9-12 (Table 15), wild-type bispecific molecules (A A, AB, BB), individual parental antibodies (AA and BB), and constructs 1 and 2 The dual feature obtained using the combination (E356K:D399K / K392D':K409D' (charge inversion pair)) Isomer molecules were used as samples in these stability assays.

[0287] All IgG was diluted to 0.2 mg / ml and tested under several stress conditions (at 50°C). The following conditions were met: 2 days at 40°C for 2 weeks, followed by 5 freeze-thaw cycles. This was done to distinguish between different samples. This is to allow for a different approach. It should be noted that these high stress levels are not all One of the parent antibodies used in the bispecific molecule (holding two 1122Fab molecules) The parent of BB becomes unstable. Under the conditions of 50°C for 2 days, this protein Aggregation was detected by UV absorption. As this suggests, under these stress conditions, It may not be possible to distinguish between the instability of Fab and CH3 in bispecific molecules. No. Therefore, data from incubation at 50°C must be handled with care. ru.

[0288] A summary of the results is shown in Table 16. The analytical methods used included the following: - Fluorescence microscopy using Nile Red (see "Nile Red particles" in Table 16); Nile Red color This is to observe the amount of particles larger than 0.5 μm after adding the element. UV spectroscopy at -350nm ("UV350nm"); at wavelengths longer than 320nm Changes in absorption provide information about the state of protein aggregation. 90° light scattering at -400nm ("LS400nm"); changes in protein aggregation, e.g., I This is a highly sensitive technique for observing the difference between monomers and dimers of gG. - Autofluorescence; the maximum fluorescence wavelength and intensity of aromatic residues in proteins vary depending on the environment (for example, (Unfolding) Change. -1,8-ANS fluorescence spectroscopy analysis; 1,8-ANS is ion-paired and cation-paired through electrostatic interactions. It binds to the ON group. Changes in protein structure and / or conformation are detected. .

[0289] (UV-Vis spectroscopy) UV-Vis absorption spectra are obtained from different quartz cuvettes by Varian (e.g., optical path length) (1.0cm black low-dose Hellma cuvette and 0.2cm x 1.0cm clear Hellma cuvette) Using a Cary 300 Bio spectrophotometer with a double beam and two monochromators, measured at 25°C. Measurements were taken. Using an optical path length of 1.0 cm, the spectrum between 220 and 450 nm was monitored. I did. The absorption around 280 nm provides information about the protein concentration. From 320 nm to 45 The region between 0 nm provides information about the aggregate state of the sample.

[0290] (90° light scattering) 90° light scattering spectroscopy was developed to study protein aggregation. This method is Ca The procedure was carried out in the same manner as described in pelle 2005 and Demeule 2007a. The 90° light scattering spectrum was obtained from Fluor Using an oMax fluorometer (Spex, Instruments SA, Inc. UK), the integrated time was calculated. A synchronized scan (λem=λex) from 400nm to 750nm with time set to 0.01 seconds. The process is run and monitored at 25°C. Different slit settings are tried to obtain optimal conditions. After optimization, all measurements were performed using that slit setting.

[0291] (Steady-state fluorescence emission) The fluorescence emission of tryptophan, tyrosine, and phenylalanine residues is fully This provides information about the local environment of the orophore. Hydrophilic and / or rigid, variable The difference or transformation is measured. Generally, environments with higher hydrophobicity and rigidity result in higher fluorescence intensity. It increases and leads to a blue shift in the value where the radiation is maximum. Autofluorescence spectroscopy analysis reveals the protein's Information on the state over time is provided, and changes in physical and chemical properties are monitored. (Tyrosine) More detailed information on tryptophan can be found in Lakowicz's book (Lakowicz, 2006). It is.

[0292] The fluorescence emission and excitation spectra were recorded in different quartz cuvettes at 25°C. The sample was excited at different wavelengths. The integration time and slit settings were optimized. After optimization, the accumulated time and slit settings were applied to all samples.

[0293] (Fluorescence microscopy using Nile Red staining) The Nile Red staining method was developed to visualize protein aggregation. This staining method is De The procedure was carried out as described in meule et al., 2007b.

[0294] Microscopic observation was performed using a Leica DM RXE microscope equipped with a mercury lamp (Leica Microsystems GmbH). The image was taken by Wetzlar, Germany. The image is of a Sony NEX-5 camera and its firmware. The images were acquired using a microscope. Objective lenses were set to 10x, 20x, and 40x. Microscopic research was conducted. So, we fixed the distance between the slide and the cover glass to 0.1 mm and used that. 4x4 grid The size of the sample is 1 mm x 1 mm, which corresponds to 0.1 μl.

[0295] (1,8-ANS fluorescence spectroscopy) 8-anilinonaphthalene-8-sulfonic acid: 1,8-ANS) is an uncharged, hydrophobic fluorescent small molecule (molecular weight 299.34 Da) that interacts with membrane surfaces and proteins. It is used to study both white matter and white matter.

[0296] 1,8-ANS does not fluoresce in water in any significant amount, and does not fluoresce in membranes (quantum yield ~0.25) or proteins (amount It emits detectable fluorescence only when it is bound to a molecule (0.7%). Due to this property, 1,8- ANS is involved in protein folding, conformational changes, and other related processes. The lobe's exposure to water serves as a highly sensitive indicator of the changing process. Information about 1,8-ANS can be found on the cular Plobes website at www.probes.com. ru.

[0297] The fluorescence emission spectrum of 1,8-ANS was recorded using a FluoroMax fluorometer. A direct comparison of the fluorescence of 1,8-ANS will not be performed. Each IgG has a different number of 1,8-ANS This is because they cannot be compared due to the presence of NS binding sites. In principle, the fluorescence of 1,8-ANS is small. I see, fewer 1,8-ANS molecules are bound to the antibody. Stress causes 1,8-ANS to be released. Changes in light intensity and emission wavelength were evaluated.

[0298] [Table 16]

[0299] In summary, these data demonstrate that various IgG samples are remarkably stable. Severe stress conditions were used to produce a measurable difference between the tested samples. A sample (for example, 2 days at 50°C) is required. Under these conditions, combination number 9 and Sample 10 appears to be more prone to aggregation than the other samples.

[0300] The most important factors that can distinguish the stability of proteins are the freeze-thaw cycle and temperature increase. Taking into account incubation at 50°C, which is a severe stress factor, T366K / L3 51E, Y349E (combination number 4) and T366K, L351K / L351D, Y349E (combination number 11 Two variants of ) are the most stable proteins in this group, with T3 being the next most stable by a small margin. 66K, L351K / L351D, Y349D (combination number 10) and T366K, L351K / L351D, L368E (combination The sequence number 12) continues.

[0301] (Example 21: Experiment to perform native MS with different ratios; transfection ratio 1: (5 to 5:1) The behavior of CH3-mutated IgG in a mixture when transfected in a biased ratio. To deepen our knowledge on this topic, we will focus particularly on the combination of T366K:L351K / L351D':L368E' (see below). We conducted more detailed experiments on the ratio of KK / DE (or DEKK).

[0302] It possesses known specificity and binding ability to known common light chain human IGKV1-39, The VH region of the antibodies used up to this point was re-clone to constructs 1, 2, 68, and 69. It was used to perform the following, and as a result, vector IV (Table 17) was obtained. Different CH3 regions Each of the nucleic acid sequences encoding the Ig heavy chain, which has different antigen specificity, and the common human light chain, respectively Vector IV, including the one shown in Table 18, is transfected in different ratios. The cells were then transfected. The results are shown in Figure 23.

[0303] [Table 17]

[0304] [Table 18]

[0305] As shown in Figures 23A and 23B, in DEKK combination mutations, the excess amount When A or C exists (A or C is on the "DE side", B is on the "KK side"), AB or BC is formed, but in all cases, the excess amount of A or C is not homodimer. It exists as a mixture of both B and half. However, when an excess amount of B is present (B (where is "KK side" and A or C is "DE side"), a clear difference can be seen. AB or BC is not Although it is formed, the excess amount of B does not actually exist as a homodimer, and only half-dimers are formed. It will be done.

[0306] It should be noted that here too, the percentage is measured by the height of the peak. 2% or For peaks detected at values ​​smaller than this, the nMS technique used here is correct. It is lower than the threshold to which accurate measurement is possible. Therefore, measurements smaller than 2% are noise in the analysis. I decided to ignore it, considering it to be within the acceptable range.

[0307] It is worth noting that when B is exceeded, only the proportion of half of B increases. In particular, A When B is in a ratio of 1:3 and 1:5, there is no homodimer BB present, and half of the higher proportion is present. Child B was observed (Figures 23A and 23B), and the CH3 mutation on the KK side resulted in a homodimer. This indicates that it is unlikely to occur. The absence of homodimers is a decisive advantage. The reason is that, in order to add a specificity, when the "KK side" of the DEKK combination is selected, This is because the previously known adverse effects may occur when it exists as a homodimer. (For example, cMET or CD3 antibodies may exist in pharmaceutical compositions as divalent homodimers.) (It is known that when this is done, it can have undesirable adverse side effects.)

[0308] Observations that DE:KK have different ratios in vectors IV and V, This is in contrast to the charge inversion CH3 mutation in . As shown in Figure 23C, E356K:D399K / In the K392D':K409D' mutation, when A is present in excess (A is "K39 2D:K409D side), in all cases, the surplus amount A is a mixture of both homodimer and half-dimer. It exists as a combination. Even when B exceeds the limit (when B is on the "E356K:D399K side"), in all cases In the mixture, the excess amount of B exists as a mixture of both homodimers and half-dimers. Even in ratios of 1:3 and 1:5, homodimers exist, but half-dimer B is not observed. As a result, the E356K:D399K side does not prefer homodimers as much as the KK side in the DEKK combination. It was shown that it does not exist.

[0309] In summary, a mutation in the DEKK combination involves one of the heterodimer chains forming a homodimer. This has a clear advantage over the charge-reversal CH3 mutation, as it does not achieve this mutation.

[0310] (Example 22: Various mixtures using the DEKK combination) Mutations in the DEKK combination lead to the formation of highly pure, bispecific IgG molecules ("AB"). It was shown that induction was possible. Next, we were able to induce "AB and AA" or "AB and We explored the feasibility of controlled production of more complex antibody mixtures, such as "AC" mixtures. The Fab molds used so far are "DE construct" or "KK construct". It was incorporated into a vector containing any of the following: And, in order to demonstrate the versatility of the technology, Various combinations of these vectors were co-expressed to produce a mixture. Fab type MF1 337 (tetanus toxin), MF1122 (fibrinogen) and MF1025 (thyrogens) Robulin) exhibits these overall stable behaviors, good expression levels, and includes these Fabs. The IgG molecules were selected based on their mass differences (see Table 19).

[0311] [Table 19]

[0312] [Table 20]

[0313] According to SDS-PAGE analysis, most samples consisted mostly of full-length IgG, and In some cases, a low proportion of half-halves were present. Furthermore, many samples were found in gels under non-reducing conditions. At approximately 150 kDa, two bands were observed. These indicate that the sample had two distinguishable bands. This reflects the presence of IgG species. Even in gels under reducing conditions, several samples Two heavy-chain bands were observed (data not shown).

[0314] Native MS was performed on all samples, and the proportion of observed species was determined by the peak height. The following was calculated (percentage of "observed species" in Table 20). The results are shown in Figure 24. In all eight samples in which the heavy chain was co-expressed, there were two major peaks corresponding to the predicted species. This was observed. Of these samples, two (transfections 2 and 4) and tra In infection 11, a small amount of DE-DE homodimer was observed as contamination. The molecules were detected in very small amounts (less than 2%) in most samples. However, as mentioned above, this is not a problem because it can be easily separated from the full-length IgG portion. I don't know.

[0315] This was discovered after nMS, but in sample 11, the observed IgG mass corresponds to A different species than the one predicted was found. This was concluded to be an error in the transfection process. It was argued that, in sample 11, 1337-KK was used instead of 1122-KK. It was revealed that it was co-expressed with 1025-DE.

[0316] To confirm the functional presence of the desired specificity, the IgG sample was subjected to sandwich ELISA. Further tests were conducted using fibrinogen or thioglobulin in an ELISA test. The rate is coated, and detection is performed using fluorescein-labeled thioglobulin or The test was performed using tetanus toxin. The detection antigen was fluorescein (Pierce NHS-f) as instructed by the manufacturer. Labeled with luorescin Antibody Labeling kit (cat. #53029). The antigen that was detected was subsequently treated with FITC-conjugated anti-fluorescein antibody (Roche diagnostics, cat. Detected at # 11426346910.

[0317] Table 21 shows a summary of the results of the bispecificity ELISA (OD450 values). (Gray cells) This indicates the species predicted for each transfection. Generally, the experimental results are the predicted results and It matches, but exceptions are shown in italics or bold. Transfection 1-3 , Seed BC (Transfection #1 and #2) or AC (Transfection # 3) is a cell that is presumed to be "negative," but shows a significant background signal. Previous studies have shown that bispecific ELISA can handle high background levels. These background levels can be a source of trouble. It may be due to half-breeding. Note that the results of the bispecificity ELISA indicated that An error was confirmed in lancefection #11. Species A was used instead of BC. C (value in bold) was detected.

[0318] [Table 21]

[0319] (Example 23: Recognition of four different epitopes (AB and CD) from a single cell) (An improved mixture of two bispecific antibodies) In Example 12, the mixture from transfection ZA or ZB was transferred to mass production. There was a hypothesis that this would become a problem when that happened. The reason for this was Nobu In Two-hole variants have been reported to be unstable, and CH3 dormancy is characterized by the presence of a "knob" or "hole". This is because the possibility of dimerization of the yin with the charge-manipulated CH3 domain cannot be ruled out. As shown in the example above, selective heterodimerization is induced, and homodimer formation is substantially Novel charge pair variants were discovered. C possessing these novel charge pair variants Polypeptide chains containing an H3 domain are known to be charge-manipulated CH It is expressed in cells along with a polypeptide chain or SEED bodies that have three domains, and two This can result in the selective formation of only bispecific molecules.

[0320] As is clear from the above example, DEKK combination mutations involve heavy chain dimerization in the CH3 domain. In cloned cells induced by , one bispecific molecule (AB) or two It is excellent for producing bispecific molecules (AB and AC). However, complementary CH3 mutations are... If there is only one set of vectors that can be used as a variety, the types of mixtures that can be produced are limited. The types are limited. If it can be used in combination with DEKK, the second "orthogonal" If there is a set of "orthogonal" vectors, then "AB and CD" or "AB and CC" It is possible to produce more complex IgG and / or bispecific molecules such as hybrids. ru.

[0321] A key requirement when combining two sets of vectors is the C of two different sets. The heavy chain expressed from the H3-modified vector does not form "cross-dimers". "Crossing" refers to the process where a heavy chain produced by one set of vectors intersects with the other set of vectors. This involves the expression of the heavy chain and dimerized IgG to form the full-length IgG.

[0322] To test whether such "cross-dimer" formation is possible, HADDOC Using K, between the wild-type CH3 domain and the CH3 domain containing the DE- or KK-mutation In silico analysis was performed to gain insight into whether pairing occurs. Similarly, Between the wild-type CH3 domain and CH3 domains containing the E356K, D399K or K392D, K409D mutations Pairing may occur, including wild-type CH3 domain and knob-into-hole mutations. Is pairing possible between the CH3 domain and any of the above combinations? We analyzed the following. Table 22 lists the combinations of CH3 mutations analyzed with HADDOCK. The results were shown. The resulting HADDOCK scores are summarized in Figure 25.

[0323] [Table 22]

[0324] As shown in Figure 25, based on these HADDOCK predictions, the DEKK combination The most successful combination is when CH3 is combined with CH3 with charge reversal. It seems so. This combination reduces the amount of by-products that are introduced when co-expressed in a single cell (especially (AC, AD, BC, BD) are absent, and two bispecific molecules (AB and It can form a CD.

[0325] As can be seen in Figure 25, these undesirable bispecific species AC, AD, BC and BD has a relatively high HADDOCK score. On the other hand, the desirable AB and CD are It has the lowest HADDOCK score. Of course, it is the DEKK or charge-reversed CH3 group. When either of the combinations is inserted into a construct that retains the same singularity (for example) "C" on the DE side, "C" on the KK side, "A" on the E356K and D399K sides, and "B" on the E356K and D399K sides. " or "A" on the DE side, "B" on the KK side, "C" and E356K, D399K side When the 9K side (with "C") is co-expressed in cells, CC and AB are mainly produced.

[0326] In contrast, when we look at the predictions for co-expression of DEKK and wild-type, we see the predictions for AC and AD. The HADDOCK score is lower than the HADDOCK score for CD. This indicates that DE A vector encoding the KK CH3 combination and a vector encoding the wild-type CH3. When attempting to produce a mixture of AB and CD by co-expressing AC and This indicates that AD is very easily mixed in.

[0327] Finally, either the DEKK or charge inversion mutant, and the knob-into-hole mutant. When co-expressing them together, the prediction shows that undesirable bispecific mutants are relatively low in HA. The result was a DDOCK score. In other words, when co-expressed, these undesirable There is a high probability that seeds will be produced.

[0328] Therefore, the combination of CH3 in the DEKK combination and CH3 in the charge reversal combination ( E356K, D399K / K392'D, K409D') are essentially pure "AB and CD" and / or "AB It was concluded that this is ideally preferable for obtaining a mixture of antibodies for "CC".

[0329] Next, to put the above results into practice, we recognize four targets / epitopes (AB and CD). A mixture of two bispecific molecules that recognize three targets / epitopes (AB and CC) A mixture of one bispecific molecule and one monospecific molecule was produced. In the formation of the substance, all four different VHs are paired with the common light chain IGVK1-39. While it is possible to combine them, each individual VH / VL combination has a different specificity.

[0330] To enable native mass analysis, the (predicted) mass difference between species is sufficient. It must be a value greater than 190 Da. The number of species predicted for co-expression is n Four individual VHs take on mass values ​​that can be identified and separated by MS. The following was selected. Furthermore, the four selected VHs were grown in the mixture in addition to the two desired species. The mass difference is large enough to identify most possible contaminants. The list is shown in Table 23.

[0331] [Table 23]

[0332] As shown in Table 24, four different VHs are either "DE" or "KK" constructs Alternatively, it can be cloned into a vector containing a charge-reversal construct, and also co-expressed in several ways. The procedure was performed. As before, all vectors encode the common light chain IGKV1-39. It also contains nucleic acids. As previously shown, when two sets of vectors are combined... A key requirement is the expression of heavy chains from two different sets of CH3-modified vectors. The condition is that they do not form a "crossing" dimer. "Crossing" means that from one set of vectors... The produced heavy chain dimerizes with the heavy chain expressed by the other vector set, resulting in the full-length IgG. This is what happens. The "interchange" between a heavy chain containing a charge inversion mutation and a heavy chain containing a DE or KK mutation. To test the possibility of difference dimer formation, control transfections were performed. It broke.

[0333] [Table 24]

[0334] Table 25 shows the predicted species masses and transfection numbers 9-11 in Table 24. This provides a further overview of possible contaminants present.

[0335] [Table 25]

[0336] All purified protein samples obtained in transfection #1-11 should be stored in SDS. -The analysis was performed by PAGE and included three control samples (Figure 26). Furthermore, To identify all species in the sample, the protein samples from transfection #9-11 were used. Therefore, nMS analysis was performed.

[0337] As can be seen in Figure 26, transfection #3 and transfection #4 This is between the "KK" construct and either "E356K:D399K" or "K392D:K409D". This resulted in the predicted mismatch. Protein samples from these transfections The amount of half of the molecule in the sample exceeded the total amount of the IgG molecule.

[0338] Transfections #7 and #8 involve transfecting protein samples with half-particles and full-length IgG. The results showed that they were present in almost equal amounts. However, SDS-PAGE revealed that this IgG was present in almost equal amounts. The dimers are DE / DE dimers, DE / E356K:D399K dimers, or DE / K392D:K409D dimers, etc. It is not possible to deduce what this represents. It is noteworthy that the transfection In samples from #9-11, virtually no half-halves were observed.

[0339] Figure 27 shows the results of the nMS analysis for transfections #9 and #11. The proportion of species measured and the presence of other species were calculated by peak height. Transfection # In 9, the predicted species "AB and CD" accounted for 97% of the mixture (30% AB and 67% CD). This represents the result. On the other hand, there is a type of BD that is present in only about 3% of cases (Figure 27A). In Infection #11, the predicted species "AB and CC" make up 94% (33% AB) of the mixture. This represents 61% CC. On the other hand, BC (4) contains only 6% contamination. 0.1% and AC (1.8%) exist (Figure 27B).

[0340] As these data show, the second set of "orthogonal" vectors combines with DEKK When used together, they create more complex mixtures such as "AB and CD" or "AB and CC". This makes it possible to produce a mixture of IgG and / or bispecific molecules. Charge reversal When using both constructs and DEKK constructs together, there are very limited options. Only the formation of "cross-" dimers occurs. Adjust the transfection ratio. Therefore, it is predicted that these by-products, which are present in small proportions, can be further reduced.

[0341] (Example 24: Single-dose pharmacokinetic study in mice) Pharmacokinetics (pK) of bispecific antibodies that retain a combination of DEKK mutations in the CH3 region. To study the behavior of ), this study measured the pK parameters of three different IgG batches. The following were determined and compared. The three batches contained the following: 1) the parent of the wild-type anti-tetanus toxin Antibody 1337:1337 (wild-type Fc main chain with two MF1337Fabs), 2) wild-type antibody Tetanus toxin parental antibody 1516:1516 (with two MF1516Fab cells in the wild-type Fc main chain) 3) CH3-modified bispecific inhibitors that retain a DEKK combination mutation in the Fc region. Scarce toxin antibodies 1516:1337 (MF1516Fab on the DE side, MF1337 on the KK side) Fab).

[0342] DEKK-Bispecific antibody products have parental antibodies 1337:1337 and 1516 It was selected to have the specificity of :1516. The reason is that in some mouse strains Based on previous studies, there has been no pre-prescription serological reaction to these antibodies. This is because it is known that... Incidentally, if a serological reaction is present before prescription, the study results will be examined. This becomes difficult. Furthermore, nMS can determine the 1337:1337 (wild) relationship between parental antibodies. Type Fc), 1516:1337 (DEKK Fc) and 1516:1516 (Wild-type F One reason is that there is a sufficient mass difference to allow for the identification of c).

[0343] The three batches of IgG were prepared as previously described. However, the transfer The DNA used in the endotoxin was designed to minimize the amount of endotoxin. The batch was prepared using a toxin-free Maxiprep kit. The concentration, aggregation amount, endotoxin content, and proportion of bispecific products were tested. It was later shown to meet the acceptable criteria for the use of IgG batches in pK research. In other words, the IgG concentration after gel filtration was higher than 0.3 mg / ml, and the amount of aggregation was higher than 5%. The endotoxin level is low, at 3 EU / mg protein, and the DEKK batch is over 90%. It contained bispecific IgG.

[0344] Native mass spectrometry of the sample after gel filtration showed that the predicted species were present at a high proportion. In samples 1516:1337, a small amount of DE:DE homodimer was estimated to be about 2%. The organism was detected (Figure 28). The three IgG batches met the criteria for use in pK studies. That was the conclusion.

[0345] To compare pK parameters between three batches, three groups of female C57BL / 6J mice were used. (Harlan, The Netherlands) 1 mg / kg human IgG (5 ml / kg immunoglobulin) A brin solution ( / kg body weight) was administered. The animals were 7-8 weeks old at the time of administration, and approximately 18- The body weight was 20 grams. Blood samples were taken before administration, 15 and 60 minutes after administration, and after administration. Serum samples were collected at 2, 4, 8, 24, 48, 96, 168, 268, and 336 hours. The samples were prepared and stored at a temperature below -20 degrees Celsius until analysis. Each group consisted of four mau. It consists of three subgroups, namely 12 mice per group. Samples were collected starting from point 6.

[0346] Regarding mouse welfare, the European Community Guidelines for Animal Experiments (Directive 86 / 609 / EEC: the gene ral principles governing the use of animals in experiments of the European Commu (cities) and Dutch legislation (Dutch Act on Animal Experiments: The Experiments on This study was maintained in accordance with the Animals Act, 1997. This research was conducted by the Department of Laboratory Animal Welfare, National Institutes of Health, USA. Standards for Humane Care, issued by the organization. (and Use of Laboratory Animals), Identification Number 45859-01 (Expiration Date: April 30, 2015) ) was also followed.

[0347] Group 1 mice received the full-length monospecific IgG antibody 1516:1516 (triangle). Group 2 mice received the full-length dose of the monospecific IgG antibody 1337:1337 (square). The mice in group 3 had their CH3 region manipulated using DEKK (DE side: 1516, KK side: 1337). The patient received a full-length dose of the bispecific IgG antibody 1516:1337 (rhomboid shape).

[0348] Quantitative human IgG ELISA (ZeptoMetrix, NY USA; ELISA kit No. 0801182) Using this method, quantitative analysis of monoclonal human antibodies in mouse serum was performed by ELISA assay. I did it. Briefly speaking, the ELISA assay is based on the following principle: 96 wells of ELISA The ISA plate is coated with anti-human IgG, and the human monoclonal antibody is the anti-human Ig It binds to G. The bound antibody then processes horseradish peroxidase (HRP). Visualization was performed using a conjugated polyclonal anti-human IgG antibody.

[0349] The absorbance (OD) of each well is directly proportional to the amount of antibody in the serum sample. See Figure 29 for the results. This shows the bispecific IgG full length that retains the DEKK mutation combination and the single specificity of the parent. Sex antibodies were observed to be remarkably similar. CH3 mutations in DEKK-bispecific antibodies. It does not alter stability or half-life. Also, the DEKK mutant behaves like wild-type IgG. Mau.

[0350] (References) Deisenhofer J., Biochemistry 1981(20)2361-2370; Miller S., J. Mol. Biol. 1990(216)965-973; Padlan, Advances in Protein Chemistry 1996 (49) 57-133 Ellerson JR., et al., J. Immunol 1976 (116) 510-517; Lee and Richards J. Mol. Biol. 1971(55)379. Gunasekaran et al J.Biol.Chem. 2010(285)19637-19646 De Vries Nature Protocols 2010(5)883 Kabat et al, (1991) Mammalian Cell Biotechnology: a Practical Approach (M. Butler, ed., IRL Press, 1991 Merchant Nature biotechnology 1998(16)677 Ridgeway Protein Engineering 1996(9)617-621. Davis JH. Et al., Protein Engineering, Design & Selection 2010(23)195-202 Papadea and Check. Crit Rev Clin Lab Sci. 1989;27(1):27-58. Tissue Culture, Academic Press, Kruse and Paterson, editors (1973) Ionescu et al., J. Pharm. Sci. 2008 (97)1414) Current protocols in Protein Science 1995, coligan JE et al., Wingfield PT, ISBN 0-471-11184-8, Bendig 1988. Capelle, M.A.H., Brugger, P., Arvinte, T.Vaccine 23 (2005), 1686-1694. Demeule, B., Lawrence, M.J., Drake, A.F., Gurny, R., Arvinte, T. Biochim. Biophy s. Acta 1774 (2007a), 146-153. Demeule, B., Gurny, R., Arvinte, T., Int. J. Pharm 329 (2007b), 37-45. Lakowicz, J.R., Principles of fluorescence spectroscopy; Second Edition Kluwer A cademic / Plenum Publishers, New York, Boston, Dordrecht, London, Moscow, (2006) I SBN 0-306-46093-9.

Claims

1. A method for producing at least two different Ig-like molecules from a single host cell, Each of the two aforementioned Ig-like molecules comprises two CH3 domains capable of forming a contact surface. The above method involves the cells, a. A first nucleic acid encoding a polypeptide chain comprising a first CH3 domain, b. A second nucleic acid encoding a polypeptide chain comprising a second CH3 domain, c. A third nucleic acid encoding a polypeptide chain comprising a third CH3 domain, and d. A fourth nucleic acid encoding a polypeptide chain comprising a fourth CH3 domain, This includes giving The nucleic acid encodes means for selective pairing of polypeptide chains comprising the first and second CH3 domains with polypeptide chains comprising the third and fourth CH3 domains, Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include, to create at least one additional charge-charge interaction at the CH3 contact surface, a substitution from an uncharged amino acid to a positively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and a substitution from an uncharged amino acid to a negatively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, or Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include substitution of an uncharged amino acid to a negatively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and substitution of an uncharged amino acid to a positively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, The means for selective pairing polypeptide chains comprising the first and second CH3 domains differs from the means for selective pairing polypeptide chains comprising the third and fourth CH3 domains, The aforementioned method, A method further comprising the steps of culturing the host cells, expressing the at least four nucleic acids, and recovering the at least two different Ig-like molecules from the culture.

2. In the method according to claim 1, A method further comprising providing the host cell with nucleic acid encoding a common light chain.

3. In the method according to claim 1 or 2, In the polypeptide chain comprising the first CH3 domain, the substitution with a positively charged amino acid comprises lysine (K) at amino acid position 366, A method for substituting a negatively charged amino acid in a polypeptide chain comprising the second CH3 domain, wherein the amino acid at position 351 is aspartic acid (D).

4. In the method according to claim 3, A method wherein the polypeptide chain comprising the first CH3 domain further comprises lysine (K) at amino acid position 351.

5. In the method according to any one of claims 1 to 4, A method wherein each polypeptide chain comprising the CH3 domain further comprises a variable region that recognizes an epitope.

6. In the method according to any one of claims 1 to 5, The method wherein the at least two different Ig-like molecules are antibodies.

7. In the method according to claim 5, A method wherein at least one of the epitopes is located on a tumor cell.

8. In the method according to claim 5, A method wherein at least one of the epitopes is located in an effector cell.

9. In the method described in claim 8, The method wherein the effector cells are NK cells, T cells, B cells, monocytes, macrophages, dendritic cells, or neutrophils.

10. A mixture of at least two different Ig-like molecules obtained by the method described in any one of claims 1 to 9.

11. In the mixture according to claim 10, A mixture in which at least two different Ig-like molecules bind to different epitopes on the same antigen and / or different epitopes on different antigens.

12. In the mixture according to claim 10 or 11, The aforementioned at least two different Ig-like molecules are a mixture comprising at least one heterodimeric Ig-like molecule.

13. In the mixture according to any one of claims 10 to 12, A mixture in which two of the at least two distinct Ig-like molecules are heterodimeric Ig-like molecules.

14. A polypeptide chain comprising at least the first, second, third, and fourth CH3 domains A recombinant host cell comprising one or more nucleic acids that encode, The nucleic acid encodes means for selective pairing of polypeptide chains comprising the first and second CH3 domains with polypeptide chains comprising the third and fourth CH3 domains, Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include, to create at least one additional charge-charge interaction at the CH3 contact surface, a substitution from an uncharged amino acid to a positively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and a substitution from an uncharged amino acid to a negatively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, or Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include substitution of an uncharged amino acid to a negatively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and substitution of an uncharged amino acid to a positively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, The means for selective pairing of polypeptide chains comprising the first and second CH3 domains is different from the means for selective pairing of polypeptide chains comprising the third and fourth CH3 domains, in recombinant host cells.

15. In the recombinant host cell according to claim 14, The host cell is a recombinant host cell further comprising nucleic acids that code for a common light chain.

16. A pharmaceutical composition comprising at least two different Ig-like molecules according to any one of claims 10 to 13 and a pharmaceutically acceptable carrier.

17. In the pharmaceutical composition according to claim 16, A pharmaceutical composition comprising at least two different Ig-like molecules produced by the recombinant host cell described in claim 14 or 15.

18. A method for creating a host cell to produce at least two different Ig-like molecules, The aforementioned method, The procedure includes introducing one or more nucleic acids encoding a polypeptide chain having at least first, second, third, and fourth CH3 domains into the host cell, The nucleic acid encodes means for selective pairing of polypeptide chains comprising the first and second CH3 domains with polypeptide chains comprising the third and fourth CH3 domains, Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include, to create at least one additional charge-charge interaction at the CH3 contact surface, a substitution from an uncharged amino acid to a positively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and a substitution from an uncharged amino acid to a negatively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, or Means for selective pairing of polypeptide chains comprising the first and second CH3 domains include substitution of an uncharged amino acid to a negatively charged amino acid at amino acid position 366 of the polypeptide chain comprising the first CH3 domain, and substitution of an uncharged amino acid to a positively charged amino acid at amino acid position 351 of the polypeptide chain comprising the second CH3 domain, A method for selectively pairing polypeptide chains comprising the first and second CH3 domains, which is different from a method for selectively pairing polypeptide chains comprising the third and fourth CH3 domains.

19. In the method according to claim 18, A method further comprising the step of introducing nucleic acids encoding a common light chain into the host cells.

20. A culture of recombinant host cells according to claim 14 or 15, or recombinant host cells obtained by the method according to claim 18 or 19, which produce at least two different Ig-like molecules.

21. A mixture of at least two different Ig-like molecules, Each of the two Ig-like molecules comprises two CH3 domains capable of forming a contact surface. The first and second CH3 domains are selectively paired, and the third and fourth CH3 domains are selectively paired. The first Ig-like molecule comprises a first CH3 domain including a substitution from an uncharged amino acid to a positively charged amino acid at amino acid position 366, and a second CH3 domain including a substitution from an uncharged amino acid to a negatively charged amino acid at amino acid position 351, or The first Ig-like molecule comprises a first CH3 domain including a substitution from an uncharged amino acid to a negatively charged amino acid at amino acid position 366, and a second CH3 domain including a substitution from an uncharged amino acid to a positively charged amino acid at amino acid position 351. The means for selective pairing of the first and second CH3 domains is a mixture different from the means for selective pairing of the third and fourth CH3 domains.

22. In the mixture according to claim 21, The substitution of the first CH3 domain with a positively charged amino acid comprises lysine (K) at amino acid position 366, The substitution of the second CH3 domain with a negatively charged amino acid is a mixture comprising aspartic acid (D) at amino acid position 351.

23. In the mixture according to claim 22, The mixture wherein the first CH3 domain further comprises lysine (K) at amino acid position 351.

24. In the mixture according to claim 21, The aforementioned at least two different Ig-like molecules are a mixture comprising at least one heterodimeric Ig-like molecule.

25. In the mixture according to claim 21, A mixture in which at least two of the aforementioned two different Ig-like molecules are heterodimeric Ig-like molecules.

26. In the mixture according to claim 21, A mixture wherein at least one of the at least two distinct Ig-like molecules is a bispecific Ig-like molecule.

27. In the mixture according to claim 21, A mixture wherein at least one of the at least two distinct Ig-like molecules is a single-specific Ig-like molecule.

28. In the mixture according to claim 21, A mixture in which at least one of the at least two different Ig-like molecules is a bispecific antibody.

29. In the mixture according to claim 21, A mixture in which at least one of the at least two different Ig-like molecules is a monospecific antibody.

30. In the mixture according to any one of claims 21 to 29, A mixture comprising a second Ig-like molecule containing third and fourth CH3 domains having complementary knob-into-hole mutations, charge mutations, or a combination thereof.

31. In the method according to any one of claims 1 to 9, 18, or 19, The means for selective pairing of polypeptide chains comprising the third and fourth CH3 domains comprises a method comprising manipulated complementary knob-into-hole mutations, charge mutations, or a combination thereof.

32. In the recombinant host cell according to claim 14 or 15, the mixture according to any one of claims 10 to 13, the pharmaceutical composition according to claim 16 or 17, or the culture according to claim 20, The means for selective pairing of polypeptide chains comprising the third and fourth CH3 domains comprises recombinant host cells, mixtures, pharmaceutical compositions, or cultures comprising manipulated complementary knob-into-hole mutations, charge mutations, or combinations thereof.

Images