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 a heterodimeric Ig-like molecule from a single cell, where the Ig-like molecule comprises two CH3 domain that can form an interface, the method comprising providing in 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 comprising a second CH3 domain, where the polypeptide chain comprising the first CH3 domain has at least one amino acid substitution from a neutral amino acid residue to a positively charged amino acid residue, the polypeptide chain comprising the second CH3 domain has at least one amino acid substitution from a neutral amino acid residue to a negatively charged amino acid residue, where the method further comprises a step of culturing the host cell, expressing the two nucleic acid molecules and recovering the heterodimeric Ig like molecule 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 relates to the field of therapeutic antibodies used in the treatment of a variety of diseases. [Background technology]

[0002] Currently, the most commonly used biologics are isolated human or humanized recombinant monoclonal antibodies. These preparations allow the body's immune system to target cells involved in disease processes. neutralizing or eliminating bacteria and / or molecules, or invading pathogens or infectious agents This will increase eradication capabilities.

[0003] Monoclonal antibodies bind to a specific site on an antigen, called an epitope; and In therapeutic antibodies, epitopes may be used to, for example, eliminate tumor cells or to bind to receptor-ligand interactions. They are selected to perform desirable functions such as inhibiting interactions or neutralizing viruses. can be.

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

[0005] Despite these advantages, monoclonal antibodies have some disadvantages. Some of these are due to the inherent monospecificity of monoclonal antibodies and their ability to target specific diseases. It is related to the complexity that

[0006] It is not uncommon for many factors to be involved in the course of a disease. The actions of factors may be redundant or synergistic, resulting in increased expression of other receptors and signaling. Crosstalk between signaling networks can occur, and multiple pathways involved in disease may be involved. Inhibiting different components or pathways of the immune system may improve the efficacy of treatment. Due to their inherent monospecificity, monoclonal antibodies are ideally suited to complex disease processes. It can only interfere with one step in the process, and may not be effective enough. be.

[0007] Apart from the multifaceted nature of disease processes, single cells, soluble proteins, or pathogens Targeting a single epitope alone is not effective enough to treat the disease. The reason is that the target epitope is no longer monoclonal. This is because the antibody may bind to the target substance and not be able to exert the desired effect. For example, tumor cells may reduce or eliminate target epitopes on growth factor receptors. Mutations are often introduced or masked to evade monoclonal antibody therapy. Yes.

[0008] By activating other receptors and / or their ligands, tumor cells can respond differently. Similarly, viruses and other pathogens target epitopes. Mutations, deletions, or masking of the target gene can be used for monoclonal antibody therapy. Dodge.

[0009] Monoclonal antibodies that bind to a single epitope are more sensitive than polyclonal antibodies. Not all effector mechanisms are generated. Examples include opsonization (making antigens more susceptible to phagocytosis), steric hindrance, (antibody-encapsulated antigens are prevented from reaching host cells or mucosal surfaces), toxins Detoxification, agglutination, or precipitation (antibodies bound to some water-soluble antigens are agglutinated and then removed) complement activation and antibody-mediated cytotoxicity (antibodies are activated by NK cells and neutrophils) more likely to kill target cells).

[0010] Polyclonal antibodies suitable for therapeutic applications can be obtained from pooled human serum. Therapeutic polyclonal antibodies derived from such serum can be used to treat rabies, cytomegalovirus, and Treatment and prevention of viral infections such as flu and respiratory syncytial virus, tetanus toxin and It is used to neutralize toxins such as botulinum toxin or to prevent anti-D immunity.

[0011] The wider use of serum-derived polyclonal antibody preparations is expected to become a reality as plasma is used as a source of antibodies for infectious diseases and However, they are hampered by their applicability to only a limited range of targets, such as bacteria and toxins. Furthermore, products are dependent on the donor blood supply, both in terms of quantity and suitability. This results in considerable batch-to-batch variability. The technology for screening has not kept up with the constant evolution of the virus, and immunoglobulin preparations Finally, blood collection, screening, and immunization are important. Due to the lengthy process of immunoglobulin purification, plasma-derived immunoglobulins are expensive to produce. Light.

[0012] A mixture of monoclonal antibodies may increase the efficacy of the monoclonal antibodies, while serum There are no limitations associated with polyclonal antibodies of two human or human origin. Combinations of homologated monoclonal antibodies have been tested in preclinical models and clinical trials ( For example, a mixture of two monoclonal antibodies against the HER2 receptor, the EGFR receptor, a mixture of two antibodies against rabies virus and two monoclonal antibodies against rabies virus) .

[0013] In this field, the combination of two monoclonal antibodies has been shown to have additive or synergistic effects. effector that can achieve a mutual effect and is not related to either antibody alone For example, two types of monoclonal antibodies against EGFR or HER2 have been shown to induce Mixtures of monoclonal antibodies have been shown to kill tumor cells more potently. In addition to promoting cytotoxicity by immune system effectors, ATP also promotes receptor internalization and receptor activation. It is based on a combination of activities that includes enhanced inhibition of downstream signaling pathways of receptors.

[0014] In combination therapy based on two monoclonal antibodies, the component antibodies are administered separately. The difficulty with this method is that the two antibodies are mixed at the protein level. Conduct clinical trials for each and then repeat the process (partially) for the combination. This makes antibody combination-based treatments unsuitable for clinical trials. The cost may be unacceptable.

[0015] Alternatively, two recombinant cell lines producing the component monoclonal antibodies have been developed. The antibodies can be mixed in a fermenter and the resulting mixture of antibodies can be purified as a single sample (WO 2004 / 024994). The drawback of this method is that it is difficult to control the composition. This has resulted in poor reproducibility of the resulting recombinant polyclonal antibody preparations. This is particularly difficult when considering the temporal changes in composition that occur as cells are cultured. .

[0016] In the last decade, bispecific antibodies have been developed as an alternative to using a combination of two antibodies. In the case of a combination of two antibodies, two different immunoglobulin molecules are present in the mixture. Bispecific antibodies bind to different epitopes on the same or different targets. This is accomplished with a single immunoglobulin molecule.

[0017] By binding to two epitopes on the same or different targets, bispecific antibodies It can exert the same effect as a combination of two antibodies that bind to the same epitope. Furthermore, bispecific antibodies in IgG format have two different monovalent binding sites on one molecule. A mixture of two IgG antibodies has two different bivalent binding domains in one sample. Due to the combination of molecules, different effects of these formats have also been observed.

[0018] This makes it difficult, from a technical and regulatory perspective, to develop a single bispecific antibody. The reason is that production, preclinical and clinical trials are performed on a single molecule. Therefore, a therapy based on a single bispecific antibody would be more effective. A hassle-free, cost-effective drug discovery process will facilitate more effective antibody therapeutics Provide the law.

[0019] Bispecific antibodies in the IgG format, consisting of two heavy chains and two light chains, can be produced in a variety of ways. For example, bispecific antibodies can be produced by fusing two antibody-secreting cell lines or by recombinant These antibodies can be produced by expressing the two antibodies in a single cell using DNA technology. According to the method of

[2004] , multiple types of antibodies are produced because the heavy chains corresponding to each antibody are However, monospecific dimers (also called homodimers) contain two identical pairs of heavy chains with the same specificity. and bispecific dimers containing two different heavy chain pairs with different specificities. This is because they can form heterodimers (also called heterodimers).

[0020] Furthermore, the light and heavy chains from each antibody pair randomly, resulting in inappropriate, non-functional combinations. This problem, known as heavy and light chain mispairing, can lead to duplexing. This can be solved by selecting antibodies that share a common light chain that is expressed heterologously. Even with common light chains, expressing two heavy chains and one common light chain in a single cell Three different antibody types will be obtained: two monospecific "parent" antibodies; 'Parental' antibodies and bispecific antibodies. Bispecific antibodies must be generated from the resulting antibody mixture.

[0021] Further increasing the proportion of bispecific antibodies in the mixture of parental and bispecific antibodies, and Several techniques have been employed to reduce heavy and light chain mispairing. However, there are bispecific antibody formats that eliminate or minimize some of the above-mentioned problems. is needed. Summary of the Invention [Problem to be solved by the invention]

[0022] In summary, the prior art has not provided any monoclonal antibodies that can be used to treat patients. antibodies, bispecific antibodies, mixtures of monoclonal antibodies, or monospecific and bispecific antibodies Various techniques and methods are available for producing mixtures of antibodies.

[0023] However, as noted above, each of these existing techniques and methods has its own drawbacks and limitations. Therefore, mixtures or dual specific drugs targeting multiple disease-modifying molecules are becoming increasingly popular. Producing, improving, and / or developing biotherapeutic agents through heterologous approaches We need to develop new technologies that are different from what we have today. [Means for solving the problem]

[0024] The present invention provides a mixture or bispecific approach that targets multiple disease-modifying molecules. Improved and / or previously undeveloped methods for producing biological therapeutic agents by The present invention further provides methods and means of the art that are different from those of the prior art. and the resulting products and uses thereof.

[0025] In the prior art, various approaches have been proposed to promote the formation of specific bispecific antibodies of interest. These methods allow for the desired properties to be obtained in the resulting mixture. The composition of undesirable antibodies can be reduced.

[0026] For 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 the amino acid residues are in contact with each other, Interface residues or interface amino acids It is known that they face each other at a protein-protein contact surface consisting of two proteins (also called phospholipase C).

[0027] The contact amino acid of the first CH3 domain is one or more contact amino acids of the second CH3 domain. The contact amino acids typically interact with each other within 5.5 Å of each other in the three-dimensional structure of the antibody. (preferably 4.5 Å or less) from the contact residues of this one CH3 domain. The interactions between contact residues of the CH3 domain are due to, for example, van der Waals forces, water hydrogen bonds, water-mediated hydrogen bonds, salt bridges or other static bonds via electrical forces, attractive interactions between aromatic side chains, disulfide bonds, or other known forces This is done.

[0028] Approximately one-third of the contact amino acid side chains at the interface of the CH3 domain of human IgG1 are in contact with the It has been shown that it contributes to the majority of the folding and assembly of proteins. (adjacent) amino acid residues can also affect the interaction at this protein-protein interface. It is expected that:

[0029] In the prior art, approaches have been adopted to interfere with the dimerization of antibody heavy chains. Specific engineering of the CH3 domain to favor heterodimerization over dimerization Examples of such manipulation of the CH3-CH3 interface are given in, for example, the International Described in Publication No. 1998 / 050431, Ridgeway et al., 1996, Merchant et al. 1998 This is also known as the Knob-into-Hole approach. This method introduces mutations that result in complementary protrusions and cavities.

[0030] Generally, this method involves introducing protuberances onto the contact surface of a first polypeptide. The protrusion is inserted into the cavity corresponding to the second polypeptide, and the cavity is then formed. This promotes the formation of heteromultimers, and homomultimers Body formation is hindered.

[0031] The "protrusion" or "knob" is a short amino acid side chain at the interface of the first polypeptide. , by substituting a long amino acid side chain (e.g., tyrosine or tryptophan) A complementary "cavity" or "hole" of the same or similar size as the protrusion is created in the second The long amino acid side chains at the interface of the polypeptide are replaced by shorter amino acid side chains (e.g., The protrusions and cavities are created by substituting phospholipids (amino acids) with phospholipids (lanine or threonine). by synthetic means, such as modification of a nucleic acid encoding the peptide, or by synthesis of the peptide. can.

[0032] Using knob-into-hole technology alone, two parent antibodies and a bispecific antibody can be synthesized. At best, the proportion of the bispecific antibody of interest in the mixture is increased to 87%. Merchant et al. added disulfide between the two CH3 domains in the CH3-CH3 interface. The introduction of sulfide bonds increased the proportion of bispecific antibodies in the mixture to 95%. However, to use such bispecific antibodies as drugs, The antibody is purified from the homodimer and is then mixed with a pharmaceutically acceptable diluent or excipient. Purification of heteromultimers from such mixtures requires homodimerization. This poses a major challenge due to the similar physicochemical properties of the dimer and heterodimer.

[0033] One of the aims of the present invention is to provide a method for the production of single cell antibodies, which further improves the proportion of bispecific antibodies in the mixture. The present invention provides a method for producing bispecific antibodies from cell clones. The hole-into-hole technique can be used as a means, alone or in combination with other means, to This further improved proportion of bispecific antibodies in the product can be achieved.

[0034] Another example of such CH3-CH3 interface engineering is the technology of heterodimeric Fc. This technology is provided by the Strand-Exchange Engineered Domain Bispecific and asymmetric fusion proteins based on the proposed (ain; SEED) CH3 heterodimer These SEED·CH3 heterodimers are involved in the design of human IgA and Human IgG and IgA are composed of segments with alternating CH3 sequences of IgG and IgA. This results in 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; WO 2007 / 110205).

[0035] Yet another approach to producing the desired bispecific antibodies involves, for example, European Patent Application Publication No. 01870459 or U.S. Patent Application Publication No. 2010 / 0015 133, International Publication No. 2007 / 147901, International Publication No. 2010 / 129304, Gunasekaran et al (2010) and described in WO 2009 / 089004 As shown above, the electrostatic manipulation of the originally charged contact residues at the CH3-CH3 interface is It is based on

[0036] The mutations in the CH3 domain of the heavy chain described in these publications alter the original charged amino acid sequence. The contact residues of the amino acids are replaced with amino acid residues of the opposite charge (charge reversal strategy). This changes the polarity of the charges on the opposing contact surfaces of the Fc dimer. Therefore, co-expression of electrostatically compatible Fc chains may result in favorable attractive interactions. This promotes the formation of desirable Fc heterodimers, while repulsive charge interactions prevent the formation of desirable Fc heterodimers. Undesirable Fc homodimer formation is inhibited.

[0037] At the CH3-CH3 interface, four pairs of characteristic charged residues are involved in the interdomain interactions. These are D356 / K439', E357 / K370', K392 / D399' and D399 / K409' (the residues of the first and second strands are separated by a ' / ' and separated by a prime (') The numbering of the residues in the second chain follows the numbering reported by Kabat (1991). Because the CH3 contact surface has two-fold symmetry, each characteristic pair of charged residues is located twice in the native IgG. appear several times (i.e., K439 / D356', K370 / E357', D399 / K392' and K409 / D399' statically (Electron interactions also exist at the interface).

[0038] Taking advantage of this two-fold symmetry, we can invert a single charge, e.g., K409 in the first chain. In the case of D or D399K mutation in the second strand, homodimerization occurs due to repulsion of the same charge. It was shown that the formation of the nuclei was reduced by applying the inversion procedure to different charges. This repulsive effect is further enhanced by the reversal of the different complementary charges. Expression of the domains induces heterodimerization, resulting in dual specificity in the mixture. It was shown that the proportion of species increased.

[0039] The above approach increased the percentage of bispecific antibodies produced from a single cell to approximately 76%. One of the objectives of the present invention is to achieve a value between about 96% and about 96% from a single cell. In the method for producing a bispecific antibody, the ratio of the desired bispecific antibody is further improved. According to the present invention, electrostatic manipulation technology is used as one of the means. or other means, such as a knob-into-hole approach, to achieve the further improvement. By using the antibody, it is possible to achieve the desired proportion of (bispecific) antibodies.

[0040] In one embodiment, the present invention provides at least two different I In the method for producing an Ig-like molecule, each of the two Ig-like molecules forms an interface. and the method further comprises injecting the cell with two CH3 domains capable of: 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 comprising a second CH3 domain; c. a third nucleic acid molecule encoding a polypeptide chain comprising a third CH3 domain, and Beauty, d. a fourth nucleic acid molecule encoding a polypeptide chain comprising a fourth CH3 domain; providing at least two of said nucleic acid molecules with said first and second CH3 domains; a polypeptide comprising the main CH3 domain and a polypeptide comprising the third and fourth CH3 domains and a means for selective pairing of the host cells with the genotype of the target gene, the method comprising culturing the host cells and and expressing at least four nucleic acid molecules and isolating said at least two different Ig-like molecules from the culture. The method further comprises a step of recovering the product from the container.

[0041] For example, complex factors involved in disease processes or pathogen invasion, replication, and / or spread Producing multiple (bispecific) antibodies to more efficiently inhibit multiple biological pathways This is often desired.

[0042] The use of a mixture of multiple bispecific antibodies for the treatment of a particular disease is particularly useful. For example, during treatment with antibodies or small molecule drugs, tumor cells often develop resistance. Resistance involves multiple cell surface receptors and water-soluble molecules. and simultaneously addressing multiple such disease- and escape-related molecules. Therefore, it is believed that it would be beneficial to develop antibody-based cancer treatments.

[0043] In cases where two or more such disease- and escape-related target molecules or epitopes are involved, In this case, mixtures of bispecific antibodies are a new and interesting therapeutic modality. It is preferable that a mixture of antibodies be produced from a single cell, as this facilitates the drug discovery process. This makes the drug discovery process hassle-free from a regulatory perspective and pharmaceutical and clinically sound. From a development perspective, it will be cost-effective and easy to implement.

[0044] Single cell-based approaches allow for controlled and efficient production of bispecific antibodies This allows for the desired mixture of bispecific IgG molecules. This reduces or eliminates the need to separate the product from undesired monospecific IgG molecules. In the prior art, monospecific and bispecific antibodies can be isolated from a single cell. Antibodies against these viruses have been produced (WO 2004 / 009618). The compound is a complex mixture of several different bispecific and monospecific antibody types. do.

[0045] A further object of the present invention is to produce a mixture of defined bispecific antibodies from a single cell. As will be described later, preferably, the monomer At least 95%, at least 97% or higher than 99%, regardless of the amount of by-products in the body The present invention provides a method that results in a mixture of dimeric IgG molecules (bispecific) in a high proportion of dimeric IgG molecules. Generally speaking, in cells where multiple, native IgG molecules are produced, half molecules are produced. Although monomeric by-products are present, they can be easily removed by known size exclusion chromatography. Be removed.

[0046] In one embodiment, the present invention provides a method for the production of a single (bispecific) antibody of interest. 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 undesired dimeric antibody species. The resulting mixture is well-defined, and its composition is determined by the design of the CH3 domain mutations. Furthermore, the expression level can be regulated by different transfection factors. The injection ratio affects the composition of the mixture.

[0047] In the method according to the invention, the CH3 domain encoded by the first nucleic acid molecule is a second A CH3 domain encoded by a nucleic acid molecule that selectively pairs with the CH3 domain encoded by the nucleic acid molecule and is encoded by a third nucleic acid molecule. The CH3 domain encoded by the fourth nucleic acid molecule is selectively interdigitated with the CH3 domain encoded by the fourth nucleic acid molecule. Furthermore, the present invention provides at least two different I pairs obtained by the method of the present invention. provides a mixture of g-like molecules.

[0048] As used herein, "selection of a polypeptide comprising said first and second CH3 domains" refers to The term "selective pairing" refers to the resulting polypeptide comprising the first CH3 domain. and / or a polypeptide comprising a second CH3 domain. Generally, all polypeptides have one first CH3 domain and one second CH3 It is a dimer consisting of a pair of polypeptides with domains.

[0049] Similarly, "selective pairing of polypeptides comprising the third and fourth CH3 domains" "SEQ ID NO: 1" means the resulting polypeptide comprising the third CH3 domain and and / or substantially all of the dimers having a polypeptide comprising a fourth CH3 domain. a polypeptide comprising one third CH3 domain and one fourth CH3 domain The term "dimer" refers to a pair of polypeptides that contain the nucleotide sequence nucleotides ....

[0050] As a result, a polypeptide with four different (A, B, C, D) CH3 domains was obtained. When the encoding nucleic acid molecule was 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), As a result, a mixture of two specific Ig-like molecules is produced.

[0051] As will be described in detail below, in a preferred embodiment of the present invention, the first CH3 domain The polypeptide chain comprising the second CH3 domain has the amino acid substitution T366K. The polypeptide chain comprising the amino acid substitution L351D. The synthesis of the first and second CH3 domains induces selective pairing of polypeptide chains comprising the first and second CH3 domains. This is a suitable means for

[0052] The polypeptide chain comprising the first CH3 domain preferably further comprises an amino acid sequence The polypeptide chain comprising the second CH3 domain preferably has a substitution L351K. Preferably, further amino acids selected from the group of Y349E, Y349D and L368E. Within this group, L368E is the most preferred. In embodiments, the polypeptide chain comprising the third CH3 domain has the amino acid substitution E35 and D399K, and the polypeptide chain comprising the fourth CH3 domain is an amino acid sequence. It has the acid substitutions K392D and K409D.

[0053] In the method according to the invention, the polypeptide chains comprising each CH3 domain are preferably or further comprises a variable region that recognizes a target epitope. The variable regions that are part of the peptide chains preferably share a common light chain. Only the VH of the variable domains differ, while the VL of all variable domains are substantially identical.

[0054] Thus, in one preferred embodiment, the method according to the invention provides for the host cell to Further comprising providing a nucleic acid molecule encoding a common light chain. In the present invention, each of the four variable regions of a polypeptide chain comprising four CH3 domains is different. For example, if the first nucleic acid molecule recognizes a variable domain specific for antigen A, If the second nucleic acid molecule encodes a heavy chain further comprising a variable domain specific for antigen B, and a third nucleic acid molecule encoding a heavy chain further comprising a variable domain specific for antigen C. and a fourth nucleic acid molecule encoding a heavy chain further comprising a variable domain specific for antigen D. The mixture then encodes a bispecific Ig-like antibody specific for A and B. The antibody produced contains a bispecific Ig-like molecule specific for both the antibody and CD4.

[0055] Formation of monospecific antibodies (AA, BB, CC or DD specific) or AC, AD, BC Alternatively, the formation of bispecific antibodies specific for BD is reduced or eliminated. The reason is that the polypeptide comprising the first and second CH3 domains and the polypeptide comprising the third and second CH3 domains By means of selective pairing of polypeptides comprising four CH3 domains. For the production of defined mixtures containing different Ig-like molecules, e.g., the fifth and sixth CH Using additional nucleic acid molecules, such as those encoding polypeptides with three domains Of course, this is also 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. The resulting ratio of expressed Ig-like molecules does not need to be 1:1. It is possible to prepare a mixture of antibodies with optimized ratios using known means. For example, the expression level of a nucleic acid molecule and, therefore, the ratio of Ig-like molecules produced The rate uses different genetic elements such as promoters, enhancers and repressors Alternatively, the copy number of the antibody-encoding DNA construct can be adjusted by controlling the genome integration site. It can be divided.

[0057] The means for selective pairing is preferably a complementary knob-into-hole mutations, disulfide bridges, charge mutations including charge reversal mutations, or combinations of these. The means for selective pairing may be able to select from a range of types of mutations. Those skilled in the art will understand that the polypeptide chain encoding the CH3 domain At least four nucleic acid molecules are identified that utilize charge mutations as a means for selective pairing. It is necessary.

[0058] Furthermore, in certain cases, the unengineered wild-type CH3 also contains the wild-type CH3 domain. In one particularly preferred embodiment, the method is used for selective pairing of two polypeptide chains. The means for selective pairing comprises at least one CH3 variable selected from Table B. The differences will be discussed later in this book.

[0059] A preferred embodiment provides a method according to the invention, wherein all four of said nucleic acid molecules are A polypeptide comprising the first and second CH3 domains and the third and fourth CH3 domains. a means for selective pairing of polypeptides comprising said first and second domains; The means for selective pairing of a polypeptide comprising a second CH3 domain may comprise the steps of: A method for selective pairing of polypeptides comprising the third and fourth CH3 domains. It is different from dan.

[0060] One aspect of the present invention provides a method according to the present invention, wherein the first and second CH3 domains The means for selective pairing of polypeptides comprising the third and fourth amino acids This is distinct from the above means for selective pairing of polypeptides comprising CH3 domains. "Different" here means that the polypeptides comprising the first and second CH3 domains The means for selective pairing of the strands is preferably selective pairing of the first and second strands. In this design, the first and third and / or The polypeptide chain comprising the fourth CH3 domain is substantially free of interaction with the In other words, a polypeptide comprising a first CH3 domain and Dimerization with a third or fourth polypeptide essentially does not occur or is nearly so. The polypeptide comprising the third and fourth CH3 domains may be wild-type or A selective pairing different from the means for selective pairing of the first and second CH3 domains A ring means is provided.

[0061] Recent studies have demonstrated that the presence of the CH3 domain in the ATP-dependent ATPases, e.g., by the knob-into-hole technique, We focused on producing a single bispecific antibody using mutations (inversions) of charged contact amino acids. However, prior to the present invention, other dimeric by-products were also significantly produced. The production of a defined mixture of at least two (bispecific) Ig-like molecules without This has not been realized.

[0062] The present invention provides well-defined Ig-like molecules, including a high proportion of bispecific ones in the mixture. The present invention provides an efficient and controlled method for the production of a mixture of two bispecific antibodies. In the system, the proportion of (two) bispecifics is at least 95%, at least A 97% or higher percentage can be achieved. This means that the percentage can be as low as 5%, or even 3%. This means that only a small number of monospecific bivalent by-products are obtained. The amount of monomeric by-products (i.e., half molecules) is not very important because These half molecules can be easily separated by taking advantage of their size difference.

[0063] In another preferred embodiment of the invention, a first and second CH3 domain is provided. The variable regions of the polypeptide chain recognize different target epitopes, while the third and fourth C The variable regions of the polypeptide chains comprising the H3 domain recognize the same target epitope. This results in mainly one type of bispecific Ig-like molecule and one type of monospecific Ig-like molecule. For example, if a polypeptide comprising a first and a second CH3 domain is If the variable regions of the thiazide chains recognize different target epitopes, and the third and fourth CH3 The variable regions of the polypeptide chains comprising the domains are both identical, with the first and second CH domains. If the target epitope is different from that recognized by the AB domain, Alternatively, a mixture of Ig-like molecules with specificity for CC is created.

[0064] Further provided in accordance with the present invention is a method for producing a polynucleotide comprising a third and a fourth CH3 domain. The target epitopes recognized by the variable regions of the peptide chains are identical, but the first or The target epitope recognized by the variable region of the polypeptide chain comprising the second CH3 domain It is different from the pu.

[0065] In another method, the variable region of the polypeptide chain comprising the first and second CH3 domains is Polypeptides Recognizing Different Target Epitopes and Comprising Third and Fourth CH3 Domains and polypeptide chains in which the variable regions of the chains both comprise a first or second CH3 domain. When recognizing the same epitope, specificity for AB and AA or AB and BB A mixture of Ig-like molecules with different properties is produced.

[0066] The method of the present invention provides a polypeptide comprising a third and fourth CH3 domain. The target epitope recognized by the variable region of the polypeptide chain is located in the first or second CH3 domain. The target epitope is identical to the target epitope recognized by the variable region of the polypeptide chain comprising the It is something.

[0067] Another object of the present invention is to develop a method for the production of bispecific and monospecific antibodies in single cell culture. The objective is to provide methods and means for producing defined mixtures. A non-limiting example of a defined mixture is a mixture of a bispecific antibody specific for AB and a monospecific antibody specific for AA. Another example is a mixture of monospecific antibodies. Yet another example is a mixture of monospecific antibodies specific for A and B. A mixture of antibodies and CC-specific monospecific antibodies. Again, at least 90% Preferably 95%, most preferably at least 97%, or even more than 99%. Suitable means and methods for producing a mixture of antibodies of interest are provided.

[0068] In another embodiment, a method according to the invention is provided, wherein the first and second CH3 domains The variable regions of the polypeptide chains having the same target epitope recognize the same target epitope, while the third and The variable region of the polypeptide chain comprising the fourth CH3 domain is Recognizes a second target epitope that is different from the target epitope recognized by the region. This results in the production of monospecific Ig-like molecules with predominantly AA or BB specificity. As a result, the formation of bispecific Ig-like molecules is reduced or eliminated.

[0069] In some embodiments, a mixture of monospecific antibodies is preferred over a mixture of bispecific antibodies. Production in a single cell is preferred. For example, if cross-linking of two identical target molecules is desired, or the two targets are too far apart to be bound by a single bispecific antibody. When a mixture of monospecific antibodies is produced in a single cell, it is not possible to produce a single therapeutic antibody. This can be advantageous as it is considered a product.

[0070] The therapeutic efficacy and safety of various monospecific antibodies have already been demonstrated in the art. Manufacturing approval has also been obtained. The production of a mixture of monospecific antibodies in a single cell has enabled several Facilitate testing of the efficacy and safety of such mixtures and facilitate regulatory approval and manufacturing. However, the formation of bispecific by-products can be reduced by 5%. - A method for producing a specific mixture of monospecific antibodies in a single cell, which can be achieved in less than 1000 cells / well. Another object of the present invention is to provide a method for the preparation of such a bispecific antibody. The means and methods for producing well-defined homodimer mixtures with less than 5% formation The purpose of this invention is to provide a method for producing such a material.

[0071] The method according to the invention allows for the desired synthesis of any bispecific and / or monospecific Ig-like molecule. Again, this method is suitable for producing a mixture of more than two different Ig-like molecules. For example, the fifth and sixth (and seventh and eighth, etc.) CH Nucleic acid molecules encoding polypeptides comprising three domains can also be used. .

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

[0073] As used herein, the term "Ig-like molecule" refers to an Ig-like molecule that contains at least one immunoglobulin (Ig) domain. The Ig-like molecule refers to a protein molecule having at least one immunoglobulin. It comprises a sequence having the function of a CH3 domain, and preferably, the sequence is the CH3 domain of IgG1. Proteinaceous molecules having at least one CH3 domain further have specific binding domains. A mating portion may be provided.

[0074] The CH3 domains of the present invention contain means for selective pairing, allowing for desirable heterodimers. To design a monomeric binding molecule or mixture of binding molecules, two nucleotides containing a CH3 domain are synthesized. It is used for selective pairing of protein molecules. The linking moiety introduced into the molecule may be any linking moiety, including those shown in the following non-limiting examples: single chain Fvs, single chain or tandem diabodies (TandAb™) )), VHH, Anticalins®, Nanobodies®, BiTE®, Fab , ankyrin repeat proteins or DARPINs®, Avimers®, DART, TC R-like antibodies, Adnectins®, Affilins®, Trans-bodies®, Affibodies®, TrimerX®, MicroProteins, Fynomers® , Centyrins® or KALBITOR®.

[0075] In one preferred embodiment, the binding moiety comprises a variable region of an antibody (i.e., a VH / VL combination). A variable region that is part of a polypeptide chain comprising a CH3 domain is preferred. or share a common light chain. In such cases, only the variable region VH differs, On the other hand, the VLs are substantially identical in all variable regions.

[0076] Additionally or alternatively, cytokines, hormones, water-soluble ligands, receptors, Receptors and / or peptides and other molecules can also be introduced into the CH3 domains of the present invention. Cut.

[0077] In a more preferred embodiment, the Ig-like molecule comprises a full-length Fc main chain. In an embodiment, the Ig-like molecules are antibodies. Preferably, the variable regions of these antibodies are covalently linked. They share a common light chain but may differ in the VH region.

[0078] As used herein, the term "antibody" refers to a proteinaceous substance that belongs to the immunoglobulin class of proteins. A molecule of the present invention, comprising a domain that binds to an epitope on one or more antigens, and The term "knockout" refers to a molecule containing a known variable region of a human antigen that is derived from or has sequence homology with the known variable region of the human antigen. The antibodies are IgG1, IgG2, IgG3, IgG4, IgA, IgD, IgE, and IgE. The antibodies of the present invention may be of any isotype, including several isotypes such as gM. or functional derivatives and / or fragments thereof. As for Ig-like molecules, antibodies of the IgG isotype are produced. The reason is that IgG antibodies , because they have a longer half-life compared to antibodies of other isotypes, for example.

[0079] The antibodies produced by the methods of the present invention may contain murine and human sequences. The antibody may have sequences derived from one or more of the following: The antibody may be composed of sequences from different origins, resulting in what are called, for example, chimeric or humanized antibodies. As will be appreciated, the sequences may have more than one origin.

[0080] The closer a therapeutic antibody is to a subject's natural antibody, the better (e.g., a human (In the case of human antibodies, this is a human antibody.) Antibody binding is expressed in terms of specificity and affinity. Affinity determines which antigen or epitope a binding domain will bind to. Specific binding is a measure of the strength of binding to a particular antigen or epitope. Affinity (K D ) at least 1 × 10 -5 M, more preferably 1×10 -7 M, Yo More preferably 1×10 -9 M is defined as binding with an affinity higher than M. Typically, therapeutic of monoclonal antibody, 1 × 10 -10 M affinity or higher is used. I can.

[0081] The term "antigen" as used in this book refers to a substance that, when taken into the body, is converted into antibodies by the immune system. Antigens are substances or molecules that cause the production of antigens. Examples include drug substances, tumor cells or other abnormal cells, haptens, or autologous tissue. At the specific level, antigens are characterized by their ability to bind to the antigen-binding site of an antibody. Those skilled in the art will recognize that tumor cell lysates and While viral particles are often considered "antigens," tumor cell lysates or Many antigenic determinants are also present in preparations of virus particles.

[0082] An antigen comprises at least one, and often two or more, epitopes. The term "group" refers to a group of proteins that are recognized by the immune system, specifically antibodies, B cells, or T cells. An epitope is a part of an antigen that is present in a specific antigen. An epitope is usually thought to be derived from a non-self protein. However, host-derived sequences 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. Each light chain consists of two domains: a variable region and a constant region (VL and CL). Each heavy chain has a variable domain (VH) and three constant domains (CH1, The heavy chain has four domains: CH2 and CH3. refers to the Fc (Fragment crystallizable) portion, Fc fragment, Fc main chain, or simply Fc It is called.

[0084] IgG molecules consist of two polymers connected by disulfide bonds (-SS-) at the hinge region. The heavy chain is a heterotetramer consisting of a heavy chain and two light chains. Dimerization occurs through interactions at the face and hinge region. Hinge region disulfide The number of bonds varies depending on the immunoglobulin subclass (Papadea and Check 1989).

[0085] The Fc fragment of an immunoglobulin is a dimer of two C-terminal constant regions. namely, the CH2 and CH3 domains of the heavy chain. Some of their physiological functions are the regulation of the complement system and and interactions with specific receptors on the surface of various cells. The interaction between these three domains is known to play an important role in inducing heavy chain dimerization. It is being done.

[0086] Therefore, the CH3 domain plays a leading role in linking the antibody heavy chain. The main interface involves over 20 contact residues from each chain, with 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 variant CH3 domains of the invention may be used in conjunction with other antibody domains to produce bispecific antibodies. Alternatively, monospecific full-length antibodies can be generated. The specificity of the antibody to be obtained is generally determined by the heavy chain dimerization behavior induced by the CH3 domain. does not affect.

[0088] As used in this document, the terms "contact residue," "contact amino acid," "interface residue," and "interface amino acid" are used interchangeably. The term "acid" generally refers to any amino acid present in the CH3 domain that may be involved in interdomain contacts. In this regard, the C The solvent accessible surface area (ASA) of residues in the H3 domain In the calculation, there was a difference in ASA (>1Å) between the calculations under the two conditions. 2 ) are considered as contact residues. by known techniques, including the method of identifying the The residues identified as contact residues are numbered 347, according to the EU numbering system. 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 interface can be either charged or neutral amino acid residues. As used herein, the term "charged amino acid residue" or "charged residue" refers to an amino acid that is electrically These refer to amino acid residues with a charged side chain. These include arginine (Arg, R), Positively charged sides such as those present in histidine (His, H) and lysine (Lys, K) It can be a chain, and exists in aspartic acid (Asp, D) and glutamic acid (Glu, E) The side chain may be negatively charged, such that

[0091] As used herein, the term "neutral amino acid residue" or neutral residue refers to an amino acid that is electrically charged. 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 ), tyrosine (Tyr, Y), and tryptophan (Trp, T).

[0092] In this document, the term "CH3-CH3 domain interface" or "CH3 interface" or "CH3-C The "H3 pairing," "domain interface," or simply "interface" is a region that contains the CH3 domain. As a result of the interaction of amino acid residues in two CH3 domains of different polypeptides The relationship refers to the amino acids of the first CH3 domain and the second CH3 domain. Such interactions include, for example, at least one interaction between amino acids of the van der Waals forces, hydrogen bonds, water-mediated hydrogen bonds, salt bridges or other electrostatic forces, This may be due to attractive interactions between aromatic side chains, disulfide bond formation, or other known forces.

[0093] As used herein, a polypeptide comprising a first and a second CH3 domain. and a polypeptide comprising the third and fourth CH3 domains, The means may be any known means.

[0094] In one embodiment, at least one nucleic acid molecule is selected from the group consisting of, for example, R, F, Y, W, I or The position of the contact residue is determined by the large amino acid residue (i.e., "knob" or "protrusion") such as L. while at least one other nucleic acid molecule encodes a CH3 domain contained in, e.g., small amino acid residues such as G, A, S, T, or V (i.e., "holes" or "cavities") ") encodes the CH3 domain containing complementary contact residues. The CH3 domains pair with each other due to the conformation of the contact amino acids. The knob-into-hole technique has already been discussed in detail here.

[0095] In a further embodiment of the invention, at least one nucleic acid molecule comprises a nucleotide sequence of naturally charged residues. At the positions of the contact residues that were originally K, H, R, D, or E, It encodes a CH3 domain containing amino acids that carry the opposite charge compared to the original form, while At least one other nucleic acid molecule has a complementary contact residue at the position where the original charged residue was located. The CH3 domain contains amino acids that carry the opposite charge compared to the wild type. The resulting engineered CH3 domains are connected to each other by the opposite charges of the contact amino acids. While the same CH3 domains tend to pair, they do not pair due to electrostatic repulsion. Difficult.

[0096] According to one embodiment, the method disclosed in European Patent Application Publication No. 01870459 (Patent Document 3), International Patent Application Publication No. The CH3 mutation described in Patent Publication No. 2009 / 089004 and Gunasekaran et al. (2010) It is used.

[0097] In one embodiment, the selection of polypeptides comprising the first and second CH3 domains is The means of selective pairing are "knob" and "hole" amino acid residues, and the third and The means for selective pairing of polypeptides comprising a fourth CH3 domain is charge engineered. Preferably, the polypeptide comprising the first and second CH3 domains and selective pairing of a polypeptide comprising said third and fourth CH3 domains with said polypeptide comprising said third and fourth CH3 domains. Both of these tools are charge-engineered amino acids.

[0098] In one embodiment, the selection of polypeptides comprising the third and fourth CH3 domains the first and second CH residues relative to the amino acid residues engineered for selective pairing. The amino acid residues engineered for selective pairing of the three-domain polypeptide are: different.

[0099] In a particularly preferred embodiment, at least the first and second nucleic acid molecules are those of the nucleic acid molecule provided herein. As described in more detail below, the present invention provides novel mutations in the CH3 domain that significantly The specific bispecific peptides of interest can be produced without producing excessive amounts of undesired (dimeric) by-products. Furthermore, the present invention provides novel CH3 mutations that allow the production of specific Ig-like molecules. The desired specific monospecific peptides can be produced without producing excessive amounts of undesired (dimeric) by-products. Thus, the present invention provides novel CH3 mutations that allow the production of specific Ig-like molecules. The use of at least one of these CH3 mutations is preferred.

[0100] As used herein, "polypeptide," "polypeptide molecule," or "polypeptide chain" The term protein refers to a chain of amino acids covalently linked through peptide bonds. Generally, a polypeptide is composed of one or more polypeptide molecules. One end, called the amino terminus, or N-terminus, has a free amino group. The other end bearing the carboxyl group is called the carboxyl terminus or C-terminus. Polypeptides from may undergo post-translational modification processes, such as glycosylation. Thus, the polypeptide chains comprising the CH3 domain of the present invention comprise at least Ig·CH3 It refers to a polypeptide chain that includes a domain, and may include those that have undergone post-translational modifications.

[0101] As used herein, the term "nucleic acid molecule" refers to a chain of nucleotides, more preferably DNA and In one embodiment, double-stranded RNA is used. In other embodiments, the nucleic acid molecules of the invention are, for example, DNA / RNA helices, peptides, or the like. peptide nucleic acid (PNA), locked nucleic acid (LNA) and and / or other types of nucleic acid structures such as ribozymes. The term refers to non-natural nucleotides, modified nucleotides and / or nucleotides identical to natural nucleotides. It also includes strands that have functional non-nucleic acid components.

[0102] Additionally, the present invention provides a method for producing a host cell that produces at least two different Ig-like molecules. In one embodiment, the method comprises injecting at least first, second, third and fourth C introducing a nucleic acid sequence encoding a polypeptide chain comprising an H3 domain, At least two of said nucleic acid sequences are polypeptides comprising said first and second CH3 domains. and a polypeptide comprising the third and fourth CH3 domains. The method comprises steps, wherein the nucleic acid sequences are introduced sequentially or simultaneously.

[0103] In a further aspect of the invention, a host cell for the production of heterodimeric Ig-like molecules is provided. In a method for producing a recombinant vector comprising: introducing a nucleic acid sequence encoding a polypeptide chain comprising an H3 domain, The polypeptide chain containing the first CH3 domain is positively charged from neutral amino acid residues a polypeptide comprising at least one substitution of an amino acid residue and comprising said second CH3 domain; The peptide chain consists of at least one substitution from a neutral amino acid residue to a negatively charged amino acid residue. The nucleic acid sequences may be introduced sequentially or simultaneously. The method preferably includes the step of introducing into said host cell a nucleic acid sequence encoding a common light chain. Equipped with.

[0104] In one embodiment further provided herein, at least first, second, third and fourth CH3 In a recombinant host cell comprising a nucleic acid sequence encoding a polypeptide chain comprising the domain and wherein at least two of said nucleic acid molecules comprise a polypeptide comprising said first and second CH3 domains. and selective pairing of a polypeptide with the third and fourth CH3 domains. Provide a means of monitoring.

[0105] Further provided herein is a polypeptide comprising at least a first and a second CH3 domain. a recombinant host cell comprising a nucleic acid sequence encoding the first CH3 domain; A polypeptide chain comprising a small number of neutral to positively charged amino acid residues The polypeptide chain comprising the second CH3 domain and comprising at least one substitution is neutral. a combination comprising at least one substitution of a neutral amino acid residue with a negatively charged amino acid residue; A recombinant host cell is provided.

[0106] The recombinant host cell according to the invention preferably comprises a nucleic acid sequence encoding a common light chain. do.

[0107] A "host cell" according to the present invention is any host cell capable of expressing a recombinant DNA molecule. For example, Escherichia (e.g., E. coli), Enterobacter, Salmonella, 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 SF-9 or SF-21 cells, and preferably CHO (Chinese hamster ovary) cells, mouse ovary cells, BHK cells, SP2 / 0 cells, and NS-0 myeloma cells human primate cells such as COS and Vero cells, MDCK cells, BRL 3A cells, Hybridoma, tumor cells, immortalized primary cells, W138, HepG2, HeLa, H Mammalian cells, such as embryonic retinal cells like EK293, HT1080, or PER.C6, are included. can be.

[0108] The choice of expression system often requires the use of mammalian expression vectors to ensure proper glycosylation of the antibody. Expression vectors and hosts of the following types of cells may be used: human cell lines, preferably PER.C6 is advantageously used to obtain antibodies that match the glycosylation pattern in humans. Conditions for growing or multiplying cells (Tissue Culture, Academic Press, Kruse and d Paterson, editors (1973)) and conditions for expression of recombinant products are somewhat different. Alternatively, the product ratio and / or Process optimization is usually performed to increase the growth of one or more cells, respectively.

[0109] General guidelines, procedures, and practical techniques for maximizing productivity in mammalian cell culture Mammalian Cell Biotechnology: a Practical Approach (M. Butler, ed., IRL Pres. The expression of antibodies in recombinant host cells has been extensively described in the literature. (For example, European Patent Application Publication No. 0120694, European Patent Application Publication No. 03141 61, European Patent Application Publication No. 0481790, European Patent Application Publication No. 0523949, (U.S. Patent No. 4,816,567, International Publication No. WO 00 / 63403). The nucleic acid molecule to be transferred may be an extrachromosomal copy and / or stably integrated into the chromosome of the host cell. However, the latter is preferred.

[0110] In a further aspect of the invention, a culture or a method for obtaining a recombinant host cell according to the invention is provided. In a culture of recombinant host cells possible or obtainable by the method according to the invention, said culture The nutrient is either at least two different Ig-like molecules or a heterodimeric Ig-like molecule. Something that produces something is provided.

[0111] In obtaining expression of a sequence of a nucleic acid molecule encoding a polypeptide comprising a CH3 domain, In this case, the sequence capable of inducing such expression is a polypeptide comprising a CH3 domain. It is known that the sequences of nucleic acid molecules encoding these peptides are functionally related. The term "associated" refers to a polypeptide comprising a CH3 domain or a precursor thereof. The nucleic acid sequence encoding the CH3 domain controls the expression of a polypeptide or a precursor thereof. It is related to a sequence that can induce expression, in that it can be induced. be.

[0112] There are many useful expression vectors available, such as the Invitrogen pcDNA vector series. The sequence encoding the polypeptide of interest is readily available. When sequences controlling transcription and translation of the peptide are properly inserted, the resulting The expression cassette is useful for producing, or in other words, expressing, a polypeptide of interest. be.

[0113] The sequence that induces expression includes promoters, enhancers, and the like, and combinations thereof. These must be able to function in the host cell, thereby providing functionally relevant nucleic acid sequences. Promoters can be constitutive or regulatable and can be viral, prokaryotic, or They can be obtained from a variety of sources, including those of eukaryotic origin, or can be engineered artificially. .

[0114] Expression of the nucleic acid of interest can be achieved by the native promoter or a derivative thereof, or by a completely heterologous promoter. This can be achieved by using a promoter that is well known and widely used in eukaryotic cells. The promoters include promoters derived from viruses such as adenovirus (e.g., E 1A promoter), promoters derived from cytomegalovirus (CMV) (e.g., CM V immediate early (IE) promoter), derived from simian virus 40 (SV40) Suitable promoters can also be obtained from eukaryotic cells, The tarothionine (MT) promoter, elongation factor 1α (EF-1α) promoter, and These include the tyrosine promoter, immunoglobulin promoter, and heat shock promoter.

[0115] Any promoter or promoters capable of directing expression of a sequence of interest in a host cell Enhancer / promoters are suitable for the present invention. In one embodiment, expression can be inducible. The sequence capable of transfecting the present invention comprises a region of the CMV promoter and preferably a region of the CMV immediate early gene enhancer. The promoter region from -735 to +95 is included. However, the expression sequences used in the present invention include insulators, matrix binding regions, Expression vectors such as STAR elements (WO 03 / 004704) Combinations of elements that stabilize or enhance expression would be preferred. and / or raise the bar.

[0116] Production of proteins in recombinant host cells is described, for example, in Current Protocols in Protein Sciences. cience, 1995, Coligan JE, Dunn BM, Ploegh HL, Speicher DW, Wingfield PT, ISBN 0- 471-11184-8, Bendig, 1988. Cell culture is Allow the cells to metabolize, and / or grow, and / or divide, and / or This is achieved by methods known to those skilled in the art. This method includes, but is not limited to, the formation of a surface and the provision of nutrients to the cells. The methods include growing in suspension, growing in suspension, or a combination thereof.

[0117] Some culture conditions can be optimized by known methods to optimize protein production. Cultivation can be carried out in batch, fed-batch, or other cultures, for example in dishes, roller bottles, or reactors. This can be achieved by cell culture, continuous culture, hollow fiber culture, etc. For the production of recombinant proteins (potentially), cells must be grown in suspension. It is known that serum derived from animals or humans or serum derived from animals or humans is preferred. It is known that it is preferable to culture cells under conditions free of serum components of origin. Thus, purification is easy because there are no additional animal or human derived proteins from the culture medium. On the other hand, synthetic media are optimal in terms of reproducibility, so the system is also very will also be reliable.

[0118] The Ig-like molecules are expressed in host cells and typically released from the cells or, preferably, the cell culture medium. After collection, these Ig-like molecules can be purified by known methods. Such methods include immunoprecipitation, centrifugation, filtration, size exclusion, and the like. Exclusion chromatography, affinity chromatography, cation and / or anion exchange These include exchange chromatography, hydrophobic interaction chromatography, etc. For antibody mixtures containing IgG, Protein A or Protein G affinity chromatography is preferred. (See, e.g., U.S. Patent 4,801,687 and U.S. Patent 5,151,504. ).

[0119] The Ig-like molecules and / or mixtures thereof produced by the method according to the invention are preferred. Thus, further provided is a method according to the invention, providing said host cell with a nucleic acid molecule encoding a common light chain, A light chain that can pair with at least two different heavy chains, thereby achieving functional antigen binding. A functional antigen-binding domain specifically binds to one antigen. can be done.

[0120] A common light chain that can pair with all heavy chains produced by the method of the present invention. This allows for a more suitable antibody to be used in the formation of a functional antigen-binding domain. In one embodiment, only one mispairing of heavy and light chains can be avoided. A common light chain of the same amino acid sequence can be used. Those skilled in the art will recognize that functionally equivalent light chains are included within the meaning of "common," even if the light chains are not identical. There are many variants of the light chain, which have mutations (deletions, substitutions, additions) However, these variants do not substantially affect the formation of functional binding domains. The antigen-binding domain can bind to a corresponding heavy chain to form a functional antigen-binding domain.

[0121] As used herein, the term "common light chain" refers to a light chain that is identical or has a different amino acid sequence. On the other hand, after pairing with the heavy chain, the resulting antibody contains light chains that retain binding specificity. For example, conservative amino acid changes and / or binding when paired with the heavy chain. Introduce and test amino acid changes in regions that do not contribute to specificity or contribute only partially to specificity. Experiments can be carried out to create light chains that are not identical but are still functionally equivalent, or It is possible to find.

[0122] The term "common light chain" includes a specific common light chain and functionally equivalent variants of such a common light chain. A detailed description of the use of common light chains is provided in WO 2004 / 024944. Preferably, the common light chain used in the present invention is a germline light chain. Preferably, the light chain is germline-derived, more preferably germline-derived, light chain. is a rearranged human germline-derived kappa light chain, most preferably a rearranged human germline-derived kappa light chain IgVκ1-39 / Jκ or IgVκ3-20 / Jκ .

[0123] Alternatively, instead of using a common light chain, one could also use mismatched heavy and light chains. To avoid pairing, the skilled artisan has been taught, for example, in WO 2009 / 080251, WO 2009 / 080252, WO 2009 / 080253, WO 2009 / 080254, WO 2009 / 080255, WO 2009 / 080256, WO 2009 / 080257, WO 2 As described in Publication No. 2009 / 080252 and / or WO 2009 / 080253 As disclosed, one may choose means for forced pairing of heavy and light chains.

[0124] In the present invention, novel engineered CH3 mutation combinations as well as novel engineered Prior to the present invention, no known CH3 domains were involved in CH3-CH3 pairing. The known charged contact amino acids of the CH3 domain are replaced by amino acids of opposite charge (charge reversal). It was substituted with an acid, which affected the CH3-CH3 pairing.

[0125] The mutations of the present invention are a separate creation from this approach because the wild-type CH3 In this case, uncharged or neutral CH3 amino acids are replaced by charged residues. In this embodiment of the invention, the charged contact amino acids are replaced with amino acids of the opposite charge. Instead of swapping, uncharged CH3 amino acids are replaced with charged ones.

[0126] The approach of the present invention provides a method for efficiently promoting CH3 domain dimerization. In addition, at least one additional charge-charge interaction is created at the CH3 interface. In addition to the charge pair at the CH3-CH3 interface, Due to the additional charge-charge interactions, the dimers according to the invention are similar to the wild-type dimers (wild-type dimers). The dimer is defined as a bispecific IgG (AB) without CH3 engineering and is a fusion protein with the parent homodimer. It is generally more stable compared to the dimer (as opposed to AA or BB).

[0127] Surprisingly, it is also possible to further increase the proportion of one or more desired Ig-like molecules in a mixture. As mentioned above, it is generally possible to preferentially produce bispecific antibodies. Known methods for producing hydroxybenzoates produce undesirable dimeric by-products. Using the two-hole technique, the percentage of desired bispecific antibodies is at best 87%. On the other hand, there is an electrostatic manipulation approach in which a charged contact amino acid is replaced by an amino acid of the opposite charge. approach, the rate is 96% (see, eg, Example 11).

[0128] Quite surprisingly, the inventors of the present invention have been able to further refine the proportion of desired Ig-like molecules in a mixture. For example, in Example 17, the mutations according to the present invention were used to introduce the The method used produced such a high percentage of the desired product that no dimeric by-products were detected in the resulting mixture. It has been shown that bispecific antibodies can be obtained by combining two heavy chains, one paired with a common light chain. Some unpaired half-molecules that are simply paired are present in the mixture, but these are the heavy chain This is the result of unequal expression of the β-glucan and can be easily separated from the mixture by size exclusion chromatography. Cut.

[0129] Therefore, such mutations according to the present invention result in a substantially free from contaminating dimer by-products. Without the need for a single cell, a high percentage of bispecific Ig-like molecules are produced, making them particularly suitable for pharmaceutical compositions. can be.

[0130] A preferred embodiment of the present invention provides a method for producing heterodimeric Ig-like molecules from a single cell. In the method for producing the Ig-like molecule, the Ig-like molecule has two CH3 domains capable of forming an interface. The method comprises the steps of: 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 comprising a second CH3 domain and a step of providing a polypeptide chain comprising the first CH3 domain, wherein the polypeptide chain is a neutral amino acid. at least one substitution of a 2-amino acid residue with a positively charged amino acid residue; The polypeptide chain containing the CH3 domain consists of neutral amino acid residues and negatively charged amino acids. and at least one substitution in a 2-amino acid residue of the ... and recovering said heterodimeric Ig-like molecule from the culture. Prepare. Preferably, the method comprises providing said host cell with a nucleic acid molecule encoding a common light chain. The advantages of this procedure have been outlined above.

[0131] The amino acid at position 366 in one CH3 domain and the amino acid at position 367 in the other CH3 domain The amino acid at position 351 is likely to be a contact residue pair at the CH3-CH3 interface. These have been reported to be the three-dimensional conformations of the resulting Ig-like molecules. are located close enough in the location that they can interact with each other. , the first CH3 domain preferentially pairs with the second CH3 domain.

[0132] In one embodiment, the threonine (T) at position 366 of the first CH3 domain is and the leucine (L) at position 351 of the second CH3 domain is replaced by a charged amino acid a second charged amino acid, said first and second charged amino acids being of opposite charge; If a polypeptide with a first CH3 domain containing a charged residue at position 366 is The peptide further comprises a variable region with specificity for antigen A, and if a polypeptide comprising a second CH3 domain having a charged residue at position 351 of If the antibody further comprises a variable region with specificity for antigen B, it is called a double antibody with AB specificity. Specific Ig-like molecules are predominantly formed.

[0133] The present invention further provides a method for producing a recombinant human avian flu virus comprising the first and second CH3 domains. or the means for selective pairing of polypeptides having the third and fourth CH3 domains. The means for selective pairing of polypeptides with the first or third CH A substitution of the threonine at position 366 in the 3 domain to a first charged amino acid and or a substitution of the leucine at position 351 of the fourth CH3 domain to a second charged amino acid. wherein the first and second charged amino acids have opposite charges.

[0134] One preferred mutation combination according to the present invention comprises a first CH3 domain and a variable region Threonine at position 366 of the polypeptide further comprising (e.g., specific for A) a (T) to lysine (K) substitution, and a second CH3 domain, At position 351 of the polypeptide further comprising a leucine (L), for example, a leucine (L) specific for B This is a substitution of aspartic acid (D). This is designated as the T366K / L351D mutation pair. can be.

[0135] As mentioned above, position 366 of one CH3 domain and position 367 of the second CH3 domain The amino acid at position 351 is reported to be a pair of contact residues in the CH3-CH3 interface. The lysine introduced at position 366 and the aspartic acid introduced at position 351 Having opposite charges, these amino acids attract each other electrostatically. The CH3 domain selectively attracts the second CH3 domain. The first CH3 domain has a nucleotide sequence corresponding to the amino acid at position 351, and the second CH3 domain has an aspartic acid at position 351. The predominant formation of Ig-like molecules is in pairs with IgG.

[0136] If the polypeptide comprising the first CH3 domain has specificity for antigen A, and Therefore, if the polypeptide comprising the second CH3 domain has specificity for antigen B, In this case, "AB" specific bispecific Ig-like molecules are predominantly formed. In some embodiments, the polypeptide chain comprising the first and second CH3 domains Both variable regions may be identical, in which case a monospecific Ig-like molecule (e.g., The formation of a nucleotide sequence (AA specificity) results.

[0137] As mentioned above, one of the advantages of the mutations according to the present invention is that the original charged amino acid Instead of replacing the interactions of the two, a new interaction between the newly introduced pair of charged amino acids was introduced. This has not been previously disclosed or suggested.

[0138] One aspect of the present invention provides at least two different Ig-like molecules from a single host cell. In the method of producing the polypeptide chain according to the present invention, comprises the amino acid substitution T366K, and the polypeptide chain comprising the second CH3 domain comprises It has the amino acid substitution L351D.

[0139] One embodiment provides a method for producing heterodimeric Ig-like molecules from a single cell, comprising: the Ig-like molecule comprises two CH3 domains capable of forming an interface, is added to the cells, - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. and wherein the polypeptide chain comprising the first CH3 domain provides an amino acid substitution T366K, and the polypeptide chain comprising the second CH3 domain comprises the amino acid substitution L 351D, the method comprising culturing the host cell, expressing the nucleic acid molecule, and The method further comprises recovering the heterodimeric Ig-like molecules from the culture.

[0140] By using the above-described amino acid mutations according to the present invention, it is possible to obtain heterodimeric I from a single cell. This allows the production of g-like molecules with less than 5% homodimer contamination. Preferably less than 2%, more preferably less than 1%, or most preferably less than can substantially eliminate homodimer contamination.

[0141] One embodiment provides a method for producing heterodimeric Ig-like molecules from a single cell, comprising: The Ig-like molecule has two CH3 domains that can form an interface and are intermixed. The presence of homodimers is less than 5%, preferably less than 2%, more preferably less than 1%. and most preferably substantially free of homodimer contamination, the method comprising: - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. wherein the polypeptide chain comprising the first CH3 domain is The polypeptide chain comprising the second CH3 domain comprises the substitution T366K and the amino acid substitution and a recombinant L351D gene, the method comprising culturing the host cell to express the two nucleic acid molecules. and recovering the heterodimeric Ig-like molecule 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 heterodimeric Ig-like molecule according to the present invention, the first CH3 domain More preferably, the polypeptide chain further comprises the amino acid substitution L351K. The polypeptide chain comprising the second CH3 domain comprises Y349E, Y349D and L 368E. Most preferably, said The polypeptide chain comprising the second CH3 domain comprises the amino acid substitution L368E.

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

[0144] As shown in Example 17, the first CH3 domain having specificity for antigen A according to the present invention a polypeptide comprising a CH3 domain specific for antigen B and a polypeptide comprising a second CH3 domain specific for antigen B Introduction of this mutation into the polypeptide resulted in a bispecific Ig-like molecule with dual AB specificity. This pair of mutations results in a particularly good ratio of homodimers. Therefore, bispecific antibodies can be obtained without the formation of α- and β-blockers.

[0145] One preferred embodiment provides a method for producing heterodimeric Ig-like molecules from a single cell. In the present invention, the Ig-like molecule comprises two CH3 domains capable of forming an interface. The presence of contaminating homodimers is less than 5%, preferably less than 2%, more preferably or less than 1%, and most preferably substantially free of contaminating homodimers, The method further comprises: - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. wherein the polypeptide chain comprising the first CH3 domain is provided with an amino acid substitution T366K, and the polypeptide chain comprising the second CH3 domain comprises the amino acid substitution L 351D and L368E, the method comprising culturing the host cell and The method further comprises expressing the molecule and recovering the heterodimeric Ig-like molecule from the culture.

[0146] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( and the leucine (L) at position 351 of the second CH3 domain was replaced with asparagine (K). A tyrosine (Y) at position 349 of the second CH3 domain is replaced by a glutamine (D). This is expressed as, for example, T366K / L351D, Y349E mutation, or For example, it may appear as T366K-L351D:Y349E or T366K / L351D,Y349E or simply T366K / L351DY349E .

[0147] The Y349 residue is a residue near the residue at position 351 that may contribute to dimer interactions. According to in silico data, Y349E contributes to the destabilization of the single dimer (higher computational speed). core) but also leads to heterodimer stability (lower calculated score), At position 349, glutamic acid (E) is more preferred than aspartic acid (D). Therefore, a polypeptide with a second CH3 domain already having an amino acid substitution at position 351 was prepared. Introducing a second amino acid substitution into the polypeptide makes it more susceptible to heterodimerization.

[0148] A particularly preferred embodiment provides for the production of heterodimeric Ig-like molecules from a single cell. In the method, the Ig-like molecule has two CH3 domains capable of forming an interface. and having less than 5% homodimer contamination, more preferably less than 2%; More preferably, the method is less than 1%, and most preferably substantially none, , - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. and wherein the polypeptide chain comprising the first CH3 domain provides an amino acid substitution T366K, and the polypeptide chain comprising the second CH3 domain comprises the amino acid substitution L 351D and Y349E, the method comprising culturing the host cell and and recovering the heterodimeric Ig-like molecule from the culture. .

[0149] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( The aspartic acid (D) at position 351 of the second CH3 domain was replaced with a lysine (K). a tyrosine (L) at position 349 of the second CH3 domain, and a glutamine and replacing the leucine (L) at position 368 of the second CH3 domain with glutamic acid ( E), which is designated as the T366K / L351D, Y349E, L368E mutation. Y349 and L368 are residues that may contribute to dimer interactions.

[0150] In silico data show that Y349E and L368E destabilize the BB dimer. (higher in silico score) as well as heterodimer stabilization (lower in silico score). score), and glutamic acids at positions 349 and 368 are replaced by aspartic acids (D ) is preferred. Therefore, the B-chain already having an amino acid substitution at position 351 is preferably Introducing a second and third amino acid substitution further promotes heterodimer formation.

[0151] A particularly preferred embodiment provides a method for producing heterodimeric Ig-like molecules from a single cell. In this method, the Ig-like molecule has two CH3 domains that can form an interface. Preferably, the homodimer contamination is less than 5%, more preferably less than 2%, and even more preferably less than 1%. Preferably less than 1%, and most preferably substantially none, said method comprising: - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. wherein the polypeptide chain comprising the first CH3 domain is provided with an amino acid substitution T366K, and the polypeptide chain comprising the second CH3 domain comprises the amino acid substitution L 351D, Y349E and L368E, and the method comprises culturing the host cell and expressing the two nucleic acid molecules and recovering the heterodimeric Ig-like molecule from the culture. Prepare further.

[0152] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( and replacing the leucine (L) at position 351 of the first CH3 domain with a lysine (K). The leucine (L) at position 351 of the second CH3 domain was replaced with an aspartic acid (D ) and a leucine (L) at position 368 of the second CH3 domain to glutamic acid (E). This is designated as the T366K, L351K / L351D, L368E mutation. As such, this mutation also increases the percentage of desired (bispecific) antibodies. Thus, bispecific antibodies can be obtained without detectable homodimer formation.

[0153] Further provided is a method for producing a heterodimeric Ig-like molecule from a single cell, comprising: The Ig-like molecule has two CH3 domains that can form an interface and form a homodimer. The contamination should be less than 5%, more preferably less than 2%, and even more preferably less than 1%. and most preferably substantially no, said method comprising: - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. wherein the polypeptide chain comprising the first CH3 domain is a polypeptide comprising the second CH3 domain and comprising the T366K and L351K mutations the chain comprises amino acid substitutions L351D and L368E, and the method comprises culturing the host cell and culturing the cells to express the two nucleic acid molecules, and recovering the heterodimeric Ig-like molecule from the culture. The method further comprises the step of:

[0154] In another preferred embodiment, the threonine at position 366 of the first CH3 domain ( and replacing the leucine (L) at position 351 of the first CH3 domain with a lysine (K). The leucine (L) at position 351 of the second CH3 domain was replaced with an aspartic acid (D ) and replacing the tyrosine (Y) at position 349 of the second CH3 domain with aspartic acid (D). and substituting an arginine (R) at position 355 of the second CH3 domain with an aspartic acid (D). This is denoted as the T366K, L351K / L351'D, Y349'D, R355'D mutation. The -L351K / L351'D-Y349'D pair was further improved by the R355'D mutation in the B-chain, resulting in BB. The in silico score for AB was slightly higher. become.

[0155] Further provided is a method for producing a heterodimeric Ig-like molecule from a single cell, comprising: The Ig-like molecule has two CH3 domains that can form an interface and form a homodimer. The mixing ratio is less than 5%, more preferably less than 2%, and even more preferably less than 1%. and most preferably substantially no, said method comprising: - a first nucleic acid molecule encoding a polypeptide chain comprising a first CH3 domain, and a second nucleic acid molecule encoding a polypeptide chain comprising a second CH3 domain. wherein the polypeptide chain comprising the first CH3 domain is a polypeptide comprising the second CH3 domain and comprising the T366K and L351K mutations The chain comprises amino acid substitutions L351D, Y349D and R355D, and the method comprises Host cells are cultured to express the two nucleic acid molecules, and the heterodimeric Ig-like molecule is cultured. The method further comprises a step of recovering the soluble matter from the culture medium.

[0156] Table B shows preferred means of selective pairing for the generation of heterodimers or homodimers. The following 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, or The method for producing a heterodimeric Ig-like molecule includes: said means for selective pairing of polypeptides and / or said third and fourth CH3 domains The means for selective pairing of polypeptides with at least one of Table B Preferably, the first and second mutations are the same or similar to those described above. said means for selective pairing of a polypeptide comprising a second CH3 domain and said third CH3 domain; and the means for selective pairing of a polypeptide comprising a fourth CH3 domain comprises at least It comprises at least two combinations of mutations shown in Table B.

[0158] The novel CH3 mutation combinations provided by the present invention allow for the expression of at least two CH3 mutations in a single cell. A mixture of monospecific Ig-like molecules can be produced, and the presence of bispecific Ig-like molecules is Less than 5%, preferably more than 2%, more preferably less than 1%, most preferably These mutations according to the present invention are particularly useful for producing mixtures of monospecific antibodies. This is particularly useful when high levels of cross-linking of two identical target molecules are desired, such as in tumor on target cells to activate specific effector mechanisms, such as complement-mediated lysis of the cells. When the antibody density needs to be high enough, or when the two targets are too far apart to form a single target, When a bispecific antibody cannot bind to a target molecule, or when regulatory approval procedures are simplified, This is particularly useful for

[0159] In such cases, the production platform for such monospecific antibodies is optimal. As shown in Example 10 and as inferred from the present invention, at position 392 of a polypeptide comprising a CH3 domain of 1 (e.g., having A specificity). Lysine (K) is replaced with aspartic acid (D), and the first CH3 domain is The aspartic acid (D) at position 399 of the peptide was replaced with a lysine (K), and the first CH The lysine (K) at position 409 of the polypeptide with the three domains was replaced with an aspartic acid (D). When substituted, at least two different monospecific Ig-like molecules containing monospecific Ig-like molecules with AA specificity are obtained. A mixture of monospecific Ig-like molecules can be produced in a single cell, and bispecific by-products (bispecific Ig-like molecules) can be produced. The formation of polyspecific Ig-like molecules is less than 5%, or less than 3%, or substantially reduced to the point where it is completely undetectable.

[0160] Therefore, the above mutation combination (denoted as K392D, D399K, K409D) is a monospecific It is particularly preferred for the production of a mixture of functional Ig-like molecules. Therefore, it can be recognized that K392E, D399R, and K409E can produce similar results. In addition, double mutants with substitutions D399K and K409D or K392D and K40 9D, D399R and K409E may also have similar effects.

[0161] The glutamic acid (E) at position 356 of the polypeptide comprising the first CH3 domain is replaced with lysine. (K) a glutamine at position 357 of the polypeptide comprising the first CH3 domain an acid (E) to a lysine (K) at position 43 of the polypeptide comprising the first CH3 domain; 9 lysine (K) to aspartic acid (D), Mutation combinations for substitution of lysine (K) at position 370 of the peptide with aspartic acid (D) The same applies to the combination of mutations (denoted as E356K, E357K, K439D, and K370D). ) are particularly preferred for producing mixtures of monospecific Ig-like molecules.

[0162] Those skilled in the art will appreciate that functional variants, i.e., K356R, E357R, K439E, K370E, result in It can be recognized that the same effect can be achieved. In addition, E356K, K439D, E357 Triple or double mutants with K and K370D substitutions, or other functional mutants, are also available. A similar effect could be achieved.

[0163] A further embodiment provides for the production of at least two different single-characterized isolates from a single host cell. In the method for producing isomeric Ig-like molecules, each of the two Ig-like molecules forms an interface. and wherein the cell has two CH3 domains capable of expressing the polypeptide, the method further comprising: a first nucleic acid encoding a polypeptide chain comprising a first CH3 domain having -A specificity; acid molecules, and a second nucleic acid encoding a polypeptide chain comprising a second CH3 domain having -B specificity; acid molecule wherein the polypeptide chain comprising the first CH3 domain is a polypeptide comprising the second CH3 domain and the mutations D399K and K409D; The peptide chains have either a wild-type CH3 domain or E356K, E357K, K4 and wherein the nucleic acid molecule is a nucleotide sequence encoding the nucleotide sequence of interest. and recovering the at least two different Ig-like molecules from the culture. can.

[0164] Another embodiment provides at least two different monospecific IgEs from a single host cell. In the method for producing an Ig-like molecule, the two Ig-like molecules can form an interface. and the method further comprises injecting the cell with two CH3 domains capable of expressing the a first nucleic acid encoding a polypeptide chain comprising a first CH3 domain having -A specificity; acid molecules, and a second nucleic acid encoding a polypeptide chain comprising a second CH3 domain having -B specificity; acid molecule wherein the polypeptide chain comprising the first CH3 domain is a wild-type C the second antibody having an H3 domain or having mutations K392D, D399K, or K409D; The polypeptide chain comprising the CH3 domain of 370D mutation, the method comprising culturing the host cell, expressing the nucleic acid molecule, and The method further comprises recovering the at least two different Ig-like molecules from the culture.

[0165] As shown in Example 10, two monospecific Ig-like molecules can be produced in a single cell, The formation of bispecific Ig-like molecules is virtually undetectable. a third nucleic acid molecule encoding a polypeptide chain comprising the CH3 domain; The host cells may be provided with antibodies such that a mixture of four monospecific antibodies is produced.

[0166] In one aspect of the present invention there is provided a method for treating a pulmonary artery disease comprising administering to a subject a subject a therapeutically effective amount of at least two different Ig-like molecules according to the present invention. In the method for producing a heterodimeric Ig-like molecule, each CH3 The polypeptide chains comprising the domains further comprise variable regions that recognize different target epitopes. , and the target epitopes are located on the same molecule.

[0167] This allows for a greater degree of targeting of the target molecule compared to when only one epitope is targeted. This allows for more efficient antagonism of biological functions. Ig-like molecules are receptors for growth factors or water-soluble molecules important for tumor cell proliferation. This allows the simultaneous binding of two epitopes present in several independent sequences. This effectively blocks signal transduction pathways, which can lead to uncontrolled proliferation. Any combination of two Ig-like molecules may be used to bind to receptors or other Ig-like molecules of such growth factors. It can simultaneously bind to two, three, or four epitopes present on a water-soluble molecule.

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

[0169] In another aspect of the invention, the production of at least two different Ig-like molecules according to the invention In the method or method for producing a heterodimeric Ig-like molecule, The polypeptide chain further comprises a variable region that recognizes a target epitope, the target epitope being In this case, each different target molecule is located on a water-soluble molecule or It may be a membrane-bound molecule.

[0170] In one embodiment, the different target molecules are water-soluble molecules. The molecule is a water-soluble molecule, while the second target molecule is a membrane-bound molecule. In one embodiment, both of the target molecules are membrane-bound molecules. In other embodiments, different target molecules are expressed in different cells. It will be revealed.

[0171] As a non-limiting example, any heterodimeric Ig-like molecule or a mixture of at least two Ig-like molecules. Any combination of molecules can simultaneously inhibit multiple membrane-bound receptors or induce cytotoxicity in tumor cells. It can simultaneously neutralize multiple water-soluble molecules, such as cytokines or growth factors, or can neutralize different It is suitable for neutralizing a specific virus serotype or virus strain.

[0172] One preferred embodiment provides a method for producing a heterodimeric Ig-like molecule comprising administering to a subject a subject the method comprising administering to a subject a method for producing a heterodimeric Ig-like molecule ... In the method of the present invention for producing a molecule similar to the above, at least one of the target epitopes is Alternatively or additionally, at least one of the target enzymes is located in a tumor cell. One of the pitopes is located on the surface of an effector cell, e.g., a tumor cell. Suitable for recruiting T cells or NK cells for killing. At least one of the produced Ig-like molecules is specific for a target molecule located on an immune effector cell. By heterologously binding to immune effector cells, preferably human immune effector cells, cells, can be mobilized.

[0173] In a further embodiment, after the Ig-like molecule binds to the target molecule, the immune effector The induction of effector mechanisms can be achieved, for example, by the method of the present invention. This includes the re-modulation of immune-regulated cytotoxicity by the Ig-like molecules produced by the Ig. g-like molecules bind to molecules that cause cytotoxicity, such as T cell receptors or Fcγ receptors can activate downstream immune effector pathways.

[0174] As used herein, the term "immune effector cell" or "effector cell" refers to an active The natural response of the mammalian immune system to oxidize and affect the viability of target cells. It refers to the repertoire of cell populations. Immune effector cells include natural killer (NK) cells. lymphoid cells such as cytotoxic T cells, T cells, or B cells rather, cells of the myeloid lineage, such as monocytes, macrophages, dendritic cells, and neutrophilic granulocytes. Cells are also considered immune effector cells. Thus, the effector cells are preferably , NK cells, T cells, B cells, monocytes, macrophages, dendritic cells or neutrophilic granulocytes. do.

[0175] Target antigens present on immune effector cells include CD3, CD16, CD25, and CD28 , CD64, CD89, NKG2D, and NKp46. Also provided are: , a method for producing at least two different Ig-like molecules or heterodimeric Ig according to the present invention A method for producing a CD3-like molecule, wherein the target epitope is CD3, CD16, CD25, CD2 8. Located on CD64, CD89, NKG2D, or NKp46 molecules. Target cell survival Competence includes the ability of the cell to survive, proliferate and / or interact with other cells.

[0176] One aspect of the present invention provides a method for producing a heterodimeric Ig-like molecule according to the present invention, comprising the steps of: Each polypeptide chain containing a CH3 domain contains a variable region that recognizes the target epitope. In one embodiment, the polypeptide chain comprising the CH3 domain further comprises a Each of the two variable regions recognizes the same target epitope with different affinities. In an embodiment, each of the two variable regions of the polypeptide chain comprising a CH3 domain recognize different target epitopes.

[0177] In other embodiments, the different target epitopes are located on the same target molecule, The molecule can be a membrane-bound molecule or a water-soluble molecule. The epitopes are located on different target molecules, and the target molecules can be on the same cell or on different cells. Alternatively, the different target molecules are water-soluble molecules, or one target molecule is expressed in a The second target molecule of the water-soluble molecule can be a membrane-bound molecule.

[0178] In a preferred embodiment, at least one target molecule of the heterodimeric Ig-like molecule is In yet another preferred embodiment, the heterodimeric Ig-like molecule is located in tumor cells. At least one target molecule of the effector cells (i.e., NK cells, T cells, B cells) cells, monocytes, macrophages, dendritic cells or neutrophilic granulocytes, and said target epitope is CD3, CD16, CD25, CD28, CD64, CD89, NKG2D or NKp Located at 46 molecules.

[0179] A preferred embodiment provides for the production of at least two different Ig-like molecules according to the present invention. In the method or method for producing a heterodimeric Ig-like molecule, the at least two The different Ig-like molecules are antibodies, most preferably antibodies of the IgG isotype, and more preferably Even more preferably, it is an antibody of the IgG1 isotype.

[0180] Ig-like molecules, heterodimeric Ig-like molecules, or at least Further provided are mixtures of two or more Ig-like molecules. The mixture of Ig-like molecules preferably contains at least one CH3 mutation listed in Table B. At least one Ig-like molecule or at least two Ig-like molecules according to the present invention as well as pharmaceutical compositions containing mixtures of (heterodimeric) Ig-like molecules or at least A mixture of two Ig-like molecules, each of which contains at least one mutation listed in Table B. is also offered.

[0181] In one embodiment, the Ig-like molecule is a bispecific Ig-like molecule, such as a bispecific antibody. In other embodiments, the Ig-like molecule is a single antibody, such as a monospecific antibody. A preferred embodiment provides a method for producing a specific Ig-like molecule according to the present invention. In a mixture of at least two different Ig-like molecules, the at least two different Ig-like molecules The term "antigen" refers to different epitopes on the same antigen and / or different epitopes on different antigens. joins to the group.

[0182] Further provided is a heterodimeric Ig-like molecule obtainable by the method according to the present invention, The heterodimeric Ig-like molecules may bind to different epitopes on the same antigen and / or different Binding to different epitopes on the antigen. Advantages and preferred uses of this mixture and antibody The law is stated above.

[0183] The present invention provides a method for producing at least two different Ig-like molecules obtained by the method of the present invention. In the mixture of molecules, the at least two different Ig-like molecules form at least one heterodimer. In one embodiment, the at least two different Ig-like molecules The molecule is a heterodimeric Ig-like molecule.

[0184] Yet another preferred embodiment provides a heterodimeric antibody having two CH3 domains. one of the two CH3 domains contains the amino acid substitutions L351D and L368 E, and the other of the two CH3 domains has the amino acid substitutions T366K and L35 These amino acid substitutions are the same as those in the two CH3 domains described above. This is the preferred means of selective pairing.

[0185] the amino acid substitutions L351D and L368E in one of the two CH3 domains; The other amino acid substitutions T366K and L351K in the two CH3 domains are In addition, "DEKK combination mutation," "DEKK mutant," "DEKK pair," "D "EKK engineered CH3 domain," "DEKK," or refer to DEKK Other names are used. CH3 domain carrying amino acid substitutions L351D and L368E The DE side is also called the "DE side," and the CH3 domain contains the amino acid substitutions T366K and L351K. The main side is also called the "KK side."

[0186] (Heterodimeric) Ig-like molecules, or at least Also provided is a pharmaceutical composition comprising a mixture of both Ig-like molecules. The Ig-like molecule, or said at least two Ig-like molecules, is preferably an antibody. The pharmaceutical composition is a monospecific or bispecific Ig-like molecule. mixtures containing monospecific and bispecific Ig-like molecules or combinations of monospecific and bispecific Ig-like molecules include.

[0187] The pharmaceutical composition according to the present invention also comprises a pharmaceutically acceptable carrier. Such "pharmaceutically acceptable carriers" include any and all solvents, salts, dispersion media, etc. , coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and The route of administration (e.g., intravenous, cutaneous) In the presence of steroids (such as steroids, intra-articular injections), Ig-like molecules are exposed to the action of acids and other substances that can inactivate Ig-like molecules. It may be coated with a material to protect the Ig-like molecule from other natural conditions.

[0188] In one embodiment, there is provided at least two In a pharmaceutical composition comprising a mixture of Ig-like molecules, the at least two different Ig-like molecules are The recombinant host cells according to the present invention are produced by the pharmaceutical compositions comprising: A method according to the present invention, comprising administering to a subject a heterodimeric Ig-like molecule obtained by any method according to the present invention, The Ig-like molecules are produced by recombinant host cells according to the invention.

[0189] A polypeptide comprising a CH3 domain comprising at least one mutation listed in Table B. and a nucleic acid molecule encoding a CH3 chain having at least one mutation described in Table B. a recombinant vector comprising at least one nucleic acid molecule encoding a polypeptide chain comprising a domain; Host cells are also provided.

[0190] The present invention is further illustrated by the following examples, which are not intended to limit the invention. 1, but is shown merely for clarity of the invention. [Brief explanation of the drawings]

[0191] [Figure 1] Figure 1A) is a schematic diagram of the construct vector MV1057. The stuffer region is the region into which the VH region of an antibody is cloned. Figure 1B) is a schematic diagram of the phage display vector MV1043. [Figure 2] This is the amino acid sequence of wild-type IgG1 Fc present in construct vector MV1057 (EU numbering system applied). [Figure 3] Nucleotide and amino acid sequences of the VH region used for cloning into the various constructs. [Figure 4A] Mass spectrum data of transfection A. [Figure 4B] Mass spectrum data of transfection G. [Figure 4C]Mass spectrum data of transfection H. [Figure 5A] Mass spectrum data of transfection M. [Figure 5B] Mass spectrum data of transfection U. [Figure 6] Mass spectrum data of transfection O. [Figure 7A] Inhibition of homodimerization by substituting neutral amino acids with charged amino acids. [Figure 7B] Inhibition of homodimerization by substituting neutral amino acids with charged amino acids. [Figure 8] Figure 8(A) shows the native mass spectrometry (MS) spectrum of the transfected sample ZO(T366K / L351'D). Figure 8(B) shows the convoluted MS spectrum of the transfected sample ZO(T366K / L351'D). The second / major peak is that of the bispecific molecule. [Figure 9] HADDOCK scores for experimentally validated mutation pairs. [Figure 10] Illustrated diagrams of CH3-CH3 interface interactions: Figure 10A) for K409D:K392D / D399'K:E356'K, Figure 10B) for D399K:E356K / D399'K:E356'K, and Figure 10C) for K409D:K392D / K409'D:K392'D. [Figure 11] HADDOCK scores of various 366 / 351' charge variants. [Figure 12] Illustrated diagrams of CH3-CH3 interface interactions. Figure 12A) is for L351D / L351'D, and Figure 12B) is for L351D:S354A:R355D / L351'D:S354'A:R355'D. [Figure 13] HADDOCK scores for additional charge mutations near position L351. [Figure 14]HADDOCK scores for additional charge mutations near position T366 in strand A and position L351 in strand B. [Figure 15] This is a diagram of the CH3-CH3 interface interactions. [Figure 16] HADDOCK scores for variants near T366 / L351. [Figure 17] HADDOCK scores for additional variants near T366 / L351. [Figure 18] Example of nMS spectra of bispecific IgG obtained after co-expression of constructs T366K,L351K with constructs L351D (left panel) or L351D,Y349E (right panel), scaled based on the monovalent full-length IgG (half not shown). [Figure 19A] Mass spectrometry results showing the relative abundance of AA, AB, BB, A, and B (total of all species is 100%). [Figure 19B] The data is the same as in FIG. 19A, but AB is omitted to make it easier to see the appearance of the undesirable AA, BB, A, and B. [Figure 20] Results of thermostability assay. Squares: wild type, triangles: charge-reversed pair E356K:D399K / K392D:K409D, circles: combinations of mutated CH3s shown above each graph. [Figure 21] Results from 10 freeze-thaw experiments. 1122 = primary parent antibody BB; 1337 = secondary parent antibody AA; wild type = AA,AB,BB; CR = bispecific antibody with charge-reversed pair E356K:D399K / K392D:K409D; 3-6 and 9-12 = bispecific molecule combinations 3-6 and 9-12 in Table 15. [Figure 22]Serum stability was measured by ELISA using fibrinogen as the coated antigen. 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. Results were normalized to 100% at T=0 (day). 1337 = second parent antibody AA; wild type = AA, AB, BB; CR = bispecific antibody with charge-reversed pair E356K:D399K / K392D:K409D; 3-6 and 9-12 = bispecific molecule combinations 3-6 and 9-12 in Table 15. [Figure 23A] nMS results for experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23A shows the DEKK combination mutant, with "A" specificity on the DE side and "B" specificity on the KK side. [Figure 23B] nMS results for experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23B shows the DEKK combination mutants, with the "C" specificity on the DE side and the "B" specificity on the KK side. [Figure 23C] nMS results for experiments with transfection ratios ranging from 1:5 to 5:1. Figure 23C shows charge reversal mutation combinations, with "A" specificity on the E356K:D399K side and "B" specificity on the K392D:K409D side. [Figure 24] nMS results for transfections #1-11 from Table 20. [Figure 25] HADDOCK scores of dimers of different CH3-engineered vectors. Grey bars: desirable species AB and CD; black bars: undesirable species AA, BB, CC, DD, AC, BC, AD, BD. [Figure 26] SDS-PAGE of transfections #1-11 from Table 20. Control samples DE / KK, DE / DE, and KK / KK are included. [Figure 27A] nMS of transfection #9. [Figure 27B] nMS of transfection #11. [Figure 28A] This is nMS of gel filtered sample 1516:1516. [Figure 28B] This is nMS of gel filtered sample 1337:1337. [Figure 28C] nMS of gel filtered sample 1516:1337. [Figure 29] Blood concentrations of the DEKK-engineered antibody and its two parent antibodies (pK study) DETAILED DESCRIPTION OF THE INVENTION

[0192] Example 1: Amino acid substitutions to generate a variety of different CH3 domains Selectively promoting or inhibiting pairing of Ig-like molecules comprising a CH3 domain; To obtain a wide variety of Ig-like molecules with different CH3 domains, we promoted heterodimer formation. Many amino acid substitutions known to be involved, as well as previously unreported or untested A number of alternative amino acid substitutions were made to the constructs that were selected to promote homodimer formation. The construct was introduced into a construct vector (construct vector MV1057; Figure 1A).

[0193] Construct vector MV1057 contains the F of normal wild-type IgG1 as shown in Figure 2. The nucleic acid sequence encoding the Fc region is shown in Table 1. This is a list of seven constructs. All constructs are Geneart Constructs 1, 2, and 3, or their surrogates, have been previously It has been reported that it promotes the dimerization of teratogenic compounds (European Patent Application Publication No. 01870459, International Patent Application Publication No. Constructs 6 and 7 have also been reported (International Constructs 4 and 5 are new and promote homodimerization. It is designed to

[0194] [Table 1]

[0195] Example 2: Cloning VH into constructs with CH3 mutations Cloning into these constructs involves the use of antibodies with known specificity 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 can be used in conjunction with other antibody domains to It can be a bispecific or monospecific full-length antibody. It is determined by the VH / VL combination. The specificity of the antibody used does not affect the behavior of heavy chain dimerization induced by the CH3 domain. No. Model VH / VL combinations, i.e., all light chains are human germline IG Based on KV1-39, various VH combinations were used throughout this study.

[0196] Figure 3 shows the complete sequences and specificities of the antibody VH regions used throughout this study. is the internal designation for various VHs at Merus, e.g., VH MF1337 is Tetanus toxoid MF1025 has specificity for porcine thyroglobulin, and MF1122 has specificity for bovine fibrinogen. .

[0197] The VH region in the phage display vector MV1043 (Fig. 1B) was purified by restriction enzyme SfiI and and BstEII (New England Biolabs / cat# R0123L and R0162L / ), according to the manufacturer's instructions. ) and the VH fragment is excised from this vector. Vector MV1057 is digested with SfiI and BstEII (according to the manufacturer's instructions). .

[0198] Fragments and vectors were gel purified (Promega / cat# V3125 / according to manufacturer's instructions) The excised vector and VH gene insert are then purified and isolated. The ligated nucleic acids were then transformed into E. coli DH5α (Invitrogen / cat# 12297-016) (following the manufacturer's instructions). The next day, single colonies were picked and The resulting vectors are then sequenced to identify those vectors containing the appropriate insert.

[0199] Example 3: Transfection and expression of full-length IgG in HEK293T cells The recloned VH variants, as well as various variants encoding the common light chain, human IGKV1-39, Transfection of the plasmid into HEK293T cells resulted in the expression of IgG. The transfer is carried out according to standard procedures such as those described in (de Kruif et al. Biotech Bioeng. 2010). After transfection, the IgG expression level in the supernatant was measured using the ForteBIO Octet-QK system. The system is based on Bio-Layer Interferometry (BLI), Enables real-time quantification and dynamic characterization of biomolecular interactions. See rtebio.com. IgG is expressed at levels greater than 5 μg / ml. Purified using affinity purification.

[0200] Example 4: Purification of IgG A Protein A column (GE Healthcare / cat# 11-0034-95 / according to the manufacturer's instructions) was used. After purifying the culture supernatant, it was eluted in 0.1 M citrate buffer at pH 3.0 and immediately Neutralize with an equal volume of 1.0 M Tris-HCl, pH 8.0, or directly pass through a desalting column. Alternatively, protein A beads (Sepharose beads) can be used to rebuffer the lysate in PBS. IgG may be purified using a ELISA kit (GE Healthcare cat#170780-01).

[0201] Example 5: Antigen-specific ELISA Antigen-specific ELISA is performed to assess the binding activity to the antigen. ELISA was performed to demonstrate the binding activity of the bispecific antibody. A 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 purified by standard procedures under reducing and non-reducing conditions. S-PAGE (NuPAGE® 4-12% bis-tris gel / Invitrogen / cat# NP0323BOX) The proteins in the gel were analyzed using a colloidal blue reagent (PageBlue™ protein). The specimen was stained with cereal staining solution / Fermentas / cat# RO571.

[0203] Example 7: Enzymatic deglycosylation of IgG1 Because IgG glycosylation is heterogeneous, it is suitable for mass spectrometry analysis. Deglycosylation was performed to produce a single product with a distinct mass. One unit of N-glycosidase F (PNGase F; Roche Diagnostics) was used for 0 μg of IgG1. The cells were then incubated overnight at 37°C in the original purification buffer ( To remove the residue, add 0.1M citrate buffer pH 3.0 / 1.0M Tris-HCl pH 8.0. Buffer exchange was performed using a 0 kDa MWCO centrifugal filter column (Millipore). The lysate was then re-buffered in PBS. A phosphate exchange procedure was performed and the solution was exchanged into 150 mM ammonium acetate at pH 7.5. The reaction was carried out at 11,000 rpm for 12 minutes with 200 μl of 150 mM ammonium acetate, pH 7.5. After rinsing, 50 μl of deglycosylated IgG was added to the wells. The filter was dosed and 450 μl of 150 mM ammonium acetate, pH 7.5, was added. Then, the mixture was centrifuged again for 12 minutes at 11,000 rpm and 4°C. Add 150 mM fresh pH 7.5 ammonium acetate buffer to a total volume of 500 μl. After the final centrifugation step, buffer was added and the centrifugation was repeated five times. Approximately 25 μl of the remaining deglycosylated IgG1 was recovered after ELISA. It was transferred to an Eppendorf tube and prepared for mass spectrometry analysis.

[0204] Example 8: Native Mass Analysis Identify the different IgG species in the purified IgG mixture and in what ratios these I Mass spectrometry is used to determine if any gG species are present. 2-3 μl of 150 mM ammonium acetate, pH 7.5, containing 1 μM IgG Gold-plated borosilicate capillary tubes (Sutter P-97 puller [Sutter Instruments Co., No. vato, CA, USA] and Edwards Scancoat six sputtering equipment [Edwards Laboratories, M ilpitas, CA, USA) and optimized for high-mass detection (Tahallah et al., RCM 2001). The analysis was performed on an LCT1 mass spectrometer (Waters Corp., Milford, MA, USA) tuned to optimize A capillary voltage of 1300 V and a sampling cone voltage of 200 V were used. However, these settings require a higher resolution signal-to-noise ratio. The source backing pressure was adjusted if necessary to promote collisional cooling. The pressure (assure) was increased to approximately 7.5 mbar. The protein was sprayed at a concentration of 1 μM in 5% formic acid.

[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 was used and the spectrum was centered. Using each charge state in the series, the mass of the species was calculated. For each state, the corresponding intensities were assigned by MassLynx and summed. allows for relative quantification for all species in a sample. Alternatively, the known area under the curve Peak quantification can also be performed using area-under-the-curve (AUC) methods. All analyses were repeated three times to calculate the standard deviation of the IgG mass and its relative abundance. It was.

[0206] Example 10: Mixture of two or three monospecific antibodies from a single cell Several antibodies with known specificity and known ability to bind human IGKV1-39 light chain. The VH region of the wild-type construct vector MV1057 or the constructs in Table 1 (Figure 3) was Construct 4 or 5, resulting in vector The resulting vectors, vectors I-III (Table 2), had different CH3 regions and different VH A vector containing nucleic acid sequences encoding a specific Ig heavy chain and a common human light chain, The transfectants I, II, and III were then transfected into cells. Infections can be performed singly to detect the formation of intact monospecific antibodies, or Alternatively, one or two other constellations may be used to obtain a mixture of two or three monospecific antibodies. Table 3 shows the transfection results. be.

[0207] [Table 2]

[0208] [Table 3]

[0209] Transfections A, G, and H allow the transfer of either vector I, II, or III. From cells transfected with either one, bivalent monospecific mAbs were obtained, which only formed homodimers. The AA, BB, or CC clones were obtained (Figure 4). This point has been predicted and proven before. Construct 4 with amino acid substitutions (i.e., K392D, D399K, K409D) or quadruple amino acids Construct 5 (i.e., E356K, E357K, K439D, K370D) containing CH3 was engineered. Homodimerization of the induced Ig heavy chain is reported (transfections G and H).

[0210] Next, co-expression experiments of the two vectors in single cells were performed. Interestingly, as shown by transfections M and N, the wild-type and C Co-expression of H3-engineered Ig heavy chains with common light chains in a single cell resulted in undesired There are no bispecific antibodies present, and only 4-5% of contaminating "other molecules" are present in the mixture. The presence of the "other molecules" alone resulted in a mixture of two monospecific antibodies. It is defined as any molecule that does not have the mass of an IgG and is composed of a single heavy and light chain pair. Importantly, the "other" category does not include bispecific products. is.

[0211] In transfection M, vector DNA was transfected in equal proportions. In transfection N, the ratio of AA:BB was almost 1:1. The A:CC ratio was approximately 10:1. Therefore, this transfection was DN The ratio of A was adjusted and the transfection was repeated (transfection U). When III was 1:5, the ratio of antibody products AA:CC in the mixture was approximately 1:1. Therefore, as shown by transfections M and U, unwanted by-products absence of AC or the presence of large amounts of half molecules A or C in single cells , two different, substantially pure, monospecific antibodies can be expressed (Figure 5 ) Heterodimer formation between wild type and construct 4 or wild type and construct 5 was observed. To prevent this, the novel CH3 modifications in constructs 4 and 5 are similar to the wild-type CH3. This is a substantial difference in application to the mass production of mixtures of monospecific antibodies from single cells. is advantageous.

[0212] Similar to these results, but with two different CH3-engineered Ig heavy chains (constructs 4 and 5) were obtained from two different monospecific antibodies without the presence of additional undesired species. The resulting mixture is expected to be a mixture of CH3 and CH4. If the CH3 modifications of construct 5 and construct 6 are substantially different, heterodimerization will not occur. In that case, the CH3 engineered heavy chains of constructs 4 and 5 were expressed in single cells. When co-expressed with wild-type CH3 heavy chain in IgG1, only three monospecific antibodies were obtained. It will be possible.

[0213] Indeed, this result was observed. When the three different Ig heavy chains were expressed in a single cell together with the common light chain, the mixture It was found that a mixture of three pure monospecific antibodies was obtained without any other contamination (transformation). Transfection 0) (Figure 6). As can be clearly seen from Table 3, transfection 0 Even if the vector DNA is in the same proportion, the antibody AA:BB:CC ratio is 1:1. :1. When transfected with different ratios of vector DNA ( 1:1:10, transfection V), the ratio of AA:BB:CC in the mixture is desirable From the above, as shown by these experiments, , two or three substantially pure monospecific antibodies in a single cell, without unwanted by-products This is advantageous for the large-scale production of mixtures of monospecific antibodies for therapeutic use.

[0214] Example 11: Mixture of two bispecific antibodies from a single cell The production of a single bispecific antibody with a CH3 engineered heavy chain has been reported elsewhere. Here, this experiment demonstrates the production of a mixture of two different bispecific antibodies from a single cell. It was designed to explore whether this was possible. VH regions of antibodies with known specificity and known binding ability to the human IGKV1-39 light chain The region (Figure 3) was recloned into vectors containing constructs 1-3 or 6-7 in Table 1. This resulted in vectors IV-X (Table 4). Vector I contains nucleic acid sequences encoding the Ig heavy chain and the common human light chain, respectively. VX was then transfected into the cells. It can be done singly to show the prevention of the formation of intact monospecific antibodies, or bispecific antibodies. Other construct vectors can be used to obtain heterologous antibodies or mixtures of two bispecific antibodies. Table 5 shows the transfection results.

[0215] [Table 4]

[0216] [Table 5]

[0217] The CH3-engineered Ig heavy chains encoded by constructs 1 and 2 were expressed in single cells. It has been shown that when expressed, it maintains the ability to form homodimers (WO 2014 / 013366). However, further reports in WO 2009 / 089004 In construct 3, a triple charge pair mutation was engineered. The CH3 domain, when expressed alone, can no longer form homodimers. Studies have only partially confirmed these findings. In sections B, C, and D, the presence of full-length IgG in addition to a high proportion of unpaired half molecules This indicates that the CH3 domains encoded by constructs 1 and 2 are homologous. Transfections E and F also contain unpaired half molecules, indicating dimerization. As a result, full-length IgG was produced. This is due to the triple charge of construct 3. This indicates that the mutation does not completely impair homodimerization. It was further shown that the "knob" and "hole" CH3 mutants of 1 and 7 also form homodimers. (18% for "knob-knob" homodimers and 42% for "hole-hole" homodimers) %).

[0218] When a second CH3 mutant was co-expressed for heterodimer formation, the unwanted by-product ( Homodimerization is completely inhibited when expressed alone, as it inhibits or minimizes homodimerization. A CH3 mutant that is impaired is preferred.

[0219] Interestingly, this experiment was the first to show that the mixture was essentially free of homodimers and contained bispecific We demonstrated that a mixture of antibodies can be expressed in a single cell. In the case of L and L, the predicted bispecific species BC+AB was indeed obtained (transfer 38% + 47% for transfection K and 16% + 60% for transfection L. A relatively high proportion of undesired half molecules was observed in the transfection (transfection In transfection K, 15% were half molecule A + half molecule C, and in transfection L, 24% were half molecule A + Half-molecules C). The half-molecules still present in relatively high proportions are those with an imbalance of matched heavy chains. This is due to the low amount of heavy chain in Vector IV due to the low expression. In transfection S and T, the ratio of vector DNA was adjusted to 2:1:1. As a result, equal amounts of IgG heavy chains from the matched pair were obtained. There was no presence of IgG half molecules and only 3% of homodimeric BB, resulting in a pure dimer. A mixture of polyspecific IgG was obtained. Ideally, this low concentration of monospecific product contaminants would be achieved. The proportion of contaminating monospecific antibodies should be reduced to virtually zero. It would be desirable to find additional CH3 variants that would result in mixtures of bispecific antibodies that reduce the do.

[0220] This study is the first to demonstrate the ability to detect single-cell specific antibodies while minimizing the presence of monospecific antibodies in the mixture. A substantially pure mixture of two bispecific antibodies that recognize three different target epitopes. It showed that things can be produced.

[0221] Example 12: Various Mixtures Production of a mixture of two bispecific antibodies recognizing three epitopes from a single cell, or The production of a mixture of two or three monospecific antibodies from a single cell is technically feasible. We next explore the feasibility of controlled production of various other mixtures. The possibilities were explored.

[0222] A fourth antibody VH with known specificity and known binding ability to the human IGKV1-39 light chain. The region was used to reclone into a vector containing constructs 1-3 or 7 in Table 1. The resulting vectors are I', II', III', or X' (' represents the corresponding vector The different CH3 regions and different V A vector containing nucleic acid sequences encoding an Ig heavy chain with H specificity and a common human light chain, respectively. The nucleotides I'-III', X' and IV-IX were then transfected into cells. The transfection was carried out using various bispecific and / or monospecific antibodies. It has been combined with other construct vectors to obtain mixtures. The mixtures consisted of two bispecific antibodies derived from a single cell that recognized four epitopes. A mixture of antibodies, two bispecific antibodies and one monospecific antibody, or one bispecific antibody The transfection time points include a mixture of two or more monospecific antibodies. Here is a table and predicted results.

[0223] [Table 6]

[0224] Although the production of all mixtures is theoretically feasible, other studies to date have shown that Previously, large-scale production of knob-into-hole mutants was hampered by stability issues. Therefore, transfections ZA, ZB, ZL, ZM and ZN The resulting mixture is predicted to be problematic when transferred to mass production.

[0225] In other words, the set of constructs in Table 1 can all be mass-produced from a single cell. The reason is that knob-into The hole mutant is reported to be unstable and has a "knob" or "hole" The H3 domain dimerizes with either charge mutant or with the wild-type CH3 domain Therefore, for co-expression in a single cell, homodimers or The constructs selectively form only homodimers or heterodimers with constructs 1-5 in Table 1. It would be desirable to design new CH3 mutants that are engineered to not dimerize.

[0226] Example 13: Identification of novel charge pair mutations The purpose of this study was to investigate the effect of expressing a mixture of different IgG heavy chains in a single cell. The CH3 region of IgG is then cleaved to result in the production of only heterodimers or only homodimers. Here, the novel engineered CH3 domain is compared with the known engineered CH3 domain. It does not homodimerize or heterodimerize with the CH3 domain or the wild-type CH3 domain. Therefore, the most promising approach to identify novel engineered CH3 domains that meet this criterion is to As a first step, we isolated the many contact residues at the interface of the IgG CH3 domain one by one. or as a group, repulsion between identical heavy chains due to electrostatic interactions - i.e., formation of homodimers The aim was to investigate whether these mutations affect the activity of different IgG heavy chains. Used to induce the formation of homo- and / or heterodimers when expressed in combination with other To obtain a list of residues that, when substituted with charged residues, cause repulsion between identical chains, This makes the resulting full-length IgG stable and allows it to be produced at a high rate.

[0227] Further investigations have been carried out to identify matched pairs of CH3 residues in one or more IgG heavy chain-CH3 regions. By engineering, bispecific antibodies or mixtures of bispecific or monospecific antibodies The identified mutations are used to purify the antibody. The newly identified charge-dependent The mutation pair is combined with an existing pair and used for expression in cells. Mixtures of only heterospecific or bispecific antibodies, or defined monospecific and A mixture of bispecific antibodies is selectively obtained. The residues tested in this study have not previously been These are the contact residues identified (Deisenhofer J., 1981; Miller S., 1990; Padlan, 1996; Gun asekaran, 2010). The rationale for this approach is that the repulsive charges of each effective pair The reason is that it is introduced into contact residues.

[0228] The samples were then analyzed by SDS-PAGE under non-reducing conditions, revealing a band of approximately 72 kD. By looking at the results, pairs that reduce dimer formation are identified. These were screened for single mutations or in combination with other single mutations. The reason is that the repulsive electrostatic interaction of a single mismatched pair is not sufficient for detection by this method. These mutations are recombinogenic because it is unclear whether they are sufficient to obtain sufficient half-molecules. It is also used in combination.

[0229] Amino acid substitutions were introduced into construct vector MV1057 using Geneart according to Table 7. HEK293T cells were also transfected according to standard procedures. The expression level of the IgG was measured by Octet. If production fails twice, the mutation is considered to be an inhibitor of expression, and the mutation No further investigation was carried out.

[0230] [Table 7]

[0231] The supernatant containing ≥5 μg / ml IgG was analyzed by SDS-PAGE and purified using Protein A. The protein was stained with colloidal blue reagent. The homodimer was approximately 150 A smaller band of approximately 75 kD was observed, indicating the presence of half a molecule. (See negative controls: K392D, K409D). is shown in Figure 7.

[0232] The results of SDS-PAGE were analyzed and scored, and are shown in the rightmost column of Table 7. Contains S354, Y349, L351, K360, T366, T394, and V397 Many residues are promising and should be tested in combination for further study. The selection of ATP was based on a high score in inhibiting homodimer formation and its association with other non-complementary charges. We considered both the availability of contact residues that could be modified so that they would not be problematic. For example, F4 The Y405 and Y407 residues are already involved in CH3-CH interactions, including interactions with charged residues. It is known that there are multiple interactions at the 3 contact surface. (see Table A) introducing multiple charge mutations can be problematic.

[0233] To test further combinatorial mutations, new constructs were added to vector MV105. 7 (Table 8), with known specificity and known binding ability to the human IGKV1-39 light chain. The antibody VH region containing the nucleotide sequence was then reassembled into vectors containing these new constructs (see Table 9). Table 10 shows the transfection summary and results.

[0234] [Table 8]

[0235] [Table 9]

[0236] [Table 10]

[0237] Combinations of CH3 mutants were expressed and analyzed by SDS-PAGE (data not shown) and naive analysis. The results are summarized in Table 10. The AC produced the highest percentage of heterodimers in the mixture (69% AC). Profoundly, in ZO transfection, AA homodimers are absent, whereas C Mass spectrometry revealed that the remaining proteins in the mixture contained a small proportion (7%) of C homodimers. It consists of half A molecules and is thought to be the result of unbalanced expression of A and C heavy chains. The raw MS data from the transfection sample ZO are shown in Figure 8. Surprisingly, The reaction ZO produced a significant amount of bispecific product, but the opposite charge pair Transfection of ZP (L351K / T366'D vs. ZO (T366K / L351'D) resulted in similar results. No results were obtained, with only 52% bispecific product observed, and a significant amount of the two Homodimers were present (30% AA and 13% CC). Since D is structurally very similar to T, T366D is strong enough to repel itself. Therefore, it was explained that T366D can still form homodimers, and this was indeed observed.

[0238] The newly discovered T366K / L351D pair of minor variants (e.g., new constructs) (by testing all variants including T366R and L351E) It is expected that a proportion of bispecific antibodies (BsAbs) can be obtained as a result.

[0239] Example 14: Novel C-terminal fragments for inducing efficient heterodimerization by HADDOCK H3 mutation design) As explained in Example 13, the newly discovered charge pair T366K / L351'D was found to be present in the mixture. Increased the proportion of heterodimers in the chromatin (69%), while also reducing the proportion of undesired C A mixture of C homodimers (7%) (L351D / L351'D) and a significant proportion of half molecules A (24%) In this example, we used an in silico approach to detect CH3 Further insights were gained into the amino acid residues involved in the interaction at the CH3 region. Complementary substitutions that inhibit efficient homodimer formation of the two heavy chains were tested, while Discover novel CH3 pairs containing complementary substitutions that further increase efficient heterodimerization .

[0240] HADDOCK (High Ambiguity Driven protein-protein DOCKing) Protein-protein docking (protein-guided docking) is a novel method for modeling biomolecular complexes. HADDOCK is a flexible docking approach that utilizes information. Unlike the docking method of the previous one, ambiguous interaction restraints :AIRs) encodes information on the protein contact surfaces identified or predicted in the The input to the HADDOCK web server is The data can be generated from protein structure files such as crystal structures, NMR structure clusters, or structural models. After docking or fine-tuning, HADDOCK generates a so-called HADDOCK score. The HADDOCK score is calculated by the van der Waals energy, the electrostatic energy, and the The HADDOCK score is a weighted average of the surface area and desolvation energy used to calculate the HADDOCK score. Although direct translation to experimental data is often difficult, binding energy or affinity metrics and solutions are often used. In addition, HADDOCK identifies the top four structures from the results of the docking calculations. These structure files can be downloaded and visualized. This allows for detailed analysis of the interactions of individual residues.

[0241] In this example, the interactions between the CH3 domains of the IgG1 heavy chain are studied. The Fc region of IgG (structure 1L6X), which has a crystal structure of 1000, is used as the starting structure. (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 homozygotes. It was found to result in the formation of contaminating dimers (Table 10). CK was used to explore additional mutations in addition to T366K / L351D that disrupt homodimerization. do.

[0243] The output of the HADDOCK score is the calculated energy and the HADDOCK score ( (weighted average of several energies) and the four lowest energy values ​​found by the program. The HADDOCK scores are calculated for different structures. Other energies are used to compare what is happening in the structure. targets (e.g., better electrostatic interactions, less covered surface, higher van der Waals The lower the HADDOCK score, the 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 in HADDOCK and the experimental data and calculated We investigated whether the energies were correlated. Table 11 shows all the theoretical energies. 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 identical. In most cases, the HADDOCK score is the same for AA, AB, and BB. As expected, the AB pair had the lowest score. In this study, the BB score was slightly better than the AB score (-210.6 vs. -212.5). The difference is within the error range of the calculation. The structure of the body is visualized. For example, the combinations 1-2, 1-1, and 2-2 are shown in Figure 1. These visualizations reveal that salt bridges are formed in the heterodimer. (Fig. 10A, left panel), and electrostatic repulsion occurs between residues of the same chain ( (Figure 10B and C, center and right panels). Higher HADDOCK relative to homodimers. The score is explained by the electrostatic repulsion of the mutated contact residues. These residues bend relative to each other. The residues on the other chains are avoided and do not interact with each other, reducing affinity.

[0247] Table 11 and Figure 9 confirm the observations in Example 13. The AC heterodimer and the CC homodimer of L351D / L351'D have similar energies and are mixed. While this paper describes the existence of both heterodimers and homodimers in the compound, The T366K / T366'K AA homodimer was barely detectable in the mixture, and the T366K half molecule A was present. Table 11 and Figure 9 show that the HADDOCK score of the AA homodimer of T366K / T366'K is , which indicates that the score is higher than that of the AC heterodimer. is energetically less favorable.

[0248] (Example 15: 366 / 351 Mutation) In Example 13, the T366K / L351'D mutation charge pair was also mixed by a method other than the above. The hypothesis is that the proportion of bispecific antibodies in a product can be designed to achieve similar results. The other method is T366R, T366D, T366E, L351E, L3 The proportion of L351D / L351'D CC homodimers is 3. This can be reduced by generating 66 / 351 pairs of mutations. All possible pairs of mutations are HAD It was performed in 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 were significantly higher than T366K / L351D It was observed that the AA homodimers in many variants have similar "patterns". The dimer was found to have a higher HADDOCK score than the AB heterodimer, but the B The B homodimer was equally favored as the AB heterodimer. Residue 351 was the same as the other chain. residues in chain A are known to be "adjacent" to the CH3-CH3 residues. It pairs with residue 351 of chain B at the interface. When the BB dimer is formed, it has the same charge. There is almost no negative effect of asparagine. The acids bend and avoid each other, at least from the naturally occurring arginine at position 355. These have a stabilizing effect, and the negative charge is further stabilized by the naturally occurring serine at position 354. Mutation of these residues (S354A and R355D) results in almost no change in the ATP-dependent cleavage pathway (see Figure 12A). There is no improvement. Figure 12B shows that the main chain hydrogen of A354 is responsible for stabilizing the homodimer. From this series, it is clear that the T366R / L351E pair is responsible for the formation of bispecific molecules. The nucleotide sequence with the lowest HADDOCK score for the nucleotide sequence is considered to be the most suitable.

[0251] Example 16: Mutations around T366K / L351'D In this example, a series of HADDOCK analyses showed that T366K / L351D or T366K / L351 The starting structure was the pair A and B. The predicted proportion of these A and B chains with dual specificity was further To identify additional mutations that could further improve the HADDOCK score and The CH3 domain structure was visualized using the molecular-level protein structure visualization method. When examined using a viewer for the chromatin analysis (YASARA, www.yasara.org), the distances between individual residues were During the study, two residues, Y349 and L368, were found to be in the dimer phase. It was observed that adjacent residues could contribute positively or negatively to the interaction. These mutations—in addition to the L351 mutation—result in the formation of homo- and heterodimers. The effect of both residues on the heterodimer formation was investigated (see Figure 13). destabilizes the BB dimer (higher HADDOCK score) core). Glutamic acid (E) is preferred over aspartic acid (D) at positions 349 and 368. Therefore, the second amino acid substitution in the B chain, which already has an amino acid substitution at position 351, It was suggested that the introduction of acidic mutations would favor heterodimerization.

[0252] In the next set of HADDOCK analyses, the T366K / L351D pair was again used as the starting structure. Substitutions in the B chain (i.e., Y349D / E and An additional mutation was made to the A chain, which already contained the substitution T366K (L368E and L368E). As such, there are several mutations that appear to favor the formation of dual-specific heterodimers. In the T366K-L351K / L351'D-Y349'D pair, all four mutated residues are involved in the heterodimer pair. This suggests that K351 is not directly involved in binding, whereas T366K-L351K / L3 This is not the case for 51'E-L368'E. However, the HADDOCK score of the latter heterodimer The value of T366K / L351'E-L368'E is -228.9, which is significantly lower than the value of T366K / L351'E-L368'E, which is -214.2. This is explained by hydrogen bonding interactions at K at position 351 (see Figure 15). The T366K-L351K / L351'D-Y349'D pair can be further improved by the R355'D mutation in the B chain. This will increase the HADDOCK score of BB, but also the HADDOCK score of AB. Therefore, the additional L mutations are significantly higher than the single T366K mutation in the A chain. 351K lowers the AB score and does not significantly change the AA and BB scores. The higher the amount of bispecific heterodimer in the sample, the higher the amount of bispecific heterodimer obtained.

[0253] As is clear from Figure 11, changing position 366 to R rather than K induces heterodimerization. Therefore, this time, T366R is used instead of T366K in the A chain, Several HADDOCK analyses shown in Figure 13 were repeated. It was shown that combining mutations is unfavorable (Figure 16). Even if the structure contains all the salt bridges with R366, the other contact surface interactions Furthermore, the HADDOCK score of the AA homodimer is higher for R366 than for K366. This also contributes unfavorably to the formation of heterodimers. Further HADDOCK analysis using R366 was discontinued.

[0254] A total of 14 best-performing pairs were selected from the HADDOCK predictions ( (See Table 13 and Figure 17). In some pairs, spontaneous interactions with L351 / L351D The R355D substitution is included to remove the stabilizing effect of R355.

[0255] [Table 13]

[0256] Example 17: In Vitro Synthesis of Bispecific Molecules Using CH3 Mutations Based on HADDOCK Predictions ro expression) As suggested by the analysis in Example 16, additional mutations around the T366K / L351'D pair Some CH3 mutants with a higher proportion of bispecific components and a lower proportion of homodimers These optimally performing pairs are then used for production. Constructs T366R and L351E were also generated. The constructed constructs and their known specificity and known human IGKV1-39 light receptors were used. The list of constructs used to re-clone antibody VH regions capable of binding to the IgG chains is shown in Table 1. 14.

[0257] The expression levels of IgG containing each construct were reported in Example 13 above, and are shown in Table 1. The same procedure was repeated for the constructs listed in 4. The aim was to to assess which constructs homodimerize in the absence of a compatible partner. Ideally, a high proportion of half molecules and a low proportion of homodimers should be formed. The group of constructs includes constructs containing previously reported charge mutations and constructs containing previously reported charge mutations. Constructs containing knob-in-hole mutations were also used to express whole IgG in recombinant cells. The supernatant purified with Protein A was analyzed by SDS-PAGE, and the results were The results are shown in Table 14 along with the scores.

[0258] [Table 14]

[0259] A common light chain and two different constructs were constructed that retained the amino acid substitutions shown in Table 14. Results of co-expression with heavy chains or heavy chains that retain the amino acid substitutions of previous constructs The two nucleotide sequences with the amino acid substitutions T366K and L351'D:L368'E, respectively, are listed in Table 15. The expression of different heavy chains resulted in approximately 87% homodimers in the mixture, with no AA or BB homodimers present. The bispecific heterodimer AB could be obtained (combination number 3 in Table 15). In addition, half molecules containing an additional amino acid sequence in the first heavy chain (half molecule A) were observed. The introduction of the amino acid substitution L351K increased the proportion of bispecific heterodimers AB. For example, the amino acid substitutions T366K:L351K and L351'D:L368'E were found. Co-expression of two different heavy chains resulted in approximately 92% bispecific heterodimer AB, On the other hand, AA and BB homodimers were virtually absent from the mixture (combinations in Table 15). (Number 12).

[0260] The combination of 10 and 11 also resulted in a high percentage of heterodimers and virtually no homodimers. The absence of homodimers is due to the presence of inter- This is advantageous because the fraction containing the intact IgG molecules is composed exclusively of the heterodimer AB. For further therapeutic applications, half the molecules are purified using standard techniques such as size exclusion chromatography. known charge variants and knob-into-hole variants. The body does not eliminate "contaminating" homodimeric antibodies. It would be advantageous to apply the newly identified charge variants of the present invention to the production process of bispecific antibodies.

[0261] In addition, the T366K / L351'D:L368'E and T366K:L351K / L351'D:L368'E charge pairs are E356K:D399K / K392'D:K409'D and E356K:D399K / K392'D:K409'D:K43 described so far It has an additional advantage over the 9'D charge-reversal pair: the charge variants described so far is based on the reversal of the original charge within the CH3-CH3 interface, while the newly identified charge The charge mutant adds an additional charge pair (charge-charge interaction) to the CH3-CH3 interface. The introduction of an additional charge pair at the CH3-CH3 interface further stabilizes the interface. Combination No. 4 The same is true for the mutations used in 1, 5, 6, 9, 10, and 11. However, the AA and BB homodimers are present in very small proportions in the mixture. , leading to a favorable proportion of bispecific heterodimers.

[0262] [Table 15]

[0263] (Native MS) Native MS was performed on all bispecific samples. The resulting graphs were analyzed. The relative ratios of the species of interest were derived in two ways: peak height and peak area. Although the area-based method is a more scientifically correct analysis method, other studies have Since all analyses were done by peak height, both methods were included in the analysis for comparison purposes. The difference between the two methods was within the measurement error range, so the following measurements were Only the peak area values ​​were used.

[0264] Two typical spectra are shown in Figure 18. A graphical summary of the results is shown in Figure 19. The values ​​can be found in Table 15. In approximately half of the samples, the total monospecific IgG The contamination level was less than 5% and exceeded 10% in only three cases. Approximately 50% monospecific IgG is expected to be found in the mixture.

[0265] Ten combinations of two different heavy chains were selected from Table 15 for further analysis. These 10 combinations include combinations 1, 2, 3, 4, 5, 6, 9, 10, 11, and These 10 selections were based on the nMS-derived mixtures. Based on the low percentage of homodimers present, as well as the production yield, SDS-PAGE, The results were also based on overall physicochemical properties, including the number of mutations in the CH3 domain.

[0266] Example 18: Analysis of IgG stability In this study, we found a high proportion of bispecific heterodimers in the intact IgG fraction and very For a series of CH3 mutation pairs where low amounts (<5%) of parental IgG were obtained, Furthermore, the stability of the Fc region of the IgG molecule will be analyzed. The introduced mutated CH3 domain has an unexpected destabilizing effect on the Fc region of IgG. This can result in reduced in vivo half-life, reduced effector function, and / or This can result in undesirable properties such as increased immunogenicity or

[0267] The newly identified charge pairs were compared with wild-type bispecific molecules and previously identified The results were compared to bispecific molecules containing modified charge mutations (chain A and B with construct 1). All bispecific molecules in this study had the same heavy chain and construct 2. Since the antibody contains the variable regions of both the Fc and light chains, the observed effects are due to mutations in the Fc portion of the molecule, not the variable regions. It is guaranteed that this is not due to differences in the region.

[0268] A series of stability studies will be conducted on these bispecific molecules. These studies include: , spectroscopic (UV- Visible light absorption, fluorescence and light scattering) and microscopic (optical and fluorescent staining with Nile Red) This includes analysis using a light microscope.

[0269] UV-Vis absorption spectra were obtained using a Cary 300 Biomicroscope with two monochromators and a double beam. Spectra are recorded at 25°C on a spectrophotometer. The spectra are measured from 250 to 400 nm with a 1 cm path length. The amount of absorbance at wavelengths of 320 nm and longer is monitored between 100 and 1500 nm. Status information is provided.

[0270] The intrinsic fluorescence spectrum is monitored at 25°C using a FluoroMax fluorometer. The optical method is optimized accordingly. Fluorescence emission provides information about conformational and aggregation properties. Provide all the information.

[0271] 90° light scattering spectra were obtained using a FluoroMax fluorometer. ) for 0.01 seconds, and a synchronous scan (λ em =λ ex ) The excitation and emission slits are optimized accordingly. For example, Light scattering at right angles can distinguish the presence of 5% dimers in an IgG sample.

[0272] For fluorescence microscopy with Nile red staining, Nile red in ethanol was added immediately before measurement. Red is added to the sample, which is then loaded onto a microscope slide and analyzed by fluorescence microscopy. Particles are counted, but the size of the smallest particle observed by fluorescence microscopy is approximately 0. 0.5μm.

[0273] Stresses on proteins such as temperature, pH, mechanical stress or denaturants can cause conflict This can lead to conformational changes (e.g., unfolding) and / or aggregation. Previously, charge-engineered bispecific antibodies have been shown to have altered CH3 molecules with lower melting temperatures. It has been reported that (Gunasekaran 2010). Therefore, these studies are based on the novel electrochemical The aim is to distinguish between charge variants and variants of charges that already exist and are known.

[0274] Thermostability with Octet using Protein A biosensor and FcRn to IgG To investigate the thermal stability of CH3-engineered IgG, a PCR machine was used. The samples were incubated at 4, 50, 55, 60, 65, 70 and 75°C for 1 hour at 100 μg / ml The sample is then slowly cooled and incubated at a concentration of 0.01% (PBS as the solvent). The temperature was raised to 25°C in 15 minutes, and the mixture was kept at this temperature for 2 hours, after which it was stored at 4°C until the next day. The remaining antibodies were removed by centrifugation, and the total IgG concentration of water-soluble antibodies was measured using Octet Protein A bioassay. Derived by Octet using Osensor (1 / 10 PBS dilution).

[0275] An assay to measure the binding of CH3-engineered IgG to FcRn was developed using Octet. After binding the IgG light chain to the protein L biosensor, , and incubate with FcRn in solution. Alternatively, use Anti-Penta-HI. The S biosensor binds to the His-tagged FcRn protein and detects the target IgG. These methods are more sensitive than the Protein A biosensor. may be used for thermal stability studies.

[0276] All samples were also subjected to serum stability analysis. ) IgG samples were incubated in human serum at 37°C, and control samples were incubated at 4°C. After 1, 2, 3, and 4 weeks, the samples were centrifuged to remove precipitated IgG. The samples are then added to an antigen-specific ELISA to determine the relative amount of functional IgG. A purified control antibody was acutely injected into human serum and used as a reference group. can be done.

[0277] Example 19: Stability analysis Previous experiments demonstrated that co-expression of two different heavy chains with CH3 mutations and a common light chain resulted in As a result, a high percentage of bispecific antibodies was obtained (Example 17).

[0278] Eight combinations of two different heavy chains were selected from Table 15 for further analysis. These eight combinations are combinations 3, 4, 5, 6, 9, 10, 11 and 12. In this study, we focused on the stability of the Fc region of IgG and compared these 8 The control group consisted of wild-type bispecific molecules (i.e. , no CH3 mutations) and / or includes previously reported CH3 charge mutations. Of note, wild-type bispecific molecules do not selectively induce heterodimerization. Without any means, the two heavy chains and the common light chain were co-expressed. The term "type of bispecific molecule" refers to a mixture of AA, AB, and BB. All bispecific molecules in this study Because specific molecules retain the same heavy and light chain combination, the observed effect is due to the It is guaranteed that the mutations are due to the Fc region of the antibody and not due to changes in the Fab region. .

[0279] It is hypothesized that the mutations used to promote heterodimer pairing of two different heavy chains The heterologous pairing may be associated with unexpected structural or other destabilizing effects on the Fc region of IgG. This could be linked to the reduced half-life in vivo due to the presence of these mutations. Further clinical consequences, such as decreased effector function and / or increased immunogenicity, may be present. This could result in undesirable problems that could impede development.

[0280] (thermal stability) When stress such as temperature increase or decrease is applied, the conformation of the protein changes. CH3 is manipulated to cause unfolding and / or aggregation. To investigate the thermal stability of the IgGs, combinations 3-6 and 9-12 (Table 15) and wild-type IgGs were used. Type 2 bispecific molecule and constructs 1 and 2 (E356K:D399K / K392D':K409D' combination) The bispecific molecules obtained using the nucleotides (also called "charge-reversed" pairs) were then subjected to PC Using an R instrument, the incubation times were 4, 60, 62.5, 65, 67.5, 70, and 72.5°C for 1 hour. The samples were incubated at a concentration of 100 μg / ml (PBS as the solvent). It is then cooled quickly to 25°C over 15 minutes, kept at this temperature for 2 hours, and then stored at 4°C until the next day. The precipitated antibody was removed by centrifugation (18,000 rpm; 4°C, 20 min). The total IgG concentration of the water-soluble antibody was measured by Octet using a Protein A biosensor (1 / 1 0 PBS dilution).

[0281] The results are shown in Figure 20. The control CH3-engineered bispecific antibody (charge-reversed The combination E356K:D399K / K392D':K409D' (triangles) was significantly higher than the wild-type bispecific molecule (squares). ) have decreased thermal stability compared to the combinations 3-6 and 9-12 (diamonds). The bispecific molecule also demonstrated decreased thermal stability compared to the wild type. It should be noted that in the three combinations, the control CH3 manipulated double specificity Combinations 9, 10, and 1 demonstrated improved stability compared to the isomeric antibodies. The bispecific molecule of 1 was significantly more stable than other CH3-engineered (charge-reversed) bispecific molecules. The temperature was constant and was as stable as the wild-type bispecific molecule at the highest temperature measurement point.

[0282] (Freeze-thaw stability) To investigate the stability of CH3-engineered IgGs under repeated freeze-thaw cycles, The bispecific molecules from sections 3-6 and 9-12 (Table 15) were compared with the wild-type bispecific molecules and Constructs 1 and 2 (E356K:D399K / K392D':K409D' combinations (charge-reversed pairs)) were used. The bispecific molecules obtained using a)) were subjected to 10 freeze-thaw cycles. The samples were placed at -80°C for at least 15 minutes until completely frozen, and then thawed at room temperature. Once completely thawed, the freeze-thaw cycle was repeated. After the lysis cycle, the precipitated antibodies were separated by centrifugation (18,000 rpm; 4°C, 20 min). The total IgG concentration of the water-soluble antibodies was measured by Octet using a Protein A biosensor (1 The freeze-thaw stability test was repeated three times.

[0283] The results are shown in Figure 21. As a control, a charge-reversed CH3-engineered bispecific antibody The antibody appeared to have slightly reduced stability compared to the wild-type bispecific molecule. In contrast, the bispecific molecules from combinations 3, 4, and 9 were significantly higher than the wild-type bispecific molecules. The stability of the product was slightly improved compared to the original product. The new situation does not raise any significant stability issues for CH3-engineered mutants. It was concluded that:

[0284] (in vitro serum stability) To examine the stability of CH3-engineered IgG in serum maintained at 37°C, Bispecific molecules from combinations 3-6 and 9-12 (Table 15) and wild-type bispecific molecules The charge-reversed bispecific molecules were incubated at 37°C in the presence of 10% human serum. Control samples were kept at 4°C. After 1, 2, or 5 days, the precipitated antibodies were centrifuged. The samples were then added to a fibrinogen-specific ELISA and functional The relative amounts of IgG were derived from the purified control antibody. and was used as the reference group.

[0285] Fibrinogen ELISA data shows that fibrinogen is soluble in 10% human serum at 37°C for 5 days. All samples were very stable even after 24 h. The lower Ig levels from combinations 4 and 5 were The G concentration of bispecific molecules is slightly less stable, especially at T=1 and T=2. However, at the end of this experiment the difference was minimal (Figure 22).

[0286] Example 20: Further stability testing A further series of analytical methods was used to assess the stability of the mutant IgG. The bispecific molecules from sections 3-6 and 9-12 (Table 15), the wild-type bispecific molecule (A A, AB, BB), individual parent antibodies (AA and BB) and constructs 1 and 2 (Double characteristic obtained using the combination of E356K:D399K / K392D':K409D' (charge-reversed pair)) The isomeric molecules were used as samples in these stability assays.

[0287] All IgGs were diluted to 0.2 mg / ml and tested under several stress conditions (50°C). The following cycles were applied: 2 days at 40°C, 2 weeks at 40°C, and 5 freeze-thaw cycles. It should be noted that these high stress levels are One of the parent antibodies used in all bispecific molecules (containing two 1122 Fabs) The BB parent becomes unstable. At 50°C for 2 days, this protein Aggregation of α-glucan was detected by UV absorption, suggesting that under these stress conditions Therefore, it may not be possible to distinguish between the instability of Fab and CH3 in bispecific molecules. Therefore, data from incubation at 50°C should be treated with caution. do.

[0288] A summary of the results is shown in Table 16. Analytical methods used included: - Fluorescence microscopy with Nile red ("Nile red particles" in Table 16); Nile red color After adding the base, observe the amount of particles larger than 0.5 μm. - UV spectroscopy at 350 nm ("UV350nm"); at wavelengths longer than 320 nm The resulting change in absorption provides information about the aggregation state of the protein. -90° light scattering at 400 nm ("LS400nm"); changes in protein aggregation, e.g., I This is a sensitive technique for observing the difference between the monomer and dimer forms of gG. - Autofluorescence: the fluorescence wavelength maximum and intensity of aromatic residues in proteins can be affected by the environment (e.g., (unfolding) changes. -1,8-ANS fluorescence spectroscopy; 1,8-ANS reacts with cations through electrostatic interactions via ion pair formation. Binds to the on group, resulting in changes in protein structure and / or conformation that can be detected .

[0289] (UV-Vis spectroscopy) UV-Vis absorption spectra were obtained using different Varian quartz cuvettes (e.g., with optical path lengths 1.0cm black low volume Hellma cuvette and 0.2cm x 1.0cm clear Hellma cuvette) at 25°C on a Cary 300 Bio spectrophotometer with a double beam and two monochromators. The spectrum between 220 and 450 nm was monitored using a 1.0 cm optical path length. The absorption around 280 nm provides information on the protein concentration. The region between 0 nm provides information about the assembly state of the sample.

[0290] (90° light scattering) 90° light scattering spectroscopy was developed to study protein aggregation. The 90° light scattering spectra were obtained by the Fluorescence Intensity Spectroscopy (FIS) method. An oMax fluorometer (Spex, Instruments SA, Inc. UK) was used to measure the integration time. A synchronous scan (λem = λex) from 400 nm to 750 nm was performed with a time (λem = λex) of 0.01 seconds. The experiment was carried out and monitored at 25°C. Different slit settings were tried to obtain optimal conditions. Once optimized, all measurements were performed using that slit setting.

[0291] (steady-state fluorescence emission) The fluorescence emission of tryptophan, tyrosine, and phenylalanine residues is It gives a picture of the local environment of the hydrophore. Generally, a more hydrophobic and rigid environment results in a higher fluorescence intensity. This leads to an increase in the fluorescence intensity and a blue shift in the emission maximum. Time-dependent state information is provided and changes in physical and chemical properties are monitored. More detailed information on tryptophan can be found in Lakowicz's book (Lakowicz, 2006). are.

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

[0293] (Fluorescence microscopy with Nile Red staining) Nile Red staining was developed to visualize protein aggregation. This was performed as described in Meule et al., 2007b.

[0294] Microscopic observations were performed using a Leica DM RXE microscope equipped with a mercury lamp (Leica Microsystems GmbH, The image was taken from a Sony NEX-5 camera and its firmware. The objective lenses were 10x, 20x, and 40x. In this study, the distance between the slide and the cover glass was fixed at 0.1 mm. The size of the tube is 1 mm x 1 mm, which corresponds to 0.1 μl.

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

[0296] 1,8-ANS is virtually non-fluorescent in water and does not bind to membranes (quantum yield ~0.25) or proteins (quantum yield ~0.25). It emits appreciable fluorescence only when bound to a 1,8- ANS is a protein stimulator that regulates protein folding, conformational changes, and other processes involved in this process. The exposure of the lobe to water is a sensitive indicator of the process of change. Information about 1,8-ANS can be found on the Probes website at www.probes.com. do.

[0297] The fluorescence emission spectra of 1,8-ANS were recorded using a FluoroMax fluorometer. A direct comparison of the fluorescence of 1,8-ANS in each IgG was not performed. This is because NS has a binding site, making it impossible to compare. In principle, the fluorescence of 1,8-ANS is small. Indeed, fewer 1,8-ANS molecules are bound to the antibody. Changes in light intensity and emission wavelength were evaluated.

[0298] [Table 16]

[0299] Taken together, these data demonstrate the remarkable stability of the various IgG preparations. Severe stress conditions were required to produce measurable differences between the specimens tested. Under these conditions, combination numbers 9 and Samples 10 and 11 appear to be more prone to aggregation than the other samples.

[0300] The biggest factors that can differentiate stability between proteins are freeze-thaw cycles and elevated temperatures. Considering the severe stress factor of incubation at 50°C, T366K / L3 51E, Y349E (combination number 4) and T366K, L351K / L351D, Y349E (combination number 11 ) are the most stable proteins in this group, followed closely by T3 66K, L351K / L351D, Y349D (combination number 10) and T366K, L351K / L351D, L368E (combination number The combination number 12) continues.

[0301] Example 21: Experiments for native MS using different ratios; Transfection ratio 1: 5 to 5:1) Behavior of CH3-mutated IgG in a mixture when transfected at biased ratios In order to deepen our knowledge about the We conducted more detailed experiments on the ratio of KK / DE or DEKK.

[0302] This has known specificity and the ability to bind to the known common light chain human IGKV1-39. The VH regions of the antibodies used up to now were recloned into constructs 1, 2, 68, and 69. This resulted in vector IV (Table 17). and nucleic acid sequences encoding Ig heavy chains with different antigen specificities and common human light chains, respectively. Vector IV, containing the vector IV, was transfected at different transfection ratios as shown in Table 18. The cells were then transfected with the vector. The results are shown in Figure 23.

[0303] [Table 17]

[0304] [Table 18]

[0305] As shown in Figures 23A and B, the DEKK combination mutations showed an excess of When A or C exists (A or C is "DE side" and B is "KK side"), AB or BC is formed, but in all cases excess amounts of A or C lead to homodimers and It exists as a mixture of both the full molecule and the half molecule. However, when an excess amount of B is present (B There is a clear difference between the two groups (A is the "KK side" and A or C is the "DE side"). However, the excess of B does not exist as a homodimer, and only half molecules are formed. will be done.

[0306] It should be noted that the percentage is again measured by peak height. Peaks detected below this value are considered to be accurate for the nMS technique used here. This is below the threshold that can be reliably measured. Therefore, measurements below 2% are considered to be noise in the analysis. I decided to ignore it, considering it within the level range.

[0307] It is noteworthy that when B is in excess, only the proportion of half molecules B increases. At ratios of 1:3 and 1:5, there was no homodimer BB and a high proportion of half The B molecule was observed (Figures 23A and 23B), and the mutation of CH3 on the KK side resulted in a homodimer. This indicates that the absence of homodimers is a crucial advantage. The reason is that when the "KK side" of the DEKK combination is chosen to add specificity, This is because known adverse effects may occur when it exists as a homodimer. (e.g., the cMET or CD3 antibody is present in the pharmaceutical composition as a bivalent homodimer. (It is known to have unwanted side effects when used in combination with other drugs.)

[0308] The observation of different DE:KK ratios is due to the control As shown in Figure 23C, the E356K:D399K / In the mutation combination K392D':K409D', when A is present in excess (A is "K39 In all cases, the excess amount of A was a mixture of both homodimers and half-molecules. Even when B exists in excess (B is on the "E356K:D399K side"), in all cases In this case, the excess B exists as a mixture of both homodimers and half molecules. At ratios of 1:3 and 1:5, homodimers are present but half molecule B is not observed. This suggests that the E356K:D399K side does not favor homodimers as much as the KK side of the DEKK combination. It was shown that there is no

[0309] In summary, the DEKK combination mutations cause one of the chains of the heterodimer to form a homodimer. This has a distinct advantage over charge-reversal CH3 mutations in that it does not produce

[0310] Example 22: Various mixtures using DEKK combinations DEKK combination mutations facilitate the formation of highly pure bispecific IgG molecules ("AB"). Next, we investigated the differentiation of "AB and AA" or "AB and AA" from a single cell. explored the feasibility of controlled production of more complex antibody mixtures, such as "AC" mixtures. The types of Fab used so far are "DE construct" or "KK construct". To demonstrate the versatility of the technique, Various combinations of these vectors were co-expressed to generate mixtures. 337 (tetanus toxin), MF1122 (fibrinogen), and MF1025 (thyrogens). globulin) have been shown to exhibit overall stable behavior, good expression levels, and They were selected due to the mass differences between the IgGs (see Table 19).

[0311] [Table 19]

[0312] [Table 20]

[0313] SDS-PAGE analysis showed that most samples consisted mostly of full-length IgG, with some In some cases, a low percentage of half molecules was present. Furthermore, many samples were gelled under non-reducing conditions. In the chromatogram, two bands were observed at approximately 150 kDa. This indicates that the sample contained two distinct bands. Even under reducing conditions, the gel showed a high percentage of IgG species. Two heavy chain bands were visible (data not shown).

[0314] Native MS was performed on all samples, and the proportions of observed species were calculated based on peak height. The results are shown in Figure 24. In all eight samples in which the heavy chain was co-expressed, two major peaks corresponding to the expected species were observed. Two of these samples (transfections 2 and 4) and transfection In transfection 11, a small amount of contaminating DE-DE homodimer was observed. The molecule was detected in very low 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. Not possible.

[0315] After nMS, it was found that in sample 11, the mass of the IgG corresponded to that observed This resulted in a transfection error. In sample 11, 1337-KK was used instead of 1122-KK. It was evident that it was co-expressed with 1025-DE.

[0316] To confirm the functional presence of the desired specificity, IgG samples were analyzed using a sandwich ELISA. Further testing was performed using ELISA probes with fibrinogen or thioglobulin. Detection was performed using fluorescein-labeled thioglobulin or The detection antigen was fluorescein (Pierce NHS-f) according to the manufacturer's instructions. The antibodies were labeled with fluorescein (fluorescein antibody labeling kit, cat. #53029). The isolated antigen was then immunofluorescently stained with FITC-conjugated anti-fluorescein antibody (Roche diagnostics, cat. # 11426346910).

[0317] The results of the dual specificity ELISA (OD450 values) are summarized in Table 21. Grey cells indicates the species expected for each transfection. In general, the experimental results were consistent with the expected results. Exceptions are shown in italics or bold. , seeds BC (transfections #1 and #2) or AC (transfections # 3) is a putative "negative" cell, but shows significant background signal. Previous studies have shown that bispecific ELISAs have high background levels. These background levels may be present in samples. Of note, the results of the bispecific ELISA indicated that It was confirmed that an error had occurred in transfection #11: species A instead of BC. C (values ​​in bold) were detected.

[0318] [Table 21]

[0319] Example 23: Recognizing four different epitopes (AB and CD) from a single cell Improved mixture of two bispecific antibodies In Example 12, the mixture from transfection ZA or ZB was transferred to large-scale production. There was a hypothesis that this would cause problems when Two-hole mutants are reported to be unstable and have a CH3 domain with either a "knob" or a "hole." This is because it cannot be excluded that cysteine ​​dimerizes with the charge-engineered CH3 domain. As shown in the examples above, the heterodimerization is selectively induced, and the formation of homodimers is substantially suppressed. Novel charge pair mutants were found. The polypeptide chain containing the H3 domain is a charge-engineered CH domain. It is expressed in cells together with three-domain polypeptide chains or SEED bodies, and two This can result in the selective formation of only bispecific molecules.

[0320] As is clear from the above examples, the DEKK combination mutations are essential for heavy chain dimerization in the CH3 domain. In clonal cells derived from the AB-1 antibody, one bispecific molecule (AB) or two It is excellent for producing bispecific molecules (AB and AC). However, it is not possible to produce complementary CH3 variants. Only one set of vectors can be used to differentiate the species of the mixture that can be produced. If there is a second "orthogonal" If there is a set of "orthogonal" vectors, "AB and CD" or "AB and CC" mixtures can be It is possible to produce more complex IgG and / or bispecific molecules such as do.

[0321] When combining two sets of vectors, an important requirement is that the C The heavy chain expressed from the H3 engineered vector does not form "crossover" dimers. "Crossover" refers to the process by which heavy chains produced by one set of vectors are transferred to other sets of vectors. The goal is to obtain full-length IgG by dimerization with the heavy chain expressed by the IgG1.

[0322] To test whether such "crossover" dimers could form, HADDOC K was used to measure the correlation 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 how pairing occurs. Between the wild-type CH3 domain and the CH3 domain containing E356K, D399K, or K392D, K409D mutations The pairing of the wild-type CH3 domain with the knob-into-hole mutation can occur. The possible pairings between the CH3 domain and the nucleotide sequence of the nucleotide sequence, including the nucleotide sequence of ... The combinations of CH3 mutations analyzed by HADDOCK are listed in Table 22. 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 combinations were CH3 with a charge-reversal combination and CH3 with a charge-reversal combination. This combination seems to reduce the amount of by-products (especially those that are contaminated when co-expressed in a single cell). There are no AC, AD, BC, or BD pairs, and the desired combination of two bispecific molecules (AB and CD) can be formed.

[0325] As can be seen in Figure 25, these undesired bispecific species AC, AD, BC and and BD have relatively high HADDOCK scores. On the other hand, the desirable AB and CD It has the lowest HADDOCK score. Of course, the DEKK or charge-reversed CH3 pair When either combination is inserted into a construct that retains the same specificity (e.g. , "C" on the DE side, "C" on the KK side, "A" on the E356K, D399K side and "B" on the E356K, D399K side " or "A" on the DE side, "B" on the KK side, E356K, D399 "C" on the K side and E356K, D39 When 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 that the correlations between AC and AD are similar. The HADDOCK score for DE is lower than that for CD. A vector encoding the CH3 combination of KK and a vector encoding the wild-type CH3 When trying to produce a mixture of AB and CD by co-expressing AC and This indicates that AD is highly likely to be mixed in.

[0327] Finally, we investigated the effect of either DEKK or charge-reversal mutants and knob-into-hole mutants. The prediction for co-expression of HA and HA2 with HA2 results in relatively low levels of undesired dual-specific variants. The results showed that the co-expression of these undesirable genes had a DDOCK score. There is a high probability that seeds will be produced.

[0328] Therefore, the combination of the DEKK combination CH3 and the charge-reversal combination CH3 ( E356K, D399K / K392'D, K409D') are substantially pure "AB and CD" and / or "AB It was concluded that this method is ideally suited to obtaining a mixture of antibodies with and without CC.

[0329] Next, to put the above results into practice, we developed a novel antibody that recognizes four targets / epitopes (AB and CD). A mixture of two bispecific molecules recognizing three targets / epitopes (AB and CC) was used. A mixture of one bispecific molecule and one monospecific molecule was generated. In the generation of the product, all four different VHs are paired with a common light chain, IGVK1-39. Although different VH / VL combinations can be combined, every individual VH / VL combination has a different specificity.

[0330] The mass difference between the (predicted) species is sufficient to allow native MS analysis. The expected species upon co-expression must be n The four individual VHs were designed to have mass values ​​that allowed them to be identified and separated by MS. Furthermore, the four selected VHs were grown in a mixture in addition to the two desired species. The mass difference is large enough to identify most of the possible contaminants. The list is shown in Table 23.

[0331] [Table 23]

[0332] As shown in Table 24, four different VHs were identified as either "DE" or "KK" constructs. or cloned into a vector containing a charge-reversal construct, and also co-expressed with several As before, all vectors encode the common light chain IGKV1-39. As shown above, the combination of the two vector sets A key requirement is that heavy chains expressed from two different sets of CH3 engineered vectors The key is to avoid the formation of "crossover" dimers. The heavy chain produced dimerizes with the heavy chain expressed by another set of vectors to form full-length IgG. The "interaction" between heavy chains containing charge reversal mutations and heavy chains containing DE or KK mutations To test for the possibility of differential dimer formation, control transfections were performed. It was said.

[0333] [Table 24]

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

[0335] [Table 25]

[0336] All purified protein samples obtained from transfections #1-11 were purified using SDS-PAGE. The results were analyzed by PAGE and three control samples were included (Figure 26). To identify all species in the samples, protein samples from transfections #9-11 were used. nMS analysis was performed.

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

[0338] Transfections #7 and #8 contained half molecules and full-length IgG in the protein samples. However, SDS-PAGE revealed that the total amount of IgG was The length of the dimer is either a DE / DE dimer, a DE / E356K:D399K dimer, or a DE / K392D:K409D dimer. It is not possible to deduce what this represents. Virtually no half molecules were observed in samples from group #9-11.

[0339] Figure 27 shows the results of the nMS analysis of transfections #9 and #11. The percentage of observed species and contaminating species was calculated by peak height. In 9, the predicted species “AB and CD” accounted for 97% of the mixture (30% AB and 67% CD). On the other hand, only about 3% of the BD was present (Figure 27A). In infection #11, the expected species "AB and CC" accounted for 94% of the mixture (33% AB and On the other hand, only 6% of the contaminating BC (4 0.1%) and AC (1.8%) (Figure 27B).

[0340] These data suggest that a second set of "orthogonal" vectors can be combined with DEKK. When used together, more complex mixtures such as "AB and CD" or "AB and CC" are possible. This allows the production of a mixture of different IgG and / or bispecific molecules. When the DEKK construct was used in combination with the IL-1 construct, very limited Only a small amount of "crossover" dimer formation occurs. It is expected that these low percentages of by-products can be further reduced.

[0341] Example 24: Single-dose pharmacokinetic study in mice Pharmacokinetics (pK To study the behavior of IgG, this study measured the pK parameters of three different IgG batches. The three batches contained: 1) the wild-type anti-tetanus toxin parent; Antibody 1337: 1337 (two MF1337 Fabs in a wild-type Fc backbone), 2) wild-type antibody The parent antibody 1516 of tetanus toxin: 1516 (wild-type Fc backbone with two MF1516Fabs) ), 3) CH3-engineered bispecific antibody carrying DEKK combination mutations in the Fc region TNF-α antibody 1516:1337 (MF1516Fab on the DE side and MF1337 on the KK side) Fab).

[0342] DEKK-bispecific antibody products are derived from the parent antibodies 1337:1337 and 1516. The specificity of the :1516 was chosen because it Based on previous studies, there is no pre-treatment serological response to these antibodies. Incidentally, if a serum reaction is present before prescription, it may be difficult to examine the research results. Furthermore, nMS allows for the identification of 1337:1337 (wild type) differences between the parental antibodies. type Fc), 1516:1337 (DEKK Fc) and 1516:1516 (wild-type Fc) c) One reason is that there is a sufficient mass difference to make it possible to distinguish between them.

[0343] Three batches of IgG were prepared as previously described, but with the addition of transferrin. The DNA used in the analysis was endotoxin-free to minimize the amount of endotoxin. The batch was then prepared using a toxin-free maxiprep kit. Concentration, aggregation, endotoxin and percentage of bispecific product were tested. It was shown to meet the acceptance criteria for use of the IgG batch in subsequent pK studies. That is, the IgG concentration after gel filtration was higher than 0.3 mg / ml, and the amount of aggregation was higher than 5%. Low endotoxin content of 3EU / mg protein, DEKK batches exceed 90% It contained bispecific IgG.

[0344] Native mass spectrometry of the gel filtration sample showed a high proportion of the predicted species. In sample 1516:1337, a small amount of DE:DE homodimer, estimated at about 2%, was present. Three IgG batches met the criteria for use in pK studies. It was concluded that

[0345] Three groups of female C57BL / 6J mice were used to compare pK parameters between the three batches. (Harlan, The Netherlands) with 1 mg / kg human IgG (5 ml / kg immunoglobulin G). The animals were 7-8 weeks old at the time of administration and approximately 18- The subjects weighed 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 temperatures below -20°C until analysis. Each group consisted of four mice. Each mouse consisted of three subgroups of mice, i.e., 12 mice per group. Samples were collected from 6 time points.

[0346] The welfare of the mice was determined in accordance with the European Community Directive on Animal Experiments (Directive 86 / 609 / EEC: the gene ral principles governing the use of animals in experiments of the European Commu nities) and Dutch legislation (Dutch Law on Animal Experiments: The Experiments on The animals were maintained in accordance with the Animal Welfare Act, 1997. This study was supported by the Office of Laboratory Animal Welfare of the National Institutes of Health. Standards for Humane Care and Use of Laboratory Animals, published by the and Use of Laboratory Animals), Identification Number 45859-01 (Expiration Date: April 30, 2015 ) was also observed.

[0347] Mice in group 1 received the monospecific IgG antibody 1516:1516 full-length (triangles). Mice in group 2 received the monospecific IgG antibody 1337:1337 full-length (squares). Group 3 mice had a DEKK-manipulated CH3 region (1516 on the DE side and 1337 on the KK side). ) and administered the full-length bispecific IgG antibody 1516:1337 (diamond).

[0348] Quantitative human IgG ELISA (ZeptoMetrix, NY USA; ELISA kit No. 0801182) was used. Quantitative analysis of monoclonal human antibodies in mouse serum by ELISA assay using Briefly, the ELISA assay is based on the following principle: The ISA plate is coated with anti-human IgG, and the human monoclonal antibody binds to the anti-human IgG. The bound antibody is then reacted with horseradish peroxidase (HRP) Visualization was performed using a conjugated polyclonal anti-human IgG antibody.

[0349] The optical density (OD) of each well is directly proportional to the amount of antibody in the serum sample. The results are shown in Figure 29. The full-length bispecific IgG carrying the DEKK mutation combination and the parental monospecific The CH3 mutations in the DEKK-diabody were observed to be remarkably similar. The DEKK mutants behave like wild-type IgG. Wow.

[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. 1. A method for producing two different monospecific bivalent antibodies comprising an IgG-like CH3 domain in a single host cell, comprising: a first nucleic acid molecule encoding a first antibody heavy chain comprising the following substitutions in the CH3 domain, according to the EU numbering system: (i) D399R or D399K and (ii) K409E or K409D; and A second nucleic acid molecule encoding a second antibody heavy chain comprising a wild-type CH3 domain. providing a host cell comprising Culturing the host cell to express the first and second antibody heavy chains, wherein the CH3 domains of the first antibody heavy chain selectively pair and the CH3 domains of the second antibody heavy chain selectively pair to produce antibodies comprising the first antibody heavy chain and antibodies comprising the second antibody heavy chain; and recovering the two different monospecific bivalent antibodies from the culture.

2. 10. The method of claim 1, The method, wherein said first antibody heavy chain further comprises the substitution K392D or K392E, according to the EU numbering system.

3. 10. The method of claim 1, The method further comprising providing said host cell with a nucleic acid molecule encoding a common light chain.

4. 10. The method of claim 1, The method, wherein each of the antibody heavy chains comprising said encoded CH3 domain further comprises a variable region that recognizes an epitope.

5. A mixture of two different monospecific bivalent antibodies obtainable by the method of claim 1.

6. A mixture of two different monospecific bivalent antibodies, a mixture wherein the first monospecific bivalent antibody comprises two heavy chains that contain the following substitutions in the CH3 domain, according to the EU numbering system: (i) D399R or D399K and (ii) K409E or K409D, and the second monospecific bivalent antibody comprises two heavy chains with wild-type CH3 domains.

7. A recombinant host cell for producing two different monospecific bivalent antibodies, comprising: the host cell comprises one or more nucleic acids encoding at least a first and a second antibody heavy chain; the first antibody heavy chain comprises the following substitutions in an IgG-like CH3 domain, according to the EU numbering system: (i) D399R or D399K and (ii) K409E or K409D; the second antibody heavy chain comprises a wild-type IgG-like CH3 domain; Recombinant host cells.

8. 8. The recombinant host cell of claim 7, The recombinant host cell, wherein said first antibody heavy chain further comprises the substitution K392D or K392E according to the EU numbering system.

9. 8. The recombinant host cell of claim 7, A recombinant host cell, wherein the host cell further comprises a nucleic acid encoding a common light chain.

10. A pharmaceutical composition comprising a mixture of two different monospecific bivalent antibodies according to claim 5 or 6 and a pharmaceutically acceptable carrier.

11. A pharmaceutical composition comprising two different monospecific bivalent antibodies produced in the recombinant host cell of claim 7.

12. 1. A method for producing a host cell for producing two different monospecific bivalent antibodies, comprising: The method comprises: introducing into the host cell one or more nucleic acids encoding at least a first and a second antibody heavy chain; the first antibody heavy chain comprises the following substitutions in an IgG-like CH3 domain, according to the EU numbering system: (i) D399R or D399K and (ii) K409E or K409D; the second antibody heavy chain comprises a wild-type IgG-like CH3 domain; method.

13. 13. The method of claim 12, The method, wherein said first antibody heavy chain further comprises the substitution K392D or K392E, according to the EU numbering system.

14. 13. The method of claim 12, The method further comprising introducing into said host cell a nucleic acid encoding a common light chain.

15. A culture of the recombinant host cell of claim 7.

Images